{
    "0209/astro-ph0209470_arXiv.txt": {
        "abstract": "Magneto-convection simulations on meso-granule and granule scales near the solar surface are used to study small scale dynamo activity, the emergence and disappearance of magnetic flux tubes, and the formation and evolution of micropores.  From weak seed fields, convective motions produce highly intermittent magnetic fields in the intergranular lanes which collect over the boundaries of the underlying meso-granular scale cells.  Instances of both emerging magnetic flux loops and magnetic flux disappearing from the surface occur in the simulations.  We show an example of a flux tube collapsing to kG field strength and discuss how the nature of flux disappearance can be investigated.  Observed stokes profiles of small magnetic structures are severely distorted by telescope diffraction and seeing.  Because of the strong stratification, there is little recycling of plasma and field in the surface layers.  Recycling instead occurs by exchange with the deep layers of the convection zone.  Plasma and field from the surface descend through the convection zone and rise again toward the surface.  Because only a tiny fraction of plasma rising up from deep in the convection zone reaches the surface due to mass conservation, little of the magnetic energy resides in the near surface layers.  Thus the dynamo acting on weak incoherent fields is global, rather than a local surface dynamo. ",
        "introduction": "We model magneto-convection in a small domain near the solar surface by solving the partial differential equations for mass, momentum and internal energy conservation and the induction equation for the vector potential.  Our goal is to make a realistic representation of the solar surface (Stein \\& Nordlund 2000).  Our domain is 6 $\\times$ 6 Mm horizontally and extends from the temperature minimum, 0.5 Mm above continuum optical depth unity, to 2.5 Mm below the visible surface, using a grid 253 $\\times$ 253 $\\times$ 163 grid points, which gives a horizontal resolution of 25 km and a vertical resolution of 15 km near the surface increasing to 35 km at the bottom.  The initial state was a snapshot of non-magnetic solar convection on which was imposed a uniform magnetic field: either a horizontal seed field of 1 G or 30 G, or a vertical field of 400 G. The boundary conditions are periodicity in horizontal directions, open boundaries for the fluid in the vertical direction, and the magnetic field at the top tends toward a potential field.  The magnetic bottom boundary condition for horizontal seed fields was that inflows advect in horizontal 1G or 30G field, while in outflows the vector potential is advanced in time from the induction equation with the current calculated using spline derivatives with the cubic spline condition that the third derivative is continuous.  The magnetic bottom boundary condition for the vertical field was that the field tend toward the vertical.  Rotation and coriolis forces are neglected, because on this small scale of mesogranulation, with a depth of only 3 Mm, the flows don't feel the rotation. ",
        "conclusions": ""
    },
    "0209/astro-ph0209193_arXiv.txt": {
        "abstract": "We investigate chemical abundance variations along the sightlines through 13 damped \\lya systems (DLA). We introduce a technique designed to identify abundance variations in multiple velocity bins along the sightlines and perform a series of Monte-Carlo simulations to derive quantitative limits and values. The majority of these DLA have very uniform relative abundances: 95$\\%$ c.l.\\ upper limits of 0.2~dex dispersion along the sightline and best fit values typically $\\lesssim 0.1$~dex. This surprising result indicates the gas comprising individual DLA has similar enrichment history, nearly identical differential depletion, and small abundance variations relating to ionization corrections.  This uniformity contrasts with stellar abundance variations observed within the Galaxy and with the differences in abundances observed between galaxies of the Local group. It also contrasts with variations in differential depletion along sightlines through the SMC, LMC, and Milky Way. The results constrain the process of metal production and dust formation in the early universe and reflect on processes of galaxy formation. In terms of depletion the observations indicate a very low porosity of significantly depleted gas, substantially lower than the filling factor observed in present-day metal-poor galaxies. Finally, the observed chemical uniformity may present a difficult challenge to scenarios which assume individual DLA are comprised of multiple protogalactic clumps. ",
        "introduction": "\\label{sec-intro}  Studies of chemical abundance ratios in Galactic metal-poor stars address processes of nucleosynthesis, star formation, and ultimately galaxy formation in the early universe. Of particular importance is revealing variations or trends in the relative abundances which identify distinct stellar populations, unique nucleosynthetic sites, and/or provide insight into the time-scales and channels of star formation. In very low metallicity ([Fe/H]~$< -3$) Galactic stars, for example, observers have identified important trends in the Fe-peak elements and large variations in several abundance ratios (e.g.\\ Co/Fe, Sr/Fe) which suggest unique nucleosynthetic processes and inhomogeneous chemical enrichment \\citep[e.g.][]{mcw95,johnson02}. In contrast, the abundance patterns of the majority of Galactic stars with $-2 < $~[Fe/H]~$< -1$ are remarkably similar \\citep{gratton91,fulbright00}, suggesting a rapid enrichment phase for the Galactic halo over this metallicity range. At higher metallicity,  there are important trends in the relative abundances, in particular the steady decline of the $\\alpha$-elements (O,S,Si) to the Fe-peak, which mark the onset of Type~Ia SN enrichment \\citep{tinsley79}. Comparisons of these relative abundances with extragalactic stellar populations help map paths of galaxy formation \\citep[e.g.][]{matteucci01}. In the few galaxies where detailed studies can be performed \\citep[the SMC and dwarf spheroidals;][]{venn98,shetrone01,bonifacio00}, one often observes significant differences with stars at the same metallicity in the Milky Way.  Future studies of entire stellar populations within these galaxies will help refine their star formation histories.   An examination of relative abundances of gas in local galaxies complements these stellar abundance studies. Sightlines through the Galactic ISM probe gas with a range of physical conditions (e.g.\\  volume density, dust-to-gas ratio, ionization state). Therefore, the observed {\\it gas-phase} abundance ratios along a sightline may vary by an order of magnitude as the sightline penetrates clouds arising in various phases of the ISM \\citep[e.g.][]{sav96}. Differences in nucleosynthetic enrichment or photoionization are likely to be small such that the variations primarily reflect differences in depletion level or dust composition. The LMC and SMC also exhibit variations in these gas-phase abundance ratios \\citep{welty99,welty01} and together the observations impact models for the formation and distribution of dust as well as the physical state of galactic interstellar media.  Observations of the damped \\lya systems (DLA) -- quasar absorption line systems with $\\N{HI} \\geq 2 \\sci{20} \\cm{-2}$ -- probe the ISM of high $z$ galaxies.  Surveys of the DLA reveal chemical evolution in the early universe \\citep{ptt94,ptt97,pw00}, and examine the dust depletion, nucleosynthetic enrichment, and ionization of these protogalaxies \\citep{lu96,vladilo98,pw02,pro02}. \\cite{pw02} recently reviewed the chemical abundances of the DLA and emphasized the uniformity in the relative abundance patterns (e.g.\\ Si/Fe, Ni/Fe, Zn/Si) of their $\\approx 30$ systems. Even though damped \\lya metallicities span over an order of magnitude, the systems show very similar abundance ratios suggesting protogalaxies have common enrichment histories. The authors do identify a mild trend of increasing Si/Fe and Zn/Fe with increasing metallicity which is well explained by the effects of dust depletion, yet even this trend spans only a factor of $\\approx 3$ in the ratios. In addition to the uniformity in abundance patterns from galaxy to galaxy, \\cite{pw02} asserted that this uniformity holds along the sightlines penetrating each protogalaxy.  That is, the individual 'clouds' comprising metal-line profiles show the same relative abundances {\\it within} a given damped \\lya system. If verified, this would have important implications for the ISM of high $z$ galaxies and the enrichment of gas in the early universe.  \\cite{pro96} first quantitatively investigated variations in the chemical abundances of a single damped \\lya system. They compared Zn, Si, Fe, Ni, and Cr ionic column densities along the observed velocity profiles using the apparent optical depth method \\citep{sav91}. Their analysis showed that the chemical abundances were uniform to within statistical uncertainties. They then argued that the absence of significant variations demonstrates a low depletion level within this protogalaxy. More recently, \\cite{lopez02} performed the first detailed line-profile analysis of the relative abundances for a different $z>2$ damped system and also found nearly constant column density ratios among the components comprising the line-profile solution (see also Petitjean, Srianand, \\& Ledoux 2002). Together, these studies argue the gas within high $z$ protogalaxies has a similar enrichment history and uniform differential depletion.  In this paper, we search for variations in the relative abundances along the sightlines of \\ndla\\ damped \\lya systems. These systems were selected to have high signal-to-noise velocity profiles with velocity widths exceeding 40~km\\,s$^{-1}$.  The latter criterion ensures a large enough baseline to examine variations within the protogalaxy. We introduce a quantitative method which can be applied to our observations as well as mock spectra derived from numerical simulations and semi-analytic models of galaxy formation.  The technique assesses changes in the physical conditions and enrichment histories of the gas within each damped system. In turn, our results place valuable constraints on the various morphological models of the damped \\lya systems and, ultimately, scenarios of galaxy formation.   \\begin{table*} \\footnotesize \\begin{center} \\caption{{\\sc OBSERVATIONAL SAMPLE \\label{tab:summ}}} \\begin{tabular}{lcccccccccccc} \\tableline \\tableline QSO &$z_{abs}$ & [M/H]\\tablenotemark{a} & $\\delv$ & Transition & [X/Y]$_T$ & $\\delta v_{bin}$ & $N_{b}$ & $\\sigma_{bin}$\\tablenotemark{b} & $\\chi^2_\\nu$ & $\\Delta_{sngl}$ & $\\Delta_{all}$ & $\\Delta_{best}$ \\\\ & & & (km\\,s$^-1$) & Pairs & (dex) & (km\\,s$^{-1}$)  & & (dex) & & (dex) & (dex) & (dex)\\\\ \\tableline \\tskip PH957     & 2.309 & --1.46 & 54 & Zn\\,II 2026 & 0.22 & 10 & 3 & 0.18 & 1.65 & 0.24 & 0.47 & 0.03 \\\\ &       &        &    & Ni\\,II 1741 \\\\ Q0201+36  & 2.463 & --0.41 & 202& Si\\,II 1808 & 0.45 & 20 & 12& 0.07 & 3.21 & 0.38 & 0.15 & 0.09 \\\\ &       &        &    & Fe\\,II 1608 \\\\ Q0347--38 & 3.025 & --1.17 & 94 & Si\\,II 1808 & 0.50 & 10 & 5 & 0.47 & 11.0 & 0.74 & 0.90 & 0.39 \\\\ &       &        &    & Fe\\,II 1608 \\\\ Q0458--02 & 2.040 & --1.19 & 86 & Zn\\,II 2026 & 0.39 & 20 & 4 & 0.48 & 4.81 & 0.35 & 0.45 & 0.15 \\\\ &       &        &    & Cr\\,II 2056 \\\\ HS0741+47 & 3.017 & --1.69 & 40 &  S\\,II 1259 & 0.24 & 10 & 5 & 0.08 & 2.23 & 0.23 & 0.22 & 0.07 \\\\ &       &        &    & Fe\\,II 1608 \\\\ PSS0957+33& 4.180 & --1.50 & 346& Si\\,II 1304 & 0.37 & 20 & 6 & 0.26 & 5.34 & 0.70 & 0.68 & 0.29 \\\\ &       &        &    & Fe\\,II 1608 \\\\ Q1223+17  & 2.466 & --1.59 & 94 & Si\\,II 1808 & 0.13 & 10 & 10& 0.11 & 1.82 & 0.40 & 0.19 & 0.09 \\\\ &       &        &    & Ni\\,II 1751 \\\\ Q1331--17 & 1.776 & --1.30 & 70 & Zn\\,II 2026 & 0.82 & 10 & 9 & 0.23 & 22.8 & 0.66 & 0.41 & 0.24 \\\\ &       &        &    & Fe\\,II 2374 \\\\ GB1759+75 & 2.625 & --0.82 & 78 & Si\\,II 1808 & 0.35 & 10 & 8 & 0.18 & 2.50 & 0.44 & 0.27 & 0.11 \\\\ &       &        &    & Ni\\,II 1751 \\\\ Q2206--19 & 1.920 & --0.42 &130 & Si\\,II 1808 & 0.40 & 20 & 7 & 0.06 & 1.17 & 0.24 & 0.15 & 0.04 \\\\ &       &        &    & Fe\\,II 1611 \\\\ Q2230+02  & 1.864 & --0.74 &177 & Si\\,II 1808 & 0.23 & 20 & 7 & 0.08 & 1.30 & 0.31 & 0.18 & 0.03 \\\\ &       &        &    & Ni\\,II 1741 \\\\ Q2231--00 & 2.066 & --0.86 &127 & Si\\,II 1808 & 0.28 & 20 & 4 & 0.08 & 0.93 & 0.26 & 0.33 & 0.03 \\\\ &       &        &    & Ni\\,II 1741 \\\\ Q2359--02 & 2.154 & --1.58 &127 & Si\\,II 1526 & 0.32 & 20 & 4 & 0.08 & 0.97 & 0.24 & 0.32 & 0.04 \\\\ &       &        &    & Fe\\,II 1608 \\\\ \\tableline \\end{tabular} \\tablenotetext{a}{Metallicity derived from Zn, Si, or S} \\tablenotetext{b}{Logarithmic RMS dispersion in X/Y} \\tablecomments{$\\Delta_{sngl}, \\Delta_{all}, \\Delta_{best}$ are measures of the intrinsic abundance variation along each sightline.  See $\\S$~3} \\end{center} \\end{table*} ",
        "conclusions": "We have performed a quantitative analysis of chemical abundance variations along the sightlines through 13 damped \\lya systems.  These systems each have velocity width $\\delv > 40 \\mkms$, i.e., large enough base-line to examine chemical uniformity. We performed several sets of Monte-Carlo simulations to place upper limits on relative abundance variations. To our surprise, the majority of DLA exhibit a high degree of chemical uniformity. In most cases, the dispersion in intrinsic abundances -- the combined effects of all abundance variations -- is less than 0.2~dex and the best values are generally $<0.1$~dex.  For the ratios examined in this paper, there are three potential sources of intrinsic abundance variation: (1) nucleosynthetic enrichment; (2) differential depletion; and (3) photoionization corrections. In terms of depletion, the observed DLA uniformity contrasts with variations in differential depletion observed along sightlines through the SMC, LMC, and Milky Way. Although the latter galaxies have higher metallicity than most DLA, the difference in depletion variations is not explained solely by differences in metallicity; there is no correlation between abundance variations and metallicity in the DLA (Figure~\\ref{fig:mtl}). For an unknown reason or reasons (perhaps related to dust destruction mechanisms), the nature of dust at high $z$ is qualitatively different from the local universe. Considering nucleosynthetic processes, the DLA abundances are more uniform than the dispersion in nucleosynthetic enrichment of the Milky Way as traced by stellar abundances. Furthermore, one observes a greater dispersion between stars in various galaxies within the local group than that observed in the gas of the DLA. Finally, photoionization calculations suggest that Si$^+$/Fe$^+$ can vary by over 0.2~dex between a neutral gas and a highly ionized gas \\citep{howk99,vladilo01,pro02}.  Our observations suggest that the cross-section of H\\,II regions or similarly ionized gas is small for galaxies with $\\N{HI} \\geq 2 \\sci{20} \\cm{-2}$.   The chemical uniformity of the DLA poses an important constraint on the nature of high $z$ protogalaxies.  We have argued that 'single-system' scenarios might reasonably account for the observed abundance invariance.  Central to this conclusion, however, is that these protogalaxies have a small filling factor of highly depleted gas.  If star formation is linked to dust-depleted molecular clouds, then our results indicate the gas relevant to star formation encompasses a very small cross-section. This conclusion is consistent with the low fraction of DLA showing molecular gas \\citep{petit00}. In contrast with the single-system models, we contend the observed uniformity presents a unique challenge to the multiple-clump scenarios favored by CDM. These protogalactic clumps or 'satellites' do not share a common gas reservoir and should have unique physical characteristics (density, metallicity) and enrichment histories. It remains to be demonstrated whether these clumps might express very similar differential depletion and nucleosynthetic enrichment patterns.   Before concluding, we wish to comment on several future observational efforts which could improve upon this paper. First, we emphasize that the results for many of the DLA in the current sample are limited by signal-to-noise. Follow-up observations of the brighter quasars might reveal uniformity at the 0.01~dex level and place qualitatively tighter constraints on processes of metal production and dust formation. Similarly, higher resolution data would enable an investigation in velocity bins of a few km\\,s$^{-1}$. Second, we would like to repeat our analysis using Zn and Si separately to better isolate the effects of depletion and nucleosynthesis. Variations in the Zn/Si ratio have been observed in at least one low $z$ DLA \\citep{pettini00} and we would like to examine similar variations at high $z$. Third, it is also important to examine abundance ratios which relate to physical processes separate from the Type~Ia vs.\\ Type~II enrichment and differential depletion emphasized in this paper.  For example, a study on the variations of the N/$\\alpha$ ratio within DLA would impact our understanding of star formation timescales and possibly the universality of the initial mass function \\citep[e.g.][]{pro02b}. Similarly, because the Ar$^0$/Si$^+$ or O$^0$/Si$^+$ ratios are particularly sensitive to photoionization, an analysis of these ions would help reveal processes of ionization within the DLA. Finally, a comparison of the C\\,II$^*$ fine-structure line with Si and Fe resonance transitions bears on the nature of the protogalactic ISM and ultimately star formation rates \\citep{wolfe02}. Together these observations would help reveal the detailed physical properties of the gas comprising DLA protogalaxies."
    },
    "0209/astro-ph0209250_arXiv.txt": {
        "abstract": "We provide the whole set of Lick indices from CN$_1$ to TiO$_2$ in the wavelength-range $4000\\la\\lambda\\la 6500$~\\AA\\ of Simple Stellar Population models with, for the first time, variable element abundance ratios, $[\\aFe]=0.0,\\:0.3,\\:0.5$, $[\\aCa]=-0.1,\\:0.0,\\:0.2,\\:0.5$, and $[\\aN]=-0.5,\\:0.0$. The models cover ages between $1$ and $15$~Gyr, metallicities between 1/200 and 3.5 solar. The impact from the element abundance changes on the absorption-line indices are taken from \\citet{TB95}, using an extension of the method introduced by \\citet{Traetal00a}. Our models are free from the intrinsic \\aFe\\ bias that was imposed by the Milky Way template stars up to now, hence they reflect well-defined \\aFe\\ ratios at all metallicities.  The models are calibrated with Milky Way globular clusters for which metallicities and \\aFe\\ ratios are known from independent spectroscopy of individual stars. The metallicities that we derive from the Lick indices \\Mgb\\ and Fe5270 are in excellent agreement with the metallicity scale by \\citet{ZW84}, and we show that the latter provides total metallicity rather than iron abundance.  We can reproduce the relatively strong CN-absorption features \\CNone\\ and \\CNtwo\\ of galactic globular clusters with models in which nitrogen is enhanced by a factor three.  An enhancement of carbon, instead, would lead to serious inconsistencies with the indices \\Mgone\\ and \\FeC. The calcium sensitive index Ca4227 of globular clusters is well matched by our models with $[\\CaFe]= 0.3$, including the metal-rich Bulge clusters NGC~6528 and NGC~6553. From our \\aFe\\ enhanced models we infer that the index [MgFe] defined by \\citet{G93} is quite independent of \\aFe\\ but still slightly decreases with increasing \\aFe. We find that the index ${\\rm [MgFe]}^{\\prime} \\equiv\\sqrt{\\Mgb\\ (0.72\\cdot {\\rm Fe5270}+0.28\\cdot{\\rm Fe5335})}$, instead, is completely independent of \\aFe\\ and serves best as a tracer of total metallicity.  Searching for blue indices that give similar information as \\Mgb\\ and \\Fe, we find that \\CNone\\ and Fe4383 may be best suited to estimate \\aFe\\ ratios of objects at redshifts $z\\sim 1$. ",
        "introduction": "\\label{sec:intro} The Lick system \\citep{Buretal84,Fabetal85} defines absorption-line indices at medium resolution ($\\sim 8$~\\AA) that can be used---through the comparison with stellar population models---to derive ages and metallicities of stellar systems.  Interestingly, the indices \\Mgb\\ and \\Mgtwo\\ of early-type galaxies yield higher metallicities (and younger ages) than the indices Fe5270 and Fe5335 (\\citealt{P89}; \\citealt*{WFG92}; \\citealt*{DSP93}; \\citealt{CD94}; \\citealt{BP95}; \\citealt*{FFI95}; \\citealt{Jor99}; \\citealt{Mehetal98}; \\citealt{Ku00}; \\citealt{Lonetal00}; and others).  The most straightforward qualitative interpretation of these strong Mg-indices and/or weak Fe-indices is that the stellar populations in elliptical galaxies have high Mg/Fe element ratios (or \\aFe\\ ratios if Mg is taken as representative of $\\alpha$-elements) with respect to the solar values \\citep{WFG92}. This finding strongly impacts on the theory of galaxy formation, as super-solar \\aFe\\ ratios require short star formation time-scales ($\\la 1$~Gyr, \\citealt{Ma94}; \\citealt*{TGB99}), that are not achieved by current models of hierarchical galaxy formation \\citep{Th99a,TK99}.  However, there exist two major caveats about this conclusion. 1) Lick indices have very broadly defined line windows ($\\sim 40$~\\AA). Each index actually contains a large number of absorption features from various elements, so that the direct translation into element abundances is not very straightforward \\citep*{Greggio97,TCB98}. 2) The stellar library \\citep{Woretal94} used in stellar population models to compute Lick indices contains only very few stars with metallicities above solar. An additional complication is that in these libraries \\aFe\\ is not independent of \\FeH\\ (see Section~\\ref{sec:bias}).  In order to resolve these ambiguities, \\citet{Maretal02} compare the Lick indices of Simple Stellar Population (SSP) models with data of metal-rich globular clusters of the Galactic Bulge \\citep{Puzetal02}, the ages and and element abundances of which are known from high-resolution stellar spectroscopy. They find that metal-rich, \\aFe\\ enhanced Bulge clusters show the same features as early-type galaxies: their Mg indices are stronger than predicted by SSP models at a given Fe index value.  This result is empirical evidence that Mg and Fe indices indeed trace \\aFe\\ element ratios. \\citet{Maretal02} verify the uniqueness of interpreting the strong Mg indices and weak Fe indices in elliptical galaxies in terms of Mg over Fe element overabundance. They show that alternative explanations like uncertainties in stellar evolution and SSP modelling, or a significant steepening of the initial mass function (IMF) do either not reproduce the observed indices or violate other observational constraints. \\citet{Maretal02} further show that the standard models reflect variable element abundance ratios at the various metallicities, in particular super-solar \\aFe\\ ratios at sub-solar metallicities.  Motivated by these results, we construct stellar population models for various and well-defined element abundance ratios. We present the whole set of Lick indices (\\CNone, \\CNtwo, Ca4227, G4300, Fe4383, Ca4455, Fe4531, \\FeC, \\Hb, Fe5015, \\Mgone, \\Mgtwo, \\Mgb, Fe5270, Fe5335, Fe5406, Fe5709, Fe5782, Na~D, \\TiOone, and \\TiOtwo) of SSP models with the \\aFe\\ ratios $[\\aFe]=0.0,\\:0.3,\\:0.5$. The models cover ages from $1$ to $15$~Gyr, and total metallicities from $[\\ZH]=-2.25$ to $0.65$. In these models, the elements nitrogen and calcium are enhanced in lockstep with the other $\\alpha$-elements, hence $[\\aN]=0.0$ and $[\\aCa]=0.0$. Additionally, we provide models with variable \\aN\\ and \\aCa\\ ratios, $[\\aN]=-0.5$ and $[\\aCa]=-0.1,\\:0.2,\\:0.5$.  The impact from the element abundance changes on the absorption-line indices are taken from \\citet[][ hereafter TB95]{TB95}, using an extension of the method introduced by \\citet[][ hereafter T00]{Traetal00a}.  The models are calibrated with the globular cluster data of \\citet{Puzetal02}.  The present models now allow for the unambiguous derivation of SSP ages, metallicities, and element abundance, in particular \\aFe, ratios.  The paper is organised as follows.  In Section~\\ref{sec:construction} we describe the construction of the models, and introduce the main input parameters. In Section~\\ref{sec:calibration} we present the model results and their calibration with globular cluster data.  If the reader is predominantly interested in the application of the present SSP models, for the first reading we recommend to skip Section~\\ref{sec:construction}, and to focus on the summary given in Section~\\ref{sec:modelsum}.  \\medskip The models for selected ages are provided in the tables in the appendix. Their complete versions are available electronically via ftp at {\\tt ftp.mpe.mpg.de} in the directory {\\tt people/dthomas/SSPs}, via WWW at {\\tt ftp://ftp.mpe.mpg.de/people/dthomas/SSPs}, or email to {\\tt dthomas@mpe.mpg.de}. ",
        "conclusions": "\\label{sec:summary} We present a comprehensive set of new generation stellar population models of Lick absorption line indices, which for the first time include abundance ratios different from solar. We computed the 21 Lick indices \\CNone, \\CNtwo, Ca4227, G4300, Fe4383, Ca4455, Fe4531, \\FeC, \\Hb, Fe5015, \\Mgone, \\Mgtwo, \\Mgb, Fe5270, Fe5335, Fe5406, Fe5709, Fe5782, Na~D, \\TiOone, and \\TiOtwo\\ in the wavelength range $4000\\la \\lambda\\la 6500$~\\AA. Models are provided with: $[\\aFe]=0.0,\\:0.3,\\:0.5$, $[\\aCa]=-0.1,\\:0.0,\\:0.2,\\:0.5$, and $[\\aN]=-0.5,\\:0.0$; ages from 1 to 15~Gyr; total metallicities from 1/200 to 3.5 solar ($-2.25\\leq [\\ZH]\\leq 0.67$).  The models are based on the evolutionary synthesis technique described in \\citet{Ma98}. The \\aFe\\ enhanced mixtures are obtained by increasing the abundances of $\\alpha$-group elements and by decreasing the abundances of the Fe-peak elements, such that total metallicity is conserved.  The impact from these element abundance variations on the absorption line indices is taken from \\citet{TB95}, using an extension of the method introduced by \\citet{Traetal00a}. Most importantly, we take into account that the empirical stellar libraries used to compute model indices follow the chemical enrichment history of the Milky Way, and are therefore biased towards super-solar \\aFe\\ ratios at sub-solar metallicities. We corrected for this bias, so that the models presented here have well-defined \\aFe\\ ratios at all metallicities.  We take particular care at calibrating the models with galactic globular clusters, for which ages, metallicities, and element abundance ratios are known from independent sources. Our \\aFe\\ enhanced models with $[\\aFe]=0.3$ (and 12~Gyr age) perfectly reproduce the positions of the globular cluster data in the \\Mgb-\\Fe\\ diagram up to solar metallicities \\citep[see also][]{Maretal02}.  The total metallicities for the sample clusters that we derive from these indices are in excellent agreement with the \\citet{ZW84} metallicity scale. We point out that the latter most likely reflects total metallicity rather than iron abundance, because it is obtained essentially by averaging the abundances derived from the Mg triplet near 5175~\\AA\\ and the Fe blend at 5270~\\AA\\ \\citep{Cohen83,ZW84}. This aspect needs to be emphasized, as with the \\aFe\\ enhanced models we are now in the position to distinguish total metallicity [\\ZH] and iron abundance [Fe/H].  By means of our calibrated \\aFe\\ enhanced models, we confirm that the index [MgFe], suggested by \\citet{G93} to balance \\aFe\\ ratio effects, is almost independent of \\aFe. As it modestly decreases with increasing \\aFe, however, we define the slightly modified index \\[ {\\rm [MgFe]}^{\\prime}\\equiv \\sqrt{\\Mgb\\ (0.72\\cdot {\\rm Fe5270}+0.28\\cdot {\\rm Fe5335})}\\ , \\] which is completely independent of \\aFe, and hence an even better tracer of total metallicity. We further show that the linear correlation between \\Mgtwo\\ and metallicity at old ages derived empirically by \\citet{BH90} is valid up to $\\sim 1/3$ solar metallicity, but underpredicts \\Mgtwo\\ indices at metallicities above that threshold.  It turns out to be hard to find indices that correlate with \\aFe\\ as well as the intensively studied indices \\Mgone, \\Mgtwo, and \\Mgb. Promising alternatives are the blue indices \\CNone\\ and \\CNtwo\\ that also increase with increasing \\aFe\\ ratio, mainly because of an anti-correlation with Fe abundance. With the caveat that \\CNone\\ and \\CNtwo\\ are additionally sensitive to C and N abundances, they can be regarded to be complementary to the indices \\Mgone, \\Mgtwo, and \\Mgb. Alternatives to the iron indices Fe5270 and Fe5335, the strengths of which decrease with increasing \\aFe\\ ratio, are easier to find. The best cases are the indices Fe4383, Fe4531, Fe5015, and Fe5709.  The indices \\CNone, \\CNtwo, and Ca4227 of globular clusters are very interesting, particular cases.  We find that the relatively strong CN features observed in globular clusters require models in which nitrogen is enhanced by a factor three relative to the $\\alpha$-elements, hence $[\\aN]=-0.5$.  This is in agreement with early suggestions by \\citet{DGC83} and \\citet{Renzini83}, that stars in globular clusters may accrete carbon and/or nitrogen enriched ejecta from the surrounding AGB stars \\citep{RV81}.  The good calibration of other indices like \\Mgone, \\Mgb\\ or \\Fe\\ is not affected by a variation of the \\aN\\ ratio, as these indices are not sensitive to nitrogen abundance.  We note that an enhancement of carbon abundance, instead, would lead to serious inconsistencies with \\Mgone.  Interestingly, also Ca4227 is sensitive to nitrogen abundance, and the globular cluster data of this index are also best reproduced by the model with increased nitrogen abundance.  \\medskip To conclude, the stellar population models presented here make it possible, for the first time, to study in detail individual element abundance ratios of unresolved stellar populations. In particular, total metallicity is now a well-defined quantity. In an accompanying paper (D.~Thomas et al., in preparation), we use these models to derive quantitatively \\aCa\\ and \\CaFe\\ ratios of the stellar populations in elliptical galaxies from their Ca4227, \\Mgb, and \\Fe\\ indices.  Interesting for galaxy formation will also be to investigate element abundance ratios of galaxies at earlier stages of their evolution. On the basis of the calibration carried out in this paper, we suggest that the combination of the blue Lick indices \\CNone\\ and Fe4383 may be best suited to estimate \\aFe\\ ratios of objects at redshifts $z\\sim 1$."
    },
    "0209/astro-ph0209299_arXiv.txt": {
        "abstract": "{We present a phenomenological approach to the study of disk galaxy evolution, based on i) a detailed modelling of the Milky Way (used as a prototype disk galaxy) and ii) an extension of the model to other disks through some simple scaling relations, obtained in the framework of Cold Dark matter models. The main conclusion is that, on average,  massive disks have formed the bulk of their stars earlier than their lower mass counterparts. It is not yet clear why the ``star formation hierarchy'' has been apparently opposite to the ``dark matter assembly'' hierarchy. } ",
        "introduction": "\\label{sec:intro}  Accordingly to the currently popular scenario for galaxy formation (pioneered by White and Rees 1978) the dark haloes of galaxies form hierarchically by the gravitational clustering of non-dissipative dark matter, while the luminous parts form through a combination of gravitational clustering and dissipative collapse (which may be affected by feedback). The major uncertainty in this scenario concerns the baryonic component, since the physics of star formation and feedback are very poorly understood at present, thus requiring a parametric approach to the problem. Semi-analytic models of galaxy formation help to explore large regions of the relevant parameter space and have produced quite encouraging results (e.g. Cole et al. 2000); still, they have not yet managed to reproduce succesfully some key observed properties of disks (van den Bosch 2002) or ellipticals (Thomas et al. 2002).  Simple (i.e. not dynamical) models of (chemical and/or photometric) galaxy evolution use, in general, fewer free parameters than semi-analytical models. If the number of observables that are successfully reproduced is (much) larger than the number of free parameters, it is reasonable to assume that the galaxian histories produced by such models may indeed match the real ones.  In that spirit, we have developed a simple approach to disk galaxy evolution, based on i) a detailed model of the Milky Way  (used as a prototype, Sec. 2) and ii) an extension to other disks trough some simple scaling relations (Sec. 3). The main conclusion (Sec. 4) is that massive disks have formed the bulk of their stars earlier than low mass ones. Whether this ``star formation hierarcy'' is compatible with the ``dark matter assembly'' hierarchy remains to be demonstrated. ",
        "conclusions": "\\label{sec:conclusion}  In the currently popular paradigm of hierarchical galaxy formation, low mass dark matter haloes form first, while more massive ones are formed later through accretion and merging; in principle, baryons are supposed to follow the dark matter, but their fate is largely unknown at present, due to a lack of a reliable theory of star formation (and feedback).  In recent cosmological hydro-simulations by Nagamine et al. (2001) it is found that star formation in small galaxies has stopped many Gyr ago, in clear contradiction with observations of local galaxies. On the other hand, van den Bosch (2002) finds that semi-analytical models, even with  feedback, produce massive disks that are systematically bluer than their lower mass counterparts, again in contradiction with observations; the reason of the failure is obviously related to the fact that the mass accretion histories of baryons are largely dictated by the hierarchical clustering of dark matter (e.g. Avila-Reese and Firmani 2000). Once gas becomes available  it forms rapidly stars; feedback can only delay star formation for a short time (shorter than the several Gyr that are observationally required to obtain small disks bluer than massive ones).  Our simple, ``hybrid'' model for disk evolution, calibrated on the MW, suggests that on average massive disks have formed the bulk of their stars several Gyr earlier than low mass ones. Their predictions match successfully most currently available observables, including data from surveys at intermediate redshifts (Boissier and Prantzos 2001).  At present, and despite claims to the contrary, there is no satisfactory explanation (at least, not a published one) for the observables presented in Fig. 3 in the framework of hierarchical galaxy formation. It remains to be shown why star formation in galaxies apparently followed an ``inverted hierarchy'' w.r.t the dark matter asembly. Feedback offers  an obvious solution to that problem, but the required delay timescales appear unphysically large."
    },
    "0209/astro-ph0209585_arXiv.txt": {
        "abstract": "The optical lightcurve of GRB~010222 exhibited one of the slowest decays of any gamma-ray burst to date. Its broadband properties have been difficult to explain with conventional afterglow models, as they either require the power law index of the underlying electron energy distribution to be low, $p<2$, or that the outflow is quasi-spherical thus reviving the energy problem. We argue that the slow decay of GRB~010222 and a linear polarization of $1.36\\pm 0.64$\\%, is naturally explained by a jet model with continuous energy injection. The electron energy distribution then has $p=2.49\\pm0.05$, fully consistent with the expectation from detailed modelling of acceleration in relativistic shocks, that $p>2$, thus alleviating the ``$p$-problem''. ",
        "introduction": "\\label{sec:intro}  GRB~010222 was a bright gamma-ray burst (GRB) localized by BeppoSAX\\\\ \\citep{piro01}. X-ray observations were reported by in 't Zand et al.\\ (2001).  The optical afterglow was discovered by Henden \\& Vrba \\cite{henden01}, 4.3 hours after the burst and a redshift of $z=1.477$ was determined by Jha et al.\\ \\cite{jha01}. Further optical/near-infrared observations have been reported by Stanek et al.\\ \\cite{stanek01}, Lee et al.\\ \\cite{lee01}, Masetti et al.\\ \\cite{masetti01}, Cowsik et al.\\ (2001), Sagar et al.\\ \\cite{sagar01} and Mirabal et al.\\ (2002). In our analysis we shall adopt the light curve fit of Henden et al.\\ (2002), that is based on data extending up to 80 days after the burst. Their best fit in the $R$-band gives a pre-break power-law slope of $\\alpha_1=-0.66\\pm 0.03$ and a post-break slope of $\\alpha_2=-1.40\\pm 0.02$, with a break time of $t_b=0.58\\pm 0.04$ days. The observed optical spectral index is about $\\beta=-1.0$ (e.g.\\ Stanek et al.\\ 2001), but as it may be strongly affected by host extinction, it can lead to ambiguous inference of the afterglow properties. In particular, the relationship between the light curve decay indices, the $\\alpha$'s, and the spectral index, $\\beta$, that is predicted by synchrotron models of afterglows (e.g.\\ Sari, Piran \\& Narayan 1998; Sari, Piran \\& Halpern 1999; Price et al.\\ 2002), can lead to inconsistent interpretations due to the unknown host extinction.  To date, interpretation of the GRB~010222 afterglow observations has mainly relied on two possible scenarios: i) A narrow sideways expanding jet propagating in a low-density medium (e.g.\\ Stanek et al.\\ 2001). ii) A wide jet or spherical fireball transiting from relativistic to non-relativistic regime (e.g.\\ Masetti et al.\\ 2001; in 't Zand et al.\\ 2001). The latter models require a very dense medium and implies a very large energy release, thus bringing back the energy problem \\citep{kulkarni99, andersen99}. Both classes of models require a hard electron energy distribution, i.e.\\ $p<2$, in the first case even as low as $p\\approx 1.4$. This is contrary to most other bursts that seem to be adequately fit with models where $p\\approx 2.3-2.5$ (van Paradijs, Kouvelioutu \\& Wijers 2000). Recently, Panaitescu \\& Kumar \\cite{pankum02} presented fits to several afterglows, concluding that $p$ is smaller than 2 in a number of them. Such small inferred values of $p$ signal a departure from the standard fireball model \\citep{mesz02}, and introduce additional free parameters into the model, such as an upper cutoff in the electron energy distribution, or additional assumptions about the electron acceleration mechanism to facilitate generation of flat energy distributions (Dai \\& Cheng 2001; Bhattacharya 2001). The $p$-problem then arises from the fact that detailed modelling of particle acceleration in relativistic shocks indicates that $p\\approx 2.2-2.3$ \\citep{achterberg01}.  In addition, if $p$ is assumed to be constant throughout the fireball evolution, the magnitude of the observed light curve break in the case of GRB~010222, $\\Delta\\alpha=\\alpha_1-\\alpha_2=0.74\\pm 0.04$, cannot be explained by the above models. The sideways expanding jet model predicts a break magnitude of $\\Delta\\alpha=1-\\alpha_1/3=1.22\\pm 0.01$, while the fireball transiting to the non-relativistic regime gives $\\Delta\\alpha=(\\alpha_1+3/5)=0.06\\pm 0.03$ for slow cooling electrons. For fast cooling electrons the latter model predicts $\\Delta\\alpha=(\\alpha_1+1)=0.34\\pm 0.03$. The break magnitudes predicted by the models are in all cases very different from the observed value that is however, in perfect agreement with the prediction of a jet model with a fixed opening angle \\citep{mesz99}, $\\Delta\\alpha=3/4$, being essentially of geometrical origin.   In this Letter we use polarization measurements to argue against interpreatations based on spherical models transiting to the non-relativistic regime. Furthermore, we demonstrate that the observations of GRB~010222 can be naturally interpreted with a jet model with a small opening angle and continuous energy injection. ",
        "conclusions": ""
    },
    "0209/astro-ph0209316_arXiv.txt": {
        "abstract": "The High Energy Transmission Grating Spectrometer (HETGS) onboard the Chandra X-ray Observatory has so far produced a large number of high resolution X-ray spectra with unprecedented spectroscopic details. Spectra from outflows in galactic and extragalactic X-ray sources indicate plasmas with a wide range of properties. Optically thick fluorescent matter and warm photoionized regions play as much a role as very hot regions where collisional ionization and scattering dominate the emission. Through the measurements of blue- and redshifts the complex dynamics of these plasmas is revealed. Quite intriguing in this respect is the case of X-ray absorption of neutral matter. In many cases spectral features are found to be of high complexity though the detection of edges from intermediate Z elements as well as absorption lines from monatomic species to molecular compounds. With the application of line diagnostic tools and more accurate atomic data bases we are now able to model the properties of these plasmas as well as measure line shifts and shapes to constrain their spatial distribution and dynamics. In various examples, i.e. plasmas from accretion disks, winds, stationary clouds as well as the ISM, the power of highly resolved X-ray spectra is demonstrated and the scientific capability of XEUS in this context is explored. ",
        "introduction": "With the launch of the Chandra X-ray Observatory in 1990 we entered a new era of high resolution X-ray spectroscopy that enriched our views throughout the entire field of high energy astrophysics. Specifically the two grating spectrometers, the HETGS and the Low Energy Transmission Grating Spectrometer (LETGS), with a spectral resolving power of over an order of magnitude larger than most previous space-born X-ray spectrometers have already quite impressively demonstrated the importance and need for high resolution instruments in the field. While focussing mostly on bright galactic X-ray sources, I will explore some aspects of X-ray spectroscopy and plasma diagnostics using the HETGS which are now state of the art in the field.  X-ray binaries are the brightest X-ray sources in the sky and their bright X-ray emission is a consequence of the accretion process of material from a close companion onto a compact object, likely a neutron star of a black hole (BH). The result is a strong X-ray continuum that can be detected and measured with fairly low resolution detectors like proportional or gas scintillation counters. The latter offer spectral resolving powers of about $\\Delta$E/E of 1 to 10 depending on energy. These continua have been modeled to be of either thermal nature like blackbody radiation from the surface of a neutron star, disk blackbody radiation from a hot inner disk surface, thermal bremsstrahlung with high energy cut-offs, power laws from synchrotron radiation and/or reprocessed radiation through inverse Compton scattering, just to name a few. These spectra are generally modified by continuum (photoelectric) absorption in the ISM and optically thick matter intrinsic to the sources. Discrete emission like Fe K lines was merely detected as local perturbations in the continuum modeling process. Lewin et al. (1995) offers a more complete review with references. With the advent of charge coupled devices (CCDs) the resolving power increased to the order to 10 to 60. Some good progress was made to detect discrete line emission and absorption predominantly from faint extragalactic source like warm absorbers and relativistic iron lines in active galactic nuclei. Spectra from very bright X-ray binaries offered quite little in this respect. Here only a few are to name like 4U 1626-67 (Angelini et al. 1995), Cyg X-3 (Kitamoto et al. 1994), and various high mass X-ray binaries like Vela X-1, Cen X-3 and GX 301-2 (Nagase et al. 1994). The problem with X-ray binaries was thought to be photon pile up in the CCD frame, which at such brightness levels wiped out potential line emission.  Why do we expect discrete emission in X-ray binaries in the first place? The answer is that there are many different plasma environments in these systems that re-process the radiation of the central source. In the following I chose a few examples of such environments and focus on the fact that we need various levels of spectral resolving powers to not only detect descrete features but also resolve them. The range of parameters in these plasmas is quite large as we observe temperatures between 10$^4$ and 10$^8$ K, densities from as low as $\\sim 10^{8}$ cm$^{-3}$ in winds to as high as $\\sim 10^{18}$ cm$^{-3}$ in accretion disks, optical depths between 0.01 and 100, ionization parameters of up to 2$\\times 10^4$ and a vast range of ionization stages. For in depths reviews I recommend Liedahl (1997) and Paerels (1997). Many times conditions are complicated by the fact that most of these plasmas are neither in ionization equillibrium nor at rest and we thus have to deal with all kinds of plasma dynamics.  The observation of X-rays in stars was discovered with \\ein, but the fact that many stars exhibit significant X-ray emission was finally established with the advent of \\ros in 1990. Observation and analysis of stellar spectra is difficult and highly complex. Although most of the stars show X-ray emission at some level, but it is primarily the emission from very young late type and early type stars or stars that are very close to the sun that dominate the observations. The reason is that the two former show X-ray luminosities that are many orders of magnitudes higher than most late type main sequence stars. On the other hand, the X-ray spectra of stars are more or less pure line spectra that need high spectral resolving power. ",
        "conclusions": ""
    },
    "0209/astro-ph0209120_arXiv.txt": {
        "abstract": "We present evidence from \\sax\\ and \\xmm\\ of extreme X-ray variability in the high luminosity radio-quiet quasar PDS 456, the most luminous known AGN at $z<0.3$. Repeated X-ray flaring is found in PDS 456, over the duration of the 340 ksec long \\sax\\ observation. The X-ray flux doubles in just 30 ksec, whilst the total energy output of the flaring events is as high as $10^{51}$~erg. Under the assumption of isotropic emission at the Eddington limit, this implies that the size of the X-ray emitting region in PDS 456 is less than 3 Schwarzschild radii, for a $10^9$M$_{\\odot}$ black hole. From the rates of change of luminosity observed during the X-ray flares, we calculate lower limits for the radiative efficiency limit between 0.06 and 0.41, implying that accretion onto a Kerr black hole is likely in PDS 456. We suggest that the rapid variability is from X-ray flares produced through magnetic reconnection above the disc and calculate that the energetics and timescale of the flares are plausible if the quasar is accreting near to the maximum Eddington rate. A similar mechanism may account for the extreme rapid X-ray variability observed in many Narrow Line Seyfert 1s. In the case of PDS 456, we show that the X-ray flaring could be reproduced through a self-induced cascade of $\\sim1000$ individual flares over a timescale of the order 1 day. ",
        "introduction": "PDS 456 is a luminous, butlow redshift ($z=0.184$) radio-quiet quasar identified in 1997 (Torres \\et 1997). The optical and infra-red spectra (Simpson et al. 1999) show broad Balmer and Paschen lines (e.g. H$\\beta$ FWHM 3000 km~s$^{-1}$), strong Fe \\textsc{ii}, a hard (de-reddened) optical continuum ($f_{\\nu} \\propto \\nu^{-0.1\\pm0.1}$), and one of the strongest `big blue bumps' of any AGN (Simpson \\et 1999, Reeves et al. 2000). It is also radio-quiet (F$_{5GHz}=8$mJy; Reeves et al. 2000), and is presumably not jet dominated or strongly beamed. PDS~456 has a de-reddened, absolute blue magnitude of M$_{B}\\approx -27$ (Simpson et al. 1999), making it as luminous as the radio-loud quasar 3C~273 ($z=0.158$, M$_{B}\\approx -26$). Indeed PDS 456 is the most luminous known AGN in the local Universe (z\\ $<0.3$), its luminosity being more typical of quasars at z=2-3, at the peak of the quasar luminosity function.  PDS 456 was first detected as the X-ray source RXS~J172819.3-141600 in the ROSAT All Sky Survey (Voges \\et 1999). Subsequent \\asca\\ and \\xte\\ observations of PDS 456 showed that it was highly X-ray variable (see Reeves et al. 2000). In particular, during the \\xte\\ observation, an X-ray flare occurred with a doubling time of just 15 ksec, implying that the X-ray emitting region was extremely compact, less than 2 Schwarzschild radii (or $2R_S$) in size. Such rapid variability is very unusual for luminous quasars, as the variability timescale is thought to increase with luminosity, and black hole mass (e.g. Turner \\et 1999). One possibility is that the accretion rate is unusually high in PDS 456, perhaps close to Eddington. An analogy might then be drawn with the extreme events observed in several Narrow Line Seyfert 1 galaxies (e.g. Boller, Brandt \\& Fink 1996, Leighly \\et 1999), thought to have smaller black hole masses ($10^6$M$_{\\odot}$ - $10^7$M$_{\\odot}$), accreting near to the Eddington rate. We report here on X-ray observations of PDS 456, conducted with \\sax\\ and \\xmm\\ in February and March 2001. The prime motivation was to study the extra-ordinary variability of PDS 456 with an imaging X-ray telescope, thus negating the possibility of source contamination which may occur within the field of view of a non-imaging instrument.  \\begin{figure} \\begin{center} \\rotatebox{-90}{\\includegraphics[width=6.5cm]{figure1.ps}} \\end{center} \\caption{Background subtracted \\sax\\ lightcurves for the (a) LECS (0.3-2 keV) and (b) MECS (1-10 keV), binned into orbital bins (96 minutes). Repeated X-ray flaring is seen throughout the observation. The largest changes are by a factor of x1.9 in the MECS in 35 ksec and a factor of x4.1 in the LECS in 40 ksec, after 230 ksec. The variations imply that for PDS 456, with a $10^9$M$_{\\odot}$ black hole, the X-ray emitting region is no larger than 3 Schwarzschild radii ($3R_S$).} \\end{figure} ",
        "conclusions": "Recent \\sax\\ and \\xmm\\ observations have shown that the luminous quasar PDS 456 exhibits rapid X-ray variability on timescales of $\\sim30$~ksec, with a total energy output of $10^{51}$~erg~s$^{-1}$ for the flaring events. This limits the size of the X-ray emitting region to $<3$ Schwarzschild radii for a $10^{9}$M$_{\\odot}$ black hole. The energetics of the of the accretion disc in PDS~456 can power its extreme X-ray variability if the black hole is massive ($\\ga 10^9$ $M_\\odot$) and is accreting close to the Eddington rate. Coronal magnetic flare events can explain the X-ray variability as long as the disc is able convert accretion energy into coronal magnetic energy efficiently, and that this energy can be released in the form of a self-induced cascade of $\\ga$ 1000 individual flare events on a timescale of the order 1 day.   \\vspace{-0.5cm}"
    },
    "0209/astro-ph0209134_arXiv.txt": {
        "abstract": "{ Primary inversions of accurately measured solar oscillation frequencies coupled with the equations of thermal equilibrium and other input physics, enable us to infer the temperature and hydrogen abundance profiles inside the Sun. These profiles also help in setting constraints on the input physics as well as on heavy element abundance in the solar core. Using different treatments of plasma screening for nuclear reaction rates, limits on the cross-section of proton-proton nuclear reaction as a function of heavy element abundance in the solar core are obtained and an upper limit on heavy element abundance in the solar core is also derived from these results.} ",
        "introduction": "The precisely measured frequencies of solar oscillations have been used to probe the solar interior. The primary inversions of these observed frequencies yield the sound speed and density profiles inside the Sun. In order to infer the temperature and chemical composition profiles, we also need to know the input physics such as opacities, equation of state and nuclear energy generation rates (Gough \\& Kosovichev \\cite{dog88}; Kosovichev \\cite{kos96}; Shibahashi \\& Takata~\\cite{st96}; Takata \\& Shibahashi~\\cite{tak98}; Antia \\& Chitre \\cite{ac98}). In all these works the heavy element abundance profile is assumed to be known; attempts to determine heavy element abundance profile from helioseismic data have not been particularly successful (Antia \\& Chitre \\cite{ac99}; Takata \\& Shibahashi \\cite{tak01}) as the resulting inverse problem becomes extremely ill-conditioned. Fukugita \\& Hata~(\\cite{fuk98}) have obtained limits on heavy element abundance in the solar core using observed solar neutrino fluxes. It would be interesting to enquire if such limits can be independently obtained from helioseismic data.  In general, the computed luminosity in a seismically computed solar model is not expected to match the observed solar luminosity. By applying the observed luminosity constraint it is possible to constrain the input physics, particularly, the cross-section of proton-proton (pp) nuclear reaction. Antia \\& Chitre (\\cite{ac98}) estimated this cross-section to be $S_{11}= (4.15\\pm0.25)\\times10^{-25}$ MeV barns. Similar values have been obtained by comparing the computed solar models with helioseismic data (Degl'Innocenti, Fiorentini \\& Ricci~\\cite{inn98}; Schlattl, Bonanno \\& Paterno~\\cite{bon99}). The main source of error in these estimates is the uncertainty in the $Z$ profile and, therefore, Antia \\& Chitre (\\cite{ac99}) attempted to find the pp reaction rate as a function of $Z$ in the solar core. In all these works the plasma screening of nuclear reaction cross-sections was calculated using intermediate screening formulation of Graboske et al.~(\\cite{gra73}). The treatment of screening in stellar nuclear reaction rates is not yet adequately understood (Dzitko et al.~\\cite{dzi95}; Gruzinov \\& Bahcall \\cite{gru98}). Wilets et al.~(\\cite{wil00}) have done a sophisticated treatment of plasma screening and compared their results with earlier prescriptions. Their results indicate that for the solar core the intermediate screening treatment due to Mitler (\\cite{mit77}) is better than that due to Graboske et al.~(\\cite{gra73}). It would thus be interesting to study the effect of different treatment of plasma screening on the helioseismically estimated pp reaction cross-section.  Antia \\& Chitre (\\cite{ac99}) included the effect of heavy element abundance $Z$ only on the opacity of the solar material. If we make the reasonable assumption that the abundances of C, N, O also increase with $Z$, then the CNO cycle will become more effective in contributing to the nuclear energy generation in the solar core. At normally accepted values of $Z$ it is estimated that less than 2\\% of energy generated in the central region is produced by the CNO cycle (Bahcall et al.~\\cite{bp01}). But if $Z$ value is increased, this proportion will clearly increase and consequently, the pp reaction rate needs to be reduced to maintain the observed solar luminosity. In this work we demonstrate that this effect can be exploited to set an upper limit on $Z$ in the solar core. ",
        "conclusions": "With the help of inverted sound speed and density profiles, it is possible to infer the $T,X$ profiles in the solar interior, provided the $Z$ profile and the input physics are known. The resulting seismic models have the correct solar luminosity, provided the heavy element abundance $Z_c$ in the solar core and the cross-section for pp nuclear reaction are within the shaded region shown in Fig.~1. It appears that the currently accepted values of $Z_c$ or $S_{11}$ need to be increased marginally to make them consistent with helioseismic constraints. The required increase is within the error estimates. The higher estimates for $S_{11}$ obtained earlier were due to differences in treatment of plasma screening. With the use of weak (Salpeter \\cite{sal54}) or intermediate screening due to Mitler (\\cite{mit77}) the theoretically estimated value of $S_{11}$ is in reasonable agreement with seismically estimated value. With a $Z$ profile in a standard solar model N0 of Brun et al.~(\\cite{bru02}), the seismically estimated value of $S_{11}$ is $4.07\\times10^{-25}$ MeV barns with the use of intermediate screening and $4.02\\times10^{-25}$ MeV barns for weak screening used while calculating the nuclear energy generation rate.  If the value of  heavy element abundance in the solar core, $Z_c$, is increased beyond 0.035 the CNO cycle generates a good fraction of solar luminosity and it is not possible to get any consistent seismic model unless $S_{11}$ is decreased substantially below the accepted value. This puts a clear upper limit on the heavy element abundance in the solar core, which is comparable to that independently obtained by Fukugita and Hata (\\cite{fuk98}). This upper limit is not very sensitive to the $Z$ profile or to the uncertainties in the CNO reaction rates. Even if the additional heavy elements in solar core do not include CNO, the effect of opacity alone will also put an upper limit on $Z_c$, but in that case the limit will depend on the assumed lower limit on $S_{11}$. For currently accepted theoretical limits, the upper limit in this case turns out to be around 0.035.  Note that the seismic model satisfies the normal stellar structure equations, though the inferred $X$ profile may not match that given by an evolutionary solar model. Further, since the seismic model is confined to the radiative interior, it will not be possible to match it to an acceptable convection zone model, unless the heavy element abundance at the top of the radiative region is close to the known surface value. Thus a relatively high value of $Z$ in the solar core can be realised only if the $Z$ profile has a significant gradient in the radiative region. From experiments with varying $Z$ or $S_{11}$ in evolutionary solar models also it is known that an increase in $Z$ can be compensated for by a reduction in $S_{11}$ to match the seismically inferred sound speed and density profiles (Brun et al.~\\cite{bru02}). Thus it is quite possible that a comparable upper limit on $Z_c$ may be obtained from these models if we restrict the range of $S_{11}$, provided the initial $Z$ profile at  the zero age main sequence stage is not homogeneous."
    },
    "0209/astro-ph0209302_arXiv.txt": {
        "abstract": "We report the detection of a new candidate exoplanet around the metal-rich star $\\tau^1$~Gruis. With M~sin~$i$~=~1.23$\\pm$0.18 M$_{\\rm JUP}$, a period of 1326$\\pm$300~d and an orbit with an eccentricity of 0.14$\\pm$0.14 it adds to the growing population of long period exoplanets with near-circular orbits. This population now comprises more than 20\\% of known exoplanets.  When the companion to $\\tau^1$~Gruis is plotted together with all exoplanets found by the Anglo-Australian Planet Search and other radial velocity searches we find evidence for a peak in the number of short-period exoplanets, followed by a minimum of planets between around 7 and 50 days and then an apparent rise in the number of planets per unit radius that seems to set in by a hundred days, indicating more planets farther from the host star. This is very different from the gaussian-like period distribution found for stellar companions. This lends support to the idea that once a clearing in the inner protoplanetary disk develops, it halts the inward migration of planets. In particular, the smooth distribution of exoplanets arising from planetary migration through a disk is altered by an accumulation of exoplanets at the point where the disk has been cleared out. ",
        "introduction": "The Anglo-Australian Planet Search (AAPS) is a long-term planet detection programme which aims to perform exoplanet detection and measurement at the highest possible precision. Together with programmes using similar techniques on the Lick 3\\,m and Keck I 10\\,m telescopes (Fischer et al. 2001; Vogt et al. 2000), it provides all-sky planet search coverage for inactive F, G, K and M dwarfs down to a magnitude limit of V=7.5. So far the AAPS has has published data for 17 exoplanets. (Tinney et al. 2001; Butler et al. 2001; Butler et al. 2002a; Jones et al. 2002a,b; Tinney et al. 2002a,b).  The AAPS is carried out on the 3.9m Anglo-Australian Telescope (AAT) using the University College London Echelle Spectrograph (UCLES), operated in its 31 lines/mm mode together with an I$_{2}$ absorption cell. UCLES now uses the AAO's EEV 2048$\\times$4096 13.5$\\mu$m pixel CCD, which provides excellent quantum efficiency across the 500--620~nm I$_2$ absorption line region. Despite this search taking place on a common-user telescope with frequent changes of instrument, we achieve a 3~m~s$^{-1}$ precision down to the V~=~7.5 magnitude limit of the survey (Butler et al. 2001; fig. 1, Jones et al. 2002a).  Our target sample, which we have observed since 1998, is given in Jones et al. (2002b). It includes 178 late (IV-V) F, G and K stars with declinations below $\\sim -20^\\circ$ and is complete to V$<$7.5. We also observe sub-samples of 16 metal-rich ([Fe/H]$>$0.3) stars with V$<$9.5 and 7 M dwarfs with V$<$7.5 and declinations below $\\sim -20^\\circ$. The sample is being increased to around 300 solar-type stars to be complete to a magnitude limit of V=8. Where age/activity information is available from log~$R$'(HK) indices (Henry et al. 1996; Tinney et al. 2002c) we require target stars to have log~$R$'(HK) $<$ --4.5 corresponding to ages greater than 3 Gyr. Stars with known stellar companions within 2 arcsec are removed from the observing list, as it is operationally difficult to get an uncontaminated spectrum of a star with a nearby companion. Spectroscopic binaries discovered during the programme have also been removed and will be reported elsewhere (Blundell et al., in preparation). Otherwise there is no bias against observing multiple stars. The programme is also not expected to have any bias against brown dwarf companions. The observing and data processing procedures follow those described by Butler et al. (1996, 2001). ",
        "conclusions": "The companion to $\\tau^1$~Gruis announced here serves to further reinforce the predominantly metal-rich nature of stars with exoplanets. It also adds to the growing population of long period exoplanets with near-circular orbits. Now more than 20\\% of exoplanets have orbital parameters within those of the Solar System. It is notable that as the Anglo-Australian Planet Search becomes sensitive to longer periods, we are continuing to find objects with longer periods, but remain limited by our first epoch observations. $\\tau^1$~Gruis is a pleasing example. Within the errors its velocity amplitude is nearly as low as any long-period single exoplanet announced by radial velocity searches and the error on its period is dominated by our first epoch observation. Thus the detection of an exoplanet around $\\tau^1$~Gruis together with our long-term stable stars (e.g., Butler et al. 2001) gives us confidence in the stability of our search as we move to longer periods and the possibility of detecting Jupiter analogues.  The radial velocity signal we measure for $\\tau^1$~Gruis suggests a planet with a minimum mass around that of Jupiter. $\\tau^1$~Gruis~b becomes the fifth exoplanet to be found with a mass around that of Jupiter with a period of greater than three years and indicates that radial velocity surveys now have significant sensitivity to Jupiter mass planets out to relatively large periods.  It is thus intriguing to look at the period distribution of exoplanets found by the AAPS and other radial velocity searches. In Butler et al. (2002b), we plotted a histogram of semimajor axes for exoplanets from the Lick, Keck and AAT searches. This showed a relatively large number of exoplanets at very short orbits and a tail of objects with longer orbits. The detection of long period planets and long-term stable stars indicated that the peak at short periods was a real feature. Two years later, with twice as many exoplanets known, Figure 3 shows the exoplanets that have been announced based on exoplanets.org by 2002 August 21. The bulk of known exoplanets now lie at relatively large periods. Although the AAPS has been operating for less time than other successful searches, Figure 3 also shows that the exoplanets published by the AAPS are dominated by companions at longer periods such as $\\tau^1$~Gruis~b. The top part of Figure 3 shows a peak at shorter periods together with a substantial fraction at longer periods. Interestingly there appears to be a gap in the distribution between periods of around 7 and 50 days. The evidence for this gap is relatively poor when considering the AAT planets alone though striking when all exoplanets are considered.  The relative lack of exoplanet candidates from around 0.2 to 0.6 AU was noted by Cummings, Marcy \\& Butler (1999) and Butler et al. (2002b) and is also evident in fig. 2 of Heacox et al.(1999), fig. 5 of Rabachnik \\& Tremaine 2001, fig. 4 of Lineweaver \\& Grether (2002) and fig. 7 of Armitage et al. (2002). Armitage et al. interpret this feature as a slight excess of exoplanets at the shortest periods and attribute it to the completeness of radial velocity surveys falling off toward longer periods. However, the observed period distribution actually appears to rise toward longer periods where the incompleteness of the radial velocity surveys falls rapidly. To allow for this incompleteness introduced by including lots of low-mass short-period exoplanets we follow Armitage et al. (2002) and consider planets in a restricted mass and period range where the surveys can be judged to be more complete. Following the analysis of Cummings et al. (1999), Armitage et al. consider the known exoplanets to be complete in the mass range 0.6--10 M$_{\\rm JUP}$ sin $i$ for periods of less than 3 AU. In the middle plot of Figure 3 we show the period distribution for a 0.6--10 M$_{\\rm JUP}$ mass cut off as well as all announced planets. The removal of the lowest mass exoplanets reduces the peak of very short period planets, however, the peak at short periods remains an order of magnitude higher than would be expected from an extension of counts at longer periods. We have also looked at the CORALIE, Keck and Lick surveys in the same manner as for the AAPS. Despite different radial velocity surveys operating with different samples, sensitivities, instruments, scheduling, strategies and techniques we do find evidence for the gap in each of the major surveys. As mentioned above, this gap is evident in a number of works by other authors though is relatively less pronounced because of the substantially smaller number of exoplanets announced when those plots were made. The pronounced nature of this peak leads us to consider Fig. 4 to show evidence for two (or more) populations of exoplanets. That is, a population of exoplanets spanning a small range of short periods (3-7 days) and an separate population increasing with number towards larger periods.  From the lower part of Figure 3, it can be seen that there is no evidence for such a gap in the stellar binary distribution. The stellar companion period distribution plotted was determined by Duquennoy \\& Mayor (1991) using the same general radial velocity method to discover binary stars as used to discover the exoplanets. Duquennoy \\& Mayor find it necessary to make corrections to the stellar binary period distribution for incompleteness at longer periods, however, no such corrections are necessary for shorter periods and they find no gap in short-period stellar binaries. Overall Duquennoy \\& Mayor find that the period distribution of stellar binaries is well fit by a gaussian. Since stellar companions are expected to form via large-scale gravitational instabilities in collapsing cloud fragments or massive disks, whereas planets are expected to form by accretion in dissipative circumstellar disks it is not surprising that stellar and planetary companions should have different period distributions.    The period distribution for exoplanets has been investigated a number of times as the number of radial velocity exoplanets has grown. The existence of a peak at very short periods has been an important motivation in the development of migration theories for exoplanets (Lin et al. 1999; Murray et al. 1998; Trilling et al. 1998; Ward 1997; Armitage et al. 2002; Trilling, Lunine \\& Benz 2002). The trend toward finding an increasing number of exoplanets with large orbital separation runs counter to the selection effects inherent in radial velocity searches and has been well reproduced by migration theories (Armitage et al. 2002; Trilling et al. 2002). Whilst selection effects start to play an increasingly important role beyond around few hundred days (e.g. Duquennoy \\& Mayor 1991; Cummings et al. 1999; Butler et al. 2002b), we do not consider them to be significant between 7 and 50 days and thus consider the 'gap' in exoplanet periods to be a feature of the period distribution not currently predicted by migration theories. It is interesting to speculate on the origin of the possible gap in the exoplanet period distribution.  The onset of the stellar wind in young stars and the magnetic clearing of a hole at the centre of the disk will lead to the evacuation of the circumstellar disk and prevent migration of planets. This is expected to happen sooner in stars of higher mass and suggests the exoplanets of stars with higher mass will lie at greater radii. So far the range of stellar masses yielding significant numbers of exoplanets is rather small and we find no clear difference in exoplanet properties for stars of different mass. Even without such evidence, migration theory does provide an attractive explanation for a range of exoplanet properties. Migration theory can already reasonably explain the progressively larger number of exoplanets at larger radii and with the inclusion of appropriate stopping mechanisms (e.g.  Lin, Bodenheimer \\& Richardson 1996; Kuchner \\& Lecar 2002) may also be able to consistently produce the peak in the period distribution of short period planets."
    },
    "0209/astro-ph0209628_arXiv.txt": {
        "abstract": "{ \\begin{center}\\begin{minipage}{0.9\\textwidth} We present Near-IR photometry of the Arches cluster, a young and massive stellar cluster near the Galactic center. We have analyzed the high resolution (FWHM $\\sim$ 0.2$''$) $H$ and $K'$ band images in the \\emph{Galactic Center Demonstration Science Data Set}, which were obtained with the Gemini/Hokupa's adaptive optics (AO) system. We present the color-magnitude diagram, the luminosity function and the initial mass function (IMF) of the stars in the Arches cluster in comparison with the $HST/NICMOS$ data. The IMF slope for the range of $1.0< \\log~ (M/M_\\odot) <2.1$ is estimated to be $\\Gamma = -0.79\\pm0.16 $, in good agreements with the earlier result based on the $HST/NICMOS$ data [Figer et al. 1999, ApJ, 525, 750]. These results strengthen the evidence that the IMF of the bright stars close to the Galactic center is much flatter than that for the solar neighborhood. This is also consistent with a recent finding that the IMFs of the bright stars in young clusters in M33 get flatter as the galactocentric distance decreases [Lee et al. 2001, astro-ph 0109258]. It is found that the power of the Gemini/AO system is comparable, with some limits, to that of the $HST/NICMOS$. % \\end{minipage}\\end{center} } ",
        "introduction": "The Arches cluster is a very unique cluster in the Milk Way, because it is a very massive and compact young cluster close to the Galactic center. It was confirmed as a star cluster  including emission-line stars by Nagata et al (1995). To date only three clusters are known to be very close to the Galactic center. The other two clusters are the Quintuplet cluster and the IRS 16 cluster at the Galactic center. The size of the Arches cluster is about $15''$ (= 0.58 pc at the distance of 8 kpc), and the total mass is estimated to be about $10^4 M_\\odot$ (Figer et al 1999). % The Arches cluster has a very high peak density $3 \\times 10^5$ $M_\\odot$ pc$^{-3}$ in the inner $9''$ (0.35 pc), showing that it is one of the densest known young clusters in the Local Group galaxies. Similar examples are R 136, the central cluster of 30 Dor in the Large Magellanic Cloud and NGC 3603 in our Galaxy. The age of the cluster is estimated to be about 2--5 Myrs (Figer et al 1999, Blum et al 2001). Very recently  Yusef-Zadeh et al (2002) detected, using the Advanced CCD Imaging Spectrometer on board Chandra X-Ray Observatory,  two X-ray sources in this cluster, and suggested that the X-ray emission from the sources arises from  stellar wind sources in the cluster.  The presence of compact young clusters like the Arches cluster and the other two clusters near the Galactic center indicates that stars are forming even in such a dense environment. Therefore a study of these clusters will provide important hints for understanding the star formation process under extreme environments.  Stars in the Arches cluster were studied in detail for the first time by Figer et al (1999) and Kim et al (2000) who used the  Hubble Space Telescope ($HST$) Near-Infrared Camera and Multiobject Spectrometer ($NICMOS$) observations. Figer et al (1999) found several interesting results on this cluster: (1) the Arches cluster is very young, with an age of only about 2 Myrs, showing that stars are forming very recently in the region close to the Galactic center; and (2) the initial mass function (IMF) of the massive stars in this cluster is derived to be significantly flat, having a slope of $\\Gamma = \\log N / \\log M = -0.7\\pm0.1$ for the mass range of $0.8 < \\log ~(M/M_\\odot) <2.1 $. Surprisingly this IMF slope is much flatter than the average for other clusters in the solar neighborhood which is close to the Salpeter value, $\\Gamma= -1.35$ (see Scalo 1998). This result shows, if confirmed, that stars with flatter IMFs are formed in the dense region like the Galactic center, while stars with steeper IMFs are formed in the low-density region like the solar neighborhood.  In spite of the importance of the study of these clusters, there are only a few studies of these clusters to date. It has been a demanding job to investigate the IMF for the clusters like the Arches from the ground-based observation, because the cluster fields are very crowded and the interstellar extinction toward the clusters is severe. Therefore the $HST$ remains to be almost the only instrument useful for these studies until recently. However, with the advent of Adaptive Optics (AO) system at the Gemini Telescope, it became possible to study the stars in the clusters like the Arches in detail with ground-based observations.  \\begin{table}[!htb] \\begin{center} {\\bf Table 1.} Observation log \\\\ \\begin{tabular}{ c c c c c } \\hline \\hline ID & Date   & Filter  & Total exp.   & FWHM \\\\ & (2000) &         &  time(sec)   & $('')$ \\\\ \\hline 1 & 07-05 & $H$  & 3    & 0.140-0.165 \\\\ 2 & 07-05 & $H$  & 720  & 0.180-0.230 \\\\ 3 & 07-09 & $K'$ & 2    & 0.120-0.140 \\\\ 4 & 07-30 & $K'$ & 16   & 0.105-0.135 \\\\ 5 & 07-04 & $K'$ & 180  & 0.185-0.250 \\\\ 6 & 07-03 & $K'$ & 240  & 0.145-0.180 \\\\ 7 & 07-30 & $K'$ & 480  & 0.125-0.145 \\\\ 8 & 07-09 & $K'$ & 1020 & 0.125-0.135 \\\\ \\hline \\end{tabular} \\end{center} \\label{obslog} \\end{table}   In this paper we present Near-IR photometry of the Arches cluster obtained for science demonstration using the Gemini/AO system in comparison with the $HST/NICMOS$ results. Preliminary results of this study were presented by Yang et al (2002). During the preparation of this paper, Stolte et al (2002) also presented at a conference a similar study to ours using the same data set. ",
        "conclusions": "We present Near-IR photometry of the Arches cluster near the Galactic center, using the data obtained for scientific demonstration with the Gemini/AO. Primary results are summarized as follows: First, the color-magnitude diagram of the Arches cluster shows a dominant blue main sequence consisting mainly of massive stars. Second, the age of the Arches cluster is estimated to be $2\\pm1$ Myrs, using the Geneva isochrones. This value is consistent with that based on the $HST/NICMOS$ data (Figer et al 1999). Third, the $K$ and $H$ luminosity functions of the bright stars in the Arches clusters are derived, showing a slow increase toward the faint end. Fourth, the initial mass function of the massive stars with $1.0< \\log~ (M/M_\\odot) <2.1$ is derived. Fitting the power law to the data, we obtain a value for the IMF slope, $\\Gamma = -0.79\\pm 0.16$. This confirms that Figer et al (1999)'s result that the IMF of the Arches cluster is much flatter than that of the solar neighborhood. Although the $HST$ provides unprecedented spatial resolution for the dense region, our results show that the ground-based AO systems in the Gemini telescope can yield comparable scientific results with some limits."
    },
    "0209/astro-ph0209591_arXiv.txt": {
        "abstract": "We present the application of the Fast Independent Component Analysis ({\\ica}) technique for blind component separation to polarized astrophysical emission. We study how the Cosmic Microwave Background (CMB) polarized signal, consisting of $E$ and $B$ modes, can be extracted from maps affected by substantial contamination from diffuse Galactic foreground emission and instrumental noise. {We implement Monte Carlo chains varying the CMB and noise realizations in order to asses the average capabilities of the algorithm and their variance.} We perform the analysis of all sky maps simulated according to the {\\sc Planck} satellite capabilities, modelling the sky signal as a superposition of the CMB and of the existing simulated polarization templates of Galactic synchrotron. Our results indicate that the angular power spectrum of CMB $E$-mode can be recovered on all scales up to $\\ell\\simeq 1000$, corresponding to the fourth acoustic oscillation, while the $B$-mode power spectrum can be detected, up to its turnover at $\\ell\\simeq 100$, if the ratio of tensor to scalar contributions to the temperature quadrupole exceeds $30\\%$. The power spectrum of the cross correlation between total intensity and polarization, $TE$, can be recovered up to $\\ell\\simeq 1200$, corresponding to the seventh $TE$ acoustic oscillation. ",
        "introduction": "\\label{introduction}  We are right now in the epoch in which the cosmological observations are revealing the finest structures in the Cosmic Microwave Background (CMB) anisotropies\\footnote{see lambda.gsfc.nasa.gov/ for the list and details of the operating and planned CMB experiments}. After the first discovery of CMB total intensity fluctuations as measured by the COsmic Background Explorer (COBE) satellite (see Smoot 1999 and references therein), several balloon-borne and ground-based operating experiments were successful in detecting CMB anisotropies on degree and sub-degree angular scales (De Bernardis et al 2002, Halverson et al. 2002, Lee et al. 2001, Padin et al. 2001, see also Hu \\& Dodelson 2002 and references therein). The Wilkinson Microwave Anisotropy Probe (WMAP, see Bennett et al. 2003a) satellite\\footnote{map.gsfc.nasa.gov/} released the first year, all sky CMB observations mapping anisotropies down to an angular scale of about $16'$ in total intensity and its correlation with polarization, on five frequency channels extending from 22 to 90 GHz. In the future, balloon-borne and ground based observations will attempt to measure the CMB polarization on sky patches (see Kovac et al. 2002 for a first detection); the {\\sc Planck}\\footnote{astro.estec.esa.nl/SA-general/Projects/Planck} satellite, scheduled for launch in 2007 (Mandolesi et al. 1998, Puget et al. 1998), will provide total intensity and polarization full sky maps of CMB anisotropy with resolution $\\gsim 5'$ and a sensitivity of a few $\\mu$K, on nine frequencies in the range 30-857 GHz. A future satellite mission for polarization is currently under study\\footnote{CMBpol, see spacescience.nasa.gov/missions/concepts.htm}.  Correspondingly, the data analysis science faces entirely new and challenging issues in order to handle the amount of incoming data, with the aim of extracting all the relevant physical information about the cosmological signal and the other astrophysical emissions, coming from extra-galactic sources as well as from our own Galaxy. The sum of these foreground emissions, in total intensity, is minimum at about 70 GHz, according to the first year WMAP data (Bennett et al. 2003b). In the following we refer to low and high frequencies meaning the ranges below and above that of minimum foreground emission.  At low frequencies the main Galactic foregrounds are synchrotron (see Haslam et al. 1982 for an all sky template at 408 MHz) and free-free (traced by H$\\alpha$ emission, see Haffner, Reynolds, \\& Tufte 1999, Finkbeiner 2003 and references therein) emissions, as confirmed by the WMAP observations (Bennett et al. 2003b). At high frequencies, Galactic emission is expected to be dominated by thermal dust (Schlegel, Finkbeiner, Davies 1998, Finkbeiner, Schlegel, Davies 1999). Moreover, several populations of extra-galactic sources, with different spectral behavior, show up at all the frequencies, including radio sources and dusty galaxies (see Toffolatti et al. 1998), and the Sunyaev-Zeldovich effect from clusters of galaxies (Moscardini et al. 2002). Since the various emission mechanisms have generally different frequency dependencies, it is conceivable to combine multi-frequency maps in order to separate them.  A lot of work has been recently dedicated to provide algorithms devoted to the component separation task, exploiting different ideas and tools from signal processing science. Such algorithms generally deal separately with point-like objects like extra-galactic sources (Tenorio et al. 1999, Vielva et al. 2001), and diffuse emissions from our own Galaxy. In this work we focus on techniques developed to handle diffuse emissions; such techniques can be broadly classified in two main categories.  The ``non-blind\" approach consists in assuming priors on the signals to recover, on their spatial pattern and frequency scalings, in order to regularize the inverse filtering going from the noisy, multi-frequency data to the separated components. Wiener filtering (WF, Tegmark, Efstathiou 1996, Bouchet, Prunet, Sethi 1999) and Maximum Entropy Method (MEM, Hobson et al. 1998) have been tested with good results, even if applied to the whole sky (Stolyarov et al. 2002). Part of the priors can be obtained from complementary observations, and the remaining ones have to be guessed. The WMAP group (Bennett et al. 2003b) exploited the available templates mentioned above as priors for a successful MEM-based component separation.  The ``blind\" approach consists instead in performing separation by only assuming the statistical independence of the signals to recover, without priors either for their frequency scalings, or for their spatial statistics. This is possible by means of a novel technique in signal processing science, the Independent Component Analysis (ICA, see Amari \\& Chichocki 1998 and references therein). The first astrophysical application of this technique (Baccigalupi et al. 2000) exploited an adaptive (i.e. capable of self-adjusting on time streams with varying signals) ICA algorithm, working successfully on limited sky patches for ideal noiseless data. Maino et al. (2002) implemented a fast, non-adaptive version of such algorithm ({\\ica}, see Hyv\\\"{a}rinen 1999) which was successful in reaching separation of CMB and foregrounds for several combinations of simulated all sky maps in conditions corresponding to the nominal performances of {\\sc Planck}, for total intensity measurements. Recently, Maino et al. (2003) were able to reproduce the main scientific results out of the COBE data exploiting the {\\ica} technique. The blind techniques for component separation represent the most unbiased approach, since they only assume the statistical independence between cosmological and foreground emissions. Thus they not only provide an independent check on the results of non-blind separation procedures, but are likely to be the only viable way to go when the foreground contamination is poorly known.  In this work we apply the {\\ica} technique to astrophysical polarized emission. CMB polarization is expected to arise from Thomson scattering of photons and electrons at decoupling. Due to the tensor nature of polarization, physical information is coded in a entirely different way with respect to total intensity. Cosmological perturbations may be divided into scalars, like density perturbations, vectors, for example vorticity, and tensors, i.e. gravitational waves (see Kodama \\& Sasaki 1984). Total intensity CMB anisotropies simply sum up contributions from all kinds of cosmological perturbations. For polarization, two non-local combinations of the Stokes parameters $Q$ and $U$ can be built, commonly known as $E$ and $B$ modes (see Zaldarriaga, \\& Seljak 1997, and Kamionkowski, Kosowsky \\& Stebbins 1997 featuring a different notation, namely gradient $G$ for $E$ and curl $C$ for $B$). It can be shown that the $E$ component sums up the contributions from all the three kinds of cosmological perturbations mentioned above, while the $B$ modes are excited via vectors and tensors only. Also, scalar modes of total intensity, which we label with $T$ in the following, are expected to be strongly correlated with $E$ modes: indeed, the latter are merely excited by the quadrupole of density perturbations, coded in the total intensity of CMB photons, as seen from the rest frame of charged particles at last scattering (see Hu et al. 1999 and references therein). Therefore, for CMB, the correlation $TE$ between $T$ and $E$ modes is expected to be the strongest signal from polarization. The latter expectation has been confirmed by WMAP (Kogut et al. 2003) with a spectacular detection on degree and super-degree angular scales; moreover, a first detection of CMB $E$ modes has been obtained (Kovac et al. 2002).  This phenomenology is clearly much richer with respect to total intensity, and motivated a great interest toward CMB polarization, not only as a new data set in addition to total intensity, but as the best potential carrier of cosmological information via electromagnetic waves. Unfortunately, as we describe in the next Section, foregrounds are even less known in polarization than in total intensity, see De Zotti (2002) and references therein for reviews. For this reason, it is likely that a blind technique will be required to clean CMB polarization from contaminating foregrounds. The first goal of this work is to present a first implementation of the ICA techniques on polarized astrophysical maps. Second, we want to estimate the precision with which CMB polarized emission will be measured in the near future. We exploit the {\\ica} technique on low frequencies where some foreground model have been carried out (Giardino et al. 2002, Baccigalupi et al. 2001).  The paper is organized as follows. In Section \\ref{simulated} we describe how the simulation of the synchrotron emission were obtained. In Section \\ref{component} we describe our approach to component separation for polarized radiation. In Section \\ref{performance} we study the {\\ica} performance on our simulated sky maps. In Section \\ref{application} we apply our technique to the {\\sc Planck} simulated data, studying its capabilities for polarization measurements in presence of foreground emission. Finally, Section \\ref{concluding} contains the concluding remarks. ",
        "conclusions": "\\label{concluding}  Forthcoming experiments are expected to measure CMB polarization\\footnote{see lambda.gsfc.nasa.gov/ for a collection of presently operating and future CMB experiments}. The first detections have been obtained on pure polarization (Kovac et al. 2002), as well on its correlation with total intensity CMB anisotropies, by the Wilkinson Microwave Anisotropy Probe (WMAP) satellite\\footnote{map.gsfc.nasa.gov/}.  The foreground contamination is mildly known for total intensity measurements, and poorly known for polarization (see De Zotti 2002 and references therein). It is therefore crucial to develop data analysis tools able to clean the polarized CMB signal from foreground emission by exploiting the minimum number of a priori assumptions. In this work, we implemented the Fast Independent Component Analysis technique in an astrophysical context ({\\ica}, see Amari, Chichocki 1998, Hyv\\\"{a}rinen 1999, Baccigalupi et al. 2000, Maino et al. 2002) for blind component separation to deal with astrophysical polarized radiation.  In our scheme, component separation is performed {both on the Stokes parameters $Q$ and $U$ maps independently and by joining them in a single dataset}. $E$ and $B$ modes, coding CMB physical content in the most suitable way (see Zaldarriaga, Seljak 1997, Kamionkowski, Kosowsky, Stebbins 1997), are then built out of the separation outputs. We described how to estimate the noise on {\\ica} outputs, on $Q$ and $U$ as well as on $E$ and $B$.  We tested this strategy on simulated polarization microwave all sky maps containing a mixture of CMB and Galactic synchrotron. CMB is modelled close to the current best fit (Spergel et al. 2003), with a component of cosmological gravitational waves at the $30\\%$ level with respect to density perturbations. We also included re-ionization, although with an optical depth lower than indicated by the WMAP results (Bennett et al. 2003a) since they came while this work was being completed, but consistent with the Gunn-Peterson measurements by Becker et al. (2001). Galactic synchrotron was modelled with the two existing templates by Giardino et al. (2002) and Baccigalupi et al. (2001). These models yield approximately equal power on angular scales above the degree, dominating over the expected CMB power. On sub-degree angular scales, the Giardino et al. (2002) model predicts an higher power, but still subdominant compared to the CMB $E$ mode acoustic oscillations. Note that at microwave frequencies, the fluctuations at high multipoles ($\\ell \\gsim 1000$), corresponding to a few arcmin angular scales, are likely dominated by compact or flat spectrum radio sources (Baccigalupi et al. 2002b, Mesa et al. 2002). Their signal is included in the maps used to estimate the synchrotron power spectrum.  We studied in detail the limiting performance in the noiseless case, as well as the degradation induced by a Gaussian, uniformly distributed noise, by considering two frequency combinations: 30 $+$ 44 GHz and 70 $+$ 100 GHz. In the noiseless case, the algorithm is able to recover CMB $E$ and $B$ modes on all the relevant scales. In particular, this result is stable against the space variations of the synchrotron spectral index indicated by the existing data. In this case, {\\ica} is able to converge to an average synchrotron component, characterized by a ``mean\" spectral index across the sky, and to remove it efficiently from the map. The output CMB map, also containing residual synchrotron due to its space varying spectral index, is mostly good as far as the frequencies considered are those where the synchrotron contamination is weaker.  By switching on the noise we found that separation, at least for what concerns the CMB $E$ mode, is still satisfactory for noise exceeding the CMB but not the foreground emission. The reason is that in these conditions the algorithm is still able to catch and remove the synchrotron component efficiently. {We implemented a Monte Carlo chain varying the CMB and the noise realizations in order to show that the performance quoted above is typical and does not depend on the particular case studied. Moreover, we studied how the foreground emission biases the recovered CMB map, by computing maps of residuals, i.e. subtracting the true CMB map out of the recovered one. In the noiseless case, the residual is just a copy of the foreground emission, with amplitude decreased proportionally to the accuracy of the separation matrix. In the noisy case, for interesting noise amplitudes the residual maps are dominated by the noise in the input data, linearly mixed with the separation matrix. The situation is obviously worse for the weaker CMB $B$ mode.  We applied these tools making reference to the {\\sc Planck} polarization capabilities, in terms of frequencies, angular resolution and noise, to provide a first example of how the {\\ica} technique could be relevant for high precision large polarization data-sets. We addressed separately the analysis of the CMB $E$, $B$, and $TE$ modes. {While this work was being completed, the Low Frequency Instrument (LFI) lost its 100 GHz channel, having polarization sensitivity. However polarimetry at this frequency could be restored if the 100 GHz channel of the High Frequency Instrument (HFI) is upgraded, as is presently under discussion. Due to the scientific content of the CMB polarization signal, the {\\sc Planck} polarization sensitivity deserves a great attention. Within our context here, it is our intention to support the importance of having polarization capabilities in all the cosmological channels of {\\sc Planck}, and in particular at 100 GHz. Our results have been obtained under this assumption.}  To improve the signal statistics, we found convenient to consider at least three frequency channels in the separation procedure, including the ones where the CMB is strongest, 70 and 100 GHz, plus one out of the two lower frequency channels, at 30 and 44 GHz. Since the latter have lower resolution we had to degrade the higher frequency maps since the present {\\ica} architecture cannot deal with maps having different resolutions. CMB $E$ and $TE$ modes were accurately recovered for both the synchrotron models considered. The $B$-mode power spectrum is recovered on very large angular scales in the presence of a conspicuous re-ionization bump. On smaller scales, where the $B$-mode power mainly comes from cosmological gravitational waves, the recovery is only marginal for a $30\\%$ tensor to scalar perturbation ratio.  On the sub-degree angular scales the contamination from synchrotron is almost irrelevant according to both models (Giardino et al. 2002, Baccigalupi et al. 2001). Moreover, it is expected that Galactic $E$ and $B$ modes have approximately the same power (Zaldarriaga 2001), while for CMB the latter are severely damped down since they are associated with vector and tensor perturbations, vanishing on sub-horizon scales at decoupling corresponding to a degree or less in the sky (see Hu et al. 1999). This argument holds also if the $B$ mode power is enhanced by weak lensing effects from matter structures along the line of sight (see Hu 2002 and references therein). Therefore, on these scales, we expect the $E$ power spectrum to be a sum of Galactic and CMB contributions, while the $B$ power comes essentially from foregrounds only. In these conditions, the CMB $E$ power spectrum is recovered by simply subtracting the $B$ power spectrum.  We also estimated the $TE$ contamination from synchrotron to be irrelevant for CMB, because of the strength of the CMB $TE$ component due to the intrinsic correlation between scalar and quadrupole modes exciting $E$ polarization. By applying these considerations on sub-degree angular scales, as well as the results of the {\\ica} procedure described above on larger scales, we show how the {\\sc Planck} instrument is capable of recovering the CMB $E$ and $TE$ spectra on all scales down to the instrumental resolution, corresponding to a few arc-minutes scales. In terms of multipoles, the $E$ and $TE$ angular power spectra are recovered up to $1000$ and $1200$, respectively.  Summarizing, we found that the {\\ica} algorithm, when applied to a {\\sc Planck}-like experiment, could be able to substantially clean the foreground contamination on the relevant multipoles, corresponding to degree angular scales and above. Since the foreground contamination on sub-degree angular scales is expected to be subdominant, the CMB $TE$ and $E$ modes are recovered on all scales extending from the whole sky to a few arc-minutes. In particular, the {\\ica} algorithm can clean the $B$-mode power spectrum up to the peak due to primordial gravitational waves if the cosmological tensor amplitude is at least $30\\%$ of the scalar one. In particular, we find that on large angular scales, of a degree and more, foreground contamination is expected to be severe and the known blind component separation techniques are able to efficiently clean the map from such contamination, as it is presently known or predicted.  Still, despite these good results, the main limitation of the present approach is the neglect of any instrumental systematics. While it is important to assess the performance of a given data analysis tool in the presence of the nominal instrumental features, as we do here, a crucial test is checking the stability of such tool with respect to relaxation of the assumptions regarding the most common sources of systematic errors, like beam asymmetry, non-uniform and/or non-Gaussian noise distribution etc., as well as the idealized behavior of the signals to recover. In this work we had a good hint about the second aspect, since we showed that {\\ica} is stable against relaxation of the assumption, common to all component separation algorithms developed so far, about the separability between space and frequency dependence for all the signal to recover. In a forthcoming work we will investigate how ICA based algorithms for blind component separation deal with maps affected by the most important systematics errors."
    },
    "0209/astro-ph0209072_arXiv.txt": {
        "abstract": "The mass-energy formula for a black hole endowed with electromagnetic structure (EMBH) is clarified for the nonrotating case. The irreducible mass $M_{\\mathrm {irr}}$ is found to be independent of the electromagnetic field and explicitly expressable as a function of the rest mass, the gravitational energy and the kinetic energy of the collapsing matter at the horizon. The electromagnetic energy is distributed throughout the entire region extending from the horizon of the EMBH to infinity. We discuss two conceptually different mechanisms of energy extraction occurring respectively in an EMBH with electromagnetic fields smaller and larger than the critical field for vacuum polarization. For a subcritical EMBH the energy extraction mechanism involves a sequence of discrete elementary processes implying the decay of a particle into two oppositely charged particles. For a supercritical EMBH an alternative mechanism is at work involving an electron-positron plasma created by vacuum polarization. The energetics of these mechanisms as well as the definition of the spatial regions in which thay can occur are given. The physical implementations of these ideas are outlined for ultrahigh energy cosmic rays (UHECR) and gamma ray bursts (GRBs). ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209558_arXiv.txt": {
        "abstract": "I analyze the properties of the first galaxies in cosmological simulations with radiative feedback. Preliminary results indicate similarities with the observed properties of the bulk of dwarf spheroidal galaxies (dSphs) in the Local Group and Andromeda. I briefly discuss observational tests that could help in understanding the impact of a population of small primordial objects on the cosmic evolution. ",
        "introduction": "}  Ricotti, Gnedin and Shull 2002a,b \\nocite{RicottiGSa:02} \\nocite{RicottiGSb:02} have studied radiative feedback processes that regulate the formation of the first galaxies. Contrary to normal galaxies, the global star formation in these objects is self-regulated on cosmological scales. This happens because internal and external radiative feedback processes are important in triggering or suppressing their ability for form stars. Typically star formation in the first galaxies is bursting and the emitted ionizing photons remain confined in the denser filaments of the intergalactic medium (IGM), preventing a complete IGM reionization. The main parameter that regulates the star formation is \\fesc: the escape fraction of ionizing photons. In this talk I present preliminary results on the properties of the first galaxies in our simulations. I try to understand the differences and similarities with observed dSphs and discuss the observational consequences of the existence of such a population of small primordial galaxies \\footnote{By definition these ``microgalaxies'' form in dark matter (DM) ``minihalos'' with masses $M_{DM} \\le 10^8$ M$_\\odot$.  If the gas is of primordial composition (metal free), molecular hydrogen is the only coolant available to form dense gas clouds and the first stars (Population~III).  There is not consent on a unique name for this population of primordial objects, sometimes also called PopIII objects or ``small-halo'' galaxies.}.  DSphs have masses comparable to those of the first galaxies, but this does not necessarily imply that they are relics of primordial objects.  It is possible that part or all of the observed dSphs are galaxies formed later from larger Dark Matter (DM) halos, subsequently stripped of part of their DM content.  Our goal is to distinguish between these two formation scenarios comparing observed properties with simulated properties. ",
        "conclusions": "I have shown preliminary results, part of a larger work currently in progress (Ricotti, Gnedin and Shull 2003, Ricotti and Gnedin 2003, in preparation), to constrain and understand the theory for the formation of the first galaxies in the universe. I have shown how observations of dwarf galaxies in the Local Group can be used to constrain theoretical results of cosmological simulations with radiative feedback. If we establish a connection between dSphs and relics of the first galaxies we can hope to learn in some detail the physics that regulates the formation of the first stars in the universe, and the importance of Population~III stars. The answer to currently popular quests on the stellar initial mass function, stellar nucleosynthesis, star formation efficiency of Population~III stars at $z \\sim 30$, could be found studying the most numerous an closest galaxies to the Milky Way.  Studies of thermal and chemical evolution of the intergalactic medium as a function of redshift and overdensity can further constrain the model once a realistic treatment of supernova feedback is included in the simulation (Ricotti, Gnedin and Shull 2003, in preparation)."
    },
    "0209/astro-ph0209244_arXiv.txt": {
        "abstract": "Supersonic turbulence fragments the interstellar medium into dense sheets, filaments, cores and large low density voids, thanks to a complex network of highly radiative shocks. The turbulence is driven on large scales, predominantly by supernovae. While on scales of the order of the disk thickness the magnetic energy is in approximate equipartition with the kinetic energy of the turbulence, on scales of a few pc the turbulent kinetic energy significantly exceeds the magnetic energy.  The scaling properties of supersonic turbulence are well described by a new analytical theory, which allows to predict the structure functions of the density and velocity distributions in star-forming clouds up to very high order.  The distribution of core masses depends primarily on the power spectrum of the turbulent flow, and on the jump conditions for isothermal shocks in a magnetized gas. For the predicted velocity power spectrum index $\\beta=1.74$, consistent with results of numerical experiments of supersonic turbulence as well as with Larson's velocity-size relation, one obtains by scaling arguments a power law mass distribution of dense cores with a slope equal to $3/(4-\\beta)$ = 1.33, consistent with the slope of the Salpeter stellar initial mass function (IMF). Results from numerical simulations confirm this scaling. Both the analytical model for the stellar IMF and its numerical estimate show that turbulent fragmentation can also explain the origin of brown dwarfs. The analytical predictions for the relative abundance of brown dwarfs are confirmed by the observations.  The main conclusion is that the stellar IMF directly reflects the mass distribution of prestellar cores, due predominantly to the process of turbulent fragmentation. ",
        "introduction": "Turbulence in the interstellar medium (ISM) of the Milky Way -- and more generally turbulence in the discs of other galaxies -- is of crucial importance for both the structure and evolution of the galaxy.  The importance of turbulence is both direct, through its influence on the pressure equilibrium and stratification, and indirect, through its influence on the star formation process.  The vertical pressure equilibrium and stratification of the ISM is determined by the level of turbulence, together with the temperature distribution of the medium (which is in turn probably tightly coupled to the turbulence), and it is likely that even the distributions of magnetic fields and cosmic ray particles, which also contribute to the pressure and stratification, are integral parts of the same process; it is unlikely that the near equipartition of the energy content of turbulence, magnetic fields, and cosmic ray particles is a mere coincidence.  It has long been realized that turbulence in the interstellar medium, in particular in the cold, molecular cloud components, is highly supersonic \\cite{% 1979MNRAS.186..479L,% 1981MNRAS.194..809L,% 1982MNRAS.200..159L,% 1990ApJ...359..344F,% 1992A&A...257..715F}. More recently, it has been realized that the supersonic nature of the turbulence is a boon, rather than a nuisance, when trying to understand the properties of the ISM, the cold molecular clouds, and star formation \\cite{% 1995MNRAS.277..377P,% 1999ApJ...526..279P,% 2000ApJ...530..277E}. It turns out that supersonic turbulence is in many respect similar to ordinary, subsonic turbulence, and that it thus has a number of generic, statistical properties.  Much like ordinary turbulence, its decay time is of the order of the dynamical time, even in the MHD-case \\cite{% MacLow_Puebla98,% 1998ApSS.261..195M,% 1999ApJ...524..169M,% Stone+98,% 1999ApJ...526..279P}. And much like ordinary turbulence, it is characterized by power law velocity power spectra and structure functions over an inertial range of scales \\cite{% Boldyrev,% Boldyrev+1,% Boldyrev+2}.  An important difference between supersonic and subsonic turbulence is the distribution of density.  A supersonic medium is, by definition, highly compressible; on average its gas pressure $P_g$ is small relative to the dynamic pressure $\\rho u^2$.  As a consequence, a supersonic medium is characterized by a wide distribution of densities.  A turbulent and isothermal supersonic medium has a log-normal density probability distribution (PDF) \\cite{% 1994ApJ...423..681V,% 1997MNRAS.288..145P,% Passot+98}, with a dispersion of linear density proportional to the Mach number \\cite{% 1997MNRAS.288..145P,% 1999intu.conf..218N,% 2001ApJ...546..980O}. Cold molecular clouds are indeed approximately isothermal, and are known to have a very intermittent density distribution, consistent with the properties of isothermal supersonic turbulence \\cite{% 1997ApJ...474..730P}.  Deviations from isothermal conditions are in general of the type where compression leads to even lower temperatures (effective gas gamma less than unity) \\cite{1998ApJ...504..835S}, resulting in a density PDF skewed towards greater probability at high densities.  The PDF may be described as a skewed log-normal, with a high density asymptote that formally tends to a power law in the limit $T \\rightarrow 0$ \\cite{% 1998ApJ...504..835S,1999intu.conf..218N}.  Effectively then, supersonic turbulence acts to fragment the ISM, causing local density enhancements also over a range of geometrical scales.  Molecular clouds themselves represent relatively large scale density enhancements, probably caused by the random convergence of large scale ISM velocity features \\cite{1999ApJ...515..286B,1999ApJ...527..285B,2001ApJ...562..852H}. Inside molecular clouds smaller scale turbulence leads to high contrast local density enhancements in corrugated shocks, intersections of shocks, and in knots at the intersection of filaments.  Such small scale density enhancements are `up to grabs' by gravity; if their density is sufficiently high, relative to their temperature and the local magnetic field strength, they form pre-stellar cores, and eventually collapse to form stars.  The decisive importance of turbulence in this process makes it possible to predict the distribution of masses of the pre-stellar cores, and hence the distribution of new borne stars, the initial mass function (IMF) \\cite{Padoan+Nordlund-IMF,Padoan_etal-IMF}.  The process of star formation is indeed crucial to understand. Only by understanding star formation, qualitatively and quantitatively, can we understand galaxy formation. We need to understand evolution effects to answer questions such as ``Was star formation different in the Early Universe?''. We need to understand environmental effects to answer questions such as ``Do other galaxies have different `Larson laws'?''  We also need to understand star formation to answer questions related to Gamma-Ray Bursts; e.g., ``Are Very Massive Stars progenitors of Gamma-Ray Bursts?'', and ``What environment does the blast wave associated with Gamma-Ray Bursts encounter''?  Finally, we would like to understand star formation as such, because it is a neat problem -- one that involves supersonic, selfgravitating MHD turbulence and thus was thought to be enormously difficult.  With access to supercomputer modeling the problem has become tractable, and it has turned out \\emph{a posteriori} that it is even partly amenable to analytical theory.  In the subsequent sections of this tutorial star formation and turbulence in the interstellar medium is discussed in more detail.  Section \\ref{sne.sec} discusses supernova driving of the ISM, Section \\ref{sst.sec} discusses properties of supersonic turbulence, Section \\ref{stas.sec} summarizes a new theory of supersonic turbulence, while Section \\ref{imf.sec} discusses star formation and the initial mass function.  Conclusions are summarized in Section \\ref{concl.sec}. ",
        "conclusions": "\\label{concl.sec}  The main conclusion from the preceding sections is that the statistics of star formation is primarily controlled by supersonic turbulence, rather than by gravity.  Star formation takes place in cold molecular clouds, which are part of a turbulent cascade in the interstellar medium.  The ultimate energy input to the cascade comes from supernovae, with a possibly significant contribution from local variations of the galactic rotation curve (density waves). The clouds owe their existence to random convergence of the interstellar medium velocity field, which creates local density enhancements over a range of scales.  The internal, supersonic and super-Alfv{\\'e}nic velocity field in molecular clouds is responsible for their fragmentation, thus preventing global collapse but triggering local collapse at the many local density maxima whose mass exceeds the local Jeans' mass. Such prestellar cores are formed as sheet corrugations and filamentary density enhancements, and are taken over by self-gravity only after they have been shaped by the turbulence.  The velocity field of the cascade is dominated by power in solenoidal (shearing) motions, even though it is supersonic and super-Alfv{\\'e}nic.  Its spectrum of kinetic energy is less steep than its velocity and magnetic field power spectra, which explains how conditions can be super-Alfv{\\'e}nic on small (molecular cloud) scales, even though there is rough equipartition between magnetic and kinetic energy density on large (disk thickness) scales.  A Salpeter like IMF is the result of the near-self-similar, power law nature of turbulence in molecular clouds, in combination with density jump amplitudes determined by MHD-shock jump conditions.  Star formation (at least in our galaxy) bites its own tail; it is driven by supernovae and at the same time the birth of massive stars gives rise to new supernovae that re-enforce the driving.  External sources of turbulence, such as kinetic energy from galaxy collisions and merging may be the primary driving agent in star-burst galaxies.  Different physical conditions (primarily higher temperatures and lower metal abundances) in the Early Universe would lead to higher mass at the low-mass cut-off, and a much weaker magnetic field would lead to a steeper IMF slope."
    },
    "0209/astro-ph0209008_arXiv.txt": {
        "abstract": "We propose the Galactic flare model for the origin of the X-ray gas in the Galactic halo. For this purpose, we examine the magnetic reconnection triggered by Parker instability (magnetic buoyancy instability), by performing the two-dimensional resistive numerical magnetohydrodynamic simulations. As a result of numerical simulations, the system evolves as following phases: Parker instability occurs in the Galactic disk. In the nonlinear phase of Parker instability, the magnetic loop inflates from the Galactic disk into the Galactic halo, and collides with the anti-parallel magnetic field, so that the current sheets are created in the Galactic halo. The tearing instability occurs, and creates the plasmoids (magnetic islands). Just after the plasmoid ejection, further current-sheet thinning occurs in the sheet, and the anomalous resistivity sets in. Petschek reconnection starts, and heats the gas quickly in the Galactic halo. It also creates the slow and fast shock regions in the Galactic halo. The magnetic field ($B\\sim 3\\ \\mu$G), for example, can heat the gas ($n\\sim 10^{-3}$ cm$^{-3}$) to temperature of $\\sim 10^6$ K via the reconnection in the Galactic halo. The gas is accelerated to Alfv\\'en velocity ($\\sim 300$ km s$^{-1}$). Such high velocity jets are the evidence of the Galactic flare model we present in this paper, if the Doppler shift of the bipolar jet is detected in the Galactic halo. \\\\ ",
        "introduction": "The X-rays from hot gas are observed in the Galactic halo. Its luminosity and temperature are $L_{\\rm X}\\sim 7\\times 10^{39}$ erg s$^{-1}$ and $T\\sim 10^6$ K (\\cite{pie98}). The volume and thermal energy are estimated to be $E\\sim 10^{55}$ erg and $V\\sim 10^{68}$ cm$^3$. To explain such hot gas in the Galactic halo, the ``Galactic fountains'' model has been proposed (i.e., supernova remnants and stellar winds heat the gas; \\cite{bre80}; \\cite{nor89}; \\cite{sha91}; \\cite{shu96}; \\cite{avi00}; \\cite{sla00}). The energy source, however, may not be explained fully by this model because the evidence is not observed adequately (see also Birk, Lesch, \\& Neukirch 1998). In this paper, we propose another mechanism. It is the Galactic flare model, i.e., the magnetic heating in the Galactic halo.   Parker (1992) pointed out the importance of magnetic reconnection for the heating of Galactic plasmas. Whenever the magnetic flux collides with another flux, which is not exactly parallel, the current density increases, and a strong dissipation sets in (e.g., via anomalous resistivity) to trigger fast reconnection (e.g., \\cite{uga86}). The magnetic reconnection is observed in solar flares by the X-ray satellites {\\it Yohkoh} (\\cite{mas94}; \\cite{shi96}; \\cite{tsu96}) and {\\it SoHO} (\\cite{yok01}). In the solar flare, the reconnection heats the plasma from a temperature of $\\sim$ several $\\times 10^6$ K to $\\sim$ several $\\times 10^{7-8}$ K, and accelerates it to Alfv\\'en velocity ($\\sim 10^{2-3}$ km s$^{-1}$) (e.g., \\cite{shi96}). The reconnection would occur also in the Galaxy (Tanuma et al.\\ 1999a, 1999b, 2001a, 1999b), which may be called ``Galactic flare'' (\\cite{stu80}; \\cite{kah93}). Total magnetic energy is $E_{\\rm mag}\\sim$ $(\\langle B\\rangle_{\\rm obs}^2/8\\pi)$ $V_{\\rm G}$ $\\sim 10^{54.4}$ erg at least, where $\\langle B\\rangle_{\\rm obs}$($\\sim 3\\ \\mu$G) is the mean observed field strength (see \\cite{bec96}; \\cite{val97}), and $V_{\\rm G}$ ($\\sim 10^{67}$ cm$^{3}$) is the volume of the Galaxy. The rotational energy of the Galaxy ($\\sim 10^{58.9}$ erg) and kinetic energy of the interstellar gas ($\\sim 10^{58.2}$ erg) are its origin (e.g., \\cite{par71}; \\cite{stu80}; \\cite{tan99a}; \\cite{tan00}).    The steady reconnection mechanisms were proposed (see \\cite{pri00}). In Sweet(1958)-Parker(1957) type reconnection, the diffusion region is so long as to occupy whole current system. It can not be applied to the solar flare phenomena, because the reconnection rate of this model is too small ($\\sim R_{\\rm m}^{-1/2}$) in the solar corona, where $R_{\\rm m}$ is the magnetic Reynolds number ($\\sim 10^{12}$). On the other hand, in Petschek(1964) type reconnection, the diffusion region is localized near an X-point, and standing slow shocks occupy whole current systems. In this case, the energy conversion via slow shocks is much larger than Ohmic heating. This can hence be applicable to the solar flare phenomena, because the reconnection rate of this model is $\\sim 0.1-0.01$. This is called ``fast reconnection''. A basic problem of Petschek model is as follows: Petschek reconnection occurs, if the anomalous resistivity sets in the current sheet (e.g., \\cite{uga86}; \\cite{yok94}; \\cite{tan00}; Tanuma et al.\\ 1999a, 2001a). The anomalous resistivity set in, when the current-sheet thickness becomes comparable with ion Lamor radius or ion inertial radius. It is, however, not fully known how the current sheet becomes thin, because the typical size of solar flare ($10^{9-11}$ cm) is much larger than these radii ($10^{2-3}$ cm). This situation is similar to that of the Galaxy ($R_{\\rm m}>10^{15}$), where typical size of magnetic field ($>10^{19}$ cm) is much larger than the ion Lamor radius ($\\sim 10^7$ cm). To solve these problems, we proposed the current-sheet thinning via the ``fractal tearing instability'' (\\cite{tan00}; \\cite{shi01}; \\cite{tan01a}).    Many two-dimensional (2D) magnetohydrodynamic (MHD) numerical simulations have been carried out for the magnetic reconnection in the solar atmosphere (\\cite{mag97}; \\cite{ods97}), and in the Galactic halo (Zimmer, Lesch, \\& Birk 1997; \\cite{bir98}), by assuming the current sheet at the initial condition (see also \\cite{nit01}). Recently, Tanuma et al.\\ (1999a, 1999b, 2001a) examined the magnetic reconnection triggered by a supernova-shock by performing the 2D MHD simulations with a high spatial resolution, and proposed that it can generate X-ray gas in the Galactic disk (e.g., \\cite{ebi01}). They found that the tearing instability (\\cite{fur63}) occurs in the current sheet long after the passage of a shock wave. Petschek reconnection occurs after further current sheet thinning via secondary tearing instability. In the present model, Parker(1966) instability creates the current sheet by itself and trigger the magnetic reconnection (\\cite{tan99b}; \\cite{tan00}; see also Shibata, Nozawa, \\& Matsumoto 1992; Yokoyama \\& Shibata 1996, 1997).    Recently, we examied three-dimensional (3D) MHD simulations of the magnetic reconnection with a low spatial resolution. Petschek reconnection occurs after the current sheet thinning by the tearing instability in both 2D (\\cite{tan99b}) and 3D models (\\cite{tan00}; \\cite{tan01b}), because we can not resolve the secondary tearing instability when we assume a rough grid (The similarities between 2D and 3D models are consistent with Ugai \\& Shimizu 1996). 3D effect such as Rayleigh-Taylor instability, however, appears when reconnection jet collides with high pressure gas and magnetic loop much after the onset of Petschek reconnection. Tanuma et al.\\ (2002) applied the results to the creation of helical magnetic field and confinement of high energy particles in the solar flare. We study the basic physics of magnetic reconnection which are common between 2D and 3D models, although 2D model examined in this paper is a toy model of limited in 2D dimension. In this paper, however, we examine the 2D model under a higher spatial resolution than the 3D model which we are able to do.     Parker instability is the undular mode ($\\mbox{\\boldmath$k$}\\parallel \\mbox{\\boldmath$B$}$) of magnetic buoyancy instability (\\cite{par66}), which occurs if a gas layer in a gravitational field is supported by the horizontal magnetic fields. Suppose that the magnetic field lines are disturbed and begin to undulate. The gas in the loop top slides down along the field lines, so that loop rises further, and the instability sets in. Parker instability is suggested to influence the motion of clouds, H{\\small II} regions, and OB associations (\\cite{tos74}), and the distribution of clouds (\\cite{mou74}; \\cite{vrb77}; \\cite{bli80}; \\cite{elm82}); for example, Perseus hump (\\cite{sof74}), Perseus arm (Appenzeller 1971, 1974), Barnard loop (\\cite{mou74}), and Sofue-Handa(1984) lobe.   The linear analysis of Parker instability were made by many researchers (\\cite{shu74}; \\cite{hor88}; \\cite{han92}; \\cite{fog94}; \\cite{cho97}; \\cite{kam97}). The 2D MHD simulations were performed for solar flares (\\cite{kai90}; \\cite{noz92}; \\cite{shi92}; Yokoyama \\& Shibata 1996), and the interstellar medium (\\cite{bas97}; \\cite{mat98}; \\cite{san00}; \\cite{ste01}). The three-dimensional (3D) simulations of Parker instability of horizontal magnetic field in solar atmosphere and Galaxy (\\cite{mat93}; \\cite{kim01}; \\cite{han02}), and the twisted flux tube in the solar atmosphere (\\cite{mat98}; \\cite{abb00}; \\cite{fan01}; \\cite{mag01}), the Galaxy (\\cite{han00}; \\cite{fra02}), and accretion disks (\\cite{zie01}) are also performed. Shibata et al.\\ (1989, 1992) and Yokoyama \\& Shibata (1994, 1996, 1997) examined the magnetic reconnection triggered by Parker instability in the solar corona. Recently, Hanasz et al.\\ (2002) examined the 3D model of the magnetic reconnection in magnetic loop created by Parker instability with Coriolis force in the interstellar medium. In the present paper, we extend Shibata et al.\\ (1989, 1992) and Yokoyama \\& Shibata (1994, 1996, 1997)'s solar flare model to the Galaxy: The Galactic flare as the origin of the X-ray gas in the Galactic halo.     In this paper, we propose a possible origin of X-ray gas in the Galactic halo. In the next section, we describe the simulation method. In sections 3 and 4, we describe the results of numerical simulations, and discuss it. In the last section, we summarize this paper. ",
        "conclusions": "\\subsection{The Origin of X-Ray Gas in the Galactic Halo\\label{dis2}}   Parker instability is initiated by a small perturbation, a supernova explosion (\\cite{kam97}; \\cite{ste01}), collision of the high-velocity clouds (HVCs; \\cite{san00}), and cosmic rays etc. The X-ray gas can be generated, if the magnetic reconnection is triggered by Parker instability or the collision of HVCs (Kerp et al.\\ 1994, 1996).   The reconnection heats the gas to \\begin{equation} T\\sim 10^6 \\left({n\\over 10^{-3}\\ {\\rm cm}^{-3}}\\right)^{-1} \\left({B\\over 3\\ \\mu{\\rm G}}\\right)^2\\ \\rm K. \\end{equation} The reconnection also accelerates the gas to \\begin{equation} v_{\\rm A} \\sim 300 \\left({n\\over 10^{-3}\\ {\\rm cm}^{-3}}\\right)^{-1/2} \\left({B\\over 3\\ \\mu{\\rm G}}\\right)\\ \\rm km\\ s^{-1}. \\end{equation} The duration of fast reconnection is \\begin{eqnarray} t&\\sim& {l\\over \\epsilon v_{\\rm A}}\\\\ &\\sim& 3000 \\left({l\\over 100\\ {\\rm pc}}\\right) \\left({\\epsilon\\over 0.1}\\right)^{-1}\\nonumber\\\\ & & \\left({n\\over 10^{-3}\\ {\\rm cm}^{-3}}\\right)^{1/2} \\left({B\\over 3\\ \\mu{\\rm G}}\\right)^{-1}\\ \\rm yr, \\end{eqnarray} where the reconnection rate is $\\epsilon$($=v_{\\rm in}/v_{\\rm A}$).  The anomalous resistivity sets in at least when the current sheet thickness becomes comparable with ion Lamor radius [$\\sim 3\\times 10^7$ $(T/10^4\\ {\\rm K})^{1/2}$ $(B/3\\mu\\ {\\rm G})^{-1}$ cm]. The radius of magnetic island (i.e., plasmoid) is larger than it. The island in 2D simulation is helically twisted magnetic tube in 3D. Its volume, mass, and luminosity are \\begin{eqnarray} V_{tube}&\\sim& 10^{61} \\left({r_p\\over 10\\ {\\rm pc}}\\right)^2 \\left({l_p\\over 100\\ {\\rm pc}}\\right)\\ \\rm cm^3,\\\\ M_{tube}&\\sim& 2\\times 10^{33} \\left({r_p\\over 10\\ {\\rm pc}}\\right)^2 \\left({l_p\\over 100\\ {\\rm pc}}\\right) \\left({n  \\over 1\\ {\\rm cm}^{-3}}\\right)\\ \\rm g\\\\ L_{tube}&\\sim& n^2\\Lambda (T)V_{tube}\\nonumber\\\\ &\\sim& 10^{34} \\left({n\\over 10^{-3}\\ {\\rm cm^{-3}}}\\right)^2 \\left({\\Lambda(T)\\over 10^{-21}\\ {\\rm erg\\ cm^3\\ s^{-1}}}\\right)\\nonumber\\\\ & & \\left({r_p\\over 10\\ {\\rm pc}}\\right)^2 \\left({l_p\\over 100\\ {\\rm pc}}\\right), \\end{eqnarray} where $r_p$ and $l_p$ are radius and length of the plasmoid, and $\\Lambda(T)$ is the cooling function. Then, the number of magnetic tubes at such size is $\\sim 10^{7-8}$.    Total released magnetic energy is \\begin{equation} E_{\\rm mag}\\sim 10^{51} \\left({B\\over 3\\ \\mu{\\rm G}}\\right)^2 \\left({\\lambda_{\\rm tot}\\over 1\\ {\\rm kpc}}\\right)^2 \\left({l\\over 100\\ {\\rm pc}}\\right)\\ \\rm erg, \\end{equation} where $\\lambda_{\\rm tot}^2$ is the total area of many reconnection regions, and $l$ are the typical thickness of magnetic loop. The energy release rate, then, is \\begin{eqnarray} {{\\rm d}E_{\\rm mag}\\over {\\rm d}t} &\\sim& 10^{40} \\left({B\\over 3\\ \\mu{\\rm G}}\\right)^3 \\nonumber\\\\ & & \\left({\\lambda_{\\rm tot}\\over 1\\ {\\rm kpc}}\\right)^2 \\left({n\\over 10^{-3}\\ {\\rm cm}^{-3}}\\right)^{-1/2}\\ \\rm erg\\ s^{-1}. \\end{eqnarray} It is also derived from eq.\\ (\\ref{dedt}). It can explain the X-ray luminosity ($\\sim 10^{39-40}$ erg s$^{-1}$; \\cite{pie98}). The heated gas is confined for $\\tau_{\\rm cond} \\sim 10^9 (n/10^{-3}\\ {\\rm cm}^{-3}) (\\lambda_{\\rm eff}/3\\ {\\rm kpc})^2 (T/10^6\\ {\\rm K})^{-5/2}$ yr (eq.\\ [\\ref{cond2}]), where $\\lambda_{\\rm eff}$ is the effective length of the helical magnetic tubes. The reconnection creates the bipolar jet, forming the high velocity gas at Alfv\\'en velocity ($\\sim 300$ km s$^{-1}$; see also \\cite{nit01}, \\cite{nit02}). The Doppler shift of the bipolar jet will be the evidence of the Galactic flare, proposed in this paper.  The plasmoid of cool gas is also created at the same time by the reconnection (\\cite{yok96}). It is also confined by magnetic field. The high velocity cool gas as well as hot gas will be the evidence of our model.    \\subsection{The Comparison With Other Numerical Simulations\\label{dis1}}  The 2D model examined in this paper is an extension from the supernova-shock driven reconnection model in the Galactic disk (Tanuma et al.\\ 1999a, 1999b, 2001a; Tanuma 2000). Some 2D numerical simulations were done for the reconnection in the Galaxy (\\cite{zim97}; \\cite{bir98}; Tanuma et al.\\ 1999a, 2001a). They assume the current sheet for the initial condition (see also \\cite{uga92}; \\cite{uga01}). Different from their models, Parker instability creates the current sheets spontaneously in our present model.    In our results, the tearing instability triggers Petschek reconnection. Sweet-Parker current sheet, however, would be created, and the secondary tearing instability will occur in the current sheet before Petschek reconnection, if we use fine grids like Tanuma et al.\\ (1999a, 2001a)s' model (Table \\ref{resmodel}). Furthermore, in the actual Galaxy ($>10^{15}$), as well as in solar corona ($\\sim 10^{12}$), the magnetic Reynolds number is much larger than that of the numerical simulations ($\\sim 10^{4-5}$), so that the current-sheet thickness must become much smaller than the current-sheet length, to set in the anomalous resistivity (e.g., through ``fractal tearing instability''; \\cite{tan00}; \\cite{shi01}; \\cite{tan01a}).    A different situation from the solar flare model is the growth of odd-mode of Parker instability. Another different result from Yokoyama \\& Shibata (1996)'s model is the time variation of reconnection rate in the uniform resistivity model (studied in section \\ref{secres}). On the other hand, in our model, Petschek reconnection occurs in the anomalous resistivity model, while Sweet-Parker reconnection occurs in the uniform resistivity model. This result is the same with Yokoyama \\& Shibata (1994). We confirmed this in more general situation.     \\subsection{Turbulence and 3D Effects}  The MHD turbulence may be also important in reconnection problem (\\cite{laz99}). The effective reconnection rate increases if the diffusion region is in a state of MHD turbulence. It is, however, difficult to resolve MHD turbulence (even if in 2D).   Recently, we revealed that the fast reconnection occurs after the current sheet thinning by the tearing instability in both 2D (\\cite{tan99b}) and 3D models (\\cite{tan00}; \\cite{tan01b}) under a low spatial resolution and the assumption of initial current sheet. We found no difference between 2D and 3D models in reconnection rate, inflow velocity, velocity of reconnection jet, temperature of heated gas, slow shock formation accompanied with Petschek reconnection, fast shock formation due to the collision between the reconnection jet and high pressure gas, and time scale of these phenomena. These results do not have a large quantitative difference from 2D models with a high spatial resolution examined by Tanuma et al.\\ (2001a) (see also \\cite{uga96}). Rayleigh-Taylor instability is, however, excited due to the collision, which occurs much after the onset of anomalous resistivity (\\cite{tan02}; in prep.\\ for study of details).  We can not examine 3D model with the same small grid with that of 2D model, so that we examine 2D model with a small grid in this paper. The secondary tearing instability, however, may be different between 2D and 3D, if we assume enough small grid. It is also important to study details of 3D structure of diffusion region, the tearing instability, plasmoid ejection, heating of X-ray gas in the Galactic halo, and creation of high velocity ``bipolar jet'', by performing 3D simulations with enough fine grid. It is our future work."
    },
    "0209/astro-ph0209522_arXiv.txt": {
        "abstract": "The Subaru Deep Field provides the currently deepest $K$-selected sample of high-$z$ galaxies ($K' \\sim 23.5$ at 5$\\sigma$). The SDF counts, colors, and size distributions in the near-infrared bands are carefully compared with pure-luminosity-evolution (PLE) as well as CDM-based hierarchical merging (HM) models. The very flat faint-end slope of the SDF $K$ count indicates that the bulk (more than 90\\%) of cosmic background radiation (CBR) in this band is resolved, even if we take into account every known source of incompleteness. The integrated flux from the counts is only about a third of reported flux of the diffuse CBR in the same band, suggesting that a new distinct source of this missing light may be required. We discovered unusually red objects with colors of $(J-K) \\ga$ 3--4, which are even redder than the known population of EROs, and difficult to explain by passively evolving elliptical galaxies. A plausible interpretation, which is the only viable one among those we examined, is that these are dusty starbursts at high-$z$ ($z \\sim 3$), whose number density is comparable with that of present-day ellipticals or spheroidal galaxies, as well as with that of faint submillimeter sources. The photometric redshift distribution obtained by $BVRIz'JK'$ photometries is also compared with the data, and the HM model is found to predict too few high-$z$ objects at $K' \\la 22$ and $z \\la 2$; the PLE model with reasonable amount of absorption by dust looks more consistent with the data. This result is apparently in contradiciton with some previous ones for shallower observations, and we discuss the origin of this.  These results raise a question for the HM models: how to form massive objects with starbursts at such high redshifts, which presumably evolve into present-day elliptical galaxies or bulges? ",
        "introduction": "The Subaru Deep Survey is a systematic project of the 8.2m Subaru telescope to study the deep extragalactic universe. The Subaru Deep Field was selected near the north Galactic pole, avoiding large Galactic extinction and nearby galaxy clusters, and the airmass of this field is smaller than the Hubble Deep Field at Mauna Kea (Maihara et al. 2000). The wide field near-infrared (NIR) camera CISCO took a very deep NIR $2'\\times 2'$ iamge in $J$ and $K'$ bands, with 5$\\sigma$ magnitude limits of 25.1 and 23.5. This is the deepest image in the $K$ band taken so far, providing a unique $K$-selected sample of galaxies which should be useful for study of faint, high-$z$ galaxies. The field was also deeply followed-up by optical instruments of FOCAS and Suprime-Cam. Here we review some interesting implications obtained by these data set, focusing on NIR galaxy counts, colors, and photometric-redshift distribution, compared with some theoretical models of galaxy formation and evolution. Although omitted here, some interesting results for the clustering of Lyman break galaxies and Lyman alpha emitters at $z \\sim 4$ have been obtained in the SDF and another project of the Subaru/XMM-Newton deep survey, thanks to the very wide field of the Suprime-Cam.  See Ouchi et al. (2001, 2002) for these. ",
        "conclusions": "From these results, we conclude that the deepest $K$-selected sample of the SDF is in overall agreement with the simple picture of PLE for early type or elliptical galaxies, but the present version of HM models has a problem in the redshift distribution of the faintest galaxies. Considering the overall success of the CDM structure formation theory against various tests ${\\it not}$ based on galaxy luminosity and star formation activities (e.g., clustering properties or abundance of galaxy clusters), it is reasonable to think that the problem identified for HM models is related with the treatment of star formation activity.  It must incorporate a population of massive and dusty starbursts at high redshift ($z \\ga 3)$, which presumably evolved into present-day elliptical galaxies or bulges without significant number evolution.  The author would like to thank all the collaborators of the SDF project, and the Subaru Telescope staffs who made this project possible. He has been financially supported in part by the JSPS Fellowship for Research Abroad."
    },
    "0209/astro-ph0209187_arXiv.txt": {
        "abstract": "The recent discovery,  based on a search of public OGLE data, of Be star candidates in the Small Magellanic Cloud showing spectacular photometric variations (Mennickent et al. 2002), has motivated further analysis  of their light curves. Here we report the results of a statistical study of the light curves of Type-1, Type-2 and Type-3 stars. Type-3 stars show a bimodal period distribution, with a main, broad peak, between 20 and 130 days, and a narrower, secondary peak, between 140 and 210 days. We also find that, among Type-3 stars, the maximum amplitude of oscillation correlates with the system luminosity, in the sense that only low luminosity systems show large amplitude oscillations. In general, the amplitude of these oscillations tends to increase with the photometric period.  A parametric study shows no correlation between the  amplitude, duration and asymmetry of Type-1 outbursts, and amplitude of Type-2 jumps, with the available stellar photometric parameters. ",
        "introduction": "Detailed  studies of Be stars in the Small Magellanic Cloud (SMC) have been performed only in recent years, being especially confined to open clusters like NGC 330 (e.g. Keller, Wood and Bessell 1999). These studies show the importance of studying Be stars in the low metallicity environment of the SMC since they serve as probes to test for the mechanisms of disk formation and of global disk oscillations (Baade et al.\\ 2001, Hummel et al.\\ 1999).  Recently, Mennickent et al. (2002, hereafter M02), based on a search of public OGLE data, reported the discovery of about 1000 Be star candidates in the SMC showing spectacular light curve variations. M02 classified their sample in Type-1 stars (showing outbursts), Type-2 stars (showing sudden luminosity jumps), Type-3 stars (showing periodic or near-periodic variations) and Type-4 stars (showing similar light curves that Galactic Be stars). M02 present examples of light curves for each one of these groups, giving some basic statistical information about colours and periodicities. Here we pursue this analysis, with a more detailed statistical study of the photometric properties of each group. ",
        "conclusions": "The first impression, after the visual inspection of Type-1 light curves, was that the outbursts were divided in hump-like and sharp outbursts. However, a clear separation between these groups is not confirmed by the distributions of the parameters $A$, $\\Delta T$ and $\\delta$, which instead suggest smooth transitions between one outburst type and the other. The above results do not help to discriminate between the possible hypotheses for these outbursts rised by M02, viz.\\, thermal instabilites in circumstellar discs or accretion by an unseen white dwarf.  The discovery of a bimodal period distribution in Type-3 stars is surprising and deserves further study. At present, we do not have an explanation for this finding, but we expect to make a deeper analysis of these stars incorporating spectroscopic data at the end of this year.  The fact that {\\it only} Type-3 systems with faint absolute magnitude show large amplitude oscillations could be the clue to understand the underlying cause of this phenomenon. M02 suggested that Type-3 variability could be due to some kind of oscillation in a circumstellar envelope. More massive envelopes could trigger larger amplitude oscillations. Perhaps low luminosity stars can host massive envelopes, whereas the higher radiation pressure of higher luminosity stars conspires against the formation and maintenance of massive envelopes. This could explain the absence of large amplitude variations in high luminosity systems."
    },
    "0209/astro-ph0209464_arXiv.txt": {
        "abstract": "{ We present high angular resolution millimeter wavelength continuum and \\cii\\ observations of the circumstellar disk surrounding the T\\,Tauri star DG~Tauri. We show that the velocity pattern in the inner regions of the disk is consistent with Keplerian rotation about a central 0.67\\,M$_\\odot$ star. The disk rotation is also consistent with the toroidal velocity pattern in the initial channel of the optical jet, as inferred from HST spectra of the first de-projected 100~AU from the source. Our observations support the tight relationship between disk and jet kinematics postulated by the popular magneto-centrifugal models for jet formation and collimation. ",
        "introduction": "The interplay between accretion and ejection of matter is believed to be a crucial element in the formation of stars. In particular, stellar jets may contribute substantially to the removal of excess angular momentum from the system, thereby allowing the central star to accrete to its final mass (e.g. Eisl\\\"{o}ffel et al.~\\cite{Eea00}, K\\\"{o}nigl \\& Pudritz~\\cite{KP00}, Shu et al.~\\cite{Sea00}). Most of the proposed models invoke the simultaneous action of magnetic and centrifugal forces in a rotating star/disk system threaded to open magnetic field lines. Even if widely accepted, these models have not yet been tested observationally on the launching scale (a few AU from the star), although this may be possible with the coming generation of interferometers. On much larger scales, there is some evidence for the requisite relationship between the jet and envelope kinematics in the HH\\,212 protostellar system (Davis et al.~\\cite{Dea00}). Hints of rotation are seen in the H$_2$ jet knots at 2$\\times$10$^3$ to 10$^4$\\,AU from the powering source; the sense of rotation is the same as that of the flattened envelope detected in NH$_3$ VLA observations (Wiseman et al.~\\cite{Wea01}). Although encouraging, these measurements probe regions too far from the central source to allow detailed comparison between the disk and jet kinematics.  Protostellar disk/jet systems are too embedded to probe the jet close to the launching region with current techniques which rely on optical and near infrared observations. Moreover, the kinematics of disk/envelope systems may encompass both rotational and infall motions, hampering tests of disk-jet interaction models. Optically visible T\\,Tauri stars, which have associated disks but little remnant envelopes, are much more suitable candidates for such studies.  The optical jet from the T\\,Tauri star DG\\,Tauri has been extensively studied at high resolution in recent years and displays properties that are in general agreement with magneto-centrifugal models for jet-launching (Dougados et al.~\\cite{Douea00}; Bacciotti et al.~\\cite{Bea00}; \\cite{Bea02}). The latter studies showed that the flow appears to have an onion-like kinematic structure, with the faster and more collimated flow continuously bracketed in a wider and slower one.  The flow becomes gradually denser and more excited from the edges toward the axis. The mass loss rate in the flow is about one tenth of the estimated mass accretion rate through the disk (Bacciotti et al.~\\cite{Bea00}). Even more interestingly, for the spatially resolved flow component at moderate velocity (peaked at $-$70\\,\\kms) systematic offsets in the radial velocity of the lines have been found in pairs of slits symmetrically located with respect to the jet axis (Bacciotti et al.~\\cite{Bea02}). If these results are interpreted as rotation, then the jet is rotating clockwise (looking toward the source) with average toroidal velocities of about 10--15\\,\\kms, in the region probed by the observations (i.e. 10-50~AU from both the star and jet axis). All of these properties, including the implied velocities and angular momentum fluxes are in the range predicted by the models, assuming a central star mass of 0.67~M$_\\odot$ (Hartigan et al.~\\cite{Hea95}). The kinematic properties of the material surrounding the star are of crucial importance in further establishing if the models apply.  DG\\,Tauri is known to be surrounded by a circumstellar disk (Beckwith et al.~\\cite{Bea90}; Kitamura et al.~\\cite{Kea96a}; Dutrey et al.~\\cite{Dea96}). Previous interferometric observations of the molecular component of the system (Sargent \\& Beckwith~\\cite{SB94}; Kitamura et al.~\\cite{Kea96b}, hereafter KKS) could not identify a clear signature for rotation around the central object. These relatively low-resolution (4--5\\arcsec), $^{13}$CO(1--0) observations could not disentangle the kinematics of the circumstellar disk from the outflow and the outer envelope velocity fields. In fact, the environment of the star on large scales appears to be dominated by outflow motions, possibly due to the interaction between the outer regions of the disk and the stellar wind (KKS). In contrast, the inner portion of the disk, closer to the jet launching region, is expected to display a Keplerian rotation pattern. We have carried out new, higher-resolution millimeter wavelength observations of the \\cii\\ transition toward the DG~Tauri system with the aim of distinguishing the velocity field close to the star and ascertaining if it is consistent with that expected for Keplerian rotation and to check that the disk and jet rotate in the same sense.  \\section {Observations and results}   Millimeter wavelength interferometric observations of the DG~Tauri system was performed using the Owens Valley Radio Observatory (OVRO) mm-array located near Big Pine, California, between Oct~1999 and Dec~2001. The six 10.4 meter dishes were deployed in configurations that provided baselines from 15 to 240~m. Continuum observations centered at $\\sim$220 and $\\sim$108~GHz used an analog correlator with a total bandwidth of 2~GHz. The digital correlator was configured to observe the \\cii\\ transition with 0.125~MHz resolution over an 8~MHz band (0.17 and 11~\\kms, respectively). Frequent observations of 0528$+$134 were used to perform phase and gain calibration. The passband calibration was obtained by observing 3C273, 3C454.3 and/or 3C84. The flux density scale was derived by observing Neptune and/or Uranus, and the calibration uncertainty is expected to be $\\sim 20\\%$. All calibration and data editing used the MMA software package (Scoville et al.~\\cite{mma}). Calibrated ($u,v$) data were then loaded into the AIPS and/or GILDAS packages for imaging, deconvolution and analysis. Continuum maps and line cubes were produced using natural weighting of the ($u,v$) data, and smoothed to a spectral resolution of 0.5~\\kms, unless specifically noted. The synthesized beam full width at half maximum is 1$\\farcs$7$\\times 1\\farcs4$. Continuum subtraction was performed on the dirty images before deconvolution using channels at the edge of the band. %   \\subsection{Continuum maps}   We detect unresolved continuum emission from DG~Tauri at both 1.3 and 2.7~mm. The peak position is the same at both wavelengths, $\\alpha$(2000)$=$04$^{\\rm h}$27$^{\\rm m}$04.$\\!\\!^{\\rm s}$66 $\\delta$(2000)$=$26$^\\circ$06$^\\prime$16$\\farcs$3, in agreement with previous measurements (e.g. Kitamura et al.~\\cite{Kea96a}; KKS). The total flux density is 215~mJy at 222~GHz and 55~mJy at 108~GHz. Within calibration uncertainties, the 3 mm value agrees with earlier interferometer measurements (KKS; Dutrey et al.~\\cite{Dea96}; Looney et al.~\\cite{Lea00}) but the 1.3~mm value is a factor of two lower than the single dish flux (Beckwith et al.~\\cite{Bea90}), probably because of spatial filtering by the interferometer. If we assume optically thin emission from dust grains at T$\\sim$40~K and a dust opacity coefficient k$_\\nu$=k$_{\\rm 230{\\rm GHz}}\\times(\\nu/230~{\\rm GHz})^\\beta$, with k$_{\\rm 230{\\rm GHz}}$=0.01 (Hildebrand~\\cite{H83}, including a gas to dust ratio of 100 by mass), and $\\beta\\sim$0.5 (Beckwith \\& Sargent~\\cite{BS91}), our measurements imply a total mass of $\\sim$0.04~M$_\\odot$, consistent with previous estimates.%  \\subsection{Line maps}  \\begin{figure} \\centerline{\\psfig{figure=fcore.eps,angle=-90,width=7.2cm}} \\caption[]{OVRO \\cii\\ integrated intensity map (contours) overlaid on the 2.7~mm continuum image (greyscale).} \\label{fcore} \\end{figure}  \\begin{figure*} \\centerline{\\psfig{figure=fchan.eps,angle=-90,width=17.2cm}} \\caption[]{\\cii\\ channel maps. Top panels: blue wing; bottom panels: red wing. Each panel is labelled with the appropriate velocity (v$_{LSR}$ in \\kms). The cross marks the position of the 1.3~mm continuum peak. Contour levels start at 3$\\sigma$ and are spaced by 1$\\sigma=60$~mJy/beam. The central velocity of the system is assumed to be $\\sim$6.0~\\kms\\ (rightmost, isolated panel). } \\label{fchan} \\end{figure*}  In Figure~\\ref{fcore} we show the \\cii\\ integrated intensity map overlaid on the 2.7~mm continuum image. Our observations are sensitive only to the compact features of the emission and resolve out most of the extended core emission seen in the KKS maps. In spite of the filtering by the interferometer, most of the \\cii\\ emission is not concentrated in the inner regions of the system close to the position of the optical star. The velocity pattern exhibited by this extended gaseous component has been interpreted by KKS as due to an expanding disk-like structure, possibly the outer edge of the disk that is being dispersed by the stellar wind. Our higher angular resolution map is in good agreement with their interpretation. In this paper however, % we will focus on the inner regions of the disk.  In Figure~\\ref{fchan} we show \\cii\\ channel maps (0.5\\,\\kms\\ resolution) of the central $\\sim$12$^{\\prime\\prime}$ region centered on the continuum peak position (marked by a cross). The core of the line (v$_{LSR}$=5.0-7.0~\\kms) is dominated by the poorly imaged extended structure discussed above. Note that the central velocity channels may also be affected by self-absorption due to cold foreground gas (Schuster et al.~\\cite{Sea93}). By contrast, the higher velocity wings, corresponding to the leftmost three blue and red channels, display compact emission arising from the inner disk. The maps show that the emission peak in the blue channels is shifted toward the south-east with respect to the continuum peak, while in the red channels it is shifted to the north-west. In order to emphasize this velocity gradient, in Fig.~\\ref{fwings} we show the wing emission integrated over the ranges v$_{LSR}$=3.5--4.5~\\kms\\ (blue wing) and 7.5--8.5~\\kms\\ (red wing). These maps were obtained with a robust weighting of the ($u,v$) data and the resulting angular resolution is 1$\\farcs4\\times$1$\\farcs$0. The red and blue wing emission peaks on opposite sides of the continuum, which is assumed to trace the stellar position (see also KKS), and is aligned along a line approximately perpendicular to the observed direction of the optical jet (p.a.$\\sim$226$^\\circ$, marked with a thick line in Fig.~\\ref{fwings}). ",
        "conclusions": "The prime goal of this study was to investigate the velocity pattern in the inner regions of the DG~Tauri disk and to relate it to the velocity pattern detected at the base of the optically visible jet by Bacciotti et al.~(\\cite{Bea02}). The channel and wing maps of Figs.~\\ref{fchan} and~\\ref{fwings} indeed show a velocity gradient across the inner regions of the circumstellar disk. If interpreted as rotation within a disk the axis of which coincides with the jet axis, the direction of the gradient is consistent with the sense of rotation inferred for the jet.  \\begin{figure} \\centerline{\\psfig{figure=fwings.eps,angle=-90,width=7.4cm}} \\caption[]{\\cii\\ wing maps (contours) overlaid on the 1.3~mm continuum image (greyscale). Blue and red wing are shown as solid and dashed contours, respectively. Contour levels start at 3$\\sigma$ and are spaced by 1$\\sigma$, dotted lines show negative contours. The thick solid line indicates the direction and extent of the initial segment of the optical jet studied by Bacciotti et al.~(\\cite{Bea02}).} \\label{fwings} \\end{figure}  Due to the contamination of the poorly imaged external regions of the disk, and self-absorption, it is not possible to study the kinematics of the inner disk using the line core within 1~\\kms\\ from the systemic velocity, assumed to be 5.8~\\kms\\ (KKS). In particular, there can be no detailed comparison of the observed velocity patterns with Keplerian rotation models such as those undertaken by Koerner et al.~(\\cite{Kea93}), Guilloteau \\& Dutrey~(\\cite{GD94}) and Mannings et al.~(\\cite{MKS97}). Nevertheless, in Fig.~\\ref{fpv} position-velocity diagrams along directions parallel and perpendicular to the jet suggest that, along the disk major axis ($\\Delta_\\perp$), higher absolute velocity (with respect to the systemic velocity) peaks are located closer to the star, while the lower absolute velocities peak further away. Moreover, blue velocities systematically peak to the south-east and red velocities peak to the north-west of the stellar position. This behaviour is qualitatively consistent with Keplerian rotation in the inner regions of the disk. A more quantitative comparison with the expected line-of-sight velocities for a disk surrounding DG~Tauri is shown in Fig.~\\ref{fpv}, bottom panel. The theoretical curves in the figure were computed for a central star with mass M$_\\star$=0.67~M$_\\odot$ (Hartigan et al.~\\cite{Hea95}), and an inclination from the line of sight of 38$^\\circ$ (Eisl\\\"offel \\& Mundt~\\cite{EM98}); these are the parameters adopted by Bacciotti et al.~(\\cite{Bea02}) to check the rotational hypothesis for the jet. The dotted lines include an uncertainty in these parameters of $\\pm$0.25~M$_\\odot$ and $\\pm$15$^\\circ$. The observed velocity pattern in the inner regions of the disk is in excellent agreement with the model predictions. A more detailed comparison, including the complete derivation of the disk rotation from molecular line observations, will require higher angular resolution and more sensitive observations of optically thinner transitions, such as \\ciii, which are possibly less affected by the external regions of the disk/envelope.  \\begin{figure} \\centerline{\\psfig{figure=fpv_3.eps,angle=-90,width=7.0cm}} \\caption[]{Top panels: \\cii\\ position-velocity diagrams parallel (top) and perpendicular (bottom) to the jet axis. The crosses mark the intensity averaged position at each velocity, computed only if the emission is above 3$\\sigma$. Errorbars in the top panel show the velocity and spatial resolutions. The thin lines mark the core average velocity (5.8~\\kms) and the position of the continuum source. Bottom panel: intensity averaged line of sight velocities in the first and third quadrant of the lower position-velocity diagram, averaged every 0$\\farcs$75 (105~AU). The dashed line marks the expected line of sight velocity across a Keplerian disk with the axis inclined by 38$^\\circ$ from the line of sight and orbiting about a 0.67~M$_\\odot$ star at a distance of 140~pc from the Sun. The dotted lines show the range of variation of the Keplerian disk predictions for a maximum uncertainty of $\\pm$15$^\\circ$ in the inclination angle and $\\pm$0.25~M$_\\odot$ in the stellar mass. } \\label{fpv} \\end{figure}  Summarizing our results, we have shown for the first time that the disk kinematics in a young T\\,Tauri system are in qualitative and quantitative agreement with the velocity pattern at the base of the jet. In other words, the simultaneous and kinematically consistent rotation of disk and jet postulated by the popular magneto-centrifugal models has been observationally inferred for the first time for the region within $\\sim$200~AU from the central source."
    },
    "0209/astro-ph0209378_arXiv.txt": {
        "abstract": "\\noindent  The study of photoionized environments is fundamental to many astrophysical problems. Up to the present most photoionization codes have numerically solved the equations of radiative transfer by making the extreme simplifying assumption of spherical symmetry. Unfortunately very few real astronomical nebulae satisfy this requirement. To remedy these shortcomings, a self-consistent, three-dimensional radiative transfer code has been developed using Monte Carlo techniques. The code, Mocassin, is designed to build realistic models of photoionized nebulae having arbitrary geometry and density distributions, with both the stellar and diffuse radiation fields treated self-consistently. In addition, the code is capable of treating ones or more exciting stars located at non-central locations.  The gaseous region is approximated by a cuboidal Cartesian grid composed of numerous cells. The physical conditions within each grid cell are determined by solving the thermal equilibrium and ionization balance equations. This requires a knowledge of the local primary and secondary radiation fields, which are calculated self-consistently by locally simulating the individual processes of ionization and recombination. The structure and the computational methods used in the Mocassin code are described in this paper.  Mocassin has been benchmarked against established one-dimensional spherically symmetric codes for a number of standard cases, as defined by the Lexington/Meudon photoionization workshops \\citep{pequignot86,ferland95, pequignot01}. The results obtained for the benchmark cases are satisfactory and are presented in this paper. A performance analysis has also been carried out and is discussed here. ",
        "introduction": "Amongst the first numerical models for photoionized gaseous nebulae were those calculated by \\citet{flower68}, \\citet{harrington68} and \\citet{rubin68}. These early models included the basic physical processes of ionization and recombination of hydrogen and helium, thermal balance and escape of the emitted photons from the nebula. However, the lack of reliable atomic data heavily limited the success of these models, as well as the fact that a number of important physical processes, such as charge exchange and dielectronic recombination \\citep{aldrovandi73, pequignot78, storey81}, were not accounted for at the time. The evolution of photoionization modelling has gone hand in hand with advances made in atomic physics and computer technology. The application of photoionization models to a wider range of ions has been aided by the photoionization cross-section calculations by \\citet{reilman79}, and, more recently, the Opacity Project \\citep{hummer93}. Compilations based on the latter's data \\citep[e.g.][]{verner95}, have made possible the inclusion of accurate photoionization cross-sections for many more ions in calculations. \\citet{mendoza82} presented a compilation of radiative and collisional data for collisionally excited ultraviolet, optical and infrared lines which was widely adopted, with some of these data still in use today, though most have been seperceded by more recent calculations such as the R-matrix calculations of the Iron Project \\citep{hummer93} and the Belfast group \\citep[e.g.][]{mclaughlin98, ramsbottom98}.  Currently, radiative and dielectronic recombination rates are still highly uncertain or unavailable for some ions; recent efforts to improve the situation have been reviewed by \\citet{nahar99} and \\citet{nahar00}. Most photoionization models include temperature-dependent analytical fits to these recombination rates, such as those of \\citet{aldrovandi73} for radiative and high temperature dielectronic recombination, and those of \\citet{nussbaumer83} for low temperature dielectronic recombination.  Available computer power has increased enormously since the dawn of photoionization modelling. This has allowed more complex models to be built, including more ions,  more frequency points, more lines and more atomic levels. Nevertheless, the fundamental assumption of spherical symmetry has always been retained. However, a glance at an image of any Galactic H~{\\sc ii} region will immediately demonstrate that these objects are neither spherically symmetric nor homogeneous. In addition, they usually have multiple exciting stars located at non-central positions in the nebula. By contrast, planetary nebulae (PNe) have only a single, centrally located, exciting star. However, even for PNe, spherical symmetry is not a realistic assumption, as demonstrated by observations with instruments such as the Hubble Space Telescope, which reveal an overwhelming variety in the shapes of planetary nebulae. These objects are very rarely circular in projection; a recent study inferred that about 50\\% of all known planetary nebulae are low eccentricity ellipticals, while only about 10\\% are circular in projection, with the remainder having more extreme elliptical or bipolar geometries \\citep{soker97, soker01}. Some objects, for example the two young planetary nebulae He~2-47 and PN~M1-37, \\citep[also dubbed the {\\it starfish twins;}][]{sahai00}, show even more complicated geometries, with multiple  lobes. Other PNe have FLIERs \\citep[fast, low ionization emitting regions;][]{balick93, balick94, balick98}, BRETS \\citep[bipolar, rotating, episodic jets; e.g.][]{lopez93}, ansae, jets, knots, filaments, tails or multiple envelopes. \\citep[see e.g.][]{perinotto00, corradi99, garcia97}.  To our knowledge, only two three-dimensional photoionization codes have been develped so far, one by \\citet{bassgen90} and the other by Gruenwald, Viegas \\& de Broguiere (1997). The first code used a fixed number of equally sized cells and the on-the-spot approximation for the diffuse radiation field, with only the six more abundant chemical elements being taken into account. The work by \\citet{gruenwald97} improves on this by allowing a more flexible spatial grid and by using an iterative technique for the determination of the diffuse field and also by including twelve chemical elements in the simulations.  Since most existing one-dimensional photoionization codes are  based on the numerical solution of the equations of radiative transfer assuming spherical symmetry, their expansion to three dimensions can be either very difficult or impractical, resulting in very large codes. The Monte Carlo approach to transfer problems provides a geometry-independent technique which can handle the radiation transport problem self-consistently. With this in mind, the Mocassin code (MOnte CArlo SimulationS of Ionised Nebulae) was developed, in order to provide a three-dimensional modelling tool capable of dealing with asymmetric and/or inhomogeneous nebulae, as well as, if required, multiple, non-centrally located exciting stars.  Section~\\ref{sec:description} contains a description of the general Mocassin architecture and of some of the main computational methods used in the code. The code has been benchmarked against established spherically symmetric one-dimensional photoionization codes for a set of standard nebulae and in Section~\\ref{sec:benchmarks} we present the results of this benchmarking, together with a performance analysis of the codes. In section~\\ref{sec:discussion} we discuss the results of the benchmarking and present some general guidelines on how to run the code efficiently. ",
        "conclusions": "A fully three-dimensional photoionization code, Mocassin, has been developed for the modelling of photoionised nebulae, using Monte Carlo techniques. The stellar and diffuse radiation fields are treated self-consistently; moreover, Mocassin is completely independent of the assumed nebular geometry and is therefore ideal for the study of aspherical and/or inhomogeneous nebulae, or nebulae having one or more exciting stars at non-central locations.  The code has been successfully benchmarked against established one-dimensional photoinization codes for standard spherically symmetric model nebulae \\citep[see][]{pequignot86, ferland95, pequignot01}.  Mocassin is now ready for the application to real astronomical nebulae and it should provide an important tool for the construction of realistic nebular models. A companion paper \\citep[][Paper~{\\sc ii}]{ercolanoII} will present detailed results from the modelling of the non-spherically symmetric PN~NGC~3918. Resources permitting, it is intended to make the Mocassin source code publicly available in the near future.  \\vspace{7mm}  \\noindent {\\bf Acknowledgments} This work was carried out on the Miracle Supercomputer, at the HiPerSPACE Computing Centre, UCL, which is funded by the U.K. Particle Physics and Astronomy Research Council. We thank the anonymous referee for useful comments. BE aknowledges support from PPARC Grant PPA/G/S/1997/00728 and the award of a University of London Jubber Studentship. We thank Dr M. Rosa for making available to us a copy of the photoionization code described by Och, Lucy and Rosa (1998)."
    },
    "0209/astro-ph0209323_arXiv.txt": {
        "abstract": "{ In this paper the two-dimensional structure of protoplanetary disks around Herbig Ae/Be stars is studied. This is done by constructing a self-consistent model based on 2-D radiative transfer coupled to the equation of vertical hydrostatics. As a simplifying assumption a grey opacity is used. It is found that the disk can adopt four different structures, dependent on the surface density distribution $\\Sigma(R)$ as a function of radius, i.e.~on radial- and vertical optical depth of the disk. For the case of high to intermediate vertical optical depth, the temperature and density structures are in agreement with the simple ``disk with inner hole'' model of Dullemond, Dominik \\& Natta (2001, henceforth DDN01). At large radii the disk adopts a flaring shape as expected, and near the dust destruction radius (located at about $0.5\\AU$ for most Herbig Ae stars) the disk is superheated and puffs up. The region directly behind this ``puffed-up inner dust wall'' is shadowed, as predicted by DDN01. For the case of intermediate to low vertical optical depth, but still high radial optical depth, the 2-D models show that the shadow can cover the entire disk. For such competely self-shadowed disks the inner rim emission in the near infrared constitutes the dominant part of the SED, since the flaring component in the mid- and far infrared is suppressed by the self-shadowing effect. When the disk is optically thin even in radial direction, it becomes unshadowed again because the inner rim can no longer block the stellar light. Such disks have relatively weak infrared excess compared to the stellar flux. Finally, for disks that flare at intermediate radii, but become too optically thin at large radii, the outer parts once again become shadowed. But this time the shadowing is caused by the flaring part of the disk, instead of the inner rim. The disk then consists of a bright inner rim, a shadow, a flaring part and finally a (dim) shadowed outer part. Different observational methods of determining the size of the disk (e.g.~from the SED, from continuum mapping or from CO mapping) may yield different results. } ",
        "introduction": "Herbig Ae/Be stars are widely regarded as the intermediate mass counterparts of the lower mass T Tauri stars (e.g.~Strom et al.~\\citeyear{stromstrom:1972}; Palla \\& Stahler \\citeyear{pallastahler:1993}; v.d.~Ancker et al.~\\citeyear{anckerwinter:1998}). But, while there is little doubt that the infrared excess of classical T Tauri stars originates in a protoplanetary disk, the case for Herbig Ae/Be stars is not so clear. On large scales the disk-like nature of the circumstellar matter around Herbig Ae/Be stars is by now well established. Millimeter CO and continuum maps (Mannings \\& Sargent \\citeyear{mannsarg:1997}), submm position-velocity maps (Qi \\citeyear{qithesis:2001}) and imaging at visible and near-infrared wavelengths (Grady et al.~\\citeyear{gradywoodbruh:1999}; Augereau \\citeyear{augereau:2001}) clearly show rotating flattened structures at scales of hundreds of AU. But on scales of tens of AU and smaller there seems to be conflicting evidence, with some authors arguing for a more or less spherical geometry (di Francesco et al.~\\citeyear{difrancescoevans:1994}; Pezzuto et al.~\\citeyear{pezzutostraf:1997}; Miroshnichenko et al.~\\citeyear{miroiveeli:1997},\\citeyear{miroivevinkeli:1999}; Millan-Gabet et al.~\\citeyear{millanschl:2001}) and others favoring the disk-like picture (Vink et al.~\\citeyear{vinkdrew:2003}; Grinin \\& Rostopchina \\citeyear{grininrostop:1996}; Natta et al.~\\citeyear{nattaprusti:2001}; Tuthill et al.~\\citeyear{tutmondan:2001}).  Unfortunately, for some time disk models have failed to explain the SEDs of Herbig Ae/Be stars, and were consequently rejected by many. In particular the conspicuous bump around 3 microns in the SEDs of Herbig Ae/Be stars remained a mystery. Standard disk models such as those of Chiang \\& Goldreich (\\citeyear{chianggold:1997}, henceforth CG97), and D'Alessio et al.~(\\citeyear{dalessiocanto:1998},\\citeyear{dalessiocalvet:1999}) failed to explain this striking feature. Several explanations were suggested, ranging from FeO dust at 800 K (v.d.~Ancker et al.~\\citeyear{vdanckerbouw:2000}) to accretion disks that are actively dissipating only beyond a certain radius (Hillenbrand et al.~\\citeyear{hillenstrom:1992}). None of these explanations were quite satisfactory. Recently, it was recognized by Natta et al.~(\\citeyear{nattaprusti:2001}) that this 3 micron bump may well originate from the inner rim of the dusty part of the disk. Since dust evaporates above about 1500 K, the inner parts of the disk are free of dust. These gaseous inner parts have a much lower optical depth, and can even be entirely optically thin, dependent on the gas surface density. At the dust evaporation radius the dust forms a wall of 1500 K that is directly irradiated by the central star. This produces an extra component in the spectrum that was not considered by the existing models of flaring disks.  Dullemond, Dominik \\& Natta (\\citeyear{duldomnat:2001}, henceforth DDN01) adapted the CG97 model to self-consistently include this inner rim, and showed that the SEDs of Herbig Ae/Be stars can be naturally explained in this way. In a recent paper (Dominik et al.~\\citeyear{domdulwatwal:2002}) the sample of 14 Herbig Ae/Be stars of Meeus et al.~(\\citeyear{meeuswatersbouw:2001}) was analyzed in the context of this model, and it was found that the SEDs of most stars were indeed consistent with the DDN01 picture.  According to this model, the inner rim of the flaring disk is much hotter than an ordinary flaring disk model would predict at that radius.  This is because the inner rim is irradiated frontally rather than at a grazing angle. As a consequence, the inner rim is puffed up and casts a shadow over part of the flaring disk behind it. This shadow can extend from the inner rim, at about 0.5 AU, out to 5 AU or more. Outward of this shadowing radius the disk adopts the usual flaring shape as described by CG97. This part of the disk is responsible for the observed emission at long wavelengths. Dependent on the height of the inner rim, the shadow can reach so far out that the 10 $\\mu$m silicate emission feature, produced by warm dust in the surface layers, is suppressed. This has been used by DDN01 as a possible explanation for the lack of 10 micron feature in several sources.  Though succesful in explaining several features of the SEDs of Herbig Ae/Be stars, the DDN01 model was based on highly simplified equations. Among other things the structure of the inner rim and the shadowing of the disk behind it need closer theoretical examination. It is unclear from the DDN01 model what happens when the optical depth of the disk becomes too low to sustain flaring. Also, the DDN01 model was based on the assumption that, if the disk {\\em can} flare outside the shadow, then it {\\em will}. It is unclear whether perhaps in addition to these flaring disks also fully self-shadowed disk solutions exist.  Because of the intrinsic 2-D nature of the problem, a closer theoretical study requires a full 2-D treatment of radiative transfer. This is done with a 2-D ``Variable Eddington Tensor'' solver. By coupling the radiative transfer to the equations of vertical hydrostatic equilibrium, the code solves the entire temperature and density structure of the disk as a function of radius and vertical height above the midplane. As a simplifying assumption a grey opacity is adopted in this work. This is consistent with the disk consisting of large grains. The advantage of this simplification is that the results are more readily understood in terms of simple radiative transfer arguments. In a follow-up paper more realistic opacities and grain size distributions will be used, which will put us in a position to compare the results directly to observations.  Using the 2-D disk structure code just described, the following issues will be addressed: \\begin{enumerate} \\item Does the overall structure and SED of the disk agree with the much simpler model of DDN01? Will there indeed be a shadowed region behind the inner rim, as predicted by DDN01? How much will radial radiative diffusion heat this shadowed region, and how large is this region in radial direction? \\item Under which conditions will the outer part of the disk collapse into the shadow of the inner rim, making the disk entirely self-shadowed? Can there be a bimodal set of solutions (self-shadowed and flared) for the same parameters? \\item What is the structure of a fully self-shadowed disk? Will it be very cool and collapsed, or does radial radiative diffusion keep the disk still relatively warm? \\item What will the outer (tenuous) part of a large disk look like: will it continue to be flaring, or will it sink into the shadow of the flaring part of the disk? \\end{enumerate}  The paper is organized as follows. In Sec.~\\ref{sec-model-equations} the equations are presented, and it is described how they are solved. In Sec.~\\ref{sec-model-haedisk} a model of a canonnical flaring disk around a Herbig Ae star is described, and in Sec.~\\ref{sec-lowtau-selfshad} it is shown that self-shadowed disk can exist and what their structure looks like. In Sec.~\\ref{sec-bimodal} it is investigated if multiple solutions exist for the same parameters.  A description of the tenuous outer parts of a flaring disk is given in Sec.~\\ref{sec-outer-parts}. ",
        "conclusions": "\\label{sec-disc-concl}  \\begin{figure*} \\mbox{}\\vspace{1em}\\\\ \\parbox[t]{8.2cm}{ \\centerline{A: Flared disk:} \\centerline{\\includegraphics[width=7.5cm]{f12a.eps}} } \\parbox[t]{8.2cm}{ \\centerline{B: Flared disk with shadowed outer region:} \\centerline{\\includegraphics[width=7.5cm]{f12b.eps}} }\\\\ \\parbox[t]{8.2cm}{ \\centerline{C: Self-shadowed disk:} \\centerline{\\parbox[b]{7.5cm}{\\mbox{}\\vspace{1.0cm}\\\\ \\includegraphics[width=7.5cm]{f12c.eps} \\\\\\vspace{1.2cm}\\mbox{}}} } \\parbox[t]{8.2cm}{ \\centerline{D: Transparent disk:} \\centerline{\\parbox[b]{7.5cm}{\\mbox{}\\vspace{0.6cm}\\\\ \\includegraphics[width=7.5cm]{f12d.eps} \\\\\\vspace{0.6cm}\\mbox{}}} } \\caption{Pictograms showing the four main kinds of solutions found.  They represent a vertical cross-section of the disk, but are not to scale. A hashed area represents a shadow. In cases A, B and C the vertical optical depth may, under some conditions, drop below zero even though the radial optical depth remains much larger than unity. Case B is ``zoomed out'' to indicate that the second shadowing happens at large radii. In case D, the empty polygon is meant to show that even the radial optical depth along the midplane is small, meaning that the temperature everywhere is set by the optically thin dust temperature. \\revised{The cases A, B, C and D are ordered according to decreasing optical depth (i.e.~mass of the disk). It should be noted, however, that such an ordering does not always strictly apply since the powelaw index $p$ of the surface density distribution also plays a role in determining the disk shape.}} \\label{fig-pictograms} \\end{figure*}  In this paper the structure of passive circumstellar disks was theoretically investigated. The problem was defined in a mathematically ``clean'' way: it is the problem of computing the temperature and density structure of rotating circumstellar matter around a star with a certain mass, radius and luminosity from basic principles of radiative transfer, radiative equilibrium and vertical hydrostatic equilibrium. The disk parameters that went into the calculation were the inner and outer radius, and the surface density distribution as a function of radius. The only mathematical approximation made here was the reduction of the hydrostatic equilibrium equations to 1-D vertical equations. From a physical point of view, many more approximations were made (related to dust-gas coupling, active accretion, dust opacities, etc). But these were necessary to keep the problem clear of uncertain physics for now.  Four different kinds of solutions were found: a flaring disk, a self-shadowed disk, a transparent disk, and a flaring disk with self-shadowed outer region. These solutions are pictographically listed in Fig.~\\ref{fig-pictograms}. The numerical models described in this paper can be downloaded from a website: {\\tt www.mpa-garching.mpg.de/PUBLICATIONS/DATA/ radtrans/grey2d/}.  The main conclusions are summarized as follows: \\begin{enumerate} \\item A flaring disk around a Herbig Ae/Be star has a hot inner rim, a shadowed region behind it, and the usual flared geometry at large radii (Fig.~\\ref{fig-pictograms}-A). These findings are in accordance with the predictions of Dullemond, Dominik \\& Natta (\\citeyear{duldomnat:2001}). But the effect of shadowing in suppressing the emission from the shadowed region is not as strong as was predicted in that paper. \\item Disks with intermediate to low vertical optical depth but high equatorial optical depth can become entirely self-shadowed (Fig.~\\ref{fig-pictograms}-C). The SED falls off more steeply at long wavelength than for flaring disks. \\item Disks with equatorial optical depths that are smaller than unity are un-shadowed again, since the inner rim can no longer stop the stellar radiation. These disks are fully optically thin (Fig.~\\ref{fig-pictograms}-D). \\item The outer regions of flared disks can become shadowed if beyond a certain radius the surface density becomes too low. This time it is the flared part of the disk that casts the shadow (Fig.~\\ref{fig-pictograms}-B). These outer parts do not contribute much to the SED, but may still be detectable using (sub-)millimeter interferometers. Measurements of the outer radius of a disk may therefore yield different results, depending on whether one uses SED-fitting, continuum mapping or CO mapping. \\item In the case of self-shadowed disks, and the shadowed outer parts of flaring disks, the usual 1+1-D approach to disk modeling breaks down, and so does the approach used by CG97 and DDN01. A full 2-D approach, such as the one used in this paper, is then necessary. It might be that the SED is still relatively well described using a 1+1-D model or a DDN01-type model up to the self-shadowing radius, but this remains to be proven using a 2-D model with more realistic opacities. \\item Bimodel solutions were not found. It seems that protoplanetary disks obey a kind of ``flaring disk principle'': If the disk {\\em can} flare, it {\\em will}. Self-shadowed disks are therefore disks which cannot be made to flare. \\end{enumerate}  Many of these conclusions will presumably still hold when more realistic opacities are included. But this will be the topic of the second paper in this series."
    },
    "0209/astro-ph0209115_arXiv.txt": {
        "abstract": "To characterise the cosmological evolution of the sources contributing to the infrared extragalactic background, we have developped a phenomenological model that constrains in a simple way the galaxy luminosity function evolution with the redshift, and fits all the existing source counts and redshift distributions, Cosmic Infrared Background intensity and fluctuations observations, from the mid-infrared to the submillimetre range. The model is based on template spectra of starburst and normal galaxies, and on the local infrared luminosity function. Although the Cosmic Infrared Background can be modeled with very different luminosity functions as long as the radiation production with redshift is the right one, the number counts, and the anisotropies of the unresolved background, imply that the luminosity function must change dramatically with redshift, with a rapid evolution of the high-luminosity sources (L$>$3 10$^{11}$ L$_{\\odot})$ from z=0 to z=1 which then stay rather constant up to redshift 5. The derived evolution of the IR luminosity function may be linked to a bimodal star formation process, one associated with the quiescent and passive phase of the galaxy evolution and one associated with the starburst phase, trigerred by merging and interactions. The latter dominates the infrared and submillimetre ouput energy of the Universe. \\\\ The model is intended as a convenient tool to plan further observations, as illustrated through predictions for {\\it Herschel}, {\\it Planck} and {\\it ALMA} observations. Our model predictions for given wavelengths, together with some usefull routines, are available for general use. ",
        "introduction": "The discovery of the Cosmic Infrared Background (CIB) (Puget et al. 1996; Fixsen et al. 1998; Hauser et al. 1998; Schlegel et al. 1998; Lagache et al. 1999; Lagache et al. 2000; see Hauser \\& Dwek 2001 for a review), together with recent deep cosmological surveys in the infrared (IR)  and submillimetre (submm) has opened new perspectives on our galaxy formation and evolution understanding. The surprisingly high amount of energy contained in the CIB showed that it is crucial to probe its contributing galaxies to understand when and how the bulk of stars formed in the Universe. Thanks to ISO (Kessler et al., 1996) -- mainly at 15~$\\mu$m with {\\it ISOCAM} (Cesarsky et al. 1996), and 90 and 170~$\\mu$m with {\\it ISOPHOT} (Lemke et al. 1996) -- and ground-based instruments -- {\\it SCUBA} (Holland et al. 1998) and {\\it MAMBO} (Bertoldi et al. 2000) at 850 and 1300 $\\mu$m respectively -- deep cosmological surveys have been carried out. It is thus now possible, to various degrees, to resolve the CIB into discrete sources (e.g.  Kawara et al. 1998; Barger et al. 1999; Elbaz et al. 1999; Carilli et al. 2000; Juvela et al. 2000; Linden-Vornle et al. 2000; Matsuhara et al. 2000; Bertoldi et al. 2001; Dole et al. 2001; Elbaz et al. 2002; Scott et al. 2002). The striking result of these surveys concerns the evolution of the IR and submm galaxy population. The source counts are high when compared to no, or moderate, evolution models{\\footnote{`No-evolution': the co-moving luminosity function remains equal to the local one at all redshift}} for IR galaxies (Guiderdoni et al. 1998; Franceschini et al. 1998). Therefore, it has been necessary to develop new models in the IR. Very recently, several empirical approaches have been proposed to model the high rate of evolution of IR galaxies (e.g  Devriendt \\& Guiderdoni 2000; Wang \\& Biermann 2000; Charry \\& Elbaz 2001; Franceschini et al. 2001; Malkan \\& Stecker 2001; Pearson 2001; Rowan-Robinson 2001; Takeuchi et al. 2001; Xu et al. 2001; Balland et al. 2002; Wang 2002) which fit all existing source counts, redshift distribution and CIB intensity and fluctuations, although often not all of them. We present in this paper a new model whose preliminary results were published by Dole et al. (2000). The originality of this model was to empirically separate the evolution of the starburst galaxies with respect to the normal galaxies, the current observations implying a strong evolution of the bright part of the Luminosity Function (LF). It was shown for the first time that only the starburst part should evolve very rapidly between z=0 and z=2, the evolution rate being much higher in the IR than in any other wavelength domain. \\\\  We present here a more sophisticated and detailed version of the first model (Dole et al. 2000). Our philosophy is to build the simplest model of the LF evolution, easily deliverable, with the lowest number of parametres but accounting for all observationnal data. We stress out the point that we include the CIB fluctuations levels as measured by {\\it ISOPHOT} (Lagache \\& Puget 2000; Matsuhara et al. 2000) and {\\it IRAS} (Miville-Desch\\^enes et al. 2002) as an extra constraint. Recent observations strongly suggest that the bulk of the optical and IR extragalactic background is made of two distinct galaxy populations (see Sect. \\ref{Nature}). Therefore, we restrict our model in the wavelength domain 10 - 1500~$\\mu$m, our goal being to quantify the evolution of IR galaxies. The paper is organised as follows: we first summarise our present knowledge on the evolution of IR galaxies, and on the nature of the sources contributing to the Extragalactic Background (Sect. 2). Then, we present the ingredients of the model (Sect. 3). In Sect. 4, we discuss the galaxy templates used in the model. We then present the parametrisation of the local LF (Sect. 5). In Sect. 6 are given the results of the model (evolution of the LF, evolution of the luminosity density, number counts, z-distribution, CIB intensity and fluctuations). And finally the model is used for predictions for the {\\it Herschel} and {\\it Planck} surveys (Sect. 7.2 and 7.3 respectively), gives requirements for future large deep survey experiments (Sect. 7.4) and predictions for {\\it ALMA} observations (Sect. 7.5). A summary is given in Sect. 8. ",
        "conclusions": "\\subsection{LF evolution derived from observations} Although the number of parametres is quite low, it is too much time consuming to do a blind search through the whole parametre space. We therefore search the best solution for the parametres near values expected from direct observationnal evidences: \\begin{itemize} \\item We fix the normal galaxy evolution (passive evolution) such as it nearly follows the number density evolution of optical counts up to z=0.4: \\\\ $\\Phi_{normal}(L,z) = \\Phi_{normal}(L,z=0) \\times (1+z)$. \\\\ We arbitrarily stop the evolution at z=0.4 and keep this population constant up to z=5 and then let the population decreases up to z=8 (see Fig. \\ref{Lum_density}) \\item An estimate of L$_{knee}$ is given by deep surveys at 15 and 850 $\\mu$m: the bulk of the CIB at 15 $\\mu$m and 850 $\\mu$m is made by galaxies with L$\\sim$1-5 10$^{11}$ L$_{\\odot}$ (e.g. Barger et al. 1999; Elbaz et al. 2002) \\item We have indications on the evolution of the luminosity density, $\\varphi(z)$, for the starburst galaxies, directly inferred from the CIB spectrum shape by Gispert et al. (2000) which is used as a starting point to adjust the model. \\item We take SB$_{norm}$=10$^7$ L$_{\\odot}$/Mpc$^3$ which is roughly the value observed locally. \\item We take SB$_{slope}$= -2.2 (Kim \\& Sanders, 1998) \\end{itemize} The best evolution of the LF that reproduces IR number counts, redshift distributions and CIB observations is shown on Fig. \\ref{LF_evol}. It is obtained with: \\begin{itemize} \\item L$_{cutoff}$ = 5 10$^{11}$ L$_{\\odot}$ \\item L$_{knee}$(z) = 8 10$^{10}$ $\\times$ (1+z)$^3$ L$_{\\odot}$ up to z=1.5 and then L$_{knee}$(z)= L$_{knee}$(z=1.5). \\end{itemize} Moreover, to avoid a 'break' in the evolved luminosity function (as in Dole et al. 2000), we modify the low-luminosity part of the starburst LF with the redshift (up to z=5) according to: \\begin{equation} \\Phi_{SB}(L,z) = \\Phi_{SB}(L=L_{max},z) \\times \\left ( \\frac{L_{max}}{L}\\right )^{(1+z)^2} \\end{equation} where $L_{max}$ is the luminosity corresponding to the maximum of $\\Phi_{SB}(L,z)$. \\\\  We see a very high rate of evolution of the starburst part which peaks at z$\\sim$0.7 (Fig. \\ref{Lum_density}) and then remains nearly constant up to z=4 (as shown for example by Charry \\& Elbaz, 2001). We compare on Fig. \\ref{Lum_density} the co-moving luminosity density distribution as derived from the model with the Gispert et al. (2000) one. There is a good overall agreement. However, the model is systematically lower than the Gispert et al. (2000) determination for redshifts between 0.5 and 2. This comes from the fact that the CIB values at 100 and 140~$\\mu$m used in Gispert et al. (2000) were slightly overestimated (Renault et al. 2001) leading in an overestimate of the luminosity density distribution at low redshift.\\\\  \\subsection{Comparison model-observations}  \\subsubsection{The number counts} Fig. \\ref{number_counts} shows the comparison of the number counts at 15~$\\mu$m (Elbaz et al. 1999), 60~$\\mu$m (Hacking \\& Houck 1987; Gregorich et al. 1995; Bertin et al. 1997; Lonsdale et al. 1990; Saunders et al. 1990; Rowan-Robinson et al. 1990), 170 ~$\\mu$m (Dole et al. 2001) and 850~$\\mu$m (Smail et al. 1997; Hughes et al. 1998; Barger et al. 1999; Blain et al. 1999; Borys et al. 2002; Scott et al. 2002; Webb et al. 2002) with the observations. We have a very good overall agreement (we also agree with the 90 $\\mu$m number counts of Serjeant et al. 2001 and Linden-Vornle et al. 2001). \\\\  \\subsubsection{The redshift distributions: Need for a normal 'cold' population}  On Fig. \\ref{z-distrib} is shown the redshift distribution of resolved sources at 15, 60, 170 and 850 $\\mu$m. The 15 $\\mu$m redshift distribution is in very good agreement with that observed by Flores et al. (1999b) and Aussel et al. (1999). At 850 $\\mu$m, we predict that most of the detected sources are at $z>2.5$. At 170 $\\mu$m, we predict that about 62$\\%$ of sources with fluxes $S>$180 mJy (4$\\sigma$ in Dole et al. 2001) are at redshift below 0.25, the rest being mostly at redshift between 0.8 and 1.2. We know from the {\\it FIRBACK} observations that the redshift distribution predicted by the model is very close to that observed at 170 $\\mu$m: it is clear from these observations that we have a bi-modal z-distribution (Sajina et al. 2002; Dennefeld et al. in prep). Moreover, very recently, Kakazu et al. (2002) published first results from optical spectroscopy of 170~$\\mu$m Lockman Hole sources. They find that 62$\\%$ of sources are at z$<$0.3 with IR luminosities (derived using Arp220 SED) lower than 10$^{12}$ L$_{\\odot}$, the rest being at redshift between 0.3 and 1, which is in very good agreement with the model prediction. It is very important to note that the agreement between the model and observations can only be obtained with the local `cold' population. A model containing only starburst-like template spectra and `warm' normal galaxies overpredicts by a large factor the peak at z$\\sim$1 (as in Charry \\& Elbaz 2001 for example).  \\subsubsection{The CIB and its anisotropies}  The predicted CIB intensity at specific wavelengths, together with the comparison with present observations are presented in Table \\ref{CIB-tbl} and in Fig. \\ref{CIB_fig}. We have a very good agreement with the estimates at 60 $\\mu$m (Miville-Desch\\^enes et al. 2002), 100 $\\mu$m (Renault et al. 2001), and 170 $\\mu$m (Kiss et al. 2001) and the {\\it FIRAS} determinations (Fixsen et al. 1998; Lagache et al. 2000). We are also in good agreement with the lower limit derived from 15 $\\mu$m counts (Elbaz et al. 2002), combined with the upper limit deduced from high energy gamma-ray emission of the active galactic nucleus Mkn 501 (Renault et al. 2001). According to the model, sources above 1 mJy at 850 $\\mu$m contribute for about 30$\\%$ of the CIB. At 15 $\\mu$m, with the deepest {\\it ISOCAM} observations (Altieri et al. 1999; Aussel et al. 1999; Metcalfe 2000) about 70 $\\%$ of the CIB has been resolved into individual sources.\\\\  Finally, we compare the model predictions for the CIB fluctuations with the present observations at 60, 90, 100 and 170~$\\mu$m (Table \\ref{Fluc-tbl}). For the comparison we need to remove the contribution of the brightest sources that make the bulk of the fluctuations (sources with fluxes $S>S_{max}$). When S$_{max}$ is quite high (at the order of 500 mJy - 2 Jy), the fluctuations are dominated by the strongest sources, making the result very dependent on the accuracy of the evaluation of S$_{max}$. This is why the comparison observations/model is very difficult{\\footnote{ We cannot compare the model predictions with the Kiss et al. (2001) results since we have no information on S$_{max}$}}. For S$_{max}$ around 50-150 mJy, the fluctuations are dominated by the faint and numerous sources that dominate the CIB and the values do not depend critically on the exact value of S$_{max}$. We see from Table \\ref{Fluc-tbl} that although for some observations the model can be lower or greater by factor 1.5 in amplitude, we have an overall very good agreement. We stress out that reproducing the CIB fluctuations gives strong constraints on the LF evolution. For example, evolution such that we better reproduce the 850 $\\mu$m counts gives too high CIB fluctuations. Future observations with better accuracy will show if these minor discrepancies disappear or are indicative that the bright submm counts are overestimated due to a high fraction of gravitationally lensed sources (Perrotta et al. 2002), or are simply indicative that the present phenomenological model is too simple ! The level of the predicted CIB fluctuations for dedicated experiments (as for example {\\it Boomerang} or {\\it Maxima}), with respect to the cirrus confusion noise and instrumental noise will be discussed in detail in Piat et al. (in prep). \\\\  At 170 $\\mu$m, it is clear from Fig. \\ref{CIB_fluc_170} that the redshift distributions of sources that are making the CIB and those making the bulk of the fluctuations are similar. The fluctuations are not dominated by bright sources just below the detection threshold but by numerous sources at higher redshift. Thus, in this case, studying the CIB fluctuations gives strong constraints on the CIB source population. This is true in the whole submm-mm range. The population of galaxies, where the CIB peaks, will not be accessible by direct detection in the coming years. For example, {\\it SIRTF} will resolve about 20$\\%$ of the background at 160 $\\mu$m (Dole et al. 2002), {\\it PACS} about 50 $\\%$ at 170 $\\mu$m (Sect. 7.2.2) and {\\it SPIRE} less than 10$\\%$ at 350 $\\mu$m (Sect. 7.2.1).\\\\  In conclusion, we have seen that our model gives number counts, redshift distributions, CIB intensity and fluctuations that reproduce all the present observations. It can be now used for future experiment predictions, in particular for {\\it Herschel}, {\\it Planck} and {\\it ALMA} observations. For {\\it SIRTF}, a complete and more detailed study, including simulations and a detailed discussion on the confusion, is done in Dole et al. (2002).   We have developped a phenomenological model that constrains in a simple way the IR luminosity function evolution with the redshift, and fits all the existing source counts and redshift distribution, CIB intensity and for the first time CIB fluctuations observations from the mid-IR to the submm range. The model has been used to give some predictions for future {\\it Herschel} deep survey observations and the all-sky {\\it Planck} survey. It comes out that the planned experiments ({\\it SIRTF}, {\\it Herschel}, {\\it Planck}) will be mostly limited by the confusion. To find out a large number of objects that dominate the LF at high redshift (z$>$2), future experiments need both the angular resolution and sensitivity. This can be achieved in the submm only thanks to interferometres such as {\\it ALMA}. However, mapping large fractions of the sky with high signal-to-noise ratio will take a lot of time (for example, 1.9 years to map 5 square degrees that resolve 50$\\%$ of the CIB at 1.3~mm with ALMA).   \\par\\bigskip\\noindent \\par\\bigskip\\noindent {\\bf APPENDIX A:} \\par\\medskip\\noindent We provide, in an electronic form through a web page{\\footnote{http://www.ias.fr/PPERSO/glagache/act/gal$_{-}$model.html}}, a distribution of the model's outputs and programs (to be used in IDL) containing: \\begin{itemize} \\item The array dN/(dlnL.dz) as a function of L and z, dS/dz as a function of L and z and S$_{\\nu}$ in Jy for each luminosity and redshift, from 10 to 2000 $\\mu$m and the evolution of the LF for both the normal and the starburst populations (for $\\Omega_{\\Lambda}$=0.7, $\\Omega_{0}$=0.3 and h=0.65) \\item Some usefull programs that compute the integral counts, detected sources, CIB and fluctuations redshift distribution, and the level of the fluctuations from the previous data cubes. \\end{itemize} \\par\\bigskip\\noindent   Acknowledgements: HD thanks the {\\it MIPS} project (under NASA Jet Propulsion Laboratory subcontract \\# P435236) for support during part of this work and the Programme National de Cosmologie and the Institut d'Astrophysique Spatiale for some travel funding."
    },
    "0209/astro-ph0209579_arXiv.txt": {
        "abstract": "An analytic model of the evolution of a rotating black hole (BH) is proposed by considering the coexistence of disk accretion with the Blandford-Znajek process.  The evolutionary characteristics of the BH are described in terms of three parameters: the BH spin $a_*$, the ratio $k$ of the angular velocity of the magnetic f\\/ield lines to that of the BH horizon and the parameter $\\lambda$ indicating the position of the inner edge of the disk. It is shown that the ratio $k$ being a little greater than 0.5 af\\/fects the evolutionary characteristics of the BH signif\\/icantly, and the BH spin increases rather than decreases in its evolutionary process provided that the initial value of the BH spin is located in an appropriate value range determined by ratio $k$. Our calculations show that the system of a BH accretion disk with $k=0.6$ might provide a much higher output energy in a shorter timescale for gamma-ray bursts than the same system with $k=0.5$. ",
        "introduction": "As is well known, the Blandford-Znajek(BZ) process was proposed originally as a possible energy mechanism of quasars and active galactic nuclei(AGNs;Blandford \\& Znajek 1977; Macdonald \\& Thorne 1982, hereafter MT82; Rees 1984). Recently, Lee, Wijers, \\& Brown (2000) proposed that the BZ process can be used as a central engine for powering gamma-ray bursts (GRBs), where the rotating energy of a stellar black hole (BH) with a magnetic f\\/ield of $10^{15} G$ is extracted along the magnetic f\\/ield lines supported by a transient magnetized accretion disk. Very recently, Lee and Kim (2000, 2002) proposed a model of the evolution of a rotating BH at the center of GRBs by considering the ef\\/fects of the BZ process with the transient disk (hereafter the LK model), and some constraints to the parameters of the BH accretion disk are given.  In this paper an analytic model of BH evolution is proposed based on the LK model and is improved in two respects: (1) A parameter $\\lambda$ is introduced for the position of the inner edge of the disk/torus, which is located between the innermost bound orbit and the last stable orbit. (2) Another parameter $k$ is used to indicate the ratio of the angular velocity of the magnetic f\\/ield lines to that of the BH horizon. It is shown that the ratio $k$ being a little greater than 0.5 af\\/fects the evolutionary characteristics of the BH signif\\/icantly, and the BH spin increases rather than decreases in its evolutionary process provided that the initial value of the BH spin is located in an appropriate value range determined by the ratio $k$. Our calculations show that the system of a BH accretion disk with $k=0.6$ might provide a much higher output energy in a shorter timescale for GRBs than the same system with $k=0.5$.  This paper is organized as follows: In \\S2 we derive the basic equations of BH evolution and the corresponding characteristic functions by introducing two parameters $\\lambda$ and $k$. In \\S\\S3 and 4 we discuss the ef\\/fects of the parameters $k$ and $\\lambda$ on BH evolution and GRBs by using the parameter space and the corresponding characteristic functions, respectively. Finally, in \\S5 we discuss some problems concerning our model. Geometrized units ($G=c=1$) are used in this paper. ",
        "conclusions": "In this paper the evolutionary characteristics of a rotating BH surrounded by a transient magnetized disk are discussed by considering the coexistence of disk accretion with the BZ process, and the two parameters $\\lambda$ and $k$ are introduced in our model to modify the LK model. A main consequence of our model is that the BH will be spun up rather than spun down provided that $k$ is greater than a critical value $k^{bot}$ with $a_*^{turn}<a_*(0)<a_*^{eq}$. Our calculations show that more energy from a BH system with $k$ above $k^{bot}$ might be extracted in less time compared to a BH system with $k$ below $k^{bot}$.  However there are still some problems related to our model. First, the dif\\/ferent values of the ratio $k \\equiv \\Omega_f/\\Omega_h$ play a very important role in our model. Unfortunately, the determination of this ratio has remained one of the main unresolved problems in the BZ process. It was argued in a speculative way that the ratio might be regulated to about 0.5, if the charged particles conspired with the BH to have just the right amount of inertia for the impedance matching (MT82). However, it is difficult to understand how the load can conspire with the BH to have the same resistance and satisfy the matching condition, since the load is so far from the BH that it cannot be causally connected (Punsly \\& Coroniti 1990).  It was pointed out in the theory of BH magnetospheres that $\\Omega_f$ can be regarded as a function of magnetic f\\/lux $\\Psi$, i.e., $\\Omega_f(\\Psi)$, since it is constant on the magnetic surface due to isorotation and axial symmetry (MT82). Very recently, Beskin and Kuznetsova (2000) discussed the stream equation describing magnetic surfaces $\\Psi (r,\\theta)$, and found that $\\Omega_f$ can be either greater or less than $\\Omega_h/2$ for the dif\\/ferent possibilities of the f\\/low, with the electric current f\\/ixed by the pair creation region in the BH magnetosphere. However, we  still have a long way to go for the determination of $\\Omega_f$ in a BH magnetosphere.  Second, we notice that the relation in equation(16) between ${\\dot M}_d$ and $B_h$ is the key to af\\/fecting the evolutionary characteristics. To illustrate this, we replace equation (16) with another relation between ${\\dot M}_d$ and $B_h$, which was proposed based on the balance between the pressure of the magnetic f\\/ield on the horizon and the ram pressure  of the innermost parts of the accretion f\\/low (Moderskin et al. 1997), i.e., \\begin{equation} \\frac{B_h^2}{8 \\pi}=P_{ram} \\sim \\rho c^2 \\sim \\frac{{\\dot M}_d}{4 \\pi r_h^2}. \\end{equation} From equation (21) we assume the relation to be \\begin{equation} {\\dot M}_d=\\frac{r_h^2B_h^2}{2}. \\end{equation} Hereafter equation (22) is referred to as the MSL relation. Substituting the MSL relation into equation (13) and (15), we obtain the following evolutionary equations: \\begin{equation} \\frac{dM}{dt}=f_{_{MSL}}(a_*,k,\\lambda) {\\dot M}_d, \\end {equation} \\begin{equation} \\frac{da_*}{dt}=M^{-1}g_{_{MSL}}(a_*,k,\\lambda) {\\dot M}_d, \\end{equation} where $f_{_{MSL}}(a_*,k,\\lambda)$ and $g_{_{MSL}}(a_*,k,\\lambda)$ are CFBHM and CFBHS corresponding to the MSL relation, respectively. In the same way, we obtain the parameter space consisting of $a_*$ and $k$ as shown in Figure 15. The parameter space is very dif\\/ferent from that depicted in Figure 2, and it is only divided into two regions by the boundary curve $g_{_{MSL}}(a_*,k,\\lambda)=0$, corresponding to only two states of BH evolution. The characteristics of BH evolution related to the MSL relation are shown in Tabel 3.  In this paper an analytic model of BH evolution is proposed for GRBs. Although this model is rather particular and speculative , the potential application is still attractive: a little variation of the angular velocity of the magnetic f\\/ield lines might af\\/fect the output energy and evolutionary timescales of a BH system remarkably. We also hope to improve and extend this model to AGNs and stellar-mass BHs in  future work, since there is more evidence to strongly suggest the existence of fast-spinning BHs and possibly energy extraction via Blandford-Znajek-like processes(Krolik 2001).  The anonymous referees are thanked for their suggestions on the improvement of our manuscript. V. S. Beskin is thanked for his encouragement and the discussion of the determination of $\\Omega_f$ in a BH magnetosphere. This work is supported by the National Natural Science Foundation of China under grant No. 10173004."
    },
    "0209/astro-ph0209053_arXiv.txt": {
        "abstract": "Radiative accelerations on Ne are calculated for the atmospheres of main sequence stars with $11\\,000 \\leq \\teff \\leq 15\\,000$\\,K.  This range corresponds to that of the mercury-manganese (HgMn) stars. The calculations take into account neon fine structure as well as shadowing of neon lines using the entire Kurucz line list, bound-bound, bound-free and free-free opacity of H, He and C as well as some NLTE effects. NLTE effects are found to modify the radiative acceleration by a factor of order $10^{2}$ in the outer atmosphere and are crucial for $dm<10^{-3}$ g\\,cm$^{-2}$. The dependence of the radiative accelerations on the Ne abundance, effective temperature and gravity is studied. Radiative accelerations are found to be well below the gravity in the entire range of \\teff\\ and it is predicted that in stable atmospheres devoid of disturbing motions, Ne should sink and be observed as underabundant. This agrees with recent observations of low Ne abundances in HgMn stars. ",
        "introduction": "Radiatively driven diffusion processes are generally accepted to be responsible for chemical peculiarities observed among upper main sequence CP stars \\citep{michaud70}. The upward radiative acceleration and downward gravity acting on different chemical species compete and drive their microscopic diffusion, resulting in the stratification of these elements in the stellar atmospheres or envelopes of these stars. If these atmospheres are sufficiently stable, chemical inhomogeneities can persevere, and are not wiped out by various mixing processes such as convective zones, stellar winds or rotationally induced mixing. In the particular case of HgMn star atmospheres, temperatures are high enough for hydrogen to be largely ionized, thus removing its convection zone.  Also, if the He ionization convection zone diminishes due to He settling in late B stars rotating slower than {\\it{c.}} 75 km\\,s$^{-1}$ \\citep{michaud82,cm88}, diffusion can operate very effectively in the atmosphere. This is the case for HgMn stars as they are also known as slow rotators. Such diffusion processes are generally time dependent problems, but the characteristic time-scales in the atmosphere are much shorter than in the envelope ($\\leq 10^{3}$ yr) due to the considerable lower density, smaller collision rates and higher diffusion velocities ({\\it{c.}}\\,1\\,cm\\,s$^{-1}$). This makes the time dependent calculations extremely difficult.  So far, diffusion can be followed in time only in the envelope, as shown in the case of neon by \\citet{Seaton97,Seaton99}; \\citet*{ldv98} and \\citet*{lmr00}. These calculations suggest that radiative accelerations on Ne are much lower than gravity below the atmosphere and Ne should sink and be observed as underabundant unless a weak stellar wind transports upwards the Ne accumulated in deep layers.  Nevertheless, observations are restricted to the atmospheres. This important stellar region bridges the effects in the envelopes with those outside the stars, such as stellar winds, and one clearly needs also to know the radiative accelerations here in order to compare the theory with observations. The case of Ne is worth studying for two reasons: it is an important element having very high standard abundance \\citep*{ag89,gns96,db00} of about A(Ne/H)=$1.23 \\times 10^{-4}$, and because one can expect similar effects as in the case of He. \\nei\\ (along with He\\,{\\sc i}) has a very high ionization potential and first excited level well above the ground state. \\nei\\ is thus the dominant ionization state in the atmospheres of these stars and all \\nei\\ resonance lines are in the Lyman continuum. This part of the spectrum is very sensitive to non-LTE (NLTE) effects and the temperature structure of the atmosphere. It is known that there is a big difference in the temperature behaviour between LTE and NLTE model atmospheres at small optical depths and that temperature can rise outwards in the NLTE models. \\nei\\ is also known to have strong NLTE effects on lines originating from excited levels in the visible part of the spectrum even in late B-type main sequence stars \\citep{am73,sigut99,db00}. Thus, the main complication is clear in doing the task.  One cannot rely upon the diffusion approximation when calculating energy flux as this is a non-local quantity in the atmosphere, in contrast to envelope calculations. One needs to solve the radiative transfer as, e.g., in \\citet*{hlh00}, \\citet*{haa96} or \\citet{hlhb02}, including NLTE effects if possible.  Unfortunately, observations of Ne in HgMn stars are very scarce and apart from recent LTE Ne abundances in $\\kappa$\\,Cnc, HR7245 and HD29647 by \\citet{ap00} and \\citet{aswesef01} the only study of Ne was undertaken recently by \\citet{db00}. Their NLTE analysis of 20 HgMn and 5 normal stars revealed that Ne is clearly underabundant in HgMn stars and the deficit reaches its peak values in the middle of the \\teff\\ domain of HgMn stars.  In this paper we deal with the radiative accelerations of neon in the atmosphere.  This might enable us to model the vertical element stratification in the future, e.g. considering the atmosphere as being in a sequence of stationary states which are to be tailored to the much more slowly varying envelope as proposed by \\citet{bzzzk93}. Nevertheless, some conclusions can be drawn already from the radiative accelerations alone. In the following, cgs units and LTE approximations (in a NLTE atmosphere model) are used if not specified otherwise. ",
        "conclusions": "\\subsection{Radiative accelerations on different Ne ions}  The LTE populations of various Ne ions throughout the atmosphere are depicted in Fig. \\ref{f2} for a representative atmosphere model with $T_{\\rm eff}=12\\,000$\\,K, $\\log g=4$. \\nei\\ is definitely dominant in the atmosphere.  Ne\\,{\\sc iv} becomes populated at the base of our models while \\neii\\ may dominate at very low densities and electron concentrations for $dm<10^{-4}$ g\\,cm$^{-2}$. But above $dm<10^{-3}$g\\,cm$^{-2}$ strong departures of Ne ionization and excitation equilibrium from LTE occur \\citep{sigut99} and our accelerations on different Ne ions should be considered as approximate above this level (see Fig.\\ref{f8} for an illustration of expected deviations from fully consistent NLTE case).  \\begin{figure} \\epsfig{file=bud-f2.eps,width=8.5cm} \\caption{Ne ionization fractions in LTE in a representative NLTE model atmosphere.} \\label{f2} \\end{figure}  Accelerations in the representative model are plotted in Fig. \\ref{f3}. One can see that none of them exceeds the gravitational acceleration except at the very deepest layers. The acceleration on \\nei\\ is strongest in the atmosphere where this ion is also most populated. The contribution of the neutral species like \\nei\\ to the total effective radiative acceleration of the element is both favoured by weighting with their diffusion coefficient (see Eq. \\ref{e10}) which is approximately two orders of magnitude larger than that for the charged species and reduced by the redistribution effect. Nevertheless, acceleration of \\nei\\ will be the most important contribution to the total acceleration not only in the atmosphere but also at rather great depth up to $dm \\simeq 10$\\,g\\,cm$^{-2}$. The low radiative accelerations found on \\nei\\ are a consequence of its atomic structure because all resonance transitions, which are usually the most important, are in the Lyman continuum where \\nei\\ atoms see little photon flux, while the rest of the lines originate from highly excited and weakly populated states. Nor can the redistribution effect discussed earlier increase the Ne I acceleration (see Eq.~\\ref{e8} and Fig.~\\ref{f6b}); just the opposite occurs. Radiative acceleration gained in the \\nei\\ state is redistributed between \\nei\\ and {\\neii}, which reduces the \\nei\\ acceleration and adds it to the \\neii\\ state where, however, the acceleration is not so effective due to the much lower diffusion coefficient of ionized species; see Eqs. \\ref{e7}, \\ref{e10}.  \\begin{figure} \\epsfig{file=bud-f3.eps,width=8.5cm} \\caption{Radiative accelerations on Ne\\,{\\sc i-iv} ions in a representative NLTE model atmosphere for the standard abundance of Ne. The downward gravitational acceleration of $10^4$ cm\\,s$^{-2}$ is at least 2-3 orders of magnitude greater than the upward radiative acceleration throughout the photosphere.} \\label{f3} \\end{figure}  We would like to stress that radiative acceleration (Eq. \\ref{e6}) on a particular ion is very sensitive to the partition functions. These are generally precise in the region where the ion dominates the element but have large uncertainties outside the region. Displaying effective accelerations overcomes the problem as partition functions drop out in Eq. \\ref{e10} outside the region. Thus, if the reader wishes to use our radiative accelerations stored separately for each ion, the same partition functions should be used also.  \\subsection{Dependence on effective temperature and surface gravity}  Now we can proceed further and explore how the situation changes with the effective temperature of the star. It is clear from Fig. \\ref{f4} that the total acceleration rapidly increases with the effective temperature. However, within our range of interest it never exceeds the gravity in the line forming region. Nevertheless, this suggests that Ne could accumulate at deeper layers in cooler stars than in hotter ones, the latter thus being more vulnerable to departures from the ideal stable atmosphere. Various mixing processes or radiatively driven stellar winds whose intensity also increases with the effective temperature \\citep{babel95,kk01} might more easily enrich Ne in the atmospheres of much hotter stars. This resembles the case of He which has atomic structure similar to Ne. The abundance of He progressively increases with effective temperature from He-weak towards He-rich stars. This suggests that it might be interesting to search for Ne-rich analogues of He-rich stars among early B stars. On the other hand, for $\\teff<11\\,000$\\,K, hydrogen becomes partially ionized, which triggers ambipolar diffusion and hydrogen superficial convective zones. This may also transport some Ne to the atmosphere from below. Our observations \\citep{db00} indicate that Ne underabundances are most pronounced in the middle of the HgMn domain and tend to be less extreme towards the cool and hot ends of the HgMn temperature region. This pattern could therefore be qualitatively expected. Observing Ne below $\\teff = 11\\,000$\\,K, however, becomes very difficult.  \\begin{figure} \\epsfig{file=bud-f4.eps,width=8.5cm} \\caption{Radiative accelerations on Ne throughout the atmosphere as a function of stellar effective temperature and depth. Calculated for standard abundance of Ne.} \\label{f4} \\end{figure}  \\begin{figure} \\epsfig{file=bud-f5.eps,width=8.5cm} \\caption{Radiative accelerations on Ne throughout the atmosphere as a function of stellar surface gravity and depth. Calculated for standard abundance of Ne.} \\label{f5} \\end{figure}  The radiative accelerations also depend on the surface gravity (Fig. \\ref{f5}). Lowering the surface gravity of the model shifts the radiative acceleration behaviour to deeper layers and reduces the radiative acceleration at the bottom of the atmosphere. However, the gravity to radiative acceleration ratio does not change very much and in this context effective temperature is a more important atmospheric parameter than $\\log g$. Apparently, within the main sequence (approximately $3.5\\leq \\log g \\leq 4.5$), radiative acceleration of Ne never overcomes gravity.  \\subsection{Effects of homogeneous Ne abundance}  Radiative acceleration on the element also depends on its abundance. This is a simple consequence of the fact that the energy flux in the integral in Eq. \\ref{e6} depends on the element abundance.  This is, however, only important in the case of strong lines where the line opacity is comparable to or larger than opacity in the continuum. One can then expect that radiative acceleration will asymptotically increase as element abundance decreases, until it reaches some maximum value corresponding to the sufficiently low abundance when all the element lines disappear and radiative acceleration will no longer be sensitive to the abundance \\citep[for more detail see][]{al00}.  The situation is illustrated in Fig. \\ref{f6}. Surprisingly, the accelerations are very little dependent on abundance in the photospheres above $dm \\simeq 0.1$ g\\,cm$^{-2}$ and do not exceed gravity for any abundance except at the extreme base of our models. Here a Ne deficit ranging from -0.5 dex for the hottest to -2 dex for the coolest model could be supported by radiation (i.e. radiative acceleration roughly equals gravity for this abundance and depth, $dm\\approx 20$\\,g\\,cm$^{-2}$).  This weak abundance sensitivity results from the fact that resonance \\nei\\ lines are not very strong because they are in the Lyman continuum where the continuous opacity is very large, effectively reducing line to opacity ratio. We have found that contributions to the radiative acceleration of lines originating from excited \\nei\\ levels is also important, but these weak lines do not introduce strong dependence on the abundance because their line-to-continuum opacity ratio is low due to the high excitation energies and low populations of the levels from which they originate. One can see from the figure that there is no abundance for which the radiative acceleration could balance the gravity in the line forming region above $dm \\simeq 0.1$\\,g\\,cm$^{-2}$. The implication is that in this region, assuming a stable atmosphere devoid of any motions, Ne should sink and should be almost completely depleted and only its strong concentration gradient could balance the downward flux of Ne atoms. The Ne atom is at least an order of magnitude heavier than the mean `molecular' mass of the gas, so the characteristic scale of the Ne abundance gradient would be an order of magnitude less than the pressure scale height, i.e.  $\\log dm \\approx 0.1$. Consequently, the fact that Ne is detected in most HgMn stars \\citep{db00} suggests the presence of some kind of instability competing with diffusion and suppressing the efficiency of diffusion processes. An example of such an instability is the weak stellar wind suggested by \\citet{ldv98}.  \\begin{figure} \\epsfig{file=bud-f6.eps,width=8.5cm} \\caption{Radiative accelerations on Ne throughout the representative NLTE model atmosphere as a function of the homogeneous Ne abundance.} \\label{f6} \\end{figure}  One may also notice a sudden drop in the radiative accelerations at approximately $10^{-4} \\leq dm \\leq 10^{-3}$ g\\,cm$^{-2}$. This is not a numerical artifact but a very interesting effect when radiative acceleration may acquire negative values; this will be discussed in more detail below.  \\subsection{Redistribution effect}  \\begin{figure} \\epsfig{file=bud-f7.eps,width=8.5cm} \\caption{An illustration of the redistribution effect on the radiative accelerations on Ne throughout the representative NLTE model atmosphere as a function of depth calculated for standard homogeneous abundance of Ne. (a) -- no redistribution -- solid line, (b) -- no redistribution and no diffusion coefficient weighting -- dashes, (c) -- \\citet{glam95} type redistribution -- dots, (d) -- our type redistribution -- dash-dots. The arrow points to the depth where $T=31000K$.} \\label{f6b} \\end{figure}  To take the redistribution effect into account properly one would need to calculate probabilities of all possible radiative and collisional transitions from the upper level to continuum including multiple cascade transitions as suggested by \\citet{glam95}. That is a very difficult task and the above authors developed a method in which they set $r_{u,i}=0$ for $n \\geq 3$ and $r_{u,i}=1$ for lower states. Thus their $r_{u,i}$ is a simple step function of level energy and does not depend on e.g. temperature. To cope with the problem and estimate the influence of the redistribution effect we propose a slightly different and more general approach. The idea is simple. As mentioned in Section~\\ref{theory}, collisions with electrons may change the ionization state before the momentum absorbed by the atom is lost in collisions with protons. Consequently, the redistribution function from a particular excited level depends mainly on the electron to proton collisional rates ratio. Higher atomic levels with energies corresponding to the electron kinetic energy $\\approx kT$ below the ionization threshold are most strongly coupled to the continuum via collisional excitation and ionization. Probabilities of such processes are strongly temperature dependent via the Boltzmann factor $e^{-\\Delta E/kT}$ \\citep{Seaton62,regemorter62} and still deeper atomic levels are affected as one goes deeper into the star. Electron to proton number density ratio almost does not change with depth in such hot stars and the Boltzmann factor thus embraces the essence of the electron to proton collision rates ratio and one may write $\\beta_{u,i}/\\beta_{i}=C e^{-\\Delta E/BkT}$. Consequently, based on Eq.~\\ref{e9b} we suggest the following expression for the $r_{u,i}$: \\begin {equation} r_{u,i}=\\frac{1}{Ce^{-U}+1}~~ {\\rm with}~~ U=(I_{i}-E_{u,i})\\left(\\frac{1}{A}+\\frac{1}{BkT}\\right) \\label{e21c} \\end {equation} where $I_{i}$ is the $i$-th ion ionization potential and $A,B,C$ are adjustable constants. It is beyond the scope of the present paper to calibrate precise values for the above mentioned constants. $B$ is most important as it parameterises the temperature dependence and is of the order of 1. The $1/A$ term was introduced for those who prefer to use a \\citet{glam95} type of redistribution but do not like its step function shape and $A$ is of the order of $kT$, but we use $A\\rightarrow \\infty$. $C$ parameterises the ratio of the total electron to proton collision rates. If $E_{u,i} \\rightarrow I_{i}$ the level will register all electron impacts and $\\beta_{u,i}/\\beta_{i}=C$. Because electrons are $\\sqrt{m_{p}/m_{e}}$ faster than protons, their collisions will be more frequent by approximately the same factor, and C is thus of the order of 40. The method of \\citet{glam95} is equivalent to the approximation of Eq.\\ref{e21c} by a temperature independent rectangle and could be approached as a special case when $B\\rightarrow \\infty$ and $A\\rightarrow 0$ or $A\\rightarrow \\infty$ depending on whether $E_{u,i}>I_{i}-E_{crit}$ or $E_{u,i}<I_{i}-E_{crit}$, respectively, where $E_{crit}$ is some threshold energy. Note that $A\\rightarrow 0$ means $r_{u,i}\\rightarrow 1$ i.e. no redistribution at all and $A\\rightarrow \\infty$ means $r_{u,i}\\rightarrow 1/(C+1)\\rightarrow$ a small value, i.e. total redistribution is approached but never fully realised.  Assuming that \\citet{glam95} took the lowest $n=3$ level of C\\,{\\sc ii} as a cutoff in their calculation for $T=31000$K which corresponds to $E_{crit}=80000$ cm$^{-1}$ below continuum, we can simulate their method (using our Eq.~\\ref{e8},~\\ref{e9} and forcing $r_{u,i}=1$ or $r_{u,i}=0$) and calibrate our method so that it would give $r_{u,i}=0.5$ at the same temperature. Assuming $C=40, A \\rightarrow \\infty$ we get from Eq.~\\ref{e21c} that $B=1.0$ -- a reasonable value indeed. Results of both methods are illustrated in Fig. \\ref{f6b} and one can see that while both methods may give similar results for $T=31000$K the \\citet{glam95} method strongly overestimates the redistribution effect for lower temperatures. A simple test of the redistribution effect treatment is recommended by calculating radiative accelerations without weighting them by diffusion coefficients (setting $D_{i}=1$ for example in Eq.~\\ref{e10}). This is because such accelerations are an invariant of the redistribution process.  \\subsection{Comparison with other work}  We calculated an atmosphere model for $T_{\\rm eff}=11\\,530$\\,K, $\\log g=4.43$, corresponding to one of the envelope models of \\citet{ldv98}. Then we calculated the radiative accelerations, but in the case of \\nei\\ we omitted all except the resonance lines, to simulate the assumption adopted by \\citeauthor{ldv98}.  This assumption is not really justified as the omitted lines contribute significantly to the radiative acceleration. Inclusion of \\nei\\ lines originating from excited levels, which absorb at the wavelengths where a star radiates much more than in the Lyman continuum, may increase the acceleration on \\nei\\ by more than 1.5 dex. However, this does not have any serious consequences on the conclusions of \\citeauthor{ldv98}, as the radiative accelerations are so small they will still remain below the gravity. The comparison is shown in Fig. \\ref{f7}. In this case we also considered $D_{i}=1$ in Eq. \\ref{e10} to follow their method. The agreement is quite good and small departures at the base of the atmosphere are mainly caused by slightly different temperature behaviour between our model atmosphere and their envelope model.  The curious gap in our curve at $10^{-4} \\leq dm \\leq 10^{-3}$\\,g\\,cm$^{-2}$ appears because the acceleration is negative here and cannot be plotted on a logarithmic scale.  \\begin{figure} \\epsfig{file=bud-f8.eps,width=8.5cm} \\caption{Comparison of our calculations in the atmosphere with those of \\citet{ldv98} in the envelope for the same model parameters. Here, for \\nei\\, only resonance lines are considered. Both were calculated for standard abundances of Ne.} \\label{f7} \\end{figure}  \\subsection{NLTE effects}  The aforementioned negative acceleration is a very interesting NLTE effect which may lower still further the radiative accelerations on some ions in this part of the atmosphere. It is a consequence of the temperature inversion in the model atmosphere and the atomic structure of \\nei. As shown in Fig. \\ref{f1}, temperatures in the outer layers rise in a NLTE model for $10^{-4} \\leq dm \\leq 10^{-3}$\\,g\\,cm$^{-2}$. At this depth, and in these stars, the atmosphere is still optically thick in the Lyman continuum.  Thus, according to the diffusion approximation, the star radiates into itself towards cooler regions at these wavelengths. If an element happens to have its important transitions in the Lyman continuum it may be pushed into the star by the radiation and its acceleration is thus negative.  Similar effects can be expected also in the cores of very strong lines not necessarily in the Lyman continuum. To explore NLTE effects on the results we also calculated radiative accelerations on \\nei\\ for three different assumptions in a representative NLTE model atmosphere with $T_{\\rm eff}=15\\,000\\,{\\rm K}, \\log g=4$ and zero microturbulence: \\begin{enumerate} \\item LTE case: neon populations in Eq. \\ref{e6} are in LTE and the neon lines' source functions for flux calculations is equal to the Planck function \\begin {equation} S_{\\nu}=B_{\\nu}; \\label{e22} \\end {equation} \\item approximate NLTE: the same, but the line source function for resonance transitions is allowed to deviate from the Planck function following second order escape probability methods \\begin {equation} S_{\\nu}=\\sqrt{\\frac{\\epsilon}{\\epsilon+(1-\\epsilon)K_{2}(\\tau)}}B_{\\nu} \\label{e23} \\end {equation} where $\\epsilon$ is the photon destruction probability and $K_{2}(\\tau)$ is the kernel function \\citep[see][the {\\sc{synspec}} manual, for more detail]{hlj94}. Note that this option was used in the calculations described in Section~\\ref{calculations} above; \\item fully consistent NLTE case: with NLTE model atmosphere, NLTE populations and NLTE source functions \\begin {equation} S_{\\nu}=\\frac{2h\\nu^{3}}{c^{2}} \\left(\\frac{g_{u,i}n_{l,i}}{g_{l,i}n_{u,i}}-1\\right)^{-1}. \\label{e24} \\end {equation} In the full NLTE case the NLTE model atmosphere and \\nei\\ level populations were calculated with {\\sc tlusty195}. As an input for {\\sc tlusty195} we considered Hubeny's \\hi\\ atom model with 9 explicit levels and continuum, the 31 level \\nei\\ atom model taken from \\citet{db00}, and a simple 4 level atom of \\neii\\ with continuum (Table~\\ref{t1}). Generally, the data for Ne bound-bound and bound-free transitions were taken from the TOPbase \\citep{cmoz93,hs94} but \\nei\\ oscillator strengths were from \\citet{Seaton98}. The {\\sc{modion}} IDL interface written by \\citet{vldhh95} was particularly useful in constructing \\nei\\ and \\neii\\ atom models. Populations of those very high \\nei\\ levels not included in our \\nei\\ atom model were considered in LTE relative to the ground state of {\\neii}, of which the population was calculated in NLTE, when calculating the energy flux and acceleration. All \\nei\\ level populations were set to LTE below $dm>0.3$\\,g\\,cm$^{-2}$ when calculating the radiative accelerations. \\end{enumerate}  \\begin{table} \\caption{\\neii\\ energy levels considered. Column 1: level designation, Column 2: ionization energy in cm$^{-1}$, Column 3: statistical weight of the level. Energies are from \\citet{persson}.} \\begin{center} \\begin{tabular}{lll} \\hline Desig. &Energy    & g  \\\\ \\hline $2p^5~^2\\!P^o$ & 330445. &     6 \\\\ $2p^6~^2\\!S$   & 113658. &     2 \\\\ $3s~^4\\!P$     & 111266. &    12 \\\\ $3s~^2\\!P$     & 106414. &     6 \\\\ \\hline \\end{tabular} \\end{center} \\label{t1} \\end{table}  \\begin{figure} \\epsfig{file=bud-f9.eps,width=8.5cm} \\caption{Comparison of radiative accelerations on \\nei\\ obtained using different assumptions: LTE, approximate NLTE and full NLTE (see text). Calculated for standard abundance of Ne.} \\label{f8} \\end{figure}  The differences between the three assumptions are exhibited in Fig. \\ref{f8} and become crucial for $dm<10^{-3}$\\,g\\,cm$^{-2}$, where the radiative accelerations in LTE and NLTE may differ by more than 3 dex. In the case of LTE we observe negative acceleration in the region of the temperature inversion. This was explained above. The approximate NLTE case is parallel to the LTE acceleration but about 1.5 dex smaller. This is caused by the difference in computed energy fluxes. While flux in the approximate NLTE case is real - lines are in absorption - and surprisingly close to the flux in the full NLTE case, the resonance \\nei\\ lines are in emission in LTE as soon as the atmosphere becomes optically thin in the Lyman continuum and acceleration soars immediately. The difference between approximate NLTE and full NLTE originates mainly in the differences in level populations."
    },
    "0209/hep-ph0209036_arXiv.txt": {
        "abstract": " ",
        "introduction": "\\setcounter{equation}{0}  Over the last few years, there has been considerable interest in models in which our  universe is a 3-brane (a hyper-surface) embedded in a higher dimensional bulk.  Much of the interest in extra-dimensional field theory is due to the hope for a solution to the hierarchy problem \\cite{An,RS,hierarchy}. In these models, the extra dimensions  are hidden from us, not necessarily by their smallness but by our confinement to a four-dimensional slice of the bulk spacetime \\cite{An}. In contrast to Kaluza-Klein scenarios, standard model interactions are confined to a brane whereas gravity propagates through the bulk perpendicular  to the brane. The hierarchy problem can be resolved by either postulating large extra dimensions (in which case the TeV scale is the fundamental scale of gravity and the Planck scale is derived in terms of the fundamental scale and the volume of the extra-dimensional space) \\cite{An}  or when the 4D metric scales exponentially throughout the bulk (the so called ``warp\" factor) \\cite{RS}.  While static brane-world models have served as a useful tool for testing ideas in higher dimensional spacetimes, their direct applicability to cosmology is limited. More realistic cosmological models may be derived by allowing a non-vanishing four-dimensional cosmological constant, or by introducing time-dependent energy-densities on the branes. Various cosmological aspects of such models have been investigated in the literature \\cite{large}-\\cite{more2}. One of the serious problems in brane models is the resulting unconventional set of Friedmann equations \\cite{LOW, LK, BDL}. The Hubble parameter, $H$, on the brane is often found to scale as $H \\sim \\rho$, rather than the standard four dimensional dependence, $H \\sim \\sqrt{\\rho}$. It has been shown however, that this problem can be solved upon the proper stabilization of the extra dimension \\cite{kkop1}-\\cite{kkop2}, that removes any unnecessary constraints between the brane energy-densities.  Both static and time-dependent generalizations of the original Randall-Sundrum solutions \\cite{RS} have been also constructed by introducing a bulk scalar field \\cite{scalars, kop1}. As a matter of fact, the task of the stabilization of the extra dimension was first accomplished by introducing a bulk scalar field, which had different non-vanishing vacuum expectation values on each of the two branes \\cite{GW}. The same topic was further elaborated in \\cite{kop1,stab-scal}. Here, we will derive two classes of brane-world solutions which include a static bulk scalar field and provide inflationary solutions in the 4D slices. The two classes correspond to either vanishing or non-vanishing bulk potential for the scalar field, and are characterized by a non-trivial warping of the metric along the extra dimension even in the case of zero bulk cosmological constant. Due to the appearance of a bulk curvature singularity, we are forced to consider two-brane-system configurations which, however, have a fixed inter-brane distance and lead to conventional FRW equations on the branes without any additional fine-tuning. By using a method developed recently in Ref. \\cite{kop}, we study the stability of those solutions under time-dependent perturbations of the radion field, and demonstrate that we can easily find parameter regimes where these solutions are {\\em stable}. Even more important is the fact that some of these stable solutions have a {\\it positive} cosmological constant on the brane - previously all known solutions of this type were unstable \\cite{dS}.   We organize this paper as follows. In section 2, we present the equations of motion of our theory and show how a factorizable (in time and the extra space coordinate $y$) scale factor can be obtained in the presence of a bulk scalar field. In section 3, we present two classes of inflationary brane solutions involving a static bulk field with vanishing or non-vanishing, respectively, bulk potential and we demonstrate that these solutions indeed lead to conventional FRW equations. We derive the conditions for the stabilization of these solutions in section 4 and investigate the parameter regimes that correspond to stable configurations.  Finally, in section 5, we summarize our results. ",
        "conclusions": "\\setcounter{equation}{0}  There has been a lot of work related to the stabilization of the radion field that parametrizes the size of the extra dimension in the context of brane-world models. Most work has focused on the derivation of solutions with a constant radion field, i.e. a static extra dimension, with only a few papers investigating whether these solutions correspond to a true minimum of the radion effective potential. In this paper, we presented two new, brane-world solutions arising in the presence of a bulk scalar field, and studied their stability under time-dependent perturbations of the radion field demonstrating the existence of phenomeno\\-logically interesting, stable solutions with a positive cosmological constant on the brane.  For our analysis, we used a factorizable ansatz for the line-element along the brane. Under the assumption that the total energy of each brane is the sum of a constant brane tension and the interaction term of a time-independent, bulk scalar field, the factorization of the 3D scale factor was shown to be directly related to the stabilization of the extra dimension. This relation holds both on and off the brane, as one can see by using the scale factor {\\it jump} conditions and the off-diagonal component of Einstein's equations, respectively.  The first class of solutions presented here corresponds to a vanishing bulk potential for the scalar field and a vanishing bulk cosmological constant. Surprisingly enough, brane-world solutions with non-trivial warping along the extra dimension did emerge, with the `warping' parameter being the expansion rate on the brane. These configurations accept a variety of  time-dependent solutions for the scale factor on the brane, with flat ($k=0$) or curved ($k=\\pm1$) 3D spacetime and a zero, positive or negative effective cosmological constant. All of the above solutions are characterized by the presence of a bulk curvature singularity, which inevitably leads to the introduction of a second brane in the model. One can show that the {\\it jump} conditions, imposed on the warp factor and the scalar field, lead to the fixing of the inter-brane distance, in terms of the fundamental parameters of the theory, and that the conventional Friedmann equation on the brane is successfully recovered.  In the second class of solutions that we derived, the presence of a non-trivial bulk potential and a bulk cosmological constant was restored. In this case, the warping of the 5D metric was governed by the bulk cosmological constant shifted by a constant quantity given in terms of the kinetic and potential energy of the bulk field. The same variety of cosmological solutions on the brane emerge here as well. The sign of the kinetic term of the scalar field reveals its nature (normal or tachyonic) and affects the behaviour of the bulk potential: a normal kinetic term leads to a potential which is unbounded from below, near the bulk singularity, whereas a tachyonic kinetic term leads to an infinitely-high potential barrier that may be used to shield the singularity in single-brane configurations. Here, we introduced a second brane in order to do so, and we demonstrated that, as in the first case, the inter-brane distance is fixed and the form of the Friedmann equation on the brane is recovered.  In the second part of our paper, we investigated the stability of our solutions under small, time-dependent perturbations of the radion field. We derived the {\\it extremization} constraints for both types of solutions and demonstrated that, as expected, they correspond to extrema of the radion effective potential. The {\\it stabilization} constraints were also derived. These revealed the stability behaviour of the solutions and the type of extrema to which they correspond, either minima or maxima. In the first class of solutions, brane configurations with positive, zero or negative effective cosmological constant were studied and it was shown that, in the absence of a scalar field potential, those solutions come out to be local maxima, saddle points or minima of the radion effective potential, respectively, in agreement with the literature. However, in our case, the presence of an extra term, involving second derivatives of the interaction terms of the scalar field on the branes, acts as a {\\it universal} stabilizing force, independent of the sign of the cosmological constant on the brane, as long as the second derivatives are positive. In this way, solutions with a positive cosmological constant on the brane may become stable, saddle-point solutions \\`a la Randall-Sundrum may turn to true minima, and AdS-type solutions on the brane may be further stabilized. The results for the second class of solutions found in this paper are even more interesting: the aforementioned term with the derivatives of the interaction terms may still act as an extra stabilizing force - upon appropriate choice of the sign of the second derivatives - nevertheless, stable solutions can arise even in the case where this term is zero. The parameter regimes that correspond to stable solutions are determined by the values of the bulk cosmological constant and the kinetic term of the scalar field. It is worth noting that the sign of the latter quantity also defines the sign of the effective cosmological constant of the stable solution: a normal (positive) kinetic term gives rise to solutions with negative cosmological constant, while a tachyonic (negative) one leads to a positive effective cosmological constant.  We may, therefore, conclude that the introduction of a bulk scalar field, in a brane-world model, may successfully lead to a variety of stable solutions, but more importantly, it may lead to cosmologically interesting, stable solutions with a positive effective cosmological constant - a type of solutions that has been difficult to derive up to now."
    },
    "0209/astro-ph0209503_arXiv.txt": {
        "abstract": "FRIIb radio galaxies provide a modified standard yardstick that allows constraints to be placed on global cosmological parameters. This modified standard yardstick is analogous to the modified standard candle provided by type Ia supernovae. The radio galaxy and supernova methods provide a measure of the coordinate distance to high-redshift sources, and the coordinate distance is a function of global cosmological parameters.  A sample of 20 FRIIb radio galaxies with redshifts between zero and two are compared with the parent population of 70 radio galaxies to determine the coordinate distance to each source.  The coordinate distance determinations are used to constrain the current mean mass-energy density of quintessence $\\Omega_Q$, the equation of state of the quintessence $w$, and the current mean mass-energy density of non-relativistic matter $\\Omega_m$; zero space curvature is assumed.  Radio galaxies alone indicate that the the universe is currently accelerating in its expansion (with 84\\% confidence); most of the allowed parameter space falls within the accelerating universe region on the $\\Omega_m - w$ plane.  This provides verification of the acceleration of the universe indicated by high-redshift supernovae, and suggests that neither method is plagued by systematic errors. It is found that $\\Omega_m$ must be less than about 0.5 and the equation of state $w$ of the quintessence must lie between -0.25 and -2.5 at about 90\\% confidence.   Fits of the radio galaxy data constrain the model parameter $\\beta$, which describes a relation between the beam power of the AGN and the total energy expelled through large-scale jets.  It is shown that the empirically determined model parameter is consistent with models in which the outflow results from the electromagnetic extraction of rotational energy from the central compact object. A specific relation between the strength of the magnetic field near the AGN, and the spin angular momentum per unit mass of the central compact object is predicted. ",
        "introduction": "There are several independent ways to determine the global cosmological parameters that describe the current state of the universe. It is important to have several complementary and independent methods that yield consistent results since any given method could be plagued by unknown systematic errors.  A particularly useful way to determine global cosmological parameters is through measurements of the coordinate distance to high-redshift sources.  This method is particularly useful because the coordinate distance depends only upon global cosmological parameters, and is independent of the clustering properties of the mass-energy components that control the expansion rate of the universe, as long as each component is homogeneous and isotropic on very large scales. The coordinate distance is also independent of whether different components cluster differently, known as biasing, and on whether the dark matter is cold, warm, or hot (though it does depend on the equation of state of each component).  Two cosmological tools that are particularly sensitive to the coordinate distance as a function of redshift are FRIIb radio galaxies, which provide a modified standard yardstick (Daly 1994, 2002a; Guerra 1997; Guerra \\& Daly 1998 [GD98]; Guerra, Daly, \\& Wan 2000 [GDW00]), and type Ia supernovae, which provide a modified standard candle (e.g. Riess et al. 1998, Perlmutter et al. 1999).  Some aspects of the methods are compared in \\S \\ref{compare}.  The coordinate distance to a source at a given redshift depends on the present value of the mean mass-energy density of each component, and on the redshift evolution of the mass-energy density of each component.  Non-relativistic matter, including baryonic matter and clustered dark matter, have a mean mass-energy density that evolves as $(1+z)^3$. There is a contribution from radiation and neutrinos left over from the big bang, currently negligible, which has an energy-density that evolves as $(1+z)^4$.  There may also be a cosmological constant, which constant energy density, and so evolves as $(1+z)^0$.  The redshift evolution of the mean mass-energy density of a component depends on the equation of state $w = P/\\rho$ of the component. There could exist a dynamical vaccum energy density called Quintessence (Caldwell, Dave, \\& Steinhardt 1998). Quintessence would have an equation of state $w$ such that the density evolves as $(1+z)^n$, where $n=3(w+1)$ (see Turner \\& White 1997; Bludman \\& Roos 2001, Wang et al. 2000; or the Appendix of this paper). For example, pressure-less dust (i.e., non-relativistic matter) is described by $P=w=0$ and $n=3$; a relativistic fluid is described by $w = 1/3$ and $n=4$, and a cosmological constant is described by $w= -1$ and $n=0$.  Two components of the universe are known to exist with certainty: there is a non-relativistic component (including baryons and clustered dark matter), with zero-redshift, normalized mean mass-energy density $\\Omega_m$, and a primordial relativistic component that is negligible for the epochs of interest here (including microwave background photons and neutrinos).  There must be at least one more component, which remains to be determined.  This third component could be a cosmological constant, quintessence, space curvature, or something else.  In the simplest case, there is only one additional component that is significant and has a measurable impact over the redshift interval where the coordinate distance is used to constrain cosmological parameters and the properties of the unknown component.  Here, the determination of the coordinate distances to 20 radio galaxies with redshifts between zero and two are used to constrain the equation of state and current mean mass-energy density of quintessence assuming a spatially flat universe as indicated by measurements of the microwave background radiation (de Bernardis et al. 2000, Balbi et al. 2000).  The radio galaxy method is briefly reviewed in \\S \\ref{radgals}. Constraints on cosmological parameters and the equation of state of quintessence obtained using FRIIb radio galaxies as a modified standard yardstick are presented in \\S \\ref{quint}. The supernovae and radio galaxy methods and results are compared in detail in \\S \\ref{compare}.  The relation between beam power and total energy for FRIIb radio jets is discussed in \\S \\ref{beta}. Here it is shown that the underlying hypotheses of the radio galaxy method is consistent with current models of large-scale jet production in AGN. The results are summarized in \\S \\ref{summary}. ",
        "conclusions": "\\label{summary}  Type IIb radio galaxies and type Ia supernovae are particularly important methods to develop to constrain the global cosmological parameters $\\Omega_m$, $\\Omega_{\\Lambda}$, and space curvature, or $\\Omega_m$, $\\Omega_Q$, and the equation of state of quintessence $w$.  These methods depend only upon the properties of global cosmological parameters, and are independent of other factors such as the index of the primordial power spectrum, the Hubble constant, the baryon fraction, the properties of the dark matter that clusters around galaxies and clusters of galaxies, any biasing of dark relative to luminous matter, etc.  The radio galaxy and  supernova methods are completely independent and have completely different potential systematic errors, as discussed in \\S \\ref{compare}.  Radio galaxies may be used to constrain the mass-energy density and equation of state of quintessence.  These results are presented here assuming a spatially flat universe, which is supported by recent measurements of fluctuations of the cosmic microwave background.  Radio galaxies alone suggest that the universe is accelerating in its expansion at present (see S \\ref{quint}), consistent with results obtained by the supernovae teams, indicating that both methods are working well and probably are not plagued by unknown systematic errors.  The implications for models of energy extraction for the cases of an outflow related to accretion and the Eddington luminosity, and electromagnetic energy extraction of rotational energy are considered.  If the outflow is produced by the electromagnetic extraction of energy from a rotating black hole, then the magnetic field strength must be related to the spin angular momentum of the rotating black hole $S$, the mass of the black hole $M$, and the gravitational radius of the black hole $m$, as described in \\S 5.  The relation is particularly simple if $\\beta=1.5$, and implies that the magnetic field strength satisfies $B \\propto (a/m)$, where $a =S/(Mc)$."
    },
    "0209/astro-ph0209029_arXiv.txt": {
        "abstract": "{ The relative orientation of clusters' major elongation axes and clusters' angular momenta is studied using a large N-body simulation in a box of 500~\\hMpc\\ base length for a standard $\\Lambda$CDM model.  Employing the technique of mark correlation functions, we successfully separated the correlations in the orientation from the well known clustering signal traced by the two-point correlation function.  The correlations in the orientation are highly significant for our sample of 3000 clusters.  We found an alignment of neighboring clusters, i.e. an enhanced probability of the major elongation axes of neighboring cluster pairs to be in parallel with each other. At 10~\\hMpc\\ separation the amplitude of this signal is $\\sim 10\\%$ above the value expected from random orientations, and it vanishes on scales larger than 15~\\hMpc. The ``filamentary'' alignment between cluster's major elongation axes and the lines pointing towards neighboring clusters shows even stronger deviations from random orientation, which can be detected out to scales of 100~\\hMpc, both in 2D and 3D analyses. Similarly, strong correlations of the angular momentum were seen. Also a clear signal in the scalar correlation of the absolute value of the angular momentum, the spin parameter and the mass was found.  They extend up to 50~\\hMpc\\ and have an amplitude of 40\\%, 15\\%, and 10\\% above a random distribution at 10~\\hMpc\\ separation, respectively. ",
        "introduction": "The study of orientation effects between galaxy clusters has a long and controversial history in cosmology.  In a seminal study {}\\citet{binggeli:shape} claimed that galaxy clusters are highly eccentric and oriented relative to neighboring clusters if lying at separations smaller than 15~\\hMpc.  Further he found anisotropies in the cluster distribution on scales up to 50~\\hMpc.  Following studies found no or weak statistical significance for orientation effects between neighboring clusters or between cluster orientation and the orientation of the central dominant galaxy, cp. {}\\citet{struble:new}, {}\\citet{flin:alignment}, and {}\\citet{rhee:mirror}. Remarkable was the apparent absence {}\\citep{ulmer:major} or weakness {}\\citep{rhee:x+opt} of orientation effects in projected X-ray contours of clusters, but it should be noted that the cluster samples at this time were small.  Analyzing a large set of 637 Abell clusters, {}\\citet{plionis:up150} found highly significant alignment effects on scales below 10~\\hMpc\\ that become weaker but extend up to 150~\\hMpc. More objectively selected, but smaller cluster samples seemed to put into question the reality of this signal, cp.  {}\\citet{fong:2d} and {}\\citet{martin:milano}. However, {}\\cite{chambers:x-contours} found significant nearest neighbor alignment of cluster X-ray isophotes using data from {\\sl Einstein} and {\\sl ROSAT}.  With the advent of new rich cluster catalogues as the optical {}\\textsc{Enacs} survey {}\\citep{katgert:enacs-i} and the X-ray based {}\\textsc{Reflex} survey {}\\citep{boehringer:reflex}, the question of orientation effects in clusters should attract renewed attention. Sufficiently large and well defined cluster samples showing only weak contamination by projection effects seem to be necessary to clarify this uncertain situation.  Strong stimulus to study orientation effects in clusters came from early ideas that a possible relative orientation between neighboring clusters or of clusters in the same supercluster should reflect the underlying structure formation mechanism.  \\cite{binney:prolat} proposed that tidal interactions of evolving protocluster systems may lead to the growth of anisotropies of clusters and to relative orientation effects.  Later, \\cite{vanHaarlem:merging} used numerical simulations of CDM models to demonstrate that clusters are elongated along the incoming direction of the last major merger.  In the same spirit, \\cite{west:merging} found that clusters grow by accretion and merging of surrounding matter that falls into the deep cluster potential wells along sheet-like and filamentary high density regions.  Therefore, the cluster formation is tightly connected with the supercluster network that characterizes the large-scale matter distribution in the universe.  High-resolution simulations showing this effect are described by the Virgo collaboration, cp.  {}\\citet{colberg:virgo}. {}\\citet{onuora:alignment} found a significant alignment signal up to scales of 30~\\hMpc\\ for a $\\Lambda$CDM model, whereas in a $\\tau$CDM model the signal extended only up to scales of 15\\hMpc.  To quantify the alignment of the galaxy clusters, we use a large $\\Lambda$CDM simulation in a box of 500~\\hMpc\\ side length. We identify a set of 3000 clusters. As statistical tools we employ mark correlation functions (MCF), as introduced to cosmology by {}\\citet{beisbart:luminosity}.  In this article we will extend this formalism to allow for vector valued marks.  The direction of the major axis of the mass ellipsoid serves as the vector mark.  Tightly connected with the elongation of clusters is its internal rotation. According to {}\\cite{doroshkevich:origin} and {}\\cite{white:angular}, the primary angular momentum of bound objects is due to tidal interaction between the elongated protostructures after decoupling from cosmic expansion and before turn-around.  More recent studies find that the angular momentum of dark matter halos is later modified by the merging history of their building blocks, cp. {}\\cite{vitvitska:origin} and {}\\cite{porciani:testingI, porciani:testingII}.  Therefore, we utilise the angular momentum as an additional mark for the study of the correlation of inner properties of simulated clusters, and we compare it with the orientation effects.  The plan of the paper is as follows.  In the next section, we describe our numerical simulation, the selection of a cluster sample and the precision with which we can determine structure parameters from it. Next we discuss the MCFs that are relevant for our studies.  In particular, we use special MCFs for vector marks to quantify correlations of orientation.  In Sect.~\\ref{sect:shape} we investigate correlations in the spatial orientation of clusters both in 3D and in the projected mass distribution.  In Sect.~\\ref{sect:angular} we present a MCF analysis using the angular momentum, mass and spin taken as vector and scalar marks, respectively. We conclude with a summary of the results. ",
        "conclusions": "Whether there exist correlations in the orientations of galaxies or galaxy clusters has been discussed for a long time. {}\\citet{binggeli:shape} reported a significant alignment of the observed galaxy clusters out to 50~\\hMpc. {}\\citet{struble:new,struble:new-erratum} claimed that this effect is small and prone to systematics and {}\\citet{ulmer:major} find no indication in their investigation.  Subsequently, several authors found, sometimes only weak, signs of alignments in the galaxy and galaxy cluster distribution (see e.g.\\ {}\\citealt{djorgovski:coherent, lambas:statistics,fuller:alignments,heavens:intrinsic}). As a novel statistical method we have used the mark correlation functions (MCFs) to quantify the alignment of cluster sized halos, extracted from a large scale simulation based on a $\\Lambda$CDM cosmology. Our sample with 3000 cluster sized halos is bigger than the currently available samples of galaxy clusters. The unambiguous signal we obtain benefits from the large statistics in our simulation.  Using two different weighting functions in the construction of the MCFs we investigate the direct alignment and the filamentary alignment. First we use the major axis of the mass ellipsoid as our direction marker. The clear signal from the direct alignment $\\CA(r)$ extends out $\\sim30~\\hMpc$. For the filamentary alignment $\\CF(r)$ we find deviations from isotropy up to $\\sim100~\\hMpc$.  Considering the projected mass distribution, the signal from the direct alignment $\\CA(r)$ already vanishes at a scale of $\\sim10~\\hMpc$. However, we find a filamentary alignment $\\CF(r)$ out to scales of $\\sim100~\\hMpc$, even for the projected data.  This scale is very similar to the size of the large scale filaments seen in our simulation. We think that the function $\\CF(r)$ is a powerful tool for exploring large scale alignment effects also in observational data.  {}\\citet{franx:elliptical} showed that the angular momentum of an ellipsoidal system tends to align with the minor axis of this system. We confirm this behavior in our simulation.  With the angular momentum as vector mark, $\\CF(r)$ shows the expected filamentary correlations: the angular momentum tends to be perpendicular to the connecting line, i.e.\\ the filament, up to separations of $\\sim40~\\hMpc$.  However, we obtain no signal for the direct alignment $\\CA(r)$. This is in concordance with the perception that the angular momenta are randomly oriented in the planes perpendicular to the filaments.  With the scalar MCFs $k_{\\rm m}(r)$ and $\\cov(r)$ we have investigated the correlations in the absolute value of the angular momentum.  Close pairs of clusters tend to have similar and also higher absolute values of the angular momentum compared to the global average. A clear signal can be detected up to $\\sim50~\\hMpc$. A further analysis of the mass and spin parameter distribution of the clusters with the MCFs has shown that this enhancement of the absolute value of the angular momentum is caused by an enhanced mass of close pairs of clusters as well as by the stronger rotational support of them. This behavior should be caused by the combined action of large-scale tidal fields and the hierarchical merging of progenitor structures and mass inflow onto the cluster. Since this mass growth follows the large scale filaments, tidal interactions and merger events are tightly connected. The mark correlation function with scalar and vector marks deliver quantitative measures of these effects.  \\subsection*"
    },
    "0209/astro-ph0209359_arXiv.txt": {
        "abstract": "We report Doppler measurements of $\\alpha$~Cen~A from time-series spectroscopy made with UCLES at the 3.9-m AAT.  Wavelength calibration using an iodine absorption cell produced high-precision velocity measurements, whose power spectrum shows the clear signature of solar-like oscillations, confirming the detection reported by Bouchy \\& Carrier (2001). ",
        "introduction": "We were awarded six nights to observe \\acena{} in May 2001 with UCLES and the iodine cell at the 3.9-m Anglo-Australian Telescope (AAT).  The run was affected by bad weather (the first night was completely lost), with an overall usability of about 50\\%.  We obtained 5169 spectra of \\acena, with one spectrum every 20\\,s.  We were also awarded four nights at the VLT to use UVES with the iodine cell.  Again, only 50\\% of the time was usable.  Data processing was postponed by our development of a new raw reduction package (which will also be used for the planet-search program).  The revised software properly treats the echelle blaze function, removes about 90\\% of the cosmic rays, and includes a telluric filter.  The net result is an improvement of about 0.5 m/s in the velocity precision of each spectrum. Velocities have been extracted from the UCLES/AAT spectra, but processing of the UVES/VLT data is not yet complete. ",
        "conclusions": ""
    },
    "0209/astro-ph0209498_arXiv.txt": {
        "abstract": "The evolution of a stellar disk under the influence of viscous evolution, photoevaporation from the central source, and photoevaporation by external stars is studied. We take the typical parameters of TTSs and the Trapezium Cluster conditions. The photoionizing flux from the central source is assumed to arise both from the quiescent star and accretion shocks at the base of stellar magnetospheric columns, along which material from the disk accretes. The accretion flux is calculated self-consistently from the accretion mass loss rate. We find that the disk cannot be entirely removed using only viscous evolution and photoionization from the disk-star accretion shock. However, when FUV photoevaporation by external massive stars is included the disk is removed in \\( 10^{6}-10^{7} \\)yr; and when EUV photoevaporation by external massive stars is included the disk is removed in \\( 10^{5}-10^{6} \\)yr.  An intriguing feature of photoevaporation by the central star is the formation of a gap in the disk at late stages of the disk evolution. As the gap starts forming, viscous spreading and photoevaporation work in resonance. When viscous accretion and photoevaporation by the central star and external massive stars are considered, the disk shrinks and is truncated at the gravitational radius, where it is quickly removed by the combination of viscous accretion, viscous spreading, photoevaporation from the central source, and photoevaporation by the external stars. There is no gap formation for disks nearby external massive stars because the outer annuli are quickly removed by the dominant EUV flux. On the other hand, at larger, more typical distances (\\( d\\gg 0.03 \\)pc) from the external stars the flux is FUV dominated. As a consequence, the disk is efficiently evaporated at two different locations; forming a gap during the last stages of the disk evolution. ",
        "introduction": "The Hubble Space Telescope (HST) has provided clear evidence of gas disks surrounding young stars in the Orion Nebula. Narrow band images reveal circumstellar disks seen in silhouette against either the background nebular light or the proplyd's own ionization front \\citep{orion observations}. These disks have been identified as {}``evaporating'' by \\citet*{j98}. Theoretically, disks should be ubiquitous. Any breaking of the spherical symmetry of the protostellar collapse will result in in-falling material being deflected from the radial direction, and disks forming around the central stars. Spherical symmetry may be broken either when the central star core is magnetized, or when the protostellar cloud has initial angular momentum. Magnetic fields tend to produce large pseudo-disks; since the material is not solely rotationally supported \\citep{magnetized}. Alternatively, even small initial rotational velocities in the protostellar cloud produce rotationally supported disks containing most of the angular momentum of the system \\citep{rotation}. For most theoretical models of the collapse of rotating clouds, the majority of the cloud material falls first onto the disk. Thus, as the molecular core collapses the disk mass increases. However, it is unlikely that the disk mass, \\( M_{d} \\), becomes larger than the superior limit, \\( M_{max}\\sim 0.3M_{\\star } \\), where \\( M_{\\star } \\) is the mass of the central star. At this superior limit the disk becomes gravitationally unstable, angular momentum is transported outward by spiral density waves, and the disk accretes material toward the central star at almost the same rate as it is receiving material from the molecular core \\citep{mmax}.  Planet formation is an exciting possible outcome of proto-stellar disk evolution. The coplanarity and circularity of the planetary orbits in our Solar System support this notion. Explaining the origin of the Solar System and extra-solar systems requires an understanding not only how the disks form; furthermore, we need to understand the disk evolution. In particular, the disk removal timescale and the timescale to assemble planets determine the possibility of planet formation. Shu, Johnstone, \\& Hollenbach (1993) proposed photoevaporation of the Solar Nebula as the gas removal mechanism that explains the differences in envelope masses between the gas-rich giants, Jupiter and Saturn; and the gas-poor giants, Uranus and Neptune. \\citet{removal mechanisms} generalized the discussion, describing the variety of possible disk removal mechanisms. The dominant disk removal mechanism at the inner parts of the disk is viscous accretion onto the central star. However, this process is incapable of removing the entire disk in a finite time because the accretion rate decreases as the viscous disk spreads, and the disk lifetime becomes infinite. Other possible disk removal mechanisms are planet formation, stellar encounters, stellar winds or disk winds, and photoevaporation by ultraviolet photons. \\citet{removal mechanisms} concluded that planet formation is a minor disk removal mechanism, and that the dominant mechanisms for a wide range of disk sizes are viscous accretion and photoevaporation, operating in concert within the disk.  Recently, \\citet*{uvswitch} have studied the observational consequences of the evolution of disks through a combination of photoevaporation and viscous disk evolution. Their study focused on photoevaporation due to ultraviolet photons produced in the disk-star accretion shock under the assumption that the accretion luminosity was constant during accretion and switched off when the inner disk was cleared. Using this model \\citet{uvswitch} were able to reproduce the observed millimeter fluxes of stars with disks as a function of the observed accretion rate. In this complimentary study, we focus on the physical properties of the disk under a variety of photoevaporation and viscous scenarios in order to understand the internal disk evolution. We use a time dependent \\( \\alpha  \\)-disk model \\citep{alpha} with the parameters of \\citet{h98} that are consistent with observed mass accretion rates in T Tauri stars (TTSs). Photoevaporation by external stars is studied using the model and parameterization of \\citet{j98}, in their study of the Orion Nebula. Photoevaporation by the central star is modelled with solutions originally found for high mass stars \\citep{pdr} and normalized to TTSs \\citep{evaporation shu}; however, in order to study disk evolution, approximations for the time dependence of evaporation are included in the model by estimating the continual change in the accretion shock emission of ultraviolet photons as the accretion rate subsides. In agreement with \\citet{removal mechanisms} and \\citet{uvswitch}, we show that it is possible to remove the entire disk in a finite time. However, we show that the rapid removal of the inner disk, described by \\citet{uvswitch} is not self-consistent. We further show that gaps in the disk are a natural outcome of the combination of viscous accretion and photoevaporation by the central star. ",
        "conclusions": "We have studied the possibility of disk removal by the combination of viscous diffusion and photoevaporation, assuming that the ultraviolet photons responsible for evaporation arises either from the quiescent stellar photosphere, the accretion shock, or external O stars. It is not possible to remove the entire disk when the only disk removal mechanism is viscous accretion: the disk spreads indefinitely in order to conserve total angular momentum while material is accreted onto the central star. Mass loss due to photoevaporation removes material along with its specific angular momentum; thus, it is possible to accrete material toward the central star and reduce the amount of disk spreading. The combination of the two mechanisms can result in finite disk lifetimes.  The distinctive features of photoevaporation by the ionizing flux from the central source are the formation of a gap around the EUV gravitational radius at late stages of the disk evolution and the lack of a finite time for the complete dispersal of the disk. The gap forms for the observed range of accretion shock temperatures (\\( 1-3\\times 10^{4} \\)K) and the disk becomes divided into an inner and an outer annulus. The inner annulus continues to be removed by the combination of viscous accretion, viscous spreading of material beyond the EUV gravitational radius, and photoevaporation at this radius. The outer annulus is removed as viscous spreading of material toward the gravitational radius and photoevaporation work in resonance. \\citet{uvswitch} considered models with photoevaporation by a constant central ionizing flux combined with viscous evolution, and showed that the timescales to remove the inner and the outer annuli are not the same. We conclude that this result is due to their assumption of a constant ionizing flux. In contrast, we calculate the ionizing flux from the accretion luminosity self-consistently.  We find that both the inner and outer disks survive to much longer times, and that the inner disk is {\\it not} removed first. It is not possible to quickly remove the inner annuli and maintain a high ionizing flux at the same time because the accretion rate decreases.  The formal disk lifetime is found to be in the range \\( 10^{12}-10^{13} \\)yr for \\( 10^{-3}<\\alpha <10^{-2} \\) and \\( 10^{-2}<M_{d}(0)<10^{-1} \\), much longer than the star lifetime. Introducing a stellar chromosphere will produce models intermediate between those presented here and those of \\citet{uvswitch}.  Stellar disk material can also be evaporated by external stars. This is a likely situation for the disks surrounded by ionization fronts in the Trapezium Cluster \\citep{j98, sh99, orion observations}. The EUV photons dominate the flux for the proplyds orbiting at a distance \\( d\\la 0.03 \\)pc, and the FUV photons dominate for the proplyds at more typical (\\( d\\gg 0.03 \\)pc) distances from the massive external stars.  We consider photoevaporation due to both the central source and external stars. There is no gap formation for disks nearby external hot stars since their ionizing flux removes all the disk material outside the EUV gravitational radius. The disk is quickly removed by viscous accretion, viscous spreading, and photoevaporation outside the gravitational radius. The disk lifetime is in the range \\( 10^{5}-10^{6} \\)yr for the same parameters as before, where the shorter lifetimes correspond to shorter viscous evolution timescales of high viscosities. The disk lifetimes are shorter by roughly two orders of magnitude when compared to the models with only photoevaporation from the central source. The short disk lifetimes are due to the strong influence of the external radiation field at the outer disk, where most of the disk mass initially resides.  The disk evolution is very different at larger, more typical distances (\\( d\\gg 0.03 \\)pc) from the external stars due to the different locations where photoevaporation is efficient. The disk material in the vicinity of the EUV gravitational radius is removed by photoevaporation from the central source. Similarly, the disk material in the neighborhood of the FUV gravitational radius is removed by external FUV photoevaporation. The effect of external photoevaporation is initially very dominant; quickly reducing disk size to the FUV gravitational radius. After this disk truncation, the disk mass is removed at two locations. At the inner edge, disk material is viscously accreted onto the central. On the other hand, viscous diffusion spreads the disk material beyond the FUV gravitational radius, and FUV photoevaporation removes the disk material at this location. The combination of this mechanisms decreases the disk surface density. At the last stages of the disk evolution, when the surface density is very low (\\( \\Sigma \\sim 10^{-4}\\textrm{g cm}^{-2} \\)), the ionizing from the central source opens a gap in the disk. In this case, it is possible to remove the inner annulus faster than the outer one self-consistently. The disk lifetime is in the range \\( 10^{6}-10^{7} \\)yr for the same parameters as before, where the shorter disk lifetimes correspond to the shorter viscous evolution timescales of high viscosities.  Observations suggest that the timescale for all the stars in young clusters to lose their disks is \\( \\sim 6\\times 10^{6} \\)yr \\citep{lifetime=6Myr}, in agreement with the disk lifetimes for the proplyds nearby external hot stars (\\( 10^{5}-10^{6}\\textrm{yr} \\)). The disk removal timescale is an important constraint on the timescale allowed for planet formation. It is very attractive to study the possibility of planet formation in star forming regions since this process follows star and disk formation. If the influence of massive external stars in star forming regions is not demolishing for planet formation, star forming regions are natural birthplaces of extrasolar planets. Numerical simulations suggest a timescale for terrestrial planet formation in the range \\( 10^{7}-10^{8} \\)yr \\citep{planet formation terrestrial planets 8, planet formation 20 Myr for terrestrial planets}, and a typical timescale for giant planet formation that range from a few million years to more than \\( 10^{7} \\)yr \\citep{giant planet formation, giant planet formation > 8}. It is thus not possible to form planets around stars in the neighborhood of massive O stars (i.e. EUV external photoevaporation) because the disk lifetime is too short (\\( 10^{5}-10^{6}\\textrm{yr} \\)). On the other hand, at typical distances from the external stars (\\( d\\gg 0.03 \\)pc), the disk lifetimes are long enough (\\( 10^{6}-10^{7}\\textrm{yr} \\)) to allow for the formation of terrestrial and giant planets. There are no constraints on planet formation in the absence of EUV or FUV fluxes from external stars."
    },
    "0209/astro-ph0209517_arXiv.txt": {
        "abstract": "The rich wealth of observational data, and matching theoretical investigations, of the transiting planet of HD~209458 stands in sharp contrast to systems for which only the radial velocity orbit is known.  In this paper, I summarize the current status of these observations, and motivate a variety of projects that should be accessible with existing instruments. I describe observational estimates of the planetary radius, and discuss the relevant sources of uncertainty.  I compare these estimates to those based on theoretical structural models. This discussion motivates the observational pursuit of three quantities that could be derived from measurements of the secondary eclipse:  These are the albedo, the temperature, and the orbital eccentricity.  I review the recent detection of the sodium D lines in the planetary atmosphere, and discuss ongoing work to search for molecular features in the near infrared.  I also outline the use of the Rossiter effect to study the alignment of the orbit with the stellar equatorial plane, and transit timing to search for additional objects in the system. ",
        "introduction": "We are fortunate to know of a transiting extrasolar planet of a nearby, relatively bright star.  Although the method of photometric transits as a detection technique is still very much under development, its application to study the planet of HD~209458 has proved immensely successful: HD~209458~b is the only extrasolar planet for which reliable estimates of the radius and mass are available. The calculated density proves that the planet is indeed a gas giant, with a composition primarily of hydrogen and helium.  In the 2.5 years since its discovery, numerous observers have seized upon the opportunities afforded by an extrasolar planet that periodically transits its parent star.  My goal in this contribution is both to detail the current state-of-the-art in such observations, and outline a series of projects that, with care, should be accessible to current instruments. This discussion serves to motivate wide-field surveys for bright transiting extrasolar planet systems (see Borucki et al. 2001, Brown \\& Charbonneau 2000, and numerous contributions in this volume).  It is only for such objects that some of the projects below will be permitted in the near future. ",
        "conclusions": ""
    },
    "0209/astro-ph0209047_arXiv.txt": {
        "abstract": "In spite of great progress over the last $\\sim 10$ years, especially thanks to HST, a number of exciting open problem still puzzle astronomers working on globular clusters in our own and other galaxies. These problems range from determining more accurate ages to assess whether massive black holes hide at the center of some clusters, from identifying the physical origin of red giant winds to assess whether there is an influence of cluster structure and dynamics on the evolution of individual cluster stars, from demonstrating whether or not some clusters possess a dark matter halo on their own to eventually understand the formation of globular clusters in the context of galaxy formation, and more. In this review, I briefly sketch how ground based telescopes and their instrumentation can help solving these problems through the present decade, 2001-2010. A glimpse to the next decade is also given. ",
        "introduction": "This conference demonstrates the wide, enduring scientific interest of globular clusters (GC) for a broad variety of astrophysical and cosmological issues. This is indeed a very active field of astronomical research, and as such it has a number of open problems, currently under investigation. In this context, I have been asked by the organizers to review the perspectives for such problems to be solved (or at least effectively attacked) in the near future using ground based facilities, with emphasis on 8-10m class telescopes. I will do so leaving implicit that space borne facilities will widely complement in several areas, suffice to mention here the enormous progress in GC research that has been achieved using HST and its instruments. I should also acknowledge that this review is going to be somewhat biased towards the ESO Very Large Telescope (VLT), because it is the ground based facility I am more familiar with. Finally, emphasis will especially be on the telescopes and instruments that will be available in the course of the present decade.  In the next Section a list of open problems in GC research is presented, while in the following Section 3 some of the tools that may solve them are briefly mentioned and commented. ",
        "conclusions": ""
    },
    "0209/astro-ph0209271_arXiv.txt": {
        "abstract": "An ideal coronagraph with a band-limited image mask can efficiently image off-axis sources while removing identically all of the light from an on-axis source. However, strict mask construction tolerances limit the utility of this technique for directly imaging extrasolar terrestrial planets.  We present a variation on the basic band-limited mask design---a family of ``notch filter'' masks---that mitigates this problem.  These robust and trivially achromatic masks can be easily manufactured by cutting holes in opaque material. ",
        "introduction": "Direct optical imaging of nearby stars has emerged as a potentially viable method for detecting extrasolar terrestrial planets, buoyed by new techniques for controlling diffracted and scattered light in high-dynamic-range space telescopes (see, e.g., the review by Kuchner \\& Spergel 2003). These techniques boost a telescope's ability to separate a planet's light from the light of its host star. At optical wavelengths, the Sun outshines the Earth by a factor of nearly $10^{10}$; this contrast ratio is $\\sim 10^{3}$ times larger than the contrast ratio in the mid-infrared \\citep{tpf,desm01}.  But to offset the higher dynamic range requirements of visible-light planet finding, optical techniques offer freedom from large, multiple-telescope arrays \\citep{wool03}, cryogenic optics, and background light from zodiacal and exozodiacal dust \\citep{kuch00}, while providing access to $O_2$ and $O_3$ biomarkers \\citep{trau01,desm01}, surface features \\citep{ford01}, the total atmospheric column density \\citep{trau03}, and even potentially the ``red edge'' signal from terrestrial vegetation \\citep{wool02}.  Of the obstacles to achieving the necessary dynamic range in a single-dish optical telescope, the diffracted light background appears relatively manageable. For example, maintaining the scattered light background at the level of the expected signal from the planet poses a greater challenge; this task requires a r.m.s. wavefront accuracy of $\\lesssim 1$~\\AA~\\citep{kuch02, trau02a} over the critical spatial frequencies. However, techniques for managing the diffracted light may dictate the general design of a planet-finding telescope and the planet search and characterization strategy.  Optical techniques for controlling diffracted light in planet-imaging telescopes have centered on two main designs: specially shaped and/or apodized pupils \\citep{sper01,nise01,kasd01,debe02,kasd03} and classical coronagraphs \\citep{lyot39, naka94, stah95, malb95, kuch02}. Shaped and apodized pupils produce a point spread function whose diffraction wings are suppressed in some regions of the image plane.  A classical coronagraph explicitly removes the on-axis light from the optical train by reflecting or absorbing most of it with an image mask and diffracting the remainder onto an opaque Lyot stop.  Recently, \\citet{kuch02} showed that a classical coronagraph performs best with a ``band-limited'' image mask. Different band-limited masks offer high performance for planet searching or planet characterization.  For planet characterization, the $\\sin^2$ amplitude transmissivity mask ($\\sin^4$ intensity transmissivity) introduced in \\citet{kuch02} can achieve 80\\% throughput for a planet at 4$\\lambda/D$.  With this high throughput, a 10 m by 4 m telescope can detect a planetary biomarker in $\\sim$1/3 of the time needed by alternative designs (e.g., an 8 m square apodized aperture).  A band-limited mask of the form $1-{\\rm sinc}$ (see Table~1) has both excellent throughput and large search area.  With any band-limited mask, an ideal coronagraph eliminates identically all of the on-axis light, though pointing errors and the stellar size contribute to a finite leakage \\citep{kuch02}.  A band-limited mask can operate with a pupil of any shape as long as it has uniform transmissivity.  But because they interact with focused starlight, all coronagraphic image masks face severe construction tolerances.  Errors in the mask intensity transmissivity of $\\sim 10^{-9}$ on scales of $\\lambda/D$ near the center of the mask can scatter enough light into the field of view to scuttle a planet search \\citep{kuch02}. Painting a graded-transmissivity mask requires a steady hand! This requirement has cast the classical coronagraph in an unfavorable light, despite its potential high performance and flexibility.  In this paper, we offer a way around this pitfall of classical coronagraphy: an easy-to-manufacture class of image masks.   We illustrate a family of binary image masks which offer a savings in construction tolerances of $\\sim5$ orders of magnitude compared to graded image masks, analogous to the advantage of using binary rather than graded pupil masks \\citep{sper01}. These ``notch filter'' masks offer the same planet search and characterization advantages as ideal band-limited masks, providing a robust, practical means of controlling diffracted light in a planet-finding coronagraph. ",
        "conclusions": "We have illustrated the use of notch filter functions to generate several kinds of image masks which should be relatively easy to manufacture.  We showed graded masks whose transmissivities are everywhere greater than zero.  We showed binary image masks, which can be cut or shaped from pieces of opaque material. These binary masks can be manufactured to the tolerances necessary for terrestrial planet finding using standard nanofabrication techniques, and can potentially be made self-supporting.  Our simulations of the performance of a coronagraph outfitted with a binary notch filter mask suggest that this technique could reveal extrasolar planets similar in brightness to the Earth around nearby stars, given foreseeable improvements in wavefront control on a highly stable space platform.  Binary notch filter masks combine many of the advantages of binary pupil masks (ease of manufacture, achromaticity, robustness) with the advantages of band-limited image masks (large search area, and small inner working distance). Using binary pupil or image masks seems to inevitably require stacking many copies of the same basic aperture shape; \\citet{kasd01} used this principle to generate binary pupil masks; we have used it to generate binary image masks.  In \\citet{kasd01}, the high-spatial frequency artifacts of this stacking procedure appear in the image plane directed away from a search sector. In notch filter masks, the high-spatial frequency artifacts are directed into the Lyot stop.  Ultimately, a space telescope for direct optical imaging of extrasolar planets may incorporate more than one diffracted-light management strategy. Having a choice of different techniques available will allow a mission to adapt to changing observing needs as our understanding of high-contrast space telescopes improves and the phenomenology of extrasolar planets unfolds."
    },
    "0209/astro-ph0209101_arXiv.txt": {
        "abstract": "Accretion disks and nuclear shell burning are present in some symbiotic stars (SS) and probably all supersoft X-ray binaries (SSXBs).  Both the disk and burning shell may be involved in the production of dramatic outbursts and, in some cases, collimated jets. A strong magnetic field may also affect the accretion flow and activity in some systems.  Rapid-variability studies can probe the interesting region close to the accreting white dwarf (WD) in both SS and SSXBs.  I describe fast photometric observations of several individual systems in detail, and review the results of a photometric variability survey of 35 SS.  These timing studies reveal the first clearly magnetic SS (Z And), and suggest that an accretion disk is involved in jet production in CH Cyg as well as in the outbursts of both CH Cyg and Z And.  They also support the notion that the fundamental power source in most SS is nuclear burning on the surface of a WD, and raise questions about the structure of disks in the SSXBs.  Finally, spectroscopic observations of RS Oph reveal minute-time-scale line-strength variations, probably due to a hot boundary layer.  Taken together, the rapid timing observations explore the connections between jet-producing WDs and X-ray binaries, as well as SS, SSXBs, and CVs. ",
        "introduction": "Accretion onto a white dwarf (WD) and nuclear shell burning probably power the observed activity in supersoft X-ray binaries (SSXBs) and many symbiotic stars (SS).  Besides the hallmark high-excitation-state emission lines (for SS) and soft X-ray spectra (for SSXBs), this activity can include outbursts and collimated jets.  But in SS, emission and absorption by the nebula can hide the spectroscopic signatures of a disk and absorb the soft X-rays from the nuclear-burning shell.  In SSXBs, the disk is optically visible, but the optical emission is dominated by reflected nuclear-burning light, and the disk is significantly heated (e.g., Popham \\& Di Stefano 1996).  One way to examine the region close to the accreting WD despite these complications is to look for rapid variations.  In the context of this work, `rapid' or `fast' variations are those which could be associated with WD-disk phenomena.  Stochastic variations from a disk, termed ``flickering'', generally occur on either a dynamical time \\begin{eqnarray*} t_{dyn} & \\sim & \\frac{1}{\\Omega_k} \\sim 4\\,{\\rm s}\\; \\left( \\frac{r}{10^{9} {\\rm cm}} \\right)^{3/2} \\left( \\frac{M_{WD}}{0.6\\,M_{\\odot}} \\right)^{-1/2} \\end{eqnarray*} (where $\\Omega_k = \\Omega_k(r)$ is the Keplerian angular velocity, $r$ is the radial position in the disk, and $M_{WD}$ is the mass of the WD), or a viscous time \\begin{displaymath} t_{visc}  \\sim  \\frac{1}{\\alpha} \\left( \\frac{r}{H} \\right)^2 \\frac{1}{\\Omega_k} \\sim  8\\, {\\rm hr}\\; \\left( \\frac{\\alpha}{0.1} \\right)^{-4/5} \\left( \\frac{\\dot{M}}{10^{-8} M_{\\odot} \\ {\\rm yr^{-1}}} \\right) ^{-3/10} \\left( \\frac{r}{10^{9} {\\rm cm}} \\right)^{5/4} \\\\ \\end{displaymath} (where $H$ is the disk height, $\\alpha$ is the viscosity parameter, and $\\dot{M}$ is the accretion rate onto the WD; Frank, King, and Raine 1992).  Disk flickering therefore depends on the size and properties of the disk, and the location of the emitting region within the disk.  In WD disks, fluctuations occur with time scales of roughly a day or less, and observations verify that the fastest stochastic variations come from the inner disk (e.g., Bruch 2000).  Brightness oscillations due to magnetic accretion should also have time scales ranging from minutes to hours.  When a WD is in spin equilibrium (neither being spun up nor spun down by torques from the disk material), it rotates with the Keplerian frequency at the radius where the ram pressure from in-falling material balances the pressure in the magnetic field.  For fields typically measured in intermediate polars (IPs; $B_S \\sim 10^5 - 10^6$ G) and accretion rates that are reasonable for SS or SSXBs ($\\dot{M} \\sim 10^{-9} - 10^{-7} {M}_{\\odot} \\ {\\rm yr^{-1}}$), the equilibrium spin periods are tens of minutes.   Changes in the luminosity from a burning shell occur on the nuclear time, $t_{nuc}$, which is roughly the time to accrete the envelope scaled by the fraction of the envelope that must be burned to heat the entire envelope to temperature $T$: \\begin{eqnarray*} t_{nuc} & \\sim & \\left( \\frac{C_P T}{E_{nuc}} \\right) \\left( \\frac{\\Delta M}{\\dot{M}} \\right) \\\\ & \\sim &  3\\, {\\rm yr}\\; \\left( \\frac{T}{3\\times 10^7 {\\rm K}} \\right) \\left( \\frac{\\Delta M}{6 \\times 10^{-5} M_{\\odot}} \\right) \\left( \\frac{\\dot{M}}{4\\times 10^{-8} M_{\\odot} \\ {\\rm yr^{-1}}} \\right)^{-1}, \\\\ \\end{eqnarray*} where $C_P$ is the specific heat at constant pressure, $E_{nuc}$ is the energy released per gram from nuclear burning of H, and $\\Delta M$ is the mass of the envelope, which is dependent on $M_{WD}$ and $\\dot{M}$ (Fujimoto 1982). Therefore, fast variations cannot be due to fundamental changes in the nuclear burning emission (changes due to obscuration or reflection of this emission, however, could happen quickly).  Even though the nuclear burning luminosity cannot change quickly, it is still an important consideration in rapid-variability studies. Since $E_{nuc}/E_{grav} \\approx 40$ (where $E_{grav}$ is the energy released per gram from accretion onto the WD), quasi-steady nuclear shell burning dominates the energetics of the hot component when present.  Nuclear burning can therefore hide or diminish rapid variations that are directly associated with accretion. The maximum and minimum accretion rates to produce quasi-steady shell burning are \\begin{eqnarray*} \\dot{M}_{steady,max} &  = & 2.8\\times 10^{-7} + \\; 5.9 \\times 10^{-7} \\left( \\frac{M_{WD}}{M_{\\odot}} -1.0 \\right) M_{\\odot} \\ {\\rm yr^{-1}} \\\\ \\dot{M}_{steady,min} & = & 1.32 \\times 10^{-7} M_{WD}^{3.57} M_{\\odot} \\ {\\rm yr^{-1}}. \\end{eqnarray*} (Paczy{\\'n}ski \\& Rudak 1980; Iben 1982).  If $\\dot{M} > \\dot{M}_{steady,max}$, the fuel cannot be burned as fast as it is accreted, and the envelope could expand or be ejected.  If $\\dot{M} < \\dot{M}_{steady,min}$, material burns unstably in nova explosions, although a period of residual burning may follow.  If the hot-component luminosity ($L_{hot}$) is greater than approximately $100\\,L_{\\odot}$, significant nuclear burning must be present, since such high luminosities cannot be produced by accretion with $\\dot{M} < \\dot{M}_{steady,min}$.  One final time scale of interest is the recombination time in a SS nebula, \\begin{eqnarray*} t_{rec} & \\sim & \\frac{1}{n_s \\alpha_B} \\sim 1 {\\rm hr}\\; \\left( \\frac{n_s}{10^{9}\\, {\\rm cm}^{-3}} \\right)^{-1} \\left( \\frac{\\alpha_B}{2.59 \\times 10^{-13}\\, {\\rm cm}^3/{\\rm s}} \\right)^{-1}\\;\\;\\; {\\rm at}\\; 10^4\\,{\\rm K} \\end{eqnarray*} (where $n_s$ is the density at the outer edge of the ionized region and $\\alpha_B$ is the case-B recombination coefficient; Fern{\\'a}ndez-Castro et al. 1995).  Any high-energy (far-UV or soft X-ray) variations faster than $t_{rec}$ will be smeared out when the nebula reprocesses them into the optical.  In \\S2, I discuss the first known case of a WD with both magnetically channeled accretion and surface nuclear burning (Z And).  I describe photometric evidence for changes in an accretion disk associated with the production of a jet in a WD accretor (CH Cyg) in \\S3, and discuss how high-time-resolution studies can reveal the relative importance of nuclear burning vs. viscous dissipation in \\S4.  Disk flickering from the supersoft source MR Vel is described in \\S5, and minute-time-scale spectral line variability that can act as a diagnostic for the physical conditions in the line-emitting regions (RS Oph) is shown in \\S6.  I discuss the overall results from these variability studies in \\S7. ",
        "conclusions": "Rapid-variability studies suggest that the strength of aperiodic variations in SS is related to the power source (nuclear shell burning or accretion alone).  Systems with high $L_{hot}$ generally do not show large-amplitude flickering, whereas SS with low $L_{hot}$ almost always do (Walker 1977; Dobrzycka et al. 1996; SBH and references therein).  The luminosity ratio of a typical high-$L_{hot}$ system ($\\sim100 - 1000\\, L_{\\odot}$) to a typical low-$L_{hot}$ system ($\\sim 1 - 10\\, {L}_{\\odot}$) is close to the ratio of the energy released per nucleon in the nuclear burning of hydrogen-rich material to that from accretion onto a WD.  Therefore, if nebular emission in a symbiotic is powered by quasi-steady nuclear shell burning on the surface of a WD, flickering or oscillations from accretion are often hidden or reduced.  In many SSXBs, on the other hand, nuclear-burning emission is probably reprocessed into the optical by the accretion disk, and the ratio of reprocessed light to direct disk emission may be low enough that some CV-like disk flickering is detectable.  Since symbiotic recurrent novae are preferentially low-$L_{hot}$ systems, the presence of flickering may also be related to the type of outburst a SS experiences.  Whether a symbiotic burns material quasi-steadily or not, observations described in \\S2 and \\S3 suggest that accretion-disk instabilities may play a role in the more common, classical SS outbursts.  Furthermore, Miko{\\l}ajewska (this volume) found that SS with ellipsoidal variations (in which the red giant is closer to filling its Roche lobe, and a disk is more likely to form due to focusing of the red giant wind) have more outburst activity.  So the presence of disks could be broadly associated with outbursts in classical SS.  In CH Cyg and also possibly in Z And (Brocksopp et al. 2003), collimated jets are sometimes produced during or after outbursts, so disks may also be related to the production of jets in SS.  As discussed in \\S3, there is evidence that the disk in CH Cyg was disrupted when a jet was produced.  Similar behavior has been reported for some transient-jet X-ray binaries, so, as suggested by Zamanov \\& Marziani (2002), disks and jets may provide a link between symbiotic and black-hole jet sources.  Finally, periodic variations provide information about magnetism. Given the very low oscillation amplitude in Z And, however, SBH could not rule out strong magnetic fields in any of their survey objects. The detection fraction for magnetic WDs of 3\\% is therefore only a crude lower limit. More sensitive observations are needed to determine the magnetic fraction in SS, and thereby test theories of the origin of magnetism in WDs and binary stellar evolution.  Identification of additional magnetic SS is also needed to clarify whether a strong WD field helps produce collimated jets (as suggested by Panferov \\& Miko{\\l}ajewski 2000 and references therein; Tomov, this volume) or inhibits their formation by truncating the inner accretion disk where the jet is launched (as may be the case in NS X-ray binaries; Fender \\& Hendry 2000).  Finally, since magnetically channeled accretion with $\\dot{M} \\ga \\dot{M}_{steady,min}$ produces a large soft X-ray spin modulation, whereas residual burning on a magnetic WD does not (King et al. 2002), comparison between the X-ray oscillation amplitudes in magnetic SS and magnetic SSXBs may provide information about the different (or similar) causes of nuclear shell burning in these two classes of systems."
    },
    "0209/astro-ph0209337_arXiv.txt": {
        "abstract": "We introduce the notion of the cosmic numbers of a cosmological model, and discuss how they can be used to naturally classify models according to their ability to solve some of the problems of the standard cosmological model. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209227_arXiv.txt": {
        "abstract": "High-resolution atmospheric flow simulations of the tidally-locked extrasolar giant planet, \\HDblah, show large-scale spatio-temporal variability.  This is in contrast to the simple, permanent day/night (i.e., hot/cold) picture.  The planet's global circulation is characterized by a polar vortex in motion around each pole and a banded structure corresponding to $\\sim$3 broad zonal (east-west) jets.  For very strong jets, the circulation-induced temperature difference between moving hot and cold regions can reach up to $\\sim$1000~K, suggesting that atmospheric variability could be observed in the planet's spectral and photometric signatures. ",
        "introduction": "Roughly 100 gaseous giant planets are currently known to orbit nearby sun-like stars\\footnote[9]{see, e.g., {\\tt http://www.obspm.fr/encycl/encycl.html} and {\\tt http://exoplanets.org/almanacframe.html}}.  Many of those planets are located at very small orbital distances from their parent stars, where tidal forces are thought to maintain a rotation rate synchronous with the orbit---thus producing permanent day and night sides on the planet.  This situation presents a new r\\'egime of atmospheric circulation, not encountered in our solar system: slowly-rotating giant planets, which are continuously exposed to intense stellar heating on the same side.  Measuring the resulting temperature structure on these planets is a major goal of current and future observational programs.  In this {\\it Letter}, we report on high-resolution, fully-turbulent global simulations of the atmospheric flow on \\HDblah, presently the only close-in extrasolar giant planet (CEGP) with a measured mass ($M_p$) and radius ($R_p$).  The parent star of \\HDblah\\ shows discernible brightness decrements every 3.5~days, due to occultations by the planet as it transits across the star.  This property has recently led to precise measurements of $M_p$ and $R_p$ (Charbonneau et al. 2000; Henry et al. 2000; Mazeh et al. 2000), confirming its giant nature.  It has also allowed the detection of sodium absorption, providing the first probe of the planet's atmosphere (Charbonneau et al. 2002).  According to the standard planetary formation picture, \\HDblah\\ is expected to have formed at a large distance ($>$~1~AU) from its parent star (see, e.g., Boss 1996) and migrated inward (Goldreich \\& Tremaine 1980; Lin et al. 1996; Murray et al. 1998), quickly ($\\simlt$~10 Myr) reaching its present distance of only 0.046~AU from the star (see e.g., Burrows et al. 2000).  There, it was forced by tidal effects to permanently present the same face to its star (see, e.g., Goldreich \\& Soter 1966)---as the Moon does to the Earth.  From this synchronization, the rotation period of the CEGP is known (same as its orbital period of 3.5 days).  However, unlike our Moon with its insignificant atmosphere, \\HDblah\\ is expected to possess vigorous meteorology and associated horizontal transport of heat and chemical species, due to the presence of a thin, stable (radiative) atmospheric region above the convective interior (Guillot et al. 1996; Seager \\& Sasselov 1998). ",
        "conclusions": "The presence of high-contrast hot and cold spots on \\HDblah\\ induces spatio-temporal variability, which may produce detectable fluctuations in observational signatures.  Sensitive enough infrared flux measurements (e.g., such as those possible with SIRTF) could reveal variability in time during an orbit of the planet as different faces are seen from Earth, as well as from orbit to orbit as the spots (polar vortices) revolve about the rotation poles.  The lack of detectable variability would thus point toward either an obscuring uniform haze overlying the modeled region or an inefficient conversion of stellar irradiation to atmospheric kinetic energy.  In the latter case, the formed spots are weak (i.e., small temperature/thickness perturbation) and thermal forcing which produces bulging in the modeled layer of more than several percent overwhelms any temperature variability due to atmospheric motion.  The equilibrium day/night temperature is then robustly maintained.  The extreme conditions inside the spots in our calculations suggest several additional potential observables since absorption levels, albedo, intrinsic thermal emission, and presence, type, or height of clouds could all be different within the different spots.  The spatially integrated spectrum, therefore, can be different from the uniform planet case.  For example, enhanced levels of CH$_4$/CO abundance inside the cold spots are possible, given the value of \\Tatm\\ used (Seager et al. 2000).  Similarly, condensates, such as MgSiO$_3$ (enstatite), may also be found inside the diffuse hot spots, where temperatures may be high enough (Sudarsky et al. 2000; Seager et al. 2000).  If the temperature at radiative equilibrium is actually higher than assumed (e.g., $\\sim$2000~K), the opposite situation may occur---MgSiO$_3$ may be found in the cold spots.  In addition, by sequestering chemically active species and periodically exposing them to the stellar irradiation (as the spots revolve around the poles), the spots could also affect atomic number densities and could be part of the explanation for the recently observed low abundance of Na~I on \\HDblah\\ (Charbonneau et al.  2002)."
    },
    "0209/astro-ph0209157_arXiv.txt": {
        "abstract": "An update of the set of low surface brightness galaxies is presented which can be used to set constraints on the otherwise ambiguous decompositions of their rotation curves into contributions due to the various components of the galaxies. The selected galaxies show all clear spiral structure and arguments of density wave theory of galactic spiral arms are used to estimate the masses of the galactic disks. Again these estimates seem to indicate that the disks of low surface brightness galaxies might be much more massive than currently thought. This puzzling result contradicts stellar population synthesis models. This would mean also that low surface brightness galaxies are not dominated by dark matter in their inner parts. ",
        "introduction": "In a previous paper (Fuchs 2002) I have described how arguments of density wave theory of galactic spiral arms can be used to set constraints on the otherwise ambiguous decomposition of the rotation curves of low surface brightness galaxies (LSBGs). For this purpose galaxies were selected which show clear spiral structure. These came mainly from the set of LSBGs for which high--resolution rotation curves have been published by McGaugh et al.~(2001). The same authors (de Blok et al.~2001) have also constructed dynamical models of the galaxies. The observed rotation curves were modeled as \\begin{equation} v_c^2(R)=v_{\\rm c, bulge}^2(R)+v_{\\rm c, disk}^2(R)+v_{\\rm c, is\\,gas}^2(R) +v_{\\rm c, halo}^2(R)\\,, \\end{equation} where $v_{\\rm c, bulge}$, $v_{\\rm c, disk}$, $v_{\\rm c, is\\,gas}$, and $v_{\\rm c, halo}$ denote the contributions due to the bulge, the stellar disk, the interstellar gas, and the dark halo, respectively. De Blok et al.~(2001) provide actually for each galaxy several models, one with zero bulge and disk mass, one model with a `reasonable' mass--to--light ratio, and a `maximum--disk' model with bulge and disk masses at the maximum allowed by the data. All fit the data equally well. Applying the density wave theory argument I confirmed essentially the maximum--disk models. This result is puzzling because the mass--to--light ratios of these models are unaccountably high in view of stellar population synthesis modeling of LSBGs (cf.~Bell \\& de Jong 2001). On the other hand, this might indicate that LSBGs are less dark matter dominated than currently thought.  At the time of writing of their paper there was no surface photometry available for some of the LSBGs in the set of McGaugh et al.~(2001), so that no dynamical models could be constructed. I have inspected the images of these galaxies and found four galaxies, ESO\\,14--40, ESO\\,206--140, ESO\\,301--120, and ESO\\,425--180, which can be used for the present purpose as well (cf.~Fig.~1), and contrived to obtain surface photometry of the galaxies. \\begin{figure} \\centerline{\\includegraphics[width=6cm]{bfuchs_fig1.ps}} \\caption{Image of ESO\\,206--140 reproduced from Beijersbergen et al.~(1999).} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209407_arXiv.txt": {
        "abstract": "Formation scenarios for polar ring galaxies are studied through N-body simulations that are compared with existing observations. It is shown that polar rings are likely to be formed by tidal accretion of the polar material from a gas rich donor galaxy. The distribution of dark matter in polar ring galaxies is studied: dark halos seem to be flattened towards the polar rings. ",
        "introduction": "Polar ring galaxies are peculiar systems that show two nearly orthogonal components. Two formation scenarios for polar rings have been proposed: galaxy mergers (Bekki, 1998) and tidal accretion (Reshetnikov \\& Sotnikova, 1997). They are studied through numerical simulations including gas dynamics, star formation and stellar mass-loss, and compared to observations. ",
        "conclusions": ""
    },
    "0209/astro-ph0209588_arXiv.txt": {
        "abstract": "The {\\it Space Infrared Telescope Facility} (\\sirtf) will revolutionize the study of dust--obscured star formation in distant galaxies.  Although deep images from the Multiband Imaging Photometer for \\sirtf\\ (MIPS) will provide coverage at 24, 70, and 160\\micron, the bulk of MIPS--detected objects may only have accurate photometry in the shorter wave\\-length bands due to the confusion noise. Therefore, we have explored the potential for constraining the total infrared (IR) fluxes of distant galaxies with solely the 24\\micron\\ flux density, and for the combination of 24\\micron\\ and 70\\micron\\ data.  We also discuss the inherent systematic uncertainties in making these transitions.  Under the assumption that distant star-forming galaxies have IR spectral energy distributions (SEDs) that are represented somewhere in the local Universe, the 24\\micron\\ data (plus optical and X-ray data to allow redshift estimation and AGN rejection) constrains the total IR luminosity to within a factor of 2.5 for galaxies with $0.4 \\lsim z \\lsim 1.6$.  Incorporating the 70\\micron\\ data substantially improves this constraint by a factor $\\lsim 6$.  Lastly, we argue that if the shape of the IR SED is known (or well constrained; \\eg, because of high IR luminosity, or low ultraviolet/IR flux ratio), then the IR luminosity can be estimated with more certainty. ",
        "introduction": "The evolution of the global, volume-averaged star formation rate (SFR) is currently a topic of intense interest \\citep[see, e.g.,][]{mad96,ste99,yan99,haarsma00}.  Observationally, a galaxy's SFR must be inferred from the luminosity density at wave\\-lengths dominated by young stars, such as the ultraviolet (UV), nebular emission lines, mid--to--far infrared (IR), or radio \\citep[see, \\eg,][]{ken98a,con92}, and by making assumptions about the form of the IMF.  Because of dust, rest--frame UV/optical indicators often underestimate the intrinsic SFR, especially for more luminous galaxies \\citep[see, e.g.,][]{cal94,cal01,bel02b}.  Based on the UV/optical spectral energy distributions (SEDs) and far--IR emission of distant galaxies and QSOs, dust appears to be prevalent in  high--redshift objects \\citep[$z\\sim 2-5$, \\eg,][]{pet98,ade00,car00,pap01,cha02}. Dust reprocesses the energy absorbed from the UV/optical into the mid/far--IR, and thus galaxy IR luminosities allow one to `balance the energy budget' and paint a much more complete picture of star formation throughout cosmic history \\citep[see, e.g.,][]{san96,bla99,flo99}.  The {\\it Space Infrared Telescope Facility} (\\sirtf), to be launched in 2003,  offers to revolutionize our view at IR wave\\-lengths with unprecedented sensitivity and resolution.  Several cosmological surveys  are being planned to obtain 24, 70, and 160\\micron\\ data with the Multiband Imaging Photometer for \\sirtf\\ \\citep[MIPS; see, e.g.,][]{lon01,dic01,rie01,rie01a}.  The combination of these three bands will span the peak of the IR SEDs and provide a fairly robust tracer of the total IR emission \\citep[\\eg,][]{dal02}. However, MIPS observations will rapidly become confusion limited due to the increasing source density at faint fluxes (\\eg, Xu \\etal\\ 2001; Dole, Lagache, \\& Puget 2002), especially for the longer wave\\-length data, which have lower sensitivity and resolution (MIPS resolution is roughly proportional to the bandpass central wave\\-length).  Thus, full coverage of galaxies' IR SEDs with accurate photometry may only be possible for relatively nearby or bright objects.  In this {\\it Letter} we consider the uncertainties inherent in translating MIPS photometry into total IR fluxes in the case that object photometry is only available from the shorter wave\\-length bands.  The relation between mid-IR and far--IR is complex.  Thus we explore the connection between the 24\\micron\\ data and the total IR emission from both the observational (\\S2) and modeling (\\S3) perspective, with the assumption that local galaxy SEDs are representative of high--redshift analogs \\citep[which has not been robustly demonstrated, although see][]{ade00,elb02}.  In \\S4, we investigate improving these constraints using 70\\micron\\ data, the galaxy luminosity, and/or UV flux.  We summarize our results in \\S\\ref{sec:conc}.  Because extensive X-ray/optical/near-IR coverage will be available for the majority of deep MIPS survey regions (see references above), we assume here that galaxies with dominant AGN contributions can be rejected using the X-ray data, and that galaxies' redshifts (spectroscopic or photometric) will be known. ",
        "conclusions": "\\label{sec:conc}  \\sirtf\\ will greatly improve our understanding of the IR emission of (and hence the star--formation processes within) distant galaxies. Because deep surveys could uncover many star--forming galaxies in the shorter--wave\\-length MIPS data with no longer--wave\\-length counterparts, we have explored the efficacy of constraining the total IR galaxy emission using the data at 24\\micron\\ only and the combination of 24 and 70\\micron\\ (and under the assumption that ancillary optical and X-ray data are available to allow redshift estimation and AGN rejection).  Assuming that distant star-forming galaxies have IR SEDs that are represented somewhere in the local Universe, the 24\\micron\\ data should  constrain the integrated IR flux to within a factor of 2.5: $F(24\\micron) / F_\\mathrm{IR} \\simeq 5-25$\\% for galaxies with $0.4 \\lsim z \\lsim 1.6$.  Including MIPS 70\\micron\\ data, the IR luminosity can be estimated with considerably more certainty (a factor $\\lsim 6$ improvement over the uncertainties for using 24\\micron\\ data only).  Lastly, if one can assume that the shape of the galaxy SED is similar to local luminous IR galaxies ($L \\gsim 10^{11}\\, L_\\odot$; \\eg, because of high IR luminosity or low UV/IR flux ratio), then the constraints should improve.  Throughout this study we have made the explicit assumption that the relationships between the mid-- and total--IR emission and the shape of the IR SEDs observed locally apply to properties of high--redshift galaxies.  As a final caveat, we emphasize that for many reasons this may not be the case.  Indeed, the population of IR--emitting galaxies must undergo strong evolution in order to match the cosmic IR background \\citep{hau01,cha01,xu01,elb02,dol02}, and it is unclear how such evolution manifests itself. The mid--IR UIBs (\\eg, PAHs) and total IR emission relation at high redshift may be different due to, \\eg, a changing dust composition, or a significantly different dust heating from older stellar populations than that observed locally. Moreover, we have assumed that the shape of the IR SED as a function of IR luminosity for local galaxies applies to high--redshift analogs, which is poorly known \\citep[although not inconsistent with observations; see, \\eg,][]{ade00}.  Only with large samples of galaxies detected in all three MIPS bands, and with spectroscopic measurements from the \\sirtf\\ Infrared Spectrograph (IRS), will these assumptions be testable.  Once such constraints are established, the data should provide a prescription for relating the MIPS data to total IR fluxes.  Given an estimate  on the total IR luminosity, it is then possible to estimate SFR \\citep[modulo the usual sources of uncertainty; see, e.g.,][]{ken98a,bel02b}."
    },
    "0209/astro-ph0209280_arXiv.txt": {
        "abstract": "We report on the results from a {\\it Chandra} ACIS observation of the young, compact, supernova remnant N103B. The unprecedented spatial resolution of {\\it Chandra} reveals sub-arcsecond structure, both in the brightness and in spectral variations. Underlying these small-scale variations is a surprisingly simple radial structure in the equivalent widths of the strong Si and S emission lines. We investigate these radial variations through spatially resolved spectroscopy using a plane-parallel, non-equilibrium ionization model with multiple components. The majority of the emission arises from components with a temperature of 1 keV: a fully ionized hydrogen component; a high ionization timescale (n$_{e}$t$ > 10^{12}$ s$\\;$cm$^{-3}$) component containing Si, S, Ar, Ca, and Fe; and a low ionization timescale (n$_{e}$t$\\sim$10$^{11}$ s$\\;$cm$^{-3}$) O, Ne, and Mg component. To reproduce the strong Fe K$\\alpha$ line, it is necessary to include additional Fe in a hot ($> 2$ keV), low ionization (n$_{e}$t$\\sim$10$^{10.8}$ s$\\;$cm$^{-3}$) component. This hot Fe may be in the form of hot Fe bubbles, formed in the radioactive decay of clumps of $^{56}$Ni. We find no radial variation in the ionization timescales or temperatures of the various components. Rather, the Si and S equivalent widths increase at large radii because these lines, as well as those of Ar and Ca, are formed in a shell occupying the outer half of the remnant. A shell of hot Fe is located interior to this, but there is a large region of overlap between these two shells. In the inner 30\\% of the remnant, there is a core of cooler, 1 keV Fe. We find that the distribution of the ejecta and the yields of the intermediate mass species are consistent with model prediction for Type Ia events. ",
        "introduction": "N103B is one of the brightest radio and X-ray sources in the Large Magellanic Cloud (LMC). As a result, this young, compact, supernova remnant (SNR) has been selected for further study in many radio, X-ray, and optical surveys of the LMC SNRs \\citep{DM95,H95,rd90,W99} and the general properties of this object are quite well known. At radio and X-ray wavelengths, the emission arises from a region 30$''$ in diameter (or 7.3 pc, assuming a distance of 50 kpc to the LMC) with the emission in the western half being $\\sim 3$ times brighter than in the eastern half \\citep{DM95,W99}. A deep H$\\alpha$ image \\citep{W99,S98} reveals several bright clumps which are located in the vicinity of the bright radio and X-ray regions. In all three bands, a partial shell is plainly visible.  The remnant is located on the northeastern edge of an H~II region, approximately 40~pc from the young, rich star cluster NGC~1850. While SNRs in the Magellanic Clouds are commonly associated with H~II regions, no other known LMC remnant is associated with a young star cluster \\citep{Chu88}. For years, it was naturally assumed that the progenitor of N103B was a massive member of this cluster. However, the {\\it ASCA} spectrum \\citep{H95} of this remnant shows strong emission features from highly ionized Si, S, Ar, Ca, and Fe, while K-shell emission from O, Ne, and Mg are relatively weak; this spectrum is more consistent with the nucleosynthesis products of a Type Ia SN \\citep{N84,I99} than the core collapse of a massive star \\citep{T95}. On the other hand, the optical spectrum is not dominated by the strong Balmer lines often associated with the remnants of Type Ia events \\citep{Balmer82, H95}. Thus the classification of this remnant is still somewhat uncertain.  N103B is believed to be only $\\sim 1000-2000$ years old \\citep{H95}, and the X-ray emission is still dominated by the ejecta. The deep, high resolution ACIS observation provides an opportunity to study not only the abundances but also the distribution of the ejecta, and to compare them with models and observations of Type Ia and Type II remnants. There are several striking differences in the ejecta profiles of Type Ia and II SNRs, which we discuss below. By studying the ejecta in N103B, we have gain further information about its progenitor state. Moreover, assessing whether or not N103B is the result of a Type Ia explosion bears on the relative number of young Ia and non-Ia SNRs in the LMC which, though based on sparse statistics, may be anomalous \\citep{H95}  The ejecta in Type Ia SNRs may retain some of the initial stratification generated in the explosion. As predicted by \\citet{N84} and \\citet{I99}, Fe should initially remain in the interior, while Si, S, Ar, and Ca should appear primarily near the rim of the remnant. The {\\it ASCA} observations of Tycho's SNR, which is commonly believed to be the result of a Type Ia SN event, showed that the radial profile of the Fe K$\\alpha$ line peaks at a smaller radius than the profiles of the other emission lines and the continuum emission \\citep{H97}. Furthermore, \\citet{Hwang98} find that the Fe K$\\alpha$ line is quite strong compared to the Fe L-shell emission. To reproduce the correct ratio of Fe K to Fe L emission, a large amount of Fe must exist in a very hot ($\\ge 2$ keV) plasma with a low-ionization timescale which produces primarily Fe K-shell emission. The authors argue that this Fe is hotter than the rest of the ejecta and has a lower ionization timescale because it is confined to the interior of the remnant and has been more recently shocked.  By contrast, in several of the Type II SNRs in the Milky Way, for example Cas A \\citep {HCasA,Hwang00} and G292.0+1.8 \\citep{P02}, the abundance structure is quite complex and the original distribution is not readily apparent.  Equivalent width maps of G292.0+1.8 show a high degree of non-radial structure in which the abundances of the ejecta in the clumps of emission vary throughout the remnant \\citep{P02}.  In Cas A, the Fe ejecta are even found {\\it exterior} to the lighter elements, indicating that the ejecta have actually over-turned \\citep{HCasA}.  Utilizing the superb spatial resolution of {\\it Chandra}, we can observe directly whether the ejecta in N103B have retained their original radial stratification, or whether it has been destroyed. While the distribution of the ejecta will not likely provide a definitive classification, it is interesting to compare this remnant to other Type Ia and Type II remnants.  To study the distribution of the ejecta in this complex remnant, we use a combination of narrow band imaging and spatially resolved spectroscopy, both of which are available for the first time at the required angular resolution with {\\it Chandra}. In \\S2, we describe the data set and reduction.  The methods used to perform the narrow band imaging and spatially resolved spectroscopy, as well as the scientific results, are presented in \\S3. In \\S4, we propose a 3D model for the distribution of the ejecta, consider their origin, and estimate their masses.  Finally, in \\S5, we summarize our findings and suggest avenues for future work on this remnant.\\ ",
        "conclusions": "The spectrum of N103B obtained through this {\\it Chandra} observation is quite similar to the ASCA spectrum \\citep{H95}, showing strong K$\\alpha$ lines of Si,S, Ar, Ca and Fe, suggesting a Type Ia origin for the remnant. The {\\it Chandra} image of N103B reveals structure at the sub-arcsecond level. The bright western side of the remnant, seen in previous X-ray images, is composed of a series of bright knots and filaments. An X-ray color image reveals that the spectral characteristics also vary dramatically throughout the remnant.  We find that despite the complex spatial and spectral morphology suggested by the false-color and X-ray color images, there are striking radial trends in the equivalent widths of the strong Si and S emission lines, as revealed by narrow band imaging.  The equivalent widths remain fairly constant within the interior, then rise rapidly at a radius of 10$''$. To investigate the cause for the increase in equivalent width through spatially resolved spectroscopy, we divided the remnant into seven concentric rings, each with approximately 35,000 counts.  The data are well fit by a plane-parallel, non-equilibrium ionization model. The continuum emission arises primarily from a 1 keV hydrogen plasma.  An additional $\\sim$ 1 keV plasma with a high ionization timescale (n$_{e}$t$ > 10^{12}$ s$\\;$cm$^{-3}$) contains Si, S, Ar, Ca, and Fe. A hot ($> 2$ keV), low ionization (n$_{e}$t$\\sim$10$^{10.8}$ s$\\;$cm$^{-3}$) Fe plasma is required to produce the strong Fe K$\\alpha$ line. Finally, the O, Ne, and Mg are located in a plasma with an ionization timescale of n$_{e}$t$\\sim$10$^{11}$ s$\\;$cm$^{-3}$ and temperature of roughly $\\sim$ 1 keV. The components are referred to as the H, Si, hot Fe, and O components, respectively.  Using this spectral model, we have determined that there are no significant radial variations in the temperatures or ionization timescales of the components.  Instead, we find that the emission measures (EM = $n_{e}n_{i}V/4\\pi D^{2}$) of the species in the Si component increase radially, mimicking the radial profiles of the Si and S equivalent width images. An exception is an enhancement of Fe in the innermost extraction region. In contrast to the Si component, the hot Fe EM has a profile which drops rapidly at radii greater than 10$''$.  The EM variations in the Si and hot Fe components are well modeled by a simple three-zone model for the ejecta. In the interior of the remnant is a sphere of Fe with a temperature of 1 keV and a high ionization timescale, which occupies only the inner 3\\% of the remnant's volume. Exterior to this is a shell of hot Fe which is plausibly in the form of hot Fe bubbles.  Finally, surrounding the hot Fe is a shell of Si, S, Ar, Ca, and Fe. The Si and hot Fe components coexist for a large fraction of the remnant volume, implying that the difference in ionization timescale, $n_{e}$t, is due to a difference in electron density.  We have limited information about the location and origin of the H and O components. The EMs of these components show no radial trends, like those seen in the Si component.  Furthermore, these components vary in concert, suggesting that the two components are physically linked.  It is likely that these two components are associated with a clumpy foreground or background structure in the ISM which has been shocked by the remnant, rather than a shell of ejecta or swept-up material. It is clear that the O component does not occupy the same volume as the Si component, and that any comparison between these two components must be performed with care.  In particular global O, Si, and Fe abundances derived from integrated spectra of this remnant cannot be directly compared to nucleosynthesis models without first taking into account the different physical locations of the different components.  Finally, we estimate the masses of Si, S, Ar, Ca, and Fe and find that they are more consistent with the yields of a Type Ia SN than a Type II SN. In particular, the large mass of Fe (0.34 M$_{\\rm \\odot}$) suggests that a Type Ia origin for N103B is more likely. Further support for a Type Ia origin is the lack of an O-rich component of ejecta. Finally, the properties of N103B are strikingly similar to Tycho's remnant.  Both require a hot Fe component and show a radial segregation of the Fe and Si components of the ejecta. The results of this analysis indicate that the properties of N103B are consistent with a Type Ia origin. However, further work must be done, particularly to determine the location and origin of the O component, before eliminating a Type II origin.  Certainly, a more realistic 3D model of the ejecta is needed, which takes into account the initial structure of the remnant, the expansion, and potentially mixing between the layers. Additionally, throughout this paper, we have ignored the large asymmetry between the eastern and western halves of the remnant; to model this remnant more accurately, this must be taken into account. Using more sophisticated models for seven radial bins is pointless however, and we suggest that the analysis be improved by combining this {\\it ACIS} dataset with the 0$^{\\rm th}$ order {\\it Chandra} LETG grating data. With this larger dataset, the remnant could be sampled with finer radial bins and the differences between the eastern and western halves could be explored. Also, by using the information from the {\\it Chandra} and {\\it XMM-Newton} gratings observations, one could restrict the parameters of the O component more effectively, thereby reducing some of the uncertainties in this analysis.  Finally, this analysis has ignored the intriguing small-scale variations in brightness and color. An exploration of these may yield more clues to the origin and distribution of the O and H components. In particular, it important to determine whether these clumps have the same composition as the rest of the remnant, or whether they are dominated by emission lines or continuum.  Again, while several of the brighter clumps have 10,000 counts, the model we have proposed cannot be safely used unless a strong Fe K$\\alpha$ line is present to remove some of the confusion between the two different sources of Fe emission. Again, combining the 0$^{\\rm th}$ order {\\it Chandra} LETG grating and {\\it ACIS} datasets should improve matters greatly."
    },
    "0209/astro-ph0209249_arXiv.txt": {
        "abstract": "Interferometers offer multiple methods for studying microlensing events and determining the properties of the lenses.  We investigate the study of microlensing events with optical interferometers, focusing on narrow-angle astrometry, visibility, and closure phase.  After introducing the basics of microlensing and interferometry, we derive expressions for the signals in each of these three channels.  For various forecasts of the instrumental performance, we discuss which method provides the best means of measuring the lens angular Einstein radius $\\thetaE$, a prerequisite for determining the lens mass.  If the upcoming generation of large-aperture, AO-corrected long baseline interferometers (e.g.\\ VLTI, Keck, OHANA) perform as well as expected, $\\thetaE$ may be determined with signal-to-noise greater than 10 for all bright events.  We estimate that roughly a dozen events per year will be sufficiciently bright and have long enough durations to allow the measurement of the lens mass and distance from the ground. We also consider the prospects for a VLTI survey of all bright lensing events using a Fisher matrix analysis, and find that even without individual masses, interesting constraints may be placed on the bulge mass function, although large numbers of events would be required. ",
        "introduction": "Gravitational lensing has been used for nearly a decade to study faint, compact masses in our galaxy.  Although a large number of microlensing events have been detected, the lens masses and distances cannot (in most cases) be determined, meaning that only statistical constraints on the lensing population may be derived from lensing surveys \\citep[e.g.][]{bulge,lmc5}.  The determination of the lens mass and distance in individual events would be of great utility towards elucidating the nature of the microlenses.  An example of this is the claim by \\citet{mao} and \\citet{bennett} that three long-duration events are likely massive black holes, with $M\\sim10-30M_\\odot$.  Since individual masses could not be measured for these events, statistical arguments were employed to support the claim for large masses.  If confirmed, \\citet{agol} have argued that these black holes would represent a new, significant population of black holes roaming the Galactic disk.  The measurement of mass in microlensing events requires the determination of two quantities describing the event: (1) the lens-source relative parallax, $\\piE$, and (2) the angular Einstein radius, $\\thetaE$ \\citep{gould}.  Measurement of the parallax, $\\piE$, requires the observation of the lensing event from viewpoints separated by $\\gtrsim 1$ AU.  One way to do this is to observe the event simultaneously from the ground and from a satellite in solar orbit; the Space Interferometry Mission (SIM) is currently planned to do precisely this for a number of future microlensing events.  It is also possible to measure $\\piE$ for long-duration events, with event timescales $\\tE\\gtrsim$few months, using the Earth's motion around the Sun to provide a distant vantage point.  Since $\\sim15-20\\%$ of lensing events towards the bulge have durations ${\\hat t}=2\\tE$ longer than 100 days, with high quality photometry the parallax may be measured for a significant fraction of events.  The angular Einstein radius is extremely difficult to measure, since typical values are $\\thetaE\\sim0.5 (M/0.3M_\\odot)^{1/2}$ milliarcsecond (hereafter mas), defying resolution by even the largest telescopes.  In very rare cases \\citep[e.g.][]{an}, it is possible to measure $\\thetaE$ during a caustic crossing, however generically $\\thetaE$ cannot be resolved by any single-aperture instrument.  Resolution of sub-milliarcsecond scales typically requires long baseline interferometers, and future instruments like SIM should be able to determine $\\thetaE$ for many events \\citep{pac98,boden,gouldsalim} using narrow-angle astrometry. Unfortunately, SIM will not fly until later than 2009. Before SIM flies, a number of long baseline, highly sensitive interferometers will come online, for example the Keck Interferometer or the Very Large Telescope Interferometer (VLTI).   These ground-based interferometers can study microlensing events in several different ways, which we will consider in this paper. \\citet{boden} discussed how Keck and VLTI can measure $\\thetaE$ using narrow-angle astrometry. \\citet{delplancke} have pointed out that for massive microlenses, $\\thetaE$ becomes comparable to the resolution $\\lambda/B\\approx 5$ mas for $B=100$m, $\\lambda=2.2\\mu$m.  This allows the study of microlensing events via the partial resolution of the lensed images. One possible method, discussed by \\citet{delplancke}, is the measurement of the decrement in fringe visibility as the microlensed images become resolved.  In this paper, we also investigate the use of closure phase to determine the angular Einstein radius $\\thetaE$.  Closure phase is free of many of the calibration issues afflicting visibility amplitude, however other concerns do arise.  In the next section, we provide a review of the basics of microlensing.  The following section gives an introduction to interferometry and closure phase.  We then show how closure phase may be used to measure $\\thetaE$, and compare our method to other proposed techniques.  Since both Keck Interferometer and VLTI have already observed first fringes, this method promises to be an exciting technique for the determination of $\\thetaE$ and thereby the lens mass, in the next few years. ",
        "conclusions": "Interferometry is poised to revolutionize the study of microlensing events.  Until now, microlensing has suffered the difficulty that masses of individual lenses cannot be measured, severely limiting the information able to be extracted from lensing surveys.  The problem has been that the two quantities needed to measure mass and distance, the relative parallax $\\piE$ and the angular Einstein radius $\\thetaE$, are not regularly measured.  The parallax may be measured for long duration events with high quality photometry, however measurement of $\\thetaE$ requires resolution on the order of a milliarcsecond, necessitating interferometers. The upcoming Space Interferometry Mission (SIM) should measure masses and distances for a large sample of lenses, answering the question of the microlenses' nature.  Well in advance of SIM, however, ground-based interferometers can also provide useful measurements of lensing events.  As mentioned earlier, microlensed sources are generally much fainter than the typical sources studied by optical interferometers, meaning that large apertures ($\\sim 8$m) are required.  Both the Keck Interferometer and the VLTI can measure visibility amplitude for microlensing events using their largest apertures, but only VLTI can measure closure phase using 3 large apertures; the Keck would be required to employ one of the 1.8 m outrigger telescopes which collect considerably fewer photons.  The signal measured by interferometers is a function of the Einstein radius in units of the resolution, $\\thetaE B/\\lambda$.  Because of this, there is great advantage to go to shorter wavelengths.  However, shorter wavelengths require the use of adaptive optics (AO) systems.  Since AO makes the accurate calibration of visibility very difficult, but has a smaller impact on the calibration of closure phase, there are obvious advantages to using closure phase.  For a numerical example, a microlensing event with $\\thetaE=0.5$ mas could be observed at 10 $\\mu$m without AO, but $\\thetaE B/\\lambda\\sim0.02$ giving a visibility signal of $V\\approx0.99$, which would be extremely challenging to distinguish from a point source.  The same event observed at 2.2 $\\mu$m using AO has $\\thetaE B/\\lambda\\sim0.11$ giving a closure phase signal $\\phi_{123}\\approx 2.5^\\circ$ which can already be measured.  Only a small fraction of events are expected to be bright enough ($K\\lesssim 14$) to be observed interferometrically. However, certain fainter events may also be accessible to interferometers.  If a bright ($K<13$) star falls within the isoplanatic angle, then phase referencing may be employed to extend the coherence time significantly.  As noted earlier, for sites with small isoplanatic patches the probability of finding a suitable bright star is poor.  Additionally, this technique is quite complex and as yet unproven, however in principle this could allow the study of microlensing events as faint as 20th magnitude, reaching the bright end of LMC events.  Phase referencing must be employed to perform narrow angle astrometry; our results indicate that events for which phase referencing is possible may be more profitably studied with visibility or closure phase.  We expect that $\\sim15$ events every year will be bright enough ($K<14$) and have sufficiently long duration ($\\tE>50$ days) to permit the measurement of mass and distance.  We have shown that a fairly large fraction of events accessible to ground-based interferometers should allow measurement of $\\thetaE$ with high signal to noise. We also investigated the prospects for a massive follow-up campaign by VLTI, and found that statistical information on the $\\thetaE$ distribution, even without individual mass measurements, can allow constraints to be placed on lens properties like their mass function.  Even if our estimates turn out to be overly optimistic, interferometers will still be able to elucidate the nature of claimed black hole candidates \\citep[e.g.][]{mao,bennett}. \\citet{agol} have suggested that current microlensing data indicate the presence of a significant population of intermediate mass black holes roaming the Galactic disk; interferometers will be able to confirm or reject this possibility.  In this paper, we have focused on ground-based interferometers, however our results apply also to space-based interferometers like the Space Interferometry Mission (SIM).  SIM is primarily an astrometric instrument, however it can also measure fringe visibilities. Nominally, the target precision expected for SIM is 1\\% in $V^2$ (M.~Shao 2002, priv.\\ comm.). SIM's baseline is 10m, and typical wavelengths are $\\lambda\\approx0.6\\mu$m, giving resolution of about 12 mas.  Hence, SIM can determine $\\thetaE$ with SNR$>10$ from visibility alone, entirely independently of the astrometric determination, for events with $\\thetaE>0.44$ mas.  From Figure~\\ref{events} we see that this comprises a large fraction of the events. The visibility measurements come for free with the astrometric measurements, and should significantly increase the precision of SIM mass measurements, as long as effects such as crowding do not pose too great an obstacle.  In addition to measuring $\\thetaE$, SIM also determines $\\piE$, the lens parallax, by measuring the time of the peak of the photometric lightcurve.  Since the peak of the visibility signal coincides with the peak of the photometric signal, SIM visibility measurements could also be used to determine $\\piE$.  However, since the variation in $1-V^2$ is so shallow near the peak this method may not prove to be as precise as ordinary photometric parallax measurement.  One of the most (potentially) exciting prospects is a topic we have not discussed in this paper, binary microlensing. For a binary lens system, complicated caustic structures can arise, leading in favorable cases to the production of 5 images of the source.  For a spectacular example of this, see \\citet{an}.  During caustic crossings, the magnification can get exceptionally large, e.g. factors of 30, making these events bright enough to observe with interferometers.  The five images are currently unresolvable from each other, however VLTI and Keck offer the prospect of imaging the multi-image pattern.  With a multi-aperture system (required for closure phase), one obtains several visibility measurements and one closure phase at the same time, possibly allowing the reconstruction of complex events such as caustic crossings.  \\bigskip"
    },
    "0209/astro-ph0209555_arXiv.txt": {
        "abstract": "Sloan Digital Sky Survey data for the field of the cluster Pal\\,5 reveal the existence of a long massive stream of tidal debris spanning an arc of 10$^\\circ$ on the sky. Pal\\,5 thus provides an outstanding example for tidal disruption of globular clusters in the Milky Way. Radial velocities from VLT spectra show that Pal\\,5 has an extremely low velocity dispersion, in accordance with the very low mass derived from its total luminosity. ",
        "introduction": "Pal\\,5 is a sparse low-mass halo cluster with peculiar structure and stellar content. Using wide-field multicolor data from the Sloan Digital Sky Survey (SDSS, see York et al.\\ 2000) we recently found clear direct evidence for strong mass loss from Pal\\,5, showing that this cluster is in the process of being tidally disrupted (Odenkirchen et al.\\ 2001, Rockosi et al.\\ 2002). At the current stage the SDSS covers a 6$^\\circ$ to 8$^\\circ$ wide band across Pal\\,5 (see Fig.1). This enabled us to extend our search for tidal debris to larger distances from the cluster. The data were filtered by applying an optimized smooth color-magnitude dependent weight function. We thus found out that the tidal tails of Pal\\,5 extend over at least 10$^\\circ$ on the sky (Fig.1), corresponding to a length of 4 kpc in space. The leading tail (southwest of Pal\\,5) is visible over 3\\fdg5 down and most likely continues beyond the border of the field. The trailing tail (northeast of Pal,5) is traced out to 6\\fdg5 from the cluster. The stellar mass seen in the tails adds up to 1.2 times the mass of stars in the cluster. The location and curvature of the tails provide unique information on the local orbit of the cluster. The clumpiness of the stream suggests that the process of tidal mass loss has been episodic, probably triggered by disk shocks. The orbit and the mass and geometry of the tails yield an estimate of the mean mass loss rate of about 5~$M_\\odot$/Myr.   \\begin{figure}[h] \\plotfiddle{p5padua_fig1.ps}{6.8cm}{270}{50}{50}{-220}{235} \\caption{Map of the weighted stellar surface density showing the tails of Pal\\,5 (contours drawn at $1.5 \\sigma$, $2 \\sigma$ and $3 \\sigma$ and higher). The thick dashed line shows the best-fit orbit of the cluster. The feature at (230\\fdg6,+2\\fdg1) is due to the cluster M\\,5 and hence not related to Pal\\,5. } \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209625_arXiv.txt": {
        "abstract": "Collapsing collisionless particle systems form gravitational bound halos with cuspy density profiles. Also hierarchical merging of these systems produce remnants with cuspy central density profiles. These results lead to the assumption of cuspy NFW \\cite{Navarro:96} profiles for the density distribution in dark matter halos. However, observed rotation curves in disk galaxies suggest dark matter halos with isothermal core. The same kind of problem can be found for globular clusters, which show cores in their density profiles but should be cuspy if they where formed through a cold collapse.\\cite{King:62,Jenkins:98} We are showing how small, massive, and compact objects can efficiently transform cuspy stellar systems into density profiles with an isothermal core. ",
        "introduction": " ",
        "conclusions": "We have shown that 10 black holes are capable of efficiently destroying a stellar cusp within a few ten dynamical timescales. In our simulations each black hole removes an equivalent of seven times its own mass from the central regions of the sytem. The resulting systems show isothermal cores."
    },
    "0209/astro-ph0209413_arXiv.txt": {
        "abstract": "We derive a non-parametric CO luminosity function using a FIR and an optical $B$-band selected sample of the galaxies included in the FCRAO Extragalactic CO Survey.  The FIR selected sample is defined using the IRAS Bright Galaxy Surveys (BGS; IRAS 60 micron flux density $\\ge 5.24$ Jy). Although our CO sample is not complete, the normalization using the BGS reproduces the IRAS 60 micron luminosity function in excellent agreement with those found in the literature.  Similarly, a $B$-band selected sample defined using the Revised Shapley-Ames (RSA) catalog is used to derive a CO luminosity function for a comparison. A Schechter function describes the both derived CO luminosity functions reasonably well. Adopting the standard CO-to-H$_2$ conversion factor, we derive a molecular gas density of $\\rho_{H_2}=(3.1\\pm 1.2) \\times 10^7h M_\\odot$ Mpc$^{-3}$ for the local volume. Combining with the measurements of the local HI mass density and the helium contribution, we estimate that the total mass density of cold neutral gas in the local universe is $\\Omega_{gas} =(4.3 \\pm 1.1)\\times 10^{-4} h^{-1}$, which is about 20\\% of the total stellar mass density $\\Omega_*$. ",
        "introduction": "}  The star formation history of the Universe is closely linked with the evolution of the gas content of the Universe.  Observations of the distribution and total gas content in galaxies and in intergalactic clouds offer some of the most important observational constraints for the cosmology and galaxy evolution models. Because stars form out of cold, dense gas clouds, the history of galaxy formation and evolution is also the history of gas accretion and conversion into stars. Here we derive the total cold gas density for the local volume by deriving the local CO luminosity function and combining with the existing estimates of the neutral atomic gas density.  A variety of luminosity functions (LFs) and mass functions are found in the literature: optical \\citep[e.g.][]{mar94}, HI \\citep[e.g.][]{zwa97,sch98}, and infrared \\citep [e.g.][]{soi87,yun01}. While several groups have investigated the local neutral hydrogen density, little information is available on the molecular gas content in galaxies because of the lack of appropriate data. In her pioneer work, \\citet{ver87} derived averaged values of CO luminosity and the CO/HI flux ratio based on a maximum likelihood probability distribution for 40 galaxies and 47 upper limits obtained by combining all available data at the time.  Since the large fraction of the CO measurements used were non-detections (upper limits), the derivation of the CO luminosity function was highly problematic.  Nevertheless, a general trend of increasing number of CO emitting galaxies per unit volume with decreasing CO luminosity was suggested by this analysis.  Nearly 15 years later, the availability of the extragalactic CO data has improved greatly.  The FCRAO (Five College Radio Astronomy Observatory) Extragalactic CO Survey \\citep{yng95} represents a particularly rich database for the investigation of CO luminosity function and molecular gas content in galaxies.  We have constructed a large, statistically significant sample of far-infrared (FIR) and optical $B$-band selected galaxies from this survey and derived a non-parametric CO luminosity function.  Our sample galaxies range over 4 orders of magnitudes in CO luminosity, and only a small fraction are non-detections. Since CO is a tracer of hydrogen molecules, molecular hydrogen mass can be derived from CO luminosity. By integrating the resulting molecular gas mass function, we then derive the molecular gas mass density and the total cold gas mass density of the local universe.  \\bigskip ",
        "conclusions": "}  \\subsection{Uncertainties and biases in the CO LF \\label{sec:bias}}  One major source of uncertainty in the derived CO luminosity function is that it is derived indirectly using the FIR and optical $B$-band selection functions.  A direct derivation from the observed CO properties is possible in principle, but the complication associated with the {\\it a priori} unknown line widths adds a significant uncertainty.  Instead we take advantage of the known tight correlation between FIR and CO luminosity in deriving the CO LF for our sample.  The 1.4 GHz radio luminosity function derived by \\citet{yun01} using the radio-FIR correlation and IRAS 60 $\\mu$m flux density agrees very well with the radio LFs derived directly \\citep{con91,con02}, giving some assurance to this technique. We have examined whether any systematic trends (thus a bias) exist in the $S_{60\\mu m}/S_{CO}$ ratio as a function of the 60 $\\mu$m flux density and luminosity. As shown in Figure~\\ref{fig:60/CO}, little trend is seen in this ratio, and the known linear correlation seems to hold well within the uncertainties for the entire range of flux density and luminosity.  Deriving a luminosity function for one wavelength using a selection function at another wavelength may be robust enough to work even if the two quantities are correlated in a non-linear way. There is a known correlation between FIR, CO, and optical $B$-band luminosity for late type galaxies, but the correlations involving the $B$-band luminosity are shown to be non-linear \\citep[see][]{yng89,per97}.  Yet, the CO luminosity function derived using the FIR selection function (Figure~\\ref{fig:COIRLF}) is in excellent agreement with the CO LF derived using the $B$-band selection function (Figure~\\ref{fig:COBLF}).  The selection bias is present as some of the most CO luminous galaxies (merger starbursts) are missed in the $B$-band selected sample.  Nevertheless the agreement is striking given the non-linear dependence between CO and $B$-band luminosity.  Another source of a significant uncertainty is the uniformity of sampling and correctly accounting for the selection function for the sample of galaxies which is not complete by the adopted selection criteria.  Both the differential source count and the $<V/V_m>$ analysis suggest a non-uniform distribution of the sample galaxies (see \\S~\\ref{sec:sample}), and a slight bias towards galaxies brighter at 60$\\mu m$ may be responsible for this effect.  The large scale structure also have influence in lowering $<V/V_m>$ value for our sample. The non-trivial nature of these effects are demonstrated by the fact that including and excluding Virgo cluster galaxies from the sample make little difference to the differential source count and the $<V/V_m>$ values.  \\subsection{Local Cold Gas Mass Density \\label{sec:gasdensity}}  CO emission is a commonly used tracer of molecular hydrogen because CO is one of the most abundant molecules in cold ISM and because its excitation in astrophysical conditions is determined by collisions with hydrogen molecules.  Using the CO LF derived above, we can for the first time estimate the total mass density of molecular gas in the local volume. Adopting $N(H_2)/I(CO)=3 \\times 10^{20}$ cm$^{-2}$ [K km s$^{-1}$]$^{-1}$ \\citep[see review by][]{ysc91}, \\begin{equation} M(H_2)=1.18\\times 10^4 S_{CO} D^2 M_{\\odot} \\label{eq:mh2} \\end{equation} \\noindent where $S_{CO}$ is total CO flux of a galaxy in Jy km s$^{-1}$ and $D$ is luminosity distance in Mpc.  The molecular mass density in the local volume contributed by each luminosity bin is shown in Figure~\\ref{fig:h2mass}.  The dominant contribution to the mass density comes from galaxies around $L^*$ as expected. The summation over the 10 bins gives $\\rho_{H_2}=\\sum M_{H_2}/V_m=(2.4 \\pm 0.7) \\times 10^7 M_{\\odot}$ Mpc$^{-3}$. Integration of the LF using the Schechter parameters obtained in previous section gives a little bit smaller values: $\\rho_{H_2}=(2.2 \\pm 1.1) \\times 10^7 M_{\\odot}$ Mpc$^{-3}$ for the fit trough all 10 bins, and $\\rho_{H_2}=(2.2 \\pm 0.9) \\times 10^7 M_{\\odot}$ Mpc$^{-3}$ for fit trough first 8 bins. The second fit is more realistic, since it fits much better bins that dominate contribution to the total mass, i.e. bins around $L^*$. The uncertainty stated is $\\pm 1\\sigma$. The  systematic uncertainty, which includes uncertainties in the flux measurements, distance determinations, and CO-to-H$_2$ conversion, is probably larger. Unless otherwise is stated, we adopt $\\rho_{H_2}=(2.3 \\pm 0.9) \\times 10^7 M_{\\odot}$ Mpc$^{-3}$ as an average value between values obtained from the fit and the direct summation. Since the dependence of the gas mass density on the Hubble constant ($h\\equiv H_0/{100}$ [km s$^{-1}$ Mpc$^{-1}$]$^{-1}$) is linear, this result can be written as $\\rho_{H_2}=(3.1 \\pm 1.2)\\times 10^7 h M_{\\odot}$ Mpc$^{-3}$.  The molecular hydrogen mass density in the local volume obtained from the $B$-band selected sample is $\\rho_{H_2}=(3.1\\pm 0.9) \\times 10^7h M_{\\odot}$ Mpc$^{-3}$ using a direct summation of the contribution of each galaxy and $(3.1 \\pm 1.5) \\times 10^7h M_{\\odot}$ Mpc$^{-3}$ using the best fit Schechter parameters. Values for the local molecular mass density obtained from the FIR selected sample and the $B$-band selected sample are in good agreement.  Since we have better statistics for the larger FIR selected sample, we adopt the local molecular gas mass density derived from the FIR selected sample here on. Using observed H$_2$/HI mass ratio for different morphological types of galaxies and fraction of each morphological type \\citet{fuk98} estimated a similar value for $\\rho_{H2}$.  In comparison, \\citet{zwa97} estimated the local atomic gas mass density of $\\rho_{HI}=(5.8 \\pm 1.2)\\times 10^7 h M_{\\odot}$ Mpc$^{-3}$ from their Arecibo HI strip survey while \\citet{rab93} derived $\\rho_{HI}=(4.8 \\pm 1.1) \\times 10^7 h M_{\\odot}$ Mpc$^{-3}$ from a sample of optically selected galaxies.  Therefore, the molecular gas mass density in the local volume is about 50-65\\% of the atomic mass density, and  molecular gas represents a significant component of the total mass density of the neutral gas. The HI masses for 176 galaxies in our sample are known \\citep{yng95}, and we have computed the HI mass density in the local volume using the same procedure as for the $H_2$ mass density. A summation over the HI mass weighted by $V_m$ gives a value consistent with the value obtain by \\citet{zwa97}.  We examined our sample for any trends in $M_{H_2}/M_{HI}$ ratio vs. $M_{HI}$ and $M_{H_2}$. For the range of HI masses, $8.8 < log~M_{HI}(M_\\odot) < 11$, the ratio of molecular to neutral atomic hydrogen masses is on average around unity with considerable scatter (see Fig.~\\ref{fig:ratios}a).  Masked in the large scatter is a possible trend in $M_{H_2}/M_{HI}$ as a function of Hubble type \\citep[see][]{yng89}.  A clearer trend of increased $M_{H_2}/M_{HI}$ ratio with increasing H$_2$ mass is seen in Figure~\\ref{fig:ratios}b.  For $log~M_{H_2}(M_\\odot) < 8.5$, the average of this ratio is below 0.5 while the ratio jumps to around 2 near $M_{H_2}\\sim 10^9 M_\\odot$.  Above H$_2$ mass of $10^9 M_\\odot$, on average galaxies have more molecular than atomic gas.  Using the molecular gas mass density derived here and the value of HI mass density from \\citet{zwa97} we derive a total cold gas mass density at present epoch as a fraction of the critical density $\\Omega_{HI+H_2}=(3.2 \\pm 0.8) \\times 10^{-4}h^{-1}$. Supposing that He contributes 25\\% of the total gas mass density, we add 33\\% to the derived value of $\\Omega_H$, which gives $\\Omega_{HI+H_2+He}\\equiv \\Omega_{gas}=(4.3 \\pm 1.1) \\times 10^{-4}h^{-1}$. These values for total neutral gas mass density are 15\\% smaller if we use \\citet{rab93} value for the HI mass density.  The present baryon density estimated from the D/H ratio and Big Bang nucleosynthesis is $\\Omega_b = 0.02 \\pm 0.002 h^{-2}$ \\citep{bur01} while the estimate from the cosmic microwave background anisotropy is also around $\\Omega_b\\simeq 0.02 h^{-2}$ \\citep{deb02}.  The stellar mass contents account for $\\Omega_* \\simeq 0.0025 h^{-1}$ (Cole et al. 2001; but also see Benson, Frenk, \\& Sharples 2002). We conclude that the cold gas content of late type galaxies is around 20\\% of the stellar mass content and about 2\\% of the total baryonic content in the current epoch.  \\bigskip  \\subsection{CO-to-H$_2$ conversion \\label{sec:Xfactor}}  The lack of electric dipole moment makes direct observations of molecular hydrogen difficult in general, and studying the spatial extent and molecular gas mass requires another tracer.  Highly abundant, chemically robust, and easily excited by collision with H$_2$ molecules, CO is the most commonly used tracer of molecular gas.  The CO (1--0) transition is optically thick under most astrophysically interesting conditions, making it relatively insensitive to metallicity and abundance effects.  The two key excitation parameters of density and temperature have a nearly canceling effect, making CO a fairly reliable tracer of H$_2$ in a broad range of physical conditions \\citep[see reviews by][]{mal88,ysc91}.  The derived ratios of $N(H_2)/I(CO)$ range between $(1-5) \\times 10^{20}$ cm$^{-2}$ [K km s$^{-1}$]$^{-1}$ \\citep{blo86,dic86,scs87}. We adopt a constant CO-to-H$_2$ conversion factor of $N(H2)/I(CO)=3\\times 10^{20}$ cm$^{-2}$ [K km s$^{-1}$]$^{-1}$ \\citep[see discussions  by][]{ysc91}.  \\citet{dey90} show that the H$_2$ mass estimates of galaxies from their CO luminosity are accurate to $\\pm30\\%$. \\citet{ysc91} show that the CO-to-H$_2$ conversion for galaxies of diverse morphology and metallicity are similar in absolute value to the conversion in the Milky Way. While the $H_2$ mass estimate for an individual galaxy may be uncertain to about 30\\%, the $H_2$ mass estimate for an {\\it ensemble} of galaxies should be more reliable.  In most galaxies, giant molecular clouds (GMCs) and cloud complexes dominate the total molecular gas mass, and adopting a conversion factor consistent with the values derived from the Galactic GMCs should yield the most robust, mass-weighted estimates of gas masses.  Among the low metallicity galaxies such as Small Magellanic Cloud, CO abundance may become low enough to affect the standard assumption of self-shielding and thermalization.  The derived conversion factors are generally larger \\citep[e.g.][]{wil95}, and low metallicity dwarf galaxies are often undetected entirely in CO. Therefore the H$_2$ mass and mass density derived from the CO luminosity are strictly lower limits since molecular gas from these galaxies is missing from our analysis. On the other hand, the contribution by the low luminosity bins to our derived $H_2$ mass density is small (see Fig.~\\ref{fig:h2mass}) as the majority of the contribution to the total mass density comes from bins around $L^*$. Therefore, we can largely neglect the effect of low metallicity, low luminosity galaxies in the derivation of the total molecular gas mass density for the local volume.  As discussed already in \\S~\\ref{sec:COIRLF}, the CO emission may be elevated for the FIR luminous galaxies with nuclear gas concentrations, and the CO-to-H$_2$ conversion factor may be smaller than the canonical value.  In the extreme environment of the nuclear starburst regions, CO emission arises from a multi-phase medium with sub-thermal excitation in the diffuse phase, and the standard conversion factor may over-estimate the molecular gas mass by a factor as large as 3 to 5 \\citep{sco97,dow98}. Such merger/starburst galaxies contribute significantly only for the top two luminosity bins in our CO LF, and these two bins contribute less than 5\\% to the the total molecular gas mass density. Therefore, the high CO luminosity conversion factor does not have a significant impact on the derived total molecular mass density.  Similarly, increased CO emission in the central 1 kpc of our Galaxy and other galaxies has been suggested by recent observations \\citep[][S. H\\\"{u}ttemeister, private communication]{pag01}. High angular resolution observations of gas-rich spiral galaxies frequently reveal a distinct component in the central 1 kpc \\citep[see][]{saka99}, and some fraction of the total CO luminosity may arise from such a component. On the other hand, among the Virgo spirals studied by \\citet{kyn88}, a simple exponential distribution without a significant central component offers a good fit for the 12 out of 14 galaxies whose CO emission is spatially well resolved (i.e., CO detected at $\\ge4$ positions along the major axis). We made no attempt to account for enhanced CO emission in the central kpc as the required information is generally not available.  If the molecular gas properties of Virgo spirals are typical of the late type field galaxies, then the possible contribution by the enhanced nuclear CO emission in some galaxies may not be substantial.  In summary, using the canonical CO-to-H$_2$ conversion relation is problematic in some cases, such as low metallicity systems or luminous nuclear starburst systems, and CO is a poor tracer of molecular gas among low metallicity dwarfs. For these reasons the local molecular gas mass density we infer is really a lower limit.  However, since the majority of the mass contribution comes from $L^*$ galaxies whose molecular gas mass is dominated by GMCs like our Galaxy, the use of the standard conversion factor still offers a fairly reliable estimate of the total molecular gas density for the local volume.  \\bigskip  }  Utilizing the largest available CO survey, the FCRAO Extragalactic CO Survey \\citep{yng95}, the CO luminosity function for the local volume is derived using (1) a FIR selected sample of 200 galaxies that satisfy $S_{60\\mu m}> 5.24 Jy$ and (2) optical $B$-band selected sample of 133 galaxies.  Although neither of the samples is complete in terms of the sample selection, a sampling function is constructed using a well defined parent sample, and a non-parametric CO luminosity function is derived from each sample.  By examining the properties of the CO luminosity functions, we conclude:  \\begin{enumerate}  \\item The CO luminosity functions derived from the FIR and $B$-band selected samples are reasonably well described by a Schechter function.  The characteristic luminosity $L^*$ is around $L_{CO} \\sim 10^7$ Jy km s$^{-1}$ Mpc$^2$. The low luminosity end of the CO luminosity function is ($\\alpha= -1.3~{\\rm to}~-0.9$).  Similar values are obtained for the two CO LFs derived using the two differently selected samples.  \\item The molecular gas mass density of the local volume is $\\rho_{H_2}=(3.1\\pm1.1) \\times 10^7 h M_\\odot$ Mpc$^{-3}$ which is about 50-65\\% of the HI gas mass density. This value is not strongly affected by variations in the CO-to-H$_2$ conversion factor for low and high luminosity galaxies since it is dominated by the $L^*$ galaxies with $L_{CO} \\sim  10^7$ Jy km s$^{-1}$ Mpc$^2$ ($M^*_{H_2}\\sim 5\\times 10^9 M_\\odot$). Its dependence on the global CO-to-H$_2$ conversion factor is linear and should be secure to within a factor of 30\\% or better.  \\item Combined with the HI gas mass density, we estimate the total cold gas mass density at the present epoch as a fraction of the critical density $\\Omega_{HI+H_2}=(3.2 \\pm 0.8) \\times 10^{-4}h^{-1}$. When the He contribution is included, the cold gas mass density increase to $\\Omega_{HI+H_2+He}\\equiv \\Omega_{gas}=(4.3 \\pm 1.1) \\times 10^{-4}h^{-1}$. Therefore, the cold gas content of late type galaxies corresponds to about 20\\% of the stellar mass content and about 2\\% of the total baryonic content in the universe.  \\end{enumerate}  \\bigskip"
    },
    "0209/astro-ph0209469_arXiv.txt": {
        "abstract": "We calculate spectral energy distributions (SEDs) of steady accretion discs at high accretion rates, as appropriate for bright QSOs, under the assumption that the outer parts are heated sufficiently to maintain marginal gravitational stability, presumably by massive stars formed within the disc.  The SED is independent of the nature of these auxiliary sources if their inputs are completely thermalized.  Standard assumptions are made for angular momentum transport, with an alpha parameter less than unity.  With these prescriptions, the luminosity of the disc is sensitive to its opacity, in contrast to standard discs powered by release of orbital energy alone.  Compared to the latter, our discs have a broader SED, with a second peak in the near-infrared that is energetically comparable to the blue bump. The energy in the second peak increases with the outer radius of the disc, provided that the accretion rate is constant with radius. By comparing our computed SEDs with observed ones, we limit the outer radius of the disc to be less than $10^5$ Schwarzschild radii ($R_{\\rm S}$), or about one parsec, in a typical QSO. We also discuss some properties of our minimum-$Q$ discs in the regions where auxiliary heating is dominant ($10^3-10^5 R_{\\rm S}$). ",
        "introduction": "The standard theoretical paradigm for the central engine in quasars and their radio-quiet kin (QSOs) is a viscous accretion disc surrounding a massive black hole. Direct evidence for accretion discs, such as double-peaked emission lines \\citep{Eracleous_Halpern94}, is important to seek but hard to come by. Perhaps the best reasons for belief in this paradigm are basic considerations of energy and angular momentum.  Such discs are the most plausible astrophysical mechanism for converting rest mass to radiation with high efficiency, and if the relics of QSOs reside in galactic nuclei, then efficiencies $\\gtrsim 10\\%$ are required \\citep{Soltan82,Chokshi_Turner92,Yu_Tremaine02}. On the other hand, the maximum specific angular momentum of a Kerr black hole, $GM/c\\approx 1.4 M_8\\kms$, is far less than that of most mass in galaxies. ($M_8=M/10^8 M_\\odot$, where $M$ is the mass of the black hole.) Accreting gas must be separated from its angular momentum, and a viscous disc is the most natural mechanism.  Most of the binding energy of the disc is in its inner parts, so that the outer radius of a standard accretion disc is almost irrelevant to its bolometric luminosity.  In contrast, most of the angular momentum is in the outer regions, so that the outer radius of the disc is sensitive to the initial angular momentum of the gas that feeds it. In a previous paper \\citep[][henceforth Paper I]{Goodman02}, it was argued that luminosity and angular momentum are coupled by the requirement that QSO discs be stable against their own self-gravity.  It is well known that gravitational stability is problematic in the outer parts of QSO discs \\citep{Shlosman_Begelman87}. The threat to the standard paradigm is that a strongly selfgravitating disc is likely to fragment completely into stars, leaving insufficient gas to fuel the QSO.  As discussed in Paper I, viable solutions to this difficulty fall into several categories: \\begin{enumerate} \\item enhanced angular momentum transport, not necessarily by viscous processes but at rates corresponding to a viscosity parameter $\\alpha\\gg 1$; \\item auxiliary heating in excess of what is provided by dissipating orbital energy, so as to reduce the gas density and selfgravity of the disc; \\item replacement of the outer disc with a very dense star cluster, whose collisional debris supply a disc of small radius and negligible selfgravity; \\item relatively low initial angular momentum for the gas, which therefore circularizes at small enough radius so as to avoid selfgravity. \\end{enumerate} A number of options in the first category were briefly considered in Paper I, including accretion driven by bars or global spiral waves \\citep[e.g.][]{Shlosman_Begelman89} or by magnetized winds \\citep[e.g.][]{Blandford_Payne82}.  But it was argued that none of these options is likely to achieve a supersonic accretion speed, and hence that discs cannot be stabilized much beyond one parsec by any of these mechanisms alone.  The third category was tentatively rejected on the grounds that remnants of such star clusters are not observed in present day galactic nuclei.  Recently, \\cite{Pariev_etal02} have proposed discs whose thickness is supported primarily by magnetic pressure.  If this is possible, it would imply lower disc densities on average, but the field might well squeeze the gas into dense clumps and thereby actually exacerbate selfgravity.  At any rate, \\cite{Pariev_etal02} do not apply their model to the cool outer regions beyond $10^3\\rs$.  The present paper will focus on some implications of the second category of solutions, which appear to be a natural compromise between a purely gaseous disc and a star cluster.  It is reasonable to suppose that part of the gas fragments into stars, and that the energy released by nuclear fusion and other stellar processes (supernovae, stellar-mass black holes) may sufficiently heat the rest of the gas so as to prevent complete fragmentation \\citep{Collin_Zahn99a,Collin_Zahn99b}.  There is a great deal of evidence that such a feedback cycle operates in the discs of spiral galaxies on kiloparsec scales.  Observed SEDs of typical quasars differ markedly from classical theoretical predictions in which the disc is assumed to be geometrically thin, optically thick, steady, and heated solely by viscous dissipation \\citep{Pringle81}.  To a first approximation, the typical SED is flat in a $\\lambda F_\\lambda$ plot over many decades in wavelength \\citep{Elvis_etal94}.  Relative to the classical predictions, there is excess emission at both X-ray and infrared wavelengths.  The former is conventionally ascribed to comptonization in a hot corona at small radii \\citep{Shapiro_Lightman_Eardley76}, and the latter to passive reprocessing in warped or flared outer parts of the disc \\citep{Sanders_etal89}.  We suppose that the infrared excess may be due to the energy inputs required to stabilize the outer disc against its own selfgravity.  We obtain a lower bound on the auxiliary heating needed to stabilize the disc for given values of the macroscopic parameters: namely, the black hole mass ($M$), accretion rate ($\\dot M$), and disc outer radius ($r_{\\max}$).  As shown in Paper I, and confirmed here with more realistic opacities, the inputs required for gravitational stability increase with the outer radius of the disc, $r_{\\rm max}$. Paper I argued for an upper limit to $r_{\\rm max}$ based on the energy available from plausible sources, such as fusion or accretion onto stellar-mass black holes.  In this paper, we find limits to $r_{\\rm max}$ from the SED.  It is not obvious that the auxiliary inputs should be completely thermalized, but if we assume this, then the SED can be predicted.  The observed infrared emission of typical QSOs may be due largely to reprocessing of light emitted from the inner parts of the disc, as conventionally supposed.  But by attributing all of the infrared light to the auxiliary energy inputs, and insisting that these inputs be sufficient for gravitational stability, we obtain bounds for $r_{\\rm max}$.  These bounds depend upon other parameters, especially the mass of the black hole, the accretion rate, and the viscosity parameter $\\alpha$.  We explore these dependencies. The meaning of $r_{\\max}$ constrained by this method is the radius within which $\\dot M$ is sensibly constant; obviously the disc can be extended indefinitely if the accretion rate and mass at large radii are sufficiently small.  Hence our limits on $r_{\\max}$ are best translated into upper limits on the initial angular momentum of the gas that is accreted.  The outline of our paper is as follows.  \\S 2 lays out the physical assumptions and governing equations for our disc models.  These are the same as for the classical steady thin disc, with the one important exception that wherever the classical model would be gravitationally unstable, we invoke just enough auxiliary heating to stabilize it.  A discussion of opacities becomes critically important, because the requirement of marginal stability fixes the density and temperature at the midplane (for given $\\dot M$ and $\\alpha$); the flux escaping from the disc, and hence the amount of auxiliary heating needed, then depend upon the optical depth. The computed radial structure of the disc is presented in \\S3 for parameters representative of bright quasars. Since our assumptions are no different from the classical ones at small radii, we emphasize the properties of our discs in the marginally selfgravitating region $r\\gtrsim 10^3\\rs\\sim 10^{-2}\\pc$, and we compare our SEDs with those presented by \\cite{Elvis_etal94} to obtain upper limits on the outer radius of the disc and the initial angular momentum of the gas.  In the final section, we summarize our conclusions, issue the necessary caveats, and discuss directions for future research. ",
        "conclusions": "We have estimated spectral energy distributions (SEDs) of bright QSOs using standard assumptions, with one addition:  where the disc would otherwise be gravitationally unstable, we have postulated additional sources of heat, other than release of orbital energy by accretion, just sufficient to maintain gravitational stability.  These sources become necessary beyond $\\sim 10^3\\rs$ for typical parameters.  Assuming their energy inputs to be locally and completely thermalized, we have calculated their contribution to the SED and luminosity of the disc, which occurs primarily in the red and near infrared. The larger the disc, the more auxiliary heating is required. For typical black-hole masses and accretion rates inferred from the blue bump, the auxiliary inputs actually exceed the power derived from accretion if the disc extends beyond $10^4-10^5\\rs$, or about one parsec.  This would be incompatible with the typical SED of bright QSOs, which is approximately flat in $\\lambda F_\\lambda$.  Paper I placed similar limits on $r_{\\rm max}$ from energetic arguments. It was assumed that the stars or small black holes that heat the disc also form within the disc, and that the mass in these objects is at most comparable to that of the disc.  The present limits do not rely on these assumptions.  That is to say, if the disc were heated by a much larger mass in stars, the limits of Paper I could be relaxed, but those based on the SED would still apply.  It has been assumed that the accretion rate is constant with radius, so that $\\dot M$ at large radii can be derived from the luminosity in the blue bump (which has no contribution from the auxiliary sources).  If $\\dot M$ at $r\\gtrsim 1\\pc$ is several orders of magnitude less than it is at $r\\lesssim 10^3\\rs$, then the disc could be much more extensive than we have supposed.  The Stefan-Boltzmann law implies that the observed luminosity at $10\\,\\mu$ must in any case come from $r\\gtrsim 2 (\\lambda L_\\lambda/10^{46}\\mbox{ erg s}^{-1})^{1/2} \\pc$, but this does not require that the surface density at that distance is as high as in a constant-$\\dot M$ disc. Therefore, it is useful to rephrase the limits in terms of the initial angular momentum of the gas supplied to the disc. At least on a time average, $\\dot M$ should be constant inside the radius corresponding to the initial angular momentum.  The relationship is \\begin{equation}\\label{J0} J_0\\approx 660 \\left(r_{\\pc}M_8\\right)^{1/2}\\pc \\kms. \\end{equation} This is quite small compared with the product of virial velocity ($\\sim 300\\kms$) and scale size ($\\sim 1\\kpc$) of QSO hosts, and it may be important to ask where gas with such low angular momentum comes from.  Furthermore, the mass of a gravitationally stable disc that obeys our constraints on the SED is generally much less than that of the black hole, so that the disc must be replenished many times over to grow the black hole by accretion.  Our treatment of the disc is highly simplified and certainly crude compared with many past efforts.  We have ignored relativistic effects, adopted a one-zone model for the vertical structure, used Rosseland mean opacities without distinguishing between scattering and absorption, taken a constant molecular weight, and represented angular-momentum transport by the usual viscous $\\alpha$ prescription. We feel that these simplifications are justified by the strong dependence of the auxiliary inputs and the SED on the outer radius of the disc.  A more detailed treatment of the physics that one actually understands seems unlikely to change our conclusions concerning $r_{\\rm max}$.  Radical enhancements in transport (equivalent to $\\alpha\\gg1$), or magnetic pressures $\\gg \\pgas$ \\citep[as suggested by][]{Pariev_etal02} could make some difference, but these are not yet understood.  Despite what has just been said, a more elaborate treatment of vertical structure and radiative transfer might point to a redistribution of the auxiliary energy inputs in wavelength, if not their total contribution to the disc luminosity. Unfortunately, one has no reliable predictions for the vertical distribution of purely viscous heating, much less of the auxiliary sources postulated here, which will limit the credibility of detailed vertical models.   Even if the actively accreting parts of QSO discs are smaller than $0.1\\pc$, it is still possible that selfgravity is important in them, and therefore that they form stars.  Quiescent galactic nuclei with black holes, even in early-type galaxies, often show kinematic evidence of compact stellar discs \\citep{Gebhardt_etal00b,Bower_etal01,deZeeuw_etal02}.  Nuclear starbursts (albeit on scales $\\sim 10^2\\pc$) are usually accompanied by AGN activity \\citep{Sanders99,Heckman99}.  The black hole in our own Galaxy, though not an AGN and estimated to have a very low accretion rate \\citep{Quataert_etal99}, is surrounded by what appear to be young high-mass stars at $r\\lesssim 0.1\\pc$ \\citep{Krabbe_etal95}.  More theoretical attention should be paid to star formation in these extreme environments, where the densities, temperatures, and tidal fields are much higher than in normal giant molecular clouds.      \\bigskip We thank Iskra Strateva and Jonathan Tan for helpful discussions."
    },
    "0209/astro-ph0209143_arXiv.txt": {
        "abstract": "We study the peculiar motion of non-relativistic matter in a fully covariant way. The exact nonlinear equations are derived and then applied to the case of pressure-free matter, moving relatively to a quasi-Newtonian Eulerian frame. Our two-frame formalism facilitates the study of the nonlinear kinematics of the matter, as the latter decouples from the background expansion and starts to ``turn around'' and collapse. Applied to second perturbative order, our equations provide a fully covariant formulation of the Zeldovich approximation, which by construction addresses the mildly nonlinear regime of structure formation. Employing a dynamical system approach, we show that, just like in the Newtonian case, the relativistic treatment also predicts that pancakes are the natural end-structures for any generic overdensity.\\\\\\\\ PACS number(s): 04.25.Nx, 98.80.Hw ",
        "introduction": "Studies of linear perturbations are crucial for understanding the way structure formation has progressed in our universe. In the standard model density fluctuations grow slowly, via gravitational instability, to form the galaxies, the clusters of galaxies, the superclusters, the filaments and the voids that we see in the universe today. Despite the simplicity of this idea, however, our analytical understanding of the advanced stages of gravitational collapse is still limited. The reason is that the linear theory is a good approximation only at the initial stages of the collapse, when the density contrast of the perturbation is well below unity. Most of the structures in the universe, however, have density contrasts well in excess of unity. To understand the evolution of these objects we need to go beyond the limits of the linear approximation. The inherent complications of nonlinear collapse, however, mean that analytical studies are possible only when certain simplifying assumptions are made. An approximate approach that extends well into the nonlinear regime, up to the virialization of bound objects, is the spherical collapse model~\\cite{GG} (see also~\\cite{Pee1}). Although the spherical model became popular because of its simplicity, in reality it stops short of explaining key features of the observed universe. Gravitational collapse does not seem to proceed isotropically. All galaxy surveys show structures with complicated triaxial shapes, which require a nonspherical analysis if they were to be explained.  The Zeldovich approximation is not restricted to spherical symmetry~\\cite{Z}. It is a kinematical, Lagrangian approach that addresses the issue of anisotropic collapse by extrapolating into the nonlinear regime a well known linear result (see~\\cite{SZ} for a review and also~\\cite{P} for related discussion). More specifically, the Zeldovich ansatz assumes that the acceleration-free and irrotational linear peculiar motion of pressure-free matter also holds during the nonlinear collapse. The consequences of this hypothesis are quite dramatic. What Zeldovich showed was that any generic (i.e.~non-spherical) overdensity will undergo a phase of anisotropic, effectively one-dimensional, collapse leading to the formation of two-dimensional flattened structures that are widely known as ``pancakes''. Over the years, the Zeldovich approximation has provided a great deal of insight into the initial nonlinear evolution of density fluctuations. It has also inspired a number of developments concerning the behavior of the velocity field in irrotational flows (see~\\cite{NDBB} for a representative list).\\footnote{The limits of the Zeldovich approximation, both linear and nonlinear, have been investigated in~\\cite{Bu1}. Some authors have suggested modifications of the standard Lagrangian approach to increase its accuracy, by adding vorticity~\\cite{BS}, or viscosity~\\cite{KPS}, while others have treated either the velocity potential as a constant~\\cite{MLMS}, or the gravitational potential itself~\\cite{BP}.} Most of the available studies, however, are Newtonian and, although there are general relativistic approaches in the literature (see \\cite{K}), a fully covariant treatment is still missing. Moreover, Newtonian and relativistic collapse seem to move further apart in the presence of anisotropy (e.g.~see~\\cite{ZN}). In the Newtonian approach pancakes are the natural attractors, at least in the Zeldovich approximation; a result that seems well confirmed by N-body simulations~\\cite{SMMPT}. The relativistic ``silent universe'' models, on the other hand, point towards spindle-like singularities~\\cite{BJ, BMP1}.\\footnote{Silent universes are inhomogeneous spacetimes, with irrotational pressure-free matter and zero magnetic Weyl tensor. By construction, they cannot support any communication through sound or gravitational wave signals, which explains the term silent. This class of models contains the well known Szekeres solution as a special case. For an introduction and further discussion on silent universes we refer the reader to~\\cite{MPS}.} In other words, the final fate of a collapsing overdensity remains an open issue. In the present article we attempt to address this question by looking at the dynamics of nonlinear collapse in a perturbed Einstein-de Sitter universe.  One issue that emerges naturally, especially when attempting a relativistic analysis of peculiar motions, is that the associated velocities should be defined relative to a preferred frame of reference. This frame is necessarily non-comoving with the matter, since by definition there are no peculiar velocities relative to the matter frame, which is the truly Lagrangian frame. It should be emphasized that in the standard literature the term ``comoving'' refers usually (and sometimes misleadingly) to a fictitious background 4-velocity, rather than to the actual velocity of the matter. In the covariant approach one seeks to define the aforementioned preferred frame in a physical way.\\footnote{An extensive presentation of the covariant approach to cosmology can be found in~\\cite{E1}. For an updated review the reader is referred to~\\cite{EvE}.} Motivated by~\\cite{Pee}-\\cite{EvEM}, we chose a 4-velocity field that is both irrotational and shear-free, thus defining a quasi-Newtonian frame that closely corresponds to Bardeen's quasi-Newtonian gauge~\\cite{B}. Our aim is to set up the general framework for the fully covariant and fully nonlinear treatment of non-relativistic peculiar motions. We then exploit the natural transparency of the covariant formalism to provide a fully covariant version of the Zeldovich approximation. As is well known, the latter addresses the early stages of nonlinear collapse. We call this period the ``mildly nonlinear regime''. Given that we deal with the post-recombination era, we assume an Einstein-de Sitter background and consider the second order evolution of the perturbed variables. In the frame of the pressure-free matter, we find that the equations describing the collapse of the non-relativistic component to second perturbative order do not include any tidal effects. They reduce to a set of ordinary second order differential equations, which can be rescaled into a planar dynamical system with one-dimensional pancakes as its natural attractors. ",
        "conclusions": "A key issue in contemporary theoretical cosmology is understanding the physics of gravitational collapse and the mechanisms that have given rise to the observed large-scale structure of the universe. Linear perturbations, about a Friedmann-Robertson-Walker model, are fairly straightforward to follow. When certain simplifying symmetries are imposed, the nonlinear collapse can also be treated analytically. The spherical top-hat model, in Newtonian theory, and the Tolman-Bondi solution, in General Relativity, are such examples. When it comes to the more realistic non-spherical collapse, however, the Zeldovich approximation has been the most influential and celebrated paradigm. It has also inspired a number of variations, all of which try to address the complexities of structure evolution beyond ``caustic'' formation. Despite the number of these different treatments, however, a fully covariant approach to the Zeldovich approximation has been missing. This is the issue that the present paper has tried to address.  We have pursued a relativistic treatment of nonlinear peculiar velocities by adopting a two-frame approach which enables us to provide a truly Lagrangian formalism. Assuming matter with non-relativistic peculiar motion relative to a quasi-Newtonian frame, we have derived the fully nonlinear equations and applied them to the case of pressure-free dust. The inherent transparency of the covariant formalism means that all the terms in our equations have a clear physical and geometrical interpretation. For a given background, it is also straightforward to identify the perturbative order of all the effects that influence the peculiar motion of the matter. Assuming an Einstein-de Sitter background and allowing for up to second order perturbative terms, we have addressed the mildly nonlinear collapse and in the process provided a fully covariant formulation of the Zeldovich approximation. On introducing the Zeldovich ansatz of acceleration-free and irrotational peculiar motion, our equations have reduced to a planar dynamical system, which has one-dimensional pancakes as its natural attractors. Thus, just like the Newtonian treatment, the relativistic approach also predicts that, as long as we consider the mildly nonlinear stage, pancakes are the final end-states of any generic collapsing overdensity. In addition, by looking at the collapse timescales associated with the stationary points, we have found that pancake collapse takes place first and point-like singularities occur last.  Our results for the fate of generic collapsing overdensities are in disagreement with studies of silent-universe dynamics, which argue for spindle-like rather than pancake singularities (see~\\cite{BJ,BMP1}). The reason for this difference lies in the role of the tidal field within the adopted cosmological model. Silent universes allow for non-zero electric Weyl tensor, but set its magnetic counterpart to zero. In the presence of this ``truncated'' tidal field the fluid element evolves towards a Kasner-type singularity, where pancakes are a set of measure zero. Our two-frame analysis of a perturbed Einstein-de Sitter model, on the other hand, shows that the influence of the long-range gravitational field is negligible at second order. Technically speaking, the covariant equations governing the mildly nonlinear collapse of a dust cloud have zero Weyl-curvature input, effectively reducing to the Newtonian ones. As a result, pancakes are reinstated as the natural attractors of a generic collapsing overdensity, this time within the relativistic framework. This is in agreement with numerical simulations which clearly favor pancake formation over all other types of structures (see~\\cite{SMMPT}). Once again we point out that our result holds at second perturbative order and refers to what is known as the mildly nonlinear regime. Although the second order equations can probe deep into the epoch of structure formation, they are expected to break down close to the singularity. Clearly, if we were to address the full picture, we need to incorporate the tide effects. For an Einstein-de Sitter background, however, the electric and of the magnetic Weyl terms are both of third perturbative order (see Eq.~(\\ref{p-shear-prop3})). In other words, the fully nonlinear collapse evolves under the influence of the full tidal field rather than the purely-electric Weyl field of the silent models. In addition, at higher than the second perturbative order, the magnetic Weyl tensor is also a source of peculiar vorticity (see Eqs.~(\\ref{p-vort-prop2}), (\\ref{p-vort-prop3})) and one can no longer consistently argue for irrotational peculiar motion. Of course, as the collapse proceeds beyond the mildly nonlinear stage, pressure gradients also become important, which means that an acceleration-free motion is no longer sustainable, and the whole idea of the Zeldovich approximation is expected to break down.  Note that, although it is usually bypassed, vorticity is an important issue, given the observed rotation of galaxies and galaxy clusters. By construction, the Zeldovich approximation cannot address this question. Newtonian modifications of the standard approach to incorporate vorticity have already been suggested in the past (see~\\cite{BS}). The covariant formalism developed here can provide the basic framework for a relativistic, mathematically rigorous and physically transparent treatment of rotating dust during the nonlinear collapse."
    },
    "0209/astro-ph0209233_arXiv.txt": {
        "abstract": "We examined the VSNET light curve of the unusual SU UMa-type dwarf nova V503 Cyg which is known to show a short (89 d) supercycle length and exceptionally small (a few) normal outbursts within a supercycle. In 1999--2000, V503 Cyg displayed frequent normal outbursts with typical recurrence times of 7--9 d.  The behavior during this period is characteristic to an usual SU UMa-type dwarf nova with a short supercycle length.  On the other hand, V503 Cyg showed very infrequent normal outbursts in 2001--2002.  Some of the superoutbursts during this period were observed shorter than usual.  The remarkable alternations of the outbursting states in V503 Cyg support the presence of mechanisms of suppressing normal outbursts and premature quenching superoutbursts, which have been proposed to explain some unusual SU UMa-type outbursts. The observed temporal variability of the suppressing/quenching mechanisms in the same object suggests that these mechanisms are not primarily governed by a fixed system parameter but more reflect state changes in the accretion disk. ",
        "introduction": "ER UMa stars are a still enigmatic small subgroup of SU UMa-type dwarf novae [for a review of dwarf novae, see \\citet{osa96review}], which have extremely short supercycle lengths ($T_s$ the interval between successive superoutbursts) of 19--50 d [for a review, see \\citet{kat99erumareview}] and regular occurrence of superoutbursts. Only five definite members have been recognized up to now: ER UMa (\\cite{kat95eruma}, \\cite{rob95eruma}, \\cite{mis95PGCV}); V1159 Ori (\\cite{nog95v1159ori}, \\cite{pat95v1159ori}); RZ LMi (\\cite{rob95eruma}, \\cite{nog95rzlmi}); DI UMa (\\cite{kat96diuma}); and IX Dra (\\cite{ish01ixdra}). Some helium-transferring cataclysmic variables have become recognized as ``helium counterparts\" of ER UMa stars [CR Boo: \\citet{kat00crboo}; V803 Cen: \\citet{kat00v803cen}, \\citet{kat01v803cen}].  From a theoretical side, ER UMa stars have been understood as a smooth extension of normal SU UMa-type dwarf nova toward higher mass-transfer rates ($\\dot{M}$) \\citep{osa95eruma}. The exact origin of such a high mass-transfer rate is still a mystery. Even considering a higher mass-transfer rate, the shortest period systems (RZ LMi and DI UMa) are difficult to explain without a special mechanism of prematurely quenching a superoutburst \\citep{osa95rzlmi}.  In recent years, there have been an alternative attempt to explain the ER UMa-type phenomenon.  \\citet{hel01eruma} tried to explain the ER UMa-type phenomenon by considering a decoupling between the thermal and tidal instabilities [see \\citet{osa89suuma} for details of the thermal-tidal instability model] under extremely small binary mass-ratio ($q$=$M_2$/$M_1$) conditions.  \\citet{hel01eruma} speculated that repeated post-superoutburst rebrightenings\\footnote{ These phenomena are sometimes referred to as {\\it echo outbursts}, but we avoid using this terminology because this idea was first proposed to describe the ``glitches\" or ``reflares\" in soft X-ray transients (SXTs) \\citep{aug93SXTecho}.  In SXTs, hard-soft transition is considered to be more responsible for the initially claimed phenomenon \\citep{min96SXTtransition}, which is clearly different from dwarf nova-type rebrightenings. } in WZ Sge-type dwarf novae (hereafter WZ Sge stars) or large-amplitude SU UMa-type dwarf novae (e.g. \\cite{kuu00wzsgeSXT}; see \\citet{kat01hvvir} for a recent observational review of WZ Sge-type stars).  \\citet{bua01DNoutburst} tried to explain ER UMa-type phenomenon by (rather arbitrary) introducing an inner truncation of the accretion disk and irradiation on the secondary star on a numerical model developed by \\citet{ham00DNirradiation}.  \\citet{bua02suumamodel} further tried to explain the unification idea by \\citet{hel01eruma} using the same scheme as in \\citet{bua01DNoutburst}.  Although the results partly reproduced the characteristics of ER UMa stars and WZ Sge stars, they failed to quantitatively reproduce the light curves of these dwarf novae.  From the observational side, the existence of a gap between distributions of ER UMa stars and ``usual\" SU UMa-type dwarf novae has been a challenge. The shortest known $T_s$ in usual SU UMa-type dwarf novae had been 130 d (YZ Cnc, see also Table 1 in \\cite{nog97sxlmi}) at the time of the initial proposition of ER UMa stars.  Although further works have slightly shortened this minimum $T_s$ [SS UMi: 84.7 d, \\citet{kat00ssumi}; BF Ara: 83.4 d, \\citet{kat01bfara}], there still remains a undisputed gap. In addition to these usual SU UMa-type dwarf novae with the shortest $T_s$'s, there exists a seemingly different population of SU UMa-type dwarf novae with short $T_s$'s, but with infrequent normal outbursts. V503 Cyg [$T_s$ = 89 d, only a few normal outbursts in a supercycle \\citep{har95v503cyg}] and CI UMa [$T_s \\sim$ 140 d, infrequent normal outbursts \\citep{nog97ciuma}; $T_s$ variable? \\citep{kat02v344lyr}] are the best-known examples.  The relation, however, between these objects and ER UMa stars (and short $T_s$ usual SU UMa-type dwarf novae) are unknown.  In most recent years, some instances of strong $T_s$ variations have been reported in ER UMa stars (\\cite{fri99diuma}, \\cite{kat01v1159ori}).  In this letter, we report on the dramatic changes in the outburst properties in V503 Cyg. ",
        "conclusions": "\\label{sec:dis}  In the standard disk instability model, the recurrence time of normal outbursts ($T_n$) is mainly governed by the diffusion process, while $T_s$ represents the increasing rate of net angular momentum in the accretion disk \\citep{ich94cycle}.  If the quiescent viscosity parameter has a fixed value between various SU UMa-type dwarf novae, both $T_n$ and $T_s$ are unique functions of $\\dot{M}$ \\citep{ich94cycle}.  This relation has been observationally confirmed in most of SU UMa-type stars \\citep{war95suuma}.  V503 Cyg apparently violates this relation in its low frequency of normal outbursts (figure \\ref{fig:lc}, upper panel), and several other stars (V344 Lyr, SX LMi) have been proposed to be analogous to V503 Cyg (\\cite{kat01sxlmi}, \\cite{kat02v344lyr}). There must be an unknown suppression mechanism of normal outbursts in these systems.  In 1999--2000, V503 Cyg showed a very frequent occurrence of normal outbursts (minimum $T_n \\sim$ 7--9 d, figure \\ref{fig:lc}, middle panel). This $T_n$ is just what is expected for a $T_s$ = 89 d usual SU UMa-type dwarf nova \\citep{war95suuma}. This fact indicates that the usually outbursting SU UMa-type state and unusually outbursting (in the sense of low frequency of normal outbursts) V503 Cyg-type state are interchangeable.  Since $T_s$ during this period was not appreciably different from the canonical $T_s$ = 89 d, there should have not been an appreciable change in the $\\dot{M}$. The suppression mechanism of normal outbursts must have been somehow ``unlocked\" during this period.  In 2001--2002, V503 Cyg showed another different aspect (figure \\ref{fig:lc}, lower panel).  During this period, the number of normal outbursts in a supercycle dramatically decreased to $\\sim$1.  There is some hint of alternating occurrence of a superoutburst and a normal outburst with a period of 40--80 d. Such a sequence of outbursts is only known in rarely outbursting SU UMa-type dwarf novae [cf. SW UMa, V844 Her cf. \\citet{kat00v844her} for a discussion], and is unprecedented in short $T_s$ systems. During this period, some normal outbursts have comparable peak magnitudes to those of superoutbursts.  Some of superoutbursts showed rather short durations, which seems to be incompatible with a high $\\dot{M}$ necessary to reproduce the short $T_s$ \\citep{osa95eruma}. These findings suggest that premature quenching of superoutbursts, as proposed by \\citet{osa95rzlmi} and \\citet{hel01eruma}, indeed occurred during this period, although V503 Cyg (orbital period = 0.0757 d) is unlikely to have a small $q$ required in \\citet{osa95rzlmi} and \\citet{hel01eruma}.  The overall light curve more or less resembles that of CI UMa (\\cite{kol79cpdraciuma}, \\cite{nog97ciuma}).  Although exact mechanisms have not been yet identified, the present remarkable alternations between the outbursting states in V503 Cyg support the presence of mechanisms of suppressing normal outbursts and premature quenching superoutbursts.  The most important finding is that the effects of these mechanisms are temporarily variable even in the same object, and are not a fixed character of a certain system. This finding suggests that the shortest $T_s$ usual SU UMa stars and unusual V503 Cyg-like stars can represent different aspects of the same system.  Among ER UMa stars, DI UMa can be a similar system with systematic state changes \\citep{fri99diuma}.  The observed temporal variability of the suppressing/quenching mechanisms in the same object suggests that these mechanisms are not primarily governed by a fixed system parameter [i.e. mass of the white dwarf \\citep{bua01DNoutburst}; $q$ \\citep{hel01eruma} etc.] but more reflect state changes in the accretion disk.  \\vskip 3mm  We are grateful to many amateur observers for supplying their vital visual and CCD estimates via VSNET. This work is partly supported by a grant-in-aid (13640239, TK) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. Part of this work is supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists (MU)."
    },
    "0209/astro-ph0209005_arXiv.txt": {
        "abstract": "We discuss a scenario in which brown dwarfs are formed like stars, except that their full collapse phases are interrupted through dynamical interactions in small multiple systems, leading to the ejection of the lightest member. This disintegration is a stochastic process, often resulting in the expulsion of newborn low mass stars, but when it occurs early enough the ejected stellar embryo will be a substellar object. This process may be so common at early ages that a large fraction of the ubiquitous brown dwarfs could have formed in this manner.  Detailed gas dynamical simulations are required in order to better understand the details of the decay of small newborn multiple systems.  We discuss the observational consequences of the ejection hypothesis, noting especially the importance of binaries with brown dwarf components as an observational test. Finally, we note that brown dwarfs that have recently been ejected may be so disturbed, by infall from the collapsing core and also by heavy accretion from perturbed circumstellar disks, that traditional spectral and luminosity criteria may fail to identify their substellar nature. ",
        "introduction": "Our understanding of the formation of brown dwarfs has long been shrouded in controversy. Two aspects of their formation appear problematical from a theoretical point of view. The first is that brown dwarf masses are at least 10 times smaller than the typical Jeans mass in star forming molecular clouds. Thus any successful theory needs to produce regions of exceedingly high gas density ($\\sim 10^7$ cm$^{-3}$ for isothermal gas at $10$K). Secondly, once such proto-brown dwarfs have formed, it is necessary that they avoid substantial further accretion of gas from their environment, in order that they remain as low mass entities.  A variety of scenarios have been proposed that address the density issue. Many of these involve the formation of brown dwarfs within disks.  For example, Lin et al. (1998) suggested that the requisite high densities would be produced during the encounter between two massive disks, when a long tidal filament could be flung out, leading to the subsequent formation of a brown dwarf. In loose T associations such encounters are likely to be extremely rare, which is in contrast to the increasing number of brown dwarfs found also in such environments. In the simulations of Bate (2002), however, where stars form in compact groups, such interactions are common and the majority of brown dwarfs form in massive disks that are highly disturbed by dynamical encounters. Li (2002), on the other hand, has shown that brown dwarfs may also form in isolated disks if these are supported by magnetic pressure. By contrast, Padoan \\& Nordlund (2002) have argued that brown dwarfs need not form in disks, but that sufficiently high densities may be produced in the shock compressed regions of turbulent flows; this process they term `turbulent fragmentation' to distinguish it from the above scenarios in which the high densities required to form brown dwarfs are the {\\it result} of self-gravity.  Once brown dwarfs have been produced by any of the above mechanisms, the remaining problem is to prevent their exceeding the hydrogen burning mass limit due to continued accretion. Recently, Whitworth \\& Zinnecker (2002) have suggested that photoionization may erode higher mass gas cores, so that the remnant core is reduced to substellar masses; evidently this mechanism is restricted to regions containing photoionizing OB stars. Otherwise, there are two possibilities - either the statistics of the density field are such that brown dwarfs collapse in isolation, well away from the gas reservoirs that are destined to form stars, or else, if brown dwarfs and stars form from common gas reservoirs, the brown dwarfs must somehow be removed from this environment.  Reipurth \\& Clarke (2001) have developed this latter line of argument, appealing to the dynamical interactions that occur within small N groupings in order to eject brown dwarfs from their natal gas reservoir. The spectacular simulations of Bate (see contribution this volume) illustrate both the propensity of molecular clouds to fragment into small N groupings, and the way that dynamical interactions indeed eject brown dwarfs.  In this contribution, we review the ejected stellar embryo scenario and its observable consequences, first considering the observational evidence that stars are indeed formed in small N groupings. We should stress, however, that the implication is not that all brown dwarfs {\\it must} form this way, but rather that a significant fraction can be expected to be formed in, and ejected from, multiple systems. It is our purpose here to set out the observational discriminants that can be used to assess what fraction of brown dwarfs form in this way. We emphasize that most of these discriminants must be sought in {\\it young} brown dwarfs: after a few hundred million years, the appearance and kinematics of substellar objects depend only on mass and age, and the particular mechanism that produced them is lost in the mist of time. ",
        "conclusions": ""
    },
    "0209/hep-ph0209195_arXiv.txt": {
        "abstract": "This talk is a status report on calculations of the flux of atmospheric neutrinos from the sub-GeV range to $E_\\nu\\sim$~PeV.   In the lower energy range ($E_{\\nu}< 1$~TeV) the primary interest is in using the atmospheric neutrino beam to study neutrino oscillations.  In the TeV range and above, atmospheric neutrinos are a calibration source and background for neutrino telescopes. ",
        "introduction": "The discovery of neutrino oscillations with atmospheric neutrinos makes it important to know the production spectrum of neutrinos as precisely as possible in order to infer the properties and parameters of the oscillations from the data.  It also means that the flux of cosmic-ray induced neutrinos is much better measured than it otherwise might have been.  In addition to the extensive measurements at Super-K,~\\cite{SuperK} there were important measurements at Soudan~\\cite{Soudan} and MACRO.~\\cite{MACRO} The measurements cover a range of energies and techniques, and they see the beam from different locations in the geomagnetic field, which exposes interesting effects at low energy.  Because of oscillations, measuring the cosmic-ray neutrino beam is an iterative process in which the oscillations and fluxes must be understood from the same data.  Fortunately, calculation of the neutrino spectrum at production is straightforward, and it can be checked by comparison to measurements of atmospheric muons. Moreover, the evidence for oscillations is robust because it is based on ratios, which are better known that the absolute normalization of the atmospheric neutrino beam. The anomalous ratio of electron-like to muon like events reveals a relative deficit of $\\nu_\\mu$ at low energy, and the ratio of upward to downward multi-GeV events reflects the pathlength dependence of $\\nu_\\mu$ oscillations and defines a range of $\\delta m^2$.~\\cite{SuperK} All this is reinforced by the low ratio of stopping to throughgoing upward, neutrino-induced muons~\\cite{M2,SK2} and by the low ratio of vertically upward to horizontal throughgoing muons.~\\cite{M3,SK3} A consistent pattern of energy and pathlength dependence emerges that clearly points to neutrino oscillations as the explanation. Because atmospheric $\\nu_e$ behave normally to the precision measured so far, the main effect must lie in the $\\nu_\\mu\\leftrightarrow\\nu_\\tau$ sector (or involve sterile neutrinos, which are now disfavored).~\\cite{SK4,MACRO}  Here I review the main features of the calculation of the atmospheric neutrino beam at production, emphasizing the simple features that provide the basis for the evidence for oscillations.  This talk is based to a large extent on a recent review~\\cite{GH}.  I organize the material here in order of increasing energy.   \\noindent \\begin{figure}[t] \\vspace{-.5cm} \\flushleft{\\epsfig{figure=gaisf2.ps,width=8.5cm}} \\label{fig1} \\vspace{-1cm} \\caption{Distribution of neutrino energies that give rise to several classes of neutrino events.} \\end{figure}  \\noindent \\begin{figure}[h] \\flushleft{\\epsfig{figure=resp-sub.ps,width=7.5cm}} \\label{fig2} \\vspace{-1cm} \\caption{Response functions for sub-GeV neutrinos under several conditions (see text).  The three pairs of curves show response functions for the distribution of neutrino energies in the solid curve (S-K $E_\\nu$), which is the same as the left-most curve of Fig.~1.} \\end{figure}  The atmospheric neutrino flux is a convolution of the primary spectrum at the top of the atmosphere with the yield ($Y$) of neutrinos per primary particle. To reach the atmosphere and interact, the primary cosmic rays first have to pass through the geomagnetic field. Thus the flux of neutrinos of type $i$ can be represented as \\begin{eqnarray} \\label{nuflux} \\phi_{\\nu_i} & = & \\phi_p\\,\\otimes\\,R_p\\,\\otimes\\,Y_{p\\rightarrow\\nu_i}\\\\ \\nonumber &  & +\\;\\sum_A \\left\\{\\phi_A\\,\\otimes\\,R_A\\,\\otimes\\,Y_{A\\rightarrow\\nu_i} \\right\\}, \\end{eqnarray} where $\\phi_{p(A)}$ is the flux of primary protons (nuclei of mass A) outside the influence of the geomagnetic field and $R_{p(A)}$ represents the filtering effect of the geomagnetic field. Free and bound nucleons are treated separately.    Each of the factors on the right side of Eq.~\\ref{nuflux} is a potential source of uncertainty.  Since the uncertainties depend on energy, one needs an estimate of the relative importance of different primary energies for a given region of neutrino energy.  Fig.~1~\\cite{Engel} shows the distributions of neutrino energies for four classes of events. Very roughly, sub-GeV events, multi-GeV events, upward stopping muons and upward throughgoing muons correspond respectively to primary cosmic-ray energies of $10^{1\\pm0.5}$, $10^{1.5\\pm0.5}$, $10^{2.0\\pm0.5}$ and $10^{3.0\\pm1}$~GeV.  Fig.~2 shows the response function in detail for the sub-GeV events, as defined at Super-K~\\cite{SuperK}.   \\noindent \\begin{figure}[t] \\flushleft{\\epsfig{figure=nulin3c-sk-fluxes.ps,width=7.5cm}} \\label{fig3} \\vspace{-1cm} \\caption{Comparison of neutrino flux for three primary spectra for the location of Kamioka.} \\end{figure}  \\noindent \\begin{figure}[t] \\flushleft{\\epsfig{figure=nulin3c-sd-fluxes.ps,width=7.5cm}} \\label{fig4} \\vspace{-1cm} \\caption{Comparison of neutrino flux for three primary spectra for the location of Soudan.} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209574_arXiv.txt": {
        "abstract": "We discuss some clues on the large tilt observed in the Horizontal Branch of the metal-rich galactic globular clusters NGC 6388 and NGC 6441. This not yet understood feature is investigated from theoretical and observational sides. ",
        "introduction": "Recently Brocato et al. (1999) discussed the presence of a tilt of the order of $\\Delta V \\simeq 0.1\\, mag$ in the HB morphology of the intermediate metallicity globular cluster NGC 6362 [$\\Delta V$ is the magnitude difference between the top of the blue HB and the fainter magnitude reached by the red HB (RHB)].  On the other hand, much larger tilts ($\\Delta V \\simeq 0.5\\, mag$) are observed in the metal rich clusters NGC 6388 and NGC 6441 of the inner Milky Way (Rich et al. 1997; Sweigart and Catelan 1998).  In the present work we investigate evolutionary predictions concerning the Color-Magnitude Diagram (CMD) of metal rich HB stars and arguments that can constraint possible explanations of this not yet understood feature. ",
        "conclusions": "The large tilt of the RHB of NGC 6441 and NGC 6388 seems to be a feature observed in many metal rich clusters. Up to now, a fully satisfactory explanation is not yet found. More precise observations (i.e. FLAMES@VLT) are urgently needed to properly constraint our understanding of metal rich stellar systems.  \\begin{figure}[t] \\plotfiddle{raimondofig1.ps}{0in}{0}{50}{40}{-150}{-280} \\vspace{7.5cm} \\caption{HBs of six well populated clusters in Piotto et al.  (2002) data-base.  The solid line is the slope $dV / d(B-V)= 1.5$ (see text). } \\end{figure}"
    },
    "0209/astro-ph0209097_arXiv.txt": {
        "abstract": "VLA, MERLIN and Hubble Space Telescope imaging observations of the extended regions of the symbiotic system CH Cygni are analysed. These extensions are evidence of a strong collimation mechanism, probably  an accretion disk surrounding the hot component of the system. Over 16 years (between 1985 and 2001) the general trend is that these jets are  seen to precess. Fitting a simple ballistic model of matter ejection to the geometry of the extended regions suggests a period of  $6520 \\pm 150$ days, with a precession cone opening angle of $35 \\pm 1$ degrees. This period is of the same order as that proposed for the orbital period of the outer giant in the system, suggesting a possible link between  the two. Anomalous knots in the emission, not explained by the simple model, are believed to be the result of older, slower moving ejecta, or possibly jet material that has become disrupted through sideways interaction with the surrounding medium. ",
        "introduction": "Symbiotic stars occupy an extreme and relatively poorly understood region of the binary classification scheme. The name was coined by \\citet{PWM41} to describe stars which appeared to  have a combination spectrum: that of high excitation lines usually associated with a hot, ionised nebula superimposed on a cool continuum with prominent absorption features consistent with a late--type star. At present they are understood to be interacting binaries (with orbital periods of a few to tens of years) consisting of a cool giant losing material mostly via the stellar wind and a hot, luminous compact object which ionises a portion of the cool component wind \\citep{ERS84}.  Such a state  of affairs represents the so--called {\\em quiescent phase}, which can be interrupted by periods of activity. The {\\em active phases} start with an eruption of the hot star, an event indicated photometrically by an increase of the star's brightness by 2-6\\,mag, and/or spectroscopically by high velocity broad emission features from the central star. Both radio and Hubble Space Telescope (HST) imaging can directly resolve the remnants of such dramatic events (\\citealt{WJH93}; \\citealt{SPSE95}; \\citealt{HTK96}; \\citealt{AMSR99}; \\citealt{SPSE01a}).  The symbiotic star CH\\,Cygni displays particularly complex behaviour. Optical spectroscopic studies by \\citet{MM87} showed that the orbits of the stars within the system are likely to be coplanar and eclipsing, with eclipses separated by around 5700 days.  Later studies (\\citealt{MM90b} and references therein) confirmed this period.  Further spectral studies of the system \\citep{KHH93} suggested that CH\\,Cyg is probably a triple-star system consisting of an inner 756-day period binary which is orbited by an unseen G-K dwarf on a 5300~day orbit. \\citet{AS96b} uncovered 756 day interval eclipses which show that all three stars in the system are likely to be in coplanar orbits.  Each of the three outbursts seen since 1978 was accompanied by high velocity broad emission features consistent with mass outflows. During 1984-85 the material was ejected at $\\sim$600-2500\\,km\\,s$^{-1}$ \\citep{MM86}, whilst the 1992--95 active phase was characterized  by sporadic and in part bipolar outflow at $\\sim$1000-1600\\,km\\,s$^{-1}$ (\\citealt{LL96}; \\citealt{AS96a}; \\citealt{TI96}) and finally, during the recent, 1998--2000, outburst mass outflows at about 1000\\,km\\,s$^{-1}$ were observed (\\citealt{SPSE01b},  hereafter Paper II). The outflows may be correlated with a significant increase of the radio emission and the radio light curves during these periods fit well with the optical ones \\citep{HTK96}. The 1984 mass ejection has been linked to the emergence of bipolar emission \\citep{ART86} which has been attributed to high velocity ($\\sim$1000\\,km\\,s$^{-1}$) jets. Non--thermal emission features have  been discovered in these jets from radio observations (\\citealt{MMC01},  hereafter Paper I), explained by shocked regions that arise when the high--speed ejecta interact with existing wind material.  This paper aims to extend our understanding of the jets in this interesting system by developing theories and models to explain the change in their morphology in the years following their emergence.  In this paper we present analysis of radio data from the VLA and MERLIN, along with HST observations. ",
        "conclusions": "It is certain that the bipolar jets of CH Cygni exhibit precession, with a period controlled by the motion of the mechanism resposible for the collimation  of the outflow. In addition, the variability of the mass--loss rate from the cool component causes the material to be ejected at a rate that is highly variable (between $\\sim 500$ and $2000$km~s$^{-1}$), leading  to complications in predicting the outflow geometry. The activity of the bipolar ejection seems to be tied to the state of the hot--component,  with fast jets seen at or soon after times of optical outburst. This activity then decreases during times of quiescence.  Although the precession period found here is similar to the orbital period of the outer giant in the system, the current time resolution of the observations does not rule out shorter periods. A precession that resulted from the motion of the inner, symbiotic, pair and had a similar period to its 756~day orbit would need observations to be taken at least every year for it to be detected.  The simple ballistic model fits the geometry of the nebula extremely well. Knots seen in the larger--scale radio maps (the VLA in C, X and U band) that do not fit the simple ballistic model can be explained as regions of jet disruption caused by sideways motion of the ejected material brought about as a result of the precession. The predicted distance at which this disruption would occur is in close agreement with the observations, although it relies upon several assumptions about the nature of the ambient material. Confirmation of this model would require simultaneous optical images in both H$\\alpha$ and [S II].  An alternative explanation is that the anomalous material was ejected at a much slower velocity and at an earlier time than the majority of the gas in the bipolar nebula. This would indicate the presence of a minor ejection event prior to the main ejection of the jets first seen by \\citet{ART86}. The variable ejection velocity model discussed here and the Echelle spectroscopy of \\citet{RLC01} suggest the emission of material with a velocity of no more than a few 100s of km$^{-1}$ just before the main ejection of the high--speed material.  The cause of the precession is unknown but, given realistic estimates for the masses and separation of the stellar components, warping of the collimating accretion disk by a magnetic white dwarf is the only mechanism that can give precession periods similar to those found by model fitting. The other possible causes give periods that are several orders of magnitude too large. The magnetic explanation also allows for the one--sided nature of the emission seen in all VLA radio maps following the initial outburst in 1984."
    },
    "0209/astro-ph0209268_arXiv.txt": {
        "abstract": "We present the results of first deep optical observations of the field of the old ($\\sim 10^8$ yr), nearby, isolated pulsar J0108$-$1431, in an attempt  to detect its optical counterpart.   The observations were performed  using the  FORS1 instrument  at the  focus of  the European Southern Observatory Antu Telescope  of the VLT. Observations with the Australia Telescope  Compact Array (ATCA) were made to  determine an accurate position for the radio pulsar  at the current epoch. The imaging data, obtained in the  $V$, $B$, and $U$ passbands  reveal no counterpart at the revised radio  position down to $V \\simeq  28$, $B\\simeq 28.6$ and $U  \\simeq  26.4$.  For  a  distance of  130  pc,  estimated from  the pulsar's dispersion  measure, our constraints on the  optical flux put an upper limit of $T=4.5\\times  10^4$ K for the surface temperature of the neutron star, assuming a stellar radius $R_\\infty=13$ km.  Our new radio position allows us to place  an upper limit on the pulsar proper motion of 82~mas~yr$^{-1}$ which, for $d=130$ pc, implies a transverse velocity $\\la 50$~km~sec$^{-1}$. ",
        "introduction": "PSR  J0108$-$1431  was discovered  by  Tauris  et  al.\\ (1994)  during observations performed as a part  of the Parkes Southern Pulsar Survey (Manchester et  al.\\ 1996).   With a period  $P=0.808$ s and  a period derivative  $\\dot{P} =  7.44 \\times  10^{-17}$ s~s$^{-1}$  (D'Amico et al.\\ 1998),  the pulsar  has a characteristic  age $P/2\\dot{P}  = 170$ Myr,  rotation  energy   loss  rate  $\\dot{E}=5.6\\times  10^{30}$  erg s$^{-1}$, and magnetic field  $B=2.5\\times 10^{11}$ G.  The dispersion measure of  $2.38 \\pm 0.01$  cm$^{-3}$~pc (D'Amico et al.\\  1998), the lowest observed so far for a radio pulsar, puts \\psr\\ at a distance of about 130 pc, estimated according to the Taylor \\& Cordes (1993) model for the Galactic electron  density distribution.  Tauris et al.\\ (1994) suggest that the local electron density may be greater than average in this direction, giving a distance as  low as 60 pc.  Regardless of the exact value,  the small distance  to PSR J0108$-$1431 implies  a radio luminosity  a factor  $10^3$ smaller  than the  average for  other old pulsars of similar ages.  This  pulsar could thus be representative of a population of sub-luminous radio pulsars (e.g., Dewey et al. 1985).  Since \\psr\\  is apparently the closest  known pulsar, it  is a natural candidate for observations in other wavelength bands.  However, it has not yet been  detected outside the radio range.   Presumably too faint to be detected  in the {\\sl ROSAT} All Sky Survey,  \\psr\\ has not been observed in  a pointing  mode with an  X-ray observatory so  far.  The field was  observed with the {\\sl Extreme  Ultraviolet Explorer} ({\\sl EUVE}), but the pulsar was  not detected (Korpela \\& Bowyer 1998).  In the   mid-infrared,  observations   with  the   {\\sl   Infrared  Space Observatory} were performed to search for a disk around the pulsar but with no conclusive result  (Koch-Miramond et al.\\ 2002).  Particularly interesting  would be  the detection  of optical-UV  thermal radiation from  the  neutron  star  (NS)  surface  to  examine  various  heating mechanisms  that can  operate in  the NS  interiors.  According  to the current models (see, e.g., Tsuruta 1998,  for a review), by the age of 170  Myr  a   NS  would  have  cooled  down   to  very  low  (surface) temperatures,  $T  < 10^4$  K,  and  its  thermal radiation  would  be virtually undetectable unless some (re)heating mechanisms operate.  Amongst several proposed heating mechanisms, two are most efficient in slowly rotating pulsars.   The first one is the  dissipation of energy of differential  rotation.  As invoked  by models of  pulsar glitches, the interior  neutron superfluid rotates  more rapidly than  the outer solid  crust.  Frictional  interaction with  the crust  dissipates the energy of  the differential rotation,  heating the star  (Shibazaki \\& Lamb 1989;  Umeda et al.~1993; Van  Riper et al.~1994;  Larson \\& Link 1999).   The  amount of  heat  released  depends  on the  differential angular momentum  $\\Delta J_s$ of the  frictionally coupled superfluid layers.  For instance, at an  age $\\tau\\sim 200$ Myr, plausible values of  $\\Delta  J_s$  can   provide  $T\\simeq  (3$--$10)\\times  10^4$  K, depending on the equation of  state of the NS interiors and properties of  nucleon  superfluidity.   The  other heating  mechanism  is  Joule heating caused by  dissipation of the magnetic field  in the NS crust. According to Miralles, Urpin \\&  Konenkov (1998), the NS luminosity at $\\tau \\gtrsim$  10 Myr is  approximately equal to the  energy released due to the field dissipation.   At this stage, the surface temperature decreases slowly, being comparable  to that produced by the frictional heating at $\\tau \\sim 100$ Myr.  Thus, if  the heating  mechanisms indeed operate,  one can  expect the surface  NS  temperature of  a  few  times  $10^4$~K, which  would  be detectable  in  the  optical-UV  range  but  undetectable  in  X-rays, traditionally used for observations of thermal radiation from NSs.  It is  also possible that  the  pulsar emits  nonthermal optical radiation.   So far, this  has only been firmly  detected from younger pulsars  (e.g., the Vela  pulsar --- Mignani \\&  Caraveo 2001; PSR B0656+14 --- Koptsevich et al.\\ 2001).  The  only  previous  observations   of  \\psr\\  in  the  optical  range ($B$,$V$,$R$,$I$  bands) were carried  out with  the 6-m  telescope of Special  Astrophysical Observatory  (Russia), but  they were  not deep enough  to detect  the pulsar  and provide  useful constraints  on the radiation mechanisms  (Kurt et al.\\ 2000).  Therefore,  we performed a deep observation  of the pulsar field  with one of the  ESO Very Large Telescopes (VLT) units in several passbands.  The most recent published  coordinates of PSR J0108$-$1431 (D'Amico et al.\\ 1998) were derived from timing data obtained in the interval 1993 April  to 1996 June,  giving a  mean epoch  of 1994.9.  Therefore, our knowledge of the actual position of the pulsar at the epoch of our VLT observations  is affected  by  a significant  uncertainty  due to  its unknown proper motion.   A pulsar distance of 130  pc and a transverse velocity  of 400  km  s$^{-1}$, a  typical  value for  a radio  pulsar (Lorimer,  Bailes \\&  Harrison 1997),  give a  proper motion  of $\\sim 0\\farcs65$ yr$^{-1}$.  For the epoch of our VLT observations (2000.6), such  a  proper  motion  would  imply a  displacement  in  an  unknown direction of $\\sim 3\\farcs5$ with respect to the original radio timing position.    This  obviously   makes   impossible  a   straightforward positional search  for the optical  counterpart.  For this  reason, we observed the pulsar with  the Australia Telescope Compact Array (ATCA) to obtain  a radio position at an  epoch close to that  of our optical observations and to measure the pulsar's proper motion.  We  describe the  radio and  optical  observations in  \\S2, while  the results and their implications are  presented and discussed in \\S3 and \\S4, respectively. ",
        "conclusions": "The upper limits on spectral flux at effective frequences of the three bands (Fukugita, Shimasaku, \\& Ichikawa 1995) are plotted in Figure 4. The deepest $B$-band limit corresponds to the  following limit on the (brightness) temperature  \\begin{equation} T_\\infty<\\frac{3.30\\times   10^4~{\\rm   K}}{\\ln\\left(1+ 1.08\\, R_{13}^2 d_{130}^{-2}\\right)}\\,, \\end{equation} where $R_{13}=R_\\infty/(13~{\\rm km})$, $d_{130}=d/(130~{\\rm pc})$, the subscript $\\infty$ denotes the quantities as measured by a distant observer: $T_\\infty=Tg_r$, $R_\\infty=R/g_r$, $g_r=(1-2GM/Rc^2)^{1/2}$ is the gravitational redshift parameter ($g_r=0.769$ for a neutron star mass $M=1.4 M_\\odot$ and radius $R=10$ km.)  In this estimate, we neglect corrections due to interstellar extinction as they are expected to be negligible at the small distance.  For the dispersion distance $d=130$ pc and radius $R_\\infty = 13$ km, we obtain $T_\\infty < 4.5\\times 10^4$ K. This value is considerably lower than $T_\\infty \\la 1.3\\times 10^5$ K, which follows from the upper limit on the {\\sl EUVE} flux of this pulsar (Korpela \\& Bowyer 1998), for the same $d$ and $R_\\infty$, and $N_{\\rm H} < 3\\times 10^{19}$ cm$^{-2}$.  It is also much lower than the lowest upper limit, $T_\\infty\\la 3\\times 10^5$ K, estimated from optical/UV observations of another nearby pulsar, B0950+08~\\footnote{The limit of $7\\times 10^4$ K in the paper of Pavlov et al.\\ (1996) was estimated for $d=127$ pc.  The distance $d=262\\pm 5$ pc was obtained by Brisken et al.\\ (2002) from improved parallax measurements.} --- see Pavlov, Stringfellow, \\& C\\'ordova (1996).  The limiting temperature is even lower for $d=60$ pc, suggested by Tauris et al.\\ (1994): $T<1.8\\times 10^4$ K. We plot the blackbody spectra at these values of $T_\\infty$ and $d$ in Figure 4.  The corresponding limits on bolometric luminosity, $L_{{\\rm bol},\\infty}=L_{\\rm bol} g_r^2 =4\\pi R_\\infty^2 \\sigma T_\\infty^4$, are $4.9\\times 10^{27}$ and $1.4\\times 10^{26}$ erg s$^{-1}$, for $d=130$ pc and 60 pc, respectively. We adopt the more conservative 130 pc estimate in the following discussion and consider $T_\\infty < 4.5\\times 10^4$ K and $L_{{\\rm bol}, \\infty} < 5\\times 10^{27}$ erg s$^{-1}$ as plausible upper limits.  The limit on thermal emission strongly constrains possible heating mechanisms. For the frictional heating mechanism (see \\S1), one can constrain the excess angular momentum $\\Delta J_s$, residing in the superfluid, and the angular velocity lag $\\bar{\\omega}$, between the superfluid and the crust, averaged over the superfluid moment of inertia. For instance, using equation (9) of Larson \\& Link (1999), we obtain $\\Delta J_s=L_{{\\rm bol},\\infty}/|\\dot{\\Omega}| < 7\\times 10^{42}$ erg s, $\\bar{\\omega} = \\Delta J_s/I_s < 1$ rad s$^{-1}$, where $\\dot{\\Omega} = -7.16\\times 10^{-16}$ rad s$^{-2}$ is the time derivative of the angular frequency of the pulsar, $I_s=7.3\\times 10^{43}$ g cm$^2$ is the moment of inertia of the portion of the superfluid that is differentially rotating, estimated for the Friedman \\& Pandharipande (1981) equation of state.  The upper limit on the lag is well below the maximum lag, $\\bar{\\omega}\\sim 10$ rad s$^{-1}$ (e.g., Van Riper et al.\\ 1995), sustainable by superfluid vortices before unpinning from the crust lattice.  The rate of Joule heating caused by dissipation of the magnetic field in the NS crust depends on the strength $B_0$ of the magnetic field generated in the crust during or shortly after NS formation, the density $\\rho_0$ of at the crust bottom, the crust impurity parameter $\\xi$ that determines the crust conductivity, and equation of state of the matter in the NS interior (Miralles et al.\\ 1998).  Comparing our limit with the results of Miralles et al., we can conclude that, for instance, if $B_0\\ga 10^{13}$ G, the values of $\\rho_0>10^{14}$ g cm$^{-3}$ are excluded for a plausible range of impurity parameter, $0.001<\\xi<0.01$, for both Friedman \\& Pandharipande (1981) and Pandharipande \\& Smith (1975) equations of state.  The upper limit on optical flux gives also a constraint on the temperature and size of hot polar cap(s) predicted by radio pulsar models (e.g., Beskin, Gurevich, \\& Istomin 1993): $T_{\\rm cap} < 5 \\times 10^6 R_{\\rm cap,1}^{-2} d_{130}^2$~K, where $R_{\\rm cap,1}$ is the effective cap radius in km. However, this constraint is not truly restrictive because the limiting temperature is improbably high, $\\sim 2\\times 10^8$~K, for the standard estimate of polar cap radius, $R_{\\rm cap} = R (2\\pi R/cP)^{1/2} \\sim 0.16$~km. X-ray observations would be more efficient for observing such small, hot polar caps.  The limiting $B$ magnitude allows one to estimate a limit on the X-ray flux assuming a non-thermal energy spectrum with slope $\\alpha=-0.5$ as observed for several middle-aged pulsars (e.g., Koptsevich et al.\\ 2001).  We obtain $F_{\\rm x} < 4.6\\times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$ in the 0.1--2.4 keV ({\\sl ROSAT}) energy band (or $<1.1\\times 10^{-14}$ erg cm$^{-2}$ s$^{-1}$ in the 0.1--10 keV band). Such a flux corresponds to an upper limit on X-ray luminosity $L_{\\rm x}< 9.4\\times 10^{27}$ erg s$^{-1}$ (in 0.1--2.4 keV), close to the value $6\\times 10^{27}$ erg s$^{-1}$ expected from the empirical dependence $L_{\\rm x}\\approx 10^{-3}\\dot{E}$ found for a sample of radio pulsars observed with {\\sl ROSAT} (Becker \\& Tr\\\"umper 1997). Even if the nonthermal radiation is a factor of a few lower than this upper limit, it can be detected with the {\\sl XMM-Newton} and {\\sl Chandra} X-ray observatories.  Clearly, it would be desirable to have a better estimate of the pulsar distance.  The pulsar's  ecliptic latitude  is about  $10\\degr$,  so a parallax  measurement using  the NRAO  Very Long  Baseline  Array, for example,   is  quite   feasible.  It   is  also   possible   that  the electron-density  model  could be  improved,  giving  a more  reliable dispersion-based distance estimate. However, this will be difficult to achieve since the pulsar is so close."
    },
    "0209/astro-ph0209118_arXiv.txt": {
        "abstract": "{ At present the possible existence of planets around the stars of a close binary system is still matter of debate. Can planetary bodies form in spite of the strong gravitational perturbations of the companion star? We study in this paper via numerical simulation the last stage of planetary formation, from embryos to terrestrial planets in the $\\alpha$ Cen system, the prototype of close binary systems. We find that Earth class planets can grow  around $\\alpha$ Cen A on a time-scale of 50 Myr. In some of our numerical models the planets form directly  in the habitable zone of the star in low eccentric orbits. In one simulation two of the final planets are in a 2:1 mean motion resonance that, however, becomes unstable after 200 Myr. During the formation process some planetary embryos fall into the stars possibly altering their metallicity. ",
        "introduction": "Planetary formation in close binary systems is still an open problem. Accretion disks have already been observed around each individual component \\citep{rod98} strongly supporting the idea that at least the initial stage of planetary formation can occur. However, it is still uncertain  whether the dust can coalesce into planetesimals and, in particular, whether the process of planetesimal accretion can continue till the completion of a planet. \\cite{msc00} showed that relative velocities between small planetesimals are low in spite of the strong gravitational perturbations due to the companion star. The combined effects of gas drag and secular perturbations induce a sharp periastron alignment of the planetesimal orbits and, as a consequence, low relative speeds at impact that favor accumulation rather then fragmentation. However, the gravitational perturbations from the companion star may still halt the planetary formation process in the last stage when planetary embryos collide together to form a planet. In this paper we numerically model this final scenario where large embryos collide and, eventually, form larger bodies that can be termed planets.  We have concentrated on the Alpha Centauri system, the prototype of close binary systems with the more massive star very similar to our Sun. It is also the same system studied by \\cite{msc00}. Planetary embryos are likely to form within 2.5 AU of the star since beyond the companion perturbations are too strong and planetesimal collisions may lead preferentially to fragmentation rather then accretion \\citep{msc00}.  Moreover, orbits farther than 2.7 AU become unstable according to \\cite{hwi99}. We model the accumulation of Lunar--size embryos into terrestrial planets to estimate the time-scale of formation and the dynamical and physical properties of the final planets. Different initial conditions are considered to have a statistical description of the possible final systems. ",
        "conclusions": "Planetary formation seems to be possible in the Alpha Centauri system and, in general, in close binary systems. \\cite{msc00} showed that planetesimals can accrete into planetary embryos; in this paper we demonstrate that planetary embryos can grow into planets in about 50 Myr. These terrestrial--size planets form within the region of planetary stability found by  \\cite{who97} and are possibly stable over the age of the system.  Assuming that the initial mass in the protoplanetary disk around the main component of the $\\alpha$ Cen system was comparable to that of the minimum solar nebula, planets with the size of Mars and Venus are typically formed with low orbital eccentricity. Some of them lay within the habitable zone of the star and, possibly, could harbor life.  During the accumulation process, some of the planetary embryos are engulfed by the stars in different amounts. However, only if the protoplanetary disk was very massive, at least five times more than the minimum mass solar nebula, we might observe a difference in the metallicity of the star related to the ingestion process.  Orbital resonances may occur between the planets. In a significant fraction of our numerical simulations we found an orbital period close to a low order commensurability ratio. In one case we also found a real 2:1 resonance between two planets, with the critical argument librating around 0. This system was stable over 200 Myr after which the outer planet entered a chaotic state and was ejected out of the system."
    },
    "0209/astro-ph0209432_arXiv.txt": {
        "abstract": "Results  of 3-month  continuous monitoring  of turbulence  profile and seeing at Cerro  Tololo (Chile) in May-July 2002  are presented.  Some 28000 low-resolution profiles were  measured by a new MASS single-star turbulence monitor, accompanied by  seeing data from DIMM.  The median seeing  was 0.95  arcseconds. The  first 500~m  contribute 60\\%  to the total seeing,  the free-atmosphere median seeing  was 0.55 arcseconds. Free-atmosphere  seeing is  almost never  better than  0.15 arcseconds because there is always some turbulence above 12~km. A 4-day period of calm upper atmosphere with  a stable free-atmosphere seeing of 0.2-0.3 arcseconds was noted.  A gain in resolution from adaptive compensation of ground layer will be 1.7  times typically and 2-3 times during such calm periods.  Correlations of the free-atmosphere turbulence with the wind  speed at  tropopause  and of  the  ground-layer turbulence  with ground  wind  are  studied.    Temporal  evolution  of  turbulence  is characterized  by recurrent bursts,  their typical  duration increases from 15 minutes  in low layers to 1-2 hours in  high layers. The large data  base of  turbulence  profiles  can be  used  to test  meso-scale modeling of astronomical seeing. ",
        "introduction": "A crucial role of  `seeing' in ground-based astronomy was recognized long time  ago.  Nowadays  it is possible  to improve the  seeing with adaptive optics  (AO), but this  technology is itself so  dependent on the properties  of turbulence  that it generated  a new  and important driver for  detailed atmospheric studies.  AO requires a  knowledge of the temporal time constant and  of the vertical turbulence profile, in addition to the overall (integrated) seeing. It is desirable to have a reliable statistics of  these parameters for a given  site in order to predict  the performance  of AO  systems.  A  real-time  monitoring of optical  turbulence would help  in optimizing  the AO  operation.  For example, the  scintillometer of Ochs.  et al.   was regularly operated at the AMOS station for this reason \\citep{Chonacky88}.  The   vertical  distribution  of   the  optical   turbulence  strength (characterized  by the  altitude  dependence of  the refractive  index structure    constant   $C_n^2$)    is   notoriously    difficult   to monitor. Balloon-born micro-thermal probes  are expensive and sample the turbulence profile (TP) only  once per flight, without any statistical averaging.   Optical remote sounding  by SCIDAR \\citep{Fuchs98}  is free from  this  drawback, but  it  requires  moderately large  telescopes, sensitive detectors, and powerful  signal processing. For these reasons SCIDAR was only used in  a campaign mode at existing observatories.  A limited  number of TPs measured world-wide  revealed that turbulence is typically concentrated in  few thin layers.  The physical mechanism generating  such  distribution  was  studied by  \\citet{Coulman}.   It inspired designers of AO systems  to add more deformable mirrors, each conjugated to its own layer, and thus to compensate seeing over a much wider  field  with  such  multi-conjugate AO  (MCAO).   The  promising potential  and wide  popularity of  MCAO added  even more  pressure to measure turbulence profiles; the Gemini site testing campaign at Cerro Pach\\'on   \\citep{Vernin2000,Avila2000}   is   an  example   of   such MCAO-driven study.  Ground-based telescopes of next  generation with apertures of 20-100~m will include  turbulence compensation already in  their designs. Sites for these telescopes are being  selected with a strong weight given to AO-related  turbulence  parameters;   site  surveys  based  on  seeing measurements alone,  as was  the case for  the previous  generation of telescopes, are  no longer sufficient.  Seeing is  very much dominated by local and orographic effects  that diminish predictive power of seeing  data.  With  modern  computers,  a  modeling of  optical turbulence becomes  feasible, giving new insights into  the physics of seeing    and    new    guidance    to    the    choice    of    sites, e.g. \\citep{Masciadri2001}.  But computer  models still need real TPs for their calibration.  A   low-resolution    turbulence   profile   monitor,   Multi-Aperture Scintillation Sensor (MASS), was developed in response to the needs of AO  and  MCAO, as  well  as a  portable  instrument  for site  testing \\citep{Marrakesh,MASS}.  MASS  was operated in 2002  for several months at the Cerro Tololo Inter-American Observatory (CTIO) jointly with the Differential Image Motion  Monitor (DIMM) \\citep{CTIODIMM}.  This paper presents  the results  of this  campaign. It  appears to  be  the most extensive data base of turbulence profiles existing to date worldwide.   Our aim was to gain  some understanding of the turbulence localization above  CTIO.   We were  specifically  interested  in  the fraction  of turbulence in the first few hundred  meters over the ground and in the seeing that  can be attained if  these low layers  were compensated by AO. {\\em  Ground-layer compensation} offers improved seeing  in a much wider field than does classical AO \\citep{Rigaut2001}.  This option is being studied  for the 4.2-m  SOAR telescope located close  to CTIO on Cerro Pach\\'on \\citep{SOAR}, as well as for extremely large telescopes of  next  generation.  Our  work  quantifies  the  gain expected  from ground-layer compensation at a specific good astronomical site, CTIO.  In Sect.~2 we briefly describe  the instrumentation used in this study and give typical examples of the data.  The statistics of the vertical turbulence  distribution  is explored  in  Sect.~3.  Sect.~4  contains summary and conclusions. ",
        "conclusions": "For a period of few months we followed with amazement the evolution of optical turbulence over Cerro Tololo, for the first time being able to know where the `seeing'  comes from and why it  changes.  The database of some  28000 low-resolution  TPs,  most of  which  are complemented  by seeing data, is unique by its volume and time coverage. The insights gained from these data can be summarized as follows:  \\begin{enumerate} \\item Ground layer turbulence (first 500~m)  at CTIO contributes 60\\% of the total  turbulence  integral  in   50\\%  of  cases.  Thus,  a  complete compensation  of this  layer  would typically  improve  the seeing $0.4^{-3/5} =1.7$ times.  \\item The  median  free-atmosphere  seeing  $\\epsilon_f$ (all  layers  above 500~m) is $0\\farcs 55$,  in 10\\% of cases it is  better than $0\\farcs 28$, but it  is practically never  better than  $0\\farcs 15$.  The  effective lower limit to  $\\epsilon_f$ is related to the  ever-present weak turbulence in the upper tropospheric layers above 12~km.  \\item The periods  of stably calm upper  atmosphere with $\\epsilon_f<0\\farcs 25$ can be as long as few days.  This occurs when the wind velocity at 12~km a.s.l. is around 20-30  m/s.  During these periods, a resolution gain from ground layer AO compensation will be 2-3.  \\item The  characteristic  time   of  turbulence  variation  increases  with increasing layer altitudes, from 15 min. (at 50\\% correlation level) at 1~km  to 1-2 hours  at 16~km.   Often the  turbulence at  altitudes of 1-8~km has a character of  recurrent strong bursts that last for $\\sim 0.5$ hour and repeat every 1-2 hours.  \\end{enumerate}  Perhaps the most  important impression from the data  is the fragility of  astronomical  seeing.  Most  of  the  seeing  results from  local orographic effects  and is  significantly influenced by  very unstable ground-layer  turbulence. A  common opinion  that all  good  sites are similar and  have a median  seeing around $0\\farcs 7$ is  in contradiction with  the  complexity  of   turbulence  phenomena  evidenced  by  this study. We believe that a  better understanding and modeling of optical turbulence is possible and will  help to choose `lucky summits' that are  much  less  affected  by turbulence  generated  near  surrounding mountains. Statistical data on TP will be essential for this work."
    },
    "0209/astro-ph0209162_arXiv.txt": {
        "abstract": "For the inflaton perturbations it is shown that the evolution of the difference between the spectral indices can be translated into information on the scale dependence of the tensor to scalar amplitudes ratio, $r$, and how the scalar field potential can be derived from that information. Examples are given where $r$ converges to a constant value during inflation but dynamics are rather different from the power--law model. Cases are presented where a constant $r$ is not characteristic of the inflationary dynamics though the resulting perturbation spectra are consistent with the CMB and LSS data.  The inflaton potential corresponding to $r$ given by a n--th order polynomial of the e--folds number is derived in quadratures expressions. Since the observable difference between the spectral indices evaluated at a pivot scale yields information about the linear term of that polynomial, the first order case is explicitly written down. The solutions show features beyond the exponential form corresponding to power--law inflation and can be matched with current observational data. ",
        "introduction": "Introduction\\protect} During the early phase of the cosmological evolution the universe could experience a period of accelerated expansion known as inflation. The simplest scenario with a time dependent equation of state yielding the required negative pressure is that of a single real scalar field, {\\em the inflaton}, with dynamics dominated by its potential energy. For extensive details and references on this scenario and on the other topics mentioned in this Introduction see book \\cite{inflation} and Ref.~\\cite{Percival:2002gq}.  The inflaton quantum fluctuations would be stretched by the expansion beyond the radius within which causal interactions take place. These curvature perturbations could reenter the causal horizon in much later epochs and, through gravitational collapse, lead to anisotropies in the cosmic microwave background (CMB) temperature and to cosmological large scale structure (LSS). During inflation primordial gravitational waves are also produced. These tensor perturbations induce a curled polarization in the CMB radiation and increase the overall amplitude of its anisotropies at large angular scales.  A hint about the physics in the very early universe could be obtained by fitting to data the result of analytical calculations of the CMB and density spectra. These calculations depend on the values of some parameters which are the ones to be pinned down. The initial conditions for the evolution of the thermal anisotropies are also characterized by several quantities. These primordial parameters are often given as the multipoles of \\begin{eqnarray} \\label{eq:SExp} \\ln A^2(k)&=&\\ln A^2(k_*) + n(k_*)\\ln\\frac{k}{k_*} \\nonumber\\\\ &+& \\frac12\\frac{d\\,n(k)}{d\\ln k}\\vert_{k=k_*}\\ln^2\\frac{k}{k_*}+\\cdots \\, , \\end{eqnarray} where $A$ stands for the normalized amplitudes of the scalar ($A_S$) or tensor ($A_T$) perturbations, the corresponding spectral indices, $n$, are defined by, \\begin{eqnarray} \\label{eq:nSDef} n_S-1&\\equiv&\\frac{d\\ln A_S^2}{d\\ln k} \\, ,\\\\ \\label{eq:nTDef} n_T&\\equiv&\\frac{d\\ln A_T^2}{d\\ln k} \\, , \\end{eqnarray} $k=aH$ is the comoving wavenumber corresponding to the wavelength matching the Hubble horizon during inflation and $k_*$ is a pivot scale.  Initially, only $A_S(k_*)$ was fitted. Since inflation predicts nearly scale--invariant spectra, the tilt given by the scalar spectral index was then taken into account. This reduces the primordial spectrum to a power--law function of the scale $k$. The only single field model exactly yielding such a spectrum is power--law inflation \\cite{PLinfl} where the cosmic scale factor, $a$, behaves like a power--law of the cosmic time, $t$, and the inflaton potential is an exponential function. Signals of nonzero `running'  of the scalar index, $dn_S /d\\ln k$, have already been reported \\cite{Hannestad:2001nu} and must be refined by near future observations, allowing, this way, to move beyond the power--law approximation.  The role of the tensor perturbations deserves also attention when determining the best--fit values of the cosmological parameters from CMB and LSS spectra. That is motivated in part by the possibility of measuring the cosmic background polarization, allowing the tensorial contribution to be indirectly determined. This contribution can be parametrized in terms of the relative amplitudes of the tensor and scalar perturbations, \\begin{equation} \\label{eq:r} r \\equiv \\alpha \\frac{A_T^2}{A_S^2}\\, , \\end{equation} where $\\alpha$ is a constant. Presently, due to measurement limitations, a constant value of $r$ is fitted.  Nevertheless, in Ref.~\\cite{Terrero-Escalante:2001du} it was shown that few inflationary models produce an exactly constant tensor to scalar ratio, and, in order to be proper scenarios of inflation, they must be observationally indistinguishable from power--law inflation, where $r={\\rm constant}$ too. Since power--law inflation is just one of many suitable final stages of the inflaton dynamics \\cite{Terrero-Escalante:2002sd}, it implies that, either a constant value for $r$ may be no characteristic of the underlying inflationary evolution or that conclusions about the inflaton potential beyond an exponential form may be no possible to be drawn \\cite{Terrero-Escalante:2002qe}. This last factor could be a strong limitation for programs of the inflaton potential reconstruction.  Taking the above into account, the aim of this letter is to analyze the relation between the functional form of the tensor to scalar ratio and the inflationary dynamics. After introducing in Sec.~\\ref{sec:keq} the relevant equations, the analysis is done in Sec.~\\ref{sec:sols} by means of several examples. Finally, the results are discussed in Sec.~\\ref{sec:disc} and it is concluded that the difference between the spectral indices can be a very useful quantity because it yields information on the scale dependence of the tensor to scalar ratio, hence allowing to observe features of the inflaton potential different from the exponential form. ",
        "conclusions": ""
    },
    "0209/astro-ph0209354_arXiv.txt": {
        "abstract": "{A general theory of homeoidally striated density profiles where no divergence occurs, is adapted to cuspy density profiles, with a suitable choice of the scaling density and the scaling radius.   A general formulation of some physical parameters, such as angular-momentum vector, rotational-energy tensor (both calculated in connection with a special class of rotational velocity fields), inertia tensor, and self potential-energy tensor, is performed.   Other potential-energy tensors involving two density profiles where the boundaries are similar and similarly placed, are also expressed.    Explicit results are attained for three special cases of physical interest: NFW (e.g., Navarro et al. 1997) and MOA (e.g., Moore et al. 1999) density profiles, which fit to a good extent the results of high-resolution simulations for dark matter haloes, and H (Hernquist 1990) density profiles, which closely approximate the de Vaucouleurs $r^{1/4}$ law for elliptical galaxies.   The virial theorem in tensor form for two-component systems is written for each subsystem, and applied to giant elliptical galaxies.   The predicted velocity dispersion along the line of sight, in the limiting case where a principal axis points towards the observer, is found to be consistent with observations except for (intrinsic) $E7$ configurations where the major axis points towards the observer. If dark matter haloes host an amount of undetected baryons about twice as massive as the stellar subsystem, and undetected baryons trace non baryonic matter therein, two main consequences arise, namely (i) velocity dispersions along the line of sight are lower than in absence of undetected baryons, and (ii) dark matter haloes are dynamically ``hotter'' than stellar ellipsoids, the transition occurring when the amount of undetected baryons is about one and a half times that of the stellar subsystem. In this view, both the observation that the temperature of the extended hot gas exceeds the central stellar temperature, and the fact that the non baryonic matter is dynamically ``hotter'' than the stars, are a reflection of the presence of undetected baryons, which trace the dark halo and are about twice as massive as the stellar ellipsoid. ",
        "introduction": "According to standard CDM or $\\Lambda$CDM cosmological scenarios, large-scale celestial objects such as galaxies and clusters of galaxies, are made of at least two components: one, baryonic and more concentrated, embedded within one other, non baryonic and dissipationless, usually named dark matter halo. After a wide number of both analytical and numerical studies (e.g., Cole \\& Lacey 1996; Navarro et al. 1995, 1996, 1997; Moore et al. 1998, 1999; Fukushige \\& Makino 2001; Klypin et al. 2001), it has been realized that dark matter haloes which virialize from hierarchical clustering show universal density profiles, $\\rho=\\rho(r; \\rho^\\dagger, r^\\dagger)$, where $\\rho^\\dagger$ is a scaling density and $r^\\dagger$ is a scaling radius.    In this view, smaller haloes formed first from initial density perturbations and then merged with each other, or were tidally disrupted from previously formed mergers, to become larger haloes.  The density profile is (i) self-similar, in the sense that it has the same expression, independent of time (e.g., Fukushige \\& Makino 2001), and (ii) universal, in the sense that it has the same expression, independent of halo mass, initial density perturbation spectrum, or value of cosmological parameters (e.g., Navarro et al. 1997; Fukushige \\& Makino 2001).   A satisfactory fit to the results of numerical simulations is the family of density profiles (e.g., Hernquist 1990; Zhao 1996): \\begin{equation} \\label{eq:runi} \\rho\\left(\\frac r{r^\\dagger}\\right)=\\frac{\\rho^\\dagger}{(r/r^\\dagger)^\\gamma [1+(r/r^\\dagger)^\\alpha]^\\chi}~~;\\quad\\chi=\\frac{\\beta-\\gamma} \\alpha~~; \\end{equation} for a suitable choice of exponents, $\\alpha$, $\\beta$, and $\\gamma$.  This family includes both cuspy profiles first proposed by Navarro et al. (1995, 1996, 1997), $(\\alpha,\\beta,\\gamma)= (1,3,1)$, and the so called modified isothermal profile, $(\\alpha,\\beta,\\gamma)=(2,2,0)$, which is the most widely used model for the halo density distribution in analyses of observed rotation curves.   It also includes the perfect ellipsoid (e.g., de Zeeuw 1985), $(\\alpha,\\beta,\\gamma)= (2,4,0)$, which is the sole (known) ellipsoidal density profile where a test particle admits three global integrals of motion.   Finally, it includes the Hernquist (1990) density profile, $(\\alpha,\\beta,\\gamma)=(1,4,1)$, which closely approximates the de Vaucouleurs $r^{1/4}$ law for elliptical galaxies.   In dealing with the formation of dark matter haloes from hierarchical clustering in both CDM and $\\Lambda$CDM scenarios, recent high-resolution simulations allow $(\\alpha,\\beta,\\gamma)= (3/2,3,3/2)$, as a best fit (e.g., Ghigna et al. 2000; Fukushige \\& Makino 2001; Klypin et al. 2001), as first advocated by Moore et al. (1998, 1999)% \\footnote{More precisely, an exponent $\\alpha=1.4$ was derived by Moore et al. (1998), while the value $\\alpha=1.5$ was established by Moore et al. (1999).}.  Though Eq.\\,(\\ref{eq:runi}) implies null density at infinite radius, the mass distribution has necessarily to be truncated for two types of reasons.   First, the presence of neighbouring systems makes the tidal radius an upper limit. On the other hand, isolated, over-dense objects cannot extend outside the Hubble sphere of equal mass. Second, the total mass, deduced from Eq.\\,(\\ref{eq:runi}) for an infinitely extended configuration, is divergent, at least with regard to the special choices of exponents, $(\\alpha, \\beta,\\gamma)=(1,3,1)$, hereafter quoted as NFW density profile, and $(\\alpha,\\beta,\\gamma)=(3/2,3,3/2)$, hereafter quoted as MOA density profile.   The region enclosed within the truncation boundary has to be intended as representative of the quasi static halo interior, leaving aside the surrounding material which is still infalling.   It is worth remembering that the total mass can be finite even for an infinitely extended configuration, provided the related density profile is sufficiently steep e.g., $(\\alpha,\\beta,\\gamma)=(1,4,1)$, hereafter quoted as H density profile.  In dealing with numerical simulations of dark matter haloes, it is usual to take into consideration spherically averaged density profiles (e.g., Cole \\& Lacey 1996; Navarro et al. 1997; Fukushige \\& Makino 2001; Klypin et al. 2001), or in other terms spherical isopycnic (i.e. of equal density) surfaces.   On the other hand, spin growth by tidal interactions with neighbouring objects, in expanding density perturbations, demands ellipsoidally averaged density profiles (e.g., Doroshkevic 1970; White 1984; Maller et al. 2002; Jing \\& Suto 2002), or in other terms ellipsoidal isopycnic surfaces.   As a best compromise between intrinsic simplicity and unavoidable necessity, our attention will be devoted to homeoidally striated configurations, i.e. the isopycnic surfaces are similar and similarly placed ellipsoids.  Galaxies and cluster of galaxies may safely be idealized as two subsystems which link only via gravitational interaction, in such a way that each component is distorted by the tidal potential induced by the other. Then the application of the virial theorem in tensor form may be performed either to the whole system or to each subsystem separately.   Towards this aim, the explicit expression of the potential-energy tensors are needed. Though some results are available in literature (e.g., Brosche et al. 1983; Caimmi \\& Secco 1992; Caimmi 1993, 1995), the related density profiles exhibit no central divergence, or ``cusp'', in contradiction with Eq.\\,(\\ref{eq:runi}) when $\\gamma>0$.  The present attempt aims mainly to (i) formulate a general theory of homeoidally striated density profiles with a central cusp; (ii) devote further investigation to a few special cases which are consistent with the results of both observations and simulations related to galaxies and cluster of galaxies; (iii) apply to galaxies, as the analogon to clusters of galaxies performed in an earlier paper (Caimmi, 2002).  The current paper is organized in the following way. The general theory of homeoidally striated, density profiles with a central cusp, is performed in Sect.\\,\\ref{teori}. The general results are particularized to NFW, MOA, and H density profiles, which provide good fits to the results from both observations (e.g., Geller et al. 1999; Rines et al. 2001) and numerical simulations (e.g., Klypin et al. 2001), in Sect.\\,\\ref{spec}. An application to elliptical galaxies is performed in Sect.\\,\\ref{apga}.    Finally, some concluding remarks are reported in Sect.\\,\\ref{core}. Further details on fitting simulated and theoretical, self-similar, universal density profiles, are illustrated in the Appendix. ",
        "conclusions": "\\label{core}  A general theory of homeoidally striated ellipsoids (e.g., Roberts 1962; Chandrasekhar 1969; C93), where no divergence occurs in the density profile, has been adapted to cuspy density profiles.   An explicit calculation of the related, physical parameters implies the specification of the density profile, which is equivalent to the knowledge of: (i) the functional dependence of a scaled density, $f=\\rho/\\rho^\\dagger$, on a scaled radius, $\\xi=r/r^\\dagger$; (ii) a boundary condition, i.e. $f(1)=1$; (iii) two independent parameters, i.e. a scaling density, $\\rho^\\dagger$, and a scaling radius, $r^\\dagger$; (iv) a truncated, scaled radius, $\\Xi$.    The latter requirement is due to the fact, that the systems under consideration exhibit a null density at an infinite radius where, on the other hand, the total mass may attain a divergent value.    In addition, an infinity of density profiles in the physical space, $({\\sf O}r\\rho)$, is represented by a single density profile in the abstract space, $({\\sf O}\\xi f)$, for any selected choice of exponents, $(\\alpha, \\beta,\\gamma)$, appearing in Eq.\\,(\\ref {eq:runi}).  Potential-energy tensors involving both one and two, homeoidally striated density profiles, where the boundaries are similar and similarly placed, have been expressed in terms of integrals on the mass distribution.   Explicit calculations have been performed for both NFW and MOA density profiles, which satisfactorily fit the results of high-resolution simulations for dark matter haloes (e.g., Fukushige \\& Makino 2001; Klypin et al. 2001), and for H density profiles, which closely approximate the de Vaucouleurs $r^{1/4}$ law for elliptical galaxies (e.g., Hernquist 1990; Holley-Bockelmann et al. 2001).  The virial theorem in tensor form, related to a two-component system, has been expressed for each subsystem, and applied to giant elliptical galaxies.   The predicted velocity dispersion along the line of sight, in the limiting case where a principal axis points towards the observer, has been found to be consistent with the data except for (intrinsic) $E7$ configurations, when the major axis points towards the observer.  The suggestion that dark matter haloes host an amount of undetected baryons as massive as about twice the stellar subsystem (Valageas et al. 2002), together with the assumption that undetected baryons trace non baryonic matter therein, has produced two main consequences, namely (i) predicted velocity dispersions along the line of sight are lower than in absence of undetected baryons, and (ii) dark matter haloes are dynamically ``hotter'' than stellar ellipsoids, the transition occurring when the amount of undetected baryons is about one and a half times the stellar subsystem.  In this view, both the observation that the temperature of the extended hot gas exceeds the central stellar temperature, and the fact that the non baryonic matter is dynamically ``hotter'' than the stars (e.g., LW99), are a reflection of the presence of undetected baryons, which trace the dark halo and are about twice as massive as the stellar ellipsoid."
    },
    "0209/astro-ph0209024_arXiv.txt": {
        "abstract": "We present families, and sets of families, of periodic orbits that provide building blocks for boxy and peanut (hereafter b/p) edge-on profiles. We find cases where the b/p profile is confined to the central parts of the model and cases where a major fraction of the bar participates in this morphology. A b/p feature can be built either by 3D families associated with 3D bifurcations of the x1 family, or, in some models, even by families related with the $z$-axis orbits and existing over large energy intervals. The {\\sf `X'} feature observed inside the boxy bulges of several edge-on galaxies can be attributed to the peaks of successive x1v1 orbits (Skokos et al. 2002a, hereafter paper I), provided their stability allows it. However in general, the x1v1 family has to overcome the obstacle of a S\\ar\\D\\ar S transition in order to support the structure of a b/p feature. Other families that can be the backbones of b/p features are x1v4 and z3.1s. The morphology and the size of the boxy or peanut-shaped structures we find in our models is determined by the presence and stability of the families that support b/p features.  The present study favours the idea that the observed edge-on profiles are the imprints of families of periodic orbits that can be found in appropriately chosen Hamiltonian systems, describing the potential of the bar. ",
        "introduction": "Disk galaxies, when observed edge-on, often exhibit a box- or peanut-like structure. Since this is confined to the inner parts of the galaxy, and since it extends in the vertical direction outside the disk, this structure has been called a boxy or peanut bulge. Yet there are many ways in which it differs from ordinary bulges. The b/p structures have their maximum thickness not at the center of the galaxy, like in usual $R^{1/4}$ spheroids, but `at two points symmetrically spaced on either side of the center' \\cite{bb59}. Another difference from $R^{1/4}$-bulges and ellipticals is that b/p structures rotate `cylindrically', i.e. their observed rotation is independent of the height above the plane, which bulges and ellipticals do not (e.g. Kormendy \\& Illingworth 1982). There is one more characteristic of b/p features related to their kinematics. Kuijken \\& Merrifield (1995) and Bureau \\& Freeman (1999) have shown that there are important differences between the position velocity diagrams of b/p structures and those of bulges. These have been used by Bureau \\& Athanassoula (1999) and Athanassoula \\& Bureau (1999) to develop diagnostics to detect the presence and orientation of a bar in edge-on disk galaxies. The method relies on the presence of x2 orbits in the bars of the galaxies. Seen these differences, we will avoid calling the b/p features bulges, unless we are refering to particular observations.  Recent statistical studies (L\"utticke, Dettmar \\& Pohlen 2000a) using 1350 galaxies from the RC3, show that 45\\% of the profiles of edge-on disc galaxies are box- or peanut-shaped. Observational studies (e.g. Bureau \\& Freeman 1999, L\"utticke, Dettmar \\& Pohlen 2000b) associate the b/p structure with the presence of a bar. L\"utticke et al. (2000b) classify the bulges according to their boxiness and conclude that galaxies with a prominent b/p shape have a large BAL/BUL ratio, where BAL is the projected bar length, and BUL the bulge length. Isolating photometrically the b/p structure from the bulge, they measure the ratio of the projected bar length to the length of the b/p structure (BAL/BPL). Unfortunately this is done only for six galaxies, and gives an average value of $2.7\\pm0.4$. This ratio indicates a structure confined close to the center of the galaxy.   Bars and edge-on b/p morphology are linked in all the above mentioned papers, as well as in many others. However, the percentage of the bar which takes part in the b/p structure, i.e. whether we have a b/p feature on top of the bar, or whether we have a b/p-shaped bar in total, is an open question.  The morphological differences encountered among the various b/p features remains also to be explained.  In a few cases (e.g. IC~4767, Hickson 87a) an `{\\sf X}'-shaped structure is found to be embedded in a boxy structure. It differs from the usual peanut in that the branches of the {\\sf `X'} feature resemble segments of nearly straight lines that give the impression of intersecting each other. On the contrary the classical peanuts have typically much rounder isophotes (see e.g. Shaw 1993) and, even in cases where these isophotes come very close to the equatorial plane of the galaxy at the center, the visual impression is better described by the symbol `\\cx' than by an `{\\sf X}' central morphology\\footnote{nice examples can be found in the web page of L. Kuchinski at \\\\{\\em http://www.astronomy.ohio-state.edu/\\~{}lek/galx.html}}. L\"utticke, Dettmar \\& Pohlen (2000c), studying a sample of b/p galaxies including cases with {\\sf `X'} features, estimated the angle between one branch of the {\\sf `X'} and the major axis to be around $40 \\degr \\pm 10 \\degr$. Studies of individual galaxies give for this angle values ranging from $22\\degr$ for IC~4767 \\cite{whib88} to $45\\degr$ for NGC~128 \\cite{do99}. Pfenniger \\& Friedli (1991), taking as an example the case of IC~4767, claim that this feature is an optical illusion obtained when one uses a particular look-up-table for viewing the image.  This view, however, is not generally shared. Thus Mihos, Walker, Hernquist et al.  (1995) used the process of `unsharp masking' to enhance the {\\sf `X'} embedded in the bulge of the galaxy Hickson 87a in order to compare this morphology with their model. In any case one can speak about characteristic kinks of the isophotes in edge-on profiles of a few galaxies and of the corresponding isodensities in snapshots of some $N$-body simulations \\cite{am01}, which are aligned in such a way as to describe an {\\sf `X'}-shaped feature.  There have been two approaches for explaining edge-on b/p profiles. The first invokes internal reasons, like disk or orbital instabilities, and the second one external reasons, like encounters with companions, soft merging etc.  Combes \\& Sanders (1981) were the first to reproduce a b/p profile in $N$-body simulations of barred galaxies. Such structures were since then found in many other simulations (e.g. Combes, Debasch, Friedli et al. 1990; Raha, Sellwood, James et al. 1991; Athanassoula \\& Misiriotis 2002) and are now considered a standard development in bar-unstable disc simulations. Pfenniger (1984a, 1985) associated the b/p morphology with the instability of the x1 family at the 4:1 vertical resonance. Later, Combes et al. (1990) suggested that the b/p shapes are due to the vertical 2:1 resonance, stressing the importance of having $\\Omega_b = \\Omega - \\kappa /2 = \\Omega - \\nu /2$. This mechanism invokes a conjunction of the two resonances, i.e. the radial 2:1 (radial ILR) and the vertical 2:1 (v-ILR), and relates it to the appearance of a b/p edge-on morphology. The 3D families introduced at higher order resonances exist typically over smaller energy intervals and thus are less probable to be populated by orbits to build the box, although in principle they could be used as well. In such a case of course, they will support a thin morphological feature extending to large distances from the center, i.e. close to corotation. The 2:1 vertical resonance is proposed as explanation of boxy structures also by Pfenniger \\& Friedli (1991), who speak about thick b/p bars. We mention that b/p features have also been found in edge-on profiles of orbital models of {\\em normal} spiral galaxies with thick, 3D spirals embedded in discs \\cite{pg96}. Berentzen, Heller, Shlosman et al. (1998) have shown that a peanut shape may disappear when there is substantial gas inflow to the center of the galaxy.  Building the peanut by accretion of material from satellite galaxies has been proposed initially by Binney \\& Petrou (1985), while Mihos et al. (1995) describe an encounter which produces the {\\sf `X'} feature in the galaxy Hickson 87a. However, even in the cases where a companion is involved, the families of orbits that trap the infalling gas have to be studied.  In the present paper we investigate the vertical structure of bars using orbital theory. We do not construct self-consistent models, but we explore changes that occur when the main parameters of the system vary. We combine families of periodic orbits found in the models of paper I and in the models of Skokos, Patsis and Athanassoula (2002b - hereafter paper II), in order to build b/p features in the edge-on profiles. We also compare the geometry and the dimensions of the resulting systems with the corresponding features of edge-on galaxies. Speaking about a `bulge' in the profiles of our models, we refer to a central enhancement of the density due to 3D orbits in our total potential. As we will see, the families of orbits we use are mainly 3D bifurcations of the planar x1 orbits, i.e. related to the families of the x1-tree that make the 3D bar in our models.  The layout of this paper is as follows: In Section~2 we describe the method we used to construct the vertical profiles of the families in the models and in Section~3 and 4 we describe the properties of these profiles and the effect of combining several families together on the morphology of the models. In Section~5 we discuss our results and compare them with observations found in the literature, and finally in Section~6 we enumerate our conclusions. ",
        "conclusions": " \\begin{enumerate} \\item The vertical profiles of our models are of `stair-type'. This means that families that offer the skeletons for the 3D bars and are bifurcated at higher energies (i.e. closer to corotation) have in general lower mean heights. \\item b/p features in vertical profiles can be supported mainly by the following families: \\begin{itemize} \\item {\\bf x1v1}. This family is particularly useful for building a boxy central structure if it does not have a complex unstable part in the critical energy. The best examples we found are in the slow rotating bar case and in the strong bar model. In these cases, for energies where the maximum $z$ of the orbits remains less than about 1~kpc, successive orbits of x1v1 have the maximal deviations of their edge-on projections from the equatorial plane aligned along almost straight segments. These `lines' are in oblique angles to the major axis, not passing thorough the center in general, and their angle with the major axis is $\\approx 50\\degr$. In the slow bar case, however, this angle is $\\approx 27\\degr$. For this family the ratio {\\em $B_L/O_{Ly}$ is larger than 2}. $B_L/O_{Ly}>3$ for x1v1 (as found in model A3), indicate that only a small part of the family is populated and thus its contribution to the vertical structure of the model is not significant. \\item {\\bf x1v4}. This family gives $B_L/O_{Ly} \\approx 1.3$, i.e. brings the end of the peanuts close to the end of the bars. It is introduced in the system after a U\\ar S transition of x1 and has stable representatives for larger energies than the energy at the bifurcating point. It exists over large energy intervals and, if populated, will provide the system with a b/p-shaped structure whose extent is near that of the bar. It supports the peanut morphology especial in composite profiles without any contribution from x1v1 orbits (Fig.~\\ref{BLSc1yz}b). \\item {\\bf z3.1s} This family gives b/p features in models without radial or vertical ILRs, and is quite important for the dynamics of these models. Profiles characterized by the presence of this family (Fig.~\\ref{SMprof}c, Fig.~\\ref{Myz}c) have the characteristic local minimum of the density at the (0,0) position, like in the case of NGC~2788~A in Fig.~\\ref{twogal}a, which is indicated by a black arrow. \\item The ansae-type profile is easiest made by orbits associated with the vertical 4:1 resonance and can be described as a stretched {\\sf `X'} (Fig.~\\ref{Myz}a). \\item Finally the 3D {\\bf x2-like} families of orbits support boxy morphologies. The latter are especially discernible in the end-on profiles of the model with the low pattern speed. \\end{itemize} We have to note that if families bifurcated at high order vertical resonances are responsible for the peanut, then they will support a b/p bar morphology altogether.  Coexistence of several 3D families should in general be expected. In such a case it is the family which is bifurcated at the lowest energy that plays the most important role for the morphology of the model.  \\item Narrow extensions appear on the sides of many profiles (see e.g. Fig.~\\ref{BLSc1yz}a) These features result from the `stair-type' character of profiles constructed with families of periodic orbits, and have their counterpart in many images of edge-on disc galaxies. The corresponding feature in the case of NGC~6771 is also indicated by white arrows in Fig.~\\ref{twogal}b. \\begin{figure} \\hspace{1.2cm} \\epsfxsize=5.0cm \\epsfbox{twogal2.ps} \\caption[]{DSS images of NGC~2788~A (a), and NGC~6771 (b). Both galaxies show a b/p profile.} \\label{twogal} \\end{figure} \\item The projection of the orbits of a family on the equatorial plane is confined within certain limits. By this we mean that moving on the characteristic towards corotation, we reach a certain distance from the center where the mean radius of the orbits increases only due to increasing of $z$. This is particularly evident in the case of the x1v1 family, which is related to the vertical 2:1 resonance and which in general is the 3D bifurcation of x1 closest to the center. \\item Families of periodic orbits (x1v3, x1v4, x1v5, z3.1s, x2v1) can build boxy, or even peanut-boxy, {\\em end-on} profiles. We would thus like to suggest that boxy bulges in galaxies having a bar length over a b/p length larger than 3, are related with the profiles of families seen end-on. \\item {\\sf `X'}-type features are found in the composite orbital profiles. They are formed by alignment of successive orbits of the family x1v1.  They are pronounced if the S\\ar \\De transition in this family does not play an important role in the dynamics of a model. The fact that {\\sf `X'} features are rare in real galaxies, indicates that, in most cases where a x1v1 family is populated in a galaxy, it has an important complex unstable part.  Adequate successive projections of the orbits of a family in large energy intervals, like what we see in model B mainly due to the z3.1s orbits, give central morphologies that can be described with the symbol `\\cx'. We note that the higher the order of the resonance of the family associated with {\\sf `X'} or `\\cxs' structure is, the smaller the angle between the {\\sf `X'}/`\\cx'-branch and the major axis we find. In the case of the x1v1 family this angle is smaller in the slower rotating bar case. \\item Discs out of the equatorial plane in the bulges are easiest made by breaking the symmetry and populating only one of the two symmetric branches of the 3D families. \\item Characteristic local enhancements of the surface density along the major axis of the bar are predicted by the models merely due to the orientation of the successive orbits in the profiles. \\end{enumerate}"
    },
    "0209/astro-ph0209212_arXiv.txt": {
        "abstract": "AGN with the so-called `double-double' radio structure have been interpreted as restarted AGN where the inner structure is a manifestation of a new phase of activity which happened to begin before the outer radio lobes resulting from the previous one had faded completely. The radio galaxy 1245+676 is an extreme example of such a double-double object --- its outer structure, measuring $970\\,h^{-1}$\\,kpc, is five orders of magnitude larger than the $9.6\\,h^{-1}$\\,pc inner one. We present a series of VLBI observations of the core of 1245+676 which appears to be a compact symmetric object (CSO). We have detected the motion of the CSO's lobes, measured its velocity, and inferred the kinematic age of that structure. ",
        "introduction": "Radio galaxies which are not beamed toward the observer are normally perceived as double structures. The majority of them are large scale objects (LSOs) with angular sizes ranging from several arcseconds to several arcminutes. These angular sizes translate to large linear sizes of the order of $10^5-10^6$\\,pc; galaxies with sizes $>1$\\,Mpc are labelled `giant radio galaxies' (GRGs). A very interesting exception to this (relatively simple) picture exists however: a few LSOs, which are also double at first sight, turn out to have so-called double-double structure, radio galaxy 1450+333 (Schoenmakers~et~al.~2000) being a prime example.  LSO radio lobes are powered by central engines for a maximum of approximately $10^7$~years (Alexander \\& Leahy~1987; Liu, Pooley, \\& Riley~1992). If the nuclear energy supply stops, the extended radio structure will stop growing, its luminosity drops and the spectrum gradually gets steeper and steeper because of radiation and expansion losses. Komissarov \\& Gubanov (1994) estimate fade-away time scales of several $10^7$ years which is comparable to the timescale of the activity itself.  After that time the radio structure should in principle disappear, however there are a number of known mechanisms of restarting activity. For example, Hatziminaoglou, Siemiginowska, \\&~Elvis~(2001) elaborated a theory of thermal--viscous instabilities in the accretion disks of supermassive black holes (SMBHs). It predicts that the activity is recurrent and the length of the activity phase as well as the timescale of the activity re-occurrence is controlled by the mass of the SMBH. In favourable circumstances a new phase of activity can start before the radio lobes resulting from the previous one have faded completely. This would lead to double-double radio structure. ",
        "conclusions": ""
    },
    "0209/astro-ph0209206_arXiv.txt": {
        "abstract": "We discuss the Oosterhoff classification of the unusual, metal-rich globular clusters NGC~6388 and NGC~6441, on the basis of new evolutionary models computed for a range of metallicities. Our results confirm the difficulty in unambiguously classifying these clusters into either Oosterhoff group, and also question the view that RR Lyrae stars (RRL) in Oosterhoff type II (OoII) globular clusters can all be evolved from a position on the blue zero-age horizontal branch (ZAHB). ",
        "introduction": "NGC~6388 and NGC~6441 are unusual in several respects, including the following. (i)~In spite of being metal-rich ($\\rm [Fe/H] \\simeq -0.55$~dex), they contain prominent extensions to their red HBs---including both RRL and extended blue tails (Rich et al. 1997). (ii)~As pointed out by Sweigart \\& Catelan (1998), the normally ``horizontal'' part of these clusters' HBs is actually strongly tilted, with a $\\Delta V \\sim 0.5$~mag from the lower part of the red HB to the tip of the blue HB---which cannot be accounted for by canonical stellar evolution models. (iii)~The mean periods of the RRL in both clusters are extremely long for their metallicity---longer, in fact, even than in metal-poor, OoII globular clusters, thus apparently breaking down the traditional Oosterhoff classification scheme and posing yet another serious challenge to the models (Pritzl et al. 2000, 2001, 2002). ",
        "conclusions": ""
    },
    "0209/astro-ph0209340_arXiv.txt": {
        "abstract": "We measure the central values (within $R_e/8$) of the CaII triplet line indices CaT$^*$ and CaT and the Paschen index PaT at 8600 \\AA\\ for a 93\\%-complete sample of 75 nearby early-type galaxies with $B_T<12$ and $V_{gal}<2490$.  We find that the values of CaT$^*$ are constant to within 5\\% over the range of central velocity dispersions $100\\le \\sigma\\le 340$ km/s, while the PaT (and CaT) values are mildly anti-correlated with $\\sigma$.  Using simple and composite stellar population models, we show that: a) The measured CaT$^*$ and CaT are lower than expected from simple stellar population models (SSPs) with Salpeter initial mass functions (IMFs) and with metallicities and ages derived from optical Lick (Fe, Mg and H$\\beta$) indices. Uncertainties in the calibration, the fitting functions and the SSP modeling taken separately cannot explain the discrepancy. On the average, the observed PaT values are within the range allowed by the models and the large uncertainties in the fitting functions. b) The steepening of the IMF at low masses required to lower the CaT$^*$ and CaT indices to the observed values is incompatible with the measured FeH index at 9916 \\AA\\ and the dynamical mass-to-light ratios of ellipticals. c) Composite stellar populations with a low-metallicity component reduce the disagreement, but rather artificial metallicity distributions are needed. Another explanation may be that calcium is indeed underabundant in ellipticals. ",
        "introduction": "\\label{introduction}  The determination of the mean ages and metallicities of local elliptical galaxies is one of the key observational tests of models for galaxy formation and evolution. The classical picture of monolithic collapse (Larson 1974) assumes high formation redshifts and passive evolution, producing large central metallicities with strong gradients. In contrast, semi-analytic models of galaxy formation embedded in the hierarchical structure formation typical of Cold Dark Matter universes (Kauffmann, 1996), produce elliptical galaxies through mergers of disks and predict a large spread in the formation ages (especially for field ellipticals), with solar mean metallicities and shallow gradients.  Although many indications support the merging scenario, a clear-cut answer from the observations has been hampered by the age-metallicity degeneracy in the spectra of (simple) stellar populations: the same broad-band colors and absorption features can be obtained for very different combinations of ages and metallicities. The combined use of indices more sensitive to age (like the Balmer lines) and those more sensitive to metallicity (like Mg$_2$ and Fe5270 and Fe5335) offers a way out (Worthey 1994). However ambiguities remain, because the presence of a small metal-poor old stellar population may bias the age estimate (Maraston \\& Thomas, 2000, MT), and the metallicity estimate based on the Mg indices are systematically higher than the ones using iron lines (the so-called Mg over Fe over-abundance problem; Worthey, Faber \\& Gonz\\'alez, 1992, Trager et al. 2000, Thomas, Maraston \\& Bender 2002, TMB).  Recent modeling of the CaII triplet line at 8600\\AA\\ (Idiart, Th\\'evenin \\& De Freitas Pacheco 1997, I97; Garc\\'{\\i}a-Vargas, Molla \\& Bressan 1998; Moll{\\'a} \\& Garc\\'{\\i}a-Vargas, 2000) concluded that this index is insensitive to age for populations older than 1 Gyr, raising hopes that it could allow a robust measure of metallicity, and, in combination with colors or indices, age. An excellent review of the literature on the subject can be found in Cenarro et al. (2001a, C01).  Here we present the results of our survey of the CaII triplet line in local elliptical galaxies in the light of the new definition of the index given by C01 and their accurate determination of its stellar calibrators (the so called fitting functions, FF, Cenarro et al. 2001b, Cenarro et al. 2002, C02). In \\S \\ref{observations} we describe the observations and the data reduction, in \\S \\ref{results} we present the data, discuss them with new stellar population models and draw our conclusions. ",
        "conclusions": "\\label{results}  Fig. \\ref{figcatsig} shows the relation between the CaT$^*$, PaT and CaT indices as a function of the central velocity dispersion. As already noted by Cohen (1979), Faber and French (1980) and Terlevich, Diaz \\& Terlevich (1990), elliptical galaxies have very similar central values of Calcium triplet index. Averaged over the galaxy sample, the CaT$^*$ has a mean of 6.93\\AA\\ and rms 0.33 \\AA , or $\\approx 5$\\%, just above the measurement errors (statistical, systematic and due to calibration). Within the derived errors, the CaT$^*$ index does not depend on $\\sigma$, while a mild anticorrelation is observed for both PaT and CaT, driven by the slightly larger PaT at lower sigmas. This contrasts with the behaviour of the Mg$_2$ and Mg$b$ line indices, known to correlate strongly with $\\sigma$ in elliptical galaxies (Bender, Burstein \\& Faber 1993, Colless et al. 1999). These indices trace the $\\alpha$-element magnesium (Tripicco \\& Bell 1995, Maraston et al. 2002), and if the CaII triplet indices were to trace the calcium abundance, also an $\\alpha$-element, a correlation with $\\sigma$ could have been expected.  Fig. \\ref{figssp} shows stellar population models of the CaT$^*$, PaT and CaT indices constructed using the FF subroutines of C02 and the updated code of Maraston (1998, M98). A detailed description of the models considered here will be given in Maraston et al. (in preparation). The black lines show simple stellar population (SSP) models with the Salpeter IMF as a function of age and metallicity. These models reproduce well the tight metallicity-CaT correlation observed for globular clusters (open blue squares, from Armandroff and Zinn 1988, transformed using the relation given by C01, [Z/H] from Harris 1996). At metallicities higher than solar the models flatten and the age dependence becomes more important. The PaT index varies strongly with age for ages less than a few Gyr.  It reaches a value of $\\approx 1$ \\AA\\ at high ages. Note, however, the large rms uncertainties of the FFs ($\\approx 0.4$ \\AA).  The blue filled circles show the subsample of the database investigated here where ages and metallicity estimates are available from the analysis of the Fe, Mg and H$\\beta$ Lick indices (TMB). The results discussed in the following do not change if the set of ages and metallicities of Terlevich \\& Forbes (2002) are used instead. The same applies to the effects of errors on the age, metallicity and index, explored using Monte Carlo simulations. Within the large uncertainties allowed by the FFs, the models reproduce the average value of the measured PaT indices.  In contrast, the models predict values of CaT$^*$ and CaT more than 1 \\AA\\ larger than the measured ones. Such a large discrepancy cannot be explained by calibration errors, which are at least a factor 5 smaller. Uncertainties in the FFs (rms$\\approx 0.5$\\AA) alone seem also unable to explain it, although at high metallicities ([Z/H]$>0.3$) the FFs are based on only a handful of stars. The CaT$^*$ of a high Z SSP is dominated by the contribution of the RGB and the red clump (RC). If we set artificially the value of the CaT$^*$ FF at [Z/H]$=0.35$ from the giant branch phase on to 8 \\AA\\ (the lowest value measured for stars in this phase, most of the stars have CaT$^*=9-10$ \\AA ), we obtain CaT$^*=7.5$ \\AA\\ for the SSP, still 0.5 \\AA\\ larger than what observed in ellipticals. The flux contribution of the RC can vary within $\\approx 30$\\%, due to the uncertainties on the lifetime of this phase (Zoccali et al. 2000a).  If we reduce the flux of the RC by this amount, we again obtain CaT$^*=7.5$ \\AA\\ for the SSP. Only a 50\\% reduction would produce CaT$^*\\approx7$ \\AA .  The SSP Salpeter models translate the averaged observed value of the CaT$^*$ into [Z/H]$=-0.5\\pm 0.1$, or Z=$0.3\\pm 0.1$ Z$_\\odot$. Taking this result at face value, if the CaT$^*$ traces the calcium abundance, this could indicate that this element is underabundant, a suggestion already put forward by Peletier et al. (1999) considering 3 ellipticals. A similar effect has been suggested by McWilliam and Rich (1994) and Rich and McWilliam (2000), who analyse a sample of metal rich bulge stars, suggesting that while Mg and Ti are overabundant with respect to Fe, Ca and Si have nearly solar ratios. In addition, modeling the Ca4227 Lick index TMB find [Ca/Mg]$\\approx -0.3$ in ellipticals (see also Vazdekis et al. 1997).  However, the issue of Ca underabundance in high-metallicity stars is still controversial (McWilliam 1997); C02 point out that the CaT index does not correlate with the [Ca/Fe] stellar overabundance; and current modeling of the yields of Type II supernovae (Woosley \\& Weaver 1995) does not allow much room in this direction. While Mg (and O) are produced in the greatest quantities in high mass stars ($\\approx 35~M_\\odot$), lower mass stars ($\\approx 15-25~M_\\odot$) are responsible for the production of Ca (and Si). An IMF biased to the high masses or extremely short ($\\approx 10^7$ yr) star-formation bursts that avoid the Ca enrichment at high metallicities, as discussed by M\\'olla and Garc\\i a-Vargas (2000), do not seem a very appealing solution.  An alternative explanation of the low value of the CaT$^*$ index could be a steeper IMF at low stellar masses, since the CaT$^*$ index decreases with increasing gravity for cool stars. The green lines of Fig. \\ref{figssp} show SSP models with IMF slope $\\alpha=-4$ for $M<0.6M_\\odot$ and $\\alpha=-2.35$ (Salpeter) at larger masses. The CaT$^*$ and CaT indices at the high metallicities have a stronger dependence on age and are indeed able to reproduce the measured values of ellipticals. However, the observational evidence points to an IMF flatter than Salpeter in the bulge of our Galaxy (Zoccali et al. 2000b). In addition, as already discussed by Carter, Visvanathan and Pickles (1986) and Couture and Hardy (1993), such dwarf-dominated models produce values of the FeH feature at 9916 \\AA\\ that are an order of magnitude larger than what is observed. Moreover, the visual mass-to-light ratios (at t=10 Gyr, Z=Z$_\\odot$, one gets $M/L_B=40~M_\\odot/L_\\odot$, M98) are incompatible with dynamical estimates (Gerhard et al. 2001, $M/L_B\\approx 6$).  Since at low metallicity the CaT$^*$ and CaT indices depend strongly on Z (see Fig. \\ref{figssp}), Composite Stellar Populations (CSPs) with a low metallicity component are bound to produce (significantly) lower CaT$^*$ and CaT indices than SSPs with the same (high) mean metallicity, while leaving the PaT values essentially unchanged.  The red lines of Fig. \\ref{figssp} explore this case, presenting CSPs where 90\\% of the mass is an SSP model and 10\\% is the [Z/H]=-2.25 SSP with the same age, as done in MT. They show that this admittedly rather artificial model is compatible with the UV constraints avaliable for ellipticals, generating at the same time higher values of the H$\\beta$ index than SSPs of the same metallicity. The CSP models of Fig. \\ref{figssp} predict almost constant values of the CaT$^*$ and CaT indices for Z$\\ge$ Z$_\\odot$. This stems from the increasing relative importance of the flux at 8600 \\AA\\ of the low metallicity component with increasing metallicity of the main SSP component. It becomes largest at Z$\\approx$ Z$_\\odot$, to decrease slightly at higher metallicities, where more flux is emitted at the near-infrared wavelengths (see M98). Models with broader metallicity distributions (for example, the closed-box one) smear out this effect, resulting in CaT$^*$ and CaT indices steadily increasing with metallicity. The CSP models match the upper third of the galaxy distribution, however on average they produce CaT$^*$ and CaT indices still $\\approx 0.5$ \\AA\\ larger than the observed ones. As discussed in MT, a low metallicity tail is expected in the projected line of sight metallicity distribution, if radial metallicity gradients are present. This is suggested by the analysis of the Lick indices (Davies, Sadler \\& Peletier 1993, Mehlert et al. 2000) and color gradients (Saglia et al. 2000). A ``halo-like'' low-metallicity population could also be in place, reminiscent of the bimodal (color) distributions observed in the globular cluster systems of many giant ellipticals (Larsen et al. 2001 and references therein). Finally, we note that CSP models with a young, metal-rich component, postulated to explain the observed high central H$\\beta$ values of some ellipticals (De Jong \\& Davies 1997) do not help here, since at high metallicities the SSPs predict increasing values of the CaT$^*$ and CaT indices with decreasing ages.  To conclude, none of the discussed options alone seems able to explain the observed distribution of the calcium triplet values and a combination of them might be at work. In particular, it could be that Ca is in fact underabundant in elliptical galaxies."
    },
    "0209/astro-ph0209389_arXiv.txt": {
        "abstract": "When the density parameter is close to unity, the universe has a large curvature radius independently of its being hyperbolic, flat, or spherical. Whatever the curvature, the universe may have either a simply connected or a multiply connected topology. In the flat case, the topology scale is arbitrary, and there is no {\\it a priori} reason for this scale to be of the same order as the size of the observable universe. In the hyperbolic case any nontrivial topology would almost surely be on a length scale too large to detect. In the spherical case, by contrast, the topology could easily occur on a detectable scale. The present paper shows how, in the spherical case, the assumption of a nearly flat universe simplifies the algorithms for detecting a multiply connected topology, but also reduces the amount of topology that can be seen. This is of primary importance for the upcoming cosmic microwave background data analysis.  This article shows that for spherical spaces one may restrict the search to diametrically opposite pairs of circles in the circles-in-the-sky method and still detect the cyclic factor in the standard factorization of the holonomy group. This vastly decreases the algorithm's run time.  If the search is widened to include pairs of candidate circles whose centers are almost opposite and whose relative twist varies slightly, then the cyclic factor along with a cyclic subgroup of the general factor may also be detected. Unfortunately the full holonomy group is, in general, unobservable in a nearly flat spherical universe, and so a full 6-parameter search is unnecessary. Crystallographic methods could also potentially detect the cyclic factor and a cyclic subgroup of the general factor, but nothing else. ",
        "introduction": "Recent cosmic microwave background (CMB) data analysis suggests an approximately flat universe. The constraint on the total density parameter, $\\Omega$\\footnote{$\\Omega$ is the ratio between the total energy density and the critical energy density of the universe.}, obtained from CMB experiments depends on the priors used during the data analysis. For example with a prior on the Hubble parameter and on the age of the universe, recent analysis~\\cite{sievers,netterfield} of the DASI, BOOMERanG, MAXIMA and DMR data lead to $\\Omega=0.99\\pm0.12$ at 1$\\sigma$ level and to $\\Omega=1.04\\pm0.05$ at 1$\\sigma$ level if one takes into account only the DASI, BOOMERanG and CBI data. Including stronger priors can indeed sharpen the bound. For instance, including information respectively on large scale structure and on supernovae data leads to $\\Omega=1.01_{-0.06}^{+0.09}$ and $\\Omega=1.02_{-0.08}^{+0.09}$ at 1$\\sigma$ level while including both finally leads to $\\Omega=1.00_{-0.06}^{+0.10}$. In conclusion, it is fair to retain that current cosmological observations only set the bound $0.9 <\\Omega< 1.1$. These results are consistent with Friedmann-Lema\\^{\\i}tre universe models with spherical, flat or hyperbolic spatial sections.  In the spherical and hyperbolic cases, $\\Omega \\approx 1$ implies that the curvature radius must be larger than the horizon radius. In all three cases -- spherical, flat, and hyperbolic -- the universe may be simply connected or multiply connected, but our chances of detecting the topology observationally depend strongly on the curvature.  The chances of detecting a multiply connected topology are worst in a large hyperbolic universe.  The reason is that the typical translation distance between a cosmic source and its nearest topological image seems to be on the order of the curvature radius, but when $\\Omega \\approx 1$ the horizon radius is less than half that distance.  For example, if $\\Omega_m = 0.34$, $\\Omega_\\Lambda = 0.64$ and $\\Omega = 0.98$, then the radius $\\chi_{\\rm LSS}$ of the last scattering surface is 0.43 radians \\footnote{A hyperbolic radian is defined to equal the curvature radius, just like a spherical radian.} (see Eq.~\\ref{chiLSS} below).  As $\\Omega$ approaches 1, $\\chi_{\\rm LSS}$ approaches 0 radians, which is much less than the typical topology scale, making the topology undetectable (see e.g. Refs~\\cite{gomero1,aurichsteiner,inoue3} for some studies on detectability of nearly flat hyperbolic universes). Note that, in spite of a widespread myth to the contrary, the topology scale in a hyperbolic 3-manifold might {\\it not} depend on the manifold's volume. It might remain comparable to the curvature radius even for large manifolds, assuming the observer sits at a generic point in the manifold. On the other hand, in the exceptional case that the observer sits near a short closed geodesic, his or her nearest translated image may be arbitrarily close, even in an arbitrarily large manifold.  In a multiply connected flat universe the topology scale is completely independent of the horizon radius, because Euclidean geometry -- unlike spherical and hyperbolic geometry -- has no preferred scale and admits similarities. A great deal of luck would be required for the topology scale to be less than the horizon radius but still large enough to accommodate the lack of obvious local periodicity\\footnote{The constraint on the size of a cubic 3-torus varies from half of the horizon size~\\cite{tore1} to about one fourth~\\cite{tore2} depending on the value of the cosmological constant. In the case of a vanishing cosmological constant, these constraints were extended to other flat topologies~\\cite{flat}}. Detecting such a topological structure would naturally raise deep questions about this ``topological\" coincidence: why is the topology scale of the order of the size of the observable universe today?  In a spherical universe the topology scale depends on the curvature radius, as in the hyperbolic case. Luckily, in contrast to the hyperbolic case, as the topology of a spherical manifold gets more complicated, the typical distance between two images of a single cosmic source decreases. No matter how close $\\Omega$ is to 1, only a finite number of spherical topologies are excluded from detection. For example, if $\\Omega_m = 0.34$, $\\Omega_\\Lambda = 0.68$ and $\\Omega = 1.02$, then $\\chi_{\\rm LSS}$ = 0.43 radians, and the only excluded topologies are those for which the cyclic factor (see Section~\\ref{SectionClassification} for explanations) has order at most ${2\\pi}/({2 \\chi_{LSS}}) \\approx 7$. As $\\Omega$ gets closer to 1, more topologies are excluded from observation; nevertheless the chances of observing a spherical universe remain vastly better than the chances of observing a flat universe, because in the flat case only a finite range of length scales are short enough to be observable while an infinite range is too large. The particular case of the detectability of lens spaces was studied in Ref.~\\cite{gomero2} (which also considers the detectability of hyperbolic topologies).  The present paper investigates how the assumption of a nearly flat ($\\Omega \\approx 1)$ spherical universe affects the strategy for detecting its topology (see Refs.~\\cite{texas,luminet,levin} for reviews on the different methods and their observational status). We find that the circles-in-the-sky method~\\cite{CornishSpergelStarkman} typically detects only a cyclic subgroup of the holonomy group, so the universe ``looks like a lens space'' no matter what its true topology is. If it looks like a globally homogeneous lens space (generated by a Clifford translation -- to be defined in Section \\ref{SectionClifford}), then the circles-in-the-sky search reduces from a 6-parameter search space to a 3-parameter space (if the radius $\\chi_{\\rm LSS}$ of the last scattering surface is known) or a 4-parameter space (if $\\chi_{\\rm LSS}$ is unknown). Assuming each parameter is tested at $\\sim 10^3$ points, the total run time decreases by a factor of $\\sim 10^9$ or $\\sim 10^6$. If the universe looks like a non globally homogeneous lens space, meaning that a larger subgroup of the holonomy group is detectable, generated by a pair of Clifford translations, then the matching circles lie in a 6-parameter space, but with strong constraints on their values, so the search is still much faster than in the general case.  We also note that the crystallographic method's pair separation histogram~\\cite{LLL} is well suited to detect a topological signal in a nearly flat spherical universe because of the abundance of Clifford translations (defined below).  Section~\\ref{SectionClassification} reviews the classification of spherical 3-manifolds. Section~\\ref{SectionClifford} defines Clifford translations and explains their cosmological significance. Section~\\ref{SectionCircles} analyzes the circles-in-the-sky method, and Section \\ref{SectionCrystallography} analyzes the crystallographic methods. ",
        "conclusions": "In this article, we have studied the detectability of topology when the universe is spherical but nearly flat, both by the CMB-based circle-in-the-sky method and the large scale structure-based crystallographic method.  Spherical topologies turn out to be the most readily detectable. After recalling their geometrical properties, we first showed that the circles-in-the-sky method typically detects only a cyclic subgroup of the holonomy group, so the universe ``looks like a lens space'' {\\it no matter what its true topology is}. If it looks like a globally homogeneous lens space, then the circles-in-the-sky search reduces from a 6-parameter search space to a 3-parameter space  or a 4-parameter space depending on whether the radius of the last scattering surface is known or not. If the universe looks like a non globally homogeneous lens space then the matching circles lie in a 6-parameter space, but with strong constraints on their values, so that the search is still much faster than in the general case.  We also showed that crystallographic methods are well suited to detect the topology of a nearly flat spherical universe thanks to the abundance of Clifford translations.  We hope our analyses and results will serve as preparation for intelligently searching the upcoming MAP satellite data for a topological signal.   \\ack{We thank Simon Prunet for discussions. JRW thanks the MacArthur Foundation for its support.} \\vskip0.5cm"
    },
    "0209/astro-ph0209176_arXiv.txt": {
        "abstract": "The 2:1 orbital resonances of the GJ\\,876 system can be easily established by the differential planet migration due to planet-nebula interaction. Significant eccentricity damping is required to produce the observed orbital eccentricities. The geometry of the GJ\\,876 resonance configuration differs from that of the Io-Europa pair, and this difference is due to the magnitudes of the eccentricities involved. We show that a large variation in the configuration of 2:1 and 3:1 resonances and, in particular, asymmetric librations can be expected among future discoveries. ",
        "introduction": "Marcy et al. (2001) have discovered two planets in 2:1 orbital resonances about the star GJ\\,876. This and possibly two other resonant pairs, one in 2:1 resonances about HD\\,82943 (Mayor et al. 2001; Go\\'zdziewski \\& Maciejewski 2001) and the other in 3:1 resonances about 55\\,Cnc (Marcy et al. 2002), and the ease of capture into these resonances from nebula induced differential migration of the orbits (e.g., Lee \\& Peale 2002), mean that such resonances are likely to be ubiquitous among extrasolar planetary systems. ",
        "conclusions": ""
    },
    "0209/astro-ph0209610_arXiv.txt": {
        "abstract": "Taking into account a torsion field gives rise to a negative pressure contribution  in cosmological dynamics and then  to an accelerated behaviour of  Hubble fluid.   The presence of torsion has the same effect of a $\\Lambda$-term. We  obtain a general exact solution which well fits  data coming from high redshift supernovae and Sunyaev-Zeldovich/X-ray method for the determination of cosmological parameters. On the other hand, it is possible to obtain observational constraints on the amount of torsion density. A dust dominated Friedmann behaviour is recovered as soon as torsion effects are not relevant. ",
        "introduction": "A generalization  of Einstein General Relativity can be obtained by considering a torsion tensor different from zero in ${\\bf U}_4$ space-time manifold where connection is not symmetric \\cite{hehl,trautman}. Such an approach is very interesting today in relation to several extended theories of gravity as Superstrings, Supergravity and Kaluza-Klein theories. In particular, torsion allows to include spin matter fields in General Relativity and the Einstein-Cartan-Sciama-Kibble (ECKS) theory is one of most serious attempt in this direction. However, not all the forms of torsion are directly connected to a spin counterpart as it is widely discussed in \\cite{classtor}.  Now if some forms of torsion allow to take into account spin in General Relativity, it seems reasonable that they could have had some role into dynamics of the early universe when   the number density of particles per volume was huge. The presence of torsion gives naturally a repulsive contribution to the energy-momentum tensor \\cite{desabbata}. In fact, it is possible to show that for densities of the order of $10^{47} g/cm^3$ for electrons and $10^{54} g/cm^3$ for protons and neutrons, torsion could give observable consequences if all the spins of the particles result aligned. These huge densities can be reached only in the early universe so that cosmology is the only viable approach to test torsion effects \\cite{desabbata}. However no relevant tests confirming the presence of torsion have been found until now and it is still an open debate if the space-time is a Riemannian ${\\bf V}_4$ manifold or not. Considering cosmology and, in particular primordial phase transitions and inflation \\cite{desabbata,kolb,peebles}, it seems very likely that, in some regions of early universe, the presence of local magnetic fields could have aligned  the spins of particles. At very high densities, this effect could influence the evolution of primordial perturbations remaining as an imprint in today observed large scale structures.  >From another point of view, the presence of torsion could give observable effects without taking into account clustered matter but resulting as a sort of cosmological constant. Recent observations  seem to point out  that the universe is accelerating. Type Ia Supernovae (SNe Ia) \\cite{perlmutter},  data coming from clusters of galaxies \\cite{cluster} and CMBR investigations \\cite{boomerang} give observational constraints from which we deduce that the universe is spatially flat, low density and dominated by some kind of non-clustered dark energy. Such an energy, which is supposed to have dynamics, should be the origin of the observed cosmic acceleration.  In terms of density parameter, we have $\\Omega_m\\simeq 0.3\\,,\\Omega_{\\Lambda}\\simeq 0.7\\,, \\Omega_{k}\\simeq 0.0\\,,$ where $\\Omega_{m}$ includes  non-relativistic baryonic and non-baryonic (dark) matter, $\\Lambda$ is the dark energy (cosmological constant, quintessence,..), $k$ is the curvature parameter of Friedmann-Robertson-Walker (FRW) metric.   Standard matter fluid as source of Einstein--Friedmann cosmological equations gives rise to expanding decelerated dynamics. To fit observations, non-standard forms of matter-energy have to be taken into account: the net effect should be to implement a sort of a {\\it cosmological constant} which naturally gives rise to a negative pressure capable of implementing an accelerated cosmic expansion. On the other hand, {\\it Quintessence} \\cite{steinhardt,rubano,curvature} generalizes this approach taking into account all the mechanisms which give rise to negative pressure regimes for cosmic fluid. In particular, scalar fields.   The cosmological constant problem is one of the main issue of modern physics since its value  should provide the gravity vacuum state \\cite{weinberg}, should be connected  to the mechanism which led the early universe to the today observed large scale structures \\cite{guth,linde}, and should predict what will be the fate of the whole universe (no--hair conjecture) \\cite{hoyle}.   In any case, we need a time variation of cosmological constant to get successful inflationary models, to be in agreement with observations, and to obtain a  de Sitter stage toward the future. In other words, this means that cosmological constant has to acquire a great value in early epoch, it has to undergo a phase transition with a graceful exit and has to result in a small remnant toward the future coinciding with the observational constraints {\\it (coincidence problem)}.   The today observed  accelerated cosmological behaviour should be the result of this dynamical process where the present  value of cosmological constant is not fixed exactly at  zero.  In this context, a fundamental issue is to select the classes of gravitational theories and  conditions which \"naturally\" allow to recover an effective cosmological constant without considering special initial data. Theories with torsion could match up this point, as we will see below, also if the coincidence problem (that is the fine tuning between huge initial values of $\\Lambda$ and very thin today observed constraints) remain essentially unsolved.  This paper is organized as follows. Sec.2 is devoted to an essential summary of gravity with torsion. In Sec.3, we show how introducing torsion in cosmological dynamics gives rise to a negative pressure extra-term acting as a cosmological constant and we found a general cosmological solution. In Sec.4, we  match such a cosmological solution  with data coming from SNe Ia surveys and Sunyaev-Zeldovich/X-ray method. Sec.5 is devoted to discussion and conclusions. ",
        "conclusions": "In this paper, we dealt with a cosmological model where torsion is present into dynamics. After the discussion of how such a contribution modifies cosmological Friedmann-Einstein equations, we have shown that the net effect of torsion is the introduction of an extra-term into fluid matter density and pressure which is capable of giving rise to an accelerated behaviour of the cosmic fluid. Being such a term a constant, we can consider it a sort of torsion $\\Lambda$-term. If the standard fluid matter is dust, we can exactly solve dynamics which is in agreement with the usual Friedmann model (to be precise Einstein-de Sitter) as soon as torsion contribution approaches to zero.  The next step is to compare the result with observations in order to see if such a torsion cosmology gives rise to a coherent picture. We have used SNe Ia data, Sunyaev-Zeldovich effect and X-ray emission from galaxy clusters. Using our model, we are capable to reproduce the best fit values of $H_0$ and $\\Omega_M$ which gives a cosmological model dominated by a cosmological $\\Lambda$-term. In other words, it seems that introducing torsion (and then spins) in dynamics allows to explain in a {\\it natural} way the presence of cosmological constant or  a generic form of dark energy without the introduction of exotic scalar fields. Besides, as we have seen in Sec.4, observations allows to estimate torsion density which can be comparable to other forms of matter energy $(\\sim 5.5\\times 10^{-30}\\,g\\,cm^{-3})$.  However, we have to say that we used only a particular form of torsion and the argument can be more general if extended to all the forms of torsion \\cite{classtor}. Furthermore, being in our case the torsion contribution a constant density, it is not possible to solve {\\it coincidence} and {\\it fine tuning} problems. To address these issues we need a form of torsion evolving with time. This will be the topic of a forthcoming paper."
    },
    "0209/astro-ph0209426_arXiv.txt": {
        "abstract": "Magnetohydrodynamic (MHD) simulations have been used to study disk accretion to a rotating magnetized star with an aligned dipole moment. Quiescent initial conditions were developed in order to avoid the fast initial evolution seen in earlier studies. A set of simulations was performed for different stellar magnetic moments and rotation rates. Simulations have shown that the disk structure is significantly changed inside a radius  $r_{br}$ where magnetic braking is significant. In this region the disk is strongly inhomogeneous. Radial accretion of matter slows as it approaches the area of strong magnetic field and a dense ring and funnel flow form at the magnetospheric radius $r_m$ where the magnetic pressure is equal to the total, kinetic plus thermal, pressure of the matter.  Funnel flows (FF), where the disk matter moves away from the disk plane and flows along the stellar magnetic field, are found to be stable features during many rotations of the disk. The dominant force driving matter into the FF is the pressure gradient force, while gravitational force accelerates it as it approaches the star. The magnetic force is much smaller than the other forces. The funnel flow is found to be   strongly sub-Alfv\\'enic everywhere. The FF is subsonic close to the disk, but it becomes supersonic well above the disk. Matter reaches the star with a velocity close to that of free-fall.  Angular momentum is transported to the star dominantly  by the magnetic field. In the disk the transport of angular momentum is  mainly by the matter, but closer to the star the matter transfers its angular momentum to the magnetic field and the magnetic field is dominant in transporting  angular momentum to the surface of the star. For  slowly rotating stars we observed that magnetic braking leads to the deceleration of the inner regions of the disk and the star spins up. For a rapidly rotating star,  the inner regions of the disk rotate with  a super-Keplerian velocity, and the star spins-down. The average  torque is found to be zero when the corotation radius $r_{cor}\\approx 1.5 r_m$.    The evolution of the  magnetic field  in the corona of the disk depends on the ratio of magnetic to matter energies in the corona and in the disk. Most of the simulations were performed in the regime of a relatively dense corona where the matter energy density was larger than the magnetic energy density. In this case the coronal magnetic field gradually opens but the velocity and density of outflowing matter are  small. In a test case where a significant part of the corona was in the field dominated regime, more dramatic opening of the magnetic field was observed with the formation of magneto-centrifugally driven outflows.  Numerical applications of our simulation results are made to T Tauri stars. We conclude that our quasi-stationary simulations  correspond to the classical T Tauri stage of evolution. Our results are also relevant to cataclysmic variables and magnetized neutron stars in X-ray binaries. ",
        "introduction": "Disk accretion to a rotating magnetized star is important in a number of astrophysical objects, including T Tauri stars (Edwards  {\\it et al.} 1994), cataclysmic variables (e.g., Warner 1995), and X-ray pulsars (e.g., Bildsten  {\\it et al.} 1997).  \\begin{figure*}[t] \\centering \\caption{Initial conditions of Types I (left panel) and II (right panel). The gray-scale and numbers show density distribution.} \\label{Figure 1} \\end{figure*}   The accretion of matter to a rotating star with a dipole magnetic field is a complicated problem still only partially solved. The important questions which  need to be answered include: (1) What  is the structure of the disk near the magnetized star? (2)  Where is the inner radius of the disk? (3) What is the nature of the funnel flows (FF)? For example, which force is dominant in lifting matter to the funnel flow? (4) How is the accretion rate influenced by the star's magnetic moment $\\mu$ and angular velocity $\\Omega_*$? (5) What is the mechanism of angular momentum transport between the star and the disk? What determines whether star spins-up or spins-down? (6) What are the necessary conditions for magneto-centrifugally  driven outflows from the disk and/or the star?   Many   of these questions have been investigated analytically, but the  conclusions reached by different authors often differ  because the simplifying assumptions are different. For example, regarding question (2), some authors conclude that the disk should be disrupted in the region where magnetic and matter stresses are comparable (e.g., Pringle and Rees 1972 - hereafter PR72; Davidson \\& Ostriker 1973; Lamb, Pethick \\& Pines 1973; Ghosh, Lamb, \\& Pethick 1977; Scharlemann 1978; Ghosh \\& Lamb 1979 a,b - hereafter GL79a,b; Camenzind 1990; K\\\"onigl 1991; Shu {it et al.} 1994 -- hereafter S94). Others, argue that the inner radius of the disk should be farther away, at the corotation radius, because the inner regions of the disk are disrupted by magnetic braking (Ostriker \\& Shu 1995 -- hereafter OS95; Branderburg \\& Campbell 1998; Elstner \\& R\\\"udiger 2000).  \\begin{figure*}[t] \\centering \\caption{Evolution of a disk with no magnetic field for initial conditions of Type I (panels a, b, c, d) and Type II (panels e, f, g, h). Panels a \\& e show initial density distribution. Other panels show density distribution for different viscosities, $\\alpha=0, ~0.01,$ and $ 0.02$ at $T=50$. The initial density for initial conditions of Type II is about 1.5 times smaller than for Type I. Consequently the density scale is slightly different in these cases. The  legend for $\\rho$ is chosen to show the density distribution in the main part  of the disk. It does not reflect the low density in the corona where $\\rho_{min}=0.003$ and the highest densities of matter in the disk near the star where $\\rho_{max}=18$ at the panel $c$, $\\rho_{max}=12$ at the panel $d$, $\\rho_{max}=2.1$ at the panel $g$, $\\rho_{max}=4.2$ at the panel $h$.} \\label{Figure 2} \\end{figure*}    Question (3) is investigated in only a few papers which use the Bernoulli integral (Lamb, Pethick \\& Pines 1977; Paatz \\& Camenzind 1996 -- hereafter PC96; Li \\& Wilson 1999; and Koldoba {\\it et al.} 2002). The authors agree that the flow should become supersonic (and slow magnetosonic) just above the disk. On the other hand opinions differ regarding the driving force which pushes matter up into the FF. Li and Wilson (1999) (see also Li et al. 1996) propose that the twisting of the magnetic field near the base of the FF should be very large, $\\gamma_\\phi=|B_\\phi|/B_p >> 1$ and the magnetic force should be the main one lifting matter to the FF.  Here,  $B_\\phi$ is the toroidal component of the magnetic field and $B_p$ is the poloidal component. Other groups (e.g., Lamb et al. 1977; PC96; Koldoba et al. 2002) argue that the FF should be  super-Alv\\'enic so that the twisting of the field is small so that the magnetic force is also small. In numerical simulations by MS97, H97 and GBW99 the magnetospheric accretion was reported, but no clear evidence of funnel flows was presented and no analysis of FFs was performed. A significant part of this paper is devoted to the FFs. Another important issue which has been discussed over the past $30$ years is question (5) concerning the transport of angular momentum between the disk and the star. What determines the sign of the torque on the star? In early papers it was supposed that a star can  only be spun-up because matter in a Keplerian disk brings positive angular momentum to the star (e.g., PR72). Later, it was recognized that a star can be spun-down due to the part of the star's magnetic flux which passes through the disk outside of the corotation radius (GL79b, Wang 1995). Recently, the idea of ``torqueless'' accretion was proposed where mass but not angular momentum is transported to a star (e.g. S94; OS95; Li, Wickramasinghe \\& R\\\"udiger 1996; Li \\& Wickramasinghe 1997). Wang (1997) presented arguments against this idea, but this still remains an open question.  Question  (6) regarding magneto-centrifugally driven outflows from the disk has been discussed by a number of authors (e.g., Camenzind 1990; K\\\"onigl 1991;  S94; OS95;  Lovelace, Romanova \\& Bisnovatyi-Kogan 1995 - hereafter LRBK95; Fendt, Camenzind \\& Appl 1995; PC96; Goodson \\& Winglee 1999; Bardou \\& Heyvaerts 1999; Agapitou \\& Papaloizou 2000). For example,  S94 proposed that poloidal magnetic flux  accumulates near the corotation radius and magnetic winds should blow from this point (X-point). LRBK95 proposed that wind may form from the entire region of the disk outside the corotation radius where the magnetic field threading the disk is open.    \\begin{figure*}[t] \\centering \\caption{Evolution of the disk and the poloidal magnetic field in the largest region studied, $R_{max}=55$. The density (gray scale background) varies from a minimum value $\\rho=0.003$ in the corona to a maximum value $\\rho=2.6$ in the disk. The field lines are labeled by their magnetic flux $\\Psi$ values which  change from $0.0009$ to $4.9$.} \\label{Figure 3} \\end{figure*}    Analytical investigations of disk accretion to a magnetized star are of course limited by the different assumptions made. For this reason robust 2D and 3D simulations are essential to further the understanding of the different phenomena. By robust we mean that the result should not depend on initial conditions, boundary conditions,  on grid resolution, and other  artificial factors. Several 2D MHD simulation studies have been made with different initial conditions aimed at disk/star outflows. In an early work Hayashi, Shibata, and Matsumoto (1996) (hereafter HSM96) investigated the interaction of a non-rotating star with a Keplerian accretion disk and observed the opening of magnetic field lines which initially thread both the star and the disk. They found single event outflows and the corresponding inward collapse of the disk on a dynamical time-scale (less than one period of rotation of the inner radius of the disk) with the radial  velocity of the disk close to free-fall. This fast evolution was the result of the magnetic braking of the disk by the  magnetic field linking the disk and non-rotating star through a conducting corona. This explosive behavior may correspond to some episodic accretion events of actual systems. However, it is important to investigate the possible quiescent behavior of the disk-star systems.  Miller and Stone (1997; hereafter MS97) investigated disk-star interaction for different geometries and stellar magnetic fields using the resistive ZEUS code. MS97 rotated the corona - which occupies all the space between the star and the disk - with the rotation rate of the star. This decreased the initial magnetic braking (compared to HSM96), so that they were able to perform simulations for several periods of rotation of the inner radius of the disk. In cases with a relatively weak magnetic field, they got results similar to those of HSM96. They also found the disk collapsing to the star with velocity $\\sim 0.5 v_{ff}$, the opening of magnetic field lines, and outflows of matter from the disk. However, in the case of a strong magnetic field, particularly in the case which included   a uniform homogeneous vertical magnetic field threading the disk, they observed diminished outflows. Instead the  matter flowed around the magnetosphere to the star. Similar results were obtained by Hirose et al. (1997; hereafter H97).  Goodson, Winglee and B\\\"ohm (1997) (hereafter GWB97) and Goodson, B\\\"ohm and Winglee (1999) (hereafter GBW99) did much longer simulations in very large simulation regions. They observed quasi-periodic matter outbursts   associated with the quasi-periodic opening of magnetic field lines and matter accretion to the star. The density in the corona was chosen to decrease in a special way, $1/R^{4}$, so that the Alfv\\'en speed $\\propto B/\\sqrt\\rho \\propto 1/R$ decreases gradually. This is  favorable for the opening of magnetic field lines, and for the generation and propagation of outflows. GWB97 and GBW99 do not investigate cases where the density falls off more slowly with distance.    \\begin{figure*}[t] \\centering \\caption{Evolution of the disk and the magnetic field in a medium size region $R_{max}=7$. The configurations at $T=0, 10, 20, 30, 40, \\& 50 $ are shown. The density (gray scale background) changes from maximum value $\\rho=2.6$ in the disk to the minimum value $\\rho=0.003$ in the corona. Contour levels of $\\Psi$, which label poloidal field lines, change from $4.9$ to $0.006$. The field lines of the strongest magnetic field near the star are not included in order to make the inner structure of the disk visible.} \\label{Figure 4} \\end{figure*}   Fendt and Elstner (1999, 2000 -- hereafter FE99, FE00) investigated disk-star interaction for thousands of rotations of the inner radius of the disk, and observed the opening of magnetic field lines and outflows. However, they treated the disk as a boundary condition so that they could not take into account the back reaction of the disk on the stellar magnetic field. Furthermore, the actual outflow of matter from the disk to the corona may be different from that assumed.   The above -- mentioned simulation studies show that outflows appear either in very non-stationary situations (HSM96, MS97) or for a special distribution of  coronal density and very fast rotation of the star (GWB97, GBW99). None of the  papers give answer to questions (1)-(5). Also, it is not clear whether or not outflows exist for more quiescent initial conditions, and for cases where the coronal density falls off slowly with distance.   In this paper we investigate disk accretion to a rotating magnetized star and the associated funnel flows. We start from initial conditions which give us the possibility to significantly reduce the initial magnetic braking between the disk and  the corona. This allows us to investigate the disk-star interaction and funnel flows for long times and to consider questions (1)-(5) in detail.    In \\S 2 we describe the numerical model, including initial and boundary conditions. We also discuss the evolution of the disk without a magnetic field. In \\S 3 we describe in detail the disk-star interaction for the case of  slowly rotating stars and in  \\S 4 cases of fast rotating stars. In \\S 5 we consider the dependence of disk-star interaction on the magnetic moment $\\mu$. In \\S 6 we analyze the physics of FFs. In \\S 7 we consider the possibility of outflows. In \\S 8 we apply our simulation results to T Tauri stars. In \\S 9 give the conclusions from this work.   \\begin{figure*}[t] \\centering \\caption{Evolution of the density in the disk (color background) and poloidal magnetic field lines  ($\\Psi=$const lines) (white lines)  in case of Type I initial conditions in the region $r< 6$. The black solid line corresponds to $\\tilde\\beta = 1$. The density changes from $0.003$ in the corona to $2.4$ in the disk. Contours of $\\Psi$ are exponentially spaced between $0.2$ and $1.2$.} \\label{Figure 5} \\end{figure*}  \\begin{figure*}[t] \\centering \\caption{Same as on Figure 4 but for initial conditions of Type II. At $T=0$ the inner radius of the disk is at $r=r_{cor}=3$. The $\\Psi=$const lines are exponentially spaced between $0.1$ and $0.7$.} \\label{Figure 6} \\end{figure*} ",
        "conclusions": "\\subsection{Funnel Flows}  We have done a wide range of MHD simulations of disk accretion to a rotating aligned dipole in order to understand the different accretion phenomena. The simulations show that funnel flows (FF), where matter flows out of the disk plane and essentially free-falls along the stellar magnetic field lines, are a robust feature of disk accretion to a dipole.  Specifically, we find that  ~(1).~ The disk truncates and a funnel flow forms near the magnetosphere radius $r_m$, where magnetic pressure of the dipole is comparable to the kinetic plus thermal pressure of the disk matter.  ~(2).~ The velocity of matter in the FF is much smaller than the Alfv\\'en velocity, $|{\\bf v}| \\sim (0.05-0.3 ) v_A$, so that matter flows along the magnetic field lines. The funnel flow accelerates and become supersonic. The Mach number is ${\\cal M}\\approx 3-4$ at the surface of the star. At the star velocity  is close to the free-fall velocity: $v_p\\approx 0.7 v_{ff}$.   ~(3).~ The angular velocity of the FF gradually varies from its value at the inner edge of the disk to the angular velocity of the star. The `twist' of the magnetic field lines in the FF is small, $|B_\\phi|/B_p < 0.1$, and it has a maximum approximately in the middle of the FF.  ~(4).~The main forces which are responsible for dragging matter to the FF are matter pressure gradient force (near the disk) and gravitational force in the rest of the FF. Magnetic force is negligibly small.   ~(5).~About $1/3$ of the magnetic flux responsible for the spin-up/spin-down the star goes through the FF, while the remainder is above the FF.       \\subsection{Disk-Star Interactions}   Regarding the interaction between the disk and the star we find that  ~(1)~ The magnetic field of the star influences the nearby regions of the disk inside a radius $r_{br}$, while viscosity dominates at larger radii. The radius $r_{br}$ depends on magnetic moment of the star $\\mu$ and density in the disk.  ~(2).~ Inside the radius $r_{br}$  the disk is strongly inhomogeneous. The density is $2-3$ times smaller than in the disk without magnetic field. Magnetically braked matter accumulates near magnetosphere  and forms a dense ring and funnel flow.   ~(3).~   The star may spin-up or spin-down depending on the ratio of its rotation rate to the rotation rate at the inner radius of the disk. We find that ``torqueless\" accretion is possible when $r_{cor}/r_m\\approx 1.5$, where $r_{cor}$ is the corotation radius.    ~(4).~ At the star's surface, the angular momentum flux is transported  mainly by the `twist' of the magnetic field. Angular momentum carried by matter in the disk at $\\sim r_m$ is transferred almost completely to the magnetic field at the star's surface.    ~(5).~ The coronal magnetic field is observed to open and close, but strong outflows  were not observed for the considered parameters and quasi-equilibrium initial conditions. In the area of the disk where the field is strong, $r < r_{\\Psi}\\approx  0.5 r_{br}$, the magnetic field lines tend to decelerate/accelerate rotation of the disk instead of being opened. Besides, opening of magnetic field loops is suppressed by matter dominated corona compared to GBW99 who accepted dependence $\\rho\\sim R^{-4}$.  ~(6).~ Strong outflows may be associated with strongly non-stationary accretion in the disk as  observed in simulations by HSM96, MS97, and H97 or when the disk is very dense and magnetic field lines are inclined to the disk as in simulations by FE99, FE00. Alternatively the outflows may occur in cases where the density of the corona decreases sufficiently rapidly with distance as in the simulations by  GWB97 and GBW99. Sporadic outflows can arise from a global magnetic instability of the disk (e.g., Lovelace {\\it et al.} 1994, 1997, 2002), but they may be absent during the quiescent evolution of the disk-star system."
    },
    "0209/astro-ph0209083_arXiv.txt": {
        "abstract": "We present medium-resolution $z$-, $J$-, and $H$-band spectra of four late-type dwarfs with spectral types ranging from M8 to L7.5.  In an attempt to determine the origin of numerous weak absorption features throughout their near-infrared spectra, and motivated by the recent tentative identification of the $E$ $^4\\Pi- A$ $^4\\Pi$ system of FeH near 1.6 $\\mu$m in umbral and cool star spectra, we have compared the dwarf spectra to a laboratory FeH emission spectrum.  We have identified nearly 100 FeH absorption features in the $z$-, $J$-, and $H$-band spectra of the dwarfs. In particular, we have identified \\nhfeatures\\ features which dominate the appearance of the $H$-band spectra of the dwarfs and which appear in the laboratory FeH spectrum.  Finally, all of the features are either weaker or absent in the spectrum of the L7.5 dwarf which is consistent with the weakening of the known FeH bandheads in the spectra of the latest L dwarfs. ",
        "introduction": "Absorption bands due to FeH are ubiquitous in the red and near-infrared (0.7 $<$ $\\lambda$ $<$ 1.3 $\\mu$m) spectra of late-type dwarfs. The most conspicuous FeH feature is the bandhead of the Wing-Ford band \\citep{wing69} at 0.99 $\\mu$m which arises from the 0$-$0 ($\\upsilon' - \\upsilon''$) transition of the $F$ $^4\\Delta- X$ $^4\\Delta$ system.  \\citet{schiavon98} obtained moderate-resolution spectra (R$\\sim$13,000) of a sample of early to mid M stars to study the dependence of the Wing-Ford band on atmospheric parameters.  Their spectra show that absorption features due to FeH dominate the spectra of the M dwarfs from 0.9850 to 1.0200 $\\mu$m.  Other bandheads of this electronic system have also been detected in the spectra of late-type dwarfs including the 2$-$0 at 0.7786 $\\mu$m \\citep{tinney98}, the 1$-$0 at 0.8692 $\\mu$m \\citep{kirkpatrick99,martin99}, the 2$-1$ at 0.902 $\\mu$m \\citep{tinney98}, the 0$-$1 at 1.1939 $\\mu$m \\citep{jones96,mclean00,leggett01, reid01}, and the 1$-$2 at 1.2389 $\\mu$m \\citep{mclean00,leggett01,reid01}.  In a recent study of infrared sunspot spectra, \\citet[hereafter WH]{wallace01} identified 68 lines common to both a sunspot spectrum and a laboratory spectrum of FeH between 1.581 and 1.755 $\\mu$m. In addition, they identified four bandheads in the FeH spectrum at 1.58263, 1.59118, 1.62457, and 1.61078 $\\mu$m.  Based on the theoretical work of \\citet{langhoff90}, they tentatively assigned this band to the 0$-$0 transition of the $E$ $^4\\Pi -A$ $^4\\Pi$ system.  WH noted that three of the bandheads, at 1.58263, 1.59118, 1.62457 $\\mu$m, could be seen in the low-resolution spectra of late M and L dwarfs \\citep[e.g.,][]{leggett01}.  These bandheads probably account for the three unidentified absorption features in the spectra of L dwarfs at 1.58, 1.613, and 1.627 $\\mu$m reported by \\citet{reid01}.  WH also obtained a high-resolution (R=50,000), 70 \\AA-wide spectrum of GJ 569B (M8.5 V) centered at 1.6578 $\\mu$m and identified 13 lines common to both the dwarf spectrum and FeH spectrum.  At least in this narrow wavelength range, the dwarf spectrum was dominated by FeH absorption lines.  In the course of conducting a 0.8 to 4.2 $\\mu$m spectroscopic survey of M, L and T dwarfs, we have identified numerous weak absorption features throughout the near-infrared spectra of the M and L dwarfs.  In an attempt to determine the origin of these features, and motivated by the work of WH, we have compared our spectra to a laboratory FeH emission spectrum to determine whether FeH produces these absorption features.  In this paper, we present medium-resolution $z$-, $J$- and $H$-band spectra of four late-type dwarfs with spectral types ranging from M8 to L7.5 along with the identification of nearly 100 FeH absorption features. In \\S2, we discuss the dwarf spectra while in \\S3 we discuss the FeH spectrum. In \\S4 we compare the FeH spectrum and the dwarf spectra and in \\S5 we discuss our results.  Our conclusions are summarized in \\S6. ",
        "conclusions": "We have presented $z$-, $J$-, and $H$-band spectra of four dwarfs with spectral types ranging from M8 to L7.5.  Using a laboratory FeH emission spectrum, we have identified nearly 100 FeH absorption features in the dwarf spectra.  The $H$-band spectra contain \\nhfeatures\\ features which have counterparts in the FeH spectrum.  Taken together with the results of \\citet{wallace01}, our results suggest that FeH absorption dominates the spectra of late-type dwarfs from \\hwrange."
    },
    "0209/astro-ph0209560_arXiv.txt": {
        "abstract": "Power spectrum estimation and evaluation of associated errors in the presence of incomplete sky coverage; non-homogeneous, correlated instrumental noise; and foreground emission is a problem of central importance for the extraction of cosmological information from the cosmic microwave background.  We develop a Monte Carlo approach for the maximum likelihood estimatation of the power spectrum.  The method is based on an identity for the Bayesian posterior as a marginalization over unknowns, and maximization of the posterior involves the computation of expectation values as a sample average from maps of the cosmic microwave background and foregrounds given some current estimate of the power spectrum or cosmological model, and some assumed statistical characterization of the foregrounds. Maps of the CMB and foregrounds are sampled by a linear transform of a Gaussian white noise process, implemented numerically with conjugate gradient descent.  For time series data with $N_{t}$ samples, and $N$ pixels on the sphere, the method has a computational expense $KO[N^{2} + N_{t} \\log N_{t}]$, where $K$ is a prefactor determined by the convergence rate of conjugate gradient descent. Preconditioners for conjugate gradient descent are given for scans close to great circle paths, and the method allows partial sky coverage for these cases by numerically marginalizing over the unobserved, or removed, region. ",
        "introduction": "Power spectrum estimation and evaluation of the associated errors in the presence of incomplete sky coverage; non-homogeneous, correlated instrumental noise; and foreground emission is a problem of central importance for the extraction of cosmological information from the cosmic microwave background. From a Bayesian point of view, power spectrum estimation involves the maximization of the posterior probability density, with error bars given by the set of cosmological parameters or power spectrum whose integrated posterior density achieves some specified level of confidence. A Bayesian approach to CMB analysis for large data sets involving a direct evaluation of the likelihood is intractable due to the $O(N^{3})$ expense associated with computing the inverse of non-sparse matrices, or their determinants ( \\cite{Borrill}, \\cite{Bond1} ). The goal of this paper is the development of alternative numerical methods, specifically Monte Carlo techniques, for  the Bayesian analysis of the CMB, including the complications of incomplete sky coverage, correlated noise and foregrounds.  Previous work has demonstrated that for a certain class of scanning strategies, the signal and inverse noise matrices are block diagonal. The block diagonal properties of these matrices give an exact $O(N^{2})$ Bayesian method, and therefore tractable for data sets as large as will be returned from Planck. The complications of this method are that it cannot easily accommodate partial sky converge or precessing scan strategies. The method of \\cite{OhSpergel} computes the maximum of the likelihood through a Newton Raphson method.  The numerical innovations of this method involve Monte Carlo simulations and the use of conjugate gradient descent, giving an overall expense $O(N^{2})$. The method was proposed and numerically demonstrated in the context of uncorrelated noise, and a region of sky coverage of azimuthal symmetry, where a good preconditioner can be constructed.  However, the algorithm is in fact more general, provided there is sufficient memory for storage of the needed matrices, and that conjugate gradient descent converges quickly enough (i.e. there is a good preconditioner).  As suggested in \\cite{WandeltRingSet}, we can use the ring set approach to supply preconditioners.  An outstanding problem to be solved is a way of retaining the mathematical advantages of a ring set scan (block diagonal inverse noise and signal matrices) while accommodating partial sky coverage. The approach formulated in this paper handles the problem of partial sky coverage by embedding the data in an azimuthally symmetric region of sky, and using a Monte Carlo Markov Chain to numerically marginalize over the unobserved part. For scans close to ring sets, we therefore inherit good preconditioners, allowing an extension of both the ring set and conjugate gradient methods to scan strategies as planned for Planck.  For observations $y = s + f + \\eta$, where $(s,f,\\eta)$ are the CMB signal, foregrounds, and noise respectively, our approach to power spectrum estimation is motivated by the identity (derived in Appendix \\ref{AppPosteriorIdentity} ) \\begin{equation} \\label{PosteriorIdentity} \\frac{p(\\Gamma | y)}{p(\\Gamma_{0} | y)} = \\int d(s,f) \\ \\left[ \\frac{p(\\Gamma|s)} {p(\\Gamma_{0}|s)} \\right] p(s,f|\\Gamma_{0},y) \\end{equation} where $\\Gamma$ is any parameterization of conclusions (such as the power spectrum or cosmological parameters), and $\\Gamma_{0}$ is any fixed guess. The Bayesian posterior ratio on the left is given as an integral over the unknown quantities which are assumed to generate the observed data. Maximization of the posterior involves computing the gradient of equation \\ref{PosteriorIdentity} which will be shown to depend on the expectation value of the power spectrum with respect to the random field $p(s , f |  \\Gamma_{0},y)$, \\begin{equation} \\label{expectation} E[C_{l}(s)|\\Gamma_{0},y] = \\int d(s,f) \\ C_{l}(s) \\ p(s,f |\\Gamma_{0},y) \\end{equation} We maximize the posterior ratio in equation \\ref{PosteriorIdentity} by the expectation maximization algorithm ( \\cite{Dempster} ) which proceeds by iteratively setting $C_{l}(\\Gamma_{n+1}) = E[C_{l}(s)|\\Gamma_{n},y]$. The algorithm converges to the posterior maximum for a uniform prior, and gives an un-biased, consistent estimator (see Appendix \\ref{AppEstimator} ).  In this paper, we focus on computation of the expectation value of the power spectrum $E[C_{l}(s)|\\Gamma_{n},y]$ given the data and some guess $\\Gamma_{n}$ under the assumption of perfect foreground separation (although we comment on how the approach can be generalized to include foregrounds later in the paper, and leave its numerical demonstration for future work).  We compute the expectation value of the power spectrum $E[C_{l}(s)|\\Gamma_{n},y]$ numerically with a Monte Carlo approach, where we sample maps of the CMB from the probability density $p(s | f, y, \\Gamma_{0} )$. Conditioning on some estimate of the foregrounds, the method exploits the fact that $p(s | f, \\Gamma_{0},y)$ is a Gaussian random field, and therefore completely characterized by the mean field map and covariance matrix of fluctuations about that map. Maps are sampled from $p(s | f, \\Gamma_{0},y)$ by first computing the mean field map with conjugate gradient descent, and then sampling fluctuations about the mean field map from a zero mean Gaussian field with covariance matrix $(N^{-1} + C^{-1} )^{-1}$ (where $N^{-1}$ is the inverse noise matrix, and $C^{-1}$ is the inverse covariance matrix for the CMB).  These fluctuation maps are sampled by a linear transformation, numerically computed with conjugate gradient descent, of a spatial white noise Gaussian process thereby generating maps with all the same statistical properties as samples from $p(s | f, \\Gamma_{0},y)$.  Each step of conjugate gradient descent involves a multiplication by the matrix $I + CN^{-1}$, which can be done very quickly by multiplication by $N^{-1}$ in the basis in which it is diagonal, followed by a transform to the spherical harmonic basis where $C$ is diagonal. For spatially uncorrelated noise and circularly symmetric beams, we only need to transform from the pixel to the spherical harmonic domain, with an expense $O(N^{3/2})$  \\cite{OhSpergel}. In order to accommodate spatially correlated noise, we transform to the time domain, followed by a transform to the spherical harmonics, giving an expense $K[O(N^{3/2}) + N_{t} \\log N_{t} ]$ where $N_{t}$ is the number of time samples, and $K$ depends on the convergence rate of conjugate gradient descent. Including the full complications of asymmetric beams, we would need to compute a convolution on the sphere. Using the convolution method of ( \\cite{Wandelt00}), the expense of our method is $K[O(N^{2}) + N_{t} \\log N_{t} ]$.  The computational feasibility of this method is limited by finding a numerical implementation of conjugate gradient descent which converges quickly so that the prefactor $K$ above is small. The strategy here is to embed the data in a region covered by an exact ring set scan, following the intuition that good preconditioners can be constructed for scan strategies close to ring sets \\cite{WandeltRingSet}.  Embedding the data in a region on the sky with no observations (or where they have been removed) is accommodated by numerically marginalizing over the missing observations.  Moreover, the same techniques can be used to marginalize over the foregrounds, and provide Monte Carlo estimates of the confidence intervals for cosmological parameters.  The paper is organized as follows.  We first review complications with a direct computation of the likelihood, and provide an overview of our approach.  We then discuss a technique we call transformed white noise sampling, which allows us to sample maps representing fluctuations about the mean field map for some guess of the power spectrum. We demonstrate the method with a flat sky $512 \\times 512$ test case, including incomplete sky coverage, with uncorrelated, non-homogeneous noise. We close with a discussion of further complications encountered in real CMB experiments, and how they can be accommodated in the framework presented here. ",
        "conclusions": "The fundamental hurdle to numerically implementing an exact Bayesian approach to CMB analysis, including complications of partial sky coverage, correlated noise, and foregrounds, is finding efficient ways to solve the linear problem $(I + C R^{T} N^{-1} R) \\xi = \\delta$ for any vector $\\delta$. Solving the linear equation has an expense $K O[N^{3/2} + N_{t} \\log N_{t}]$ for circularly symmetric beams, and the algorithm provides a tractable approach provided $K$ can be made small enough.  The strategy we presented in this paper allows the data to be embedded in an azimuthally symmetric region of the sky covered by a Wandelt ring set scan, with the intuition that, provided the true scan of the instrument is close enough to the exact scan, we inherit good preconditioners.  We also commented on how the method of transformed white noise sampling can be used in Monte Carlo Markov Chain for the entire Bayesian posterior.  The feasibility of this approach depends on a good approximation to the posterior itself. Previous work has demonstrated several computationally feasible, unbiased estimates of the power spectrum and associated error covariance matrix. Any of these methods could therefore be used, in principle, to give an approximate posterior, so that a Markov chain approach can be used as a final consistency check.  Future work will incorporate the foregrounds in the algorithm presented here, generalized for multifrequency data. Maximization of the likelihood of the power spectrum given the data again leads to the computation of the expectation value $E[C_{l} | d, \\Gamma_{0}]$, but now the marginalization includes the foregrounds as well.  If the prior for the foregrounds is Gaussian, then we can also use transformed white noise sampling to sample a new foreground map while conditioning on the CMB. If the prior used is non-Gaussian, other sampling schemes can be used, including Gibbs sampling or the Metropolis algorithm."
    },
    "0209/astro-ph0209396_arXiv.txt": {
        "abstract": "We present an analysis of the integrated properties of the stellar populations in the {\\it Universidad Complutense de Madrid} (UCM) Survey of $\\mathrm H\\alpha$-selected galaxies. In this paper, the first of a series, we describe in detail the techniques developed to model star-forming galaxies using a mixture of stellar populations, and taking into account the observational uncertainties. We assume a recent burst of star formation superimposed on a more evolved population. The effects of the nebular continuum, line emission and dust attenuation are taken into account. We also test different model assumptions including the choice of specific evolutionary synthesis model, initial mass function, star formation scenario and the treatment of dust extinction.  Quantitative tests are applied to determine how well these models fit our multi-wavelength observations for the UCM sample. Our observations span the optical and near infrared, including both photometric and spectroscopic data. Our results indicate that extinction plays a key role in this kind of studies, revealing that low- and high-obscured objects may require very different extinction laws and must be treated differently. We also demonstrate that the UCM Survey galaxies are best described by a short burst of star formation occurring within a quiescent galaxy, rather than by continuous star formation. A detailed discussion on the inferred parameters, such as the age, burst strength, metallicity, star formation rate, extinction and total stellar mass for individual objects, is presented in paper II of this series. ",
        "introduction": "One of the main issues in today's Astrophysics is how present day galaxies formed and how they have evolved over time. A considerable observational effort is being made to study galaxies from the earliest possible times to the present. Our knowledge of the faint galaxy populations over the $0<z<5$ range has experienced remarkable progress in a relatively short period of time (see the reviews by \\citealt{1997ARA&A..35..389E} and \\citealt{2000ARA&A..38..667F}). One of the main aims of these studies is to find the  progenitors of the local galaxy population. While the majority of local galaxies seem to fit reasonably well into the Hubble sequence, this morphological classification scheme breaks down at surprisingly low redshifts \\citep[$z\\sim0.3$-$0.5$; see][and  references therein]{2002dgkdp}. Moreover, new classes of distant objects have been discovered, such as {\\it Ultraluminous Infrared Galaxies}, \\citep{1983ApJ...269..352S}, {\\it Extremely Red Objects}, \\citep{2000AJ....120..575Y} as well as bright UV galaxies \\citep[{\\it Lyman Break Galaxies}, LBGs,][] {1996ApJ...462L..17S,1999ApJ...519....1S}. The luminosity of these objects, both the reddest and the bluest, is mainly dominated by massive knots of newly-formed stars (starbursts), with different amounts of dust extinction.  A complementary approach to understand how present-day galaxies came into being is to study in detail the properties of local galaxies, and in particular their star-formation histories.  In this respect, it is important to quantify the relative importance of the current episode of star formation in comparison to the underlying older stellar populations. Indeed, even in high redshift objects, stars formed before the currently-observed star formation episode must have been present in order to produce the observed metal and dust content. Examples of such high-$z$ objects include SCUBA sources \\citep{1998Natur.394..241H} and LBGs (see \\citealt{2001PASP..113.1449C} and references therein). Moreover, an accurate determination of the total stellar mass in both local and distant galaxies is a necessary step towards understanding their formation (see, e.g., \\citealt{1998ApJ...508..539P,2001ApJ...554..981P}).  Our group is actively working on the detailed study of galaxies in the Local Universe so that their properties can be compared with distant ones.  The  techniques developed and tested with local galaxies  will have direct application in high-{\\it z} studies.  The {\\it Universidad Complutense de Madrid\\/} (UCM) Survey was carried out in order to  perform a comprehensive study of star-forming galaxies in the Local Universe (\\citealt{1994ApJS...95..387Z,1996ApJS..105..343Z}; see also \\citealt{1999ApJS..122..415A}).  This $\\mathrm H\\alpha$-selected galaxy sample has been extensively studied at optical and near infrared wavelengths (see next section). It has also been used to determine the local $\\mathrm H\\alpha$ luminosity function and star formation rate density \\citep{1995ApJ...455L...1G}, providing a low-$z$ benchmark for intermediate and high-{\\it z} studies \\citep[see, for example,][]{2000PASJ...52...73I,2000A&A...362....9M,2000A&A...362..509V,astro-ph/0111390}. Recently, the UCM Survey is being extended to higher redshifts \\citep{2001A&A...379..798P}.  The present series of papers aims at determining the main properties of the stellar populations in the UCM Survey galaxies, accounting both for the newly-formed stars and the underlying evolved populations. We make use of the extensive multi-wavelength data available for the sample. A direct precursor of the current study was presented in \\citet[hereafter \\gil]{2000MNRAS.316..357G}, where we characterized the stellar content of a smaller subsample of 67 UCM galaxies, constraining their ages, metallicities and relative strength of the current star-formation episode.  A sophisticated statistical technique was developed by \\gil\\ to compare measurements and model predictions. We now present results for virtually all the UCM galaxies, increasing the sample by a factor of 2.5. We have also included additional photometry in the $B$-band \\citep{2000A&AS..141..409P}, and use a more elaborated spectral synthesis method.   The present paper will focus on the modelling technique and the observational data used to test it.  We will discuss the model input parameters that best describe the observed properties of the UCM galaxies, including the initial mass function (IMF), star formation scenarios and extinction prescriptions. We will also study in detail how well our modelling techniques are able to reproduce the observations.  In \\citet[\\pii\\, hereafter]{2002MNRASnotyetII}, the second paper of the series, we will present the derived properties of the UCM galaxies using these data and techniques. Paper~II will discuss the young and newly-formed stars in the galaxies, together with the underlying population of evolved stars, the total stellar masses, etc.  This paper is structured as follows: section~\\ref{data} introduces the sample, the observations and the data measurements. Section \\ref{method} describes our modelling techniques, including the main features of the stellar and nebular emission models, the star formation scenarios and the extinction prescriptions. Section~\\ref{discussion} discusses the goodness of the fits and possible correlations with the input data. Finally, Section~\\ref{summary} summarizes our conclusions. Throughout this paper we use a cosmology with $\\mathrm H_{0}=70$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$, $\\Omega_{\\mathrm M}$=0.3 and $\\Lambda$=0.7. ",
        "conclusions": ""
    },
    "0209/astro-ph0209443_arXiv.txt": {
        "abstract": "Braneworld models of dark energy are examined in the light of observations of high redshift type Ia supernovae.  Braneworld models admit several novel and even exotic possibilities which include: (i) The effective equation of state of dark energy can be more negative than in LCDM: $w \\leq -1$; (ii) A class of braneworld models can encounter a `quiescent' future singularity at which the energy density and the Hubble parameter remain well behaved, but higher derivatives of the expansion factor ($\\stackrel{..}{a}$, $\\stackrel{...}{a}$ etc.) diverge when the future singularity is reached; (iii) The current acceleration of the universe is a {\\em transient feature} in a class of models in which dark energy `disappears' after a certain time, giving rise to a matter dominated universe in the future.  Since horizons are absent in such a space-time, a braneworld model with {\\em transient acceleration} might help reconcile current supernova-based observations of an accelerating universe with the demands of string/M-theory.  A maximum likelihood analysis reveals that braneworld models satisfy the stringent demands imposed by high redshift supernovae and a large region in parameter space agrees marginally better with current observations than LCDM. For instance, models with $w < -1 \\ ( \\ > -1 )$ provide better agreement with data than LCDM for $\\Omega_m \\ggeq 0.3 \\ ( \\ \\lleq 0.25 )$. ",
        "introduction": "The idea that the universe may have more than three spatial dimensions first appeared in the works of Kaluza and Klein over 70 years ago. The notion of extra dimensions gradually became popular because of the hope that one might succeed in relating the gauge symmetries of particle physics to the isometries of a compact higher dimensional manifold \\cite{kk}.  All such models, however, assumed that the compactification scale is small ($l \\sim l_{\\rm Pl} \\sim 10^{-33}$ cm) hence unobservable. More recently, the many variants of superstring theory allow for the possibility that at least some of the extra dimensions of nature may be macroscopic. For example, the eleven -dimensional supergravity model of Horava and Witten \\cite{hw} assumes that ordinary matter fields are confined to a submanifold (brane) which is embedded in a higher dimensional space (bulk).  In an important recent development, Randall and Sundrum (RS) examined a simplified variant of this model consisting of a three dimensional brane embedded in a four dimensional anti-de Sitter (AdS) bulk \\cite{rs}.  Their results showed that gravitational excitations are confined close to the brane giving rise to the familiar $1/r^2$ law of gravity, and suggesting that one could identify the brane as our observable universe.  Subsequently the RS ansatz was generalised to incorporate both expanding FRW-type models \\cite{brane} and anisotropic space-times \\cite{anis}.  Of particular importance to the present study is the observation that simple extensions of the RS scenario can give rise to a universe which is {\\em accelerating}, in agreement with studies of high redshift supernovae \\cite{dgp,ss02a}. It is now well established that high redshift type Ia supernovae appear fainter than expected in a spatially flat matter dominated (Einstein-de Sitter) universe \\cite{riess,perl}.  One way of explaining this discrepancy is to postulate that the universe is filled with a smooth component carrying large negative pressure (dark energy). Although several possible candidates for dark energy have been suggested (the cosmological constant, quintessence etc.)  none is entirely problem free (see \\cite{ss00,sahni02} for recent reviews).  In this paper we shall focus on a new form of dark energy based on the braneworld model examined in \\cite{ss02a,ss02b} (see also \\cite{CH,Shtanov1}).  Braneworld models of dark energy have interesting new properties including the fact that, depending upon the form of bulk-brane embedding, the effective equation of state of dark energy can be $w \\geq -1$ or $w \\leq -1$.  In addition, for an appropriate parameter choice, the acceleration of the universe can be a transient phenomenon, thus helping reconcile high-$z$ supernova observations of an accelerating universe with the requirements of string/M-theory. ",
        "conclusions": "This paper examines braneworld models of dark energy in the light of recent supernova observations which indicate that the universe is accelerating.  The braneworld models which we examine in this paper have several interesting properties which distinguish them both from the cosmological constant as well as from scalar field based `tracker' models of dark energy. Like the latter, braneworld models presently accelerate, and possess a longer age than the standard cold dark matter model (SCDM).  However in marked contrast to both LCDM and tracker models, the luminosity distance in one class of braneworld models, B1, can be {\\em greater} than the luminosity distance in LCDM (for identical values of $\\Omega_m$): $d_L^{~\\rm dS}(z) \\geq d_L^{~\\rm BRANE1}(z) \\geq d_L^{~\\rm LCDM}(z)$, where $d_L^{~\\rm dS}(z)$ refers to the luminosity distance in spatially flat de Sitter space.  In terms of the effective equation of state $w$, this is equivalent to the assertion that $w \\leq -1$.  This result is particularly surprising since matter in the braneworld model never violates the weak energy condition $\\rho + p \\geq 0$.  A maximum likelihood analysis which compares braneworld model predictions with high redshift type Ia supernovae data, shows that B1 models provide good agreement with observations if $\\Omega_m \\ggeq 0.3$.  These results broadly support the analysis of \\cite{phantom} in which `phantom' dark energy models, having the property $w=P/\\rho \\leq -1$, were compared against supernova observations.  (It should be pointed out, however, that `phantom models' invariably run into a physical singularity in the future when $\\rho_{\\rm phantom} \\to \\infty$, such singularities are absent in the B1 model which remains well behaved at all future times.)  The second braneworld model we consider (B2) has properties which complement those of B1, since $d_L^{~\\rm LCDM}(z) \\geq d_L^{~\\rm BRANE2}(z) \\geq d_L^{~\\rm SCDM}(z)$.  This is equivalent to the assertion that $-1 \\leq w \\leq 0$.  Results of a maximum likelihood analysis show that B2 models are in excellent agreement with SNe data for smaller values of the density parameter $\\Omega_m \\lleq 0.25$.  Finally braneworld models also permit the dark energy to be a transient phenomenon. In models of this kind (called disappearing dark energy: DDE) the acceleration of the universe takes place during a transient regime separating past and future matter dominated epochs. In these braneworld models, the universe does not possess an event horizon and so it may be possible to reconcile a universe which currently accelerates with the demands of string/M-theory.  Comparison with Sne bounds shows that the Disappearing Dark Energy models marginally satisfy existing supernova data provided $\\Omega_m$ is sufficiently small: $\\Omega_m \\lleq 0.23$. For larger values of $\\Omega_m$, this class of models may be on the verge of being ruled out."
    },
    "0209/astro-ph0209169_arXiv.txt": {
        "abstract": "We present an extended theoretical library of over 800 synthetic stellar spectra, covering energy distribution in the optical range ($\\lambda = 3500-7000$~\\AA), at inverse resolution R=500\\,000. The library, based on the ATLAS\\,9 model atmospheres, has been computed with the SYNTHE code developed by R.\\ L.\\ Kurucz. The grid spans a large volume in the fundamental parameters space (i.e.\\ $T_{\\rm eff}$, $\\log{g}$, [M/H]), and can be profitably applied to different research fields dealing both with the study of single stars and stellar aggregates, through population synthesis models. A complementary project, in progress, will extend the wavelength range to the ultraviolet, down to 850~\\AA, at an inverse resolution of R=50\\,000. ",
        "introduction": "New-generation spectrographs, at the major ground-based telescopes, have begun to pour in the hard-disks of astronomers' computers an increasing mass of high-quality spectroscopic data. Inverse resolutions as high as $\\lambda / \\Delta\\lambda = 100\\,000$ can now be easily attained, at least for the brightest ($V \\la 15$ mag) objects in the sky, and this is pushing the observation of local and extragalactic stellar systems to an ever unimagined resolution level. The outstanding performance of instruments like UVES and VIRMOS at the ESO Very Large Telescope, or HIRES and SARG at the Keck Observatory and Telescopio Nazionale Galileo, respectively, urges therefore theoretical tools of comparable accuracy level in order to consistently match and analyse such a huge amount of observational data.  In this framework, and to help filling the gap, we undertook a long-term project aimed at providing the community with a systematic theoretical library of high-resolution stellar spectra (virtually the largest sample currently available in the literature) in the optical range ($\\lambda = 3500 \\to 7000$~\\AA) and at an inverse resolution of $R=500\\,000$. ",
        "conclusions": ""
    },
    "0209/astro-ph0209219_arXiv.txt": {
        "abstract": "Observational data on the short GRBs obtained with the GGS-Wind Konus experiment in the period from 1994 to 2002 are presented. The catalog currently includes 130 events, detailing their appearance rate, time histories, and energy spectra. Evidence of an early X-ray and gamma-ray afterglow for some of the short GRBs is discussed. The catalog is  available electronically at \\url{http://www.ioffe.ru/LEA/shortGRBs/Catalog/}. ",
        "introduction": "The gamma-ray burst (GRB) duration has a bimodal distribution (Mazets et~al. 1981; Norris et~al. 1984; Hurley 1992; Kouveliotou et~al. 1993). This indicates the existence of two distinct morphological classes of events, namely short-duration (\\textless~2~s) bursts  and long-duration (\\textgreater~2~s) bursts. Approximately 20 per cent  of the observed bursts are short. Their  energy spectra are usually harder than the spectra of long bursts (Kouveliotou et al. 1993).  The catalog contains data  on 130 short GRBs observed  with the Konus-Wind experiment on the Wind spacecraft in 1994--2002. The catalog presents  time histories, energy spectra, fluences, peak fluxes, spectral parameters, and hardness ratios. Most of hardness ratios reveal spectral variability. The catalog is  available electronically at \\url{http://www.ioffe.ru/LEA/shortGRBs/Catalog/}.  Searches for the optical and radio afterglow of short GRBs have been carried out in only a few cases. Four GRBs were localized with  high accuracy by the Interplanetary Network~(IPN) (Hurley et al. 2002). No optical and radio afterglow emission was detected for these four events. No X-ray counterparts have been detected  so far for the short bursts localized by Beppo-SAX (Gandolfi et al. 2000) or by HETE-2 (Lamb et al. 2002). The rapid follow-up observations have resulted in only upper limits on the brightness of the afterglows from these GRBs.  At the same time early X-ray and gamma-ray afterglows of short bursts were detected in a bumber cases by the GRB detectors themselves, in time intervals from seconds to tens of seconds after the trigger. BATSE observations showed that such a weak afterglow exists for some of the short bursts, lasting several tens seconds (Burenin, 2000; Lazatti et~al. 2001; Connaughton, 2002). The afterglows of short GRBs were also detected by the Konus-Wind experiment. Afterglow emission in the energy range bellow 1~MeV is seen for about 10 per cent of events. These bursts are included in the catalog. A statistical analysis of burst sampling reveals that the afterglow is a more common feature of short GRBs. These results will be discussed in more detail elsewhere. ",
        "conclusions": "Some statistical distributions of the main characteristics of the short GRBs are presented in Figures~165--172. The electronic version of the catalog is available at \\url{http://www.ioffe.ru/LEA/shortGRBs/Catalog/}. It contains detailed information about the characteristics of the short GRBs archived as ASCII files.  This work was supported by Russian Aviation and Space Agency Contract, and RFBR grant N~01-02-17808.  \\clearpage"
    },
    "0209/astro-ph0209505_arXiv.txt": {
        "abstract": "Anisotropy data analysis leaves a significant degeneracy between primeval spectral index ($n_s$) and cosmic opacity to CMB photons ($\\tau$). Low--$l$ polarization measures, in principle, can remove it. We perform a likelihood analysis to see how cosmic variance possibly affects such a problem. We find that, for a sufficiently low noise level ($\\sigma^P_{pix}$) and if $\\tau$ is not negligibly low, the degeneracy is greatly reduced, while the residual impact of cosmic variance on $n_s$ and $\\tau$ determinations is under control. On the contrary, if $\\sigma^P_{pix}$ is too high, cosmic variance effects appear to be magnified. We apply general results to specific experiments and find that, if favorable conditions occur, it is possible that a 2--$\\sigma$ detection of a lower limit on $\\tau$ is provided by the SPOrt experiment. Furthermore, if the PLANCK experiment will measure polarization with the expected precision, the error on low--$l$ harmonics is adequate to determine $\\tau$, without significant magnification of the cosmic variance. This however indicates that high sensitivity might be more important than high resolution in $\\tau$ determinations. We also outline that a determination of $\\tau$ is critical to perform detailed analyses on the nature of dark energy and/or on the presence of primeval gravitational waves. ",
        "introduction": "Data on cosmic microwave background (CMB) anisotropy and polarization can provide effective constraints on cosmological parameters. However, while significant anisotropy observations are available (see, e.g., Smoot et al. 1992; de Bernardis et al. 2000; Hanany et al. 2000; Halverson et al. 2002; Sievers et al. 2002; Scott et al. 2002) and further anisotropy data are expected from experiments in progress, detailed data on the weaker polarization signals may become available only within a few years. In turn, dark energy (DE) parameters or the amplitude of primeval gravitational waves (GW) are just mildly constrained by anisotropy data, while polarization would significantly depend on them. A peculiar situation then holds for the cosmic optical depth to CMB photons ($\\tau$), due to reionization. Anisotropy data only mildly constrain it, while polarization data would provide more stringent constraints; of course, greater $\\tau$ values would ease its detection, but, in turn, greater $\\tau$ values would also ease the determination of DE and/or GW parameters.  This paper tries to investigate which polarization experiment(s) can determine $\\tau$. The success of an experiment, however, will also depend on the value of $\\tau$ itself. In particular, we shall consider the SPOrt experiment (Macculi et al. 2000; Carretti et al. 2000; Peverini et al. 2001), planned to perform a (nearly) full--sky polarization measure aboard of the ISS in 2004, with an angular resolution similar to COBE.  When only anisotropy data are considered, there is a degeneracy between the effects of varying $\\tau$ or the primeval spectral index $n_s$ (Jungman et al. 1996; Zaldarriaga, Spergel \\& Seljak 1997; Eisenstein, Hu \\& Tegmark 1999; Efstathiou \\& Bond 1999). The most stringent limits on $\\tau$, based on anisotropy data, are provided by Stompor et al. (2001), who used the outputs of recent balloon experiments to set $\\tau\\lesssim 0.4$ at the 2--$\\sigma$ level, once $n_s$ is assumed $\\le1.2$. (Such limits will be included in a number of figures here below and denominated {\\it Stompor limits}). Full--sky anisotropy measures from space, like MAP\\footnote{\\url{http://map.gsfc.nasa.gov}} or PLANCK\\footnote{\\url{http://astro.estec.esa.nl/Planck}}, might improve the {\\it Stompor limits} by a factor $\\sim 2$. In turn, a residual uncertainty ($\\sim 0.05$) on the value of $n_s$ is related to our scarce knowledge of $\\tau$. Of course, a more precise determination of $n_s$ would really be welcome, first of all to shed new light on the mechanism generating primeval fluctuations. Furthermore, within a generic inflationary model, only for $n_s < 1$ primeval GW may be expected to exist.  The impact that a variation of $\\tau$ may have on CMB polarization can be seen from Fig.~1; here, for a spatially flat cosmological model (density parameters: $\\Omega_m = 0.35$, $\\Omega_b = 0.05$; Hubble constant in units of 100 km/s/Mpc: $h=0.65$; $n_s = 1$; dark energy is an ordinary cosmological constant) and for different values of $\\tau$, we plot the dependence on $l$ of the angular spectra $C_l^{A}$ with $A=T,~E,~B,~X$, standing for anisotropy, $E$-- and $B$--mode polarization and anisotropy--polarization cross correlation respectively; only $E$--mode polarization is considered through this paper. Fig.~1 shows that the main dependence on $\\tau$ actually occurs at low $l$ and for the polarization and cross--correlation spectra; comparatively, the $\\tau$ dependence of spectra, at $l \\gtrsim 30$, is milder. This point is important to devise the right polarization experiment, which should aim to great sensitivity, while, for $\\tau$ determination, high resolution data might not bear a direct relevance. Accordingly, the angular resolution of the SPOrt experiment does not prevent it from achieving cosmologically significant results.  Here, as well as in most recent literature, we assume that the whole cosmic opacity arises from an almost complete ionization of hydrogen, after a suitable redshift. More complex reionization histories may have to be taken into account to discuss future data (Bruscoli, Ferrara \\& Scannapieco 2002; Venkatesan 2002), but would not modify general conclusions.  For low--$l$ spectral components, varying $n_s$ has a milder effect than varying $\\tau$. If the whole angular spectrum is considered, other parameters, like $\\Omega_{tot}$, $\\Omega_m$, $\\Omega_b$ (total, matter and baryon density parameters), may be determined independently from low--$l$ features. Here we shall assume that they are already known from such high--$l$ anisotropy measures.  Dealing with low $l$'s, a significant part of our analysis will be devoted to inspect cosmic variance. Its impact on $C_l$ determination is known. Here, however, we are interested in measuring different quantities and a knowledge of $C_l$'s is not even required. We shall explore this point by simulating and analyzing a large number of artificial data sets. Among other things, we shall see that the SPOrt experiment, although far from granting a stringent $\\tau$ determination, may improve the {\\it Stompor limit} in a significant number of cases.  In the next section we shall first debate which range of values can be expected for $\\tau$. We shall then describe how artificial data sets are produced. In section 4 we shall show the results of their analysis, in a number of different experimental conditions and considering different $\\tau$ values. In section 5, conclusions will be drawn. ",
        "conclusions": "In this paper we performed a likelihood analysis, to determine how a polarization experiment could break the degeneracy between $n_s$ and $\\tau$ determinations, found in pure anisotropy analyses. To this aim we built artificial data, consistent with an experiment in progress to detect polarization, but pushing noise even well below the expected experimental level. The experiment considered will cover 80$\\, \\%$ of the whole sky and artificial data were built assuming such sky coverage. If data were available for the full sky, a slight improvement of the signal can be expected. Artificial data were then analyzed. In respect to future observers, we have however the critical advantage to be able to analyze as many skies (realizations) we need. This is significant, in our case, as the $n_s$--$\\tau$ degeneracy, in principle, is readily overcome from low--$l$ spectral components, where, however, cosmic variance can be critical. Our results, however, indicate that cosmic variance is not such a severe limitation, if a sufficiently low noise level is attained. This is one of the most important conclusions of our analysis, as it confirms that, to detect $\\tau$, polarization experiments should aim to high sensitivity, while high resolution does not bear a direct relevance.  The critical issue, in this respect, is whether and when firm lower limits on $\\tau$ are obtainable. The noise level to be attained, to implement such aim, obviously depends on the physical value of $\\tau$ itself. Available constraints on $\\tau$, besides that from CMB anisotropy data, can be also related to the detection of hydrogen lines in high--$z$ QSO spectra. This however leaves us a still significant range of possible cosmic opacities to CMB photons.  If $\\tau$ is not negligibly low, the noise level below which we expect that lower limits on $\\tau$ can be obtained is $\\sigma^P_{\\rm pix} \\simeq 1$--2$\\, \\mu$K, for an instrument with an angular resolution similar to COBE. A detection, at higher $\\sigma^P_{\\rm pix} $, cannot be however excluded. We compared such general conclusion with the actual expected features of the SPOrt experiment and found that, for some of the greatest values for $\\tau$ allowed by the cosmic reionization physics, and assuming the best possible performance of the experiment, a detection of $\\tau$, at the 2--$\\sigma$ level, is possible.  Acknowledgements -- The public programs CMBFAST, by U. Seljak \\& M. Zaldarriaga, was widely used here, together with its generalization to dynamical dark energy models due to R. Mainini. The public program HEALPix, by K.M. G\\`orski et al. was also widely used in the preparation of this work. Thanks are also due to Stefano Cecchini and Marco Tucci for discussions and comments."
    },
    "0209/astro-ph0209263_arXiv.txt": {
        "abstract": "The study of gas and dust at high redshift gives an unbiased view of star formation in obscured objects as well as the chemical evolution history of galaxies.  With today's millimeter and submillimeter instruments observers use gravitational lensing mostly as a tool to boost the sensitivity when observing distant objects.  This is evident through the dominance of gravitationally lensed objects among those detected in CO rotational lines at $z>1$. It is also evident in the use of lensing magnification by galaxy clusters in order to reach faint submm/mm continuum sources.  There are, however, a few cases where millimeter lines have been directly involved in understanding lensing configurations.  Future mm/submm instruments, such as the ALMA interferometer, will have both the sensitivity and the angular resolution to allow detailed observations of gravitational lenses. The almost constant sensitivity to dust emission over the redshift range $z \\approx 1-10$ means that the likelihood for strong lensing of dust continuum sources is much higher than for optically selected sources. A large number of new strong lenses are therefore likely to be discovered with ALMA, allowing a direct assessment of cosmological parameters through lens statistics.  Combined with an angular resolution $<0\\ffas1$, ALMA will also be efficient for probing the gravitational potential of galaxy clusters, where we will be able to study both the sources and the lenses themselves, free of obscuration and extinction corrections, derive rotation curves for the lenses, their orientation and, thus, greatly constrain lens models. ",
        "introduction": "Rapid progress in the development of millimeter astronomical facilities, such as the increase of antennae sizes and/or of the number of array elements, or as the continuing improvement in the sensitivity of detectors, have now made it possible to explore the high redshift universe in this window and therefore to exploit the potentialities of gravitational lensing effects.  Why is it so important to explore this wavelength domain? In one short statement: the presence of cold dust and of molecular material can be traced in this window and both components witness the formation of heavy elements. If they are detected in galaxies at high redshifts, they allow us to probe star formation in the early universe. They reveal as well processes related to the startup of active galactic nuclei (AGN) and signal the presence of massive black-holes. Large amounts of molecular gas are encountered in the close environment of the central engine in AGN. This material is often regarded as the fuel which allows to activate the AGN. An evolutionary scenario would then connect IR-luminous galaxies rich in molecular material and with intense star formation to the formation/feeding of massive black holes.  Several fundamental questions are therefore underlying the search for dust and molecular gas at high redshift: the redshift of galaxy formation? the chemical evolution of the universe with time? the evolution of the dust content in the universe at early ages? the epoch and the scenario of the formation of massive black holes? the startup and evolution of AGN activity? Do we have already some clues to answer these questions? The most powerful AGNs, quasars, are now detected up to redshift around 6.5, and galaxies up to redshift 6. So we know that in this redshift range the universe already hosted galaxies and massive black holes and that its metal content was substantial since the spectra of high redshift AGNs are very similar to those of low redshift AGNs. Yet, only a few objects are known at these high redshifts and this may provide a biased view. It is therefore mandatory to enlarge the sample and of course, the goal is also to push the redshift limit.  Pushing the redshift limit also means that we are investigating sources with lower and lower flux density. This can be achieved through technical improvements, using larger collectors and better detectors. The ALMA project is showing the way. Another manner is to take advantage of the effects induced by gravitational lensing (for a review of its theoretical basis, see the comprehensive book by Schneider et al.~\\cite{schneider92}). Firstly, image magnification allows us to detect more distant sources of cold dust and molecular gas of a given intrinsic luminosity, or to detect at a given redshift sources of fainter intrinsic luminosity. The latter in particular is important for good determinations of luminosity functions. Secondly, differential magnification effects can be used as an elegant tool to probe the size of molecular and dusty structures in the lensed source, as long as the lensing system provides the appropriate geometry. In this case, it is imperative to have an excellent model of the lensing system, as any structural information about the source itself for example is recovered by tracing the image back through the lensing system. Both aspects will be discussed at length in this paper. In some cases molecular absorption lines allow us to obtain information about the lensing galaxy itself.  Apart from hydrogen and helium, carbon and oxygen are the heavy elements with highest abundance in the universe. Therefore, the CO molecule is the most suitable candidate for the detection of molecular gas in emission at high redshift. The CO molecules can be detected directly through their thermal line emission in the source or as silhouetted absorbers along the line of sight to a background source. The latter may occur for example for the lensing galaxy. Several other molecules have been detected at high redshift, HCO$^+$, HCN, H$_2$CO..., while dust is detected essentially through its thermal emission.  \\medskip  We provide in Sect.~\\ref{molemission} an overview of the CO line emission and of the high redshift CO sources detected so far. The role played by gravitational lensing in studying CO sources at high redshift is highlighted. In Sect.~\\ref{molabs0} we review molecular absorption and the importance of such measurements to investigate the properties of the lensing galaxies. Sect.~\\ref{dustcont} introduces the dust continuum emission and the use of differential magnification effects which can be made to probe the dust content of the lensed objects. Three cases particularly well studied are discussed in detail in Sect.~\\ref{casestudies}. In Sect.~\\ref{pks1830model} existing lens models for PKS1830-211 are reviewed and a new one introduced. Finally, in Sect.~\\ref{futureprosp} the future of this type of investigations is presented in the perspective of new instrumental developments in general and of ALMA in particular. ",
        "conclusions": "\\label{summary}  The study of gas and dust at high redshift is important for several reasons. It gives us an unbiased view of star formation activity in obscured objects and it tells the story of the chemical evolution and star formation history in galaxies through the amount of processed gas (and dust) it contains. With today's millimeter and submillimeter facilities, this research area has used gravitational lensing mostly as a tool to boost the sensitivity. This is evident through the preponderance of gravitationally lensed objects among those which have been detected at $z > 2$ in the lines of the CO molecule. It is also evident in the use of lensing magnification by galaxy clusters in order to reach faint submm/mm continuum sources. There are, however, a few cases where millimeter lines have been directly involved in understanding lensing configurations. The best example of this is the highly obscured PKS1830-211, where the lens was identified through molecular absorption lines and where these lines give a velocity dispersion measure by originating in two different regions of the lens. The molecular absorption lines in this system have also been used to derive the differential time delay between the two main components, the main objective being to determine the Hubble constant, but also adding to the constraints in modeling this particular lens system.  With future millimeter and submillimeter instruments, such as ALMA, coming on-line, the situation is likely to change drastically. The sensitivity of ALMA will be such that it does not need the extra magnification from lensing to observe very distant objects. Instead it will be used to study the lensing itself. The more or less constant sensitivity to dust emission over a redshift range stretching from $z \\approx 1$ to $z \\approx 10$ means that the likelihood for strong lensing of dust continuum detected sources is much larger than for optically selected sources. ALMA will therefore discover many more lenses and allow a direct assessment of cosmological parameters through lens statistics. Weak lensing will also be an area where ALMA can successfully contribute. Again, the high sensitivity to dust emission out to very high redshifts, combined with an angular resolution $< 0\\ffas1$, and a more beneficial `PSF' will make ALMA more efficient for probing the potential of galaxy clusters than present day optical/IR telescopes. In addition we will be able to study both the sources and the lenses themselves, free of obscuration and extinction corrections, derive rotation curves for the lenses, their orientation and, thus, greatly constrain lens models.  \\bigskip  \\noindent {\\bf Acknowledgments.}\\ \\ T.W. thanks Fran\\c{c}oise Combes for allowing the use of unpublished material on millimeter wave absorption line systems. Many thanks to F. Combes, D. de Mello, P. Cox and F. Courbin for careful reading of the manuscript and for valuable comments."
    },
    "0209/gr-qc0209012_arXiv.txt": {
        "abstract": "Gravitational waves generated by the final merger of compact binary systems depend on the structure of the binary's members.  If the binary contains neutron stars, measuring such waves can teach us about the properties of matter at extreme densities.  Unfortunately, these waves are typically at high frequency where the sensitivity of broad-band detectors is not good.  Learning about dense matter from these waves will require networks of broad-band detectors combined with narrow-band detectors that have good sensitivity at high frequencies.  This paper presents an algorithm by which a network can be ``tuned'', in accordance with the best available information, in order to most effectively measure merger waves.  The algorithm is presented in the context of a toy model that captures the qualitative features of narrow-band detectors and of certain binary neutron star merger wave models.  By using what is learned from a sequence of merger measurements, the network can be gradually tuned in order to accurately measure the waves.  The number of measurements needed to reach this stage depends upon the waves' signal strength, the number of narrow-band detectors available for the measurement, and the detailed characteristics of the waves that carry the merger information.  Future studies will go beyond this toy model, encompassing a more realistic description of both the detectors and the gravitational waves. ",
        "introduction": "\\label{sec:intro}  Much of the promise of gravitational-wave (GW) observation is in its potential as a novel probe of physics and astrophysics.  Because GWs couple very weakly to matter and arise solely from gravitational interactions, sources that are completely dark electromagnetically may be strong GW emitters.  By tracking and measuring the waves generated in violent astrophysical events, we may gain insight into physical processes that cannot be easily measured in other ways.  One example of such hard-to-measure physics is the late merger of compact binary systems.  GW emission carries energy and angular momentum out of the binary, driving the compact bodies ever closer together.  During the early {\\it inspiral} portion of this coalescence process, the structure of the binary's members plays little role; they can be usefully approximated as point masses, or spinning point masses.  As the bodies come closer together, their internal structure becomes very important.  The GWs generated in the final stages of inspiral and {\\it merger}, when the bodies collide and merge into some coalesced state, will carry information about the bodies' structure.  It has long been recognized that, if the binary contains at least one neutron star, merger waves will depend on the nature of neutron star matter (cf.\\ Refs.\\ {\\cite{clark_eardley,300yrs,3minutes,vallis}} and references therein).  This opens the exciting possibility that GW measurements by detectors such as LIGO {\\cite{ligo}} could study the properties of very dense matter, such as its equation of state (EOS) {\\cite{3minutes,klt,lee_eos}}.  Of particular interest will be testing whether neutron stars contain a core of ``exotic'' matter, such as a free quark state of some kind.  Models of compact stars comprised of a free quark fluid, or with a free quark core, can have a structure rather different from ordinary neutron stars --- they are often of smaller radius, and there may exist a sharp density transition at some finite radius from the star's core {\\cite{alford,arrw}}.  If one or both members of a binary system had such a structure, there could be an observable imprint on the merger GW signature.  (We should emphasize, though, that much work remains to evaluate whether such a signature exists, and if so, what is its nature.)  Simulations of binary neutron star merger using Newtonian gravity and a polytropic EOS (see, e.g., {\\cite{zcm,rs}}) have found that the merger waves indeed carry information about the EOS, but that these waves are at very high frequency ($f\\sim 1500 - 3000\\,{\\rm Hz}$) where broad-band LIGO-type detectors do not have good sensitivity.  More recent work using irrotational matter configurations in the conformal approximation to GR {\\cite{fgrt}} shows that the EOS-dependent structure is likely to come out at somewhat lower frequencies, $f\\sim 1000\\,{\\rm Hz}$; this is still high enough that broad-band detector sensitivity is not very good.  If one member of the binary is a black hole, the EOS-dependent information will come out at still lower frequencies, $f\\sim 400 - 1000\\,{\\rm Hz}$ {\\cite{vallis}} --- the larger system mass shifts all frequencies downward.  In this case, broad-band detectors are more useful for studying the merger waves, but still may not be ideal.  To learn as much as possible from the waves generated during the merger, broad-band GW detectors should be supplemented by {\\it narrow-band} detectors.  Acoustic detectors in existence today {\\cite{blair,cerdonio}} and special interferometer topologies under development such as signal recycling {\\cite{sr}} and resonant sideband extraction {\\cite{rse}} have good sensitivity in a narrow band at high frequencies, $\\delta f/f\\sim 0.1 - 0.2$ for $f\\sim 500 - 2500\\,{\\rm Hz}$.  Narrow-band detectors answer essentially a yes-no question: ``Did the binary radiate in my frequency band?''  A ``xylophone'' of narrow-banded detectors would probe gross features of the merger waveform, such as a sharp cutoff in the GW spectrum (seen in recent simulations {\\cite{fgrt}}), or the waves generated by a transient bar that forms in the merger detritus (seen in some Newtonian simulations {\\cite{zcm,rs}}).  Such simple measurements should be robust in the sense that they wouldn't require detailed modeling of the waves' phasing --- very important, since theoretical uncertainites in the merger waveforms are likely to be significant even when these measurements can be made.  Practical considerations such as cost and available facility space will limit the number of narrow-band detectors that can be used for each measurement.  To make best use of these detectors, the network should be designed in a way that is in some sense optimal: the narrow-band detectors should be configured, in concert with the broad-band detectors, so that the network of all detectors is most likely to provide new information about merger waves, given our best present knowledge of the waves' properties.  How one ``tunes'' a detector network in this manner is the subject of this paper.  We assume that an inspiral has already been measured, so that we know merger waves must be present in the data.  For the purposes of this analysis, we assume further that the waveform depends on a single parameter $\\lambda$ that grossly characterizes the merger waves.  This $\\lambda$ could be the frequency of a sharp cutoff in the wave spectrum, or the frequency at which a short-lived bar may radiate for several cycles.  Theoretical modeling allows us to phenomenologically relate this parameter to a description of the binary's stars.  For example, in models in which the wave spectrum sharply cuts off, $\\lambda$ is most strongly related to the compactness of the stars: smaller stars exhibit a cutoff at higher frequencies {\\cite{fgrt}} since they spiral in further before the spectrum cuts off.  Describing the merger features with a single parameter is no doubt an oversimplification, but is useful for demonstrating how network tuning works to zoom-in on gross features of the waves.  Tuning the network means finding the configuration which measures $\\lambda$ with as little error as possible.  We develop a tuning algorithm that does just this: it configures the network to measure $\\lambda$ as accurately as possible, given our uncertainty in $\\lambda$'s value, and updates (``retunes'') the network as measurements teach us about merger waves.  This algorithm is based on the maximum likelihood GW measurement formalism developed by Finn {\\cite{finn92}}.  Finn defines two probability distributions which play a major role here: the {\\it prior} probability, $p_0(\\lambda)$, summarizing all that is known about $\\lambda$ before measurement; and the {\\it posterior} probability, $P_{\\rm post}(\\lambda |\\hat\\lambda)$, summarizing what is known afterward. The posterior distribution is built from the measured datastream, and so explicitly depends upon the detector network's characteristics and on the measured waveform $h(\\hat\\lambda)$, where $\\hat\\lambda$ is the unknown, true value of $\\lambda$.  Following measurement, the posterior probability is the tool one uses to estimate the values of the parameters describing GWs, and also to estimate the error in those values {\\cite{finn92,finnchernoff,cf}}. It is not quite the tool needed here: we need to estimate the accuracy with which $\\lambda$ is likely to measured, but we need this estimate {\\it before} measurement.  To this end, we introduce an additional probability, the {\\it anticipated} distribution of $\\lambda$.  This is a marginal distribution found by integrating out the dependence of the posterior probability upon the unknown true parameterization $\\hat\\lambda$: \\begin{equation} P_{\\rm ant}(\\lambda) = \\int p_0(\\hat\\lambda)P_{\\rm post}(\\lambda |\\hat\\lambda)\\,d\\hat\\lambda\\;. \\end{equation} This is the distribution that we we anticipate will describe $\\lambda$ after a measurement.  Like the posterior probability, it depends on the detector network, so we can tune the network's adjustable parameters to find the network which we anticipate will measure $\\lambda$ as accurately as possible, given our current ignorance of the merger waves.  From the anticipated distribution's definition, it is simple to update and improve the network as we learn more about $\\lambda$.  Following a measurement, we construct the posterior distribution $P_{\\rm post}(\\lambda |\\hat\\lambda)$ from the datastream.  We then use this posterior distribution as the prior distribution for the next measurement: $p_{0}^i(\\lambda) = P_{\\rm post}^{i-1} (\\lambda |\\hat\\lambda)$.  We update the anticipated distribution with the new priors, and then update the network.  The detector network is thereby adjusted and improved following each measurement, so that our ability to measure $\\lambda$ is incrementally improved by each measurement. Our gradually improving knowledge of $\\lambda$'s value is manifested as a gradual peaking of this distribution: we begin with a prior describing complete ignorance (uniform distribution) and find that after several merger measurements the distribution begins to peak around the true value $\\hat\\lambda$.  The more we learn about the merger waves, the more sharply peaked becomes this distribution.  In this paper, we demonstrate the concept and principles of network tuning using a toy model for the merger and for the narrow-band detectors.  In Sec.\\ {\\ref{sec:formal}}, we describe in more detail the probability distributions introduced above, and then discuss how we use them to tune our network designs.  Our tuning procedure is given explicitly at the end of this section.  We next present our toy model in Sec.\\ {\\ref{sec:toy}}: we approximate narrow-band detectors as zero bandwidth (delta function) GW detectors with adjustable center frequency, and treat the waveform $h(t;\\lambda)$ as a quadrupole inspiral chirp up to a merger frequency $f_m\\equiv\\lambda\\times 1000\\,{\\rm Hz}$.  This highly simplified description throws away important physics, particularly the finite bandwidth expected in real narrow-band detectors and the slower frequency rolloff seen in recent merger computations (e.g., Ref.\\ {\\cite{fgrt}}).  It is simple enough, though, that many of the calculations needed can be done analytically, and is close enough to the real problem that it should give a good sense of how network tuning is likely to proceed in practice.  We test our tuning algorithm in Sec.\\ {\\ref{sec:results}}.  We show that measurements can converge on an accurate value for $\\lambda$ after measuring some number of binary merger events.  The size of that number depends on the signal strength (a few strong signals can drive convergence rather quickly) and the number of narrow-band detectors available for the measurement (having at least two available can speed up convergence quite a bit).  The good behavior of our tuning procedure can be taken as an indication that this algorithm is robust. However, we cannot pretend that our analysis is in any way definitive: the toy description of the detectors and the waveform neglects several important effects.  This analysis should therefore be regarded as a proof-of-concept presentation; future work, discussed in Sec.\\ {\\ref{sec:future}}, will put the various complications neglected here back where they belong.  In particular, we plan to use more realistic descriptions of the narrow-band detectors (cf.\\ Refs.\\ {\\cite{whitepaper,hhs}}), and merger waveforms taken from recent computational models. ",
        "conclusions": "\\label{sec:future}  The results of the previous section show that network tuning works effectively, at least within the context of the toy model: in all cases, we find that after some number of measurements the probability distribution describing what is known about $\\lambda$ will become sharply peaked about $\\hat\\lambda$.  The number of measurements that are needed depends quite a bit on the particulars of the waveform and of the detector network used.  In particular, we can draw two conclusions that are not at all surprising:  \\begin{itemize}  \\item Multiple narrow-band detectors helps the measurement process greatly.  In particular, the improvement in going from one narrow-band detector to two detectors can be significant.  \\item The rate at which measurements converge onto an accurate description of the merger waves depends strongly on those waves' measured SNR.  These can be seen in the drastic difference in the convergence of measurement sequences for $\\hat\\lambda = 0.8$ and $\\hat\\lambda = 1.2$ at fixed inspiral SNR (Fig.\\ {\\ref{fig:snr10_nb1_lh0.8}} versus Fig.\\ {\\ref{fig:snr10_nb1_lh1.2}}) and in the drastic difference seen when the wave amplitude is increased by a factor of 3 (Fig.\\ {\\ref{fig:snr10_nb1_lh1.2}} versus Fig.\\ {\\ref{fig:snr30_nb1_lh1.2}}).  \\end{itemize}  From the first of these conclusions, we advocate investigating the possibility of running at least two detectors in the world-wide GW detector network in a narrow-band configuration.  For example, LIGO already has 3 running interferometers (a 4 km broad-band detector and a 2 km broad-band detector at Hanford, Washington, plus a 4 km broad-band detector at Livingston, Louisiana).  Room has been made in the facilities for additional interferometers, though cost is likely to limit the number that can actually be installed.  If a total of 4 interferometers can be used in the LIGO facilities, it may be worthwhile to reconfigure one of the broad-band detectors as a narrow-band instrument in order to search for merger waves, assuming the loss of a broad-band detector would not seriously impact other science goals (e.g., if other detectors worldwide are able to fill the gap).  Second, we strongly advocate continued theoretical efforts to model and understand the properties of merger waves.  Of particular interest are robust characteristics such as spectral breaks and features that should be measureable without needing detailed models of the waves' phasing.  This kind of understanding will make it possible to choose a parameterization of the waves, similar to our parameter $\\lambda$, that leaves a strong mark on the waveform and can be measured reasonably well.  A robust theoretical foundation for the merger waves will make it possible to choose our priors and configure our network usefully so that measurements will teach us about the merger waves relatively quickly.  As we have repeatedly emphasized, this analysis should be considered a first, proof-of-concept presentation of how network tuning can work. We believe it is very important that, having presented the principles and a simple example of how they work, this effort be followed by a detailed followup analysis that uses a description of the waves and the detectors that is more sophisticated than our toy model.  In wave modeling, we must investigate merger waveforms that are not as trivially simple as those used in the toy model.  In principle, this is not too difficult even now --- some groups have already produced examples of merger waves that show the influence of the dense matter EOS {\\cite{rs,fgrt}}.  Those waveforms could be built into this analysis without too much difficulty.  In the detector description, it is very important that the true bandwidth behavior of the narrow-band interferometers be included --- our infinitesimal bandwidth description is clearly inadequate.  Real detectors will be far more complicated than the toy description given here, and many of these complications are actually coupled (for example, changing the center frequency of a narrow-band detector impacts its bandwidth).  A realistic assessment of how well network tuning can work and how it should be implemented must take into account these various coupled complications.  It would also be useful to include acoustic narrow-band detectors in this analysis {\\cite{hhs}}, in order to assess what role they could play in concert with the broad-band interferometric detectors.  It would not be too surprising if including a non-zero bandwidth detector in this description improved the performance of network tuning --- finite bandwidth detectors will sample a moderate span of frequency, and can thus look for merger power more broadly than do the toy detectors considered here.  This comes at a bit of cost: the broader the bandwidth of the detector, the less amplitude sensitivity it has at its center frequency.  A combination of very narrow-band and moderately narrow-band detectors may turn out to provide the best of both worlds.  It would also be interesting to explore how well dynamical tuning might work: if it is possible to detect an inspiral in real time (plausible for strong sources measured in advanced detectors), we may be able to adjust the center frequency of the narrow-band interferometer to follow the gravitational waves as they evolve through the late inspiral and merger.  This could significantly speed up convergence of the parameter distributions.  Other simplifications that we have used here should also be removed in a follow-up analysis.  In particular, in simulating a sequence of measurements we have assumed that all measurements are at some fixed SNR.  This is obviously incorrect.  In setting the measured SNR, it would be more appropriate to assume a uniform distribution in volume out to some distance (say 500 Mpc), and to set the distance to each source according to that distribution.  Then, each measured event should have its angular function $\\Psi$ [defined in the text following Eq.\\ (\\ref{eq:probability5})] taken from the appropriate distribution (cf.\\ Ref.\\ {\\cite{finnchernoff}}).  Also, when we construct the posterior distribution following a simulated measurement, it would be much more appropriate to construct a noise instance ${\\vec n}$ and hence simulate the data stream ${\\vec g}$, rather than ensemble averaging.  We advocate undertaking this kind of follow-up analysis soon, so that these algorithms are well-understood when it become possible to actually construct the relevant detector networks, and so that what is learned from them can impact the design of future detectors."
    },
    "0209/astro-ph0209055_arXiv.txt": {
        "abstract": "The BTC40 Survey for high-redshift quasars is a multicolor search using images obtained with the Big Throughput Camera (BTC) on the CTIO 4-m telescope in $V$, $I$, and $z$ filters to search for quasars at redshifts of $4.8 < z < 6$.  The survey covers 40 deg$^2$ in $B$, $V$, \\& $I$ and 36 deg$^2$ in $z$.  Limiting magnitudes (3$\\sigma$) reach to $V = 24.6$, $I = 22.9$ and $z = 22.9$.  We used the ($V-I$) vs. ($I-z$) two-color diagram to select high-redshift quasar candidates from the objects classified as point sources in the imaging data.  Follow-up spectroscopy with the AAT and CTIO 4-m telescopes of candidates having $I \\leq 21.5$ has yielded two quasars with redshifts of $z = 4.6$ and $z = 4.8$ as well as four emission line galaxies with $z \\approx 0.6$. Fainter candidates have been identified down to $I = 22$ for future spectroscopy on 8-m class telescopes.  \\vskip 1.5in ",
        "introduction": "Surveys for faint quasars at $z > 4.5$ and the subsequent constraints they place on the quasar luminosity function (QLF) will eventually determine how luminosity evolution and density evolution each contribute to the declining space density of quasars established for $3 \\lesssim z \\lesssim 4.3$ (Warren, Hewett, \\& Osmer 1994, hereafter WHO; Schmidt, Schneider, \\& Gunn 1995, hereafter SSG; Kennefick, Djorgovski, \\& DeCarvalho 1995) and now observed out to redshifts of $z\\sim5$ (Fan \\etal\\ 2001b).  The QLF in turn is an important input to models of structure formation in the early Universe (Haehnelt, Natarajan \\& Rees 1998).  Additionally, high-redshift quasars provide insight into the nature of quasars and their environments in the early Universe (Haehnelt \\& Kauffmann 2000), contribute to the ionizing UV background (Madau, Haardt \\& Rees 1999), and act as background illumination for absorption-line studies of the intergalactic medium.  It is therefore important to continue to search for quasars of all luminosities at the highest possible redshifts, and thus the earliest possible epochs.  We began this survey to address the shape of the QLF at redshifts $z \\gtrsim 5$. WHO found evidence that the positive evolution in the QLF at $0 < z < 2.2$ continues to $z\\approx 3.3$ and then the space density declines by a factor of 6.5 at $z = 4$.  SSG found that space densities have a maximum between $z=1.7$ and $2.7$ and then decrease by a factor of 2.7 per unit redshift beyond $z=2.7$.  Extrapolations of these QLF's to $5 < z < 6$ predict 0.02 (WHO) to 0.6 (SSG) quasars per deg$^2$ to $I=22$.  The major effort to find quasars at higher redshifts is the Sloan Digital Sky Survey (SDSS), which continues to be remarkably successful at finding $z > 4$ quasars.  SDSS has discovered more than 100 such objects including one at $z = 6.3$, the most distant published (see, e.g. Fan \\etal\\ 2000, 2001c; Anderson \\etal\\ 2001).  However, SDSS is limited to $z > 4.5$ quasars with M$_B \\lesssim -26$ and misses the bulk of the population, which is found at lower luminosities.  Thus, there is a need for surveys to find less luminous $z > 4.5$ quasars to address the shape of the faint end of the QLF.  Sharp \\etal\\ (2001) have presented initial results from one such survey, finding two $z > 4.5$ quasars to $i \\approx 21.5$ in 10 deg$^2$ of $griz$ data.  The BTC40 survey is a deep, 40 deg$^2$ survey in $BVIz$ filters undertaken to search for clusters of galaxies, morphologically-selected gravitational lenses, and quasars at $z \\gtrsim 4.8$.  The results on clusters and gravitational lenses will be presented elsewhere.  In this paper we present results of our efforts using a 4-m telescope and large format camera to complement the SDSS quasar search and extend it to lower luminosities.  We used the $VIz$ imaging data to select quasar candidates over 36 deg$^2$ of sky down to $I \\leq 22$, corresponding to absolute magnitudes of $M_B \\lesssim -24.7$ at $z=4.8$.  The selection process compared the expected colors of quasars at redshifts $4.8 < z < 6$ to the locations of catalogued stellar objects in ($V-I$) vs. ($I-z$) color space. Follow-up spectroscopy at the CTIO 4-m and AAT was attempted to $I < 21.5$ and has resulted in the discovery of two quasars with redshifts of $z = 4.6$ and $z = 4.8$, as well as several emission-line galaxies at $z \\approx 0.6$.  Spectroscopy of the fainter candidates, down to $I=22$, will be the focus of our future efforts on larger telescopes.   We describe the survey imaging data in $\\S$2, the candidate selection in $\\S$3, and the follow-up spectroscopy in $\\S$4.  We discuss our results in $\\S$5. ",
        "conclusions": "In our high-redshift quasar survey we have found two $I<21.5$ quasars with $z = 4.6$ and $z = 4.8$ in 36 deg$^2$, proving the validity of the selection technique.  Although our candidate selection is designed to be most sensitive for $z \\gtrsim 4.8$, when the \\lya\\ emission line has moved into the $I$ filter, slightly lower-redshift ``lyman-break'' quasars may also enter the sample.  This was the case with the BTC2340$-$3949.  To compare our results with SDSS, Anderson \\etal\\ (2001) found 29 quasars with $z \\gtrsim 4.5$ in $\\approx$ 700 deg$^2$ of the SDSS commissioning data, for a surface density of 1 quasar per 24 deg$^2$ to i$^*$ $\\leq$ 20.5.  Four of these quasars had $z > 5$, or 1 per 175 deg$^2$.  In the fall equatorial stripe of the SDSS commissioning data, Fan et al. (2001a) found 5 quasars with  $4.5 \\leq z \\leq 4.77$  to $i^*$ $\\lesssim$ 20 in 182 deg$^2$, or 1 in 36 deg$^2$.  Thus, our findings from the BTC40 survey are consistent with the early SDSS results, although the BTC40 statistical base is admittedly very small.  More formally, the expected number of quasars predicted by a given survey may be computed by numerically integrating the QLF determined by the survey over the redshift and magnitude ranges of interest, and multiplying by the efficiency of the survey. Although we have not found enough quasars to derive a luminosity function, our results can be compared to predictions based on the QLF determined by the SDSS team.  Fan \\etal\\ (2001b) used the 39 quasars from the SDSS commissioning data to derive a QLF over the range $3.6 < z < 5$ and $-27.5 <$ \\Mftf $< -25.5$, where \\Mftf\\ is the absolute continuum magnitude measured at $\\lambda = 1450$(1+$z$) \\AA\\ and calculated in the AB system.  Since our quasar spectra are not spectrophotometric we scaled them to return the $I$ magnitudes previously measured from the imaging data and then determined AB(1450(1+$z$)).  Assuming the continuum follows a power law with slope $\\alpha$ = $-0.5$, then \\Mftf = $-26.6$ for BTC40 J2340$-$3949 and \\Mftf = $-26.8$ for BTC40 J1429+0119, and both quasars occupy the parameter space probed by SDSS.  When integrated over $4.5 < z < 5$, the luminosity function of Fan \\etal\\ (2001b) predicts a surface density of $\\approx$0.026 quasars per square degree down to $i^{\\prime} \\approx 20.3$, or $I=19.89$ using the conversion between the AB and conventional magnitude systems from Fukugita \\etal\\ (1996).  In the 36 deg$^2$ of the BTC40 survey, therefore, we would expect to find $\\approx$1 quasar with redshift $4.5 < z < 5$ and $I < 19.9$, and in fact the two quasars we found fall into this category. Furthermore, the absence of z $>$ 5 quasars in our sample to date is understandable, given their scarcity in the SDSS fields.  The SDSS luminosity function predicts 0.015 quasars per deg$^2$ with $5 < z < 6$ to $I\\approx20$, or $<1$ quasars in the 36 deg$^2$ of our survey.  The goal remains to determine the quasar luminosity function at fainter magnitudes. From the SDSS luminosity function over $4.5 < z < 5$ we would expect to find 10 quasars down to $I=21.5$ and 20 quasars to $I=22$ in our survey, while the QLF of SSG predicts 15 quasars to $I=21.5$ and 35 to $I=22$.  Although we implemented candidate selection to $I=21.5$, in the end we attempted spectroscopy of only five candidates with $I > 21$.  None of the resulting spectra were good enough to allow identification, but most likely the objects are stars given the absence of any obvious emission lines.  Constraining the number of faint quasars in our survey may require additional $z$-band imaging data to reduce the scatter in the color-color diagram at faint magnitudes and enable candidate selection closer to the stellar locus.  Follow-up spectroscopy on the $I > 21$ candidates will also necessitate the use of 8-10m telescopes to achieve a reasonable efficiency.  In addition to the selected candidates down to $I = 21.5$ remaining to be observed, we have identified 275 potential candidates with $21.5 < I < 22$. Of course, effective use of time on the largest telescopes will require additional work to keep cool stars from ending up in the candidate list.  As an example, Fan \\etal\\ (2001c) used $J$-band photometry to pare L- and T-type dwarfs from their list of $i$-band dropouts in a search for quasars at $z \\simeq 6$.  In summary, to date we have obtained spectroscopic observations of the brightest candidates in our survey for high-redshift quasars. We have not yet found any quasars with $z > 5$, and in hindsight this is not surprising given the results from SDSS.  But the SDSS results also predict our survey area will contain $\\sim$5 quasars with $z > 5$ to $I = 21.5$, and 11 to $I = 22$. The next important step is to follow up on the fainter objects, which will require observations with 8-m class telescopes."
    },
    "0209/astro-ph0209325_arXiv.txt": {
        "abstract": "{We report the firm discovery of solar-like oscillations in a giant star. We monitored the star $\\xi$~Hya (G7III) continuously during one month with the CORALIE spectrograph attached to the 1.2m Swiss Euler telescope. The 433 high-precision radial-velocity measurements clearly reveal multiple oscillation frequencies in the range 50 -- 130 $\\mu$Hz, corresponding to periods between 2.0 and 5.5 hours. The amplitudes of the strongest modes are slightly smaller than $2\\ms$. Current model calculations are compatible with the detected modes. ",
        "introduction": "Doppler studies with high-precision instruments and reduction algorithms have been refined dramatically, mainly in the framework of the search for exo\\-planets.  These refinements have led to a breakthrough in the observations of solar-type oscillations, which have now been found repeatedly (Procyon, Martic et al.\\ \\cite{martic}; $\\beta$~Hyi, Bedding et al.\\ \\cite{bedding}; $\\alpha$~Cen\\,A, Bouchy \\& Carrier \\cite{bouchy3}; $\\delta$~Eri, Carrier et al.\\ \\cite{carrier}).  The signal-to-noise ratio (S/N) in the oscillation frequency spectra is, for each of these cases, so good that the resemblance with the solar oscillation spectrum is obvious.  Observations of solar-like oscillations in the giant star $\\alpha\\,$UMa have been claimed by Buzasi et al.\\ (\\cite{buzasi}), based upon space photometry gathered with the star tracker of the WIRE satellite.  However, the interpretation of these reported oscillations frequencies is not straightforward. Guenther et al. (\\cite{guenther}) find a possible solution in terms of a sequence of radial modes with a few missing orders for a star of 4.0--4.5 \\Msun. The interpretation is not supported by theoretical calculations by Dziembowski et al. (\\cite{dziembowski}). Velocity observations of Arcturus provide evidence for solar-type oscillations with periods from 1.7 to 8.3 days and a frequency separation of evenly spaced modes of $\\Delta \\nu \\sim 1.2\\ \\mu$Hz (Merline, 1999). WIRE data (Retter et~al. \\cite{retter}), however, points to an excess power at 4.1~$\\mu$Hz and a frequency spacing of $\\Delta \\nu = 0.8~\\mu$Hz.  In this {\\it Letter\\/}, we provide clear evidence for the presence of solar-type oscillations in the giant star $\\xi$~Hya ($m_V=3.54$). This star has a mass close to $M = 3.0 \\Msun$, and is thus considerably heavier than the Sun. Moreover, its luminosity amounts to $L \\sim 61 \\Lsun$ and its effective temperature $\\teff = 5000$~K, which places the star among the giants. In the current {\\it Letter\\/} we present the first results of our study. Detailed modelling will be presented, when completed, in a subsequent paper. ",
        "conclusions": "The main conclusions of this study are as follows.  Solar-like oscillations have been firmly discovered in the bright G7III star $\\xi$~Hya. The amplitudes of the strongest modes are somewhat below $2\\ms$.  The observed frequency distribution of the modes detected (Table 1) is in agreement with theoretically calculated frequencies both in terms of the spacing and the absolute values.  The modes with the largest amplitudes can be well matched with radial modes that have an almost equidistant separation around 7.1~$\\mu$Hz.  A most important and exciting result of our study is the confirmation of the possibility, suggested by the results reported on $\\alpha$~UMa and Arcturus, to observe solar-like oscillations in stars on the red giant branch. This opens the red part of the HR\\,diagram for detailed seismic studies. The latter require an accuracy within the range of current and future ground-based instruments.  Such future studies will only be successful if an extremely high stability of the instrument is achieved and if one performs multisite observing campaigns covering several months in order to resolve the frequency spectrum of the oscillations and to eliminate the aliasing problems."
    },
    "0209/astro-ph0209439_arXiv.txt": {
        "abstract": "We present $K$-band imaging of all 49 radio galaxies in the 7C--I and 7C--II regions of the 7C Redshift Survey (7CRS). The low--frequency (151 MHz) selected 7CRS sample contains all sources with flux-densities $S_{151} > 0.5$ Jy in three regions of the sky.  We combine the $K$-band magnitudes of the 7CRS radio galaxies with those from the 3CRR, 6CE and 6C$^\\star$ samples to investigate the nature of the relationship between $K$-magnitude and redshift and whether there is any dependence upon radio luminosity. We find that radio galaxies appear to belong to a homogeneous population which formed the bulk of their stars at high redshifts ($z_{\\rm f}>5$) and evolved passively from then until they reach a mean present-day luminosity of $3\\,L_{\\star}$. We find a significant difference between the $K$-magnitudes of the 7CRS and 3CRR radio galaxies with the 7CRS galaxies being $\\approx 0.55$ mag fainter at all redshifts. The cause of this weak correlation between stellar and radio luminosities probably lies in mutual correlations of these properties with the central black hole mass. We compare the evolution-corrected host luminosities at a constant radio luminosity and find that the typical host luminosity (mass) increases by approximately $1\\,L_{\\star}$ from $z \\sim 2$ to $z \\sim 0.5$ which, although a much smaller factor than predicted by semi-analytic models of galaxy formation, is in line with results on optically-selected quasars. Our study has therefore revealed that the small dispersion in stellar luminosity of radio galaxies around $3\\,L_{\\star}$ includes subtle but significant differences between the host galaxies of extreme- and moderate-power radio sources at fixed redshift, and between those of high- and low-redshift radio sources at fixed radio luminosity. ",
        "introduction": "The luminous radio and narrow-line emission from radio galaxies provides a convenient method for selecting at least a subset of the most massive galaxies at a wide range of redshifts. At low redshifts, powerful radio sources are known to be triggered only within massive elliptical galaxies (Bettoni et al. 2001). The near-infrared magnitudes of radio galaxies out to beyond $z=5$ (De Breuck et al. 2002) suggest that radio galaxies at all redshifts have extreme stellar masses. Standard CDM-based galaxy formation models predict that structures grow in a hierarchical fashion. The formation epoch and evolution of the most massive galaxies is therefore of relevance to such models.  The near-infrared $K$-band emission from radio galaxies is dominated by stellar light in most cases (Best, Longair \\& R\\\"ottgering 1998; Simpson, Rawlings \\& Lacy 1999). Although narrow emission lines can dominate in some objects (Eales \\& Rawlings 1993), this is confined to only the most radio-luminous objects at $z>2$ (e.g. Jarvis et al. 2001a).  The $K$-band magnitudes of radio galaxies from the bright 3CRR sample (Laing, Riley \\& Longair 1983) follow a tight correlation with redshift (Lilly \\& Longair 1984). The nature of this `$K-z$ relation' showed that the stellar luminosities of high redshift radio galaxies are considerably more luminous than the curve representing no stellar evolution and consistent with a passively evolving population which formed at high redshifts. The small dispersion in $K$-magnitudes at a given redshift evident in the 3CRR sample is also found in other complete samples and continues up to at least $z=3$ (Jarvis et al. 2001a). Together these facts suggest that radio galaxies are a homogeneous population which formed the bulk of their stars at high-redshifts ($z_{\\rm f}>5$) and evolved passively from then until the present day.  Eales et al. (1997) presented $K$-band data for radio galaxies in the 6CE sample which has a flux-density limit a factor of about 5 times lower than the 3CRR sample. They found that the less radio-luminous 6CE galaxies tended to have fainter $K$-band magnitudes than 3CRR galaxies at similar redshifts. They determined a difference of 0.6 mag for the two samples at $z>0.6$, but found that at $z<0.6$ there was no difference between the samples. Such an evolutionary change in the radio-luminosity dependence of stellar luminosities was interpreted by Best et al. (1998) as due to radio luminosity being more closely related to the mass of the black hole at high redshifts than at low redshifts where availability of fuel would dominate. Lacy, Bunker \\& Ridgway (2000) used near-infrared imaging of sources from the 7C--III region of the 7C Redshift Survey (7CRS; selected at a flux-density limit a factor of 20 times lower than 3CRR) to show that radio galaxies at $0.8<z<1.4$ from both the 7CRS and 6CE samples are fainter by 0.4 mag than 3CRR galaxies. They suggest two possible reasons for this correlation between host and radio luminosities: (i) AGN-related contamination of the stellar light; (ii) a correlation of both luminosities with the mass of the central black hole.  It is important in studies of the properties of radio galaxies that the samples used should be close to complete in terms of secure optical identifications and redshifts. There are several biases which can appear for incomplete samples, such as missing objects with the weakest emission lines because spectra are harder to obtain. Similarly, spectroscopically targeting the faintest sources at $K$-band (e.g. with the aim of finding the highest redshift galaxies, De Breuck et al. 2001) will bias the sample in terms of its $K-z$ relation.  In this paper we present near-infrared imaging of a completely-identified sample of radio sources selected at 151 MHz from the 7C catalogue. This sample (regions 7C--I and 7C--II of the 7CRS) has 90\\% spectroscopic redshift completeness (Willott et al. 2002) with reasonable estimates of the redshifts for the remaining few sources from multi-colour photometry (Willott, Rawlings \\& Blundell 2001). We combine these data with those from the 3CRR, 6CE (Rawlings et al. 2001), 6C$^\\star$ (Jarvis et al. 2001a,b) and 7C--III (Lacy et al. 2000) samples to define the $K-z$ relation from over 200 radio galaxies in samples with high spectroscopic completeness.   In Section 2 we present the $K$-band imaging observations for all the radio galaxies in 7C--I and 7C--II. In Section 3 we discuss corrections made to the observed $K$-magnitudes from all the samples to produce a uniform dataset and then discuss the form of the $K-z$ relation and its evolutionary interpretation. Section 4 investigates evidence for a correlation between the stellar luminosity (mass) of the host galaxy and the radio luminosity and also considers the evolution in stellar luminosity at a fixed radio luminosity.  The conclusions are given in Section 5. In the Appendix we present $K$-band images as finding charts for the 7C--I and 7C--II quasars and broad-lined radio galaxies whose properties were discussed in Willott et al. (1998). We assume throughout that $H_0=70~ {\\rm km~s^{-1}Mpc^{-1}}$, $\\Omega_{\\mathrm M}=0.3$ and $\\Omega_\\Lambda=0.7$. ",
        "conclusions": "We have presented $K$-band imaging for the 49 radio galaxies in regions 7C--I and 7C--II of the 7C Redshift Survey. This sample is selected at a low radio frequency of 151 MHz and a flux-density level a factor of 20 lower than the 3CRR sample. The 7C--I and 7C--II data have been combined with other complete samples to define the near-infrared Hubble diagram for a total of 205 radio galaxies at redshifts ranging from 0.05 to 4.4 and investigate the radio-luminosity dependence of the near-infrared magnitudes.  The $K-z$ relation is well fit by a second order polynomial between $K$-magnitude and $\\log_{10} z$: \\begin{equation} K=17.37+4.53 \\log_{10} z - 0.31 (\\log_{10} z)^{2} \\end{equation} This curve is very close to the expected $K$-magnitude evolution of a passively evolving instantaneous starburst model galaxy which formed at $z_f=10$ with present day luminosity of $3\\,L_\\star$. The rms dispersion about this relation is found to be 0.58 magnitudes at all redshifts up to $z=3$. These results are consistent with a high formation redshift ($z_f>5$) for radio galaxies.  Using samples of radio galaxies selected at different flux-density limits, we have investigated the radio luminosity dependence of the $K-z$ relation. We find that there is a significant difference between the $K$-magnitudes of the 3CRR and the fainter 7C radio galaxies over all redshifts. This is best interpreted as being due to a correlation of both properties with black hole mass. The typical evolution-corrected host luminosities decrease at higher redshifts by a factor in the range 1.3-2. This corresponds to a small decrease in the masses of radio galaxies at higher redshifts. The weakness of these correlations of host properties with radio luminosity and redshift aid the interpretation of the strong cosmic evolution of the radio source population. The evolution is due to a higher accretion rate in galaxies of a given mass, presumably due to an increased available mass of gas as fuel supply."
    },
    "0209/astro-ph0209113_arXiv.txt": {
        "abstract": "The ultraluminous compact X-ray sources (ULXs) generally show a curving spectrum in the 0.7--10 keV {\\it ASCA} bandpass, which looks like a high temperature analogue of the disk dominated high/soft state spectra seen in Galactic black hole binaries (BHBs) at high mass accretion rates. Several ULXs have been seen to vary, and to make a transition at their lowest luminosity to a spectrum which looks more like a power law. These have been previously interpreted as the analogue of the power law dominated low/hard state in Galactic BHBs. However, the ULX luminosity at which the transition occurs must be at least 10--50 per cent of the Eddington limit assuming that their highest luminosity phase corresponds to the Eddington limit, while for the Galactic BHBs the high/soft--low/hard transition occurs at a few per cent of the Eddington limit. Here we show that the apparently power law spectrum in a ULX in IC342 can be equally well fit over the {\\it ASCA} bandpass by a strongly Comptonised optically thick accretion disk with the maximum temperature of $\\sim 1$ keV. Recent work on the Galactic BHBs has increasingly shown that such components are common at high mass accretion rates, and that this often characterises the very high (or {\\it anomalous}) state. Thus we propose that the power law type ULX spectra are not to be identified with the low/hard state, but rather represent the Comptonisation dominated very high/{\\it anomalous} state in the Galactic BHBs. ",
        "introduction": "Since the {\\it Einstein} era, many ultraluminous compact X-ray sources (ULXs; Makishima et al. 2000) with X-ray luminosities $L_{\\rm X} \\ge 10^{39-40}~{\\rm erg~s^{-1}}$ have been found in spiral arms of nearby spiral galaxies (e.g., Fabbiano 1989).  Such high values of $L_{\\rm X}$ exceed the Eddington limit, $L_{\\rm E}$, for a neutron star by several orders of magnitude, and instead suggest massive (30--100 $M_\\odot$) accreting black holes (BHs). However, it is difficult to understand how such massive BHs could form: single massive stars have extreme mass loss throughout their life, and the maximum BH mass expected is of order 10--15 $M_\\odot$ (e.g. Fryer \\& Kalogera 2001). Since there are no nearby systems either in our galaxy or in M31 which could be easily studied then the nature of the ULX remained a mystery.  {\\it ASCA} (Tanaka, Inoue, \\& Holt 1994) data led to a breakthrough for these objects by providing the first moderate resolution energy spectra.  As reported by many authors (e.g., Makishima et al. 2000, and references therein), it is clear that the majority of the most luminous ULXs exhibit spectra which are well fitted by the multicolor disk model (MCD model; Mitsuda et al. 1984), similar to the case of Galactic/Magellanic BH binaries (BHBs) at high mass accretion rates, $\\dot{M}$ (e.g. Makishima et al. 1986; the review by Tanaka \\& Lewin 1995).  This, together with the variability of these systems including periodic variation (Bauer et al. 2000; Sugiho et al. 2001) and their general association with regions of ongoing star formation (Zezas, Georgantopoulos, \\& Ward 1999; Roberts \\& Warwick 2000; Fabbiano, Zezas, \\& Murray 2001) has led to their identification with BHBs.  There are then three possibilities, first that these really are intermediate mass BHBs, formed perhaps via mergers of massive stars/BHs in a compact star cluster (e.g. Ebisuzaki et al. 2001). Alternatively, they could be ``normal'' mass ($\\sim 10 M_\\odot$) stellar BHs accreting beyond the critical accretion rate at which the disk luminosity $L_{\\rm disk}$ reaches $L_{\\rm E}$. Recent work suggests that the Eddington limit can be violated by the disk becoming clumpy (e.g., Krolik 1998; Gammie 1998; Begelman 2002 and references therein), and it has long been known observationally that super critical accretion can happen (e.g. Cir X-1). A third alternative is that these are ``normal'' mass BHs, but that  their X-ray luminosity is strongly anisotropic (beamed), so that the bolometric luminosity is overestimated. This beaming is highly unlikely to be the relativistic beaming seen in jet sources as this generally leads to a power-law (PL) type of spectrum (e.g. blazars), very unlike the curving spectrum seen by {\\em ASCA} for the majority of the ULX.  However, strong anisotropy of the disk flux might be produced if the disk is geometrically thick in its inner regions.  The radiation can then be strongly collimated by a funnel (e.g. Madau 1988), although the factor of 10--100 required (King et al. 2001) seems extreme. Such an extreme thick disk might be expected to form only under a super critical accretion rate. So this is merely an addendum to the high $\\dot{M}$ scenario, rather than an independent alternative.   All these alternatives involve an extreme of one kind - either mass, radiation luminosity or disk shape. One way to test these is to compare the ULX spectra and spectral variability with those of Galactic BHBs. This has become much more feasible in recent years with the unprecedented volume of data from the Galactic BHBs gathered by {\\it RXTE}. Here we use the bright Galactic BHB transient XTE~J$1550-564$ to observationally determine the spectra and spectral variability of high $\\dot{M}$ disks, and show that the ULX are indeed compatible with being massive (30--100 $M_\\odot$) BHs accreting at close to the critical accretion rate. ",
        "conclusions": "In \\S3, we have shown that the apparently PL-type spectra seen from IC342 in 2000 is most probably {\\em not} related to the low/hard spectra in Galactic BHBs. It shows significant deviations from a PL shape, in the sense that the spectrum softens at higher energies. This is opposite to the behaviour of the Galactic BHBs in the low/hard state. % We propose that the PL-type spectra are instead the analogue of the {\\it anomalous regime} (also termed the very high state) spectra seen in the Galactic BHs at high luminosities. We demonstrate this by fitting Comptonisation models, {\\it thcomp}, to the PL-type data, and show that they can indeed give as good a fit to the data as a PL continuum plus ionized edge (Paper I). If the ionized edge is not incorporated, the PL fit is significantly inferior to the {\\it thcomp} fit (table 1), because of the intrinsic concaveness of the spectral shape.  Additionally, validity of this interpretation can be tested by investigating the location of the obtained result on the $L_{\\rm disk}$-$T_{\\rm in}$ plane (Fig.1). There, we also show the 0.1--100 keV {\\it thcomp} luminosity, $L_{\\rm thc}=1.1\\times10^{40}~{\\rm erg~s^{-1}~cm^{-2}}$, extrapolated from the best fit model assuming isotropic emission. If the Comptonisation is from overheating of the disk (Beloborodov 1998) then this represents the disk luminosity. If instead it is from a separate corona, then the disk luminosity is amplified by Comptonisation. However, at this low compton $y$ parameter the amplification is rather low (less than a factor of 2), so we use the {\\it thcomp} luminosity as our estimate for the disk luminosity and plot this with the seed photon temperature on Fig~\\ref{fig:t-l}. The PL-type spectrum then lies nicely on the same luminosity-temperature relation as defined by the MCD-type spectra. This predicts a ULX mass between $30~M_\\odot$ ($a=0$) and $150~M_\\odot$ ($a=0.998$) if the PL-type spectrum marks the break between the {\\it standard} and {\\it apparently-standard} (slim disk) regimes. This break then appears at $\\sim 1$--$2 L_{\\rm E}$. If instead the break is at $\\sim 0.1$--$0.2L_{\\rm E}$ as in J1550 then this implies a mass of 50--250 $M_\\odot$.    Based on these observational results, including spectral softening at higher energies, X-ray luminosity, and disk temperature within the framework of the disk Comptonisation, we conclude that the PL-type spectra of IC 342 source 1 in 2000 are related to the standard MCD-type disk with strong disk Comptonisation state rather than to the low/hard state, and that the transition between 1993 and 2000 is likely an {\\it apparently standard} to {\\it anomalous} transition, rather than the canonical high/soft to low/hard transition.     \\subsection*{Acknowledgements}  A.K. is supported by Japan Society for the promotion of Science Postdoctoral Fellowship for Young Scientists. The present work is supported in part by Sydney Holgate fellowship in Grey College, University of Durham."
    },
    "0209/astro-ph0209331_arXiv.txt": {
        "abstract": "{It is becoming clear that `self--pollution' by the ejecta of massive asymptotic giant branch stars has an important role in the early chemical evolution of globular cluster stars, producing CNO abundance spreads which are observed also at the surface of unevolved stars. Observing that the ejecta which are CNO processed must also be helium enriched, we have modelled stellar evolution of globular cluster stars by taking into account this possible helium enhancement with respect to the primordial value. We show that the differences between the main evolutionary phases (main sequence, turn--off and red giants) are small enough that it would be very difficult to detect them observationally. However, the difference in the evolving mass may play a role in the morphology of the horizontal branch, and in particular in the formation of blue tails, in those globular clusters which show strong CNO abundance variations, such as M13 and NGC 6752. ",
        "introduction": "The chemical inhomogeneities (spread in the abundances of CNO, O -- Na and Mg -- Al anticorrelation) observed in globular cluster (GC) members from the main sequence to the red giant branch impinge on problems such as cluster formation and early evolution (see, e.g., Kraft 1994, Gratton et al. 2001). In some clusters the peculiarities are particularly wide spread and strong, as in M13 and NGC 6752 (Da Costa \\& Cottrell 1980, Norris et al. 1981, Smith \\& Norris 1993, Pilachowski et al. 1996, Kraft et al. 1997, Sneden 2000, Gratton et al. 2001). The fact that these clusters present also the ``anomaly\" of an extended and exclusively blue horizontal branch has been noticed by Catelan \\& de Freitas Pacheco (1995), who observe that clusters with super oxygen--poor giants have all very blue HBs. On the other hand, it seems most likely that there is not a direct relation between the two features (i.e., blue tail and anomalies), but rather that they may share the same origin, at least in part.  The most common hypotheses for the origin of the chemical peculiarities are either deep mixing in the cluster members of material nuclearly processed in their interior, or pollution -- partial or total -- by external material (see, e.g., Bell et al. 1981, Cottrell and Da Costa 1981, D'Antona et al. 1983, Kraft 1994, Cannon et al. 1998). The confirmation that turn--off stars show peculiarities quite similar to those observed in giants casts severe doubts on the deep mixing scenario (Gratton et al. 2001).  The primordial pollution hypothesis requires either contamination of intracluster material, out of which new stars form with a chemical composition different from a preceding star generation, or partial contamination (superficial or deep) of pre--existing main sequence structures. The contaminating objects are generally identified with intermediate mass asymptotic giant branch (AGB) stars (4 -- 6 \\Msun\\ stars, see D'Antona et al. 1983, Ventura et al. 2001, 2002; Thoul et al. 2002), which evolve rapidly enough to shed their envelopes in about $10^8$ yr.  The detailed composition of these envelopes is not easy to predict, as it depends on the uncertain physics of convection and mass loss in the intermediate mass AGB stars, but {\\it it will surely show an enrichment in helium}, in part as a consequence of the second dredge--up, and in part due to the third dredge--up: the helium mass fraction should be $\\simgt$0.29 instead of $\\sim$0.24, the cosmic value (Ventura et al. 2001)\\footnote{Notice that the value Y=0.29 for the ejecta of the most massive AGBs is substantially conservative, as the Ventura et al. models make very conservative assumptions on the modalities of the third dredge up.}. When these envelopes are expelled at low velocity  by the stellar bodies during the AGB evolution, various consequences can arise. Given the depth of the potential well, a significant fraction of this material is not lost by a globular cluster, unless there is a significant interaction with external systems (e.g. another cluster, or a giant molecular cloud, or the galactic bulge). New stars can form out of this matter, or pre--existing diffuse material can be polluted, or low mass unevolved stars can be polluted to various degrees.  \\begin{figure} \\centering \\resizebox{8.8cm}{!}{\\rotatebox{0}{\\includegraphics{h3781f1.eps}}} \\caption{Comparison of two isochrones of age 14 Gyr, Z=$2 \\times 10^{-4}$\\ and different helium abundance. The ZAHBs for the two Y values are also shown, down to a minimum HB mass of M=0.6 \\Msun.} \\label{f1} \\end{figure} \\begin{figure} \\centering \\resizebox{8.8cm}{!}{\\rotatebox{0}{\\includegraphics{h3781f2.eps}}} \\caption{The same comparison as in Fig. \\ref{f1}, for Z=$10^{-3}$. In this case, we have extended the ZAHB down to masses as small as M=0.5 \\Msun for Y=0.24 and M=0.491 \\Msun\\ for Y=0.28. These additional (bluer) models below 0.6 \\Msun\\ are necessary for the simulations presented in Sect. 2.1. } \\label{f2} \\end{figure}  In order to evaluate the impact of a possible population having a different chemical composition on the color magnitude diagram of a globular cluster, in this article we compare isochrones having primordial helium abundance \\footnote{We take Y=0.24 as standard value for the metallicity Z=$10^{-3}$. Implicitly, we are assuming that the Big Bang value was Y=0.23, and that there is a small increase in Y for the matter reaching a metallicity Z=$10^{-3}$. Alternatively, we may regard Y=0.23 and Y=0.24 both as values indicative of the helium abundance emerging from the Big Bang. The value Y=0.228$\\pm$0.005 was derived by Pagel et al. (1992), whereas today a larger value, Y=0.244$\\pm$0.002 results from Izotov and Thuan (1998) analysis of the helium emission lines in HII regions of metal poor dwarf galaxies.}, which we set at Y=0.23 for Z=2$\\times 10^{-4}$, and at Y=0.24 for Z=$10^{-3}$, with models having a main sequence enhanced helium Y=0.28. We compute models for all evolutionary phases and compare the location of the main sequence (MS), turn--off (TO), red giants (RGB) and horizontal branch (HB). The lower main sequence models, and the detailed evolution along the RGB are considered in a forthcoming paper (Montalb\\'an et al. 2002). In Figs. 1 and 2, the low mass cutoff of each main sequence is 0.6 \\Msun. Notice that the Y=0.28, 0.6 \\Msun\\ models are much more luminous (by more than 0.4 mag) than the standard helium models. Nevertheless, the relative locations of the standard and enhanced helium isochrones show that it is not obvious how to discriminate between a population with uniform helium content and one with a spread in helium, for what concerns TO and RGB evolution. On the other hand, {\\it the morphology of the HB is affected by a helium variation, directly through the difference in the TO mass for a given age} and we argue that a helium variation may be the main reason for the occurrence of extreme blue tails in clusters such as NGC 6752 and M13.  This conclusion differs from the common hypothesis put forward in recent years: a helium enrichment in the envelope as a consequence of deep mixing in today evolving red giants has in fact been suggested as a way to obtain blue and very blue HBs (Langer \\& Hoffman 1995, Sweigart 1997). As a matter of fact, an increased helium abundance can help in reaching a HB position bluer than the RR Lyrae variables (at the expense of a more or less substantial increase in luminosity), but the extremely blue locations {\\it always} require  an extreme mass loss, because the envelope mass has to be in any case {\\it smaller} than 0.01 \\Msun\\ (see the discussion in Caloi 2001). In addition, Weiss et al. (2000) have shown that the helium enhancement needed to explain the exclusively blue HB morphology would produce related chemical inhomogeneities in red giants much stronger than observed. So an envelope helium content larger than the cosmological one does not help in obtaning the structures which make up the long blue tails in, e.g., M13 and NGC 6752. We show that the situation is different if helium is larger in the whole stellar structure, that is, if cluster stars form out of material with enhanced helium. We considered this hypothesis at the light of the present state of the art in stellar modelling. ",
        "conclusions": "The problem of chemical inhomogeneities in GCs is part of the wider question of their chemical history, which appears more and more complex, and differing from cluster to cluster. As put in evidence, e.g., by Sneden (2000), the neutron--capture elements Ba and Eu have variable abundances without obvious connection to overall cluster metallicity. Besides, variations in Si content are also observed from cluster to cluster.  The fact that field stars do not show evidence of the substantial abundance re--shuffling between CNO elements and Na, Mg and Al strongly suggests some intra cluster processes at the origin of the phenomenon. Here we examine one possible aspect of these (admittedly) rather vague processes.  We discussed the evolutionary properties of GC stars in which variations of the primordial helium content are allowed. These variations are attributed to self--pollution by an initial population of intermediate mass stars, and are expected on the basis of the same stellar models which are thought to produce the CNO variations among members of several GCs.  The observation of chemical peculiarities (dichotomy in the strength of CN lines, Na and Al enhancement) in some GC giants suggested already a long time ago (Cottrell and Da Costa 1981, Norris et al. 1981) that  intermediate mass stars may have played a role in creating some inhomogeneity in the primordial cloud of certain GCs. In particular, Norris et al. (1981) put in relation chemical peculiarities with blue tails (NGC 6752), and suggested precisely the mechanism which is at the basis of the present investigation.  We find that a spread in the helium content does not affect the morphology of the MS, TO and RGB  in an easily observable way. The simulations in Figs. \\ref{f5} -- \\ref{f8} show that also the HB luminosity level is affected in a limited way, since only an RR Lyrae region populated exclusively by structures with the maximum Y allowed (0.28) would stand out clearly as peculiarly luminous. On the other hand, such an occurrence should not be verified with standard values for the mean mass loss ($\\Delta$M $\\simgt 0.2$ \\Msun, see Lee et al. 1994).  So age determination should not be affected in a substantial way. There may be a small effect if we use as distance indicator the subdwarf main sequence, since it is reasonable to assume that subdwarfs have a helium content not affected by self--pollution.  On the other hand, the helium spread may constitute part of the ''second parameter\" problem. The presence of very blue extensions in many clusters with predominantly blue HB, and sometimes also in HBs with RR Lyrae variables and red members --NGC 2808 (Walker 1999), M62 (Caloi et al. 1987, Brocato et al. 1996)-- may be related to the second parameter problem, but this side of the question will not be considered here. The extremely small envelope masses necessary to populate these blue tails require mass losses, and/or ages, much larger than the average values required to explain also exclusively, but not extended, blue HBs, in case of uniform cosmological helium abundance. As we saw before, the smaller TO masses (for a given age) of a helium enriched structure are crucial to obtain the tiny envelopes of very blue HB stars.  The difficulties with the present hypothesis are mainly due to the enrichment scenario, since it is not easy to pollute a large fraction of original intracluster matter in a substantial way (Cannon et al. 1998). We note that the requirement of the pollution of a large fraction of the cluster population comes from the roughly bimodal distribution of CN abundances and from the well developed blue tails in many clusters, if they are related to the helium enhancement phenomenon (M13, NGC 6752).  In the self-enrichment scenarios, if the extreme blue tails can indeed be attributed to the difference among the total masses of the stars evolving in the RGB, due to their different helium content, the helium content must be larger also in the center of the stars, to affect their hydrogen burning lifetimes, and therefore we should conclude that the self--pollution has occurred either on the gas from which the helium rich stellar generation was formed, or it has occurred during the phases of the stars evolution in which they were fully convective. It is also possible that the ejecta of AGBs, collected at the center of the cluster, {\\it directly form other stellar generations.} If this is the case, some gaps in the HB stellar distributions could reflect discontinuities in the helium content of the constituent matter, due to the difference in the helium abundance in the ejecta of AGBs of different mass (Ventura et al. 2002) and to some peculiar modalities of star formation from these ejecta."
    },
    "0209/gr-qc0209006_arXiv.txt": {
        "abstract": "s#1#2#3{{ \\centering{\\begin{minipage}{4.5in}\\footnotesize\\baselineskip=10pt \\parindent=0pt #1\\par \\parindent=15pt #2\\par \\parindent=15pt #3 \\end{minipage}}\\par}}  \\def\\keywords#1{{ \\centering{\\begin{minipage}{4.5in}\\footnotesize\\baselineskip=10pt {\\footnotesize\\it Keywords}\\/: #1 \\end{minipage}}\\par}}  \\newcommand{\\bibit}{\\nineit} \\newcommand{\\bibbf}{\\ninebf} \\renewenvironment{thebibliography}[1] {\\frenchspacing \\ninerm\\baselineskip=11pt \\begin{list}{\\arabic{enumi}.} {\\usecounter{enumi}\\setlength{\\parsep}{0pt} \\setlength{\\leftmargin 12.7pt}{\\rightmargin 0pt} % \\setlength{\\itemsep}{0pt} \\settowidth {\\labelwidth}{#1.}\\sloppy}}{\\end{list}}  \\newcounter{itemlistc} \\newcounter{romanlistc} \\newcounter{alphlistc} \\newcounter{arabiclistc} \\newenvironment{itemlist} {\\setcounter{itemlistc}{0} \\begin{list}{$\\bullet$} {\\usecounter{itemlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{romanlist} {\\setcounter{romanlistc}{0} \\begin{list}{$($\\roman{romanlistc}$)$} {\\usecounter{romanlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{alphlist} {\\setcounter{alphlistc}{0} \\begin{list}{$($\\alph{alphlistc}$)$} {\\usecounter{alphlistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newenvironment{arabiclist} {\\setcounter{arabiclistc}{0} \\begin{list}{\\arabic{arabiclistc}} {\\usecounter{arabiclistc} \\setlength{\\parsep}{0pt} \\setlength{\\itemsep}{0pt}}}{\\end{list}}  \\newcommand{\\fcaption}[1]{ \\refstepcounter{figure} \\setbox\\@tempboxa = \\hbox{\\footnotesize Fig.~\\thefigure. #1} \\ifdim \\wd\\@tempboxa > 5in {\\begin{center} \\parbox{5in}{\\footnotesize\\smalllineskip Fig.~\\thefigure. #1} \\end{center}} \\else {\\begin{center} {\\footnotesize Fig.~\\thefigure. #1} \\end{center}} \\fi}  \\newcommand{\\tcaption}[1]{ \\refstepcounter{table} \\setbox\\@tempboxa = \\hbox{\\footnotesize Table~\\thetable. #1} \\ifdim \\wd\\@tempboxa > 5in {\\begin{center} \\parbox{5in}{\\footnotesize\\smalllineskip Table~\\thetable. #1} \\end{center}} \\else {\\begin{center} {\\footnotesize Table~\\thetable. #1} \\end{center}} \\fi}  \\def\\@citex[#1]#2{\\if@filesw\\immediate\\write\\@auxout {\\string\\citation{#2}}\\fi \\def\\@citea{}\\@cite{\\@for\\@citeb:=#2\\do {\\@citea\\def\\@citea{,}\\@ifundefined {b@\\@citeb}{{\\bf ?}\\@warning {Citation `\\@citeb' on page \\thepage \\space undefined}} {\\csname b@\\@citeb\\endcsname}}}{#1}}  \\newif\\if@cghi \\def\\cite{\\@cghitrue\\@ifnextchar [{\\@tempswatrue \\@citex}{\\@tempswafalse\\@citex[]}} \\def\\citelow{\\@cghifalse\\@ifnextchar [{\\@tempswatrue \\@citex}{\\@tempswafalse\\@citex[]}} \\def\\@cite#1#2{{$\\null^{#1}$\\if@tempswa\\typeout {IJCGA warning: optional citation argument ignored: `#2'} \\fi}} \\newcommand{\\citeup}{\\cite}  \\def\\pmb#1{\\setbox0=\\hbox{#1} \\kern-.025em\\copy0\\kern-\\wd0 \\kern.05em\\copy0\\kern-\\wd0 \\kern-.025em\\raise.0433em\\box0} \\def\\mbi#1{{\\pmb{\\mbox{\\scriptsize ${#1}$}}}} \\def\\mbr#1{{\\pmb{\\mbox{\\scriptsize{#1}}}}}  \\def\\fnm#1{$^{\\mbox{\\scriptsize #1}}$} \\def\\fnt#1#2{\\footnotetext{\\kern-.3em {$^{\\mbox{\\scriptsize #1}}$}{#2}}}  \\def\\fpage#1{\\begingroup \\voffset=.3in \\thispagestyle{empty}\\begin{table}[b]\\centerline{\\footnotesize #1} \\end{table}\\endgroup}   \\def\\thefootnote{\\fnsymbol{footnote}} \\def\\@makefnmark{\\hbox to 0pt{$^{\\@thefnmark}$\\hss}}\t%  \\def\\ps@myheadings{% \\let\\@oddfoot\\@empty\\let\\@evenfoot\\@empty \\def\\@evenhead{\\slshape\\leftmark\\hfil}% \\def\\@oddhead{\\hfil{\\slshape\\rightmark}}% \\let\\@mkboth\\@gobbletwo \\lets{Theoretical considerations of fundamental physics, as well as certain cosmological observations, persistently point out to permissibility, and maybe necessity, of macroscopic modifications of the Einstein general relativity. The field-theoretical formulation of general relativity helped us to identify the phenomenological seeds of such modifications. They take place in the form of very specific mass-terms, which appear in addition to the field-theoretical analog of the usual Hilbert-Einstein Lagrangian. We derive and study exact non-linear equations of the theory, along with its linear approximation. We interpret the added terms as masses of $spin-2$ and $spin-0$ gravitons. The arising finite-range gravity is a fully consistent theory, which smoothly approaches general relativity in the massless limit, that is, when both masses tend to zero and the range of gravity tends to infinity. We show that all local weak-field predictions of the theory are in perfect agreement with the available experimental data. However, some other conclusions of the non-linear massive theory are in a striking contrast with those of general relativity. We show in detail how the arbitrarily small mass-terms eliminate the black hole event horizon and replace a permanent power-law expansion of a homogeneous isotropic universe with an oscillatory behaviour. One variant of the theory allows the cosmological scale factor to exhibit an `accelerated expansion' instead of slowing down to a regular maximum of expansion. We show in detail why the traditional, Fierz-Pauli, massive gravity is in conflict not only with the static-field experiments, but also with the available indirect gravitational-wave observations. At the same time, we demonstrate the incorrectness of the widely held belief that the non-Fierz-Pauli theories possess `negative energies' and `instabilities'.}{}{}   \\vspace*{1pt}\\textlineskip ",
        "introduction": "Presently, there seems to be no pressing need in devising theories of gravitation, alternative to the existing Einstein's general relativity. General relativity (GR) is an internally consistent theory, and it has passed all the performed experimental tests with flying colors \\cite{Willbook}, \\cite{Fock}, \\cite{LL}, \\cite{MTW}, \\cite{W}. And yet, there are some clouds on the horizon. On the theoretical side, the M/string theory considerations persistently point out to possible macroscopic modifications of GR, particularly in the form of various ``mass-terms\". On the observational side, there exists some discomfort in understanding the large-scale structure and evolution of the Universe, including some indications to the possibility of its present ``accelerated expansion\". So far, theorists enjoy playing with the cosmological $\\Lambda$-term and various highly speculative forms of matter, but the credibility of these models can soon be exhausted. The old question arises again, whether there do exist well-motivated consistent alternative theories of macroscopic gravity, with non-trivial observational consequences.  At the first sight, general relativity is an isolated theory with no immediate neighbours. In particular, it seems that GR cannot be modified without raising the order of differential field equations. Indeed, in the geometrical formulation of GR, which operates with the curved space-time metric tensor $g_{\\mu \\nu}$, there is no structure that can be added to the usual Hilbert-Einstein Lagrangian. The only possibility is the $\\Lambda$-term: ${\\sqrt {-g}} \\Lambda$, but this structure can be included in the definition of GR, and we know all the phenomenological consequences of the $\\Lambda$-term, and in any case the $\\Lambda$-term is not a ``mass-term\". The situation changes drastically when one looks at GR from the field-theoretical perspective. The old remark of Feynman \\cite{F} on the intrinsic value of equivalent formulations of a fundamental theory proves to be very profound. One gets the possibility to analyse the problems which otherwise could not be even properly formulated. We believe that general relativity does indeed contain the seeds of its own modification, and the field-theoretical formulation of GR helped us to identify these seeds. The modification of GR, which looks almost unavoidable from the viewpoint of the field-theoretical approach, leads to the appearance of very specific mass-terms. The resulting theory is a fully consistent finite-range gravitational theory. General relativity is a smooth limit of this theory when the range of gravity tends to infinity. The theory is in perfect agreement with all local weak-field experiments, such as experiments in the Solar system, and satisfies the requirement, formulated long ago \\cite{BD}, of ``physical continuity\". However, some other consequences of the theory are truly striking. It is surprising to see that some of the crucial conclusions of GR are so much vulnerable to pretty innocent modifications of GR. For instance, the existence of a black hole event horizon, and a permanent power-law expansion of the matter-dominated Universe, get invalidated by the arbitrarily small mass-terms. We introduce and explain this finite-range gravitational theory in the present paper.  The fundamental quantity in the field-theoretical GR is a symmetric second-rank tensor field $h^{\\mu \\nu}(x^{\\alpha})$. The gravitational field $h^{\\mu \\nu}(x^{\\alpha})$ is defined in a flat space-time with the line-element \\be {\\rm d} \\sigma^2 = \\gamma_{\\mu \\nu} {\\rm d}x^{\\mu} {\\rm d} x^{\\nu}. \\label{sigma} \\en The curvature tensor constructed from $\\gamma_{\\mu \\nu}(x^{\\alpha})$ is identically zero: \\be \\label{R} \\breve R_{\\alpha\\beta\\mu\\nu}(\\gamma_{\\rho\\sigma}) = 0. \\en In flat space-time, one is always free to choose Lorentzian coordinates, in which case Eq. (\\ref{sigma}) takes on the Minkowski form \\be \\label{Mi} {\\rm d} \\sigma^2 = \\eta_{\\mu \\nu} {\\rm d}x^{\\mu} {\\rm d} x^{\\nu} = c^2{\\rm d}t^2 - {\\rm d}x^2 - {\\rm d}y^2 - {\\rm d}z^2. \\en The flat space-time is not a choice of some artificial ``prior geometry\", but is a reflection of experimental facts. As far as the present-day physics knows, the intervals of space and durations of time, in absence of all fields including gravity, satisfy the relationships of the Minkowski 4-dimensional interval (\\ref{Mi}). If there existed any observational evidence to something different, we would have started from a different metric.  The Lagrangian of the field-theoretical GR depends on the gravitational field variables $h^{\\mu \\nu}(x^{\\alpha})$ and their first derivatives. (We present more details in Sec.~2) The variational principle gives rise to the dynamical field equations, which are fully equivalent to the Einstein equations. The transition to the geometrical formulation of GR proceeds through the introduction of the tensor $g^{\\mu \\nu}(x^{\\alpha})$ and the inverse tensor $g_{\\mu \\nu}(x^{\\alpha})$: $g^{\\mu \\rho}g_{\\nu \\rho} = \\delta^{\\mu}_{\\nu}$. The quantities $g^{\\mu \\nu}$ are calculable from the gravitational field variables $h^{\\mu \\nu}$ and the metric tensor $\\gamma^{\\mu \\nu}$ according to the rule \\bea \\label{g} \\sqrt{-g}g^{\\mu\\nu}= \\sqrt{-\\gamma}(\\gamma^{\\mu\\nu} + h^{\\mu\\nu})~, \\ena where $g=det|g_{\\mu\\nu}|$, $\\gamma=det|\\gamma_{\\mu\\nu}|$, and $\\gamma^{\\mu \\rho}\\gamma_{\\nu \\rho}=\\delta^{\\mu}_{\\nu}$. The tensor density $\\sqrt{-g}g^{\\mu\\nu}$ participates in the matter Lagrangian, realizing the universal coupling of gravity to all other fields, but apart of that, it is simply a short-hand notation for the quantity in the right-hand-side (r.h.s.) of Eq.~(\\ref{g}). In the geometrical formulation of GR, tensor $g_{\\mu \\nu}(x^{\\alpha})$ is interpreted as the metric tensor of a curved space-time: \\be \\label{ds} {\\rm d}s^2 = g_{\\mu \\nu} {\\rm d}x^{\\mu} {\\rm d} x^{\\nu}. \\en In terms of $g_{\\mu \\nu}$, the field equations acquire the familiar form of the geometrical Einstein's equations. From the viewpoint of the field-theoretical formulation, the tensor $g_{\\mu \\nu}(x^{\\alpha})$ is the effective metric tensor; it defines the intervals of space and time measured in the presence of the universal gravitational field $h^{\\mu \\nu}(x^{\\alpha})$. The field-theoretical approach to GR has a long and fruitful history. For a sample of references, see \\cite {Pap}, \\cite{Gupta}, \\cite{Kr}, \\cite{Thirring}, \\cite{Ros}, \\cite{weinb}, \\cite{Deser}, \\cite{Feyn}, \\cite{GPP}, \\cite {GZ}, including a history review \\cite{lastreview}, and many papers cited therein.  It was shown \\cite{BG} that the gravitational Lagrangian of the field-theoretical GR must include, in addition to the field-theoretical analog of the Hilbert-Einstein term, the extra term \\be \\label{ad} \\sqrt{-\\gamma}\\left[-\\frac{1}{4}\\breve R_{\\alpha\\rho\\beta\\sigma} (h^{\\alpha\\beta}h^{\\rho\\sigma} - h^{\\alpha\\sigma} h^{\\rho\\beta})\\right]. \\en This term does not affect the field equations, but is needed for the variational derivation of the gravitational energy-momentum tensor $t^{\\mu \\nu}$. The variational (metrical) energy-momentum tensor is the response of a physical system to variations of the metric tensor $\\gamma_{\\mu \\nu}$, caused by arbitrary coordinate transformations. Obviously, such variations of the metric tensor should obey the constraint (\\ref{R}). The variational procedure incorporates the constraint (\\ref{R}) by adding to the Lagrangian an extra term: $\\Lambda^{\\alpha \\beta\\rho\\sigma} \\breve R_{\\alpha\\rho\\beta\\sigma}$, where $\\Lambda^{\\alpha \\beta\\rho\\sigma}$ are undetermined Lagrange multipliers. The constraint (\\ref{R}) has to be enforced at the end of the variational derivation of the field equations and the energy-momentum tensor. It has been proven \\cite{BG} that the Lagrange multipliers must have the unique form \\[ \\Lambda^{\\mu \\nu \\alpha \\beta} = - \\frac{1}{4}(h^{\\alpha\\beta} h^{\\mu\\nu} - h^{\\alpha \\nu} h^{\\beta\\mu}), \\] in order for the derived energy-momentum tensor $t^{\\mu \\nu}$ to satisfy all the necessary mathematical and physical requirements, including the absence of second derivatives of the field variables in the $t^{\\mu \\nu}$.  As was explained above, the quantity $\\breve R_{\\alpha\\rho\\beta\\sigma}$ in Eq.~(\\ref{ad}) is the curvature tensor of a flat space-time. If it were something other than that, the theory would not be GR. However, it is natural to assume that the Lagrangian may also include an additional term similar to (\\ref{ad}), but where the quantity $\\breve R_{\\alpha\\rho\\beta\\sigma}$ is the curvature tensor of an abstract space-time with a constant non-zero curvature. Space-times of constant curvature are as symmetric as flat space-time, but contain a parameter $K$ with dimensionality of $[length]^{-2}$: \\be \\label{cc} \\breve R_{\\alpha\\beta\\mu\\nu}=K(\\gamma_{\\alpha\\mu}\\gamma_{\\beta\\nu}- \\gamma_{\\alpha\\nu}\\gamma_{\\beta\\mu}). \\en If one uses (\\ref{cc}) in (\\ref{ad}), the generated additional term in the Lagrangian is \\be \\label{fp} \\sqrt{-\\gamma}\\frac{K}{2}(h^{\\alpha \\beta}h_{\\alpha\\beta} - h^2). \\en Clearly, the new theory is not GR, but what this theory is ? Quite surprisingly, one recognizes in (\\ref{fp}) the Fierz-Pauli \\cite{FP} mass-term. Having discovered that the structure (\\ref{ad}) generates mass-terms, we have asked about the most general form of such terms. It is easy to show that there exist only two independent quadratic combinations: $h^{\\alpha \\beta}h_{\\alpha\\beta}$ and $h^2$. Therefore, we arrive at a 2-parameter family of theories with the additional mass-terms in the gravitational Lagrangian: \\be \\label{two} \\sqrt{-\\gamma}\\left[ k_1h^{\\rho\\sigma}h_{\\rho\\sigma} + k_2 h^2\\right], \\en where $k_1$ and $k_2$ have dimensionality of $[length]^{-2}$. Fierz and Pauli, as well as many other authors after them, were considering the (internally contradictory) ``linear gravity\", whereas in our case the tensor $h^{\\mu\\nu}$ is the full-fledged non-linear gravitational field. The 2-parameter class of theories with the additional mass-terms (\\ref{two}) is what we shall study in the present paper. We consider the mass-terms as phenomenological, even though their deep origin can be quantum-mechanical or multi-dimensional.  The structure of the paper and its conclusions are as follows.  In Sec.2 we derive exact non-linear equations, as well as gravitational energy-momentum tensor, for the gravitational field in absence of any matter sources. Since almost all calculations in gravitational physics are performed in geometrical language, and we will need some of the results, we introduce the notion of a quasi-geometrical description of the finite-range gravity.\\footnote{Geometry in physics, like communism in politics, is not very dangerous, if introduced in well-measured doses.} Specifically, we retain the usual presentation of the Einstein part of the equations in terms of $g_{\\mu \\nu}$, but in the massive part, which originates from (\\ref{two}) and cannot be written in terms of $g_{\\mu \\nu}$ only, we trade $h_{\\mu \\nu}$ for $g_{\\mu \\nu}$ and $\\gamma_{\\mu \\nu}$, according to the rule (\\ref{g}). The important point is the symmetries of the theory. Equations of the field-theoretical GR enjoy two different symmetries. The first one (general covariance, or diffeomorphism) is the freedom to use arbitrary coordinates and the associated transformations of, both, the metric tensor $\\gamma^{\\mu \\nu}$ and the field tensor $h^{\\mu \\nu}$. The second symmetry is the freedom to use the (true) gauge transformations, which do not touch coordinates and the metric tensor, but transform the field variables only \\cite{GPP}. It is this second symmetry that gets violated by the mass-terms, while the first symmetry survives.  In Sec.3 we formulate exact equations for the gravitational field in the presence of matter sources. Again, we are often using the quasi-geometrical description. This means, in particular, that in the matter part of the field equations we retain the geometrical energy-momentum tensor $T_{\\mu \\nu}$, i.e. the matter energy-momentum tensor defined as the variational derivative of the matter Lagrangian with respect to $g^{\\mu \\nu}$, as opposed to the field-theoretical energy-momentum tensor $\\tau_{\\mu\\nu}$, defined as the variational derivative of the matter Lagrangian with respect to $\\gamma^{\\mu\\nu}$. The content of Sec.3 will be needed in Section 7 and, partially, in Section 5.  In Sec.4 we discuss the linearised approximation of the theory and give physical interpretation to the parameters $k_1$ and $k_2$. In accord with the analysis of Ogievetsky and Polubarinov \\cite{OP}, and Van Dam and Veltman \\cite{VDV}, these parameters give rise to the two fundamental masses: the mass $m_2$ of the $spin-2$ graviton, and the mass $m_0$ of the $spin-0$ graviton. Strictly speaking, the corresponding wave-equations contain two fundamental lengths, rather than two fundamental masses. Concretely, the equations contain two parameters, $\\alpha^2$ and $\\beta^2$, with dimensionalities of $[length]^{-2}$: \\be \\label{ab} \\alpha^2 = 4k_1, ~~~~~ \\beta^2 = -2k_1 \\frac{k_1+4k_2}{k_1+k_2}, \\en but $\\alpha$ and $\\beta$ can be thought of as inverse Compton wavelengths of the two gravitons with the masses \\be \\label{mm} m_2 = \\frac{\\alpha \\hbar}{c}, ~~~~~m_0 = \\frac{\\beta \\hbar}{c}. \\en The interpretation of the free parameters in terms of masses implies that $\\alpha^2$ and $\\beta^2$ are strictly positive quantities. However, the Lagrangian itself does not require this restriction, and we will exploit this freedom in the cosmological Section 7.  One very special choice of the parameters $k_1$ and $k_2$ is $k_2 = -k_1$. This choice of the parameters brings the Lagrangian (\\ref{two}) to the Fierz-Pauli form (\\ref{fp}). It is this case that has led to a lively debate on the unacceptability of a ``massive graviton\". Although the Lagrangian (\\ref{fp}) itself does smoothly vanish in the limit $k_1 \\rightarrow 0$, the corresponding solutions and local weak-field physical predictions (for instance, the deflection angle of light propagating in the gravitational field of the Sun) do not approach those of GR. In other words, this particular massive theory disagrees with the original massless theory even in the limit of vanishingly small mass $m_2$ and, hence, in the limit of arbitrarily long Compton wavelength $1/\\alpha$. The finite, and independent of the mass $m_2$, difference in local predictions became known as the Van Dam-Veltman-Zakharov discontinuity \\cite{VDV}, \\cite{Zakh}, \\cite {FadSl}, \\cite{Iwas}, \\cite{Vain}, \\cite{Visser}, \\cite{KogMP}, \\cite{Por}, \\cite{DDGV}, \\cite{CarG}, \\cite{DesT}, \\cite{Gruz}. This puzzling conclusion about discontinuity is described \\cite{Velt} as something that seems counter-intuitive to certain physicists. We have to confess that the usual presentation of this conclusion seems counter-intuitive to us, as well. We believe that the issue should be looked upon from a different angle. When taking the massless limit of a massive theory, one should do what the logic requires to do, namely, to send both masses to zero. Then, both Compton wavelengths tend to infinity, and one recovers, as expected, the local weak-field predictions of GR. If, instead, one takes $k_2 = -k_1$ (whatever the motivations behind this choice might be), the mass $m_0$ becomes infinitely large (see Eqs. (\\ref{ab}), (\\ref{mm})) and the corresponding Compton wavelength $1/\\beta$ is being sent to zero. Any local experiment is now supposed to be performed at scales much larger than one of the characteristic lengths, $1/\\beta$. In this situation, the deviations from GR should be expected on the grounds of physical intuition. There is no wonder that the subsequent limit $1/\\alpha \\rightarrow \\infty$ does not cure these deviations. This situation may look like a counter-intuitive discontinuity.  To explore the difference in local predictions, there is no need to propagate light in the Solar system. It is sufficient to consider the geodesic deviation equation for free test bodies separated by small distances. We do this study below in the paper. In particular, the geodesic deviation equation illustrates the difference between GR and finite-range gravity in the domain of gravitational-wave predictions.  In Sec.5 and Appendices B, C, we study weak gravitational waves. Certain modifications of GR are well anticipated. In the field-theoretical GR, the $spin-0$ gravitational waves (represented by the trace $h=h^{\\mu \\nu} \\eta_{\\mu \\nu}$) exist as gauge solutions. They contribute neither to the gravitational energy-momentum tensor $t^{\\mu \\nu}$, nor to the deformation pattern of a ring of test particles in the geodesic deviation equation. The same is true for the $helicity-0$ polarization state (represented by the spatial trace $h^{ij} \\eta_{ij}$) of the $spin-2$ graviton. In the finite-range gravity, as one could expect, both these degrees of freedom become essential. They provide additional contributions to the energy-momentum flux carried by the gravitational wave, and the extra components of motion of the test particles. However, gravitational wave solutions, their energy-momentum characteristics, and observational predictions of GR are fully recovered in the massless limit $\\alpha \\rightarrow 0$, $\\beta \\rightarrow 0$ of the theory. We show that the Fierz-Pauli case is very peculiar and unacceptable. Even in the limit of $\\alpha \\rightarrow 0$, there remains a nonvanishing ``common mode\" motion of test particles in the plane of the wave front. The extra component of motion is accounted for by the corresponding additional flux of energy from the source; typically, of the same order of magnitude as the GR flux. This analysis, together with the Solar system arguments, leads to the important conclusion. Whatever the sophisticated ``brane world\" motivations of the M/string theory may be, if they lead to the phenomenological mass-term of the Fierz-Pauli type, the corresponding variant of the theory should be rejected as being in conflict with the static-field experiments and with the already available indirect gravitational-wave observations of binary pulsars. We do not think that this conclusion can be invalidated by any ``non-perturbative effects\". At the same time, by doing concrete calculations, we dispel the deeply-rooted myth that the non-Fierz-Pauli theories should suffer from ``negative energies\" and ``instabilities\".  The fully non-linear finite-range gravity is considered in the next two Sections. In Sec.6 we analyse static spherically-symmetric solutions. We summarise the weak-field approximation in Appendix~\\ref{bhb}. There, we demonstrate that the GR solutions and physical predictions are recovered in the massless limit $\\alpha \\rightarrow 0$, $\\beta \\rightarrow 0$, and, with the help of the geodesic deviation equation, we confirm the observational unacceptability of the Fierz-Pauli coupling. The main thrust of Sec.6 is the non-linear (would-be black hole) solutions. The case of arbitrary relationship between  $\\alpha$ and $\\beta$ is difficult to analyse in full generality. The equations are somewhat simpler when the masses are assumed to be equal, i.e. $\\alpha = \\beta$. We call this choice the Ogievetsky-Polubarinov (OP) case. We analyse this case in great detail, and present more general considerations whenever possible, demonstrating that the qualitative conclusions remain valid for $\\alpha \\neq \\beta$. A single dimensionless parameter in the OP case is $\\alpha M$, where $M$ is the Schwarzschild mass (using $G =1$ and $c=1$), which is supposed to be a very small number.  We start with intermediate scales, that is, with Schwarzschild distances $R$ which are much larger than $2M$, but much smaller than $1/\\alpha$. Combining analytical and numerical techniques, we demonstrate that the solution of the massive theory is practically indistinguishable from that of GR for all $R$ sufficiently larger than $2M$, but, obviously, smaller than $1/\\alpha$. As expected, for $R$ larger than $1/\\alpha$, the solution takes on the form of the Yukawa-type potentials; this is why the theory is called finite-range gravity. However, the massive solution strongly deviates from that of GR not only at very large distances, but also in the vicinity of $R=2M$. This is a consequence of the non-linear character of the field equations. The hypersurface $R=2M$ is the location of the (globally defined) event horizon of the Schwarzschild black hole in GR. We carefully explore the vicinity of $R=2M$, as well as $0 \\le R < 2M$, the region that would have been the interior of the Schwarzschild black hole. We show that the smaller the parameter $\\alpha M$, the closer to $R=2M$ one can descend (from large $R$) along the essentially Schwarzschild solution. We show that the deviations from GR near $R=2M$ are so radical that the event horizon does not form, and the solution smoothly continues to the region $R < 2M$. The further continuation of the solution terminates at $R=0$, where the curvature singularity develops. Since the $\\alpha M$ can be extremely small, the redshift of the photon emitted at $R=2M$ can be extremely large, but it remains finite. In contrast to GR, the infinite redshift is reached at the singularity $R=0$, and not at $R=2M$. The conclusion of this study is quite dramatic. In the astrophysical sense, the resulting solution still looks like a black hole; in the region of space just outside the $R=2M$, the gravitational field is practically indistinguishable from the Schwarzschild solution. However, all conclusions that rely specifically on the existence of the black hole event horizon, are likely to be abandoned. It is very remarkable and surprising that the phenomenon of black hole should be so unstable with respect to the inclusion of the tiny mass-terms (\\ref{two}), whose Compton wavelengths can exceed, say, the present-day Hubble radius.  Section 7 is devoted to cosmological solutions for a homogeneous isotropic universe. Matter sources are taken in the simplest form of perfect fluids with fixed equations of state. First, we show that if the mass of the $spin-0$ graviton is zero, i.e. $\\beta^2 = 0$, the cosmological solutions of the massive theory are exactly the same as those of GR, independently of the mass of the $spin-2$ graviton, that is, independently of the value of $\\alpha^2$. This result could be expected due to the highest spatial symmetry of the problem under consideration; the $spin-2$ degrees of freedom have no chance to reveal themselves. Then, we proceed to cases with $\\beta^2 \\neq 0$. Since we prefer to deal with technically simple equations, we consider a particular case $4\\beta^2 = \\alpha^2$. This case is studied in full details, but we also show that the qualitative results are general and are valid for $4\\beta^2 \\neq \\alpha^2$. Combining analytical approximations and numerical calculations, we demonstrate that the massive solution has a long interval of evolution where it is practically indistinguishable from the Friedmann solution of GR. However, the deviations from GR are dramatic at very early times and very late times. The unlimited expansion is being replaced by a regular maximum of the scale factor, whereas the singularity is being replaced by a regular minimum of the scale factor. The smaller $\\beta$, the higher maximum and the deeper minimum. In other words, astonishingly, the arbitrarily small mass-terms (\\ref{two}) give rise to the oscillatory behaviour of the cosmological scale factor.  Following the logic of interpretation of the theory in terms of masses, we assume in the most of the paper that the signs of $\\alpha^2$ and $\\beta^2$ are positive. However, as mentioned above, the general structure of the Lagrangian (\\ref{ad}) does not imply this. It is interesting to observe that if we allow $\\alpha^2$ and $\\beta^2$ to be negative (which would probably require to think of the massive gravitons in terms of ``tachyons\"), the late time evolution of the scale factor exhibits an ``accelerated expansion\", instead of slowing down towards the maximum. This behaviour of the scale factor is similar to the one governed by a positive cosmological $\\Lambda$-term. The physical significance of this result is presently unclear, but the problem deserves further study. In any case, cosmological modifications proposed here are justified better, than in many inconsistent ``ad hoc\" models that appeared in the literature.  We briefly summarise our results in the concluding Sec.8 and relegate some technical details of the paper to Appendices A, B, C, D. ",
        "conclusions": "\\label{conc}  The internal logic of the field-theoretical formulation of the Einstein's general relativity suggests certain modifications of GR, which take place in the form of very specific mass-terms. These terms appear in addition to the field-theoretical analog of the usual Hilbert-Einstein Lagrangian. The arising finite-range gravity is a theory fully acceptable, both, from mathematical and physical points of view. Probably, the resulting theory can also be viewed as a phenomenological realisation of some macroscopic modifications of the 4-dimensional gravity, suggested by the M/string theory (see, for example, \\cite{DDGV}, \\cite{GRS}, \\cite{KMP}, \\cite{BKK}).  We have derived and studied the exact non-linear equations of the theory, along with its linear approximation. The added terms have been interpreted as masses of $spin-2$ and $spin-0$ gravitons. We have shown that the local weak-field predictions of GR are fully recovered in the massless limit, that is, when both masses are sent to zero. At the same time, the traditional (Fierz-Pauli) way of including the mass-terms was shown to be very peculiar and unacceptable. It corresponds to sending the mass of the $spin-0$ graviton to infinity. As a result, the Fierz-Pauli theory contradicts the performed static-field experiments, as well as the indirect gravitational-wave observations. The contradiction stays even in the limit when the mass of the $spin-2$ graviton tends to zero. This fact rules out those variants of the candidate fundamental theories which suggest the macroscopic modifications of GR in the Fierz-Pauli form. At the same time, we have shown that the non-Fierz-Pauli theories are free from ``negative energies\", ``instabilities\", etc.  The most surprising deviations of the finite-range gravity from GR occur in strongly non-linear regime. We have considered static spherically-symmetric configurations and homogeneous isotropic cosmologies. We demonstrated that the mass-terms modify the Schwarzchild solution not only at very large distances (these are the expected Yukawa-type modifications that explain the name: finite-range gravity) but also in the vicinity of the Schwarzchild sphere $R= 2M$. The deviations near $R=2M$ are so radical that the event horizon does not form, and the massive solution smoothly continues up to $R=0$, where the curvature singularity develops. The result of this study is quite dramatic. In the astrophysical sense, the resulting massive configuration is still similar to the black hole configuration of GR. Namely, in the region of space just outside the $R=2M$, the gravitational field of the massive gravity solution is practically indistinguishable from the Schwarzschild solution. However, all conclusions of GR that rely specifically on the existence of the black hole event horizon, are likely to be abandoned. One can distinguish the two configurations observationally. For instance, the gravitational waveforms emitted by a body inspiralling toward the center of configurations are expected to be different. In the finite-range gravity, in contrast to GR, the body continues to emit observable gravitational waves even from distances $R < 2M$.  Finally, we have considered cosmological solutions for homogeneous isotropic universes. We have shown that there is a long interval of evolution where cosmological solutions of the finite-range gravity are practically indistinguishable from those of GR. However, the arbitrarily small mass-terms lead to strong deviations from GR at very early and very late times. We show in detail how the unlimited expansion is being replaced by a regular maximum of the scale factor, while the singularity is being replaced by a regular minimum of the scale factor. In other words, the arbitrary small mass-terms give rise to the oscillatory behaviour of the model universe. We show that when the gravitons are traded for the ``tachyons\", the cosmological scale factor exhibits an interval of accelerated expansion instead of slowing down toward the maximum of expansion. This may explain the ``cosmic acceleration\", if it is observationally confirmed.  We believe that the solid theoretical motivations for the finite-range gravity, as well as its highly interesting conclusions derived so far, warrant further investigations in this area of research."
    },
    "0209/astro-ph0209088_arXiv.txt": {
        "abstract": "We present near-infrared $H$-band observations of the hosts of three $z\\sim 1$ quasars from the Sloan Digital Sky Survey made with the  adaptive optics system at Lick Observatory. We derive a PSF for each quasar and model the host plus quasar nucleus to obtain magnitudes and approximate scale sizes for the host galaxies. We find our recovered host galaxies are similar to those found for $z\\sim 1$ quasars observed with the {\\em Hubble Space Telescope}. They also have, with one interesting exception, black hole mass estimates from their bulge luminosities which are consistent with those from emission-line widths. We thus demonstrate that adaptive optics can be successfully used for the quantitative study of quasar host galaxies, with the caveat that better PSF calibration will be needed for studies of the hosts of significantly brighter or higher redshift quasars with the Lick system. ",
        "introduction": "All massive elliptical galaxies, and all the bulge components of spiral galaxies, seem to contain black holes whose masses correlate well with the depth of the potential wells of the stellar systems containing them (Ferrarese \\& Merritt 2000; Gebhardt et al.\\ 2000). This implies that all such systems were probably quasars at some stage in their lifetimes, and indeed the luminosity density produced by quasars is probably consistent with this (Yu \\& Tremaine 2002). The importance of the evolution of quasars as diagnostics of galaxy formation models was recognized by Kauffmann \\& Haehnelt (2000), who produced predictions for the nature of quasar host galaxies as a function of quasar luminosity and redshift, based around semi-analytic models of galaxy formation.  Two main problems exist with current studies of quasar hosts. First, the samples are small. The hosts of only $\\sim 30$ quasars with $z\\sim 1-3$ have been observed with NICMOS on the {\\em Hubble Space Telescope} (HST) (Ridgway et al.\\ 2001; Kukula et al.\\ 2001; Rix et al.\\ 2001). Comparison with, e.g., the Kauffmann \\& Haehnelt models using samples of this size is difficult, as the models predict a wide range in host galaxy magnitudes for a given quasar luminosity and significant redshift evolution in the host magnitudes. Second, quasar samples are selected using very different techniques, with different selection biases. For example, UV excess selection may select against spiral hosts, which are dustier than ellipticals, and radio selection probably selects the biggest black holes, in giant elliptical hosts (e.g.\\ Lacy et al.\\ 2001). Studying large samples of high redshift quasar hosts with HST is impractical due to the large amount of time required. Therefore our best hope of improving our knowledge of the nature of quasar hosts at $z\\stackrel{>}{_{\\sim}} 1$ is to use adaptive optics (AO) from the ground. So far, only small samples (such as the one in this paper) have been studied with AO. The quasar survey component of the Sloan Digital Sky Survey (SDSS), however, contains a large enough number of quasars that a significant number of quasars have good AO guide stars nearby. Furthermore, because this survey selects quasars on the basis of their being unresolved optically and and having colors lying off the stellar locus (Fan 1999; Richards et al.\\ 2001), it is much more sensitive than most previous optical surveys to quasars with a modest amount of dust reddening. We thus expect that host galaxies of quasars from this survey should give a more complete picture of the quasar host galaxy population.  AO is, however, is a new technique with problems of its own. Previous attempts at studying host galaxies with AO,  although largely successful in detecting both close companions and diffuse emission around the quasars, have encountered problems with PSF characterization and stability, which has limited the amount of quantitative information obtainable from the images (Stockton, Canalizo \\& Close 1998; Hutchings et al.\\ 1999; M\\'{a}rquez et al.\\ 2001). In this paper we present a pilot study of a small sample of $z\\sim 1$ quasars from the SDSS Early Data Release (Stoughton et al.\\ 2002; hereafter EDR) observed with the Lick Adaptive Optics system in Natural Guide Star mode. We discuss issues related to PSF stability and subtraction, and present the results and a brief discussion of our observations and prospects for future quantitative studies of larger samples of quasar hosts with AO. We assume a cosmology with $H_0=65 {\\rm kms^{-1}Mpc^{-1}}$, $\\Omega_{\\rm M} = 0.3$ and $\\Omega_{\\Lambda} = 0.7$ ",
        "conclusions": "Our AO observations have proved successful at detecting hosts around three $z\\sim 1$ quasars. The accuracy of our host galaxy flux density measurements are, as expected, limited most by the accuracy of our PSFs. For our $z\\sim 1$ quasars, errors in the PSF have not been large enough to prevent us obtaining scientifically-useful results. It will, however, be necessary to make a better PSF calibration if we wish to study systems with significantly brighter quasars relative to their hosts, or higher redshift quasars where surface brightness dimming makes extended emission from the host harder to detect. This may simply involve taking more frequent PSF observations, or better matching the PSF star/PGS pair to the quasar/QGS pair, but it may turn out that better PSF calibration will require a different approach than the one described here.  Although a substantial improvement over the natural seeing at Lick was obtained through the use of adaptive optics, our resultant image quality is only comparable to the natural seeing at a first-class astronomy site such as Mauna Kea. Similar techniques could, however, be used at such sites to deliver even better quality images. One complication not addressed in this paper is that of altitude-azimuth mounted telescopes. In these, the component of the PSF due to the telescope optics rotates with the field, and thus will vary rapidly with time, though the off-axis component, determined by the atmosphere, should not be affected. How useful our technique would be for such telescopes will depend on the details of the optical system, although given that a large contribution to the PSF width comes from the off-axis component our technique may still be useful.  The magnitudes and scale sizes of the quasar hosts are comparable to the nine $z\\sim 1$ quasar hosts from the HST/NICMOS study of Kukula et al.\\ (2001) (Figure 5), although our hosts are on average a little fainter. They are therefore closer to the predictions of Kauffmann \\& Haehnelt (2000). Though our sample is too small to make a definitive statement, if this trend continues to be seen in a larger sample, it may be a result of the SDSS quasar survey having fewer selection biases than conventional optical surveys. Previous optical surveys tended to select very blue quasars and are thus sensitive to small amounts of reddening in the host. One might therefore expect them to favour quasars in less dusty hosts, such as giant ellipticals.  A comparison between the black hole mass estimates from the galaxy luminosities and the Mg{\\sc ii} linewidths shows that for SDSS~1727+5946 and SDSS~0050+0113, the agreement is very good, within a factor of two, but for SDSS~2323+0040 the black hole mass estimated from the linewidth is an order of magnitude less than that derived from the host. Dunlop et al.\\ (2002) also find that their black hole mass estimates for $z\\sim 0.2$ quasars imaged with HST usually agree well, but a few objects also have order-of-magnitude discrepancies between black hole mass estimates, in the same sense, that the emission line estimates are too low. They ascribe these to disk-like broad-line regions being seen face-on. However, it is also true that the three most discrepant black hole mass estimates in the Dunlop et al.\\ sample all show evidence for interactions, and include the two most spectacular mergers in their sample. SDSS~2323+0040 is also the only one of our hosts to show signs of interactions. Perhaps the bulge mass luminosity estimates in merger systems are too high. Possible reasons for this include a starburst in the merger system lowering the mass-to-light ratio of the stellar population, or a delay between the galaxy merger and the merger of the black holes of the two galaxies and the subsequent accretion of significant amounts of matter onto the merged black hole. Large differences between black hole mass estimates derived from emission-line widths and host bulge luminosities may thus be indicators of a quasar formed from a recent merger event. All our quasars are accreting at or below the Eddington rate using either black hole mass estimate.  The advent of large samples of quasars from the SDSS and the Anglo-Australian 2dF surveys means that significant numbers of quasars near bright AO guide stars have already become available. Even larger samples will be possible using laser guide stars. The image quality with laser guide stars will also be better as the AO corrections will be made on-axis. Thus we expect to be able to form statistically-useful samples of high quality quasar host images in the near future."
    },
    "0209/astro-ph0209041_arXiv.txt": {
        "abstract": "A small group of X-ray binaries currently provides the best evidence for the existence of stellar-mass black holes.  These objects are interacting binary systems where the X-rays arise from accretion of material onto a compact object (i.e.\\ an object with a radius of less than a few hundred km).  In some favourable cases, optical studies of the companion star lead to dynamical mass estimates for both components.  In 17 cases, the mass of the compact object an X-ray binary has been shown to exceed the maximum mass of a stable neutron star (about $3\\,M_{\\odot}$), which leads to the conclusion that these objects are black holes.  In this contribution I will review the basic properties of these black hole binaries. ",
        "introduction": "Black holes represent an extreme manifestation of Einstein's theory of general relativity.  As an observational astronomer, I will not consider any of the detailed theory of black holes but will instead consider only the practical question of how to find them.  The usual route is by the process of elimination.  In the current universe, black holes must form via the gravitational collapse of a dying star. Three outcomes are possible, depending on the mass of the degenerate core: a white dwarf, a neutron star, or a black hole.  The mass of a white dwarf cannot exceed the well-known Chandrasekhar limit, and a neutron star has a somewhat analogous upper mass limit, generally thought to be on the order of $3\\,M_{\\odot}$ (Rhoades \\& Ruffini 1974; Kalogera \\& Baym 1996).  Once the mass of the degenerate core exceeds about $3\\,M_{\\odot}$, no known force can halt the gravitational collapse, and a black hole must be formed.  Given this, we have a relatively straightforward observational definition of a black hole: {\\em A black hole is a compact object with a mass greater than three solar masses}, where ``compact'' in this context means an object with a radius smaller than about 100 km (i.e.\\ much less than the radius of normal stars).  Since black holes are dark, the only way one could hope to observe them is through their gravitational influence on surrounding matter. Early attempts to search catalogs of spectroscopic binaries to look for single-lined binaries with massive and undetected companions were not successful (Zel'dovich \\& Guseynov 1966; Trimble \\& Thorne 1969). A far more efficient approach for source selection has its roots in the 1960s when Zel'dovich and others realized that if a black hole accreted material (either from a nearby companion star or the interstellar medium), then it might shine brightly in X-rays and $\\gamma$ rays.  Today we know of a few hundred bright X-ray sources which must be powered by accretion onto a compact object, either a neutron star or a black hole (i.e.\\ rapid variability in many cases indicates a size scale of a few hundred km or less).  All of the black hole candidates discussed below were selected on the basis of their X-ray activity, although two of the sources, V404 Cyg and V4641 Sgr, were previously known (optically) variable stars that were essentially ignored until they were associated with bright X-ray sources.  Thus, one should keep in mind that owing to special circumstances needed to observe them (i.e.\\ interacting binaries), the list of presently known stellar-mass black holes is probably a very biased sample and may not be representative of the general population of black holes. ",
        "conclusions": ""
    },
    "0209/astro-ph0209277_arXiv.txt": {
        "abstract": "An infrared-selected, narrow-line QSO has been found to exhibit an extraordinarily broad \\halpha\\ emission line in polarized light.  Both the extreme width (35,000~\\kms\\ full-width at zero intensity) and 3,000~\\kms\\ redshift of the line centroid with respect to the systemic velocity suggest emission in a deep gravitational potential.  An extremely red polarized continuum and partial scattering of the narrow lines at a position angle common to the broad-line emission imply extensive obscuration, with few unimpeded lines of sight to the nucleus. ",
        "introduction": "If the Unified Scheme for AGN (e.g., Antonucci 1993) is to be equally successful for objects of high as well as low luminosity, there should exist a large number of QSOs whose optical/UV spectra are dominated by narrow emission lines (Type 2 QSOs or QSO-2s).  Like their Seyfert counterparts, the broad emission-line regions of such objects are expected to lie within and behind a nominally toroidal obscuring structure, so that broad lines are most readily visible in light polarized by scattering off particles above and below the plane of obscuration. Though a few narrow-line objects with quasar luminosities have been found with broad permitted lines in infrared and/or polarized light (esp. {\\it IRAS\\/} QSOs: Hines et al. 1995, 1999; Young et al. 1996), it is clear that such highly-obscured nuclei are generally not present in classic lists of objects selected via UV excess.  Far more promising is the AGN catalog being compiled from the Two Micron All Sky Survey (2MASS; Cutri et al. 2001).  The fundamental selection criterion of this sample is a color limit ($J-K_S>2$) that weeds out stars while at the same time favoring extragalactic sources with substantial reddenning toward their nuclei.  This expectation has received strong support from the discovery of optical linear polarization in a sizeable fraction of 2MASS QSOs (in some cases $P\\gtrsim10\\%$; Smith et al. 2002) as well as in the detection of large absorbing columns toward the associated nuclear X-ray sources (Wilkes et al. 2002).  The above studies found a surprisingly large number of 2MASS QSOs with the seemingly paradoxical characteristics of polarization/absorption coupled with prominent Type 1 (broad emission-line) total-flux spectra, properties that are not easily explained by the simplest models. Nevertheless, the narrow-line 2MASS AGN provide a number of high-luminosity analogues to Seyfert 2 galaxies where one might seek the broad, polarized emission lines that signal the presence of an accretion source powering the nucleus. This paper reports the first such discovery: a narrow-line QSO with an extraordinarily broad scattered \\halpha\\ emission line and a polarized flux spectrum that may yield new information on the inner structure of AGN. ",
        "conclusions": "Much has been made of the illumination of particles situated above and below the toroidal plane by the brilliant nucleus.  For edge-on perspectives, the scattering of broad-line emission by these particles can provide an indirect view of the nucleus that we detect as polarized lines like the enormous \\halpha\\ feature in 2M130005.  The radius of the inner edge of the torus is typically inferred by molecular/maser and hot dust emission at $R\\sim1$~pc. Further reprocessing occurs at mid- and far-infrared wavelengths in a possibly distinct structure of much larger size.  Clouds in the inner torus are thought to be individually optically thick at most wavelengths (Krolik \\& Begelman 1988), but the filling factor may be low, so it is unclear what the overall vertical optical depth is to visible-light photons in the inner few hundred pc.  If this region has significant transparency, some of the narrow-line emission from one lobe of a bipolar narrow line region will pass through the central plane to be scattered off particles in the other lobe.  With narrow lines of [\\ion{O}{3}], [\\ion{N}{2}], and \\halpha\\ showing a reduced, but significant, fractional polarization at the same position angle as the scattered nuclear emission, it is useful to estimate the importance of this effect.  We write the fraction of the narrow-line emission emitted on one side of the nucleus that is subsequently scattered by the opposite lobe as \\begin{equation} {L_{\\lambda,{\\rm SC}}\\over L_{\\lambda,{\\rm EM}}} \\sim {\\Omega R e^{-\\tau} \\over 4\\pi} \\end{equation} where $\\Omega$ is the solid angle of the scattering lobe as seen from the emitting lobe, $\\tau$ is the effective vertical optical depth of the nuclear region, and $R$ is the effective scattering efficiency of the lobe.  We have assumed isotropic emission for this crude exercise.  Assuming that the scattering process is also isotropic implies that the observed fluxes are in the same ratio.  At the end of \\S3, we found that the observed ratio was $\\sim$0.1. The quantity $R$ includes variables such as the optical depth in a lobe and the ratio of scattering to total extinction, and we have difficulty imagining an overall value $R>0.5$.  For this limit and $\\tau$ small, we find $\\Omega\\gtrsim\\pi$, i.e., a half-opening angle for a scattering lobe $\\beta\\gtrsim60^\\circ$.  While the opening angle is uncomfortably large in comparison to the observed angular extents of scattering and [\\ion{O}{3}]-emitting lobes in resolved AGN, an ionization gradient is expected to exist within each cloud, with high-ionization species like O$^{++}$ predominately confined to the surface facing the nuclear engine. Hence, a scattering cloud may see brighter narrow lines than an edge-on observer. An internal ionization structure also suggests that the backscattering effect would be greater for high-ionization lines than for low-ionization lines, possibly explaining the absence of [\\ion{O}{1}]$\\lambda\\lambda$6300,6363 and [\\ion{S}{2}]$\\lambda\\lambda$6717,6731 in the polarized light of 2M130005 (see Figure 1, inset). Finally, because this explanation places no requirement on the radial location of the scattering particles, the broad-line emission and continuum could be scattered so close to the toroidal plane that the light has difficulty emerging over the ``lip'' of the dusty torus.  The resulting extinction could account for the unusually red polarized continuum depicted in Figure 1.  A more attractive explanation for the observed polarization spectrum may be scattering in an extended narrow-line region that contains a global ionization gradient.  Stratification of the narrow-line region has been established in radio-loud AGN (Hes, Barthel, \\& Fosbury 1993) and Seyfert galaxies (Filippenko \\& Halpern 1984), among others, and can naturally lead to polarization differences between species (e.g., Hines et al. 1999 and Tran, Cohen \\& Villar-Martin 2000 for the [\\ion{O}{3}] and [\\ion{O}{2}] lines in IRAS P09104+4109).  Unfortunately, this mechanism provides no natural explanation for the extremely red continuum in polarized light.  The $5000-8000$~\\AA\\ portion of the latter can be matched to the spectrum of a typical QSO using a Galactic reddening law with $A_V\\sim4$~mag. At shorter wavelengths the dereddened spectrum turns up too rapidly, suggesting the onset of a prominent 3000~\\AA\\ bump. Of course, if dust grains are the scattering particles, this amount of extinction should be considered a lower limit, since the spectrum might be bluened by the scattering.  A somewhat smaller continuum suppression factor is inferred from a comparison of the measured equivalent width of [\\ion{O}{3}]$\\lambda$5007, 330~\\AA, with the average for low-$z$ QSOs from Boroson \\& Green (1992), $\\sim$30~\\AA. However, this result is subject to possible extinction of the narrow-line emission.  Heavy extinction of a scattered nuclear spectrum is apparently rare for AGN (see Young et al. 1996 for IRAS 2306+0505), and one is left with the impression that the nucleus of 2M130005 is obscured along most sightlines.  As noted above, the broad \\halpha{} emission is redshifted by about 3000~\\kms\\ with respect to the narrow lines and stellar features.  This is substantially greater than the mean redward shift of \\hbeta\\ with respect to the systemic velocity measured in the large samples of McIntosh et al. (1999) and Zamanov et al. (2002). In fact, the shift observed in 2M130005 is three times larger than the largest value reported in either study.  This may indicate that scattering occurs in an outflowing medium (see also Hines et al. 1995).  However, if we assume that the redshift has a gravitational origin, the ratio of the black hole mass to the characteristic radius of the broad line-emitting region can be estimated as ${\\rm M}_{\\rm BH} / {\\rm R}_{\\rm BLR} \\simeq 10^{6}~{\\rm M}_{\\odot} / {\\rm AU}$.  It is interesting to note that this is similar to the result of an independent derivation, ${\\rm M}_{\\rm BH} / {\\rm R}_{\\rm BLR} \\simeq 4 \\times 10^{5}~{\\rm M}_{\\odot} / {\\rm AU}$, made under the assumption that the broad-line clouds are in Keplerian orbits about the central black hole with a maximum orbital velocity equal to the half width at zero intensity of scattered \\halpha\\ ($\\sim$18,000~\\kms).  An estimate for the black hole mass in 2M130005 therefore lacks only a size for the broad-line region from, e.g., reverberation mapping (e.g., Kaspi et al. 2000) or the overall accretion luminosity.  The latter will be possible using wide-band photometry with {\\it SIRTF\\/}.  An X-ray spectrum will soon be obtained, and infrared spectroscopy will be essential for defining the intrinsic spectrum and obscuration of the nucleus of this new Type 2 QSO."
    },
    "0209/astro-ph0209511_arXiv.txt": {
        "abstract": "To further enhance our understanding on the formation and evolution of bars in lenticular (S0) galaxies, we are undertaking a detailed photometric and spectroscopic study on a sample of 22 objects. Here we report the results of a 2D structural analysis on two barred face--on S0's, which indicate that presently these galaxies do not possess disks. We discuss two possibilities to explain these surprising results, namely strong secular evolution and bar formation without disks. ",
        "introduction": "At the very beginning of galactic astronomy, barred galaxies were considered as anomalies in the realm of the nebulae. Later, N-body numerical calculations (e.g., \\opencite{hoh71}) as well as analytic studies (e.g., \\opencite{kal72}) gave us an idea on how bars are formed in galaxies, namely via a global instability in dynamically cool disks. Then, the problem was how we could explain why not {\\em all} galaxies have bars, until a solution was found based on increasing the stellar velocity dispersion in the disk, and/or adding a halo (see, e.g., \\opencite{ost73}).  After a period of calmness, some recent studies brought up issues that have put us back again almost at the starting point! \\inlinecite{too81} already argued that a high central density disk will not form a bar, which was confirmed numerically by \\inlinecite{sel99}. Nonetheless, we observe barred galaxies with dense centres (see \\opencite{sel00}). Thus, how galaxies form bars remains an open question. The answer is not only interesting for academic reasons, for it can give us important clues and constraints to study the formation and evolution of galaxies (see, e.g., \\opencite{gad01}).  The problem gets worse if we consider S0 galaxies, since these are not dynamically cool systems. To tackle this problem, we have collected optical (B, V, R, I) and near-infrared (Ks) images of a sample of 22 galaxies. Furthermore, we have also taken spectra along the major and minor axes of the bar with a S/N high enough to obtain line of sight velocity distribution (LOSVD) profiles. We report here partial results on two galaxies which have led us to surprising conclusions. ",
        "conclusions": ""
    },
    "0209/astro-ph0209457_arXiv.txt": {
        "abstract": "We present a secure redshift of $z=0.944\\pm0.002$ for the lensed object in the Einstein ring gravitational lens \\blazar\\ based on five broad emission lines, in good agreement with our preliminary value announced several years ago based solely on the detection of a single emission line. ",
        "introduction": "\\blazar\\ is a strong flat-spectrum radio source which was discovered to be gravitationally lensed by \\cite{pat93}. The object consists of two point sources which are radio loud and variable, with similar flat spectra in the radio regime.  In addition to the point sources, there is an Einstein ring of 0.33 arcsec diameter. This object is the smallest known Einstein radio ring (Patnaik \\etal\\ 1993). \\cite{big99} (see also Cohen \\etal\\ 2000) have measured a time delay for \\blazar\\ of 10.5$\\pm0.4$ days. HST optical images by \\cite{kee98} and by \\cite{leh00} and NICMOS images by \\cite{jac00} reveal the lensing galaxy clearly.  \\blazar\\ is an extremely important object for studies of gravitational lensing.  The Einstein ring provides a strong constraint on the mass distribution in the lens. The simple source structure and small angular size of \\blazar, the constraints from sub-structure in the images which are well mapped with VLBI \\citep*[see][]{biggs2002}, and the apparent absence of significant external shear make this object easier than most to model. As the error on the time delay is currently estimated to be 3 percent (1$\\sigma$), (Biggs, private communication), \\blazar\\ is a prime target for determining the Hubble constant and so it is vital to obtain secure and accurate lens and source redshifts.  In order to fulfill the promise of \\blazar\\ for these purposes, both the source and the lens redshift are required.  \\cite{bro93}, and independently \\cite{sti93}, established the redshift of one of these, which they assumed was the lens, as $z=0.6847$, detecting several emission and absorption features.   Subsequently absorption arising in the lensing galaxy has been detected at 21 cm \\citep{car93}, in CO \\citep{wik95}, and in formaldehyde \\citep{men96} against the background source.   However, not surprisingly, the source redshift for \\blazar\\ proved more elusive, since the source is a blazar \\citep{odea92}.  We therefore started an effort to find the source redshift in 1994 using the Low Resolution Imaging Spectrograph \\citep{oke95} at the Keck Observatory. We were able to detect one definite weak emission feature fairly rapidly, but could not find a secure second line.  Assuming the detected line at 5460\\AA\\ is the 2800 \\AA\\ Mg II line, the redshift of the lensed object in \\blazar\\ is then $z \\sim0.95$, further supporting \\cite{bro93} and \\cite{odea92}, who offer $z=0.94$ as a ``tantalizing possibility, rather than a firm claim''. This result was announced at two conferences in 1996 \\citep{coh96,law96}. It has taken longer than expected, but we have finally succeeded in detecting with confidence multiple emission lines from the source. We present in this brief paper the secure redshift of the lensed object in \\blazar\\ based on the detection of five broad emission lines, $z=0.944\\pm0.002$. ",
        "conclusions": ""
    },
    "0209/astro-ph0209382_arXiv.txt": {
        "abstract": "We update our analysis of recent exoplanet data that gives us a partial answer to the question: How does our Solar System compare to the other planetary systems in the Universe? Exoplanets detected between January and August 2002 strengthen the conclusion that Jupiter is a typical massive planet rather than an outlier. The trends in detected exoplanets do not rule out the hypothesis that our Solar System is typical. They support it. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209310_arXiv.txt": {
        "abstract": "Planetary transits of bright stars, $V<10$, offer the best opportunity for detailed studies of extra-solar planets, such as are already being carried out for HD209458b.  Since these stars are rare, they should be searched over the entire sky.  In the limits of zero read-out time, zero sky noise, and perfect optics, the sensitivity of an all-sky survey is independent of telescope aperture: for fixed detector size and focal ratio, the greater light-gathering power of larger telescopes is exactly cancelled by their reduced field of view.  Finite read-out times strongly favor smaller telescopes because exposures are longer so a smaller fraction of time is wasted on readout.  However, if the aperture is too small, the sky noise in one pixel exceeds the stellar flux and the field of view becomes so large that optical distortions become unmanageable.  We find that the optimal aperture is about $1''$. A one-year survey using such a $1''$ telescope could detect essentially all hot-jupiter transits of $V<10$ stars observable from a given site. ",
        "introduction": "In Pepper, Gould \\& DePoy (astro-ph/0208042), we argued that all-sky surveys are the best way to find planets transiting bright $V<10$ stars. Such transits offer the best opporunity for detailed studies of planets. We showed that the number of systems probed is given by \\begin{equation} \\label{equfin} N_{p} = \\frac{4}{3} \\pi n \\eta \\, d_{0}^{3} \\, \\left( \\frac{R_{0}}{a_{0}} \\right) \\left( \\frac{L}{L_{0}} \\right)^{3/2} \\left( \\frac{R}{R_{0}} \\right)^{-7/2} \\left( \\frac{a}{a_{0}} \\right)^{-5/2} \\left( \\frac{r}{r_{0}} \\right)^{6} , \\end{equation} where $d_0$ is the maximum distance at which an $i = 90^{\\rm o}$ transit can be detected for a star of luminosity $L_{0}$, radius $R_{0}$, with a planet at semi-major axis $a_{0}$ and radius $r_{0}$, and where $n$ is the local number density of such stars and $\\eta=0.719$ is a numerical factor.  We then showed that the sensitivity of a given survey will depend almost entirely on $\\gamma$, the number of photons collected from a fiducial star of some fixed designated magnitude, which we arbitrarily chose to be $V=10$. That is, it will not depend on the details of the all-sky observing program. We then normalized equation (\\ref{equfin}) in terms of $\\gamma$, \\begin{equation} \\label{equfin2} N_{p} = 600 F(M_{V}) \\left( \\frac{a}{a_{0}} \\right)^{-5/2} \\left( \\frac{r}{r_{0}} \\right)^{6} \\left( \\frac{\\gamma}{\\gamma_{0}} \\right)^{3/2} \\left( \\frac{\\Delta \\chi_{\\rm min}^{2}}{36} \\right)^{-3/2} \\end{equation} where we adopted $\\gamma_{0} = 2 \\times 10^{7}$, $a_{0}=20 \\, R_{\\odot}$, $r_{0} = 0.10 \\, R_{\\odot}$, and where we have made our evaluation at $M_{V} = 5$ (i.e. $R = 0.97 \\, R_{\\odot}$, $V_{\\rm max} = 10$, $d_{0} = 100 \\, $pc, and  $n = 0.004 \\, {\\rm pc}^{-3}$). Here, $F(M_V)$ is a function, which is shown in Figure \\ref{figF}, and $\\Delta \\chi^2_{\\rm min}$ is the minimum $\\chi^2$ improvement of a transit-model relative to a constant-flux model, which is set to avoid excessive false positives. Note that $\\gamma_0=2 \\times 10^{7}$ corresponds approximately to 1000 20-second exposures with a 5 cm telescope and a broadened $(V+R)$ type filter for one $V = 10$ mag fiducial star.  Our goal was to apply this formula to the problem of telescope design in a subsequent paper.  However, in light of the referee report, we decided to append our work on telescope design as an additional chapter of the original paper.  The following is that chapter. ",
        "conclusions": ""
    },
    "0209/astro-ph0209126_arXiv.txt": {
        "abstract": "Soon after launch, the Advanced CCD Imaging Spectrometer (ACIS), one of the focal plane instruments on the Chandra X-ray Observatory, suffered radiation damage from exposure to soft protons during passages through the Earth's radiation belts. The ACIS team is continuing to study the properties of the damage with an emphasis on developing techniques to mitigate charge transfer inefficiency (CTI) and spectral resolution degradation. A post-facto CTI corrector has been developed which can effectively recover much of the lost resolution. Any further improvements in performance will require knowledge of the location and amount of sacrificial charge - charge deposited along the readout path of an event which fills electron traps and changes CTI. We report on efforts by the ACIS Instrument team to characterize which charge traps cause performance degradation and the properties of the sacrificial charge seen on-orbit. We also report on attempts to correct X-ray pulseheights for the presence of sacrificial charge. ",
        "introduction": "\\label{sect:intro}  The Chandra X-ray Observatory, the third of NASA's great observatories in space, was launched just past midnight on July 23, 1999, aboard the space shuttle {\\it Columbia}\\cite{cha2}.  After a series of orbital maneuvers Chandra reached its final, highly elliptical, orbit.  Chandra's orbit, with a perigee of 10,000~km, an apogee of 140,000~km and an initial inclination of 28.5$^\\circ$, transits a wide range of particle environments, from the radiation belts at closest approach through the magnetosphere and magnetopause and past the bow shock into the solar wind.  The Advanced CCD Imaging Spectrometer (ACIS), one of two focal plane science instruments on Chandra, utilizes charge-coupled devices (CCDs) of two types, front- and back-illuminated (FI and BI).  Soon after launch it was discovered that the FI CCDs had suffered radiation damage from exposure to soft protons scattered off the Observatory's grazing-incidence optics during passages through the Earth's radiation belts\\cite{gyp00}.  Since mid-September 1999, ACIS has been protected during radiation belt passages and there is an ongoing effort to prevent further damage and to develop hardware and software strategies to mitigate the effects of charge transfer inefficiency on data analysis.  One such strategy, post-facto correction of event pulseheights based on knowledge of the charge history along the transfer direction, is described here.  This paper begins by describing the characteristics of ACIS radiation damage in Section~\\ref{sect:raddamage} and of sacrificial charge in Section~\\ref{sect:saccharge}.  Section~\\ref{sect:data} defines the calibration data used to test correction algorithms.  The basic CTI correction algorithm and the additional sacrificial charge correction algorithm are outlined in Sections~\\ref{sect:cticorr} and \\ref{sect:saccorr}.  The spectral resolution performance of the correction algorithms are compared in Section~\\ref{sect:perf} and further discussed in Section~\\ref{sect:disc}. ",
        "conclusions": "\\label{sect:disc}  While our preliminary CTI plus sacrificial charge correction algorithm does offer improved performance, we believe that further improvement is possible.  Figure~\\ref{fig:ideal} compares the FWHM at 5.9 and 1.5~keV versus row number resulting from our sacrificial charge correction with an estimated theoretical limit on performance.  This limit was determined by assuming that the fluctuations in charge loss and in sacrificial charge are Poissonian and includes an estimate of the additional noise from split events.  At 1.5~keV the performance has reached this theoretical limit, while at 5.9~keV the sacrificial charge corrected data are still substantially worse than the limit.  \\begin{figure} \\vspace{4.0in} \\special{psfile=fitrow2.eps angle=0 hscale=80 vscale=80 voffset=0 hoffset=+20} \\caption{FWHM of the spectral lines at 5.9 and 1.5~keV as a function of row number for sacrificial charge corrected data.  The solid line is an estimate of the theoretical limit on performance.} \\label{fig:ideal} \\end{figure}  One source of excess noise in the sacrificial charge correction is the presence of cosmic ray events which interact in the framestore region of the CCD.  These events appear in raw frames and event lists as sacrificial charge and have been included in the calculation of $Z_X$, but do not actually influence the pulseheight of X-ray events.  If a framestore event is present, particularly if it appears to be the closest precursor charge to the X-ray event, the value of $Z_X$ will be overestimated and the charge loss underestimated.  These X-ray events will be poorly corrected and should broaden the spectral resolution.  While it is possible to run the CCD such that only framestore events are read out to determine general population characteristics, it is not possible to know which precursors of a given X-ray are sacrificial charge and which are not.  Another possible source of performance degradation that is ignored in the current treatment is the influence of sacrificial charge from previous integration frames of the CCD.  The typical exposure time for an ACIS CCD frame is 3.2~sec, while the longest measured emission time constant on ACIS is estimated at 4 - 6~sec.  While events with sacrificial charge in the same frame should be relatively unaffected, some events with $Z_X$ = 0 may have their charge loss overestimated and should again broaden the spectral resolution.  To check the efficacy of further work on the sacrificial charge algorithm alone, Figure~\\ref{fig:znoz} is a comparison of the FWHM versus row number for events with $Z_X > 0$ (sacrificial charge present) to events with $Z_X = 0$ (no sacrificial charge) after CTI plus sacrificial charge correction.  The results are not significantly different, indicating that the events with sacrificial charge have been corrected as well as events that do not have sacrificial charge within the statistical errors of the data. Since the results in Figure~\\ref{fig:ideal} argue that more improvement is possible, at least at higher energies, we must examine the assumptions in both the CTI and sacrificial charge correction.  \\begin{figure} \\vspace{4.0in} \\special{psfile=fitrow3.eps angle=0 hscale=80 vscale=80 voffset=0 hoffset=+20} \\caption{FWHM of the spectral lines at 5.9 and 1.5~keV as a function of row number for events corrected for CTI and sacrificial charge for the case of $Z_X > 0$, events with sacrificial charge, and of $Z_X = 0$, events with no sacrificial charge.} \\label{fig:znoz} \\end{figure}  The primary difference between the low and high energy events is the importance of events split into multiple pixels; at 1.5~keV 79\\% of the events are single pixel, 19\\% are doubly split and 2\\% are multiply split, while at 5.9~keV 47\\% are singly split, 33\\% are doubly split and 20\\% are multiply split.  The additional spectral resolution degradation at 5.9~keV over 1.5~keV may be a result of incorrect treatment of split events.  There are at least two means by which split events could be poorly corrected.  The treatment of self-shielding within the event island assumes that the emission time constants are much longer than the event transfer time (40~$\\mu$sec), however there are a small number of traps with a time constants close to the transfer time (2\\% with 60~$\\mu$sec).  This should be a small effect, since the fraction of relevant traps is so small.  The energy dependence of charge loss, which is currently parameterized as a power law, has not yet been confirmed at the lowest energies ($<$~670~eV).  Split events will often include pixels with small amounts of charge, so an error in the energy dependence could produce a additional noise in the corrected pulseheights.  This possibility is currently under investigation.  If a sacrificial charge correction scheme does provide significant performance improvement, its implementation will require changes to the flight software to telemeter additional information about sacrificial charge history with each event without making a significant impact on telemetry bandwidth.  Such a patch is under development by the ACIS team.  The strategies currently under study include encoding precursor information into bits normally used to telemeter corner pixel (in 3x3 faint mode) or outer pixels (in 5x5 very faint mode) pulseheight."
    },
    "0209/astro-ph0209256_arXiv.txt": {
        "abstract": "Comparisons of the kinematics of {MgII} absorbing gas and the stellar rotation curves in $0.5 \\leq z \\leq 1.0$ spiral galaxies suggests that, at least in some cases, the extended gaseous envelopes are dynamically coupled to the stellar matter.  A strong correlation exists between the overall kinematic spread of {MgII} absorbing gas and {CIV} absorption strength, and therefore kinematics of the higher--ionization gas.  Taken together, the data may suggest a ``halo/disk connection'' between $z\\sim 1$ galaxies and their extended gaseous envelopes.  Though the number of galaxies in our sample are few in number, there are no clear examples that suggest the gas is accreting/infalling {\\it isotropically\\/} about the galaxies from the intergalactic medium. ",
        "introduction": "For $0.5 \\leq z \\leq 1$, there are observed correlations between galaxy luminous properties and {MgII} absorption properties that support a view in which metal--enriched extended ($\\sim 40$~kpc) gaseous envelopes of normal bright galaxies are coupled to galaxies (e.g.\\ \\cite{ref:bb91}; \\cite{ref:sdp94}; \\cite{ref:steidel95}).  An alternative view, extracted from numerical simulations of cosmic structure growth, is that the gas is concentrated along intergalactic filaments, where matter overdensities also give rise to mergers and normal bright galaxies.  By $z \\sim 1$, do galaxies remain coupled to the cosmic flow of baryons driven by matter overdensities or have they decoupled?  If the latter, they likely sustain their gaseous envelopes via mechanical means within the galaxies.  In this contribution, we present data that suggest the {MgII} absorption and the emission line kinematics are coupled in some galaxies.  We also discuss the kinematic relationship between {CIV} and {MgII} and present the first galaxy for which data of the emission and {MgII} and {CIV} absorption kinematics are available. ",
        "conclusions": "If galaxy evolution to the present epoch is governed by the accretion of gas from the IGM, the gas would provide a tracer of the structure, kinematics, and chemical enrichment of the cosmic web.  The gas would not necessarily be coupled to galaxy emission--line kinematics in the majority of cases; neither merging events nor IGM accretion predict strong coupling between the gas kinematics and the stellar kinematics. A large statistical sample is needed to discern the veracity of this expectation.  What scenario, then, can predict the observed coupling between the kinematics of the extended gas envelopes and the galaxy stars? Following a merging event, star formation rates are elevated long after the stellar system has relaxed.  Supernovae inject gas into the halos of their host galaxies.  This scenario naturally provides for the expulsion of gas from galaxies that is metal enriched and harbors some memory of the dynamical state of the stellar component of the galaxies."
    },
    "0209/astro-ph0209583_arXiv.txt": {
        "abstract": "{We report on a dramatic flux ($\\sim50\\%$ increase in the LECS and MECS band) and spectral variation between two BeppoSAX observations of the Circinus Galaxy performed almost three years apart. Through the analysis of all $Chandra$ observations available in the archive, including a new DDT observation on May 2001, we show that a high flux state of an extremely variable Ultra Luminous X-ray source \\citep[CG~X-1: ][]{Bauer01}, which is within the adopted BeppoSAX source extraction region of $2\\arcmin$, is the most likely explanation for most of the observed variation. However, the presence of a high flux 6.7 keV line and the spectral variation of the PDS in the new BeppoSAX data could be partly due to intrinsic variation of the nucleus. Comparing the longest Chandra observation and the BeppoSAX one, we find that the long-term flux variability of CG~X-1 is not accompanied by a significant spectral variability. We also re-analysed the $Chandra$ HEG nuclear spectra and report on the presence of a Compton shoulder with a flux of about $20\\%$ the line core, in agreement with theoretical expectations for Compton-thick matter. ",
        "introduction": "The Circinus Galaxy hosts one of the closest (3.8 Mpc) and X-ray brightest ($F_{\\rm 2-10\\,keV} \\sim 1.5\\times10^{-11}$ erg cm$^{-2}$ s$^{-1}$) Seyfert~2. The detailed HST images clearly reveal a compact ($<2$ pc) active nucleus  seen through high obscuration, surrounded by extended and complex structures \\citep{wil00}.  The first X-ray observation was performed during the $ROSAT$ All Sky Survey \\citep{brink94}. A reflection dominated spectrum was revealed by $ASCA$, together with a prominent neutral iron K$\\alpha$ line and a number of other lines from lighter elements \\citep{Matt96}. The BeppoSAX observation added a precious piece of information, detecting the direct nuclear emission above $\\sim10$ keV, absorbed at lower energies by a Compton--thick ($4\\times10^{24}$ cm$^{-2}$) material, usually identified with the torus envisaged in unification models \\citep{Matt99, Guainazzi99}.  The line spectrum above $\\sim2$ keV clearly originates from low ionized matter \\citep{Matt96,net98,sako00}. It was shown that it is fully compatible with reflection from the inner surface of a mildly ionized torus, the same matter likely responsible for the absorption of the nuclear radiation: this interpretation leads to an estimate of the inner radius of the torus of $\\sim0.2$ pc \\citep{bmi01}. The spectrum below $\\sim2$ keV is probably contaminated by an extended emission and/or off-nuclear sources within $ASCA$ and BeppoSAX extraction regions.  This scenario was basically confirmed by $Chandra$. Two different regions are clearly observed: one compact and spatially unresolved ($<$15 pc) which corresponds to the nucleus, where the reflection spectrum and the iron line is produced; the other, extended over about 50 pc, where most of the soft emission is produced \\citep{Sambruna01a,Sambruna01b}. Furthermore, a number of off-nuclear sources were detected, mostly concentrated within $2\\arcmin$ of the nucleus \\citep{sw01,Bauer01,Sambruna01a}: at least one of them is likely to have contaminated $ASCA$ and SAX observations during its high flux states (see below).  In this paper we report on a dramatic flux and spectral variation detected in a second BeppoSAX observation of the Circinus Galaxy performed on January 2001, almost three years later than the first one. In order to discriminate between a nuclear variation and the contamination of an off--nuclear source, we will make extensive use of all $Chandra$ observations available in the archive, together with unpublished Director's Discretionary Time (DDT) data taken on May 2001. We adopt the names after \\citet{Bauer01} (from now on B01) for the two sources discussed in this paper, CG~X-1 ($\\alpha_{\\rm 2000}=14^h13^m12.^s3$, $\\delta_{\\rm 2000}=-65\\degr20\\arcmin13\\arcsec$) and CG~X-2 ($\\alpha_{\\rm 2000}=14^h13^m10.^s0$, $\\delta_{\\rm 2000}=-65\\degr20\\arcmin44\\arcsec$). ",
        "conclusions": "The most likely explanation of at least most of the flux and spectral variation observed between two BeppoSAX observations taken almost three years apart is in terms of a high-flux state of CG~X-1, which is located well within the adopted BeppoSAX source extraction region of $2\\arcmin$. There are several pieces of evidence:  \\begin{itemize} \\item CG~X-1 shows a strong, long-term variation, reaching a flux which is comparable with that of the nucleus and is consistent with that measured in the residuals of the two SAX observations. \\item The spectral shape of CG~X-1, when averaged over more than a period, is fully consistent with the spectrum of the LECS and MECS BeppoSAX residuals. \\item Finally, the strongest evidence comes from the short-term variation of CG~X-1, on a period of 27 ks: the MECS lightcurve of the new observation clearly varies with the same period, indicating that a significant part of the observed flux originates from CG~X-1. The periodic behaviour was also present in the data from the old SAX observation, but the variation amplitude was much less, being consistent with contamination from a low-flux state of CG~X-1.  \\end{itemize}  The best interpretation for the nature of CG~X-1 is in terms of a $\\geq$ 80 M$_{\\sun}$ black hole in an accreting binary system in Circinus, as previously suggested by \\citetalias{Bauer01}. Our analysis of all $Chandra$ observations and the BeppoSAX spectrum shows that the long-term variability of this source is not associated to spectral variations, indicating that it is not due to changes of the inner radius of the disk.  However, CG~X-1 cannot be the cause of the variation of the PDS and the presence of an ionized iron line in the residuals between the two BeppoSAX observations. In both cases, a variation of the properties of the circumnuclear matter in the AGN environment is a possible explanation: an increase of the column density of the torus, for example, would cause the observed decrease of the PDS count rates at lower energies. However, it is difficult to imagine the physical cause to support such an increase.  As for the ionized iron line, it should be noted that the SAX residuals must include a line from CG~X-2, which was likely not present in the old observation. Nevertheless, the contribution of the flux of this line is too low to exclude a variation of the 6.7 keV iron line originating from the nucleus.  Finally, our new analysis of the $Chandra$ HEG nuclear spectrum has led to the detection of a Compton shoulder in the 6.4 keV iron line. Its flux is about 20$\\%$ the line core, in agreement with theoretical expectations for Compton-thick matter, providing one more argument in favour of the association of the matter producing the iron line with the $4\\times10^{24}$ cm$^{-2}$ neutral absorber."
    },
    "0209/gr-qc0209027_arXiv.txt": {
        "abstract": " ",
        "introduction": "The idea of using a pair of twin satellites, denoted as S1 and S2, in identical orbits with the same semimajor axes $a$ and eccentricities $e$, except for the inclinations $i$ of their orbital planes, which should be supplementary, in order to measure the general relativistic Lense--Thirring effect (Lense and Thirring 1918) in the gravitational field of the Earth\\footnote{In (Ciufolini {\\it et al} 1998) an experimental check of such prediction of General Relativity in the field of the Earth by using the laser data of LAGEOS and LAGEOS II satellites is reported. The claimed accuracy is of the order of 20$\\%$.} was put forth for the first time by Ciufolni with the proposed LAGEOS--LAGEOS III mission (Ciufolini 1986). The proposed observable is the sum of the rates of the longitudes of the ascending nodes \\eqi\\Sigma\\dot\\Omega\\equiv\\dot\\Omega_{\\rm S1}+\\dot\\Omega_{\\rm S2}.\\lb{sumn}\\eqf Indeed, it turns out that while the Lense--Thirring secular nodal rates are independent of the inclinations of the satellites and add up in \\rfr{sumn}, the classical secular nodal rates induced by the oblateness of the Earth, which would mask the relativistic effect due to the uncertainties in the even zonal coefficients $\\delta J_2,\\ \\delta J_4,\\ \\delta J_6,...$ of the multipolar expansion of the terrestrial gravitational field, are equal and opposite for supplementary orbital planes because they depend on odd powers of $\\cos i$, so that they would be cancelled out by \\rfr{sumn}. Later on, the orbital and physical configuration of LAGEOS III slightly changed: the eccentricity of its orbit was increased in order to be able to perform other general relativistic tests, its mass was reduced so to reduce the mission--launch costs, and the area was reduced in such a way to guarantee the same area--to--mass ratio of the older LAGEOS, so to reduce the impact of the non--gravitational perturbations. Thus LARES was born (Ciufolini 1998). In Table 1 we quote the orbital parameters of some existing or proposed laser--ranged satellites which are used, or could be used, in general relativistic tests. The accuracy available with the originally proposed version of the LAGEOS--LARES mission should amount to 2$\\%$--3$\\%$ (Ciufolini  1998). \\begin{table}[ht!] \\caption{Orbital parameters of LAGEOS, LAGEOS II, LARES, S1 and S2.} \\label{para} \\begin{center} \\begin{tabular}{lllllll} \\noalign{\\hrule height 1.5pt} Orbital parameter & LAGEOS & LAGEOS II & LARES & S1 & S2\\\\ \\hline $a$ (km) & 12270 & 12163 & 12270 & 12000 & 12000\\\\ $e$ & 0.0045 & 0.014 & 0.04 & 0.05 & 0.05\\\\ $i$ (deg) & 110 & 52.65 & 70 & 63.4 & 116.6\\\\ \\noalign{\\hrule height 1.5pt} \\end{tabular} \\end{center} \\end{table} Very recently, some modifications of the observable to be adopted in the LARES mission have been suggested (Iorio {\\it et al} 2002). The total error should then become $\\sim\\ 1\\%$.  The concept of satellites in identical and supplementary orbits have been recently extended also to the perigees (Iorio 2002; 2003). In particular, it has been noticed that also the difference of the rates of the perigees \\eqi\\Delta\\dot\\omega\\equiv\\dot\\omega_{\\rm S1}-\\dot\\omega_{\\rm S2}\\lb{diffp}\\eqf could be considered, in principle, for measuring the gravitomagnetic field of the Earth. Indeed, the Lense--Thirring secular apsidal rates depend on $\\cos i$ and add up in \\rfr{diffp}, while the classical secular apsidal rates due to the oblateness of the Earth, which depend on $\\cos^2 i$ and on even powers of $\\sin i$, are equal and cancel out in \\rfr{diffp}. Of course, such an observable could not be considered for the LAGEOS--LARES mission since the eccentricity of LAGEOS is too small and the perigee of its orbit is badly defined. On the contrary, launching an entirely new pair of LAGEOS--type satellites in rather eccentric orbits would allow to adopt both \\rfr{sumn} and \\rfr{diffp} and also \\eqi\\dot X\\equiv\\Sigma\\dot\\Omega-\\Delta\\dot\\omega.\\lb{sumdiff}\\eqf In (Iorio 2002) it has been noticed that it should be better to adopt the critical inclinations $i_{\\rm S1}=63.4^{\\circ}$ and $i_{\\rm \\rm S2}=116.6^{\\circ}$ because, in this way, the periods of many time--dependent harmonic orbital perturbations of gravitational and non--gravitational origin would be not too long. So, it would be possible to adopt an observational time span $T_{\\rm obs}$ of a few years. This fact would be important not only from the point of view of reducing the data analysis time, but also because certain relevant and very useful assumptions on the surface properties of the satellites and on their spins motion, which would affect certain subtle but important non--gravitational perturbations, could be safely done by adopting just the first years of life of both satellites for the data analysis.  In this paper we wish to analyze quantitatively the impact of many systematic errors induced by gravitational and non--gravitational perturbations on the proposed observables so to yield realistic error budgets for such new proposed gravitomagnetic experiments and clarify if the alternative proposed observables are really competitive with the sum of the nodes.  The paper is organized as follows. In section 2 we will deal with the systematic error due to the mismodelling in the even zonal harmonics of geopotential and its sensitivity to the orbital injection errors in the inclinations of the satellites. In section 3 we will focus our attention on the impact of the non--gravitational perturbations. Section 4 is devoted to the conclusions. ",
        "conclusions": "In this paper we have quantitatively analyzed the scenarios offered by the proposal of launching a pair of new twin LAGEOS--like satellites in identical orbits and critical supplementary inclinations (CSOC satellites) in order to measure the gravitomagnetic Lense--Thirring effect not only by means of the sum of their nodes but also with the difference of their perigees. We have so intended to establish, on one hand, if the use of the perigees would be able to yield some benefits to the measurement of the gravitomagnetic frame dragging with respect to the sum of the nodes, and, on the other, if the launch of a new pair of SLR satellites would be justified also from the point of view of the node-only observable with respect to the LAGEOS-LARES project.  The future improvements in our knowledge of the Earth's gravitational field thanks to the CHAMP and GRACE missions has led us to draw our attention mainly on the impact which the mismodelling of the non--gravitational perturbations (NGP) could have on the proposed gravitomagnetic observables.  The sum of the nodes would yield by far the most accurate results. The obtainable precision should be realistically considered at the level of the order of\\footnote{Here we do not consider the impact of measurement errors like plate motion, atmosphere and polar motion. Moreover, also the impact of the ocean  tidal perturbations has not been addressed. In (Watkins {\\it et al} 1993) six full simulations of LAGEOS--LAGEOS III data yielded a 7$\\%$--8$\\%$ error.} $1\\%$.  The difference of the perigees would be an independent, less accurate observable. It should be noticed that the practical data reduction of the perigee rates should be performed very carefully in order to account for possible, unpredictable changes in the physical properties of the satellites' surfaces which may occur after some years of their orbital life, as it seems it has happened for LAGEOS II. Such effects may yield a not negligible impact on the response to the direct solar radiation pressure. However, the great experience obtained in dealing with the perigee of LAGEOS II in the LAGEOS--LAGEOS II Lense--Thirring experiment could be fully exploited  for the proposed measurement as well. The obtainable precision for the difference in the perigees should be of the order of $5\\%$.  The combination involving the sum of the nodes with the difference of the perigees would lie at an intermediate level of accuracy\\footnote{It should be considered that such results have been obtained by using the force models and the approximations which have proven to be valid for the existing LAGEOS satellites. The new satellites could be suitably built up in order to reduce the impact of the non-gravitational accelerations with respect to the existing LAGEOS satellites.}.  These estimates are based on the fact that, thanks to the chosen critical inclinations, over an observational time span of about 6 years (2187 days), all the time--dependent harmonic perturbations would complete some full cycles. Then, they could be viewed as empirically fitted quantities to be removed from the analyzed temporal series: this fact should yield further improvements in the error budget. Moreover, the impact of the orbital injection errors on the gravitational systematic error should be probably reduced well below $1\\%$ by the new results from CHAMP and GRACE.  Finally, we must conclude that, although appealing, the use of the alternative observable represented by the difference of the perigees of the proposed CSOC satellites would not yield any significant improvement with respect to the sum of the nodes as far as the detection of the Lense-Thirring effect is concerned. Moreover, the advantages of analyzing only the sum of the nodes of the proposed CSOC satellites with respect to the corresponding observable of the LAGEOS-LARES project would perhaps not justify the expense of the construction and the launch of such entirely new satellites, especially in view of the present-day budget restrictions of many space agencies and of the difficulties already encountered with the LARES."
    },
    "0209/astro-ph0209530_arXiv.txt": {
        "abstract": "An effect of rotation on a developed turbulent stratified convection is studied. Dependencies of the hydrodynamic helicity, the alpha-tensor and the effective drift velocity of the mean magnetic field on the rate of rotation and an anisotropy of turbulent convection are found. It is shown that in an anisotropic turbulent convection the alpha-effect can change its sign depending on the rate of rotation. The evolution of the alpha-effect is much more complicated than that of the hydrodynamic helicity in an anisotropic turbulent convection of a rotating fluid. Different properties of the effective drift velocity of the mean magnetic field in a rotating  turbulent convection are found: (i) a poloidal effective drift velocity can be diamagnetic or paramagnetic depending on the rate of rotation; (ii) there is a difference in the effective drift velocities for the toroidal and poloidal magnetic fields; (iii) a toroidal effective drift velocity can play a role of an additional differential rotation. The above effects and an effect of a nonzero divergence of the effective drift velocity of the toroidal magnetic field on a magnetic dynamo in a developed turbulent stratified convection of a rotating fluid are studied. Astrophysical applications of the obtained results are discussed. ",
        "introduction": "Turbulent transport of particles and magnetic fields was intensively studied for the Navier-Stokes turbulence (see, {\\em e.g.,} \\cite{MY75,ZMR88,ZRS90,F95}). However, there are a number of applications with other kinds of turbulence, {\\em e.g.,} turbulent convection. For instance, in the Sun and stars there is a developed turbulent convection that is strongly influenced by a fluid rotation.  The mean-field theory of magnetic field was in general developed for the Navier-Stokes turbulence without taking into account turbulent convection (see, {\\em e.g.,} \\cite{M78,P79,KR80,ZRS83,RSS88,S89,RS92,R94}). In particular, the dependencies of the the $ \\alpha $-effect, the effective drift velocity and the turbulent magnetic diffusion on the rate of rotation were found only for the Navier-Stokes turbulence (see, {\\em e.g.,} \\cite{K91,RK93,KPR94,RKR00}) in spite of that in many astrophysical applications there are turbulent convection regions. A turbulent convection in different situations has been studied mainly by numerical simulations (see, {\\em e.g.,} \\cite{ZB89,BN90,TB97,BGT98,B2000,OSB01,OSB02}).  In this paper we study an influence of rotation on a developed turbulent stratified convection. This allows us to find the dependencies of the hydrodynamic helicity, the alpha-tensor and the effective drift velocity of the mean magnetic field on the rate of rotation.  This study has a number of applications in astrophysics. In particular, the evolution of the mean magnetic field in the kinematic approximation (without taking into account a two-way coupling of mean magnetic field and turbulent fluid flow) can be described in terms of propagating waves with the growing amplitude, i.e., the magnitude of the mean magnetic field $ B $ is given by \\begin{eqnarray} B \\propto B_{0} \\exp(\\gamma_{_{B}} t) \\cos(\\omega_{_{B}} t + {\\bf k} \\cdot {\\bf r}) \\;, \\label{P1} \\end{eqnarray} where $ B_{0} $ is a seed magnetic field, $ \\gamma_{_{B}} $ is the growth rate of the mean magnetic field, $ \\omega_{_{B}} $ and $ {\\bf k} $ are the frequency and the wave vector of a dynamo wave. In the Sun, {\\em e.g.,} according to the magnetic field observations these dynamo waves with the $ \\sim 22 $ years period propagate to the equator (see, {\\em e.g.,} \\cite{M78,P79,KR80,ZRS83,S89}). The magnetic field is generated in the turbulent convective zone inside the Sun. The growth of the mean magnetic field is a combined effect of a nonuniform fluid rotation (the differential rotation, $ \\bec{\\nabla} (\\delta \\Omega)) $ and helical turbulent motions (the $ \\alpha$-effect). The direction of propagation of the dynamo waves is determined by a sign of the parameter $ \\alpha [\\partial (\\delta \\Omega) / \\partial r] ,$ where $ r, \\theta , \\varphi $ are the spherical coordinates, and $ {\\bf \\Omega} $ is the angular velocity. When the parameter $ \\alpha \\, [\\partial (\\delta \\Omega) / \\partial r] $ is negative the dynamo waves propagates to the equator. The helioseismology shows that in the solar convective zone $ \\partial (\\delta \\Omega) / \\partial r > 0 $ and the existing theories yield $ \\alpha > 0 .$ This results in that the dynamo waves should propagate to the pole in contradiction to the solar magnetic field observations (see, {\\em e.g.,} \\cite{M78,P79,KR80,ZRS83,S89}).  In this study we found that in a developed turbulent convection the $\\alpha$-effect can change its sign depending on the rate of rotation and an anisotropy of turbulence. In the lower part of the solar convective zone the fluid rotation is very fast in comparison with the turnover time of turbulent eddies. In this region $ \\alpha > 0 .$ In the upper part of the solar convective zone the fluid rotation is very slow and $ \\alpha < 0 .$ This explains the observed properties of the solar dynamo waves. The growth of the mean magnetic field is saturated by nonlinear effects (see, {\\em e.g.,} \\cite{KRR94,F99,KR99,RK2000,KMRS2000}). The $ 22 $-years solar magnetic activity is also poorly understood. A characteristic time of the turbulent magnetic diffusion in the solar convective zone is of the order 2-3 years and it cannot explain the characteristic time of solar magnetic activity. We found that the fast rotation causes an additional effective drift velocity of a mean magnetic field that can increase the period of the dynamo waves provides the $ 22 $-years solar magnetic activity. ",
        "conclusions": "In this paper we studied an effect of rotation on a developed turbulent stratified convection. This allowed us to determine the dependencies of the hydrodynamic helicity, the alpha-tensor and the effective drift velocity of the mean magnetic field on the rate of rotation and an anisotropy of turbulence. We demonstrated that in a turbulent convection the alpha-effect can change its sign depending on the rate of rotation and an anisotropy of turbulence. We found different properties of the effective drift velocity of the mean magnetic field in a rotating turbulent convection. In particular, a poloidal effective drift velocity can be diamagnetic or paramagnetic depending on the rate of rotation. There is a difference in the effective drift velocities for the toroidal and poloidal magnetic fields which increases with the rate of rotation. We found also a toroidal effective drift velocity which can play a role of an additional differential rotation.  Some of the results obtained in our paper using the $\\tau$-approximation are observed in the direct numerical simulations of the stratified turbulent convection (see \\cite{OSB02}).  In particular, it was found in \\cite{OSB02} that the alpha-effect can change its sign depending on the rate of rotation. It was also demonstrated in \\cite{OSB02} that there is a difference in the effective drift velocities for the toroidal and poloidal magnetic fields, and that an observed toroidal effective drift velocity in \\cite{OSB02} can play a role of an additional differential rotation.  Now we apply the obtained results for the analysis of an axisymmetric $\\alpha \\Omega$-dynamo. The mean magnetic field in an axisymmetric case is given by $ {\\bf B} = B {\\bf e}_{\\varphi} + \\bec{\\nabla} {\\bf \\times} (A {\\bf e}_{\\varphi}) ,$ where $ A $ is the vector potential. The equations for $ B $ and $ A $ in dimensionless form are given by \\begin{eqnarray} {\\partial B \\over \\partial t} + r_{\\perp} \\, \\bec{\\nabla} {\\bf \\cdot} ( {\\bf V}^{(B)} \\, r_{\\perp}^{-1} \\, B) &=& D \\, (\\hat \\Omega A) + \\Delta_{s} B \\;, \\label{L10} \\\\ {\\partial A \\over \\partial t} + r_{\\perp}^{-1} \\, ({\\bf V}^{(A)} {\\bf \\cdot} \\bec{\\nabla}) \\, (r_{\\perp} \\, A) &=& \\alpha B + \\Delta_{s} A \\;, \\label{L11} \\end{eqnarray} where the length is measured in units of the thickness of the convective zone $ L_{c} ,$ the time is measured in units of $ L_{c}^{2} / \\eta_{_{T}} ,$ the velocity is measured in units of $ \\eta_{_{T}} / L_{c} ,$ the turbulent magnetic diffusion $ \\eta_{_{T}} = l_{0} u_{0} / 3,$ and $u_{0}$ is the characteristic turbulent velocity in the scale $ l_{0} ,$ $ \\,  D = R_\\alpha R_\\omega $ is the dynamo number, $ R_\\alpha = L_{c} \\alpha_\\ast / \\eta_{_{T}} $ and $ R_\\omega = L_{c}^{2} (\\delta \\Omega)_\\ast / \\eta_{_{T}} .$ Here $ \\alpha $ is measured in units of the maximum value $ \\alpha_\\ast $ of the $ \\alpha $ effect, $ (\\delta \\Omega)_\\ast $ is the characteristic differential rotation in the scale $ L_{c} ,$ $ \\hat \\Omega A = [\\bec{\\nabla} (\\delta \\Omega) {\\bf \\times} \\bec{\\nabla} (r_{\\perp} \\, A)] \\cdot {\\bf e}_{\\varphi} ,$ $ \\, \\Delta_{s} = \\Delta - r_{\\perp}^{-2} ,$ $ r_{\\perp} = r \\sin \\theta $ and we used the induction equation for the mean magnetic field (see, {\\em e.g.,} \\cite{M78,P79,KR80,ZRS83,RSS88}) and Eqs. (\\ref{E9}) and (\\ref{E10}). When $ {\\bf V}^{(A)} = {\\bf V}^{(B)} $ and $ \\bec{\\nabla} {\\bf \\cdot} {\\bf V}^{(B)} = 0 ,$ Eqs. (\\ref{L10}) and (\\ref{L11}) coincide with that given in \\cite{M78}. Now we seek for a solution of Eqs. (\\ref{L10}) and (\\ref{L11}) in the form $ A, B \\propto \\exp(\\hat \\gamma t + i {\\bf k} {\\bf \\cdot} {\\bf x}) ,$ where $ {\\bf k} = k {\\bf e}_{k} ,$ $ \\, {\\bf e}_{k} = {\\bf e}_{\\varphi} {\\bf \\times} {\\bf e}_{_{\\Omega}} ,$ the unit vector $ {\\bf e}_{_{\\Omega}} $ is directed opposite to $ \\bec{\\nabla} (\\delta \\Omega) $ and \\begin{eqnarray} \\hat \\gamma &=& \\kappa / 2 - k^{2} - i k U^{(1)} \\nonumber \\\\ & & \\pm [(\\kappa / 2 + i k U^{(3)})^{2} + i k D]^{1/2} \\;, \\label{L12} \\end{eqnarray} $ \\kappa = - \\bec{\\nabla} {\\bf \\cdot} {\\bf V}^{(B)} ,$ $ \\, U^{(1,3)} = {\\bf V}^{(1,3)} \\, {\\bf \\cdot} \\, {\\bf e}_{k} $ and $ \\hat \\gamma = \\gamma_{_{B}} + i \\omega_{_{B}} .$ In the limit of large dynamo number $ | D | $ the maximum growth rate of the mean magnetic field $ \\gamma_{_{B}} $ is given by \\begin{eqnarray} \\gamma_{_{B}} = (3/4) (| D | / 4)^{2/3} + \\kappa / 2 \\;, \\label{L14} \\end{eqnarray} which is achieved at the wave number $ k_{m} = (1/2) (| D | / 4)^{1/3} .$ At this wave number the frequency $ \\omega_{_{B}} $ of the dynamo wave is \\begin{eqnarray} \\omega_{_{B}} = - (| D | / 4)^{2/3}  - (1/2) U^{(1)} (| D | / 4)^{1/3} \\; \\label{L15} \\end{eqnarray} (see \\cite{KRS01}). The negative sign of $ \\omega_{_{B}} $ implies that the dynamo waves propagate to the equator in agreement with the solar magnetic field observations. On the other hand, the divergence of the effective drift velocity $ {\\bf V}^{(B)} $ of the toroidal magnetic field can cause an increase of the growth rate of the mean magnetic field when $ \\kappa > 0 .$ The change of the sign of the $\\alpha$-effect depending on the rate of rotation and anisotropy of turbulent convection (see Section III-B) can explain the observed direction of propagation of the solar dynamo waves.  Note that a meridional circulation in the solar convective zone can also cause an equatorward drift of the solar dynamo wave (see, e.g., \\cite{M78,DC94,CSD95}). However, it was shown recently in \\cite{DBE02} that the meridional velocity, which is required for the equatorward propagation of the solar dynamo wave with the period $ \\sim 22 $ years, should be of the order of $ \\sim 10-12 $ m/s. Such large meridional velocities are not observed on the solar surface. On the other hand, we found that the effective drift velocities of the mean magnetic field have a meridional component (along $ {\\bf e}_\\theta )$. This velocity has the maximum $ (V_\\theta^{(1)})_{\\rm max} \\sim 10-12 $ m/s in the upper part of the solar convective zone. Therefore, this meridional effective drift velocity of the mean magnetic field can cause the equatorward propagation of the solar dynamo wave in the upper part of the solar convective zone. Note that the meridional circulations in the solar convection zone and the meridional component of the effective drift velocities of the mean magnetic field are different characteristics, because the first velocity describes large-scale fluid motions (which may cause advection of the mean magnetic field by the large-scale fluid motions, i.e., by the mean flow), and the second velocity determines the drift velocity of the mean magnetic field (which is originated from the mean electromotive force $ \\bec{\\cal E} = \\langle {\\bf u} \\times {\\bf b} \\rangle ).$  We found also that in the upper part of the solar convective zone the $\\alpha$ effect does not change its sign, i.e., it is positive. But in the lower part of the solar convective zone the $\\alpha$ effect changes its sign, because the parameter $\\Omega \\tau_{_{0}}$ increases with the increase of the depth the solar convective zone, and the $\\alpha$ effect becomes negative. Therefore, in the lower part of the solar convective zone the negative $\\alpha$ effect is responsible for the equatorward propagation of the solar dynamo waves. On the other hand, the meridional effective drift velocity of the mean magnetic field in the lower part of the solar convective zone is very small and, thus, it cannot be used for the explanation of the equatorward propagation of the solar dynamo wave.  Therefore, both effects, the meridional effective drift velocity of the mean magnetic field in the upper part of the solar convective zone and the sign reversal of the $\\alpha$ effect in the lower part of the solar convective zone, can cause the equatorward propagation of the solar dynamo wave.  Note that in the present study we did not discuss the magnetic buoyancy effects which play an important role in a creation of strongly inhomogeneous magnetic structures (see, e.g., \\cite{P79,KMR96,KR94,MFS92,FSS94})."
    },
    "0209/astro-ph0209476_arXiv.txt": {
        "abstract": "We describe an experiment to measure the polarization of the Cosmic Microwave Background (CMB) with the Degree Angular Scale Interferometer (DASI), a compact microwave interferometer optimized to detect CMB anisotropy at multipoles $l \\simeq $~140\\dash900.  The telescope has operated at the Amundsen-Scott South Pole research station since 2000 January.  The telescope was retrofit as a polarimeter during the 2000\\dash2001 austral summer, and throughout the 2001 and 2002 austral winters has made observations of the CMB with sensitivity to all four Stokes parameters.  The telescope performance has been extensively characterized through observations of artificial sources, the Moon, and polarized and unpolarized Galactic sources.  In 271 days of observation, DASI has differenced the CMB fluctuations in two fields to an rms noise level of $2.8\\muK$. ",
        "introduction": "\\footnotetext[1]{Current address: Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109}   The DASI experiment, previously described in~\\markcite{leitch02a}Leitch {et~al.} (2002) (hereafter Paper I), is an interferometric array designed to measure anisotropy in the cosmic microwave background radiation.  The telescope was deployed to the Amundsen-Scott South Pole Station in the 1999\\dash2000 austral summer and made total intensity measurements of the CMB during the 2000 austral winter.  These observations were described in Paper I. The angular power spectrum of the CMB derived from these data were reported by \\markcite{halverson02}{Halverson} {et~al.} (2002) (hereafter Paper II) and the constraints on cosmological parameters derived from the power spectrum were presented by \\markcite{pryke02}{Pryke} {et~al.} (2002) (hereafter Paper III).  During the 2000\\dash2001 austral summer, the DASI receivers were each fitted with broadband achromatic polarizers to allow polarization-sensitive observations of the CMB.  In addition, a large reflecting ground screen was installed to reduce the sensitivity to terrestrial sources of emission. Throughout the 2001 and 2002 austral winters, the telescope has observed the CMB in all four Stokes parameters.  This paper (Paper IV in the continuing series) describes the design of the DASI CMB polarization experiment, the polarization response and calibration of the instrument, and the CMB observations made during the 2001 and 2002 seasons. The analysis of the CMB polarization data obtained is presented by \\markcite{kovac02}Kovac {et~al.} (2002) (hereafter Paper V). ",
        "conclusions": "We have described an experiment to measure polarization in the CMB, using the Degree Angular Scale Interferometer (DASI).  New broadband polarizers were installed in the telescope in 2001, and careful optimization of these polarizers prior to installation resulted in achromatic performance across DASI's frequency band to $\\lesssim1\\%$, as confirmed by astronomical observations.  We have developed techniques which allow us to calibrate the polarimeter end-to-end using only unpolarized sources, and an extensive campaign of observation has resulted in a characterization of the instrumental polarization response to levels well below what is required for a detection of CMB polarization.  Observations of the Moon and various Galactic sources demonstrate that DASI can map degree-scale polarization to high accuracy.  During 2001\\dash2002, the telescope acquired 271 days of data in all four Stokes parameters on two fields identified from previous observations with DASI as containing no detectable point sources.  The data show no evidence for contamination by point sources, polarized or unpolarized, and both the total intensity and polarization data are consistent with a thermal spectrum.  These observations show structure from the CMB detected with an unprecedented \\snr\\ of $\\sim25$."
    },
    "0209/astro-ph0209195_arXiv.txt": {
        "abstract": "Our recent theoretical work (Townsley and Bildsten 2002) on the thermal state of white dwarfs (WDs) in low mass transfer rate binaries allows us to predict the broadband colors of the binary from those of the WD and companion when the disk is dim. The results based on standard CV evolution are presented here. These will aid the discovery of such objects in field surveys and proper-motion selected globular cluster surveys with \\emph{HST}; especially for the largely unexplored post period minimum Cataclysmic Variables (CVs) with the lowest accretion rates and degenerate companions. We have also calculated the fraction of time that the WD resides in the ZZ Ceti instability strip thus clarifying that we expect many accreting WDs to exhibit non-radial oscillations.  The study of these will provide new insights into the rotational and thermal structure of an actively accreting WD. ",
        "introduction": "As summarized elsewhere in these proceedings our recent work has demonstrated that at the low accretion rates appropriate for dwarf novae (DN), the core temperature, $T_c$, of the WD is \\emph{set} by the long-time average accretion rate, $\\langle \\dot M\\rangle$, of the binary \\cite{TownBild02}.  With $T_c$ set, we determine the WD luminosity, which depends on $\\langle \\dot M\\rangle$, $M_{\\rm WD}$, and the mass of the freshly accreted layer.  This luminosity is directly observable for DN when the disk is in quiescence and the surface of the WD has cooled after outburst. ",
        "conclusions": ""
    },
    "0209/cond-mat0209043_arXiv.txt": {
        "abstract": "A self-consistent approach to nonequilibrium radiation temperature is introduced using the distribution of the energy over states. We begin rigorously with ensembles of Hilbert spaces and end with practical examples based mainly on the far from equilibrium radiation of lasers. We show that very high, but not infinite, laser radiation temperatures depend on intensity and frequency. Heuristic ``temperatures'' derived from a misapplication of equilibrium arguments are shown to be incorrect. More general conditions for the validity of nonequilibrium temperatures are also established. \\\\ \\centerline{Original: August 26, 2002} \\centerline{Final: June 8, 2003} \\centerline{Published in {\\bf American Journal of Physics}} \\centerline{Vol. 71, No. 10, October 2003, pp. 969-978} \\centerline{PACS numbers: 05.70.Ln, 42.50.-p, 42.50.Ar, 42.55.Ah} ",
        "introduction": "\\label{intro}  The standard definitions of intensive thermodynamic parameters, such as temperature, seem to require the system in question to be in thermodynamic equilibrium. In this paper, we explore the entropy and temperature of radiation out of equilibrium and show that, within limited restrictions which do not require equilibrium, the radiation temperature is well-defined and distinct from any associated matter temperature and from plausible but incorrectly applied equilibrium definitions. We use laser radiation as our major example.  Laser radiation is a fascinating example of a highly organized quantum system of quasi-coherent bosons.\\cite{sargent74,silf96,mandel95} A laser beam is supported by external pumping, which keeps the beam far from thermodynamic equilibrium. The laser shares this feature with other steady-state systems that are kept from equilibrating by external constraints. We find that the temperature of laser radiation far exceeds the temperatures of the laser cavity and the lasing atomic transition. Misidentifying the radiation with the matter temperature leads to erroneous estimates of the laser radiation temperature that are as much as ten orders of magnitude too small.  Photon number, unlike energy, is not conserved. The Gibbs-Duhem relation for radiation, $SdT - VdP = 0$, implies that the two intensive thermodynamic parameters, pressure $P$ (conjugate to volume) and temperature $T$ (conjugate to energy), reduce to one independent intensive parameter, which is usually identified as $T$. This feature of radiation thermodynamics, like the photon's zero mass and lack of rest frame, makes radiation thermodynamics much simpler than that of matter, which has conserved particle numbers and nonzero chemical potentials.\\cite{essex99,reichl98} It also makes generalizing intensive thermodynamic parameters out of equilibrium much easier. Thus radiation is a natural context in which to introduce nonequilibrium temperature.  A properly defined nonequilibrium temperature has physical meaning.  It occurs in the {\\it entropy production rate} $\\Sigma$, an important measure of both how far a system is from equilibrium and how fast it is approaching equilibrium.\\cite{degroot62} $\\Sigma$ has a generic form rooted in the equilibrium expression for the entropy differential, $dS = dQ/T$, where $dQ$ is the differential of heat or random energy, the change in system energy while holding volume and particle number constant.  The form $\\Sigma\\sim J_Q(12)(1/T_1 - 1/T_2)$ expresses the entropy produced by two subsystems (1,2) at temperatures $T_1, T_2$ as they exchange a heat flux $J_Q(12)$. Subsystem temperatures occur naturally in expressions for entropy production. $\\Sigma$ is positive semidefinite and vanishes if and only if $T_1 = T_2$, the condition for thermal equilibrium. The heat flux $J_Q(12)$ vanishes in this case as well.  The difference $1/T_1 - 1/T_2$ of the inverse temperatures is a measure of how far out of equilibrium the two subsystems are. The heat flux $J_Q(12)$ is a measure of how fast the subsystems are approaching equilibrium with one another, assuming no external pumping of the system. (With external pumping, $J_Q(12)$ is a measure of how much power has to be injected into the system to keep it from equilibrating.) The product of these two quantities, given by $\\Sigma$, combines the two measures into a single quantity characteristic of a nonequilibrium process.  The books by Reichl\\cite{reichl98} and De Groot and Mazur\\cite{degroot62} explain in detail the significance and role of entropy production in nonequilibrium matter systems. Section~\\ref{otherphysconseq} and Refs.~\\onlinecite{essex99} and \\onlinecite{bludman97} explore entropy production in radiation and radiation-matter systems. ",
        "conclusions": ""
    },
    "0209/astro-ph0209181_arXiv.txt": {
        "abstract": "We discuss the amplification dispersion in the observed luminosity of standard candles, like supernovae (SNe) of type Ia, induced by gravitational lensing in a Universe with dark energy (quintessence). We derive the main features of the magnification probability distribution function (pdf) of SNe in the framework of on average Friedmann-Lema\\^{\\i}tre-Robertson-Walker (FLRW) models for both lensing by large-scale structures and compact objects. Analytic expressions, in terms of hypergeometric functions, for luminosity distance--redshift relations in a flat Universe with homogeneous dark energy have been corrected for the effects of inhomogeneities in the pressureless dark matter (DM). The magnification pdf is strongly dependent on the equation of state, $w_Q$, of the quintessence. With no regard to the nature of DM (microscopic or macroscopic), the dispersion increases with the redshift of the source and is maximum for dark energy with very large negative pressure; the effects of gravitational lensing on the magnification pdf, i.e. the mode biased towards de-amplified values and the long tail towards large magnifications, are reduced for both microscopic DM and quintessence with an intermediate $w_Q$. Different equations of state of the dark energy can deeply change the dispersion in amplification for the projected observed samples of SNe Ia by future space-born missions. The ``noise\" in the Hubble diagram due to gravitational lensing strongly affects the determination of the cosmological parameters from SNe data. The errors on the pressureless matter density parameter, $\\Omega_M$, and on $w_Q$ are maximum for quintessence with not very negative pressure. The effect of the gravitational lensing is of the same order of the other systematics affecting observations of SNe Ia. Due to the lensing by large-scale structures, in a flat Universe with $\\Omega_M =0.4$, at $z=1$ a cosmological constant ($w_Q=-1$) can be interpreted as dark energy with $w_Q <-0.84$ (at $2$-$\\sigma$ confidence limit). ",
        "introduction": "During the last years, two independent groups, the High-$z$ SuperNova Search Team \\cite{sm&al98} and the Supernova Cosmology Project \\cite{pe&al99} have given strong evidences of the acceleration of the Universe's expansion \\cite{ri&al98,pe&al99}. Several other observational and theoretical evidences, like  measurements of the anisotropy of the Cosmic Microwave Background Radiation \\cite{deb&al00} and inflationary theories, strongly support a flat or nearly flat Universe. On the other hand, direct measurements of $\\Omega_M$ from dynamical estimates or X-ray and lensing observations of clusters of galaxies indicates that $\\Omega_M$ is significantly less than unity, $\\Omega_M \\simeq 0.3$ \\cite{tur00}. To solve this puzzle, a new type of energy component in the Universe, now called dark energy or quintessence, was proposed; dark energy with strongly negative pressure is required to explain acceleration ($w_Q\\equiv p_Q /\\rho_Q <-1/3$, where $p_Q$ and $\\rho_Q$ are, respectively, the pressure and energy density of the dark energy).  Observations of SNe Ia, which at low redshifts are sensitive to the deceleration parameter $q_0=(\\Omega_M + (1+3w_Q)\\Omega_Q)/2$ where $\\Omega_Q$ is the density parameter of the dark energy, rely on several properties of these sources. SNe Ia are very luminous and have a small intrinsic dispersion in their peak absolute magnitude, $\\delta M \\stackrel{<}{\\sim} 0.3$ \\cite{fi&ri99}. These features make them the long expected standard candles for cosmology.  A standard candle is a source with known intrinsic luminosity ($L$) (or absolute magnitude). Measurements of its apparent flux (${\\cal{F}}$) allow us to determine the photometric distance $D_L$ to the source via equation \\begin{equation} \\label{eq:1} D_L \\equiv \\sqrt{\\frac{L}{4\\pi {\\cal{F}}}}. \\end{equation} Using standard candles, it is possible to plot the Hubble diagram (that is, the redshift of an object versus cosmological distance to it or vice versa) with very high precision and estimate the global cosmological parameters.  There are several candidates for the dark energy. The oldest one, initially introduced by Albert Einstein as a new fundamental constant of nature, is the cosmological constant ($w_Q=-1$). After the formulation of inflationary theory, cosmologists found that a $\\Lambda$ term can be introduced dynamically \\cite{dol90,zel92,sa&st00}; a dynamical $\\Lambda$ term by a scalar field slowly rolling down its potential ($w_Q \\geq -1$) \\cite{pe+ra88,wett88,os&al95,ca&al98,zl&al99,rit&al00,rub00,ru&sc01} can support a static energy component with positive energy density but negative pressure. Other possibilities for the quintessence are represented either by networks of light, non-intercommuting topological defects \\cite{vi84,sp+pe97} ($w_Q =-m/3$, where $m$ is the dimension of the defect: for a string  $w_Q=-1/3$; for a domain wall $w_Q=-2/3$) or by the so called $X$-matter \\cite{ch&al97,tu&wh97}. Alternatively to quintessence, a Universe filled with Chaplygin gas \\cite{ka&al01} is an additional alternative to obtain a negative pressure. Generally, the equation of state $w_Q$ evolves with the redshift, and the feasibility of reconstructing its time evolution has been investigated \\cite{co&hu99,ch&na00,ma&al00,sa&al00,gol+al01,hu&tu01,na&ch01,pa+al01,wa&ga01,wa+lo01,we+al01a,we+al01b,ya&fu01,ge+ef02}. Since in flat FLRW models the distance depends on $w_Q$ only through a triple integral on the redshift \\cite{ma&al00}, $w_Q (z)$ can be determined only provided a prior knowledge of the matter density of the Universe \\cite{gol+al01}. In what follows, we will consider only the case of a constant equation of state.  Astrophysical sources other than SNe have been long investigated to build the Hubble diagram. Two independent luminosity estimators, the first one based on the variability of Gamma-Ray Bursts (GRBs) \\cite{rei01a,rei01b} and the second one derived from the time lag between peaks in hard and soft energies \\cite{nor+al00}, have been recently proposed to infer the luminosity distance to these sources. On the other hand, standard rods, as compact radio sources \\cite{gur+al99} or double radio galaxies \\cite{gue+al00}, have been long studied to evaluate the angular diameter distance to cosmological sources. With no regard to their different physical origins, all these observations are affected by gravitational lensing of the sources. In this paper, we want to quantify the effect of inhomogeneities in the pressureless matter on the determination of the distance. In particular, we will study SNe Ia, whose importance in the determination of the cosmological parameters makes necessary a complete study of all systematics. For light beams propagating in the inhomogeneous Universe, the expression of the luminosity distance in terms of the cosmological parameters, as obtained from Eq.~(\\ref{eq:1}), changes with respect to the corresponding FLRW model. Mathematical considerations for non-flat models of Universe are done in Sereno et al. \\shortcite{ser+al01}. In this paper, we will discuss the simple case of the flat Universe and the influence of clumpiness on the Hubble diagram. We consider inhomogeneous pressureless matter and smooth dark energy. For the more general case of inhomogeneous quintessence see Linder \\shortcite{li88} and Sereno et al. \\shortcite{ser+al01}.  The paper is as follows. In Sect. 2, we introduce the on average FLRW models; we discuss the Dyer-Roeder (DR) equation and its analytical solution in terms of hypergeometric functions. In Sect. 3, we consider the case of the homogeneous Universe. Section 4 contains the discussion on the statistical nature of the lensing dispersion induced either by large-scale structures or compact objects; we study the magnification pdf induced by gravitational lensing and the connected systematic errors in the estimate of $\\Omega_M$ and $w_Q$. Section 5 is devoted to some final considerations. ",
        "conclusions": "Observations of SNe Ia are strongly affected by inhomogeneities in the Universe. For redshifts $z \\stackrel{>}{\\sim} 1$, the variation in the distance modulus from a standard flat FLRW model to a clumpy Universe with the same content of pressureless matter can be considerably greater than other systematic effects. The effect of amplification dispersion by gravitational lensing must be accurately considered. The prospects of future space-born missions, like the SuperNova Acceleration Probe (SNAP - Http://snap.lbl.gov) and the Next Generation Space Telescope, of determining properties of the dark energy have been discussed \\cite{gol+al01,we+al01a,we+al01b,ge+ef02}. According to these studies, SNAP data should only distinguish between a cosmological constant and quintessence with $w_Q$ relatively far from $-1$. When SNe observations are combined with an independent estimate of $\\Omega_M$, for example from galaxy clustering \\cite{ver+al01}, the degeneracies among the quintessence models can be significantly reduced and some constraints on the time evolution of the equation of state can be put \\cite{we+al01b,ge+ef02}. However, these studies only consider measurement errors and intrinsic dispersion of the sources, neglecting the systematic and redshift dependent error induced by gravitational lensing. We have shown how, also assuming an exact knowledge of $\\Omega_M$, in the redshift range covered by future missions a cosmological constant can be interpreted as dark energy with $w_Q>-1$. For $\\Omega_M=0.4$ and $z=1$, a $\\Lambda$ constant may be interpreted as quintessence with $w_Q <-0.84$, only due to the lensing by large-scale structure. A fraction of DM in form of compact objects will make the situation even more dramatic. So, also with a prior knowledge of the remaining cosmological parameters, gravitational lensing can make the statements on the properties of dark energy based on SNe data significantly less certain.  The effect of inhomogeneities dominates at high redshifts and should be one of the main systematics in attempting to build the Hubble diagram with GRBs \\cite{nor+al00,rei01a,rei01b,sch+al01}. The physical origin of GRBs is still uncertain, but recent observations suggest that they are related to the violent death of massive stars. Under the hypothesis that GRBs trace the global star formation history of the Universe, their assumed rate is strongly dependent on the expected evolution of the star formation rate with the redshift \\cite{po+ma01}. While some scenarios prefer a redshift distribution of the GRB comoving rate peaked between $z=1$ and $2$, according to other ones the comoving rate remains roughly constant at $z\\stackrel{>}{\\sim} 2$ and out to very high redshift \\cite{po+ma01}. Furthermore, the lack of strong lensing events in the fourth BATSE GRBs catalog \\cite{hol+al99} suggests that, at the $95\\%$ confidence level, the upper limit to the average redshift of GRBs is $\\stackrel{<}{\\sim} 3$ in a flat, low-matter density Universe with cosmological constant. According to these considerations, the effect of gravitational lensing would be really dominant in the Hubble diagram built with GRBs.  As an example, we consider the GRB redshift distribution derived from a combined analysis of two independent luminosity indicators \\cite{sch+al01}. Examining a sample of 112 GRBs from the BATSE catalog, Schaefer et al. \\shortcite{sch+al01} found redshifts varying between $0.25$ and $5.9$ with a median of $1.5$. At $z=1.5$, gravitational lensing by large-scale structures, in a model with $\\Omega_M =0.4$ and $w_Q=-2/3$, induces a magnification distribution with $\\mu_{peak}=1.25$, $\\mu_{low}=1.20$ and $\\mu_{high}=1.46$. Assuming $w_Q=-2/3$, we will estimate $\\Omega_M=0.43^{+0.05}_{-0.16}$; assuming $\\Omega_M =0.4$, we will estimate $w_Q<-0.51$.  Although the lensing dispersion on the luminosities of standard candles represents a noise in the determination of the cosmological parameters, it can also be considered as a probe of the clustering properties of the DM. Lensing dispersion has been investigated to search for the presence of compact objects in the Universe \\cite{lin+al88,rauc91,me+si99,se+ho99}. The possibility of determining the fraction of macroscopic DM using future samples of SNe Ia has also been explored \\cite{mor+al01}. SNAP should intensively observe SNe up to $z \\sim 1.7$. In one year of study, this space-born mission should be able to discover $\\sim 2350$ SNe, most of which in the region $0.5 \\stackrel{<}{\\sim} z \\stackrel{<}{\\sim} 1.2$. The discrimination of models of Universe with different fractions of compact objects is mainly based on the shift in the peak of the lensing dispersion \\cite{se+ho99,mor+al01}: a shift of $\\sim 0.01$ mag in the peak of the lensing dispersion in the projected SNAP sample towards lower amplifications corresponds to a growth of $20\\%$ in the fraction of macroscopic DM in a flat Universe with $\\Omega_M = 0.3$ and a cosmological constant (see figure (4) in M\\\"{o}rtsell et al. \\shortcite{mor+al01}). In Fig.~(\\ref{mu_pdf_compact_snap}), we plot the dispersion in amplification, for the projected redshift distribution of SNe according to the SNAP proposal, in a Universe with $\\Omega_M=0.3$ filled in with macroscopic DM . High de-amplification are preferred in the case of a cosmological constant, when the maximum of the distribution is depleted and the mode is shifted away from the mean with respect to dark energy with $w_Q>-1$. Changing from $w_Q =-1$ to $w_Q=-1/2$, the peak of the distribution moves for $\\sim 0.015$ mag towards higher amplifications. So, a significant reduction in the fraction of compact object can be mimed by quintessence with $w_Q >-1$. Since quintessence reduces the dispersion of gravitational lensing, it also reduces the ability to distinguish between microscopic and macroscopic DM from the shape of the amplification dispersion. Both quintessence and microscopic DM reduce the bias towards the empty beam value and the high magnification tail and their effect is of the same order. A Universe with an high fraction of macroscopic objects can be misleadingly interpreted as one with dark energy with large negative pressure."
    },
    "0209/astro-ph0209462_arXiv.txt": {
        "abstract": "Following the recent ideas of Dryzek, Kato, Mu\\~noz and Singleton \\cite{R1} (henceforth we call this paper as R1), that the composite system consisting of an electron along with its screened electric field and within the sphere of influence, the trapped  magnetic field of white dwarf can behave like a boson, we have argued that such an exotic transition (bosonization) of electronic component in strongly magnetized  neutron star matter in $\\beta$-equilibrium can make the existence of magnetars physically viable. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209524_arXiv.txt": {
        "abstract": "In this talk we present a model of the universe in which dark energy is modelled explicitely with both a dynamical quintessence field and a cosmological constant. Our results confirm the possibility of a collapsing universe (for a given region of the parameter space), which is advantageous  for an adequate formulation  of both perturbative quantum field and string theories. We have also reproduced the measurements of modulus distance from supernovae with good accuracy. ",
        "introduction": "From 1998 to date several important discoveries in the astrophysical sciences have being made, which have given rise to the so called New Cosmology \\cite{turner1,turner2}. Amongst its more important facts we may cite:  - Flat, critical density accelerating universe  - Early period of rapid expansion (inflation)  - Density inhomogeneities produced from quantum fluctuations during inflation  - Composition: 2/3 dark energy; 1/3 dark matter; 1/200 brigh stars  - Matter content: $(29\\pm 4)\\%$ nonbarionic dark matter; $(4\\pm 1)\\%$ baryons, $(0.1-5)\\%$ neutrinos  - $T_0=2.275\\pm 0.001 K$  - $t_0=14\\pm 1 Gyr$  - $H_0=72\\pm 7 km.s^{-1} Mpc^{-1}$  It is a fact that the standard LCDM model, though rather simple from the theoretical point of view, can accomodate most of today's astrophysical data. However, it still has several open questions (see, for instance, \\cite{lahav,pr}), one of them being: could other (yet unknown) models fit the data equally well?  Alternative models should obviously consider the main components of the universe. There is much concern about unveiling the dark part of our universe, which implies that we don't lack candidates. So, for dark matter we have neutrinos, axions, neutralinos; for dark energy: the cosmological constant, scalar fields (for example, quintessence), cosmic field defects, etc. So far, most models of dark energy have a rather phenomenological character, though a few proposals concerning the possible role of the dark energy field in the context of fundamental physics have appeared \\cite{fhsw,wetterich,abrs,gasperini,gpv,pietroni}.  The recent dark energy scalar field research has several interesting features (see \\cite{pr} for an extense review). Many models have attractor or tracker behaviour, allowing, for a wide range of initial  conditions, a subdominant field energy density at high redshifts (radiation and matter  dominated eras).  In the simplest versions, scalar fields models of dark energy have a scalar field kinetic term, and the scalar field is coupled only to itself and gravity. So, the scalar field part of the model is fully characterized by the scalar field potential, with some broad constraints on the initial conditions for the field, if the attractor behaviour is realized.  Many different potentials have being used (see reviews \\cite{pr,sahni}): \\begin{table*}[tbh!] \\begin{center} \\begin{tabular}{lll} \\hline Quintessence Potential & Reference\\\\\\hline & \\\\ $V_0\\exp{(-\\lambda\\phi)}$ & Ratra \\& Peebles (1988), Wetterich (1988), \\\\ & Ferreira \\& Joyce (1998)\\\\ & \\\\ $m^2\\phi^2, \\lambda\\phi^4$ &  Frieman et al (1995)\\\\ & \\\\ $V_0/\\phi^\\alpha, \\alpha > 0$ &  Ratra \\& Peebles (1988) \\\\ & \\\\ $V_0\\exp{(\\lambda\\phi^2)}/\\phi^\\alpha, \\alpha > 0$ & Brax \\& Martin (1999,2000)\\\\ & \\\\ $V_0(\\cosh{\\lambda\\phi} - 1)^p$, & Sahni \\& Wang (2000)\\\\ & \\\\ $V_0 \\sinh^{-\\alpha}{(\\lambda\\phi)}$, & Sahni \\& Starobinsky (2000), Ure\\~{n}a-L\\'{o}pez \\& Matos (2000)\\\\ & \\\\ $V_0(e^{\\alpha\\kappa\\phi} + e^{\\beta\\kappa\\phi})$ & Barreiro, Copeland \\& Nunes ( 2000)\\\\ & \\\\ $V_0(\\exp{M_p/\\phi} - 1)$, & Zlatev, Wang \\& Steinhardt (1999)\\\\ & \\\\ $V_0[(\\phi - B)^\\alpha + A]e^{-\\lambda\\phi}$, & Albrecht \\& Skordis (2000)\\\\  & \\\\ \\hline \\end{tabular} \\caption{} \\end{center} \\end{table*}  In this talk I want to call the attention to exponential potentials, which have being often discarded on fine tunig arguments or (the simplest exponential) because they can not produce the wanted transition from subdominant to dominant energy density (\\cite{pr}). However, as shown in \\cite{rs,rse}, they have proved useful in describing several features in the history of the universe, from radiation decoupling to nowadays. Also, several authors have recently pointed out that the degree of fine tuning needed in these scenarios is no more than in others usually accepted \\cite{cline,kl,rs}.  Especially interesting results are obtained if we model dark energy using both a scalar field and a cosmological constant.  The cosmological constant can be incorporated into the quintessence potential as a constant which shifts the potential value, especially, the value of the minimum of the potential, where the quintessence field rolls towards. Conversely, the height of the minimum of the potential can also be regarded as a part of the cosmological constant. Usually, for separating them, the possible nonzero height of the minimum of the potential is incorporated into the cosmological constant and then set to be zero. The cosmological constant can be provided by various kinds of matter, such as the vacuum energy of quantum fields and the potential energy of classical fields and may also be originated in the intrinsic geometry. So far there is no sufficient reason to set the cosmological constant (or the height of the minimum of the quintessence potential) to be zero, especially when the ultimate fate of our universe is more sensitive to the presence of the cosmological constant (or the nonzero height of the minimum of the quintessence potential) than any other matter content, even though the cosmological constant may be extremely tiny and undetectable at all in present time (\\cite{hwang}. In particular, some mechanisms to generate a negative cosmological constant have been pointed out, in the context of spontaneous symmetry breaking \\cite{ss,gh}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209238_arXiv.txt": {
        "abstract": "{We present a comprehensive spectroscopic study of the integrated light of metal-rich Galactic globular clusters and the stellar population in the Galactic bulge. We measure line indices which are defined by the Lick standard system and compare index strengths of the clusters and Galactic bulge. Both metal-rich globular clusters and the bulge are similar in most of the indices, except for the CN index. We find a significant enhancement in the CN$/\\langle$Fe$\\rangle$ index ratio in metal-rich globular clusters compared with the Galactic bulge. The mean iron index $\\langle$Fe$\\rangle$ of the two metal-rich globular clusters NGC~6528 and NGC~6553 is comparable with the mean iron index of the bulge. Index ratios such as Mgb$/\\langle$Fe$\\rangle$, Mg$_2/\\langle$Fe$\\rangle$, Ca4227$/\\langle$Fe$\\rangle$, and TiO$/\\langle$Fe$\\rangle$, are comparable in both stellar population indicating similar enhancements in individual elements which are traced by the indices. From the globular cluster data we fully empirically calibrate several metallicity-sensitive indices as a function of [Fe/H] and find tightest correlations for the Mg$_2$ index and the composite [MgFe] index. We find that all indices show a similar behavior with galactocentric radius, except for the Balmer series, which show a large scatter at all radii. However, the scatter is entirely consistent with the cluster-to-cluster variations in the horizontal branch morphology.} ",
        "introduction": "Stars in globular clusters are essentially coeval and -- with very few exceptions -- have all the same chemical composition, with only few elements breaking the rule. As such, globular clusters are the best approximation to {\\it simple stellar populations} (SSP), and therefore offer a virtually unique opportunity to relate the integrated spectrum of stellar populations to age and chemical composition, and do it in a fully empirical fashion. Indeed, the chemical composition can be determined via high-resolution spectroscopy of cluster stars, the age via the cluster turnoff luminosity, while integrated spectroscopy of the cluster can also be obtained without major difficulties. In this way, empirical relations can be established between integrated-light line indices \\citep[e.g. Lick indices as defined by][]{faber85} of the clusters, on one hand, and their age and chemical composition on the other hand (i.e., [Fe/H], [$\\alpha$/Fe], etc.).  These empirical relations are useful in two major applications: 1) to directly estimate the age and chemical composition of unresolved stellar populations for which integrated spectroscopy is available (e.g. for elliptical galaxies and spiral bulges), and 2) to provide a basic check of population synthesis models.  Today we know of about 150 globular clusters in the Milky Way \\citep{harris96}, and more clusters might be hidden behind the high-absorption regions of the Galactic disk. Like in the case of many elliptical galaxies \\citep[e.g.][]{harris01}, the Galactic globular cluster system shows a bimodal metallicity distribution \\citep{freeman81,zinn85,ashman98,harris01} and consists of two major sub-populations, the metal-rich bulge and the metal-poor halo sub-populations.  The metal-rich ($\\mathrm{[Fe/H]}>-0.8$ dex) component was initially referred to as a ``disk'' globular cluster system \\citep{zinn85}, but it is now clear that the metal-rich globular clusters physically reside inside the bulge and share its chemical and kinematical properties \\citep{minniti95,barbuy98,cote99}. Moreover, the best studied metal-rich clusters (NGC 6528 and NGC 6553) appear to have virtually the same old age as both the halo clusters and the general bulge population \\citep{ortolani95n, feltzing00, ortolani01, zoccali01, zoccali02, feltzing02}, hence providing important clues on the formation of the Galactic bulge and of the whole Milky Way galaxy.  Given their relatively high metallicity (up to $\\sim Z_\\odot$), the bulge globular clusters are especially interesting in the context of stellar population studies, as they allow comparisons of their spectral indices with those of other spheroids, such as elliptical galaxies and spiral bulges. However, while Lick indices have been measured for a representative sample of metal-poor globular clusters \\citep{burstein84, covino95, trager98}, no such indices had been measured for the more metal-rich clusters of the Galactic bulge. It is the primary aim of this paper to present and discuss the results of spectroscopic observations of a set of metal-rich globular clusters that complement and extend the dataset so far available only for metal-poor globulars.  Substantial progress has been made in recent years to gather the complementary data to this empirical approach: i.e. ages and chemical composition of the metal-rich clusters. Concerning ages, HST/WFPC2 observations of the clusters NGC~6528 and NGC~6553 have been critical to reduce to a minimum and eventually to eliminate the contamination of foreground disk stars (see references above), while HST/NICMOS observations have started to extend these studies to other, more heavily obscured clusters of the bulge \\citep{ortolani01}.  High spectral-resolution studies of individual stars in these clusters is still scanty, but one can expect rapid progress as high multiplex spectrographs become available at 8--10m class telescopes. A few stars in NGC~6528 and NGC~6553 have been observed at high spectral resolution, but with somewhat discrepant results. For NGC~6528, \\cite{carretta01} and \\cite{coelho01} report respectively [Fe/H]$ =+0.07$ and $-0.5$ dex (the latter value coming from low-resolution spectra). For [M/H] the same authors derive $+0.17$ and $-0.25$ dex, respectively. For NGC~6553 \\cite{barbuy99} give [Fe/H]$ =-0.55$ dex and [M/H]$ =-0.08$ dex, while \\cite{cohen99} report [Fe/H]$ =-0.16$ dex, and \\cite{origlia02} give [Fe/H]$ =-0.3$ dex, with [$\\alpha$/Fe]$ =+0.3$ dex. Some $\\alpha$-element enhancement has also been found among bulge field stars, yet with apparently different element-to-element ratios \\citep{mcwilliam94}.  Hopefully these discrepancies may soon disappear, as more and better quality high-resolution data are gathered at 8--10m class telescopes. In summary, the overall metallicity of these two clusters (whose color magnitude diagrams are virtually identical, \\citealt{ortolani95n}) appears to be close to solar, with an $\\alpha$-element enhancement [$\\alpha$/Fe] $\\simeq +0.3$ dex.  The $\\alpha$-element enhancement plays an especially important role in the present study. It is generally interpreted as the result of most stars having formed rapidly (within less than, say $\\sim 1$ Gyr), thus having had the time to incorporate the $\\alpha$-elements produced predominantly by Type II supernovae, but failing to incorporate most of the iron produced by the longer-living progenitors of Type Ia supernovae. Since quite a long time, an $\\alpha$-element enhancement has been suspected for giant elliptical galaxies, inferred from the a comparison of Mg and Fe indices with theoretical models \\citep{peletier89,worthey92, davies93, greggio97}. This interpretation has far-reaching implications for the star formation timescale of these galaxies, with a fast star formation being at variance with the slow process, typical of the current hierarchical merging scenario \\citep{thomas99}. However, in principle the apparent $\\alpha$-element enhancement may also be an artifact of some flaws in the models of synthetic stellar populations, especially at high metallicity \\citep{maraston01}. The observations presented in this paper are also meant to provide a dataset against which to conduct a direct test of population synthesis models, hence either excluding or straightening the case for an $\\alpha$-element enhancement in elliptical galaxies. This aspect is extensively addressed in an accompanying paper \\citep{maraston02}.  The main goal of this work is the measurement of the Lick indices for the metal-rich globular clusters of the bulge and of the bulge field itself. Among others, we measure line indices of Fe, Mg, Ca, CN, and the Balmer series which are defined in the Lick standard system \\citep{worthey97,trager98}. In \\S2 we describe in detail the observations and our data reduction which leads to the analysis and measurement of line indices in \\S3. Index ratios in globular clusters and the bulge are presented in \\S4. Index-metallicity relations are calibrated with the new data in \\S5 and \\S6 discusses the index variations as a function of galactocentric radius. \\S7 closes this work with the conclusions followed by a summary in \\S8.  \\begin{table*}[t!] \\centering \\caption{General properties of sample Globular Clusters. If not else mentioned, all data were taken from the 1999 update of the McMaster catalog of Milky Way Globular Clusters \\citep{harris96}. $R_{\\rm gc}$ is the globular cluster distance from the Galactic Center. $r_h$ is the half-light radius. E$_{(B-V)}$ and $(m-M)_V$ are the reddening and the distance modulus. $v_{\\rm rad}$ the heliocentric radial velocity. Note, that our radial-velocity errors are simple {\\it internal} errors which result from the fitting of the cross-correlation peak. The real {\\it external} errors are a factor $\\sim3-4$ larger. HBR is the horizontal-branch morphology parameter \\citep[e.g.][]{lee94}.} \\label{tab:gcprop} \\begin{tabular}{l|cccccrrr} \\hline \\noalign{\\smallskip} GC &$R_{\\rm gc}$ [kpc] & [Fe/H] &$r_h$ [arcmin] &E$_{(B-V)}^{\\mathrm{a}}$ &$(m-M)_V$ &$v_{\\rm rad}^{\\mathrm{b}}$ [km s$^{-1}$] &$v_{\\rm rad}$ [km s$^{-1}$] & HBR$^{\\mathrm{c}}$ \\\\ \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} NGC 5927      & 4.5&$-0.37$&1.15&0.45& 15.81&$ -130\\pm12$&$-107.5\\pm 1.0$ & $-1.00^d$ \\\\ NGC 6218 (M12)& 4.5&$-1.48$&2.16&0.40& 14.02&$  -46\\pm23$&$ -42.2\\pm 0.5$ & $ 0.97^d$ \\\\ NGC 6284      & 6.9&$-1.32$&0.78&0.28& 16.70&$    8\\pm16$&$  27.6\\pm 1.7$ & $ 1.00^e$ \\\\ NGC 6356      & 7.6&$-0.50$&0.74&0.28& 16.77&$   35\\pm12$&$  27.0\\pm 4.3$ & $-1.00^d$ \\\\ NGC 6388      & 4.4&$-0.60$&0.67&0.40& 16.54&$   58\\pm10$&$  81.2\\pm 1.2$ & $-0.70^e$ \\\\ NGC 6441      & 3.5&$-0.53$&0.64&0.44& 16.62&$  -13\\pm10$&$  16.4\\pm 1.2$ & $-0.70^f$ \\\\ NGC 6528      & 1.3&$-0.17$&0.43&0.56& 16.53&$  180\\pm10$&$ 184.9\\pm 3.8$ & $-1.00^d$ \\\\ NGC 6553      & 2.5&$-0.34$&1.55&0.75& 16.05&$  -25\\pm16$&$  -6.5\\pm 2.7$ & $-1.00^d$ \\\\ NGC 6624      & 1.2&$-0.42$&0.82&0.28& 15.37&$   27\\pm12$&$  53.9\\pm 0.6$ & $-1.00^d$ \\\\ NGC 6626 (M28)& 2.6&$-1.45$&1.56&0.43& 15.12&$  -15\\pm15$&$  17.0\\pm 1.0$ & $ 0.90^d$ \\\\ NGC 6637 (M69)& 1.6&$-0.71$&0.83&0.16& 15.16&$    6\\pm12$&$  39.9\\pm 2.8$ & $-1.00^d$ \\\\ NGC 6981 (M72)&12.9&$-1.40$&0.88&0.05& 16.31&$ -360\\pm18$&$-345.1\\pm 3.7$ & $ 0.14^d$ \\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\begin{list}{}{} \\item[$^{\\mathrm{a}}$] taken from \\cite{harris96} \\item[$^{\\mathrm{b}}$] this work \\item[$^{\\mathrm{c}}$] horizontal branch parameter, (B$-$R)/(B+V+R), for details see e.g. \\cite{lee94} \\item[$^{\\mathrm{d}}$] taken from \\cite{harris96} \\item[$^{\\mathrm{e}}$] taken from \\cite{zoccali00} \\item[$^{\\mathrm{f}}$] Due to very similar HB morphologies in CMDs of NGC~6388 and NGC~6441 \\citep[see][]{rich97}, we assume that the HBR parameter is similar for both globular clusters and adopt HBR$=-0.70$ for NGC~6441. \\end{list} \\end{table*} ",
        "conclusions": "\\label{ln:conclusions} For the first time the complete set of Lick indices have been measured for a sample of metal-rich globular clusters belonging to the Galactic bulge. In combination with data for metal-poor globular clusters this data set has allowed us to establish an empirical calibration of the Lick indices of old stellar populations from very low metallicities all the way to near solar metallicity. On the one hand, these empirical relations can be directly used to get age and chemical composition information for the stellar populations of unresolved galaxies. On the other hand, they can be used to submit to most stringent tests of population synthesis models, an aspect which is the subject of an accompanying paper \\citep{maraston02}.  The comparison of the Lick indices for the Galactic bulge with those of globular clusters shows that the bulge and the most metal-rich globular clusters have quite similar stellar populations, with the slightly deviating values of some of the bulge indices being the likely result of the metallicity distribution of bulge stars, which extends down to [Fe/H]$\\simeq -1.0$ \\citep{mcwilliam94, zoccali02}. Within the uncertainties, both the metal-rich clusters and the bulge appear to have also the same index ratios, in particular those sensitive to [$\\alpha$/Fe]. This implies similar enhancements for individual $\\alpha$-elements in clusters as in the field. Existing spectroscopic determinations of the $\\alpha$-element enhancement in clusters and bulge field stars are still scanty, but extensive high-resolution spectroscopy at 8--10m class telescopes will soon provide data for a fully empirical calibration of the Lick indices at the [$\\alpha$/Fe] values of the bulge and bulge globular clusters.  Some other line index ratios, such as CN$/\\langle$Fe$\\rangle$, show clear exceptions. In these cases the bulge indices are definitely below the values for the metal-rich clusters. Several possibilities have been discussed for the mechanism responsible for the CN index offset between the bulge and the clusters, the environmental-pollution being active in clusters (but not in the field) appearing as the most likely explanation. In this scenario, globular cluster stars would have experienced accretion of materials lost by cluster AGB stars, early in the history of the clusters (i.e., when clusters were $\\sim 10^8-10^9$ years old)."
    },
    "0209/astro-ph0209597_arXiv.txt": {
        "abstract": "\\noindent We present ISOPHOT P32 oversampled maps and P37/39 sparse maps, of three bright elliptical galaxies in the Virgo Cluster. The maps reach the limiting sensitivity of the ISOPHOT instrument at 60, 100, 170 and 200$\\mu$m. Two elliptical galaxies show no emission at all far-IR ISOPHOT wavelengths at a level of few tens of mJy. The null detection provides a test of the evolution of dust in elliptical galaxies and its size distribution and composition. \\noindent As previous studies have shown, in many elliptical galaxies both IRAS and ISO have detected mid-IR excess 6-15 micron emission relative to the stellar continuum indicating emission from circumstellar dust. Under the assumption that these dusty outflows from evolving red giant stars and planetary nebulae are continuously supplying dust to the interstellar medium, we have computed the infrared luminosity at the ISOPHOT bands appropriate for NGC4472. The null far-IR ISOPHOT observations exceed the far-IR flux expected from dust expelled from a normal old stellar population. ",
        "introduction": "In recent years the traditional view of elliptical galaxies as simple systems of non interacting stars, devoid of interstellar matter, has radically changed. Observations across the electromagnetic spectrum have demonstrated that the interstellar medium (ISM) in elliptical galaxies contains substantial amounts of cold gas and dust in addition to hot gas, the dominant component. Far-IR emission detected by the IRAS satellite and improved optical imaging of elliptical galaxies provide clear evidence for the presence of dust. However, estimates of the total amount of dust, as well as its origin and spatial distribution, remain uncertain and controversial.  Optical observations indicate dust masses that are one order of magnitude  less than those inferred by IRAS. We have recently embarked on a program to study the infrared emission from early-type galaxies using the large database of observations  taken by the ISO satellite. Our principal goals are to observe the  spatial location and emission spectrum of dust throughout elliptical  galaxies, and with this information, to determine the origin, evolution, and physical properties of dust in massive elliptical galaxies. ",
        "conclusions": ""
    },
    "0209/astro-ph0209074_arXiv.txt": {
        "abstract": "Using a sample of 43 suitable local subdwarfs with newly acquired $BVI_{C}$ photometry, we apply our main sequence fitting method to the metal rich Globular Cluster 47 Tucanae. Fitting in 2 colour planes, we find an apparent distance modulus of $(m-M)_{V} = 13.37^{+0.10}_{-0.11}$, leading to a dereddened distance modulus of $(m-M)_{0} = 13.25^{+0.06}_{-0.07}$.  Consideration of the Red Clump in the cluster produces a distance modulus fully consistent with this result.  The implied cluster age is $11\\pm1.4$ Gyr. ",
        "introduction": "Recent empirical distance estimates to 47~Tuc are very discrepant.  MS-fitting by several studies (e.g. Carretta et al. 2000) results in an apparent distance modulus of $(m-M)_{V} > 13.5$, whilst the recent White Dwarf fitting study of Zoccali et al. (2001) finds a much shorter distance, $(m-M)_{V}=13.27$.  The results are not consistent within their quoted errors and the discrepancy in distance moduli implies an uncertainty in age of $\\sim$ 3 Gyr.  Previous MS-fitting studies have suffered from a severe lack of suitable, homogeneous, subdwarf (SD) data with which to determine the cluster distance, which in turn leads to questions concerning the accuracy of the method.  We have identified a large sample of suitable local SDs which enables us to investigate the MS-fitting method thoroughly. Our multi-waveband data also allows us to fit simultaneously in $V/(B-V)$ and $V/(V-I)$ and because of the different sensitivities of the 2 colours to metallicity and reddening, obtaining consistent results is a strong test of the reliability of the derived distance. ",
        "conclusions": ""
    },
    "0209/astro-ph0209242_arXiv.txt": {
        "abstract": "In the ``natural inflation'' model, the inflaton potential is periodic. We show that Planck scale physics may induce corrections to the inflaton potential, which is also periodic with a greater frequency. Such high frequency corrections produce oscillating features in the primordial fluctuation power spectrum, which are not entirely excluded by the current observations and may be detectable in high precision data of cosmic microwave background (CMB) anisotropy and large scale structure (LSS) observations. ",
        "introduction": "In the past decade, inflation theory has successfully passed several non-trivial tests. In particular, recent cosmic microwave background (CMB) observations show that the spatial geometry of the observable Universe is very close to flat \\cite{de Bernardis00,Lange00,Hanany00,Balbi00,Jaffe00}, just as inflation theory predicts. Inflation theory also offers an elegant way of generating the primordial fluctuations which seed the formation of galaxies and large scale structures (LSS)(see e.g., Ref.~\\cite{Lyth-Riotto99} and references therein). In particular, slow-roll inflation models predict that the perturbations are adiabatic, gaussian, and nearly scale-invariant (i.e with a power index $n_s \\simeq 1$). It has long been known that these predictions are also in broad agreement with the observed properties of large scale structures and CMB anisotropy, although at present the data is still not very restrictive \\cite{Wang01}: $n_s=0.91^{+0.15}_{-0.07}$.  This general success of inflation theory brings up the hope of extracting even more detailed information of the inflaton potential\\cite{Lyth-Riotto99,Lidsey97} from high quality observational data. There are two complementary approaches to this problem. In the first approach, one tries a model-independent (subject to some standard assumptions) reconstruction of the primordial power spectrum, then the inflaton potential from the data \\cite{Hannestad01,MSY01,Wang-Mathews02}. Alternatively, one can look for specific features in the power spectrum and study their observational consequences. In particular, there have been many investigations on inflationary models with broken scale invariance\\cite{KLS85,Starobinsky92,ARS97,LPS98,Chung00,WK00,L00}. Such features have been invoked to explain the tentatively observed feature at $k \\sim 0.05 Mpc^{-1}$\\cite{GSZ00,HHV01,BGSS00,GH01}, or even to solve the small scale problem of the CDM model \\cite{KL00}.\\footnote{For other solutions to this problem, see Refs.~\\cite{Spergel PRL}-\\cite{Bode APJ} and for a recent review on this issue see Ref.\\cite{Tasitsiomi}}  In the present work, we consider a new type of feature, periodic in the primordial power spectrum. This type of feature is interesting from both a theoretical and phenomenological point of view. Theoretically, if discovered, it gives a strong hint on the nature of the inflaton field. Phenomenologically, it might change the position, shape or even the number of acoustic peaks in the CMB power spectrum.  This paper is organized as follows: in Sec. II, we describe our model, and show how this type of feature could arise from Planck scale physics by constructing a toy model. Our toy model, which is based on the ``natural inflation'' model, is by no means the only possibility, but in this context it is particularly easy to see how this might happen. In Sec. III we derive the power spectrum in this model, and then consider how it would affect CMB and large scale structure in Sec. IV.  The final section, Sec.V is on the summary and discussions of our results. ",
        "conclusions": "In this paper we present a  model which is a variation of natural inflation. We have shown two features of the primordial spectrum of this model, oscillating and scale-dependence and studied the implications on CMB and LSS. In the presence of the $\\delta$ term the parameter space allowed for a successful natural inflation will be enlarged relative to the original natural inflation model\\cite{MT01}. When the parameter space is enlarged and extended to $\\delta N^2 > 1$ for example due to some other physical motivations\\cite{FLZZ}, there are several additional interesting effects. Although the SR approximation is violated and the spectral index oscillates with a large scale variation, there could be a large parameter space not ruled out by the observations. As we can see from Fig.\\ref{fig:tmp} when we gradually increase the value of $\\delta$ the effects on CMB firstly take place on the first peak which can be slightly split, meanwhile wiggles on the matter power spectrum are gradually enhanced. The effects on CMB first peak and large scale structure are potentially observable and can be tested by future precise experiments.  In summary we have studied in this paper a model which is a variation of natural inflation and show that there are some interesting phenomenological features of this  model, such as oscillating and scale-dependence in the primordial spectrums. And we have also discussed their implications on CMB and LSS.   {\\bf Acknowledgments} We thank  R. Brandenberger  for comments and suggestions on the manuscript. This work is supported in part by National Natural Science Foundation of China under Grant No.90303004 and by Ministry of Science and Technology of China under Grant No.NKBRSF G19990754.    \\newcommand\\AJ[3]{~Astron. J.{\\bf ~#1}, #2~(#3)} \\newcommand\\APJ[3]{~Astrophys. J.{\\bf ~#1}, #2~ (#3)} \\newcommand\\APJL[3]{~Astrophys. J. Lett. {\\bf ~#1}, L#2~(#3)} \\newcommand\\APP[3]{~Astropart. Phys. {\\bf ~#1}, #2~(#3)} \\newcommand\\CQG[3]{~Class. Quant. Grav.{\\bf ~#1}, #2~(#3)} \\newcommand\\JETPL[3]{~JETP. Lett.{\\bf ~#1}, #2~(#3)} \\newcommand\\MNRAS[3]{~Mon. Not. R. Astron. Soc.{\\bf ~#1}, #2~(#3)} \\newcommand\\MPLA[3]{~Mod. Phys. Lett. A{\\bf ~#1}, #2~(#3)} \\newcommand\\NAT[3]{~Nature{\\bf ~#1}, #2~(#3)} \\newcommand\\NPB[3]{~Nucl. Phys. B{\\bf ~#1}, #2~(#3)} \\newcommand\\PLB[3]{~Phys. Lett. B{\\bf ~#1}, #2~(#3)} \\newcommand\\PR[3]{~Phys. Rev.{\\bf ~#1}, #2~(#3)} \\newcommand\\PRL[3]{~Phys. Rev. Lett.{\\bf ~#1}, #2~(#3)} \\newcommand\\PRD[3]{~Phys. Rev. D{\\bf ~#1}, #2~(#3)} \\newcommand\\PROG[3]{~Prog. Theor. Phys.{\\bf ~#1}, #2~(#3)} \\newcommand\\PRPT[3]{~Phys.Rept.{\\bf ~#1}, #2~(#3)} \\newcommand\\RMP[3]{~Rev. Mod. Phys.{\\bf ~#1}, #2~(#3)} \\newcommand\\SCI[3]{~Science{\\bf ~#1}, #2~(#3)} \\newcommand\\SAL[3]{~Sov. Astron. Lett{\\bf ~#1}, #2~(#3)}"
    },
    "0209/astro-ph0209132_arXiv.txt": {
        "abstract": "We have developed a bolometric detector that is intrinsically sensitive to linear polarization which is optimized for making measurements of the polarization of the cosmic microwave background radiation. The receiver consists of a pair of co-located silicon nitride micromesh absorbers which couple anisotropically to linearly polarized radiation through a corrugated waveguide structure. This system allows simultaneous background limited measurements of the Stokes $I$ and $Q$ parameters over $\\sim 30$\\% bandwidths at frequencies from $\\sim 60$ to 600 GHz. Since both linear polarizations traverse identical optical paths from the sky to the point of detection, the susceptibility to systematic effects is minimized. The amount of uncorrelated noise between the two polarization senses is limited to the quantum limit of thermal and photon shot noise, while drifts in the relative responsivity to orthogonal polarizations are limited to the effect of non-uniformity in the thin film deposition of the leads and the intrinsic thermistor properties. Devices using NTD Ge thermistors have achieved NEPs of $2 \\cdot 10^{-17} ~ \\mathrm{W}/\\sqrt{\\mathrm{Hz}}$ with a $1/f$ knee below 100 mHz at a base temperature of 270 mK. Numerical modelling of the structures has been used to optimize the bolometer geometry and coupling to optics. Comparisons of numerical results and experimental data are made. A description of how the quantities measured by the device can be interpreted in terms of the Stokes parameters is presented. The receiver developed for the \\boom and \\planckhfi focal planes is presented in detail. ",
        "introduction": "\\label{sect:intro}  %  Observational cosmologists have yet to detect polarization in the cosmic microwave background radiation (CMB), and upper limits are still well above the level expected due to Thompson scattering of quadrupole anisotropies in the background radiation during the epoch of recombination.\\cite{staggs99,odell02}  The small amplitude of this polarized signal, peaking at perhaps $5 \\mu$K at $\\sim 10'$ scales, demands not only extremely high raw sensitivity, but also exquisite control of systematics.  \\noindent Over the last twenty years, nearly all published efforts to detect polarization in the CMB have used coherent receivers.\\footnote{To the author\\rq s knowledge, the pioneering efforts of Caderni, et al., are the only published CMB polarization limits set by a bolometric system.\\cite{caderni}} Heterodyne, quasi total-power, and correlation receivers with front-end RF low noise amplifier blocks based on HEMTs are mature technologies at millimeter wavelengths. The fundamental design principles of these of receivers are well established and have been used to construct polarized receivers at radio to mm-wave frequencies for many years.\\cite{gaier,spiga02} Although cryogenic bolometric receivers achieve far higher instantaneous sensitivities over wider bandwidths than their coherent analogs, the intrinsic polarization sensitivity of coherent systems has made them the choice of the first generation of CMB polarization experiments.  \\noindent In this paper we describe a new bolometric system which combines the sensitivity, bandwidth, and stability of a cryogenic bolometer with the intrinsic polarization capability traditionally associated with coherent systems. In addition, the design obviates the need for orthogonal mode transducers (OMTs), hybrid tee networks, waveguide plumbing, or quasi-optical beam splitters whose size and weight make fabrication of large format arrays impractical. Finally, unlike OMTs or other waveguide devices, these systems can be relatively easily scaled to $\\sim 600$ GHz, limited at high frequencies only by the ability to reliably manufacture sufficiently small single-moded corrugated structures. Polarization sensitive bolometers (PSBs) are fabricated using the proven photolithographic techniques used to produce \\lq spider web\\rq ~ bolometers, and enjoy the same benefits of reduced heat capacity, negligible cross section to cosmic rays, and structural rigidity.\\cite{yun}  \\noindent Polarization sensitivity is achieved by controlling the vector surface current distribution on the absorber, and thus the efficiency of the ohmic dissipation of incident Poynting flux. This approach requires that the optics, filtering, and coupling structure preserve the sense of polarization of the incident radiation with high fidelity. A multi stage corrugated feed structure and coupling cavity has been designed which achieves polarization sensitivity over a 33\\% bandwidth. A next generation of sub-orbital, ground based, and orbital bolometric CMB polarization experiments, including \\boomn, BICEP, QUEST, and the \\planckhfi are basing their receiver designs around the PSB concept.  \\begin{figure}[tbp] \\centering \\rotatebox{0}{\\scalebox{1}{\\includegraphics{fig1.eps}}} \\caption{An instantaneous image of the field distribution in \\boomn\\rq s corrugated coupling feed. The radiation is incident from the right where low-pass filters and, in some applications, additional optical elements are located. The two bolometers are symmetrically spaced at $\\lambda_\\mathrm{g} / 4 ~ +(-) ~ 30\\mu$m from the backshort in order to maximize coupling efficiency. Similar feed structures have been designed for \\planckn, QUEST and BICEP\\cite{kiwon,bicep} at 100, 150, 217, and 350 GHz.} \\label{fig:feedgeom} \\end{figure} ",
        "conclusions": "We have demonstrated a 300 mK bolometric receiver which is intrinsically sensitive to linear polarization over a 33\\% bandwidth.  The general design is scalable from $\\sim 60-600$ GHz. This design benefits from reduced susceptibility to systematic effects due to the common filtering, matched beams on the sky, matched time constants, stable relative responsivities, and matched end-to-end efficiencies of each sense of linear polarization.   Unlike coherent correlation polarimeters, this receiver simultaneously measures the polarized and unpolarized components of the signal with comparable sensitivity. The design minimizes the size and weight of the receiver, making it especially appropriate for orbital and sub-orbital compact feedhorn arrays. A general method of reliably calculating the optimal absorber impedance for a bolometric detector is presented. The measured performance of the system is in good agreement with the results of the numerical modelling.  \\newpage"
    },
    "0209/astro-ph0209418_arXiv.txt": {
        "abstract": "WeBo~1 (PN~G135.6+01.0), a previously unrecognized planetary nebula with a remarkable thin-ring morphology, was discovered serendipitously on Digitized Sky Survey images.  The central star is found to be a late-type giant with overabundances of carbon and \\sprocess\\ elements. The giant is chromospherically active and photometrically variable, with a probable period of 4.7 days; this suggests that the star is spotted, and that 4.7~days is its rotation period. We propose a scenario in which one component of a binary system became an AGB star with a dense stellar wind enriched in C and \\sprocess\\ elements; a portion of the wind was accreted by the companion, contaminating its atmosphere and spinning up its rotation. The AGB star has now become a hot subdwarf, leaving the optical companion as a freshly contaminated barium star inside an ionized planetary nebula. ",
        "introduction": "In this paper we report the discovery of a new planetary nebula (PN) in Cassiopeia.  The PN appears nearly perfectly elliptical, and its central star is a late-type star with enhanced abundances of carbon and \\sprocess\\ elements.  The system thus appears to represent the immediate aftermath of the formation of a barium star.  In the following sections we present the serendipitous discovery of the nebula, the classification of the nucleus as a \\BaII\\ star, a study of the star's photometric properties and variability, estimates of the distance to the system, and an evolutionary scenario for the origin of this remarkable object.  We conclude with suggestions for follow-up studies. ",
        "conclusions": "WeBo~1 has several properties in common with the class of ``Abell~35''-type planetary nuclei. This class was defined by Bond, Ciardullo, \\& Meakes (1993), and has been discussed by Bond (1994), Jasniewicz et~al.\\ (1996), Jeffries \\& Stevens (1996), and Gatti et al.\\ (1997).  In the three known A~35-type nuclei (A~35, LoTr~1, and LoTr~5), a rapidly rotating late-type giant or subgiant is seen optically, while a hot companion is detected at UV wavelengths.  The cool components vary photometrically with periods of a few days, corresponding to their rotation periods.  A definitive orbital period has not been found for any of these three objects from radial-velocity studies, suggesting that the orbital periods may be long.  This suspicion is confirmed in the case of the field star HD 128220, which lacks a PN but is otherwise similar in all respects to the A~35-type nuclei: its orbital period is 872~days (Howarth \\& Heber 1990).  These systems, then, have almost certainly not undergone common-envelope interactions, which would have decreased their orbital periods by substantial amounts.  As discussed by Jeffries \\& Stevens (1996), there is a closely related class of wide binaries containing hot white dwarfs and cool, rapidly rotating, magnetically active {\\it dwarfs}.  These authors propose a mechanism in which an AGB star in a wide binary develops a dense stellar wind, part of which is accreted by the companion star.  Their calculations suggest that significant spin-up of the companion may occur, along with accretion of chemically enriched material from the AGB star.  Although Jeffries \\& Stevens considered their suggestion somewhat speculative (in the absence of actual 3-D hydrodynamical simulations of the accretion and spin-up), observational support has arisen in the past several years.  This includes the finding of mild Ba enhancements in the rapidly rotating dK component of 2RE~J0357+283 (Jeffries \\& Smalley 1996), and in the nuclei of A~35 and LoTr~5 (Thevenin \\& Jasniewicz 1997). WeBo~1, with its pronounced Ba and C overabundances, now provides further support.  A scenario that emerges to explain the properties of WeBo~1 is thus as follows. The progenitor system was a fairly wide binary whose components had nearly equal masses (initial mass ratio $\\sim$0.98).  The more massive star evolved to the AGB stage, at which point the less massive component had also begun to ascend the red-giant branch. The AGB star was constrained to rotate with the orbital period, so that, as it developed a dense wind, the wind was ejected preferentially in the orbital plane, leading to the ring-like nebular morphology.  The wind was enriched in Ba and other \\sprocess\\ elements, and had $\\rm C/O>1$. A portion of the wind was accreted by the companion giant, spinning it up to the observed 4.7-day rotation period, and contaminating its photosphere with Ba, C, and other pollutants.  At present, the AGB star has completely shed its envelope, exposing its hot core whose UV radiation ionizes the ejected ring.   Several follow-up studies are clearly warranted. (1)~Ultraviolet spectra would confirm the expected presence of a hot companion; a determination of its surface gravity from its Ly$\\alpha$ profile would provide a mass determination, and thus an estimate of the progenitor's initial mass and  constraints on evolutionary scenarios.  (2)~Radial velocities from ground-based spectra would allow a search for the orbital period, or set a lower limit if, as we suspect, the orbital period is long. (3)~High-dispersion spectra of the barium star should be obtained, to determine its rotational velocity and atmospheric elemental abundances.  In particular, it would be of great interest to search for lines of technetium, which should be present if the star is really a very recently created barium star.  (4)~An abundance analysis of the nebula would also be of interest, since it may be possible to detect lines of heavy \\sprocess\\ elements."
    },
    "0209/astro-ph0209304_arXiv.txt": {
        "abstract": "Isothermal models and other simple smooth models of dark matter halos of gravitational lenses often predict a dimensionless time delay $H_0\\Delta t$ much too small to be comfortable with the observed time delays $\\Delta t$ and the widely accepted $H_0$ value ($\\sim 70$ km/s/Mpc). This conflict or crisis of the CDM has been highlighted by several recent papers of Kochanek, who claims that the standard value of $H_0$ favors a strangely small halo as compact as the stellar light distribution with an overall nearly Keplerian rotation curve. In an earlier paper we argue that this is not necessarily the case, at least in a perfectly symmetrical Einstein cross system (Paper I, astro-ph/0209191). Here we introduce a {\\it new mass degeneracy} of lens systems to give a realistic counter example to Kochanek's claims.  We fit the time delay and image positions in the quadruple image system PG1115+080. Equally good fits are found between lens models with flat vs. Keplerian rotation curves.  Time delays in both types of models can be fit with the standard value of $H_0$. We demonstrate that it may still be problematic to constrain the size of lens dark halos even if the data image positions are accurately given and the cosmology is precisely specified. ",
        "introduction": "Gravitational lensing provides a powerful tool to constrain the dark matter halos of galaxies.  One of the promises of gravitational lenses is to constrain the Hubble constant.  However, this has been hampered to some extent by the intrinsic degeneracies in models of the dark matter potential of the lens (Williams \\& Saha 2000; Saha 2000; Saha \\& Williams 2001; Zhao \\& Pronk 2001). Now that the value of $H_0$ is fairly well constrained by independent methods, e.g., $H_0=72\\pm 8$~km/s/Mpc from the HST key project (Freedman et al. 2001), and the cosmological model has been determined at more and more precision, it is interesting to ask whether we can reverse the game and set more stringent constraint on the dark matter potential.  To this end, we would like to understand whether the Hubble constant and the lensed images could uniquely specify the dark matter content, or whether there are very different lens models with identical $H_0$ value.  It is well-known that isothermal models and other simple smooth models of dark matter halos of gravitational lenses often predict a dimensionless time delay $H_0\\Delta t$ much too small to be comfortable with the observed time delays $\\Delta t$ and the widely accepted $H_0$ value ($\\sim 70$ km/s/Mpc).  Models with isothermal dark halos tend to yield an $H_0$ around 50 km/s/Mpc.  This conflict has been highlighted by several recent papers (Kochanek 2002a,b,c).  Kochanek (2002a) found that it is difficult to reconcile the time delays measured for five simple and well-observed gravitational lenses with $H_0 \\sim 70$ km/s/Mpc unless the lens galaxy has a nearly Keplerian rotation curve with the halo following the stellar mass profile by a constant mass-to-light ($M/L$) ratio.  If the lenses had a more plausible flat rotation curve (isothermal mass distributions) he found $H_0=48_{-4}^{+7}$~km/s/Mpc, which is grossly inconsistent with the HST Key Project.  Kochanek (2002c) argued that more realistic models with a CDM halo plus adiabatically cooled baryons behave like isothermal models.  They produced a still too low $H_0$ unless one adopts a problematically high baryon fraction $\\Omega_b/\\Omega_m > 0.2$ of the universe, and require all these baryons to cool.  His conclusion was that either $H_0 \\sim 70$ km/s/Mpc is too high or any lens mass models for the observed time delay systems must follow a compact distribution, nearly like that of the stellar light, hence has very little extended dark matter halo.  This argued for the first time a {\\it new problem} for dark matter halos, and a particularly serious problem for current CDM paradigm of galaxy formation.  Here we discuss the effect of a {\\it new degeneracy} in strong lensing models in resolving this {\\it new problem}.  It is shown that the observed time delays and image positions cannot uniquely determine the extent of the lens mass distribution.  In particular, a system with a very extended dark matter distribution could minic a system without any dark matter as far as strong lensing data are concerned.  Here we give an analytical explicit illustration of the degree of degeneracy in lens models. ",
        "conclusions": "As we can see, it is possible to construct many very different models with positive, smooth and monotonic surface densities to fit the image positions.  There are also no extra images.  These models also fit the same time delay and time delay ratios using a Hubble constant and cosmology consistent with $\\Lambda$CDM cosmology. Hence the models are truely indistinguishable for lensing data. They fit the flux ratios equally well, and produce nearly indistinguisable Einstein rings, which is the region of minimal gradient of the time delay surface.  The models have identical light profiles, undistinguishable by data.  Among the acceptable models to PG1115+080, there are models with a Keplerian rotation curve and models with a nearly flat rotation curve.  So lensing data plus $H_0$ cannot uniquely specify the mass-to-light ratio of this system.  We conclude that strong lensing data may not uniquely determine the Hubble constant, even if we fix the cosmology, the lens and source redshifts, and the time delays and amplification ratios of the four images.  There are at least important degeneracies in inverting the data of a perfect Einstein cross to the lens models and the Hubble constant.  The relation between the value of $H_0$ and the size of the halo is not straightforward: a high $H_0$ does not necessarily mean no dark halo, and models with a flat rotation curve do not always yield a small $H_0$.  We also comment that it would be difficult to determine the cosmology from strong lensing data alone because the non-uniqueness in the lens models implies that the combined parameter $h_0^{-1}\\zeta(\\Omega,z_l,z_s)$ is poorly constrained by the lensing data, even if $h_0$ and the redshifts $z_l,z_s$ are given.  This work was supported by the National Science Foundation of China under Grant No. 10003002 and a PPARC rolling grant to Cambridge.  HSZ and BQ thank the Chinese Academy of Sciences and the Royal Society respectively for a visiting fellowship, and the host institutes for local hospitalities during their visits."
    },
    "0209/astro-ph0209022_arXiv.txt": {
        "abstract": "Procedure and results of computations of stellar model atmospheres and spectral energy distributions are discussed. Model atmospheres of some chemically peculiar stars are computed taking into account detailed information about their abundances:  --- R CrB-like stars of Teff $\\sim$ 7000 K,  --- Sakurai's object (V4334 Sgr) of 4000 $<$ \\Tef $<$ 7000 K  --- Przybylski's star of Teff $\\sim$ 6500 K.  We show that our self-consistent approach provides a unique possibility to investigate the temporal changes of physical parameters of chemically peculiar stars.  Some issues of computation of model atmospheres of M and C-giants are also considered. ",
        "introduction": "In many aspects, the existence of the irregular hydrogen-deficient (Hd) variables remains puzzling so far. R CrB is the most known member of the post-AGB group. Sakurai's object (SO, V4334 Sge) provides another, extreme case of stellar evolution.  It has been firmly established that the most abundant elements in atmospheres of R CrB-like stars are helium and carbon. Determination of abundances in their atmospheres is possible only in the frame of self-consistent approach (Asplund et al. 1998). Still even in the case of the ``normal'' red giants that approach should be used. Otherwise, abundance determination results might be affected by significant errors ($>$ 0.2-0.3 dex, see Pavlenko \\& Yakovina 1994 for more details).   HD~101065 (V816 Cen) presents another case of the peculiar stellar spectrum (see Cowley et al. 2000) ---- the strongest spectral lines in the spectrum of HD~101065 generally belong to lanthanides. ",
        "conclusions": ""
    },
    "0209/astro-ph0209508_arXiv.txt": {
        "abstract": "A {\\it shadow} is an exact solution to a chaotic system of equations that remains close to a numerically computed solution for a long time, ending in a {\\it glitch}.  We study the distribution of shadow durations at low dimension and how shadow durations scale as dimension increases up to 300 in a slightly simplified gravitational $n$-body system.  We find that ``softened'' systems are shadowable for many tens of crossing times even for large $n$, while in an ``unsoftened'' system each particle encounters glitches independently as a Poisson process, giving shadow durations that scale as $1/n$. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209487_arXiv.txt": {
        "abstract": "We have observed Markarian~421 in January and March 2001 with the CANGAROO-II imaging Cherenkov telescope during an extraordinarily high state at TeV energies. From 14~hours observations at very large zenith angles, $\\sim$70$^\\circ$, a signal of 298\\,$\\pm$\\,52 gamma-ray--like events (5.7~$\\sigma$) was detected at $E>10$~TeV, where a higher sensitivity is achieved than those of usual observations near the zenith, owing to a greatly increased collecting area. Under the assumption of an intrinsic power-law spectrum, we derived a differential energy spectrum $dN/dE = (3.3\\,\\pm\\,0.9_{stat.}\\,\\pm\\,0.3_{syst.})\\times10^{-13}~(E/10~\\mbox{TeV})^{-(4.0\\,^{+0.9}_{-0.6}\\,_{stat.}\\,\\pm\\,0.3_{syst.})}$  ph./cm$^2$/sec/TeV, which is steeper than those previously measured around 1~TeV, and supports the evidence for a cutoff in the spectrum of Markarian~421. However, the 4\\,$\\sigma$ excess at energies greater than 20~TeV in our data favors a cutoff energy of $\\sim$8~TeV, at the upper end of the range previously reported from measurements at TeV energies. ",
        "introduction": "Markarian~421 (Mrk~421, J1104+3812) is a nearby BL Lacertae object ($z=0.031$) and was the first extragalactic TeV gamma-ray source discovered \\citep{punch92}. The TeV gamma-ray flux is variable, with flaring behavior observed on time-scales of less than an hour \\citep{gaidos96}. Extensive measurements have been performed by several experimental groups based on the imaging Cherenkov technique~\\citep{aharonian99_1,krennrich99,piron01}. Multi-wavelength observations support the Synchrotron--Self--Compton (SSC) mechanism for the production of TeV gamma-rays from this source ~\\citep[see, e.g.,][]{takahashi00,krawczynski01}.  TeV gamma-rays from extra-galactic sources suffer absorption due to photon-photon interactions with the inter-galactic infrared background radiation~\\citep{hikishov62,gould67,stecker92}. According to recent measurements of the infrared background~\\citep[see, e.g.,][and references therein]{hauser01} and predictions of the optical depth for TeV gamma-rays~\\citep{primack99,dejager01,totani02}, gamma-rays at energies above 10~TeV from Mrk~421 are expected to be suppressed, since they interact with mid- to far-infrared photons of $\\sim$100~$\\mu$m.  Mrk~421 became active in 2000 and 2001, especially at the beginning of 2001~\\citep{boerst01}. During this period, northern hemisphere observers measured the energy spectrum with good statistics in the region from several hundred GeV to $\\sim$10~TeV and reported cutoffs at 3--6~TeV \\citep[]{krennrich01,aharonian02b}. The cutoff energy is consistent with, or slightly smaller than, that measured for Mrk~501 during its flaring state in 1997~\\citep{aharonian99_2,aharonian01}. As Mrk~501 has a similar redshift ($z=0.034$) to Mrk~421, this suggests the cutoffs may be due to infrared absorption of TeV gamma-rays.  We observed Mrk~421 during the 2001 high state with the CANGAROO-II 10~meter telescope, at very large zenith angles of $\\sim$70$^\\circ$. Similar observations have been reported by the Durham group for Mrk~501 in the high state of 1997~\\citep{chadwick99}. For these observations, an effective collecting area $\\sim$10 times larger than that for observations near the zenith is obtained, with an accompanying increase in the gamma-ray energy threshold to $\\sim$10~TeV. ",
        "conclusions": "Figure~\\ref{fig:energyflux} (inserted panel) shows the raw energy spectrum of the observed gamma-ray events from Mrk~421. The gamma-ray energy was assigned from the pulse-height sum of the individual pixels, using a relation obtained from the simulations. This method is similar to that described in \\citet{mohanty98}, and an energy resolution of $\\sim$31~\\% is estimated. The excess events are distributed in the energy range 7--45~TeV, however one must take care of the spill-over effect from the lower energies due to the finite energy resolution. In order to take this effect into account, simulated gamma-ray spectra, with the spectral indexes and cutoff energies varied, were compared to the data and the observed spectral parameters were determined from the values which minimized the value of $\\chi^2$. With the assumption of a power-law spectrum, the differential flux was fitted by  $$ \\frac{dN}{dE} = (3.3\\,\\pm\\,0.9_{stat.}\\,\\pm\\,0.3_{syst.}) \\times 10^{-13}\\ \\left(\\frac{E}{10~\\mbox{TeV}}\\right)^{-(4.0\\,^{+0.9}_{-0.6}\\,_{stat.}\\,\\pm\\,0.3_{syst.})} \\ ~~~~\\mbox{ph.}/\\mbox{cm}^{2}/\\mbox{sec}/\\mbox{TeV} $$  \\noindent with  $\\chi^2$=2.5/2~$d.o.f.$ The cut dependence on $R_{prob}$ and $alpha$ parameters, and the trigger conditions in the simulation, were considered as sources of the systematic uncertainties. The systematic errors giving rise to uncertainty in the energy scale such as Cherenkov photon scattering in the atmosphere are not included here, but are considered in more detail later. The derived spectrum is steeper than those observed at lower TeV energies. The spectral shape was tested with a cutoff spectrum of $E^{-1.9}\\exp(-E/4\\mbox{TeV})$, as was derived from the measurements by the Whipple and HEGRA-CT groups, with the spectral index being the hardest one observed during the strong flaring period~\\citep{aharonian02b,krennrich02}. The fitting result did not improve compared to that with the power-law assumption ($\\chi^2$=5.0/3~$d.o.f.$), as an excess of events above 20~TeV is apparent, as shown in Fig~\\ref{fig:alpha}~(b). An excess of 103\\,$\\pm$\\,26 (4.0~$\\sigma$) was observed with $alpha<20^\\circ$, while 11~events are expected for the cutoff spectrum, based on an estimation using the event ratio between 10--20~TeV and over 20~TeV. However, if a cutoff energy of 8~TeV is assumed, the consistency with the data becomes better (48 events expected for $E^{-1.9}\\exp(-E/8\\mbox{TeV})$). This cutoff energy is at the high end of the range allowed for Mrk~501~(\\citealt{aharonian99_2}, see also \\citealt{aharonian01}). Since these two AGNs have similar redshifts, the cutoff energies in both spectra are expected to be similar, assuming the attenuation is predominantly due to infrared absorption. As there is only a 2\\,$\\sigma$ difference between our observations and this prediction, our result falls in the acceptable range of the absorption hypothesis due to the cosmic infrared background.  Figure~\\ref{fig:energyflux} (main panel) shows the measured energy flux, assuming the power-law spectrum. Data for the Whipple~\\citep{krennrich01} and HEGRA-CT groups~\\citep{aharonian02b}, observed during a similar period of the flaring state (January--March 2001) are also shown. The observation periods were not exactly the same and the source varied significantly during this high state, therefore the absolute fluxes are expected to differ at some level. The absolute flux level determined from the CANGAROO-II data is within the observed range of the flux variation reported by the Whipple group~\\citep{krennrich02}, and the spectral slope around 10~TeV is consistent with that of these two groups, supporting the roll-over from the flatter spectrum measured at lower energies.  For large zenith angle observations, a large uncertainty in the energy scale, due to the absorption of Cherenkov photons in the atmosphere, is inevitable. Only Rayleigh scattering was considered in the simulation code to avoid over-estimating the gamma-ray energies. The inclusion of Mie scattering and ozone absorption would affect the energy scale by $\\sim$30~\\% and $\\sim$3~\\%, respectively, based on numerical estimations using the program code of \\citet{kneizys96}. We stress that these effects increase the energy scale. The use of the  ``flat-Earth'' approximation for the atmosphere in the simulations requires a $\\sim$6~\\% correction which has already been taken into account in the discussion above.  The measurement of spectra at large zenith angles was verified by observations of the Crab nebula up to the zenith angles of $\\sim$55$^\\circ$, although calibration using the Crab nebula at the same zenith angles as the Mrk~421 observations ($\\sim$70$^\\circ$) is unfortunately impractical with the current instrumental sensitivity. The strong gamma-ray emission of Mrk~421 ($\\sim$~3 times that of Crab nebula) enabled us to detect the source in only 14~hours. In order to detect the Crab nebula at the same significance level, more than 150~hours observations would be required.  In summary, owing to the large effective area and the high resolution performance of the Cherenkov imaging camera, E$>$10~TeV gamma-rays from Mrk~421 were detected at a high confidence level at zenith angles of $\\sim$70$^\\circ$ with 14~hours of observations. The derived spectrum in the region of 10--30~TeV is steeper than that around 1~TeV, which supports the cutoff spectrum of Mrk~421 measured in the 0.2--10~TeV range by other groups. The excess observed above 20~TeV is strongly suggestive of a higher cutoff energy, $\\sim$8~TeV, compared to the lower energy observations. These observations confirm, with the support of detailed simulations, the viability of the large zenith angle technique. Large zenith angle observations provide a unique method of measuring the spectrum in the important energy range above 10~TeV with a relatively short observation time."
    },
    "0209/astro-ph0209352_arXiv.txt": {
        "abstract": "Recently, from the {\\it Hubble Space Telescope} (HST) images of one of the Large Magellanic Cloud (LMC) events taken 6.3 years after the original lensing measurement, Alcock et al.\\ were able to directly image the lens. Although the first resolved lens was identified for an LMC event, much more numerous lenses are expected to be resolved for Galactic bulge events. In this paper, we estimate the fraction of Galactic bulge events whose lenses can be directly imaged under the assumption that all bulge events are caused by normal stars.  For this determination, we compute the distribution of lens proper motions of the currently detected Galactic bulge events based on standard models of the geometrical and kinematical distributions of lenses and their mass function.  We then apply realistic criteria for lens resolution, and the result is presented as a function of the time elapsed after an original lensing measurement, $\\Delta t$. If followup observations are performed by using an instrument with a resolving power of $\\theta_{\\rm PSF}=0''\\hskip-2pt .1$, which corresponds to that of HST equipped with the new Advanced Camera for Surveys, we estimate that lenses can be resolved for $\\sim 3\\%$ and $22\\%$ of disk-bulge events and for $\\sim 0.3\\%$ and $6\\%$ of bulge self-lensing events after $\\Delta t=10$ and 20 years, respectively.  The fraction increases substantially with the increase of the resolving power.  If the instrument has a resolution of $\\theta_{\\rm PSF}=0''\\hskip-2pt .05$, which can be achieved by the {\\it Next Generation Space Telescope}, we estimate that lenses can be resolved for $\\sim 22\\%$ and $45\\%$ of disk-bulge events and for $\\sim 6\\%$ and $23\\%$ of bulge self-lensing events after $\\Delta t=10$ and 20 years, respectively. ",
        "introduction": "Following the proposal of Paczy\\'nski (1986), experiments to search for lensing-induced light variations of stars (microlensing events) located in the Galactic bulge and the Magellanic Clouds have been or are being conducted by several groups (MACHO: Alcock et al.\\ 1993; EROS: Aubourg et al.\\ 1993; OGLE: Udalski et al. 1993; MOA: Bond et al.\\ 2001; DUO: Alard \\& Guibert 1997).  These experiments have successfully detected a large number of events ($\\sim 1,000$), most of which are detected towards the Galactic bulge.   Despite a large number of event detections, the nature of the lenses is still poorly known.  This is because the Einstein ring radius crossing time $t_{\\rm E}$ (Einstein timescale), which is the only observable providing information about the physical parameters of the lens (lens parameters), results from a combination of the lens parameters, i.e.\\ \\begin{equation} t_{\\rm E} = {r_{\\rm E}\\over v};\\qquad r_{\\rm E} = \\left[ {4GM\\over c^2}{D_{\\rm OL}(1-D_{\\rm OL})\\over D_{\\rm OS}} \\right]^{1/2}, \\end{equation} where $r_{\\rm E}$ is the Einstein ring radius, $M$ is the lens mass, $v$ is the lens-source transverse speed, and $D_{\\rm OL}$ and $D_{\\rm OS}$ are the distances to the lens and the source from the observer, respectively.  Under this circumstance, the only approach one could pursue would be identifying the major lens population by statistically determining the lens mass function based on the observed timescale distribution.  However, this approach requires {\\it a prior} knowledge about the geometrical distribution of the lens, the lens kinematics, and the functional form of the lens mass function, which are all poorly known.  In addition, even if all lensing objects were of the same mass, they would give rise to a broad range of timescale. As a result, it is difficult to identify the major lens population from this approach (Mao \\& Paczy\\'nski 1996; Gould 2001).   Recently, from the {\\it Hubble Space Telescope} (HST) images of one of the Large Magellanic Cloud (LMC) events (MACHO LMC-5) taken 6.3 years after the original lens measurement, Alcock et al.\\ (2001) were able to resolve the lens from the lensed source star.  By directly imaging the lens, they could identify that  the event was caused by a nearby low mass star located in the Galactic disk.   Besides the identification of the lens as a normal star, direct lens imaging is of scientific importance due to following reasons.  First, by directly and accurately measuring the lens proper motion with respect to the source, $\\mu$, one can better constrain the physical parameters of the individual lenses.  The previous method to determine $\\mu$ is based on the analysis of the lensing light curves of events affected by the finite source effect, such as source-transit single lens events and caustic-crossing binary lens events (Gould 1994; Witt \\& Mao 1994; Nemiroff \\& Wickramasinghe 1994).  By analyzing the part of the light curves near the source transit or the caustic crossing of these events, one can measure the source star angular radius normalized by the angular Einstein ring radius $\\theta_{\\rm E}$, i.e.\\ $\\rho_\\star=\\theta_\\star/ \\theta_{\\rm E}$, where $\\theta_\\star$ is the angular source star radius. Then, the lens proper motion is determined by \\begin{equation} \\mu = {\\theta_{\\rm E}\\over t_{\\rm E}} = {\\theta_\\star/\\rho_\\star \\over t_{\\rm E}}. \\end{equation} For the proper motion determination by using this method, however, one should know the source star angular radius, which can only be deduced from an uncertain color-surface brightness relation.  As a result, the proper motions determined in this way suffer from large uncertainties. By contrast, if the lens is resolved, the proper motion can be directly and thus accurately measured from the observed image.  Measuring the proper motion is equivalent to measuring the angular Einstein ring radius because $\\theta_{\\rm E}=\\mu t_{\\rm E}$, where the event timescale is determined from the light curve.  While $t_{\\rm E}$ depends on three lens parameters of $M$, $D_{\\rm OL}$, and $v$, $\\theta_{\\rm E}$ does not depend on $v$, and thus the lens mass can be better constrained.  Second, if the lens is resolved for an event where the lens-source relative parallax, $\\pi_{\\rm rel} = {\\rm AU}/(D_{\\rm OL}^{-1}-D_{\\rm OS}^{-1})$, was previously measured during the lensing magnification, one can completely break the lens parameter degeneracy and the lens mass is uniquely determined by \\begin{equation} M = {\\mu^2 t_{\\rm E}^2\\over \\kappa \\pi_{\\rm rel}}, \\end{equation} where $\\kappa = 4G/(c^2{\\rm AU})\\sim 8.144\\ {\\rm mas}/M_\\odot$ (Gould 2001).  Third, if the source of an event was resolved via either a source transit or a caustic crossing and thus $\\rho_\\star$ was precisely measured, one can determine the angular source star radius by reversing the process of the classical method of the proper motion determination, i.e.\\ $\\theta_\\star=\\mu\\hskip2pt t_{\\rm E} \\hskip2pt \\rho_\\star$.  By measuring $\\theta_\\star$, one can determine the effective temperature of the source star, which is important for the accurate construction of stellar atmosphere models (e.g., Alonso et al.\\ 2000).   Although the first directly imaged lens was identified for an LMC event, much more numerous direct lens identifications are expected if high resolution followup observations are performed for events detected towards the bulge.  There are several reasons for this expectation. First, compared to the total number of LMC events, which is $\\sim 20$, there are an overwhelmingly large number of bulge events.  Second, while the majority of LMC events are suspected to be caused by dark (or very faint) objects, most bulge events are supposed to be caused by normal stars, for which imaging is possible.  Third, an important fraction of lenses responsible for bulge events are believed to be located in the Galactic disk with moderate distances, and thus more likely to be imaged due to their tendency of being bright and having large proper motions.   The goal of this work is to estimate the fraction of Galactic bulge events whose lenses can be directly imaged.  For this estimation, we first compute the expected distribution of the lens-source proper motions of the currently detected Galactic bulge events based on standard models of the geometrical and kinematical distributions of lenses and their mass function (\\S\\ 2).  We then apply realistic detection criteria for lens resolution and the result is presented as a function of the time elapsed after the original lensing measurement, $\\Delta t$ (\\S\\ 3).  Based on the result in \\S\\ 3, we discuss some of the observational aspects of the future high resolution followup lensing observations aimed for direct lens imaging (\\S\\ 4).  We summarize the result and conclude in \\S\\ 5. ",
        "conclusions": "We have estimated the fraction of Galactic bulge microlensing events for which the lenses can be directly imaged from future high resolution followup observations by computing the distribution of proper motions of the currently detected bulge events and imposing realistic criteria for lens resolution.  From this computation, we find that lens identification will be possible for a significant fraction of bulge events from followup observations using NGST under the assumption that most bulge events are caused by normal stars.  Besides identifying lenses as stars, direct lens imaging will allow one to accurately determine the lens proper motion, from which the physical parameters of the individual lenses can be better constrained.  If lenses are imaged for events where lens parallaxes were measured, the lens parameter degeneracy can be completely broken and the lens mass can be uniquely determined.  In addition, high resolution followup observations will provide a valuable chance to measure the angular radii of remote bulge source stars involved with events for which the source was previously resolved via either a source transit or a caustic crossing.   We would like to thank A.\\ Gould for proving useful comments about the work.  This work was supported by a grant (R01-1999-00023) of the Korea Science \\& Engineering Foundation (KOSEF)."
    },
    "0209/astro-ph0209288.txt": {
        "abstract": "{We present a multi-wavelength study of the star forming region ISOSS J 20298+3559, which was identified by a cross-correlation of cold compact sources from the 170 $\\mu$m ISOPHOT Serendipity Survey (ISOSS) database coinciding with objects detected by the MSX, 2MASS and IRAS infrared surveys. ISOSS J 20298+3559 is associated with a massive dark cloud complex (M $\\sim$ 760 M$_{\\odot}$) and located in the Cygnus X giant molecular cloud. We derive a distance of 1800 pc on the basis of optical extinction data. The low average dust temperature (T $\\sim$ 16 K) and large mass (M $\\sim$ 120 M$_{\\odot}$) of the dense inner part of the cloud, which has not been dispersed, indicates a recent begin of star formation. The youth of the region is supported by the early evolutionary stage of several pre- and protostellar objects discovered across the region: I) Two candidate Class 0 objects with masses of 8 and 3.5 M$_{\\odot}$, II) a gravitationally bound, cold (T $\\sim$ 12 K) and dense (n(H$_{2}$) $\\sim$ 2 $\\cdot$ 10$^{5}$ cm$^{-3}$) cloud core with a mass of 50 M$_{\\odot}$ and III) a Herbig B2 star with a mass of 6.5 M$_{\\odot}$ and a bolometric luminosity of 2200 L$_{\\odot}$, showing evidence for ongoing accretion and a stellar age of less than 40000 years. The dereddened SED of the Herbig star is well reproduced by an accretion disc + star model. The externally heated cold cloud core is a good candidate for a massive pre-protostellar object. The star formation efficiency in the central cloud region is about 14 \\%. ",
        "introduction": "It is a challenge to identify massive young stellar objects during their early evolution. The youngest protostars form deeply embedded in their cold (T $\\sim$ 10-20 K) parental clouds (Pudritz 2002). The association with dense ambient material makes such objects best detectable as cold condensations at far-infrared and (sub)millimeter wavelengths. The short evolutionary timescales (Palla \\& Stahler 1993) and low spatial density of massive objects require large scale surveys for their identification. Many of the known intermediate- and high-mass protostellar candidates have therefore been discovered by follow-up studies towards IRAS sources (eg. Shepherd et al. 2000, Cesaroni et al. 1997, Molinari et al. 1998, Beuther et. al 2002), which were selected on the basis of color and flux density criteria (e.g. by Wood \\& Churchwell 1989, Palla et al. 1991)  The earliest stages of massive star formation are characterized by the initial conditions of their parental cloud cores with spectral energy distributions peaking beyond 100 $\\mu$m (Evans et al. 2002). In order to unveil such young objects we are using the ISOPHOT (Lemke et al. 1996) 170 $\\mu$m Serendipity Survey (ISOSS) (Bogun et al. 1996), which is the largest high spatial resolution survey performed beyond the IRAS 100 $\\mu$m band. We selected bright and compact sources detected by ISOSS and IRAS with a flux ratio [F170$\\mu$m/F100$\\mu$m] $>$ 2, implying a low dust temperature T $<$ 18 K and a large mass of the cold ISM in these objects. Since the clustered mode of massive star formation commonly involves young stellar objects of different evolutionary stages, we require the presence of embedded sources with thermal infrared excess as indicated by the 2MASS (Cutri et al. 2000) and MSX (Price et al. 2001) infrared surveys. The latter criterium also avoids confusion with cold interstellar cirrus. Here, we present the results of follow-up observations of the cold star forming region ISOSS J 20298+3559 and show evidence for its early evolutionary stage. ",
        "conclusions": "We have identified the young star forming region ISOSS J 20298+3559 performing a cross-correlation of cold compact far-infrared sources from the ISOPHOT 170 $\\mu$m Serendipity Survey database with the 2MASS, MSX and IRAS surveys. Multi-wavelength follow-up observations of this region yield:  \\begin{enumerate} \\item The star forming region is associated with a complex of four optical dark clouds C1..C4 which have a total mass of 760 M$_{\\odot}$.  \\item We derived a distance of 1800 $\\pm$ 300 pc based on optical extinction data. This associates the region with the Cygnus\\,X Giant Molecular Cloud in agreement with our molecular line kinematics.  \\item The cold ISOSS source FIR1 corresponds to the dense inner region of the central dark cloud C1 and contains a total mass of 120 M$_{\\odot}$ gas and dust with an average temperature of 16 K.  \\item We have identified two candidate Class 0 objects SMM1 and SMM3. The sources have masses of 8 and 3.5  M$_{\\odot}$ which makes them precursors of intermediate mass stars.  \\item The externally heated cloud core of C1 has a total mass of 50 M$_{\\odot}$ and a central dust temperature as low as 11K. Ammonia in the NH$_{3}$(1-1) transition has been detected. The object is gravitationally bound as derived from our ammonia molecular line observations, which makes it a candidate massive pre-protostellar core.  \\item The most luminous object in the vicinity is the Herbig B2 star IRS1. The object has a mass of 6.5 M$_{\\odot}$ and a bolometric luminosity of 2200 L$_{\\odot}$. Inverse P Cygni profiles in the higher HI Balmer series and SiII lines indicate ongoing accretion. The spectral energy is well described by a Kurucz model for the stellar photosphere and a viscous reprocessing accretion disk. The stellar age inferred from pre-main-sequence evolutionary tracks is less than 40000 yr.  \\item Several embedded near-infrared sources have been identified. One of them (IRS7) is surrounded by a mid-infrared reflection nebula.  \\item The star formation efficiency in the dense and cold region FIR1 is about 14 \\%.  \\end{enumerate}"
    },
    "0209/astro-ph0209434_arXiv.txt": {
        "abstract": "{ I use literature data and a new temperature calibration to determine the Li abundances in the globular cluster M 92. Based on the same data, Boesgaard et al. have claimed that there is a dispersion in Li abundances in excess of observational errors. This result has been brought as evidence for Li depletion in metal-poor dwarfs. In the present  note I argue that there is no strong evidence for intrinsic dispersion in Li abundances, although a dispersion as large as 0.18 dex is possible. The mean Li abundance, A(Li)=2.36, is in good agreement with recent results for field stars and TO stars in the metal-poor globular cluster NGC 6397. The simplest interpretation is that this constant value represents the primordial Li abundance. ",
        "introduction": "The present  note is motivated by the recent analysis of the metal-poor globular cluster NGC 6397 by \\citet{Bonifacio}. We have confirmed that all the cluster stars share the same Li abundance that is found exactly at the level of the {\\em Spite plateau} defined for field stars. This  supports the notion that Li in metal-poor stars is of primordial origin and good agreement is found with the other primordial nuclei D and $\\rm ^3He$. However, in this picture, the observations of \\citet{boe98} of Li in the globular cluster M 92 are troublesome. After a careful analysis of the best available data, the above authors conclude that there is a dispersion in Li abundances by a factor of three, unexplained by observational errors. The most likely explanation suggested for this dispersion is differential Li depletion due to rotational mixing; the stars would deplete more or less Li depending on their initial angular momentum. While this is a reasonable explanation, one is left to wonder why such effect should be observable in M 92 but not in NGC 6397, which has a very similar metallicity. Moreover, recent models that predict Li depletion either through rotational mixing \\citep{pin01}, or diffusion \\citep{sal01}, or a combination of both plus composition gradient \\citep{theado}, predict a mild depletion of 0.1 -- 0.2 dex accompanied by  a very tiny dispersion. It  is doubtful that such a tiny dispersion could be detected in M 92 with the presently available data. The experience with NGC 6397 has shown that use of a photometric colour such as $b-y$ or $B-V$ to estimate effective temperatures is not well suited to the issue of the scatter in Li abundances. The Li abundance is very sensitive to errors in \\teff , therefore quite small photometric errors in the colours, translate into large errors on Li abundances. In the case of NGC 6397, \\citet{Bonifacio} took advantage of the fact that above the TO, where the observed stars lie, there is a tight relation between V magnitude and $b-y$; after correcting for the cluster reddening, this may be calibrated against temperature. The advantage is that an error of 0.05 mag in V translates into an error of only 20 -- 70 K in Teff. A possible drawback is the vulnerability of the method to binarity and \\citet{boe98}  note that SH 18 may have a variable radial velocity, indicating the presence of a companion. In this  note I use the V -- \\teff ~ calibration to reanalyse the Li data in M 92.  \\begin{figure}[t!] \\resizebox{\\hsize}{!}{\\includegraphics[clip=true]{pl_v_teff_M92.eps}} \\caption{The adopted V -- \\teff ~calibration: the solid line is the fiducial sequence for M92,  filled symbols correspond to the star in Table \\ref{liabun}} \\label{vteff} \\end{figure}  \\begin{table*} \\caption{Equivalent widths and Li abundances for TO stars in M 92} \\label{liabun} \\begin{center} \\begin{tabular}{lrccrrrrrr} \\hline \\\\ star  & EW &  $\\sigma_{EW}$ & T$\\rm _{eff}$ & A(Li) & $\\sigma_{Li}$ & A(Li) &  A(Li)  & A(Li) & A(Li)\\\\ & pm & pm             & K           &       &               &  TC            &  NLTE & B98 (K93) &  B98 (C83) \\\\ (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) (10) \\\\ \\\\ \\hline \\\\ SH 18  & 5.93   &  0.69 & 6050  & 2.53 & 0.07 & 2.55 & 2.57  & 2.57 & 2.45   \\\\ SH 21  & 2.50   &  0.62 & 5921  & 2.00 & 0.13 & 2.04 & 2.06  & 2.06 & 1.96   \\\\ SH 34  & 5.40   &  1.41 & 5956  & 2.38 & 0.17 & 2.42 & 2.43  & 2.36 & 2.22   \\\\ SH 46  & 2.87   &  0.74 & 6295  & 2.34 & 0.14 & 2.34 & 2.36  & 2.17 & 2.04   \\\\ SH 60  & 4.08   &  1.79 & 6171  & 2.51 & 0.23 & 2.52 & 2.54  & 2.42 & 2.30   \\\\ SH 80  & $<2.20$   &  1.10 & 6557  & $<2.39$ & 0.34  &  &    & $<2.23$ & $<2.16$   \\\\ SH 350 & 2.88   &  1.17 & 6102  & 2.20 & 0.25 & 2.22 & 2.23  & 2.13 & 2.01   \\\\ \\\\ \\hline \\end{tabular} \\end{center} \\end{table*}   \\begin{figure}[t!] \\resizebox{\\hsize}{!}{\\includegraphics[clip=true]{plot_li_teff_M92.eps}} \\caption{Li abundances, corrected for standard depletion and NLTE effects, versus effective temperature for TO stars in M92. The points are labelled with the star's SH number. } \\label{li_teff} \\end{figure} ",
        "conclusions": "From this reanalysis of existing data of M92 I conclude that there is no strong support for an intrinsic dispersion in Li abundances; however, a dispersion of the order of 0.18 dex would go unnoticed with the present errors. The mean Li abundance is in good agreement with that found for field stars and for TO stars of NGC 6397, which has a metallicity comparable to that of M92. I believe that the simplest hypothesis is that Li is constant in M92 and at the level of the {\\em Spite plateau}. However, the main purpose of this  note is to encourage observers to obtain spectra with higher S/N ratio in order to check if this is really the case. It is clear that statistical analysis and Monte Carlo simulations cannot make up for inadequate observations. The presently-available observations of M 92 cannot prove nor disprove the existence of an ``intrinsic'' dispersion. In particular, if SH 80 is really Li-depleted, it would be the first Li-depleted dwarf so-far detected in a metal-poor globular cluster and could testify in favour of Li-depletion mechanisms in metal-poor dwarfs. It is also obvious that an instrument like HIRES cannot tackle the problem without a prohibitive investment of telescope time. The new generation of medium resolution spectrographs operating on 8m telescopes, like ESI at Keck \\citep{1998SPIE.3355...48E}, or proposed, like AVES for VLT \\citep{2000SPIE.4008..167P}, have the capability to solve the problem. For example, ESI, with an entrance slit of 0\\farcs{3} which projects to 1.95 pixels, provides a resolving power of 0.032 nm per resolution element, which does not fully resolve the Li doublet, but is enough to measure an accurate equivalent width. There is no exposure time calculator for ESI, but a V=10.55 star provides 56650 $e^{-}{\\mathrm n \\mathrm m} ^{-1} {\\mathrm s}^{-1}$; scaling this number for a V=18.278 star (SH 80) one obtains 46 $e^{-}{\\mathrm n \\mathrm m} ^{-1} {\\mathrm s}^{-1}$, therefore in one hour of exposure  the S/N $ \\sim 73$ per resolution element. Assuming one can reach a S/N $\\sim  140$ adding 4 one hour exposures the Cayrel formula \\citep{cayrel88} provides an error on the equivalent width of 0.24 pm, i.e. comparable to what was obtained by \\citet{Bonifacio} for the much brighter stars in NGC 6397. AVES has been proposed for the VLT and could not obviously observe M92; however the design of AVES can easily be modified for a different telescope and its performances without adaptive optics would be similar to those of ESI and better with adaptive optics. Therefore it is from these medium resolution spectrographs that we must expect accurate abundances for TO stars in Galactic Globular Clusters in the near future."
    },
    "0209/astro-ph0209328_arXiv.txt": {
        "abstract": "Detailed Monte Carlo Inversion analysis of the spectral lines from three Lyman limit systems (LLS)\\, [$N$(H\\,{\\sc i}) $\\ga 1.0\\times10^{17}$ \\cm] and nine lower $N$(H\\,{\\sc i}) systems [$2\\times10^{14}$ \\cm $\\la N$({\\rm H}\\,{\\sc i}) $\\la 2\\times10^{16}$ \\cm] observed in the VLT/UVES spectra of Q0347--3819 (in the range $2.21 \\leq z \\leq 3.14$) and of APM BR J0307--4945 (at $z = 4.21$ and 4.81) is presented. Combined with the results from a previous work, the analyzed LLSs show that they are a {\\it heterogeneous} population originating in different environments. A functional dependence of the line-of-sight velocity dispersion $\\sigma_{\\rm v}$ on the absorber size $L$ is confirmed: the majority of the analyzed systems follow the scaling relation $\\sigma_{\\rm v} \\sim (N_{\\rm H}\\,L)^{0.3}$ (with $N_{\\rm H}$ being the total gas column density). This means that most absorbers may be related to virialized systems like galaxies or their halos. Previously noted enhancement of the metal content in small size systems is also confirmed: metallicities of $Z \\sim (1/3-1/2)\\,Z_\\odot$\\, are found in systems with $L \\la 0.4$ kpc, whereas we observe much lower metal abundances in systems with larger linear sizes. For the first time in LLSs, a pronounced [$\\alpha$-element/iron-peak] enrichment is revealed: the absorber at \\zabs = 2.21 shows [O/Fe]  = $0.65\\pm0.11$, [Si/Fe] = $0.51\\pm0.11$, and [Mg/Fe] = $0.38\\pm0.11$. Several absorption systems exhibit characteristics which are very similar to that observed in high-velocity clouds in the Milky Way and may be considered as high-redshift counterparts of Galactic HVCs. ",
        "introduction": "With the present  paper we continue to study  the chemical composition and the kinematic characteristics of quasar absorption systems using a new computational procedure, -- the Monte Carlo Inversion algorithm (MCI), -- developed earlier in a series of papers [see Levshakov, Agafonova \\& Kegel (2000); hereafter LAK]. The MCI technique allows us to recover self-consistently the physical parameters of the intervening gas cloud (such as the average gas number density $n_0$, the column densities for different species $N_{\\rm a}$, the kinetic temperature $T_{\\rm kin}$, the metal abundances $Z_{\\rm a}$, and the linear size $L$), the statistical characteristics of the underlying hydrodynamical fields (such as the line-of-sight velocity dispersion $\\sigma_{\\rm v}$, and the density dispersion $\\sigma_{\\rm y}$), and the line of sight density $n_{\\rm H}(x)$ and velocity $v(x)$ distributions (here $x$ is the dimensionless coordinate in units of $L$). Having this comprehensive information we are able to classify the absorbers more reliably and hence to obtain important clues concerning the physical conditions in intervening galaxies, galactic halos and large scale structure objects at high redshifts. Besides, it will also be possible to constrain the existing theories of the origin of galaxy formation since the observed statistics of the damped Ly$\\alpha$ (DLA) and Lyman limit (LLS) systems is believed to be a strong test of different cosmological models (e.g. Gardner et al. 2001; Prochaska \\& Wolfe 2001).  In the first part of our study (Levshakov et al. 2002a, hereafter Paper I) we reported results on the absorption systems at \\zabs = 1.87, 1.92 and 1.94  toward the HDF-South quasar J2233--606. These systems  exhibit many metal lines with quite complex structures. It was found that all profiles can be well described with an assumption of a homogeneous metallicity and a unique photoionizing background. According to the estimated sizes, velocity dispersions and  metal contents the absorbers at \\zabs = 1.92 and 1.87 were related to the galactic halos whereas the system at \\zabs = 1.94 was formed, more likely, in an irregular star-forming galaxy. It was also found, that the linear size and the line-of-sight velocity dispersion for all three absorbers obey a scaling relation of the same kind that can be expected for virialized systems.  The present paper deals with absorbers observed in the spectra of  Q0347--3819 ($z_{\\rm em} = 3.23$) and  APM BR J0307--4945 ($z_{\\rm em} = 4.75$, see \\S~2.1). Both spectra include several dozens of systems containing metals, but most of them are weak and severely blended and hence do not allow to estimate the underlying physical parameters with a reasonable accuracy. After preliminary analysis only 12 systems were chosen for the inversion with the MCI and their properties are described below.  The structure of the paper is as follows. \\S~2 describes the data sets. In \\S~3 our model assumptions and basic equations are specified. The estimated parameters for individual systems are given in \\S~4. The implication of the obtained results to the theories of LLS origin are discussed in \\S~5 and our conclusions are reported in \\S~6. Appendix contains a table with typical parameters of different absorbers which are referred to in the present study. ",
        "conclusions": "\\subsection{The origin of metal systems}  Metal systems with $N$(\\ion{H}{1}) $< 5\\times10^{16}$ \\cm\\, are usually believed to originate in galactic halos at different galactocentric distances. At low redshifts ($z < 1$) the galaxies associated with certain metallic absorptions (e.g. \\ion{C}{4}) can be in most cases identified directly (e.g. Chen et al. 2001a). Our results on absorption systems with $z \\ga 2$ also support this assumption: absorbers with \\zabs = 1.87 (Paper~I), 2.54, 2.65, 2.962, 2.98 (present paper) are produced by metal-enriched ($Z < 0.1\\,Z_\\odot$), hot ($T_{\\rm kin} \\ga 20000$ K), rarefied ($n_0 \\simeq 10^{-4} - 10^{-3}$ \\cmm) gas clouds which have typical linear sizes of $L > 10$~kpc. These parameters are consistent with contemporary models of galactic halos (e.g. Viegas, Friaca \\& Gruenwald 1999).  The nature of Lyman limit absorbers is less understood. Mo \\& Miralda-Escud\\'e (1996) associate them with cold photoionized clouds randomly moving in hot spherical halos. The clouds are supposed to form from the initial density inhomogeneities in the accreting intergalactic gas during its cooling. Both the cloud and the halo  obviously reveal the equal metallicity since they are formed from the same gas. In our study two LLSs with \\zabs = 1.92 (Paper~I) and 2.81 (present paper) can be related to the absorbers of this type.  However, this scenario obviously fails to explain metal abundant ($Z > 0.1\\,Z_\\odot$) systems since it is hard to understand how the whole halo can be metal-enriched to such a high level. It was shown by hydrodynamic simulations (e.g. Katz et al. 1996; Gardner et al. 2001) that LLSs can also arise on lines of sight that pass through small protogalaxies. We found two systems with \\zabs = 1.94 (Paper~I) and 4.21 (present paper) that can be explained within this framework. These metal-rich ($Z \\simeq 1/3\\,Z_\\odot$) absorbers with the sizes of several kpc are probably hosted by objects that may be akin to the local compact blue galaxies.  Some absorbers in our present study (\\zabs = 2.21, 2.965, and, possibly, 2.89) reveal small linear sizes ($L < 1$ kpc) together with very high metal content ($Z \\simeq 1/2\\,Z_\\odot$). These three systems may be explained in the framework of the process known as a galactic fountain~: metal-enriched (supernova-heated) gas arises from the inner region of a galaxy and condenses into the clouds within the hot galactic halo. After formation, clouds fall back toward the galaxy centre because of their higher density. It is supposed that high-metallicity HVCs observed in the Milky Way halo are formed by this mechanism (Bregman 1980). The HVCs are common objects in our Galaxy and are detected in every longitude and latitude region. If galactic fountain works also in distant galaxies, it would be quite probable to encounter such a cloud on the line of sight which intersects the galactic halo, as also discussed by Charlton et al. (2001). Another type of HVCs -- hot, highly ionized clouds with sizes of several kiloparsecs -- is represented by the absorption system at \\zabs = 2.848. The origin of this type of HVCs is uncertain, but they may be produced by the intergalactic metal-enriched gas falling onto metal-poor galactic halos.  Measured abundances of C and Si are depicted versus logarithmic sizes of the studied systems in Fig.~17. Systematically higher metal abundances are seen in compact systems with linear sizes $L < 4$ kpc. This result seems to indicate that the more effective metal enrichment occurs within relatively compact regions.  Our results show that Lyman limit systems are a {\\it heterogeneous} population which is formed in at least three different environments. This should be taken into account when statistics of LLSs is used to verify different models in hydrodynamic cosmological simulations.   \\subsection{$\\sigma_{\\rm v} - N_{\\rm H}\\,L$ relation}  If QSO metal systems are formed in gas clouds gravitationally bound with intervening galaxies, the internal kinematics of the QSO absorbers should be closely related to the total masses of the host galaxies. In case of galactic population, different types of galaxies show different scaling relations between the linear size and the velocity width of emission lines (e.g., Mall\\'en-Ornelas et al. 1999). Possible correlation between the absorber linear size $L$ and its line-of-sight velocity dispersion $\\sigma_{\\rm v}$ was also mentioned in Paper~I.  The correlation between $\\sigma_{\\rm v}$ and $L$ stems from the virial theorem which states~: \\begin{equation} \\sigma^2_{\\rm v} \\sim \\frac{M}{L} \\sim n_0\\,L^2 = N_{\\rm H}\\,L\\; . \\label{eq:E1} \\end{equation} Assuming that the gas systems are in quasi-equilibrium, one can expect $\\sigma_{\\rm v} \\sim \\sqrt{N_{\\rm H}\\,L}$.  In Fig.~18 we examine our systems by comparing their kinematics ($\\sigma_{\\rm v}$) with measured sizes ($L$) and total gas column densities ($N_{\\rm H}$). Shown are the data for all QSO absorbers studied in Paper~I and in the present paper except for the systems at \\zabs = 2.848 and 2.899 (Table~3) which show inhomogeneous metallicities. It is seen that in the $\\log (\\sigma_{\\rm v})$ versus $\\log (N_{\\rm H}\\,L)$ diagram, most systems with linear sizes $L > 1$ kpc lie along the line with the slope $\\kappa = 0.30\\pm0.03$ (1 $\\sigma$ c.l.).  Taking into account that we know neither the impact parameters nor the halo density distributions, this result can be considered as a quite good fit to the relation (1). Hence we may conclude that most absorbers with $L > 1$ kpc are gravitationally bound with systems that appear to be in virial equilibrium at the cosmic time when the corresponding Ly$\\alpha$ absorbers were formed. The possible consequence of this conclusion is that since the most metal rich absorbers identified in the QSO spectra arise in the galactic systems the question whether the intergalactic matter is metal enriched or pristine remains still open.   \\subsection{[$\\alpha$-element/iron-peak] ratio}  The metal abundances measured in the \\zabs = 2.21 LLS (Table~1) can be used to estimate the $\\alpha$-element to the iron-peak group ratio which is a good indicator of the chemical evolutionary status of high redshift gas clouds. During the chemical evolution, heavy elements produced in stars show different nucleosynthetic histories so that their relative abundances vary with cosmic time.  Oxygen and other $\\alpha$-chain elements are mainly produced by Type~II SNe, while iron is also a product of Type~Ia SNe which have longer evolution scales. In the early stages of the chemical evolution of galaxies ($\\Delta t \\la 2\\times10^7$ yr) the interstellar gas is likely enriched by Type~II SNe products, while at $\\Delta t \\ga 10^8$ yr, the [$\\alpha$/Fe] ratio should decline. Observations reveal both low [e.g. $\\simeq 0.1-0.2$ in the \\zabs = 3.390 dust-free DLA (Q0000--2621; Molaro et al. 2001) and in the \\zabs = 3.386 DLA (Q0201+1120; Ellison et al. 2000)], and high [e.g. $\\simeq 0.7$ in the DLA I~Zw 18 (Levshakov, Kegel \\& Agafonova 2001) and $0.68\\pm0.08$ in the \\zabs = 3.025 DLA (Q0347--3819; Levshakov et al. 2002b)] ratios of [$\\alpha$-element/iron-peak].  Oxygen with its weak affinity with dust grains is a good tracer of the $\\alpha$-element abundances. Nevertheless, the intrinsic [$\\alpha$/Fe] ratio may be affected by depletion of iron since being a refractory element iron may be partly locked into dust grains. The dust content in the \\zabs = 2.21 LLS may not, however, be too high. The relative abundances of the $\\alpha$-elements O, Mg and Si are [Si/O] $= -0.14\\pm0.11$ and [Mg/O] = $-0.27\\pm0.11$. In Galactic stars the $\\alpha$-elements show the same behaviour relative to iron-peak elements (oversolar at [Fe/H] $\\la -1$; see, e.g., Gaswami \\& Prantzos 2000). We thus expect to find solar $\\alpha$-element ratios in dust-free absorbing regions, as observed, e.g., in the mentioned above \\zabs = 3.390 DLA toward Q0000--2620. A negative value of [Mg/O] found in this LLS may indicate the presence of some amount of dust with a depletion factor of about 0.2 dex for the magnesium abundance. If, however, only the gas-phase abundances of O and Fe are taken, the upper bound on the [O/Fe] ratio is $0.65\\pm0.11$, which is comparable with that found, for instance, in the \\zabs = 3.025 DLA toward Q0347--3819 where the dust-to-gas ratio is $\\simeq 1/30$ of the mean Galactic interstellar medium value (Levshakov et al. 2002b). The enrichment of the $\\alpha$-elements in the \\zabs = 2.21 LLS is also supported by the relative abundances of Si, Mg to Fe: [Si/Fe] = $0.51\\pm0.11$ and [Mg/Fe] = $0.38\\pm0.11$. Thus, the absorbing cloud at \\zabs = 2.21 appears to be a chemically young object."
    },
    "0209/astro-ph0209602_arXiv.txt": {
        "abstract": "We derive constraints on cosmological parameters and the properties of the lensing galaxies from gravitational lens statistics based on the final Cosmic Lens All Sky Survey (CLASS) data. For a flat universe with a classical cosmological constant, we find that the present matter fraction of the critical density is $\\Omega_{\\rm m}=0.31^{+0.27}_{-0.14}$ (68\\%) $^{+0.12}_{-0.10}$ (systematic). For a flat universe with a constant equation of state for dark energy $w = p_x({\\mbox{pressure}})/\\rho_x({\\mbox{energy density}})$, we find $w < -0.55^{+0.18}_{-0.11}$~(68\\%). ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209058_arXiv.txt": {
        "abstract": "The process of gravitational scattering of planetesimals by a massive protoplanetary embryo is explored theoretically. We propose a method to describe the evolution of the disk surface density, eccentricity, and inclination caused by the embryo-planetesimal interaction. It relies on the analytical treatment of the scattering in two extreme regimes of the planetesimal epicyclic velocities: shear-dominated (dynamically ``cold'') and dispersion-dominated (dynamically ``hot''). In the former, planetesimal scattering can be treated as a deterministic process. In the latter, scattering is mostly weak because of the large relative velocities of interacting bodies. This allows one to use the Fokker-Planck approximation and the two-body approximation to explore the disk evolution. We compare the results obtained by this method with the outcomes of the direct numerical integrations of planetesimal orbits and they agree quite well. In the intermediate velocity regime an approximate treatment of the disk evolution is proposed based on interpolation between the two extreme regimes. We also calculate the rate of embryo's mass growth in an inhomogeneous planetesimal disk and demonstrate that it is in agreement with both the simulations and earlier calculations. Finally we discuss the question of the direction of the embryo-planetesimal interaction in the dispersion-dominated regime and demonstrate that it is repulsive. This means that the embryo always forms a gap in the disk around it, which is in contrast with the results of other authors.  The machinery developed here will be applied to realistic protoplanetary systems in future papers. ",
        "introduction": "\\label{sect:intro}    This paper continues the line of investigation started in our previous work (Rafikov 2001, 2002a; hereafter Papers I and II) which was devoted to the treatment of planetesimal-planetesimal gravitational interactions. Here we consider the interaction between the growing protoplanetary embryo and the planetesimal disk. By embryo we imply in the present context a single body with mass $M_e$ much larger than the masses of individual planetesimals $m$. There are several reasons for studying this important problem separately from the mutual gravitational scattering of planetesimals.  First, planetesimal-planetesimal encounters in realistic protoplanetary disks usually occur in the dispersion-dominated regime, which applies when the relative approach velocity of two particles is bigger than the differential shear in the disk across the Hill (or tidal) radius. The Hill radius is defined as \\begin{eqnarray} r_H=a\\left(\\frac{m_1+m_2}{M_c}\\right)^{1/3}, \\label{eq:Hill_radius} \\end{eqnarray} where $a$ is a value of the semimajor axis at which the interaction takes place, and $m_1,m_2,$ and $M_c$ are the masses of interacting planetesimals and of the central star.  However, in the same protoplanetary disk gravitational interaction between the embryo and planetesimals could be in the opposite velocity regime --- shear-dominated --- when the planetesimal random motion is small compared to the shear across the Hill radius, simply because the embryo mass and therefore the Hill radius is much larger. Indeed, in the case of embryo-planetesimal interactions Hill radius $R_H=a_e(M_e/M_c)^{1/3}\\gg r_H$ (here $a_e$ is a semimajor axis of the embryo) as a consequence of $M_e\\gg m_i$. Thus, reduced (normalized in Hill coordinates) values of random velocities in the embryo-planetesimal case are smaller by a factor $[(m_1+m_2)/M_e]^{1/3}\\ll 1$ than those corresponding to the planetesimal-planetesimal interactions.  Of course, it might be that the planetesimal disk has  already been so  excited dynamically that even embryo-planetesimal encounters are in the dispersion-dominated regime. Thus, we consider here both regimes of the embryo-planetesimal interaction --- shear- and dispersion-dominated.  Second, embryo-planetesimal interactions are complicated by the presence of a special type of orbits in a $3$-body problem --- the so-called horseshoe (or librating) orbits (H\\'enon \\& Petit 1986; Murray \\& Dermott 1999). Planetesimals on these orbits do not perform the usual circulating motion which is characteristic of passing orbits (the most important case for planetesimal-planetesimal scattering) but a librating one. This horseshoe motion can only occur when the difference in semimajor axes of interacting bodies is smaller than their Hill radius. For planetesimal-planetesimal interactions $r_H$ is negligible compared to the scale of surface density variations or the radial epicyclic excursion of an individual planetesimal. Thus horseshoe orbits are unimportant in this case. However, the Hill radius of the embryo-planetesimal interaction $R_H$ can be comparable to the length scale of the disk inhomogeneities caused by the embryo. Thus the phenomenon of horseshoe motion can be crucial for the planetesimal dynamics near the embryo (see below \\S \\ref{subsubsect:hs_separation}).  Third, as we have already mentioned in Paper II, planetesimal-planetesimal scattering is described in terms of the disk properties {\\it averaged} over some region of the disk, which diminishes the importance of the details of spatial distributions of disk properties. On the contrary, in the case of the embryo-planetesimal interaction we are interested in details of the {\\it spatial behavior} of various quantities characterizing the state of the disk and they are the primary goal of our present study.   All these complications preclude the direct application of the results obtained in Paper II to the present consideration. However the general analytical approach to the treatment of planetesimal disk evolution developed there remains valid and we will employ it in this paper.   Numerical orbit integrations (Tanaka \\& Ida 1996, 1997) and N-body simulations (Ida \\& Makino 1993) provide an alternative and important route of studying embryo-planetesimal interactions. Their drawback is their intrinsically low speed and inability to treat large number of planetesimals. However, since the physics is incorporated in them on a very basic level with the minimum of additional assumptions they can provide us with robust predictions. To use this advantage of numerical methods and to avoid their handicaps we employ the following methodology: we provide a self-consistent {\\it analytical} description of the embryo-planetesimal interaction in different velocity regimes. To check this description and to verify the validity of the simplifying assumptions utilized in its development we have used numerical orbit integrations performed for several sets of typical planetesimal disk parameters. After we make sure that our theory works well for these sets of parameters we can use it for others as well and be confident of its reliability. What we gain by this approach is the speed of computation and ability to explore the whole space of important physical variables.   The condition on the embryo's mass, $M_e\\gg m$, has important dynamical  implications. In many applications it justifies the neglect of the embryo's recoil resulting from planetesimal  scattering. Also, dynamical friction between the embryo and planetesimals will tend to produce random energy equipartition (Stewart \\& Wetherill 1988; Wetherill \\& Stewart 1989) which means that embryo's eccentricity and inclination are most likely to be negligibly small. Thus, we will assume in this paper that the embryo moves on a fixed circular orbit and its eccentricity and inclination are zero. We will also consider the embryo to be isolated from the gravitational effects of other massive bodies which may be growing nearby, an assumption which can easily be abandoned in future work. Throughout the paper we neglect the presence of any resonant effects. This is justifiable if frequent planetesimal-planetesimal encounters can destroy any commensurabilities with the embryo's rotation period. The embryo's recoil in the course of planetesimal scattering and distant embryo-embryo interactions would also help to do that. We leave for the future the clarification of conditions necessary for employing this simplification.   It will be convenient to use the embryo's Hill radius $R_H$ as a unit of length in our study. We introduce the Hill orbital elements of a planetesimal evaluated at large azimuthal distance from the embryo --- the difference in semimajor axes $H$, eccentricity $\\tilde e$, and inclination $\\tilde i$ relative to the embryo's orbit at semimajor axis $a_e$: \\begin{eqnarray} H=\\frac{a-a_e}{R_H},~~~\\tilde e=e\\frac{a_e}{R_H},~~~\\tilde i =i\\frac{a_e}{R_H}, \\label{eq:new_vars} \\end{eqnarray} and use them as planetesimal coordinates. The distribution function of the orbital elements $\\tilde e$ and $\\tilde i$ will be assumed to have a Rayleigh form with dispersions $\\tilde \\sigma_e$ and $\\tilde \\sigma_i$: \\begin{equation} \\psi(\\tilde e,\\tilde i)d\\tilde e d\\tilde i=\\frac{\\tilde e d\\tilde e ~\\tilde i d\\tilde i}{\\tilde \\sigma_{e}^2 \\tilde \\sigma_{i}^2}\\exp\\left[-\\frac{\\tilde e^2}{2\\tilde \\sigma_{e}^2} -\\frac{\\tilde i^2}{2\\tilde \\sigma_{i}^2}\\right]. \\label{eq:Railey} \\end{equation} We will also be using the dimensionless surface number density of guiding centers $N(H)$ to characterize the planetesimal spatial distribution.  Although we focus on a single embryo, for many applications we can treat the embryo using a continuous form of the evolution equations. Fro example, we may assume that the discrete surface number density of the embryo is given by \\begin{eqnarray} N_{em}(H)=\\frac{1}{2\\pi}\\delta(m-M_e)\\delta(H). \\label{eq:em_surf_density} \\end{eqnarray} Since planetesimal-planetesimal interactions are not important here it will be enough to consider a single-mass planetesimal population.  The three-body interaction in the Hill approximation preserves a certain combination of relative orbital elements of interacting bodies called the Jacobi constant (Goldreich \\& Tremaine 1980; H\\'enon \\& Petit 1986): \\begin{eqnarray} J=\\tilde e^2+\\tilde i^2-\\frac{3}{4}H^2+2\\phi_e, \\label{eq:Jacobi} \\end{eqnarray} where $\\phi_e$ is the gravitational potential of the embryo, which can be neglected far from the embryo. For embryo-planetesimal scattering one can introduce the concept of integrated Jacobi constant of the whole planetesimal population: \\begin{eqnarray} J^{tot}=\\int\\limits_{-\\infty}^{\\infty} \\left[2N(H)\\tilde \\sigma_{e}^2+2N(H)\\tilde \\sigma_{i}^2 -\\frac{3}{4}N(H)H^2\\right]dH. \\label{eq:int_Jacobi} \\end{eqnarray} This quantity should be conserved  because (1) each individual planetesimal scattering off the embryo conserves the Jacobi constant of the relative motion, and (2) embryo's random motion is negligible, which means that relative eccentricity, inclination, and difference in semimajor axes are determined by planetesimal orbital parameters only. We will use the conservation of this quantity and of the total number of planetesimals  (we neglect their coagulation at this point) \\begin{eqnarray} N^{tot}=\\int\\limits_{-\\infty}^{\\infty}N(H)dH. \\label{eq:int_number} \\end{eqnarray} as checks of our evolution equations.   The orbit integrations that we use are performed by solving Hill equations numerically. We have integrated the evolution of the system [equations of osculating orbital elements evolution (11) of Paper II] using fourth order Runge-Kutta integrator (Press \\etal 1988). Unlike similar calculations of Tanaka \\& Ida (1996, 1997) our orbit integrations  do not employ additional analytical simplifications to avoid possible biases. In a typical integration the Jacobi constant is conserved with fractional accuracy $10^{-8}-10^{-12}$. The results of these orbit integrations and their comparisons with theoretical predictions will be presented in the following sections.   We devote \\S \\ref{subsect:em_shear} to studying the shear-dominated case and \\S \\ref{subsect:em_dispersion} to exploring the dispersion-dominated case. The velocity regime intermediate between them is addressed in Appendix \\ref{app:intermediate_velocity}. We discuss some general features of the embryo-planetesimal interaction in \\S \\ref{sect:repulsion}. Some auxiliary results are presented in appendices: Appendix \\ref{app:scat_erobab} contains the derivation of the probability distribution of scattered semimajor axes in the dispersion-dominated regime, while in Appendix \\ref{app:accr_rate} we calculate the embryo's accretion rate in different velocity regimes. ",
        "conclusions": "\\label{sect:conclusion}   We have studied the embryo-planetesimal interaction in the gravitational field of a central star. Two different cases were explored: when the interaction between the embryo and planetesimals occurs in the shear-dominated  and in the dispersion-dominated regimes. The treatment of the first case parallels that of Paper I but is complicated by the fact that now we explicitly include the evolution of not only the surface density but also the eccentricity and inclination of planetesimals in the disk.  Our study of the dispersion-dominated regime relies on the methods of kinetic theory, and it uses many of the results obtained in Paper II. However our present treatment is more refined since the description of embryo-planetesimal scattering requires clarifying many details which were not important for the planetesimal-planetesimal interactions. In particular we have to study not only passing but also horseshoe orbits of planetesimals to determine the spatial distribution of the disk properties. To do this we propose a condition which separates the horseshoe and passing orbits and check its viability using numerical orbit integrations. Angular momentum exchange between the embryo and planetesimal long before and after their closest approach turns out to be important for the scattering on passing orbits near the horseshoe boundary. We illustrate this point by comparing the analytical scattering probability function with the one obtained from numerical integrations. A simple method to  account for this effect in our Fokker-Planck approach is proposed. Taken together all these refinements are shown to provide rather good agreement with the results of the numerical orbit integrations. Thus we hope to have grasped the most important features of the embryo-planetesimal interaction by our theoretical approach.  This does not mean that our treatment of the embryo-planetesimal interaction is complete. We have only focussed on the most important, dominant effects, and there is certainly room for additional refinements, which would farther improve the agreement with numerical results.  The calculation of the scattering coefficients to the next order in $1/\\ln\\Lambda$ would provide us with subdominant corrections which might be important for modest values of $\\tilde\\sigma_e$ and $\\tilde\\sigma_i$. One can certainly do a better job in treating distant encounters, calculating various coefficients entering formulae (\\ref{eq:my_cond}), (\\ref{eq:h_hprime}), etc. or the interpolating functions of Appendix \\ref{app:intermediate_velocity}, using a larger set of numerical orbit integrations. On a somewhat deeper level, one can try to come up with a more sophisticated treatment of the horseshoe-passing orbits separation (instead of the complete spatial separation of these two types of orbits assumed in this paper). Our purely deterministic treatment of the shear-dominated regime can also be improved, which would ameliorate the comparison with numerical results in the intermediate velocity regime (see Appendix \\ref{app:intermediate_velocity}). The recoil of the embryo and the excitation of the embryo's eccentricity and inclination in the course of the planetesimal gravitational scattering can be important in some applications, such as the embryo's migration (Tanaka \\& Ida 1999) or its interaction with the embryos nearby (Tanaka \\& Ida 1997). Our treatment relies on the use of the Schwarzschild velocity distribution which was demonstrated to be applicable in the dispersion-dominated regime (Greenzweig \\& Lissauer 1992), but the deviations from this assumption could be important e.g. in the intermediate velocity regime, and this subject can also be pursued farther. All these refinements would better the quantitative agreement between the analytical theory and numerical results. But reasonably good accord is provided even by our basic treatment developed here, especially in the dispersion-dominated case where the assumptions we make are the most justifiable.    We also dwell upon the question of the direction of the embryo-planetesimal interaction, namely whether it leads to the repulsion of planetesimal orbits from the embryo or to their attraction. The latter outcome has been favored in some scenarios (Ida \\etal 2000) and is based on the fact that in the dispersion-dominated regime embryo {\\it on average} tends to attract planetesimal semimajor axes toward its orbit. We demonstrate, however, that the average change of planetesimal semimajor axes cannot serve a standard for determining the direction of the embryo-planetesimal interaction because the transport of the angular momentum (associated with the changes in semimajor axes) is not the same as the bulk motion of the disk material. We propose our own criterion for judging the embryo-planetesimal scattering outcome, and show that the embryo always {\\it repels} planetesimals rather than attracts them in an initially homogeneous disk thereby carving out a gap in the distribution of the planetesimal semimajor axes [which is in contrast to claims by Ida \\etal (2000)].  Our own numerical results support this conclusion.  The combination of our results for planetesimal-planetesimal gravitational scattering presented in Paper II and the theory for the embryo-planetesimal interaction developed here now allows us to study the planetesimal disk dynamics with these processes operating simultaneously. It also provides a framework to which various refinements and additional physical mechanisms can be naturally added. Our results are not restricted in applications to the problem of the formation of planetary systems but can also be used for studying their present day evolution, e.g. the dynamics of asteroid and Kuiper belts affected by massive bodies inside or near them. Our results for the accretion rate of massive body immersed in inhomogeneous planetesimal disk (\\S \\ref{sect:accretion} \\& Appendix \\ref{app:accr_rate}) allow us also to include self-consistently the embryo's mass growth into consideration. We will examine protoplanetary disk evolution with all these effects working together in a future work (Rafikov 2002b)."
    },
    "0209/astro-ph0209572_arXiv.txt": {
        "abstract": "The effects of radiation drag force on the structure of relativistic electron-positron and electron-proton outflows are considered within the one-fluid approximation for quasi-monopole cold outflow. It is shown that for a Poynting-dominated case the drag force does not change the particle energy inside the fast magnetosonic surface. In this region the action of the drag results in a diminishing of the Poynting flux, not the particle flux. Outside the fast magnetosonic surface, for intermediate photon density the drag force may result in additional acceleration of the plasma. This acceleration is a result of the disturbance of magnetic surfaces under the action of the drag. At even larger distances particles are not frozen into the magnetic field and the drag force decelerates them efficiently.  In the case of extreme photon densities, the disturbance of magnetic surfaces becomes large and the drag force changes the total energy flux significantly, the particles becoming nonrelativistic.  We find that for Active Galactic Nuclei the photon density is too low to disturb the parameters of an ideal MHD outflow. The drag action may result in additional acceleration of outgoing plasma only for central engines with very high luminosities. For cosmological gamma-ray bursts the drag force can strongly affect the process of formation of a Poynting-dominated outflow. ",
        "introduction": "Magnetohydrodynamic (MHD) models are now developed intensively in theories of the magnetospheres of rotating supermassive black holes ($M \\sim 10^{8}\\hbox{--}10^{9}M_{\\odot}$, $B_0 \\sim 10^4$ G), which are believed to reside in central engines of Active Galactic Nuclei (AGNs) and quasars (Begelman, Blandford \\& Rees 1984; Thorne, Price \\& Macdonald 1986). In particular, it is the MHD model that is the most promising in the problem of the origin and stability of jets. Indeed, the MHD approach explains both the energy source (the rotational energy of the compact object) and the mechanism of the energy and angular momentum loss (for an overview, see e.g. Blandford 2002). Observational evidence in favor of MHD models was recently found in the possible presence of toroidal magnetic fields in jets (Gabuzda et al 1992; Gabuzda et al 1999). Magnetically-dominated outflows are also believed to be responsible for the energy transport in cosmological gamma-ray bursts (M\\'esz\\'aros \\& Rees 1997; Lee, Wijers \\& Brown 1998; van Putten \\& Levinson 2003), when energy is released in the merging of black holes or neutron stars ($M \\sim M_{\\odot}$, $B_0 \\sim 10^{15}$ G).  It has been suggested that the density of photons in the vicinity of the central engine is so high that they may drastically change the characteristics of the ideal MHD outflow. For example, they may result in extensive $e^+e^-$ pair creation (Svensson 1984), acceleration of low-energy pairs by the radiation drag force (Phinney 1982; Turolla, Nobili \\& Calvani 1986; Beloborodov 1999) and deceleration of high-energy particles (Melia \\& K\\\"onigl 1989; Sikora et al 1996). In other words, a self-consistent consideration should take the drag force into account.  So far the two processes  -- the ideal MHD acceleration and the action of external photons -- have been considered independently. The first step to combine them was made by Li, Begelman \\& Chiueh (1992). In particular, it was shown how the equations can be integrated in a conical geometry (which is impossible in the general case). On the other hand, the consideration was performed within the approximation of a fixed poloidal magnetic field. Under this assumption the fast magnetosonic surface of a cold flow is shifted to infinity (Michel 1969; Kennel, Fujimura \\& Okamoto 1976; Lery et al 1998). As a result, it was impossible to analyze the effects of radiation drag on the position of the fast magnetosonic surface and the properties of the outflow outside this surface.  The main goal of this paper is to determine more carefully the radiation drag effects on a magnetically-dominated outflow. To describe analytically the effects of radiation drag, including simultaneously the disturbance of the magnetic surfaces we consider here a quasi-monopole outflow. For AGNs, such geometry in the immediate vicinity of the central engine was recently confirmed by direct observations (Junior, Biretta \\& Livio 1999). In other words, in the zeroth approximation (i.e., without drag) we use the analytical solution for a magnetically-dominated MHD outflow (Beskin, Kuznetsova \\& Rafikov 1998, hereafter Paper I), in which the fast magnetosonic surface is located at a finite distance from the origin.  For simplicity we consider the following model of the radiation field in the vicinity of the central engine. First, we notice that for ultra-relativistic particles the energy of a photon propagating nearly along the particle trajectory remains almost the same after a collision. This means that the drag force from these photons is small. Thus, only the isotropic component of the photon field contributes substantially to the drag force. Hence, in our geometry with a strong central source of photons and a monopole outflow of particles, only a small fraction of photons (the isotropic component of the photon field) interacts efficiently with the particles, producing inverse Compton photons with energies ${\\cal E}_{\\rm IC} \\sim \\gamma^2{\\cal E}_{\\rm ph}$.  The isotropic component can be produced, firstly, by the outer part of the accretion disk and, secondly, by external sources. It can be modeled as \\begin{equation} U = U_{\\rm iso} = U_{\\rm A}\\left(\\frac{r}{R_{\\rm L}}\\right)^{-n} + U_{\\rm ext}, \\label{u_large} \\end{equation} where $R _{\\rm L} = c/\\Omega$ is the radius of the light cylinder, $U_{\\rm A} = U(R_{\\rm L}) = L_{\\rm tot}/(4\\pi R_{\\rm L}^{2}c)$, and $n \\approx 3$ (for more details see, e.g., Sikora et al 1996). Here the first term describes the radiation from the outer parts of the disk, $ r_{\\rm rad} > r$, while the second one corresponds to the homogeneous external radiation. For AGNs this can be due to clouds located at a distance $r_{\\rm cloud} \\sim 1$pc from the central engine and reradiating $k L_{\\rm tot}$ of the total luminosity ($k \\sim 10\\%$). In this case \\begin{equation} U_{\\rm ext} = k\\frac{L_{\\rm tot}}{4\\pi r_{\\rm cloud}^2c}. \\label{u1} \\end{equation} However, this model only makes physical sense at distances less than $r_{\\rm cloud}$, and the term vanishes at larger distances.  Finally, as some arguments exist both in favor of (Reynolds et al 1996; Hirotani et al 1999) and against (Sikora \\& Madejski 2000) the leading role of $e^+e^-$ plasma in relativistic jets, in what follows we consider both electron-positron and electron-proton outflows.  In Section 2 we formulate the basic equations describing a quasi-monopole outflow of relativistic plasma in two-fluid and one-fluid approximations. Then in Section 3 we analyze the main properties of an electron-positron outflow. A similar analysis for electron-proton plasma is produced in Section 4. Finally in Section 5 we consider the effects of radiation drag for real astrophysical objects. ",
        "conclusions": "We have demonstrated how for a simple geometry it is possible to determine a small radiation-drag-force correction to the one-fluid ideal MHD outflow. The disturbance of magnetic surfaces was self-consistently taken into consideration. As a result, it is possible to characterize the general influence of the drag action on the magnetic field structure for an ideal magnetically-dominated quasi-monopole cold outflow and to determine under what circumstances radiation drag is important.  As demonstrated above, the characteristics of the flow are determined by two main parameters, namely the compactness parameter $l_{\\rm A}$ (\\ref{A}) (which is proportional to the photon density) and the magnetization parameter $\\sigma$ (\\ref{sigma}). If the photon density is low, so that the compactness parameter is small $l_{\\rm A} \\ll l_{\\rm cr}(\\sigma)$, the action of the drag force is negligible, while for a high photon density $l_{\\rm A} \\gg l_{\\rm cr}(\\sigma)$, particles are additionally accelerated outside the fast magnetosonic surface.  In particular, for $l_{\\rm cr} \\ll l_{\\rm A} \\ll l_{\\rm max}$ the increase of the drag force results in an increase of the outgoing plasma energy ${\\cal E}_{\\rm max} \\approx \\gamma_{\\rm max}m_{\\rm e, p}c^2$, but the disturbance of magnetic surfaces is small ($\\varepsilon f \\ll 1$). For $l_{\\rm A} \\sim l_{\\rm max}$ an increase of the photon density results in the increase of collimation up to values $\\varepsilon f \\sim 1$, but the particle energy remains near the saturation value ${\\cal E}_{\\rm sup}$. Finally, for a very high photon density $l_{\\rm A} \\gg l_{\\rm max}$ an effective collimation of magnetic surfaces becomes possible, but in this case the drag force substantially diminishes the flux of electromagnetic energy inside the fast magnetosonic surface. As a result, for $l_{\\rm A} \\gg l_{\\rm max}$ almost all the energy of the electromagnetic field is lost via the inverse Compton interaction  of particles with external photons. For this reason, the very existence of a magnetically-dominated flow becomes impossible. The dependence of the maximum particle energy ${\\cal E}_{\\rm max} = \\gamma_{\\rm max}m_{\\rm e, p}c^2$ on the compactness parameter $l_{\\rm A}$ is shown in Fig. 3.  We now consider several astrophysical applications.  \\subsection{Active Galactic Nuclei}  For AGNs (the central engine is assumed to be a rotating black hole with mass $M \\sim 10^9M_{\\odot}$, $R \\sim 10^{14}$ cm, the total luminosity $L \\sim 10^{45}$ erg s$^{-1}$, $B_0 \\sim 10^4$ G) the compactness parameter $l_{\\rm A}$ (\\ref{A}) can be evaluated as \\begin{equation} l_{\\rm A} \\approx 30M_9^{-1}\\left(\\frac{\\Omega R}{c}\\right)L_{45}. \\end{equation} In the Michel magnetization parameter $\\sigma$ (\\ref{sigma}) \\begin{equation} \\sigma \\approx 10^{14}\\lambda^{-1}M_9B_4\\left(\\frac{\\Omega R}{c}\\right), \\end{equation} the main uncertainty comes from the multiplication parameter $\\lambda$, i.e., in the particle number density $n$. Indeed, for an electron-positron outflow this value depends on the efficiency of pair creation in the magnetosphere of a black hole, which is still undetermined. In particular, this process depends on the density and energies of the photons in the immediate vicinity of the black hole. As a result, if the hard-photon density is not high, then the multiplication parameter is small ($\\lambda \\sim 10 - 100$; Beskin, Istomin \\& Pariev 1992; Hirotani \\& Okamoto 1998). In this case for $(\\Omega R/c) \\sim 0.1$--$0.01$ we have $\\sigma \\sim 10^9 - 10^{12}$, so that $l_{\\rm cr} \\sim 10^3$--$10^4$. On the other hand, if the density of photons with energies ${\\cal E}_{\\gamma} > 1$MeV is high enough, direct particle creation $\\gamma + \\gamma \\rightarrow e^+ +e^-$ results in an increase of the particle density (Svensson 1984). This gives $\\sigma \\sim 10 - 10^3$, and hence $l_{\\rm cr} \\sim 10$ for an electron-positron outflow.  From a theoretical point of view, the most interesting result here is the possibility of an additional acceleration of particles outside the fast magnetosonic surface. Indeed, for a high enough photon density ($l_{\\rm A} \\sim 10 - 100$, i.e., for $L \\sim 10^{46} - 10^{48}$ erg s$^{-1}$) and a small magnetization parameter $\\sigma \\sim 10 - 100$, the compactness parameter $l_{\\rm A}$ can exceed the critical value $l_{\\rm cr}$ for an electron-positron outflow. In this case, according to Fig. $1b$, our analysis suggests that the kinetic luminosity of the relativistic jet should be proportional to $l_{\\rm A}^{2/3}\\propto L_{\\rm tot}$, where $L_{\\rm tot}$ in the total luminosity of the central engine. Kinetic luminosity is not easily determined from observations. However, observational evidence suggests that the radio luminosity of the jets is positively correlated with the luminosity of the central engine and the scatter of this correlation decreases towards larger luminosities (Baum, Zirbel \\& O'Dea 1995).  For an electron-proton outflow the magnetization parameter (\\ref{sigma}) can be rewritten in the form (Camenzind 1990) \\begin{equation} \\sigma = \\frac{m_{\\rm p}}{m_{\\rm e}}\\left(\\frac{\\Omega R}{c}\\right)^2 \\frac{B_0^2R^2}{c{\\dot M}} \\approx  3 \\times 10^4 \\left(\\frac{\\Omega R}{c}\\right)^2 B_4^2M_9^2\\left(\\frac{{\\dot M}}{0.1\\,M_{\\odot}/{\\rm yr}}\\right)^{-1}. \\end{equation} Here ${\\dot M} = 4\\pi nm_{\\rm p}R^2c$ is the mass ejection rate. Hence, for a high ejection rate (${\\dot M} > 0.1\\,M_{\\odot}$~yr $^{-1}$) the magnetization parameter $\\sigma < m_{\\rm p}/m_{\\rm e}$. In this case there is no acceleration of plasma. On the other hand, for low ejection rate ${\\dot M} < 0.1\\,M_{\\odot}/{\\rm yr}$ the magnetization parameter becomes too large for the drag force to be efficient.  Thus, the drag force can substantially disturb the MHD parameters of a Poynting-dominated outflow only for a very high luminosity of the central engine ($L_{\\rm tot} \\gg 10^{45}$~erg~s $^{-1}$) and only for an electron-positron outflow. In all other cases the action of the drag force remains negligible. In particular, the additional acceleration of particles outside the fast magnetosonic surface is not efficient.  \\subsection{Cosmological Gamma-Ray Bursts}  For cosmological gamma-ray bursts (the central engine is represented by the merger of very rapidly orbiting neutron stars or black holes with $M \\sim M_{\\odot}$, $R \\sim 10^{6}$ cm, total luminosity $L \\sim 10^{52}$ erg s$^{-1}$, $B_0 \\sim 10^{15}$ G; see, e.g., Lee et al 2000 for details) the compactness parameter $l_{\\rm A}$ is extremely large: \\begin{equation} l_{\\rm A} \\sim 10^{17}\\left(\\frac{\\Omega R}{c}\\right)L_{52}. \\end{equation} On the other hand, even for a superstrong magnetic field of $B_0 \\sim 10^{15}$ G (which is necessary to explain the total energy release) the magnetization parameter $\\sigma$ is small ($\\sigma < 1 - 10$), because within this model the magnetic field itself is secondary and its energy density cannot exceed the plasma energy density. Thus, one can conclude that for these characteristics of cosmological gamma-ray bursts the density of photons is very high so that $l_{\\rm A} \\gg l_{\\rm max}$ and the drag force can make it difficult to form a Poynting-dominated outflow. A self-consistent analysis should include into consideration other physical processes such as high optical thickness resulting in the diminishing of the photon density, radiation and particle pressure, {\\it etc.} Nevertheless, in our opinion, our conclusion may substantially restrict some recent models of cosmological gamma-ray bursts.  \\subsection{Radio Pulsars} For radio pulsars the central engine is a rotating neutron star with $M \\sim M_{\\odot}$, $R \\sim 10^{6}$ cm, total luminosity of the surface $L_{\\rm X} \\sim 10^{33}$--$10^{37}$ erg s$^{-1}$, and $B_0 \\sim 10^{12}$ G. In this case the magnetization parameter $\\sigma \\sim 10^4$--$10^6$, corresponding to relativistic electron-positron plasma, is known with rather high accuracy (see, e.g., Bogovalov 1997). This gives $l_{\\rm cr} \\sim 10^2$--$10^3$, and the compactness parameter \\begin{equation} l_{\\rm A} \\sim \\left(\\frac{\\Omega R}{c}\\right)L_{35} \\end{equation} remains small ($< 1$) even for the most energetic ($L_{\\rm X} \\sim 10^{37}$ erg s $^{-1}$ ) fast ($\\Omega R/c \\sim 10^{-2}$) pulsars like Crab and Vela. Thus, one can conclude that the drag force does not substantially disturb the magnetically-dominated outflow from radio pulsars.  Thus, the drag force does not affect the wind characteristics (particle energy, magnetic field structure, etc.) of pulsars. However, interaction of outflowing relativistic particles with thermal photons can be important in other ways. In the wind region ($r \\gg R_{\\rm L}$) even a weak interaction with photons can result in a detectable flux of inverse Compton gamma-ray photons (Bogovalov \\& Aharonian 2000). On the other hand, near the surface of the star ($r \\ll R_{\\rm L}$), inverse Compton photons are important in the pair creation process (Kardashev, Mitrofanov \\& Novikov 1984; Zhang \\& Harding 2000)."
    },
    "0209/astro-ph0209091_arXiv.txt": {
        "abstract": "Several high-frequency peaked BL Lac objects such as Mrk 501 are strong TeV emitters. However, a significant fraction of the TeV gamma rays emitted are likely to be absorbed in interactions with the diffuse IR background, yielding electron-positron pairs. Hence, the observed TeV spectrum must be steeper than the intrinsic one. Using the recently derived intrinsic $\\gamma$-ray spectrum of Mrk 501 during its 1997 high state, we study the inverse-Compton  scattering of cosmic microwave photons by the resulting electron-positron pairs, which implies the existence of a hitherto undiscovered GeV emission. The typical duration of the GeV emission is determined by the flaring activity time and the energy-dependent magnetic deflection time. We numerically calculate the scattered photon spectrum for different intergalactic magnetic field (IGMF) strengths, and find a spectral turnover and flare duration at GeV energies which are dependent on the field strength. We also estimate the scattered photon flux in the quiescent state of Mrk 501. The GeV flux levels predicted are consistent with existing EGRET upper limits, and should be detectable above the synchrotron -- self Compton (SSC) component with the {\\em Gamma-Ray Large Area Space Telescope} ({\\em GLAST}) for IGMFs $\\lesssim 10^{-16}$ G, as expected in voids. Such detections would provide constraints on the strength of weak IGMFs. ",
        "introduction": "Blazars including high-frequency peaked BL Lac objects (HBLs) are the most extreme and powerful sources among active galactic nuclei. The standard blazar model consists of a supermassive black hole ejecting twin relativistic jets, one of which is close to the line of sight. Several HBLs such as Mrk 501, Mrk 421, PKS 2155-304, 1ES 2344$+$514, H1426$+$428 and 1ES1959$+$650 are of particular interest because they emit TeV photons (see Catanese \\& Weekes 1999; Horns et al. 2002). The detection and study of such photons can provide new insights on the energetics and physical conditions in the emission regions of such blazars (Katarzy\\'nski, Sol \\& Kus 2001; Kino, Takahara \\& Kusunose 2002). Also, constraints on the spectral energy distribution of the intergalactic infrared background can be inferred from the observations on TeV photons (for a review see Hauser \\& Dwek 2001). Stecker, De Jager \\& Salamon (1992) have emphasized that the high-energy gamma photon spectra from these blazars will be modified by strongly redshift-dependent absorption effects due to interactions of such photons with the intergalactic infrared-UV background, and indicated that the  intrinsic spectrum of  an observed TeV blazar can be derived by evaluating the optical depth to TeV photons. Such calculations were made for Mrk 501 during the 1997 flaring activity, leading to an inferred intrinsic high-energy spectrum with a broad, flat peak that is much higher than the observed one in the $\\sim 5-10$ TeV range (Konopelko et al. 1999; De Jager \\& Stecker 2002, hereafter DS). The physical reason for this difference is that a significant fraction of the original high-energy gamma rays have been absorbed in $\\gamma\\gamma$ interactions with photons of the intergalactic infrared-UV background, leading to electron/positron pairs.  The purpose of this Letter is to suggest that inverse Compton (IC) scattering of the resulting electron/positron pairs against cosmic microwave background (CMB) photons may produce a new GeV emission component in TeV blazars. For gamma-ray bursts, such Compton scattering leads to an observable, delayed MeV-GeV emission component if the intergalactic magnetic fields (IGMFs) are very weak (Plaga 1995; Cheng \\& Cheng 1996; Dai \\& Lu 2002). A similar phenomenon is also expected from gamma-ray burst proton interactions with the CMB (Waxman \\& Coppi 1996). Here we discuss the well-studied blazar Mrk 501, both because the high-energy spectrum up to 20 TeV of strong flares of this blazar in 1997 has been observed by the HEGRA air Cerenkov telescope system (Aharonian et al. 1999, 2001), and because the intrinsic spectrum of Mrk 501 over two decades of energy has been derived based on the consistency between the Whipple telescope and HEGRA spectra (DS).  The strength of IGMFs has not been determined so far. Faraday rotation  measures imply an upper limit of $\\sim 10^{-9}$ G for a field with  1 Mpc correlation length (Kronberg 1994 for a review). Other methods were proposed to probe fields in the range $10^{-10}$  G to $10^{-20}$ G (Lee, Olinto \\& Sigl 1995; Plaga 1995). To interpret the observed $\\mu$G magnetic fields in galaxies and X-ray clusters, the seed fields required in dynamo theories could be as low as $10^{-20}$ G (Kulsrud et al. 1997; Kulsrud 1999). Furlanetto \\& Loeb (2001) argued that quasar outflows may pollute the intergalactic medium, but the possible weak IGMFs in voids may remain  uncontaminated. Theoretical calculations of primordial magnetic fields show that these fields could be of order $10^{-20}$ G or even as low as $10^{-29}$ G, generated during the cosmological QCD or electroweak phase transition respectively (Sigl, Olinto \\& Jedamzik 1997). In this Letter we propose that by observing a hitherto undiscovered GeV emission component from flares of TeV blazars such as Mrk 501, one may be able to obtain important information or constraints on the poorly known IGMFs. ",
        "conclusions": "Using the intrinsic $\\gamma$-ray spectrum of Mrk 501 during its 1997 flaring activity derived by DS as an example, we have predicted a new GeV emission component for this and similar HBLs, if their radiation reaches us through regions with a low magnetic field as expected in under-dense voids. This GeV emission is due to the IC scattering of CMB photons by the electron/positron pairs produced in interactions of high-energy photons with the cosmic infrared-UV background photons. We summarize our findings: First, the GeV flux of the scattered CMB photons is higher than that from the SSC process in the blob for IGMFs $\\lesssim 10^{-16}$\\,G. For an IGMF $\\gtrsim 10^{-20}$ G, there is an observable spectral turnover, whose position is strongly dependent on the field strength. This spectral turnover is due to the fact that the lower energy emission would have a longer delay. Second, the typical duration of the GeV emission is always given by the variability timescale of the TeV gamma-ray flux for photon energes larger than $E_{\\rm turn}$, and it is given by $\\Delta t_{\\rm B}\\propto E_{\\gamma,1}^{-5/2}$ for photon energies smaller than $E_{\\rm turn}$. Third, the GeV emission predicted here is consistent with existing EGRET observations, and is strong enough to be detected by {\\em GLAST}, for IGMFs $\\lesssim 10^{-16}$ G and flare times $\\sim 1$ day. Fourth, we have derived a generic formula for the optical depth due to pair production in a relativistic blob, $\\tau_{\\gamma\\gamma}^{\\rm in}\\propto \\delta^{-2\\alpha-4}$, where $\\delta$ is the Doppler factor and $\\alpha$ is the index of the softer photon energy spectrum. This optical depth is not only consistent with the previously used formula in the $\\alpha=1$ case but can also be generalized to broader cases.  We note that even if a source such as Mrk 501 is not in a void, primary TeV photons emitted from the source have such a long mean free path that electron/positron pairs may be produced in void regions. Thus, detections on the GeV emission would provide a sensitive probe for weak IGMFs, of consequence for cosmogonical and galactic dynamo theories.  Our results are also relevant for the quiescent state of HBLs such as Mrk 501. Aharonian, Coppi, \\& V\\\"olk (1994) and Coppi \\& Aharonian (1997) discussed a similar model for very high energy emission ($>100$ GeV) from blazars in the quiescent state. We here use the recent models for the intergalactic infrared-UV background radiation to discuss lower energy emission from TeV blazars. According to Catanese et al. (1997), the Whipple telescope detected a flux of $\\phi(E_\\gamma\\simeq 1\\,{\\rm TeV})= 8^{+2}_{-3}\\times 10^{-12}\\,{\\rm ergs}\\,{\\rm cm}^{-2}\\,{\\rm s}^{-1}$ in the quiescent state of this HBL. The external pair-production optical depth $\\tau_{\\gamma\\gamma}^{\\rm ex}(E_\\gamma=1\\,{\\rm TeV})\\simeq 0.5$ and $0.6$, and thus the flux of the externally scattered photons with an energy of $\\sim 0.6$ GeV is estimated as $\\phi(E_\\gamma\\simeq 0.6\\,{\\rm GeV})\\simeq \\phi(E_\\gamma\\simeq 1\\,{\\rm TeV})\\times [\\exp (\\tau_{\\gamma\\gamma}^{\\rm ex})-1]\\sim 5 \\times 10^{-12}\\,{\\rm ergs}\\,{\\rm cm}^{-2}\\,{\\rm s}^{-1}$ and $7\\times 10^{-12}\\,{\\rm ergs}\\,{\\rm cm}^{-2}\\,{\\rm s}^{-1}$ for the ``baseline\" and ``fast-evolution\" models of DS, respectively. These values of the GeV emission flux are much larger than $\\sim 1.0\\times 10^{-13}\\,{\\rm ergs}\\,{\\rm cm}^{-2}\\,{\\rm s}^{-1}$, the sensitivity of {\\em GLAST} at energy $\\sim 0.6$ GeV for steady sources in a one-year survey. Therefore, even in the quiescent state of Mrk 501, its externally scattered emission may also be detected by {\\em GLAST}. It should be pointed out that such detections are possible for a magnetic deflection angle that is less than the blazar jet opening angle. This also requires that the field strength be below $\\sim 10^{-16}$ G, similar to the flare case, for an opening angle $\\sim 0.1$. Above this value of the magnetic field, the steady GeV emission may be suppressed."
    },
    "0209/hep-ph0209053_arXiv.txt": {
        "abstract": "\\noindent\\small We calculate the cross section of the inverse beta decay process, $\\nu_e+n\\to p+e$, in a background magnetic field which is much smaller than $m_p^2/e$.  Using exact solutions of the Dirac equation in a constant magnetic field, we find the cross section for arbitrary polarization of the initial neutrons.  The cross section depends on the direction of the incident neutrino even when the initial neutron is assumed to be at rest and has no net polarization.  Possible implications of the result are discussed. ",
        "introduction": "\\label{in} The interactions of elementary particles show novel features when they occur in non-trivial backgrounds.  Study of particle propagation in matter has proved pivotal in the understanding of the solar neutrino problem.  Similar studies of particle processes in background magnetic fields are also important since stellar objects like neutron stars are expected to possess very high magnetic fields, of the order of $10^{12}$\\,G or higher.  Analysis of these processes might be crucial for obtaining a proper understanding of the properties of these stars.   In this paper, we calculate the cross section of the inverse beta-decay process in a magnetic field.  We consider the possibility that the neutrons may be totally or partially polarized in the magnetic field, and find the cross section as a function of this polarization.  The neutrinos are assumed to be strictly standard model neutrinos, without any mass and consequent properties.  The presence of the magnetic field breaks the isotropy of the background, and a careful calculation in this background reveals a dependence of the cross section on the incident neutrino direction with respect to the magnetic field.  Considerable work has been done on the magnetic field dependence of the URCA processes which have neutrinos in their final states \\cite{DRT, MOc, Gvozdev:1999md, Arras:1998mv, Baiko:1998jq}.  An angular dependence obtained in the differential cross section of these reactions imply that in a star with high magnetic field, neutrinos are created asymmetrically with respect to the magnetic field direction.  The process that we consider, on the other hand, have neutrinos in the initial state. So this process influences the neutrino opacity in a star.   Some calculations of this process exist in the literature.  Roulet \\cite{Roulet:1997sw}, as well as Lai and Qian \\cite{Lai:1998sz} performed the calculation by assuming that the magnetic field effects enter only through the phase space integrals, whereas the matrix element remains unaffected.  Gvozdev and Ognev \\cite{Gvozdev:1999md} considered the final electron to be exclusively in the lowest Landau level.  Arras and Lai \\cite{Arras:1998mv} calculated only the angular asymmetry, and only to the first order in the background magnetic field.  The earlier calculations of the present authors \\cite{Bhattacharya:1999bm,KP2} did not take neutron polarization into account.  In this paper, we consider the problem in full detail --- i.e., we calculate the matrix element using spinor solutions of the electron in a magnetic field, take all possible final Landau levels into account, include the possiblity of neutron polarization, and perform the calculations to all orders in the background field in the 4-fermi interaction theory.    The paper is organized as follows. In Sec~\\ref{so}, we provide some background for the calculation. Most of this section contains no original material, but we provide it for the sake of completeness, as well as for setting up the notation that will be used in the later sections. In Sec.~\\ref{fo}, we define the fermion field operator and show how it acts on the states in the presence of a magnetic field. Sec.~\\ref{ib} contains the calculation of the cross section for a monochromatic neutrino beam, which contains the main results of our paper. In Sec.~\\ref{es}, we discuss the realistic case where the initial neutrino beam has a finite energy spread.  Sec.~\\ref{co} contains our discussions and conclusions. ",
        "conclusions": ""
    },
    "0209/astro-ph0209085_arXiv.txt": {
        "abstract": "Virtually every ``serious'' place where professional astronomy is done has a librarian, even if shared with the physics or math department. Since its creation in 1994 of {\\it Departamento de Astronom\\'{\\i}a} (DA) of Universidad de Guanajuato (UG) it was neither provided with a librarian, nor with proper space for its holdings, nor with a budget allowing institutional journal subscriptions. I~describe my experience of now five years as ``amateur'' librarian, and present information on other small astronomy institutions in Mexico in a similar situation. ",
        "introduction": "Since 1997 the DA of UG (DAUG) is housed at 2200\\,m above sea level, overlooking the city of Guanajuato, close to the ``geometrical center'' of Mexico, and declared as ``heritage of mankind'' by UNESCO. The research staff of eight \\linebreak[4] astronomers participates in undergraduate teaching within the Physics and other programs at UG, and hopes to offer a postgraduate program in astrophysics soon. The DA maintains the {\\it Observatorio La Luz} with a 57-cm optical reflector, used for public outreach purposes and being prepared as a student laboratory. ",
        "conclusions": "The changes in the job market and the Internet have affected radically not only the way astronomers work, but also how an astronomy library is run, especially at small and ``poor'' places. Today small groups of astronomers are established independently of favorable sky conditions and rely mainly on an adequate Internet connection, but often have to work without a professional librarian. While this may work ``well'', i.e.\\ with little effect on research output as in our case, it certainly relies heavily on the services provided by a few professional librarians thinking far beyond their own institution. I see a dangerous trend for a future 2-class system of astronomy institutions: those with professional librarians working almost ``behind the scenes'' and those which have to survive without a local librarian altogether. \\\\[-1ex]"
    },
    "0209/astro-ph0209420_arXiv.txt": {
        "abstract": "Parallel computing has turned  out to be the enabling technology to solve complex physical systems. However, the transition from shared memory, vector computers to massively parallel, distributed memory systems and, recently, to hybrid systems poses new challenges to the scientist. We want to present a cook-book (with a  very strong, personal bias) based on our experience with parallization of our existing codes. Some of the general tools and communication libraries are discussed. Our approach includes a mixture of algorithm, domain and physical module based parallization. The advantages, scalability and limitations of each are discussed at some examples. We want show that it becomes easier to write parallel code with increasing complexity of the physical problem making stellar atmosphere codes beyond the classical assumptions very suitable. ",
        "introduction": "The last few years saw a shift in high performance computing from the paradigm of vector computing  to  massively parallel systems which allow  to solve increasingly complex, physical problems. We want to concentrate on general concepts to write parallel code  which may be helpful to get started and to estimate/choose the {\\sl right(!)} number of CPUs for a given numerical problem.   Due to the rapid advances in the hardware, any number for the performance or the hardware mentioned (e.g. Myrinet, Gbit-Ethernet) will  be outdated long before this contribution goes into print. To judge the performance of parallel codes, the relevant quantity is the ratio between communication and CPU-speed and, thus, the numbers may still be useful.  Ideally, code for massively parallel computers should be designed and written from scratch, and it should not be based on any legacy code. However, this approach is both costly and very time consuming. Development cycles may be  longer than changes in the computational landscape, and the flexibility may be reduced to answer problems in astronomy. Sometimes, it may be beneficial to adopt a current code base because of faster development cycles, higher flexibility  and, more important, to use  well tested codes. Here, we want to report our attempt of the latter approach to demonstrate problems and possible solutions. Most of our code is written in FORTRAN with some C routines. Our  HYDdrodynamical RAdiation transport code is based  on a number of individual programs which have been used to carry out many of the analyzes of SNIa and Core Collapse Supernovae ({\\cite{H88}, ..., \\cite{H95}, \\cite{HHWW01}, ...}).  All components have been  written or adopted to a  modular form with well defined interfaces which allows an easy coupling (see Fig.  1) and code verification  by exchanging modules. The modules consist of physical units to provide a solution for e.g. the nuclear network, the statistical equations to determine the atomic level population, equation of states, the  opacities, the hydro or the  radiation transport problem. The individual modules are coupled explicitly. Consistency between the solutions is achieved iteratively by perturbation methods  (see H\\\"oflich 2002, this volume). \\begin{figure}[ht] \\vskip -0.5cm \\hskip 0.5cm \\includegraphics[width=8.9cm,angle=270]{pahpar4.ps} \\vskip -0.02cm \\caption{ Block diagram of our  hydrodynamical radiation transport code (HYDRA) which includes detailed equation of states, nuclear and atomic networks. The  modules are high-lightened which are used to calculate NLTE-light curves of Type Ia Supernovae. Parallization is done on the level of physical modules using the spatial coordinates (dashed), photon packages (dotted) and frequencies (solid). Within the modules {\\sl Opacities} and {\\sl Statistical Equations}, sub-slaves are created to distribute the load according to  line transitions and elements, respectively. } \\label{module} \\end{figure}  The goal of this paper is to provide some help in the transition from existing scalar or vector  to  parallel codes. After some basics and the general concept, we will discuss various approaches to create parallel code, available tools, and limitations of our approach.  The terms  {\\sl scalability} and {\\sl efficiency} characterize how well the execution time decrease with the number of processors N(CPU) and the effective speed per processor compared to N(CPU) one-processor systems. The actual numbers have been obtained using a 20 node PC-cluster with dual-Pentium IIs with 400Mhz, 512MB per node and fast Ethernet interconnections. In addition, we used a  16 node  IBM-SP3 system with 4 CPUs and 1GB per node, interconnected by a  cross-bar. ",
        "conclusions": "Parallelization may allow to solve problems previously not feasible. Stellar atmospheric programs and radiation hydro problems are most suitable because they use very complex physics which, in general allow a (physical) module based approach. This enables to the use of legacy code. Nevertheless, the efforts to change from scalar and vector-computers are non-trivial and time-consuming.  Currently, three architectures are used for parallel computing: 1) shared memory and 2) distributed memory systems, and 3) distributed systems. For most problems, only the first two are a valuable option. Inhomogeneous clusters  of workstations (e.g. PCs + Suns) should be avoided. Often,  the scalability of a parallel code can be increased significantly by employing both the shared and distributed memory model. Nowadays, systems become rather common which bundle clusters of multi-processor nodes.  Parallization tools on shared  memory systems are directive based (e.g. Open MP) or or done by the compiler. Typically, communication within multi-processor nodes is sufficient fast for task distribution on a loop level. These systems are highly preferred for algorithm based problems such a linear equations, matrix inversions etc.  The domain and physical module based approach is very suitable for distributed memory systems, and load balancing can be implemented in a 'straight forward' way. In particular, parallization on a module basis shows a very good scalability and it is easy to implement. Explicit message passing may be a bonus in large program packages because the memory space is well splitted which greatly  reduces the problem of memory leaks between physical modules. Dynamical task allocation as a crucial feature in radiation-hydro problems which include statistical equations  because it provides a 'fail-save' mode (see sect. 3.3). Fortunately, it is realized in both PVM and the new standard MPI2. New projects should use the communication libraries based on the MPI2 standard as soon as full implementations become common place.   For high performance, it is critical to use the 'right' number of processors. Communication overhead may actually decrease the overall performance significantly. Optimization and dynamical load balancing is critical for good performance."
    },
    "0209/astro-ph0209493_arXiv.txt": {
        "abstract": "The CHIANTI atomic database contains atomic energy levels, wavelengths, radiative transition probabilities and electron excitation data for a large number of ions of astrophysical interest. Version~4 has been released, and proton excitation data is now included, principally for ground configuration levels that are close in energy. The fitting procedure for excitation data, both electrons and protons, has been extended to allow 9 point spline fits in addition to the previous 5 point spline fits. This allows higher quality fits to data from close-coupling calculations where resonances can lead to significant structure in the Maxwellian-averaged collision strengths. The effects of photoexcitation and stimulated emission by a blackbody radiation field in a spherical geometry on the level balance equations of the CHIANTI ions can now be studied following modifications to the CHIANTI software. With the addition of \\ion{H}{1}, \\ion{He}{1} and \\ion{N}{1}, the first neutral species have been added to CHIANTI. Many updates to existing ion data-sets are described, while several new ions have been added to the database, including \\ion{Ar}{4}, \\ion{Fe}{6} and \\ion{Ni}{21}. The two-photon continuum is now included in the spectral synthesis routines, and a new code for calculating the relativistic free-free continuum has been added. The treatment of the free-bound continuum has also been updated. ",
        "introduction": "The CHIANTI database was first released in 1996 \\citep{dere97} and it contains energy levels, radiative data and electron excitation rates for virtually all astrophysically important ions. In addition there are a number of computer routines written in IDL which allow a user to compute synthetic spectra and study plasma diagnostics. The database was originally focussed towards reproducing collisionally-excited emission line spectra at ultraviolet wavelengths from 50 to 1150\\,\\AA. Version 2 \\citep{landi99} introduced many minor ion species to the database as well as routines to compute free-free and free-bound continua. The most recent version (v.3) of the database \\citep{dere01} extended coverage of CHIANTI to X-ray wavelengths (1--50\\,\\AA) principally through the addition of hydrogen and helium-like ions, and dielectronic recombination lines.  CHIANTI has seen applications to many different areas of astrophysics since its inception. It has been extensively used in solar physics, in particular for the analysis of spectra obtained from the CDS, SUMER and UVCS spectrometers on board the SOHO satellite \\citep[e.g.,][]{young97, landi02, akmal01}. CHIANTI is also used to model the instrument responses of the EIT \\citep{dere00} and TRACE imaging instruments in order to convert measured fluxes into physical parameters such as temperature and emission measure. The wide  coverage of many different ions allowed CHIANTI to be a useful aid in the verification and definition of ultraviolet spectrometers' flux calibrations through the use of emission line ratios that are insensitive to the plasma conditions. Examples include the SERTS rocket flights \\citep{young98, brosius98}, and the Normal Incidence Spectrometer and Grazing Incidence Spectrometers on CDS \\citep{delzanna01}.  Beyond the Sun, CHIANTI has seen application to analyses of the wind emission from the Arches cluster of massive stars \\citep{raga01}, warm gas in galaxy clusters \\citep{dixon01} and analyses of a number of cool stars including AB Doradus \\citep{brandt01}, AU Microscopii \\citep{pagano00} and $\\epsilon$ Eridani \\citep{jordan01}. \\citet{delzanna02} present a review of various spectroscopic diagnostic techniques that can be applied to XUV observations of active stars. They use CHIANTI to illustrate the severe limitations that some commonly-used methods and atomic data have. \\citet{delzanna02} obtain results in terms of stellar transition region densities, emission measures and elemental abundances that are significantly different from those of other authors. Their results suggest that a large body of work on cool star atmospheres will have to be revisited and stress the importance of using assessed and up-to-date atomic data. Laboratory work also plays a vital role in the assessment of cool star results, with work by \\citet{beiers99}, \\citet{brown98} and \\citet{fournier01} providing valuable insights into plasma processes affecting EUV and X-ray spectra.   CHIANTI also forms a significant part of other atomic database packages. APED \\citep{smith01} supplements CHIANTI with data from several other sources and is focussed towards modeling X-ray spectra. XSTAR \\citep{bautista01} is a photoionization code that uses CHIANTI data for modelling the level balance within individual ions. CHIANTI also forms a significant part of the Arcetri Spectral Code \\citep{landi98, landl02}.  The present work describes the latest updates to CHIANTI, including the addition of the new physical processes of proton and photon excitation of ion levels, the addition of new ions and revisions of existing ion data-sets. ",
        "conclusions": "The previous sections have described the latest updates to the CHIANTI atomic database that will continue to make CHIANTI a vital tool for interpreting astrophysical data. The database and the associated IDL software package are freely available at three websites in the US and Europe:  \\begin{itemize} \\item \\anchor{http://wwwsolar.nrl.navy.mil/chianti.html}{http://wwwsolar.nrl.navy.mil/chianti.html} \\item \\anchor{http://www.arcetri.astro.it/science/chianti/chianti.html}{http://www.arcetri.astro.it/science/chianti/chianti.html} \\item \\anchor{http://www.damtp.cam.ac.uk/user/astro/chianti/chianti.html}{http://www.damtp.cam.ac.uk/user/astro/chianti/chianti.html}. \\end{itemize}  In addition, both the database and software package are available through the Solarsoft system (\\anchor{http://www.lmsal.com/solarsoft/}{http://www.lmsal.com/solarsoft/})."
    },
    "0209/astro-ph0209036_arXiv.txt": {
        "abstract": "We consider the idealized expansion of an initially self-gravitating, static, singular, isothermal cloud core.  For $t\\ge 0$, the gas is ionized and heated to a higher uniform temperature by the formation of a luminous, but massless, star in its center.  The approximation that the mass and gravity of the central star is negligible for the subsequent motion of the \\hii region holds for distances $r$ much greater than $\\sim 100$ AU and for the massive cloud cores that give rise to high-mass stars.  If the initial ionization and heating is approximated to occur instantaneously at $t=0$, then the subsequent flow (for $r \\gg 100$ AU) caused by the resulting imbalance between self-gravity and thermal pressure is self-similar.  Because of the steep density profile ($\\rho \\propto r^{-2}$), pressure gradients produce a shock front that travels into the cloud, accelerating the gas to supersonic velocities in what has been called the ``champagne phase.'' The expansion of the inner region at $t > 0$ is connected to the outer envelope of the now ionized cloud core through this shock whose strength depends on the temperature of the \\hii gas. In particular, we find a modified Larson-Penston (L-P) type of solution as part of the linear sequence of self-similar champagne outflows.  The modification involves the proper insertion of a shock and produces the right behavior at infinity ($v \\rightarrow 0$) for an outflow of finite duration, reconciling the long-standing conflict on the correct (inflow or outflow) interpretation for the original L-P solution.  For realistic heating due to a massive young central star which ionizes and heats the gas to $\\sim$ 10$^4$ K, we show that even the self-gravity of the ionized gas of the massive molecular cloud core can be neglected.  We then study the self-similar solutions of the expansion of \\hii regions embedded in molecular clouds characterized by more general power-law density distributions: $\\rho \\propto r^{-n}$ with $ 3/2 < n < 3$.  In these cases, the shock velocity is an increasing function of the exponent $n$, and diverges as $n \\rightarrow 3$.  We show that this happens because the model includes an origin, where the pressure driving the shock diverges because the enclosed heated mass is infinite. Our results imply that the continued photoevaporation of massive reservoirs of neutral gas (e.g., surrounding disks and/or globules) nearby to the embedded ionizing source is required in order to maintain over a significant timescale the emission measure observed in champagne flows. ",
        "introduction": "\\label{intro}  For a spherically symmetric molecular cloud core, initially at rest, the size $r_S$ of the region that can be ionized, is given by the standard formula (Str\\\"omgren 1939): \\be \\int_{r_0}^{r_S} n_e n_p \\alpha_2 4 \\pi r^2 dr = \\dot N_\\ast. \\label{rs} \\ee Eq.~(\\ref{rs}) assumes ionization equilibrium and the ``on the spot'' approximation.  In the above, $n_e$ is the electron density; $n_p$ is the ion density; $\\alpha_2$ is the recombination coefficient to the second energy level of hydrogen; $\\dot N_\\ast$ is the rate of ionizing photons from the star, assumed to be a constant; and  $r_0$ is the radius below which all of the gas in the original cloud core may be considered to have fallen into the center (perhaps via a disk) to make a star of mass $M_\\ast$.\\footnote{In the case when $\\dot N_\\ast \\propto t^3$, Newman \\& Axford (1968) found self-similar solutions for the expansion of an ionization bounded \\hii region in a uniform H{\\sc I} cloud.}.  If the virial velocity (thermal, turbulent, or magnetohydrodynamic) supporting the original (neutral) cloud core before star formation is denoted by $a_1$, order of magnitude arguments yields $r_0 \\sim r_1$, the Bondi-Parker radius of this neutral gas, \\be r_1 \\equiv {GM_\\ast \\over 2a_1^2}. \\label{Bondione} \\ee The square of the sound speed in the \\hii gas $a_2^2$ is generally appreciably larger than $a_1^2$; thus, the equivalent Bondi-Parker radius of the ionized gas, \\be r_2 \\equiv {GM_\\ast \\over 2 a_2^2}, \\label{Bonditwo} \\ee will be considerably smaller than $r_1$.  For typical numbers, $M_\\ast \\simeq 25 \\; M_\\odot$, $a_1 \\simeq 1$ km s$^{-1}$, $a_2 \\simeq 10$ km s$^{-1}$, we have $r_1 \\simeq 10^4$ AU $\\gg r_2 \\simeq 10^2$ AU, with both $r_1$ and $r_2$ much bigger than the physical radius of the star.  Much interior to $r_2$, the ionized gas will empty into the star (or more likely, into a disk if it has even a slight amount of angular momentum); whereas much exterior to $r_2$, the gravitationally unbound \\hii gas will expand outward, if it has not already reached pressure equilibrium with the surrounding cloud.  Since $r_0 \\gg r_2$, we may henceforth ignore the gravitational field of the star on the flow of the \\hii region beyond $r_0$, although for purposes of making contact with earlier theoretical work, we shall begin by not ignoring the self-gravity of this gas. Since the material inside $r_0$ of the initial density profile should have fallen into the star, the observed presence of appreciable amounts of ionized gas at intermediate radii, $\\sim 10^3$ AU in typical ultracompact \\hii regions, is awkward to explain.  We defer until \\S 6 the discussion of the special kinds of models that are probably required to explain ultracompact \\hii regions.  Assume now that the molecular cloud core initially had a power-law distribution of gas density that extends essentially to infinity: \\be \\rho(r) = K r^{-n}. \\label{powerlaw} \\ee If $n < 3/2$, the ultraviolet radiation is trapped within a finite radius $r_S$, and the \\hii region is said to be ``ionization bounded'' (see Osterbrock 1989).  If $n > 3/2$, the \\hii region can be either ionization bounded or ``density bounded.''  In the latter case, a finite output of ultraviolet radiation can ionize an infinite volume of gas beyond $r_0$.  The dividing line between being ionization bounded and density bounded arises when the density constant $K$ equals a critical value $K_{\\rm cr}$: \\be K_{\\rm cr} =  2 \\mu_i m_H \\left[ { {(2n-3) r_0^{2n-3} \\, \\dot N_\\ast} \\over{ 4 \\pi \\alpha_2 } }\\right ]^{1/2}, \\label{Kcrit} \\ee where $\\mu_i$ is the mean weight per particle of the ionized gas, $m_H$ is the hydrogen mass, and $n_p=n_e=\\rho/2 \\mu_i m_H $.  We wish to compare the value of $K_{\\rm cr}$ with the value $K_\\ast$ implied by the assumption that the power law (\\ref{powerlaw}) initially extended inward from $r_0$ as well as outward, but that the part inward of $r_0$ has fallen into the center (perhaps via a disk) to make a star of mass $M_\\ast$: \\be K_\\ast = {(3-n)M_\\ast\\over 4\\pi r_0^{3-n}}. \\label{Kstar} \\ee Taking the ratio of eq. (\\ref{Kstar}) to eq. (\\ref{Kcrit}), we get \\be {K_\\ast\\over K_{\\rm cr}} = \\left[{(3-n)M_\\ast\\over 2\\mu_i m_H}\\right] \\left[{\\alpha_2\\over (2n-3)4\\pi r_0^3\\dot N_\\ast}\\right]^{1/2}. \\label{Kratio} \\ee For $M_\\ast \\simeq 25~M_\\odot$, $r_0 \\simeq 10^4$~AU, $\\dot N_\\ast \\simeq 10^{49}$ s$^{-1}$, $\\alpha_2 \\simeq 2.6 \\times 10^{-13}$~cm$^3$ s$^{-1}$, $K_\\ast/K_{\\rm cr} \\simeq 23\\, (3-n)/(2n-3)^{1/2}$.  It is remarkable that factors of such disparate orders of magnitude as the dimensionless quantities in the two square brackets of eq. (\\ref{Kratio}) combine to give a ratio within two orders of unity. Nevertheless, since $K_\\ast$ represents a rough estimate of $K$ and $K_\\ast > K_{\\rm cr}$, this calculation formally indicates that the \\hii regions of 25 $M_\\odot$ (and lower mass) stars are likely to be ionization bounded, at least initially before any expansion occurs. However, if we assume that $\\dot N_\\ast$ scales roughly as $M_*^3$ (as indicated by the results of Vacca et al.~1996), the expression on the right-hand side scales as $M_\\ast^{-2}$, indicating that the \\hii regions of the most massive O stars may be density bounded from the start, especially if such stars are born in regions with density gradients close to $n=3$.  They will then develop champagne flows as follows.  When $K \\sim K_\\ast < K_{\\rm cr}$, the ionization front (IF) created by the idealized instantaneous appearance of a star at $t=0$ rapidly moves to infinity and establishes an isothermal structure with $T \\simeq 10^4$ K.  After the passage of the IF, the cloud remains out of mechanical balance and the pressure gradients will produce an expansion of the whole cloud. Due to the density gradient the inner regions expand faster than the outer regions and a shock travels through the cloud, accelerating the gas to supersonic velocities. This is known as the ``champagne phase'' (e.g.  Bodenheimer, Tenorio-Tagle \\& Yorke 1979).  Franco, Tenorio-Tagle \\& Bodenheimer (1990; hereafter FTB) studied the evolution of \\hii regions embedded in molecular clouds with steep density gradients.  High spatial resolution infrared and radio recombination line observations toward several sources have found ionized gas accelerating away from the central source in the manner expected of champagne flow models (e.g. Garay et al. 1994; Keto et al. 1995; Lumsden \\& Hoare 1996; Lebr\\'on et al. 2001).  Note that in several of the observed compact \\hii regions (e.g., 29.96-0.02, G32.80+0.19B, G61.48+0.09B1) the inferred rate of ionizing photons imply excitation by central stars with masses $M_* > 30 M_\\odot$.  Density profiles in massive molecular cores have also been extensively studied observationally (e.g.  Garay \\& Rodr\\'\\i guez 1990; Caselli \\& Myers~1995; Van der Tak et al. 2000; Hatchell et al. 2000; for a review see Garay \\& Lizano 1999). Even though the environment is possibly clumpy on scales of tenths of pc, density profiles are well approximated by power laws with $1 \\simlt n \\simlt 2$.  Theoretical models of the formation of massive stars within dense and massive cores have assumed power law exponents in this range (Osorio et al. 1999; McKee \\& Tan 2002).  Recently, Franco et al. (2000) have argued that radio continuum spectra of ultracompact \\hii regions indicate initial density gradients with $2 \\simlt n \\simlt 3$.  Clearly, more observations with high spatial resolution are necessary to reliably establish the density profiles of the sites of massive star formation.  The purpose of this paper is to study by similarity techniques the ``champagne phase'' of expansion of \\hii regions with power law density distributions.  In \\S 2, we formulate the outflow problem in the case of the singular isothermal sphere (SIS) that has $\\rho \\propto r^{-2}$, including the effect of self-gravity. Tsai \\& Hsu (1995) found the outflow analogue of the inside-out collapse solution (Shu 1977), but in which the SIS is sent into expansion by an outward propagating shock. In \\S 3 we show that the Tsai \\& Hsu (1995) solution is actually the limit of a family of outflow solutions when $(a_1/a_2)^2 \\rightarrow 1$ from below.  Furthermore, the outflow solution with the particular ratio of $(a_1/a_2)^2 =0.75$ corresponds to a piece of the time-reversed Larson-Penston (L-P) collapse solution (Larson 1969; Penston 1969), but with a shock inserted to obtain the correct asymptotic behavior for large distances (or early times). For realistic heating after the passage of an IF, i.e., for realistic values of $(a_1/a_2)^2 \\ll 1$, we show that the self-gravity of the \\hii gas can be neglected.  In \\S 4 we extend our study to the evolution of champagne flows in the case of density distributions with power law exponents in the range $3/2 < n < 3$, neglecting self-gravity.  In \\S 5 we find that the self-similar models have a shock propagating at constant velocity into the ionized gas, in good detailed agreement with the models of FTB. In particular, the shock velocity diverges as $n \\rightarrow 3$.   We show that this happens because the formal treatment extends the inner radius of the calculation to the origin. In such a treatment, the mass of driving \\hii gas diverges when $n \\ge 3$.  We perform more realistic calculations in such cases that cut holes in the gas distribution for $r < r_0$.  In \\S 6 we summarize our conclusions, and we discuss the implications of our results for the problem of ultracompact \\hii regions. Finally, in the appendices we show that, for the scales relevant to molecular cloud cores, the isothermal assumptions for the gas and the shock are valid. ",
        "conclusions": "We have obtained self-similar champagne flow solutions for the expansion of power law  gas density distributions after the gas has been uniformly heated out of mechanical equilibrium by the birth of a star at the center of a molecular cloud core. These solutions attach via a shock to upstream breeze solutions.  In the case of the isothermal sphere with $\\rho \\propto r^{-2}$, the ``inside-out expansion''  found by Tsai \\& Hsu (1995) is the limit of the family of self-similar outflow solutions when the sound speed $a_2$ of the \\hii region is the same as the sound speed $a_1$ of the original molecular cloud core.  The case $(a_1/a_2)^2=1$ must include the effect of the self-gravity of the gas, and the outflow solution then attaches to the unperturbed static SIS upstream of the shock.  Another member of this family, for $(a_1/a_2)^2=0.75$, is a time reversed piece of the L-P collapse solution, with a shock allowing the upstream solution to have a correct outflow (breeze) asymptotic behavior.  For the high values of the gas temperature expected after the passage of an ionization front, $(a_1/a_2)^2 \\ll 1$, and the self-gravity of the gas can be neglected.  The solution then approaches a shape invariant form.  In the approximation that the self-gravity of the ionized gas can be ignored, we computed the self-similar champagne flows of \\hii regions formed in molecular clouds characterized by power law density distributions with exponents $3/2 < n < 3$.  These self-similar solutions behave as in the numerical models of FTB: in the ``champagne phase'' a shock moves with a constant speed into the ionized medium with a shock speed that increases with increasing density gradient. The speed of the shock diverges as $n \\rightarrow 3$ because the mass of the driving gaseous piston diverges, if the origin is included. Instead, if the origin is excluded, the shock front velocity reaches an asymptotic constant value for $n<3$. For $n \\ge 3$ the shock accelerates to infinite velocity as $r_s \\rightarrow \\infty$, but only because finite outward momentum is inputted into spherical shells of ever decreasing mass.  These results may help explain astrophysical champagne flows where expansion velocities are seen that are considerably larger than the sound speed $a_2\\simeq 10$ km s$^{-1}$ associated with conventional \\hii regions.  (Driving by fast stellar winds may contribute to the perceived motions.) Despite the large expansion velocities produced in the case of the steepest pressure gradients ($n \\rightarrow 3$), we show in the appendices that the isothermal assumption for the gas and for the shock are valid for the scales of interest in molecular clouds.  The supersonic expansion of the ionized gas creates a severe lifetime problem for champagne flows.  A natural solution is the photoevaporation of circumstellar disks and/or remnant neutral globules which would help maintain the high observed emission measures in these sources."
    },
    "0209/hep-ph0209244_arXiv.txt": {
        "abstract": "} \\nc{\\eab}{ We review the cosmological implications of the flat directions of the Minimally Supersymmetric Standard Model (MSSM). We describe how field condensates are created along the flat directions because of inflationary fluctuations. The post-inflationary dynamical evolution of the field condensate can charge up the condensate with $B$ or $L$ in a process known as Affleck-Dine baryogenesis. Condensate fluctuations can give rise to both adiabatic and isocurvature density perturbations and could be observable in future cosmic microwave experiments. In many cases the condensate is however not the state of lowest energy but fragments, with many interesting cosmological consequences. Fragmentation is triggered by inflation-induced perturbations and the condensate lumps will eventually form non-topological solitons, known as $Q$-balls. Their properties depend on how supersymmetry breaking is transmitted to the MSSM; if by gravity, then the $Q$-balls are semi-stable but long-lived and can be the source of all the baryons and LSP dark matter; if by gauge interactions, the $Q$-balls can be absolutely stable and form dark matter that can be searched for directly. We also discuss some cosmological applications of generic flat directions and $Q$-balls in the context of self-interacting dark matter, inflatonic solitons and extra dimensions. \\noindent ",
        "introduction": "The interplay between particle physics and cosmology plays an increasing role in understanding the physics beyond the Standard Model (SM)~\\cite{sm} and the early Universe before the era of Big Bang Nucleosynthesis (BBN) \\cite{sarkar96,olive00333}. On both fronts we currently lack hard data. Above the electroweak scale $E\\sim {\\cal O}(100)$~GeV, the particle content is largely unknown, while beyond the BBN scale $T\\sim {\\cal O}(1)$~MeV, there is no direct information about the thermal history of the Universe. However, there are some observational hints, as well as a number of theoretical considerations, which seem to be pointing towards a wealth of new physics both at small distances and in the very early Universe. Perhaps most importantly, new data is expected soon from accelerator experiments such as LHC and from cosmological measurements carried out by satellites such as MAP~\\cite{MAP} and Planck \\cite{PLANCK}.    In cosmology the recent observations of the cosmic microwave background (CMB) radiation, which has a temperature $\\sim 2.728\\pm 0.004$~K \\cite{peebles93}, have given rise to an era of precision cosmology. The Cosmic Background Explorer (COBE) satellite~\\cite{COBE} first detected in a full-sky map a temperature perturbation of one part in $10^{5}$ at scales larger than $7$~degrees~\\cite{smoot91}. The irregularities are present at a scale larger than the size of the horizon at the time when the microwave photons were generated and cannot be explained within the traditional hot Big Bang model~\\cite{liddle-lyth00}. The recent balloon experiments BOOMERANG~\\cite{BOOMERANG} and MAXIMA~\\cite{MAXIMA}, together with the ground-based DASI \\cite{pryke01} experiment have established the existence of the first few acoustic peaks in the positions predicted by cosmic inflation \\cite{guth81,linde82108,linde90,liddle-lyth00}. Inflation, a period of exponential expansion in the very early Universe, is a direct link to physics at energy scales that will not be accessible to Earth-bound experiments for any foreseeable future. Inflation could occur because a slowly rolling scalar field, the inflaton, dynamically gives rise to an epoch dominated by a false vacuum. During inflation quantum fluctuation are imprinted on space-time as energy perturbations which then are stretched outside the causal horizon. These primordial fluctuations eventually re-enter our horizon, whence their form can be extracted from the CMB (for a review, see \\cite{mukhanov92,liddle-lyth00}).    Inflation can be considered as a model for the origin of matter since all matter arises from the vacuum energy stored in the inflaton field. However the present models do not give clear predictions as to what sort of matter there is to be found in the Universe. From observations we know that baryons constitute about 3\\%\\ of the total mass ~\\cite{olive00333}, whereas relic diffuse cosmic ray background virtually excludes any domains of anti-baryons in the visible Universe \\cite{cohenetal97}. Almost 30\\%\\ of the total energy density is in non-luminous, non-baryonic dark matter~\\cite{jungman96}. Its origin and nature is unknown, although various simulations of large scale structure formation suggest that there must be at least some {\\it cold dark matter} (CDM), comprising of particles with negligible velocity, although there may also be a component of {\\it hot dark matter} (HDM), comprising of particles with relativistic velocities~\\cite{liddle93}. The rest of the energy density is in the form of dark energy~\\cite{perlmutter97,reiss98}.   The striking asymmetry in the baryonic matter has existed at least since the time of BBN and plays an important role in providing the right abundances for the light elements. The present Helium ($^3He$), Deuterium ($D$) and Lithium abundances suggest a baryon density and an asymmetry relative to photon density of order $10^{-10}$~\\cite{olive00333}. Such an asymmetry is larger by a factor of $10^{9}$ than what it should have been by merely assuming a initially baryon symmetric hot Big Bang \\cite{kolbturner90}. Therefore baryon asymmetry must have been created dynamically in the early Universe.   The origin of baryon asymmetry and dark matter bring cosmology and particle physics together. Within SM all the three Sakharov conditions for baryogenesis~\\cite{sakharov67} are in principle met; there is baryon number violation, $C$ and $CP$ violation, and an out-of-equilibrium environment during a first-order electroweak phase transition. However, it has turned out that within SM the electroweak phase transition is not strong enough \\cite{cohen93,yaffe95,rubshap96,kajantie96}, and therefore the existence of baryons requires new physics. Regarding HDM, light neutrinos are a possible candidate \\cite{liddle93,dolgov02}, but there is no candidate for CDM in the SM. HDM alone cannot lead a successful structure formation because of HDM free streaming length~\\cite{bonometto84,liddle93}. Therefore one must resort to physics beyond the SM  also to find a candidate for CDM \\cite{jungman96}.   The tangible evidence for small but non-vanishing neutrino masses as indicated by the neutrino oscillations observed by the Super-Kamiokande~\\cite{superk} and SNO collaborations~\\cite{sno} is definitely another indication for new physics beyond the SM. The main sources of neutrino mass could be either Dirac or Majorana. A Dirac  neutrino would require a large fine tuning in the Yukawa sector (one part in $10^{11}$) while a Majorana mass would appear to require a scale much above the electroweak scale together with an extension of the SM gauge group $SU(3)_{C}\\times SU(2)_{L}\\times U(1)_{Y}$. In the Majorana case the lightness of the neutrino could be explained via the see-saw mechanism \\cite{see-saw,mohapatra80}.   A theoretical conundrum is that the mass scale of SM is $\\sim{\\cal O}(100)$~GeV, much lower that the scale of gravity $M_{\\rm P}=(8\\pi G_{N})^{-1/2}=2.436\\times 10^{18}$~GeV, and not protected from quantum corrections. The most popular remedy is of course supersymmetry~(for a review, see \\cite{nilles84,haber85,bailin94}), despite the fact that so far supersymmetry has evaded all observations \\cite{lepsusy}. The minimal supersymmetric extension of the SM is called the MSSM. Supersymmetry must be broken at a scale $\\sim {\\cal O}(1)$~TeV, presumably in some hidden sector from which breaking is transmitted to the MSSM, e.g., by gravitational \\cite{nilles84,haber85} or gauge interactions~\\cite{giudice98}.   In the MSSM the number of degrees of freedom are increased by virtue of the supersymmetric counterparts of the SM bosons and fermions. One of them, known as the lightest supersymmetric particle (LSP), could be absolutely stable with a mass of the order of supersymmetry breaking scale. LSP would be a natural candidate for CDM (see e.g. \\cite{jungman96}). In addition, because of the larger parameter space, electroweak baryogenesis in MSSM in principle has a much better chance to succeed. However, there are a number of important constraints, and lately Higgs searches at LEP have narrowed down the parameter space to the point where it has all but disappeared~\\cite{cline00,quiros01,carena02}.   Electroweak baryogenesis within MSSM thus appears to be heading towards deep trouble. Moreover, although MSSM can provide CDM, there is no connection between dark matter and electroweak baryogenesis. On the other hand, by virtue of supersymmetry, MSSM has the intriguing feature that there are directions in the field space which have virtually no potential. They are usually known as {\\it flat directions}, which are made up of squarks and sleptons and therefore carry baryon number and/or lepton number. The MSSM flat directions have been all classified \\cite{gherghetta96}.   Because it does not cost anything in energy, during inflation squarks and sleptons are free to fluctuate along the flat directions and form scalar condensates. Because inflation smoothes out all gradients, only the homogeneous condensate mode survives. However, like any massless scalar field, the condensate is subject to inflaton-induced zero point fluctuations which impart a small, and in inflation models a calculable, spectrum of perturbations on the condensate. After inflation the dynamical evolution of the condensate can charge the condensate up with a large baryon or lepton number, which can then released into the Universe when the condensate decays, as was first discussed by Affleck and Dine \\cite{affleckdine85}.   The potential along the MSSM flat direction is not completely flat because of supersymmetry breaking. In addition to the usual soft supersymmetry breaking, the non-zero energy density of the early Universe also breaks supersymmetry, in particular during inflation when the Hubble expansion dominates over any low energy supersymmetry breaking scale \\cite{dine95,dine96}. Flatness can also be spoiled by higher-order non-renormalizable terms, and the details of the condensate dynamics depend on these.   In most cases, the MSSM condensate along a flat direction is however not the state of lowest energy. The condensate typically has a negative pressure, which causes the inflation-induced perturbations to grow. Because of this the condensate fragments, usually when the Hubble scale equals the supersymmetry breaking scale, into lumps of condensate matter which eventually settle down to non-topological solitons dubbed as $Q$-balls by Coleman \\cite{coleman85}. $Q$-balls carry a global charge, which in the case of MSSM is either $B$ or $L$.  The properties of $Q$-balls depend on supersymmetry breaking. If transmitted to MSSM by gravity, the $Q$-balls turn out to be only semistable but nevertheless long-lived compared with the time scales of the very early Universe \\cite{enqvist98}. When they decay, they may provide not only the baryonic matter but also dark matter LSPs \\cite{enqvist99}. If supersymmetry breaking is transmitted from the hidden sector to MSSM by gauge interactions, the resulting $Q$-balls would be stable and could exist at present as a form of dark matter \\cite{kusenko97405}. In this case one can make direct searches for their existence \\cite{arafune00}. In both cases there is a prediction for the relation between the baryon and dark matter densities. Moreover, the condensate perturbations are inherited by the $Q$-balls, and can thus be a source of both isocurvature and adiabatic density perturbations \\cite{enqvist9983,enqvist0062,kawasaki01}.   This review is organized as follows. In Section $2$, we recapitulate some basic cosmology, and in particular baryogenesis. We briefly discuss various popular schemes of baryogenesis and describe the original Affleck-Dine baryogenesis. In Section $3$, we present some background material for inflation, mainly concentrating on supersymmetric  models. Quantum fluctuations and reheating are also discussed. In Section $4$, we present the MSSM flat directions and discuss their properties. Various contributions to the flat direction potential in the early Universe are listed. Low energy supersymmetry breaking schemes, such as gravity and gauge mediation, are also discussed. In Section $5$, we discuss the dynamical properties of flat directions and the running of the flat direction potential due to gauge and Yukawa interactions. Leptogenesis along $LH_{u}$ flat direction, and the condensate evaporation in a thermal bath, is also described. We discuss fragmentation of the condensates for both gravity and gauge mediated supersymmetry breaking and present the relevant numerical studies. In Section $6$, $Q$-ball properties are presented in detail. We describe various types of $Q$-balls, their interactions and their behavior at finite temperature. We discuss surface evaporation, diffusion, and dissociation of charge from $Q$-balls in a thermal bath. In Section $7$, we focus on the cosmological consequences of $Q$-balls. We consider $Q$-ball baryogenesis and non-thermal dark matter generation through charge evaporation for different types of $Q$-balls. We discuss $Q$-balls as self-interacting dark matter and present experimental and astrophysical constraints on stable $Q$-balls. In Section $8$, we briefly survey beyond-the-MSSM-condensates by considering inflatonic $Q$-balls and Affleck-Dine mechanism without MSSM flat directions. We also describe solitosynthesis, a process of accumulating large $Q$-balls in a charge asymmetric Universe.     \\newpage ",
        "conclusions": ""
    },
    "0209/astro-ph0209547_arXiv.txt": {
        "abstract": "The massive star formation properties of 55 Virgo Cluster and 29 isolated S0-Scd bright ($M_B \\leq $ -18) spiral galaxies are compared via analyses of R and H$\\alpha$ surface photometry and integrated fluxes as functions of Hubble type and central R light concentration (bulge-to-disk ratio).  In the median, the total normalized massive star formation rates (NMSFRs) in Virgo Cluster spirals are reduced by factors up to 2.5 compared to isolated spiral galaxies of the same type or concentration, with a range from enhanced (up to 2.5 times) to strongly reduced (up to 10 times).  Within the inner 30\\% of the optical disk, Virgo Cluster and isolated spirals have similar ranges in NMSFRs, with similar to enhanced (up to 4 times) median NMSFRs for Virgo galaxies.  NMSFRs in the outer 70\\% of the optical disk are reduced in the median by factors up to 9 for Virgo Cluster spirals, with more severely reduced star formation at progressively larger disk radii.  Thus the reduction in total star formation of Virgo Cluster spirals is caused primarily by spatial truncation of the star-forming disks.  The correlation between HI deficiency and R light central concentration is much weaker than the correlation between HI deficiency and Hubble type. The previously observed systematic difference in HI spatial distributions and kinematics between early- and late-type spirals in the Virgo Cluster is at least partially due to the misleading classification of stripped spirals as early-types. ICM-ISM stripping of the gas from spiral galaxies is likely responsible for the truncated star-forming disks of Virgo Cluster spirals.  This effect may be responsible for a significant part of the morphology-density relationship, in that a large fraction of Virgo Cluster galaxies classified as Sa are HI-deficient galaxies with truncated star forming disks rather than galaxies with large bulge-to-disk ratios. ",
        "introduction": "What role does the environment play in the evolution of cluster galaxies? The observation that the morphological mix of galaxies varies in different nearby environments was qualitatively noted even in the early studies of the Virgo Cluster by Hubble and Humason (1931) and has been confirmed in many studies (e.g., Oemler 1974; Dressler 1980; Postman \\& Geller 1984; Dressler et al. 1997). In addition, many studies of nearby galaxies have detailed how cluster galaxies differ from field galaxies within the same Hubble type, including redder colors (Kennicutt 1983a; Oemler 1992), less HI gas (Chamaraux, Balkowski, \\& G\\'erard 1980; Giovanelli \\& Haynes 1983), and truncated outer HI gas disks (Giovanelli \\& Haynes 1983; Warmels 1988; Cayatte et al. 1990). Studies of higher redshift cluster galaxies show that evolution in cluster galaxy morphology and star formation properties has occurred over the last several billion years. Butcher \\& Oemler (1978) showed that distant clusters of galaxies have a higher proportion of blue galaxies. More recently, Dressler et al. (1997) found an excess of spirals and a lack of S0 galaxies in about the same proportion in dense clusters at redshifts of 0.5 compared to local dense clusters. These results suggest that many of the S0's in local clusters were actively star-forming spirals at z=0.5.  There is a rich literature about the types of environmental processes which could affect the evolution of galaxies in clusters. There are processes which affect mainly the gaseous content of a galaxy, such as ICM-ISM interactions (reviewed by van Gorkom 2004), starvation (Larson, Tinsley, and Caldwell 1980), and gas accretion (Kenney et al., in prep). Gravitational processes, which affect both the gaseous and stellar properties of a galaxy, range from low-velocity tidal interactions and mergers, to high velocity interactions between galaxies and/or the cluster (reviewed by Struck 1999 and Mihos 2004). Despite a number of recent studies of nearby and distant clusters, it is not yet clear which processes, if any, are dominant. Indeed the properties of cluster galaxies may be determined by a variety of environmental interactions over a Hubble time (Miller 1988; Oemler 1992; Moore et al. 1998).  The rate of ongoing star formation is an important measure of the evolutionary state of a galaxy, and a sensitive indicator of some types of environmental interactions. Previous studies of the star formation rates of cluster galaxies have reached varying and sometimes opposite conclusions. Some authors have found reduced star formation rates in clusters (Kennicutt 1983a; Bicay \\& Giovanelli 1987; Kodaira et al. 1990; Moss \\& Whittle 1993; Abraham et al. 1996; Balogh et al. 1998; Koopmann \\& Kenney 1998; Hashimoto et al. 1998; Gavazzi et al. 2002), others similar rates (Kennicutt, Bothun, \\& Schommer 1984; Donas et al. 1990; Gavazzi, Boselli, \\& Kennicutt et al. 1991, Gavazzi et al. 1998), and others enhanced rates (Moss \\& Whittle 1993, 2000; Bennett \\& Moss 1998). This confused situation on cluster galaxy star formation rates is one of the motivations for the present work. In addition, many previous studies of star formation rates have been based on aperture or integrated galaxy spectral observations. Spatial studies of star formation can probe the types of environmental interactions at work in nearby clusters, therefore revealing what may have influenced galaxies in richer clusters earlier in the history of the Universe.  This work describes results from an imaging survey of Virgo Cluster and isolated spiral galaxies in both broadband R and the H$\\alpha$ emission line. The intent of this study is to compare the amounts and distributions of massive star formation in Virgo Cluster galaxies to those of a relatively undisturbed isolated sample of galaxies. We base our comparisons on the H$\\alpha$ emission from galaxies, which is a good tracer of the massive star formation rate (Kennicutt 1983b) in relatively dust-free regions. It can be used to estimate the total star formation rate by making standard assumptions for the initial mass function. We use the H$\\alpha$ surface brightness and the H$\\alpha$ flux normalized by the R flux as distance-independent tracers of the massive star formation rate. The data we have gathered thus allow a quantitative radial comparison of massive star formation rates in the two environments. The Virgo Cluster is a particularly good laboratory to study environmental effects on star formation since it is the nearest moderately rich cluster, it has significant ICM, it is dynamically young with a current infall of spirals (Tully \\& Shaya 1984), and its galaxies are subject to a variety of environmental effects.  An important consideration in addressing these issues is the objective comparison of galaxies with different morphologies. van den Bergh (1976) made the important observation, based on visual inspection of images on plates, that many cluster spiral galaxies have low rates of star formation relative to field spirals, and that the traditional Hubble classification does not work well in nearby clusters because the bulge-to-disk (B/D) ratio is not well correlated with the disk star formation rate. Koopmann \\& Kenney (1998) confirmed and quantified this effect, using objective measurements from CCD images, showing that a one-dimensional classification scheme such as the Hubble classification is not applied in the same way to field and cluster galaxies, and that it is not adequate to describe the wider range in morphologies of cluster galaxies. For example, Koopmann \\& Kenney find that a significant fraction ($\\sim$ 50\\%) of Virgo Cluster spirals classified as Sa are small-to-intermediate concentration (B/D) galaxies with reduced global star formation rates, presumably due to environmental effects. This effect contributes to the excess of `early-type' spiral galaxies in the Virgo Cluster and therefore to the local morphology-density relationship.  This paper presents the main results of a program which is published in separate papers. We present the observational data for the Virgo galaxies in Koopmann et al. (2001, hereafter PI) and that for the isolated galaxies in Koopmann \\& Kenney (2004, in prep, hereafter PII). These papers include H$\\alpha$ and R images and radial profiles for all galaxies. Koopmann \\& Kenney (1998) present comparisons of Hubble type and central R concentration for isolated and Virgo galaxies. The present paper concentrates on the massive star formation properties, with comparisons between integrated H$\\alpha$ fluxes, H$\\alpha$ radial profiles, and the relative concentrations of H$\\alpha$ emission. A comparison of the different types of H$\\alpha$ morphologies and a discussion of environmental effects is given in Koopmann \\& Kenney (2004, hereafter PIV). ",
        "conclusions": "The results of this survey of star formation rates and morphologies of Virgo Cluster spirals compared to isolated spirals include the following:  1. H$\\alpha$-based estimates of the total massive star formation rates of Virgo Cluster galaxies span a range from strongly reduced (up to 10 times) to enhanced (up to 2.5 times) compared to the isolated sample. In the median, Virgo total star formation rates are reduced by factors up to 2.5 for different Hubble types and concentrations.  2. For most Virgo Cluster galaxies with reduced total star formation, it is truncation rather than anemia (low H$\\alpha$ surface brightness across the disk) which causes the reduced total star formation rates. Median inner rates are similar or enhanced up to a factor of 1.7, while outer star formation rates are reduced in the median by factors of 1.5 - 9. In the outermost parts of the optical disks, star formation of all types and concentrations of galaxies are reduced by factors greater than 3 times in the median. Reductions in individual galaxies range to factors greater than 100. No galaxies have star formation rates in the outer half of the disk enhanced above isolated rates. Thus, most galaxies with reduced total star formation have inner star formation rates which are similar to or enhanced with respect to isolated galaxies of similar central light concentration or Hubble type. Virgo Cluster galaxies have more concentrated star forming disks than isolated counterparts, largely due to truncation.  3. The larger mean HI deficiency of Virgo cluster Sa's as compared to cluster Sb and Sc's (Solanes et al 2001) is  predominantly due to strongly stripped galaxies being classified as Sa's, rather than galaxies with large bulge-to-disk ratios being preferentially stripped. One reason for the previously observed systematic difference in spatial distributions and kinematics between early- and late-type Virgo spirals (Dressler 1986) is that stripped spirals, which are preferentially on radial orbits (Solanes et al 2001), tend to be misleadingly \\it classified \\rm as early-types. Likewise, the excess of Sa galaxies in the Virgo cluster, and perhaps by extension other clusters, is in large part due to strongly stripped galaxies being classified as Sa's, and is not simply an excess of spiral galaxies with large bulge-to-disk ratios.  The funding for the research on the Virgo cluster and isolated spiral galaxies was provided by NSF grants AST-9322779 and AST-0071251. Martha Haynes, Christopher Springob, Karen Masters, and the Cornell Extragalactic Group are gratefully acknowledged for their aid in derivation of updated HI deficiencies. We thank Judy Young, Vera Rubin, Yasuhiro Hashimoto, and Shardha Jogee for valuable discussions, and our referee, Alessandro Boselli, for helpful comments which improved this paper. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration."
    },
    "0209/astro-ph0209292_arXiv.txt": {
        "abstract": "We report the discovery of  a new ultraluminous X-ray source (ULX) in the nearby galaxy NGC~4244  from {\\em Chandra} archival data. The source, \\1wga, is one of the least luminous and softest ULXs discovered so far. Its X-ray spectrum is the best available for a representative of the soft ULXs, a class of sources recently discovered by {\\em Chandra} and {\\em XMM-Newton}. \\1wga  appears  point-like in the {\\em Chandra} image and has a complex  spectral shape: a multicolor disk model, suitable for brighter ULXs, is inadequate for this source. \\1wga spectrum is heavily absorbed ($N_H \\sim 1-4 \\times 10^{21}$ cm$^{-2}$), and very soft. The best-fit power-law model gives $\\Gamma \\sim 5$ and implies a luminosity $L_{(0.5-10\\, {\\rm keV)}} \\sim 10^{39}$ erg s$^{-1}$. A comparison with previous detections shows that, despite the  variability displayed by the source  during the  {\\em Chandra} observation, \\1wga \\/ count rate, spectral shape and absorption are practically unchanged over a 9-year period.  We performed also deep optical imaging of the field containing the X-ray source, and found a possible ${\\rm R}\\sim 23.7$ counterpart. ",
        "introduction": "X-ray observations of nearby galaxies have shown that their X-ray emission comes from different classes of sources: X-ray binaries, supernova remnants, hot interstellar medium in addition to possible background active galactic nuclei (AGNs; see e.g. Fabbiano 1989 for a review). Some of the point-like sources appear to radiate well in excess of the Eddington limit for a $1\\, M_\\odot$ object, with inferred luminosities in the range $10^{39}-10^{40}$ \\es. These objects  are often referred to as ultra luminous X-ray sources (ULXs) or intermediate-luminosity X-ray objects (IXOs), inasmuch their luminosity is in between those of ``normal'' X-ray binaries and AGNs. Up to now about 90 ULXs have been detected in more than 50 galaxies (see Colbert \\& Ptak 2002 for a recent catalog). ULXs seem to be preferentially located in the outskirts of the host galaxy, although some are found in the central regions (e.g. IXO 95 in NGC 6949). As pointed out by Colbert \\& Ptak, starburst galaxies contain a comparatively large number of ULXs, but spirals seem not to be  favoured hosts with respect to ellipticals. In a very recent paper, moreover, evidence was found for a ULX associated with a globular cluster in NGC 4565 (Wu et al. 2002). ULXs are often variable sources and their X-ray spectrum may be quite soft (e.g. Makishima et al. 2000). No certain optical counterpart has been found for these sources yet, implying an extreme X-ray--to-optical flux ratio. Very recently Pakull \\& Mironi (2002) reported the presence of emission nebulae at the position of some ULXs and suggested that they could actually be related to the episode which led to the ULX formation.  Thus far no generally accepted model has been presented to explain the huge energy output of ULXs. Assuming that the Eddington limit is not exceeded and emission is isotropic implies that ULXs are powered by accretion onto a $\\approx 50-100\\, M_\\odot$ black hole. Present evolutionary scenarios do not preclude such a possibility and the fact that these are rare sources may be compatible with the mean number of ULX per galaxy being only 1-2 (Colbert \\& Ptak 2002). Alternatively ULXs may be interpreted as conventional black hole binaries ($M_{BH}\\lesssim 10\\, M_\\odot$) with modest beaming coming from a collimated X-ray emission ($b\\sim 0.1$, King et al. 2001) or to a relativistically beamed emission. In this picture the Galactic analogues of ULXs should be the microquasars, such as  GRS 1915+105 and GRO 1655-40.  This paper presents a multi-wavelength study of a new ULX, \\1wga, found in NGC~4244. This source is one of the 16 peculiar {\\em ROSAT} PSPC sources selected for their extremely high X-ray-to-optical flux ratio (Cagnoni et al. 2002). \\1wga \\/ is a  bright ($F_{(0.1-2.4\\, {\\rm keV})} > 10^{-13}$ \\cgs ) WGACAT (White, Giommi \\& Angelini 1994) source  with blank fields, i.e. no optical counterparts on the Palomar Observatory Sky Survey to O=21.5. The extreme \\fxv \\/ ratio that follows is incompatible with all major and common classes of extragalactic sources, including normal quasars, AGNs, normal galaxies and nearby clusters of galaxies (Maccacaro et al. 1988).  Possibilities for the nature of these `blanks' (Cagnoni et al. 2002)  include: (a) Quasar-2s, i.e.  high luminosity, high redshift heavily obscured quasars, the bright analogs of Seyfert~2s; (b) Low Mass Seyfert-2s, AGNs powered by a low-mass obscured black hole (i.e. obscured Narrow Line Sy~1); (c) AGNs with no big blue bump, e.g. low radiative efficiency flows; (d) Isolated Neutron Stars (e.g. Treves et al. 2000); (e) $\\gamma$-ray burst X-ray afterglows or fast variable/ transient sources; (f) failed clusters, in which a large overdensity of matter has collapsed but has not formed galaxies  (Tucker, Tananbaum \\& Remillard 1995); (g) high redshift clusters of galaxies and, most relevant to this paper, (h) ULXs in nearby galaxies.  Using X-ray archival data ({\\em ROSAT} and {\\em Chandra}) and the information obtained from optical and IR follow-ups, we present strong evidence that \\1wga \\/ is indeed a ULX in NGC~4244, a B=-18.4 (Olling 1996) edge-on spiral galaxy (Hubble type Scd) at a distance of $\\sim 3.6$~Mpc (Fry et al. 1999).  We present the  X-ray, optical and IR observations in \\S ~\\ref{sec_observations} and discuss the results in \\S ~\\ref{sec_discussion}.\\\\ Errors in the paper  represent 90\\% confidence levels, unless explicitly stated otherwise. ",
        "conclusions": "\\label{sec_discussion}  The properties of \\1wga \\/ indicate that it is an ULX in  NGC~4244. The source absorbed flux in the 0.5--2.0 keV band derived from {\\em Chandra} data ($\\sim 1.8\\times 10^{-13}$ \\ecs) is consistent within the errors with that measured with {\\em ROSAT} ($\\sim 1.2\\times 10^{-13}$ \\ecs), so the source appears not to have significantly varied between the two observations ($\\sim 9$ yrs). The unabsorbed luminosity of \\1wga \\/ at NGC~4244 distance is $\\sim 2\\times 10^{39}$ \\es for the absorbed power law model Note however, that the 50~ks {\\em Chandra} observation provides evidence for variability on a much shorter timescale ($\\sim 5000$ s). We also found that the spectral shape did not change from the {\\em ROSAT} observation and, although no simple emission model gives a satisfactory fit, the X-ray continuum appears to be very soft and can be roughly described by a very steep absorbed power-law with $\\Gamma\\sim 5$ but not with thermal emission. A complex spectral shape was also found for the stacked spectra obtained from the ULXs in the Antennea galaxies by Zezas et al. (2002a) (see their Figure~2): the spectrum presented here for \\1wga \\/ suggests that the complexity might not be the effect of the superposition of different types of ULX spectra, but it is intrinsic to each source.  Using the X-ray flux at 1 keV from the absorbed power-law fit, the R-band magnitude and the  upper limit  at 1.4 GHz we computed \\1wga \\/ broad band spectral indices according to the formula \\begin{equation} \\alpha_{x_1 x_2} = - \\frac{\\log(f(x_2)/f(x_1))}{\\log(\\lambda(x_1)/ \\lambda(x_2))} \\end{equation} We used the canonical 1~keV, V-band (5500 \\AA ) and 5~GHz points, by extrapolating the R-band flux density to the V-band and the 1.4 GHz flux density to 5 GHz assuming a flat spectral shape, and we obtain $\\alpha_{RO} \\geq 0.60 $,  $\\alpha_{OX} = 0.23$ and $\\alpha_{RX} \\geq 0.47$. Such values place \\1wga \\/ out of the region occupied by AGNs in the $\\alpha_{RO}$--$\\alpha_{OX}$ plane (e.g. Caccianiga et al. 1999) since not even the most extreme BL Lac objects can reach such values (e.g. Costamante \\& Ghisellini 2002). Taking absorption into account would make the situation even more extreme (e.g. an absorbing column of $\\sim 4 \\times 10^{21}$ cm$^{-2}$ with Galactic dust to gas ratio would enhance the X-ray flux by a factor of $\\sim 10$ and the optical flux of a factor $\\sim 6$). We can thus exclude the  possibility of \\1wga \\/ being a background AGN.  We can also exclude the possibility of \\1wga \\/ being a supernova remnant for the short timescale variability displayed in the {\\em Chandra} observation and by the lack of a significant fading in the 9 years between the {\\em ROSAT} and the {\\em Chandra} observations.  Even if the best fit model for \\1wga, a convex absorbed broken power-law, is usually used to  describe the synchrotron peak observed in beamed sources such as blazars, the sharp change in the power-law shape of \\1wga \\/  ($\\Gamma _1 =0.22$, $\\Gamma _2 = 3.85$) is difficult to reconcile with the smoothly curving blazar synchrotron peak (e.g. $\\Gamma _1 = 2.1$, $\\Gamma _2 = 2.8 $ measured for Mrk~421 by Guainazzi et al. 1999).   \\1wga \\/ luminosity falls at the lower end of the ULX range. Some emission models implies $L_X < 10^{39}$ \\es (see Table 1). In particular, the absorbed power-law plus Raymond-Smith model, and the absorbed broken power-law models, which have the lowest $\\chi^2$, give a luminosity of only few $10^{38}$ \\es. This would make \\1wga \\/ an ordinary X-ray binary in NGC~4244. However, optically thin bremsstrahlung emission is typical of extended sources, like clusters of galaxies, and we found no evidence for a diffuse nature of this source in {\\em Chandra} data and the broken power-law model does not have a straightforward physical interpretation. The relatively low luminosity of \\1wga  may still suggest that this source is an X-ray binary with a $M\\approx 10\\, M_\\odot$ black hole, of the Cyg X-1 type. The X-ray spectrum however strongly argues against this possibility, being much softer of those of Galactic black hole candidates (BHCs).  BHCs spectra in the high state are thermal and soft, but peak around a few keV and extend up to $\\sim 10$ keV.  The extreme softness of the X-ray spectrum makes this source rather peculiar among the ULXs. In fact, while very soft spectra have been already detected from other ULXs with {\\em ROSAT}, successive observations over a wider energy range showed a  harder component. An example of this behaviour is MS 0317.7$-$6647 which has been first associated with an isolated neutron star by Stocke et al. (1995) on the basis of its soft, thermal spectrum. The source was later identified with a ULX by Makishima et al. (2000) when {\\em ASCA} data convincingly showed a hard tail ($\\Gamma \\sim 2$), possibly associated with a multicolor disk blackbody. Before the advent of {\\em Chandra} the ULXs seen by {\\em ROSAT} and {\\em ASCA} were consistent with emission from a multicolor disk blackbody (e.g. Makishima et al. 2000); {\\em Chandra} is now discovering a new class of steep ULX (e.g. out of the 30 ULX of Zezas et al. 2002b, 9 have a steep spectrum). \\1wga \\/ is the source for which the best {\\em Chandra} spectrum is available so far. In the study of the Antennae galaxies Zezas et al. (2002a) find steeper spectra for the least luminous ULX: \\1wga seem to fit in this picture, interpreted with the possible presence of undetected diffuse hot interstellar medium in the proximity of the source. Alternatively one could consider that  the indication of a luminosity dependent spectral shape is due to different accretion properties. The steep ULXs  could be the ``galactic'' analog  of Narrow Line Seyfert galaxies (a steep version of the Seyfert galaxies) which are thought to be powered by a lower mass black hole accreting at a higher rate compared to ``normal'' Seyferts.   Follow-up optical observations allowed us to discover a possible counterpart of \\1wga \\/ at $0.6^{\\prime \\prime}$ from {\\em Chandra} position. The association is based only on the positional coincidence. The {\\em Chandra} error box is quite small (95\\% confidence radius $< 1''$, Kim et al. in preparation) and the background field is not very crowded (see Fig.~\\ref{fig3}), so the possibility of a chance alignment appears unlikely, albeit it can not be ruled out on the basis of present data. The optical counterpart has  $R\\sim 23.7$; if this source  is a star in NGC~4244, it has to be a red star or its bolometric correction would imply a luminosity not compatible with even the most luminous stars known. The most likely possibility is that the suggested counterpart is a late type giant/supergiant, like a M5 II-III; for this spectral range the source luminosity, corrected for extinction, would be $L\\approx 7\\times 10^{37}$ \\es, about $10^4\\, L_\\odot$. The R-K$ < 5.7$ derived from the IR non-detection is not stringent and it is satisfied  by the red stars.  ULXs are one of the most interesting class of sources which are now being investigated by the X-ray satellites {\\em Chandra} and {\\em XMM-Newton}. We presented in this paper the best spectrum ($\\sim 2000$ photons) available so far for a steep-spectrum ULX. Steep ULXs, like \\1wga, , appear to be related to the least luminous objects of the ULX class (e.g. Zezas et al. 2002a) and were essentially unknown before the launch of  {\\em Chandra} and {\\em XMM-Newton}. Future investigations are needed to understand the emission mechanisms powering these sources and to confirm and explain a possible luminosity dependence  of the spectral shape."
    },
    "0209/astro-ph0209401_arXiv.txt": {
        "abstract": "{We present new observations of copper and zinc abundances in 90 metal-poor stars, belonging to the metallicity range --3$<$[Fe/H]$<$--0.5. The present study is based on high resolution spectroscopic measurements collected at the Haute Provence Observatoire (R= 42000, S/N$>$100). The trend of Cu and Zn abundances as a function of the metallicity [Fe/H] is discussed and compared to that of other heavy elements beyond iron. We also estimate spatial velocities and galactic orbital parameters for our target stars in order to disentangle the population of disk stars from that of halo stars using kinematic criteria. In the absence of a firm a priori knowledge of the nucleosynthesis mechanisms controlling Cu and Zn production, and of the relative stellar sites, we derive constraints on these last from the trend of the observed ratios [Cu/Fe]and [Zn/Fe] throughout the history of the Galaxy, as well as from a few well established properties of basic nucleosynthesis processes in stars. We thus confirm that the production of Cu and Zn requires a number of different sources (neutron captures in massive stars, $s$- processing in low and intermediate mass stars, explosive nucleosynthesis in various supernova types). We also attempt a ranking of the relative roles played by different production mechanisms, and verify these hints through a simple estimate of the galactic enrichment in Cu and Zn. In agreement with suggestions presented earlier, we find evidence that Type Ia Supernovae must play a relevant role, especially for the production of Cu. ",
        "introduction": "The introduction of efficient high-resolution spectrographs on modern telescopes has greatly enhanced our ability to reconstruct the chemical enrichment of the Galaxy, which derives from a gradual melding of the outcomes by various nucleosynthesis processes in stars. This is obtained by the return of new elements into the interstellar medium (ISM) through slow and fast mass loss phenomena, whose relative importance is controlled by several parameters, like the initial stellar mass distribution, the physics of stellar winds, star formation rates, stellar lifetimes,  etc. This picture is made even more complex by the different dynamical behavior of the various galactic subsystems in a very clumpy galactic structure, which both sets the time scales of mixing processes in the ISM, and controls the hierarchy of cluster and individual star formation. A galactic evolution model describing all these processes in detail, including the dynamical interactions between subsystems, is still lacking, because of the enormous complexities inherent in such a construction and of the high number of free parameters inevitably involved in it. Unfortunately, for many heavy elements we infer that the mechanisms through which their galactic enrichment was achieved do involve all these ill-modeled complexities, so that for them galactic astrophysics has reached a stage in which observations lead theory by a large distance.  An example of this is provided by Cu and Zn, two elements immediately following the iron peak, for which we present in this paper new measurements for a very large sample (90) of metal-poor stars. In our previous works (Mishenina \\& Kovtyuk~2001, hereafter Paper I; Mishenina et al.~2001, hereafter Paper II) we investigated the behavior of several heavy elements at different metallicities. The present study on copper and zinc continues those efforts.  >From the observational point of view, Cu and Zn abundances were early addressed by  Gratton \\& Sneden~(1988) and by Sneden \\& Crocker~(1988). Their data favored a primary-like type of production for Zn (i.e. mechanisms yielding constant enrichment relative to iron) and a secondary-like process for Cu (i.e. one requiring iron seeds from previous stellar generations, giving rise to an enrichment proportional to the iron content). Later, further abundances for the two elements in halo and disk stars were provided by Sneden, Gratton, \\& Crocker~1991, and subsequently by Primas et al.~2000 and Blake et al.~2001. Recently, also some data for $\\omega$ Centauri have become available (see e.g. Smith et al.~2000; Cunha et al.~2002; Pancino et al.~2002), together with abundances in single very low-metallicity galactic stars (Westin et al.~2000; Cowan et al.~2002; Hill et al.~2002). Measurements of Zn in damped Ly-$\\alpha$ systems (see e.g. Pettini et al.~1999; Molaro et al.~2000) complete the list, offering an opportunity to address the production of heavy elements in the Universe at an epoch immediately following Galaxy formation.  Despite those recent integrations, more than a decade has passed since the work by Sneden et al.~(1991), who provided the only large sample of Cu and Zn abundances in metal-poor stars of our Galaxy. That work was used repeatedly to investigate the nucleosynthesis of these elements. With the present study we aim at providing a significant update of available database, and at establishing more reliable constraints on the still open problems involved in Cu and Zn formation.  The first schematic description of the chemical evolution of Cu and Zn was proposed by Sneden et al.~(1991), who suggested that they might be ascribed mainly to the weak $s$-process. Their conclusions were subsequently questioned by Raiteri et al.~(1992) and by Matteucci et al.~(1993). In this last work evidence was presented in favor of a large contribution from relatively long-lived processes, tentatively identified as Type Ia supernovae. Contrary to this, Timmes et al.~(1995), using the copper and zinc yields of Type II supernova explosion from Woosley \\& Weaver~(1995), suggested that these elements might be synthesized in significant amounts by the major nuclear burning stages in massive stars. These contrasting explanations are an example of the large uncertainties one meets when an incomplete picture of stellar yields and a simplified chemical evolution scheme have to be used for interpreting the data. Simplified, and sometimes purely analytical, chemical evolution models were very useful in the past, before the advent of high resolution spectroscopy (Lynden-Bell~1975; Tinsley \\& Larson~1978; Tinsley~1980; Clayton~1984). However, they are no longer sufficient today, after more sophisticated and precise measurements have become available.  Waiting for a revision in the nucleosynthesis models, what one can do is to provide a homogeneous set of data, derived with the same methods for many stars, and to compare them with known results for other neutron-rich elements. This is actually the scope of the present work, based on abundances obtained in a homogeneous way for metal-poor stars belonging either to the halo or to the thick disk of the Galaxy. Using the data as a guideline, we shall examine which scenarios for the stellar synthesis of these elements are compatible with the observed trends. As a consistency check, we shall then verify our conclusions through a simple computation of the ensuing chemical enrichment of the Galaxy, making use of the model adopted by Travaglio et al.~(1999). ",
        "conclusions": "In this paper we presented a large sample study of Cu and Zn in metal-poor stars, after verifying the population of the sample objects through estimates of their kinematic parameters. In the absence of clear indications on the stellar origin of Cu and Zn from the present status of nucleosynthesis theories, we tried to derive suggestions on the relevant mechanisms from an inspection of the observed data and from the evolutionary trends expected for the outcomes of various nuclear processes. In this way we inferred that several nucleosynthesis phenomena are involved, suggesting that about 25\\% of Cu is produced by secondary phenomena in massive stars, and only 7-8\\%  is due to primary phenomena in the same environment (either explosive or from a primary $n$-process). The bulk of Cu abundance (at least 62 -- 65\\%) should be instead contributed on long time scales by type Ia supernovae, in agreement with suggestions by Matteucci et al. (1993). For Zn, its trend with respect to iron implies similar percentage yields for the two elements: 1/3 from primary processes in massive stars and 2/3 from type Ia supernovae. These rough indications were shown to roughly account for the Cu and Zn enrichment in the Galaxy. This approach was however found to be too schematic for interpreting the details of the database, including the scatter at very low metallicity. We argued that this last might be due to poorly mixed different contributions from massive stars in a non-homogeneous early stage (perhaps also complicated to the non-uniqueness of the $r$-process). We also concluded that, while a simple separation of the yields for Cu and Zn in primary and secondary mechanisms, and in long-lived and short-lived parents, is roughly possible, and gives a preliminary guideline for future nucleosynthesis models, this cannot be pushed too far, and we found no reliable tool for disentangling the contributions of the two $r$-process mechanisms from the rest. A more quantitative analysis must therefore wait for a clarification of the underlying physical processes in evolved stars."
    },
    "0209/astro-ph0209198_arXiv.txt": {
        "abstract": "I discuss the evolution of the magnetic field of an accreting white dwarf. I show that the ohmic decay time is $7$--$12$ billion years for the lowest order decay mode, almost independent of core temperature or mass. I then show that the magnetic field structure is substantially altered by accretion if the white dwarf mass increases at a rate $>\\dot M_c\\approx (1$--$5)\\times 10^{-10}\\ M_\\odot\\ {\\rm yr^{-1}}$. I discuss the implications of this result for observed systems, including the possible evolutionary link between AM Hers and intermediate polars. ",
        "introduction": "Calculations of ohmic decay in an isolated, cooling white dwarf show that the magnetic field changes little over its lifetime (Chanmugam \\& Gabriel 1972; Fontaine, Thomas, \\& van Horn 1973; Wendell, van Horn, \\& Sargent 1987). However, an accreting white dwarf may substantially increase its mass on a timescale much shorter than the ohmic decay time, raising the question of what happens to its magnetic field. Here, I describe calculations which show that the field structure may be substantially affected by accretion if the accretion is rapid enough, $\\dot M>\\dot M_c\\approx (1$--$5)\\times 10^{-10}\\ M_\\odot\\ {\\rm yr^{-1}}$. At lower accretion rates, ohmic diffusion allows the field to ``keep up'' with the accretion flow. This critical accretion rate lies in the middle of observed rates, with interesting implications for observations. A detailed account of this work may be found in Cumming (2002).  A major uncertainty in our understanding of accreting white dwarfs is whether the white dwarf mass increases or decreases with time. The answer depends on the amount of mass ejected by classical novae, an uncertain quantity both observationally and theoretically. In the calculations presented here, I assume the white dwarf mass increases with time. ",
        "conclusions": ""
    },
    "0209/astro-ph0209151_arXiv.txt": {
        "abstract": "{ Recent estimates of the properties of the Crab nebula are used to derive constraints on the moment of inertia, mass and radius of the pulsar. To this purpose, we employ an approximate formula combining these three parameters. Our ``empirical formula'' $I\\simeq a(x) M R^2$, where $x=(M/M_\\odot) ({\\rm km}/R)$, is based on numerical results obtained for thirty theoretical equations of state of dense matter. The functions $a(x)$ for neutron stars and strange stars are qualitatively different. For neutron stars $a_{\\rm NS}(x)=x/(0.1+2x)$ for $x\\le 0.1$ (valid for $M>0.2~{\\rm M}_\\odot$) and $a_{\\rm NS}(x)={2\\over 9}(1+5x)$ for $x>0.1$. For strange stars  $a_{\\rm SS}(x)={2\\over 5}(1+x)$ (not valid for strange stars with crust and $M<0.1~M_\\odot$). We obtain also an approximate expression for the maximum moment of inertia $I_{\\rm max,45}\\simeq (-0.37 + 7.12\\cdot x_{\\rm max}) (M_{\\rm max}/M_\\odot)(R_{M_{\\rm max}}/{10~{\\rm km}})^2$, where $I_{\\rm 45} = I/10^{45}~{\\rm g\\cdot cm^2}$, valid for both neutron stars and strange stars. Applying our formulae to the evaluated values of $I_{\\rm Crab}$, we derive constraints on the mass and radius of the pulsar. { A very conservative evaluation of the expanding nebula mass, $M_{\\rm neb}=2~M_{\\odot}$, yields $M_{\\rm Crab}>1.2~M_{\\odot}$ and $R_{\\rm Crab}=10-14~{\\rm km}$. Setting the most recent evaluation (``central value'') $M_{\\rm neb}=4.6~M_{\\odot}$ rules out most of the existing equations of state, leaving only the stiffest ones: $M_{\\rm Crab}>1.9~M_{\\odot}$, $R_{\\rm Crab}=14-15~{\\rm km}$. } ",
        "introduction": "The moment of inertia of neutron stars plays a crucial role in the models of radio pulsars. In the standard case of rigid rotation the total energy expenditure per unit time $\\dot{E}_{\\rm tot}$ is related to the measured pulsar angular frequency $\\Omega$ and its time derivative $\\dot{\\Omega}$ by $\\dot{E}_{\\rm tot}=-I\\Omega{\\dot{\\Omega}}$. However, even in the simplest  case of slow rotation ($\\Omega^2\\ll (c/R)^2GM/Rc^2$, $R\\Omega\\ll c$) characteristic of  observed radio pulsars, the relation of $I$ to the matter distribution within the star is complicated by the general relativistic effects such as the dragging of the local inertial frames. Among all global neutron star parameters the moment of inertia is the most sensitive to the dense matter equation of state (EOS). The ``theoretical maximum mass'' of neutron stars increases by a factor of two when going from the softest to the stiffest EOS. The ``theoretical maximum moment of inertia'' however increases then by a factor of seven (Haensel 1990). In view of this particular sensitivity of $I$ to the largely unknown equation of state (EOS) of dense matter at supra-nuclear density, it is of interest to look for observational evaluations $I$, which could then be used to constrain theoretical models.  In the present paper we use recent evaluations of the parameters of the expanding Crab nebula, in order to derive constraints on the Crab-pulsar neutron star. In Sect. 2 we reconsider the energetics of the Crab nebula, and using expanding-shell model of the  nebula we derive a constraint on the moment of inertia of the Crab pulsar. In Sect. 3 we derive approximate formulae for $I$ of neutron stars and strange stars, based on a statistical analysis of a numerical data sample obtained for thirty theoretical  EOSs of dense matter. These formulae are then used in Sect. 4 to derive constraints in the $M-R$ plane, implied by the value of $I_{\\rm Crab}$, deduced in Sect.\\ 2. Appendix presents results of an analysis of the correlation between the maximum value of $I$ and the mass and radius  of the configuration with maximum allowable mass. ",
        "conclusions": ""
    },
    "0209/astro-ph0209221_arXiv.txt": {
        "abstract": "We present the SHEEP survey for serendipitously-detected hard X-ray sources in ASCA GIS images. In a survey area of $\\sim 40$~deg$^{2}$, 69 sources were detected in the 5-10 keV band to a limiting flux of $\\sim 10^{-13}$~erg cm$^{-2}$ s$^{-1}$. The number counts agree with those obtained by the similar BeppoSAX HELLAS survey, and both are in close agreement with ASCA and BeppoSAX 2-10 keV surveys. Spectral analysis of the SHEEP sample reveals that the 2-10 and 5-10 keV surveys do not sample the same populations, however, as we find considerably harder spectra, with an average $\\Gamma\\sim1.0$ assuming no absorption. The implication is that the agreement in the number counts is coincidental, with the 5-10 keV surveys gaining approximately as many hard sources as they lose soft ones, when compared to the 2-10 keV surveys. This is hard to reconcile with standard AGN ``population synthesis'' models for the X-ray background, which posit the existence of a large population of absorbed sources. We find no evidence of the population hardening at faint fluxes, with the exception that the few very brightest objects are anomalously soft.  53 of the SHEEP sources have been covered by ROSAT in the pointed phase. Of these 32 were detected. An additional 3 were detected in the RASS. As expected the sources detected with ROSAT are systematically softer than those detected with \\asca\\ alone, and of the sample as a whole. Although they represent a biased subsample, the ROSAT positions allow relatively secure catalog identifications to be made. We find associations with a wide variety of AGN and a few clusters and groups. At least two X-ray sources identified with high-z QSOs present very hard X-ray spectra indicative of absorption, despite the presence of broad optical lines. A possible explanation for this is that we are seeing relatively dust-free ``warm absorbers'' in high luminosity/redshift objects.  Color analysis indeed indicates that many of the spectra are not consistent with a simple, absorbed power law. The spectra are likely to be complex, with an absorbed hard power law and scattered or ``leaky'' component in the soft X-rays. Many are also consistent with a reflection dominated spectrum. Our analysis defines a new, hard X-ray selected sample of objects - mostly active galactic nuclei - which is less prone to bias due to obscuration than previous optical or soft X-ray samples. They are therefore more representative of the population of AGN in the universe in general, and the SHEEP survey should produce bright examples of the sources that make up the hard X-ray background, the majority of which has recently been resolved by Chandra. This should help elucidate the nature of the new populations. ",
        "introduction": "\\label{Sec:Introduction}  ROSAT observations have shown that the diffuse X--ray background (XRB) in the soft X-ray band (0.5-2 keV) is made up of discrete sources, primarily standard, broad-line QSOs (Shanks et al. 1991; Hasinger et al. 1998; Schmidt et al. 1998).  It is puzzling, however, that the X-ray spectra of AGN above $\\sim 2$ keV, which typically show an intrinsic power law index of $\\sim 1.9$ (Nandra \\& Pounds 1994), is so much steeper than the observed background in the same band, with $\\Gamma\\sim 1.4$ (Marshall et al. 1980). This ``spectral paradox'' has received much attention, and a consensus appears to be emerging as to its resolution.  Setti \\& Woltjer (1989) suggested that the spectral paradox can be solved by hypothesizing that large numbers of AGN are heavily absorbed in the X-ray band, consistent with Seyfert unification schemes (Lawrence \\& Elvis 1982; Antonucci \\& Miller 1985). Comastri et al. (1995), and other authors (e.g. Madau, Ghisselini \\& Fabian 1994; Gilli et al. 1999, 2001) have shown that AGN spectra with a range of absorbing column densities can indeed be made to fit the XRB spectrum consistently with the number counts. Such models of the XRB are the most promising to date, and agree with many observables, including the range of spectra seen in nearby, bright AGN. They predict that the major contributors to the XRB are a large population of highly-absorbed AGN at moderate-high redshift. This population of objects has never been directly observed. This is perhaps not surprising, given that traditional UV-excess and soft X-ray surveys - which have detected most of the AGN we know of so far - are biased against strongly-absorbed objects.  The best method of uncovering obscured AGN is in the hard X-ray band, and a large population of such sources is implied. The ROSAT 0.5-2 keV number counts can be converted into 2-10 keV counts by extrapolating the ROSAT flux assuming the mean spectrum of the ROSAT sources of $\\Gamma=2$. This exercise under-predicts the number counts observed by \\asca\\ by a factor $\\sim 2$ (Georgantopoulos et al. 1997; Cagnoni et al. 1998; Ueda et al. 1999a). This immediately implies the presence of a large population with flat or absorbed spectra.  Optical spectroscopic identifications of ASCA sources have identified a few examples of obscured AGN at high redshift. For example, Boyle et al. (1998) have reported the discovery of an X-ray obscured quasar at z=0.67 and Georgantopoulos et al. (1999) have described the properties of an even more extreme z=2.35 quasar (see Akiyama, Ueda \\& Ohta 2002 for another example). Chandra has now also uncovered a few examples of type II QSOs (Norman et al. 2001; Stern et al. 2002).  The inferred column densities for these sources are $>10^{23}$~cm$^{-2}$ which makes them extremely difficult to detect or identify in soft X-ray surveys and, if the absorbing material is dusty, heavily redenned and weak in the optical.  There may be a vast population of such hidden beasts lurking in the hard X-ray sky.  A promising start towards uncovering this population has been made using the BeppoSAX HELLAS survey. Fiore et al. (1999) presented the first results of a survey in the 5-10 keV with the BeppoSAX MECS instrument. Their survey covered an area of 50 deg$^{2}$ and detected $\\sim$150 sources above a limiting flux of about $5\\times 10^{-14}$~erg cm$^{-2}$ s$^{-1}$. The updated HELLAS survey of Fiore et al. (2001) covers a larger area of 85 deg$^{-2}$, with 147 sources. The number count distribution, logN-logS, presents a Euclidean slope with $\\gamma=1.56\\pm0.14$, in apparent agreement with the population synthesis models (Comastri et al. 2001). At the survey's limiting flux $\\sim30$ per cent of the 5-10 keV XRB has been resolved.  Catalog cross-correlations and the results of an optical ID program showed a high proportion of AGN, many of which are heavily obscured and some of which are at moderately high redshifts (Fiore et al. 1999; Fiore et al. 2001).  This survey contains several examples of obscured AGN both because of its large area and of the very hard X-ray band employed. Preliminary results from {\\it XMM} on the Lockman hole field (Hasinger et al. 2001) has extended the logN-logS a factor of twenty deeper (see also Baldi et al. 2001). It appears to present a Euclidean slope all the way down to the limiting {\\it XMM} flux of $2.4\\times10^{-15}$. At this flux limit 60 per cent of the 5-10 keV XRB has been resolved.  New data from Chandra have now provided a breakthrough in our understanding of the XRB, but have also raised further questions. Mushotzky et al.  (2000) presented observations of a deep Chandra-ACIS field.  They detected a few tens of sources most of which have been spectroscopically followed up with the Keck telescope (Barger et al. 2001).  Surprisingly there were no numerous examples of the long sought obscured ``type II QSO'' population (note we adopt the traditional definition of type I and type II objects based on their optical, rather than X-ray properties). Instead two other distinct populations appear.  One is of bright early-type galaxies.  These galaxies are ``passive'', in that they present no clear sign of AGN activity in their optical spectra. The other population of X-ray sources is of hard sources with very faint or non-existent optical counterparts (even in deep Keck images, implying $B>28$ in some cases). This immediately highlights a problem with the Chandra population - that they are too faint for effective optical or X-ray followup.  Deeper observations in the Chandra Deep Field South (CDFS), and in the Hubble Deep Field North (HDFN) respectively confirm the above findings (Tozzi et al. 2001; Brandt et al.  2001a) .  For example, Brandt et al. (2001a) detect 12 sources in the area of the HDF (at a flux limit of $\\sim2\\times10^{-16}$ $\\rm erg~cm^{-2}~s^{-1}$ in the 2-8 keV band) of which 4 are passive early-type galaxies while only 3 are broad-line AGN. The early, tentative suggestion is that the bulk of the X--ray background may arise from relatively low luminosity, low redshift AGN of uncertain character.  Further progress is possible by undertaking a large area hard X-ray survey, which can in principle reveal bright examples of the faint Chandra sources, to help understand their nature. HELLAS has already made some progress in this regard. Here we present a similar survey with the \\asca\\ GIS instruments, called SHEEP (Search for the High Energy Extragalactic Population). ",
        "conclusions": "We have performed an X-ray survey with the ASCA GIS in the 5-10 keV band in a $\\sim 40$~deg$^{2}$ area to a flux level of $\\sim 10^{-13}$~erg cm$^{-2}$ s$^{-1}$. 69 sources were detected, with a logN-logS distribution consistent with HELLAS, and 2-10 keV BeppoSAX and \\asca\\ surveys. We resolve $\\sim 15$\\% per cent of the hard X-ray background.  Of the 69 sources, 35 have ROSAT counterparts and 19 of these have optical counterparts in catalogs.  The classifications show that 11 of the ROSAT-detected sources are associated with type-1 AGN (i.e. Seyfert-1 and QSOs). We have shown, however, that the sources with ROSAT counterparts are preferentially softer than the remainder of the survey, and are therefore not an unbiased sample. A relatively large fraction (40\\%) of our sources were observed by ROSAT in the pointed phase, but not detected, and must therefore have extremely hard spectra.  The mean spectrum of the entire sample, as determined by the mean hardness ratio, depends on the method adopted and can be affected by statistical bias. However, we find that it is at least as hard as the X-ray background, and what should be the least biased estimate gives an equivalent $\\Gamma=1.1\\pm 0.1$.  Unlike previous 2-10 keV surveys (e.g. Ueda et al. 1999b; Della Ceca et al. 1999; Giommi et al. 2000; Giacconi et al. 2001), we find no systematic hardening of the spectra to faint fluxes, although direct spectral fitting shows that the brightest few objects in our sample are anomalously soft.  The spectra of the bulk of the individual sources are best described by a composite model, in which the power law is heavily absorbed, but some fraction of it either leaks through the absorber, or is scattered back into the line of sight. Some of the sources present very hard X-ray colors in the 2-10 keV band, much harder than the spectrum of the XRB. These may be very heavily absorbed, Compton thick sources such as NGC 6552 (Reynolds et al. 1994) and the Circinus galaxy (Matt et al. 1996).  This type of object could be quite common, and be a major contributer to the peak of the XRB spectrum at $\\sim 30$~keV.  \\subsection{Comparison with HELLAS and 2-10 keV surveys}  A crucial question for our survey is that of whether it selects different objects than surveys in the 2-10 keV band. As the 2-10 keV observations are generally more sensitive, there would be no point in performing harder surveys such as ours if this were indeed the case. This issue was not addressed in the analysis of the HELLAS data.  A related question is whether the 5-10 keV survey simply picks out hard (e.g. flat or absorbed) objects and misses soft ones, or whether 5-10 keV detection selects object in a less biased way. In the standard AGN synthesis picture where the objects are harder due to absorption we expect the latter to be the case. At the same equivalent flux limit, the 5-10 keV survey should pick up all of the unabsorbed objects, but it should also find absorbed objects that are missed in the softer surveys. This holds in part because the intrinsic spectral index of AGN is approximately $\\Gamma=2.0$, and therefore equal intrinsic flux is emitted per unit energy. If the only modifier is absorption, then at the same flux limit the 5-10 keV survey should pick up all objects in the 2-10 keV surveys, and in addition all objects missed by the 2-10 keV surveys because their flux is depressed by absorption in the 2-5 keV band.  These issues are partially resolved by our analysis, although our data are somewhat contradictory. The fact that we (and HELLAS) find a very similar logN-logS function to the 2-10 keV surveys suggests that we are sampling the same populations. Two effects indicate that this is not the case, however. First, the hardness ratio analysis clearly indicates that the SHEEP sources have significantly harder spectra than those obtained in 2-10 keV surveys. As we have discussed above, the mean hardness ratios is rather difficult to calculate in a robust manner, and is subject to statistical bias, but we can compare our preferred value of $\\Gamma=1.1\\pm 0.1$ with that from 2-10 keV surveys.  Both Della-Ceca et al. (1999) and Ueda et al. (1999) have presented mean values for 2-10 keV index derived from direct spectral fitting of stacked spectra. The former find $\\Gamma=1.74\\pm 0.07$, taking the weighted average of their ``bright and ``faint'' subsamples, and the latter $\\Gamma=1.49 \\pm 0.10$ from a 2-10 keV sample which was flux-selected to ignore the brightest sources. The ``faint'' subsample of Della Ceca et al. (1999) is marginally consistent with our preferred spectrum, with $\\Gamma=1.36 \\pm 0.14$  Our analysis has highlighted the difficulty in determining the average spectral properties of sources detected in different ways in different surveys. To provide what is perhaps the fairest possible comparison, we have computed the HR1 hardness ratio for a subsample of the ASCA sources of Ueda et al. (2001). We considered only serendipitous sources (i.e. not the targets) and truncated their sample at a detection level of 4.5$\\sigma$ (i.e. our detection threshold) in the 2-10 keV band, which resulted in a total of 601 sources. The mean, unweighted HR1 value of this sample is HR1=$-0.02 \\pm 0.01$, corresponding to $\\Gamma=1.75 \\pm 0.02$.  The corresponding HR1 for our sample is stated in Table~\\ref{tab:hr} and corresponds to $\\Gamma=1.11\\pm 0.11$.  These surveys were performed with the same instrument, and the hardness ratios were calculated in the same band and by the same method. The only substantive difference should therefore be the selection band (2-10 keV in the Ueda et al. subsample vs. 5-10 keV for SHEEP). We further note that the Eddington bias should artificially harden the value from the 2-10 keV survey (because the sources were selected in that band) but not the mean SHEEP spectrum. Thus we can firmly conclude that, based on the spectral form, our 5-10 keV survey selects a different and much harder population than 2-10 keV surveys.  Although the logN-logS functions are similar in the 2-10 keV and 5-10 keV bands, a few more objects can be accomodated in the 5-10 keV counts, particularly given the uncertainty in spectral shape and therefore the conversion of counts to flux. To put this on a more quantitative footing, we have taken the maximum difference in the logN-logS normalization comparing our and Cagnoni et al.'s 2-10 keV survey of 15 per cent. Could this additional 15 per cent of objects harden our average spectrum sufficiently to cause the difference between our survey and the 2-10 keV samples?  We have tested this by excluding the hardest 9 sources (i.e. 15\\%) based on their HR1 value) from the SHEEP sample and recomputing the hardness ratio. We find a mean hardness for these 60 sources corresponding to $\\Gamma=1.33\\pm 0.07$, still considerably flatter than the 2-10 keV value. Similar results are found when excluding objects based on their HM and HS hardnesses.  We can therefore further conclude that the 5-10 keV survey does not simply pick up a few {\\it additional} hard objects compared to the 2-10 keV surveys, but rather samples a different population.  Additional supporting evidence for this conclusion comes from the fact that we find no trend for the source population to harden at faint fluxes, which is found in the 2-10 keV surveys.  HELLAS similarly fails to find such a correlation (Fiore et al. 2001), and we also note that the Chandra survey of Moretti et al. (2002) reveals no correlation between hardness and 2-10 keV flux, although such a correlation is observed with the flux in the softer 0.5-2 keV band (Giacconi et al. 2001; Moretti et al. 2002). While there is some evidence that the brightest objects in our survey are softer than the mean, it appears that the 5-10 keV selection methods digs into the faint, hard populations which make up the X-ray background much more quickly than the 2-10 keV surveys. Thus, despite the good agreement between the number counts, we cannot be sampling the same populations as the 2-10 keV surveys.  Presumably this is also true of HELLAS, although no mean spectrum has been given for these sources.  We are then led to the conclusion that the agreement between the 2-10 keV and 5-10 keV number counts is largely coincidental, with our 5-10 keV survey picking up many additional hard objects, but almost exactly compensating for this in terms of numbers by losing softer ones.  The fact that the number counts from the 5-10 and 2-10 keV surveys agree so well, but that the populations are clearly different spectrally is troubling for the population synthesis models (e.g. Madau, Ghisselini \\& Fabian 1994; Comastri et al. 1995, 2001; Gilli et al. 1999, 2001). The problem is that these models generally assume an intrinsic spectrum for AGN of the form $\\Gamma=1.9$, typical of local AGN, or even flatter. The intrinsic flux of such a power law per unit energy is larger in the 5-10 keV band than in the 2-5 keV band or for that matter the 2-10 keV band.  Furthermore, absorption is invoked for a very large fraction of these sources which further depresses the 2-5 and 2-10 flux relative to the 5-10 keV flux. For example 75~per cent of sources in the model of Gilli et al. (1999) have $N_{\\rm H} > 10^{23}$~cm$^{-2}$. At z=0, this column density suppresses the 2-10 keV flux by a factor $\\sim 2$, while the 5-10 keV flux only changes by $\\sim 15$ per cent. These numbers are almost identical for a more typical AGN synthesis source with $N_{\\rm H}=10^{24}$~cm$^{-2}$ at z=1.5. Thus, if the populations synthesis models are correct, and the absorbed populations have the same luminosity function and evolutionary properties as the unabsorbed ones, we expect much higher number counts (perhaps by a factor $\\sim 2$) in the 5-10 keV logN-logS than for the converted 2-10 keV. Such a conclusion is grossly incompatible with our data (Fig.~\\ref{fig:xnc}), and those from HELLAS. We note, however, that Comastri et al. (2001) have claimed consistency of both the 5-10 keV and 2-10 keV number counts with the synthesis models. This may in part be due to the fact that the proportion of absorbed sources in the synthesis models depends on the flux limit. Specifically, a higher proportion of absorbed objects is expected at fainter fluxes. In this scenario, the apparent agreement in number counts means that at the flux limits probed by SHEEP and HELLAS, the number of heavily obscured sources must be very small. It is very hard to see how this can be the case given the strong difference in spectra we find between the 5-10 and 2-10 keV populations.  An alternative possibility is that there is a large population of very soft sources (unobscured and with $\\Gamma > 2.0$ ) which are missed in the 5-10 keV surveys and picked up at 2-10 keV.  This is not postulated typically in the synthesis models, where the obscured populations dominate and where ``soft excesses'', if present, never affect the spectra above 2 keV.  We await further detailed modelling of the number counts to address these issues, but note that there have been some tentative suggestions that much of the X-ray background may be produced at low redshift ($z<1$; Tozzi et al. 2001; Rosati et al. 2002), in agreement with our finding that the luminosity function and/or evolution of the absorbed populations is likely to differ from that of standard QSOs.  \\subsection{The nature of the X-ray background sources}  Chandra deep surveys have resolved most of the X-ray background into discrete sources (Mushotzky et al. 2000; Giacconi et al. 2001; Brandt et al. 2001b). The astrophysical nature of these sources remains mysterious, however. Many of them are extremely faint in the optical (Mushotzky et al. 2000; Barger et al. 2001; Alexander et al. 2001) making spectroscopic identification impossible. What is required to make further progress, then, is to find bright, nearby examples of these objects that can be detected and studied in more detail. To do this requires surveys with much larger area than those possible currently with Chandra. The SHEEP survey, like HELLAS, with its large area and hard X-ray selection criterion, provides such a sample. Indeed it is very clear from the failure of ROSAT to detect a large fraction of our sources (which are nonetheless very bright in the hard X-ray band), that there are a large number of hard and probably obscured sources. While these are likely AGN, how their astrophysics is related to the more familiar classes of Seyfert 1s, Seyfert 2s and QSOs in the local and soft X-ray universe remains an open question. Optical identification and detailed X-ray spectroscopy of our sample can resolve this issue.  One key question is whether our sources present hard spectra due to a large amount of intrinsic absorption or whether they are intrinsically hard. The population synthesis models predict the former, but it is possible that the sources which make up the hard X-ray background have flat spectra for other reasons, such as the radiation mechanism. For example, photon-starved Comptonization or bremsstrahlung emission from an ADAF would produce a spectrum similar to that of the X-ray background.  A pure reflection spectrum, e.g. in the case of a Compton-thick Seyfert galaxy could produce an even flatter spectrum (Reynolds et al. 1994; Matt et al. 1996, 2000). We find no clear answer to this in our hardness ratio analysis, but the most likely situation is that the X-ray spectra are composite, with an absorbed, hard power law and soft emission that may be either scattered nuclear light, or a separate thermal component (see also Della Ceca et al. 1999; Giommi et al. 2000; Vignali et al. 2001). Followup observations of the HELLAS and SHEEP sources with XMM will determine this unambiguously.  One early indication from HELLAS was that there may be a population of ``red quasars'' (Webster et al. 1995), with hard and possibly absorbed X-ray spectra (Fiore et al. 1999; Vignali et al. 2000). Complete optical followup of the SHEEP sample will confirm this, but here we highlight another potentially interesting class, of hard QSOs (see also Comastri et al. 2001). Our cross-correlation with the NED catalog shows two sources which are classified optically as QSOs, but whose hardness ratios indicate extremely hard spectra that correspond to $\\Gamma<1.0$ if they are unabsorbed. We do not have complete optical spectra or spectral energy distributions of these sources, so these may also be red quasars and absorbed in the optical. The QSO classification, however, implies that we are seeing the nuclear broad lines directly. Again these objects may simply have intrinsically hard spectra, but to flatten a more typical QSO spectrum of $\\Gamma \\sim 1.9$ to the hard value observed requires a column density $\\gg 10^{23}$~\\pcmsq. The dust associated with such gas would likely obliterate the optical/UV emission, including the broad lines, causing the source to appear as a type II quasar. That the broad lines are in fact observed implies that the line of sight is not particularly dusty. One possibility is that the gas-to-dust ratio in these objects differs substantially from Galactic values (Maiolino et al. 2001). Another is that we are seeing hot and /or photoionized gas - or ``warm absorbers'' (Halpern 1984) - at high redshift. Such a gas component is commonly observed in low redshift Seyferts (e.g. Nandra \\& Pounds 1994; Reynolds 1997; George et al. 1998). The low redshift analogue of these QSOs is the famous Seyfert galaxy NGC 4151, which shows strong UV emission and broad optical/UV emission lines, but which is heavily absorbed in the X-ray band. The most obvious explanation for this is that there is dust-free gas near the nucleus, and this interpretation is supported by the fact that the X-ray column NGC 4151 is apparently mildly ionized (Yaqoob, Warwick \\& Pounds 1989; Weaver et al. 1994).  \\subsection{A complete, hard X-ray selected sample}  Our survey has defined a new, hard X-ray selected sample of AGN. Along with the HELLAS AGN, these will be the first complete samples since the HEAO-1 survey (Piccinotti et al. 1982). Many of the objects in the Piccinotti sample are the most heavily observed AGN and their detailed study has revealed much of what we know about nuclear activity in galaxies. Those sources were almost exclusively nearby Seyferts, however, and our survey has already revealed a large number higher-redshift and higher-luminosity AGN. Follow-up observations of these hard X-ray bright objects could revise our opinions of the central regions of AGN. If, as is suggested by the Chandra data, the majority of AGN have been missed by optical surveys, much of what we think we know about their properties could be misleading. With hard X-ray selection we avoid the biases against obscuration inherent in most other methods of selection, and if we can follow up these observations with high quality data in other wavebands, our opinions about AGN phenomenology could change dramatically.  \\subsection{Future work}  Optical followup of our sources is already in progress, with an imaging program and some spectroscopy, particularly for the northern sources.  A significant problem, however, is that the GIS positions are not alone sufficient to identify unambiguously the optical counterpart. ROSAT PSPC positions are better and HRI positions the best available, but as we have shown, the ROSAT detected sources represent a biased subsample, and these sources probably do not represent the population providing the bulk of the energy density of the X-ray background at 30 keV.  What is really required is to obtain Chandra and/or XMM observations of our sources, which will give us the optical counterparts without ambiguity. Such observations have the added advantage that they will allow us to determine the X-ray extent and spectra of the source populations, providing a crucial piece in the puzzle of how the bulk of the extragalactic background light at hard X-ray energies is produced."
    },
    "0209/astro-ph0209235_arXiv.txt": {
        "abstract": "Before the end of 2002 will be launched the GALEX satellite (a NASA/SMEX project) which will observe all the sky in Ultraviolet (UV) through filters at 1500 and 2300 $\\AA$ down to m(AB)$\\sim 21$.\\\\ In 2004 will be launched the ASTRO-F satellite which will perform an all sky survey at Far-Infrared (FIR) wavelengths.\\\\ The cross-correlation of both suveys will lead to very large samples of galaxies for which FIR and UV fluxes will be available. Using the FIR to UV flux ratio as a quantitative tracer of the dust extinction we will be able to measure the extinction in the nearby universe (z$<$0.2) and to perform a statistically significant analysis of the extinction as a function of galactic properties.\\\\ Of particular interest is the construction of pure FIR and UV selected samples for which the extinction will be measured as templates for the observation of high redshift galaxies. ",
        "introduction": "The problem of the dust extinction in galaxies is crucial for the study of galaxy formation and evolution. Indeed the light emitted by the stars is partially absorbed by the dust before escaping galaxies: the determination of the true stellar content and of the star formation activity inside galaxies can only be done once the amount of extinction is known. The situation is dramatic at ultraviolet wavelengths which are observed in the visible at high redshift and where the extinction is particularly severe (e.g. Meurer et al. 1999, Steidel et al. 1999). ",
        "conclusions": ""
    },
    "0209/astro-ph0209003_arXiv.txt": {
        "abstract": "We present a variational formalism for describing the dynamical evolution of an oscillating star with a point-mass companion in the linear, non-relativistic regime. This includes both the excitation of normal modes and the back-reaction of the modes on the orbit. The general formalism for arbitrary fluid configurations is presented, and then specialized to a homentropic potential flow. Our formalism explicitly identifies and conserves both energy and angular momentum. We also consider corrections to the orbit up to 7/2 post-Newtonian order. ",
        "introduction": "When a star orbits a companion, there can be a resonant excitation of stellar oscillations if the orbital period becomes close to a multiple of the oscillation period. This can happen as a result of gravitational radiation or interaction with a third body, for example. This is a deceptively complex dynamical system because there will be a back-reaction of the oscillation on the orbit. In this paper we consider a variational approach to treating this problem.  Variational principles for linear, adiabatic, non-radial stellar oscillations have been considered by several people \\citep[e.g.][]{cha64,lyn67}. Subsequently, a variational approach to the excitation of tides in binary systems was considered by \\citet{Gin80}. However, their description was limited to polytropic equations of state, and did not account for orbital evolution due to general relativistic effects which are significant for the long-term evolution of systems involving compact stars. In this paper, we consider a more general variational approach to tidal excitation. Although we limit our detailed discussion to irrotational flow, the formalism is presented in a way so as to allow generalisation to more complicated flows.  In \\S 2, we provide an overview of one variational approach to the dynamics of perfect fluids. We then derive the most general Lagrangian for tidal excitation, and consider the special case of homentropic, irrotational flow. In \\S 3, we derive the equations for tidal excitation in terms of the normal mode amplitudes, and obtain expressions for the conserved energy and angular momentum. We also consider corrections to the orbit due to general relativistic effects. In \\S 4, we discuss the applicability of our formalism. Finally, in \\S 5, we reprise our conclusions. ",
        "conclusions": "In this paper, we have described and developed a variational approach for handling a specific dynamical problem---the oscillation-orbit interaction of a non-rotating star with a compact object companion. Our treatment is presented in some generality because we believe that this method can also be employed in other problems associated with accretion disks and extra-solar planets. We recover the standard, Newtonian equations of fluid dynamics for irrotational flow as wells as normal mode equations for small oscillations. The power of the Lagrangian approach is made manifest in the identification of algebraic expressions for the modal energy and angular momentum, and for the equations of motion. It also suggests an approach to handling more complex problems, including those involving strong-field gravity and rotational flow.  In a forthcoming paper, we will use this approach to compute the dynamical evolution of a white dwarf in orbit about a compact object as it spirals inward under the action of gravitational radiation."
    },
    "0209/astro-ph0209373_arXiv.txt": {
        "abstract": "The {\\sc Planck} mission, originally devised for cosmological studies, offers the opportunity to observe Solar System objects at millimetric and submillimetric wavelengths. We concentrate in this paper on the asteroids of the Main Belt, a large class of minor bodies in the Solar System. At present more that 40000  of these asteroids have been discovered and their detection rate is rapidly increasing. We intend to estimate the number of asteroids that can can be detected during the mission and to evaluate the strength of their signal. We have rescaled the instrument sensitivities, calculated by the LFI and HFI teams for sources fixed in the sky, introducing some degradation factors to properly account for moving objects. In this way a detection threshold is derived for asteroidal detection that is related to the diameter of the asteroid and its geocentric distance. We have developed a numerical code that models the detection of asteroids in the LFI and HFI channels during the mission. This code perfoprms a detailed integration of the orbits of the asteroids in the timespan of the mission and identifies those bodies that fall in the beams of Planck and their signal stenght. According to our simulations, a total of 397 objects will be observed by {\\sc Planck} and an asteroidal body will be detected in some beam in 30\\% of the total sky scan--circles. A significant fraction (in the range from $\\sim 50$ to 100 objects) of the  397 asteroids will be observed with a high S/N ratio.\\\\ Flux measurements of a large sample of asteroids in the submillimeter and millimeter range are relevant since they allow to analyze the thermal emission and its relation to the surface and regolith properties. Furthermore, it will be possible to check on a wider base the two standard thermal models, based on a nonrotating or rapidly rotating sphere.\\\\ Our method can also be used to separate Solar System sources from cosmological sources in the survey. This work is based on {\\sc Planck} LFI activities. ",
        "introduction": "The {\\sc Planck}~ESA mission\\footnote{http://astro.estec.esa.nl/SA-general/Projects/Planck/} will perform a high-angular resolution mapping of the microwave sky over a wide range of frequencies, from 30 to 900 GHz. These data will have important implications on different fields, from cosmology and fundamental physics to Galactic and extragalactic astrophysics. The Solar System astronomy will also benefit from the Planck mission since it will offer the opportunity to perform a survey of Solar System objects at millimetric wavelengths. Planck will be sensitive to the millimetric emissions from planets and from a significant fraction of the asteroidal population. In this paper we focus on the detection of asteroid radio emissions and on the identification of the asteroidal targets that will be observed during the mission. The relevant parameters for the possible detection  of a minor body by {\\sc Planck} are the geocentric distance and the diameter. We will concentrate in this paper on Main Belt asteroids that can be more easily detected since their orbits are at relatively small geocentric distances and several objects have diameter larger than 50~km.\\\\ Radio observations of asteroids from the Earth were performed by \\citeasnoun{Redman1998} at seven different frequencies ranging from 150 to 870 GHz. Redman focused on the information that radio data can give to the thermal models of asteroids. Their set of targets included 5 asteroids already observed by \\citeasnoun{Altenhoff1994} and two new objects. \\citeasnoun{Altenhoff1994} observed, at 250 GHz, 15 among the largest asteroids, but only to determine the diameter and to compare their measurements with other methods.\\\\ The emissivity obtained at radio frequencies may probe the temperature of the body at different depths on the surface allowing to derive precise values of the average temperature. Moreover, relevant physical data concerning the density of the regolith layer covering the surface of the asteroids can also be determined. \\\\ Radio observations of asteroids by {\\sc Planck} offer a unique opportunity to improve our knowledge on the thermal emission in the millimetric and submillimetric domain, increasing our knowledge of the physical nature of the surface layers of the objects. Considering that radio data provide an estimate of the density on a scale related to the wavelength of the observations, the advantage of using Planck will be in the possibility of gathering data on a large frequency spectrum and then to probe the temperature and regolith density at different depths below the surface.\\\\ Furthermore, the {\\sc Planck} data can help to refine the thermophysical models using also physical studies of the thermal properties of stony and FeNi meteorites as well as inferences about surface physical properties derived from asteroid radar studies. It will be possible to relate the {\\sc Planck} data to the two standard models which are commonly invoked to predict the thermal radiation from asteroids, considering the asteroid as a nonrotating sphere or a rapidly rotating sphere. \\\\ For instance Vesta has been observed on a wide range of wavelengths, from submillimeter to centimeter, and the ratio of the observed flux, at each wavelength (from 12$\\mu$m to 6cm), divided by the flux expected from a blackbody at the temperature of a nonrotating asteroid shows a behavior different from the standard models \\cite{Redman1992}. A successful model for the thermal emission must explain why the apparent flux remains below the rapidly rotating sphere value, and much below the nonrotating sphere value, over the entire submillimeter range. Several mechanisms can lower the apparent temperature at radio wavelengths as reflections at the outer surface of the regolith, reduction in the actual temperature of the deeper layers of the regolith due to diurnal temperature variations at the surface, or scattering by grains within the regolith which can reduce the emissivity in a wavelength dependent fashion. The actual emission from an asteroid may combine all three effects in differing amounts at each wavelength, but only very few objects have been observed in the submillimeter and millimeter range and it is not possible to identify a typical behavior or to relate the results to different asteroid taxonomic classes.\\\\ It is worth to point out that the main belt asteroids may represent a source of background for the {\\sc Planck} mission and, as a consequence, they should be considered in the definition of the scientific and technical aspects of the mission. \\\\ The thermal emission of asteroids, apart from their intrinsic scientific relevance, has also to be considered as a potential source of spurious detections in the analysis of Galactic or extragalactic sources. The precise determination at each epoch of the eventual presence of an asteroid, with its point emission, in some of the {\\sc Planck} beams is a critical requirement to avoid systematic errors in mapping the astrophysical sources of radiation. The high nominal accuracy of the {\\sc Planck} instruments suggests a tight control even for the contamination from asteroids below the detection threshold.\\\\ Section~2 is devoted to the discussion of the {\\sc Planck} sensitivity for the detection of moving sources at different frequency channels. The averaged final {\\sc Planck} sensitivities will be rescaled in order to take into account the positions of the different antenna beams in the telescope field of view and the main properties of the {\\sc Planck} scanning strategy. These results are used in Section~3 to determine the threshold to detect asteroids in the {\\sc Planck} data streams, according to their typical temperature and size. In Section~4, we describe our numerical code to compute the asteroid transits on the different {\\sc Planck} beams and to estimate their signals.  The results of this analysis are presented in Section~5. Finally, we discuss the results and draw our main conclusions in Section~6. ",
        "conclusions": "The {\\sc Planck} ESA mission will perform a high-angular resolution mapping of the microwave sky over a wide range of frequencies, from 30 to 900 GHz. During the two years planned for the mission the surveyor will frame a large number of main-belt asteroids. In this paper we investigate in detail for the first time the possibility for {\\sc Planck} to detect main-belt asteroids. The main parameter in the asteroid detection process by the {\\sc Planck} horns is the $R/d$\\ ratio. The instrumental sensitivity to such objects has then been defined as the minimum $R/d$\\ value required to safely observe them. This value is given by the {\\sc Planck} sensitivity at different frequencies rescaled to take into account the integration time and the confusion noise due to the background. An accurate evaluation of the noises and the sensitivity degradation, related to the high proper motion of the asteroids, yielded a minimum $R/d$ ratio of $(R/d)_{\\mathrm{min}} \\sim 10^{-7}$.\\\\ A numerical simulation of a two-years mission has been performed in order to estimate the number of objects whose $(R/d)$\\ is greater than about  $10^{-7}$\\ and that will then be observed by {\\sc Planck}. The simulation uses updated catalogs for the orbital elements of the main-belt asteroids and their diameters. The orbit of each asteroid has been integrated with a very short timestep to compute the relative position of the body with respect to the {\\sc Planck} horns and the value of $(R/d)$. An accurate mission simulation has been obtained. Up to 397 asteroids are expected to be detected in the various {\\sc Planck} channels (Table 3). Detectable asteroids will appear in about 30\\% of the total sky circles scanned by the mission. \\\\ The previous survey of Main Belt Asteroids was performed by the Infrared Astronomical Satellite (IRAS) on 1983 in four wavelength bands centered near 12, 25, 60 and 100$\\mu$m, much lower than the minimum wavelength observed by {\\sc Planck} of 350$\\mu$m, corresponding to the highest frequency of 857 GHz. It surveyed approximately 96\\% of the sky and 2228 different multiply observed asteroids were associated to IRAS sources \\cite{Tedesco2002}, providing a good estimate of diameter and albedo for most of them.\\\\ IRAS's 12$\\mu$m limiting sensitivity, for S/N=3, was about 150 mJy \\cite{Tedesco2002b}, and assuming a Main Belt Asteroid as a black body at 150 K this flux can be translated to about 56 mJy at 350$\\mu$m (857 GHz), well below the {\\sc Planck} noises reported in Table 2.\\\\ {\\sc Planck} will provide flux measurements for a smaller sample of asteroids ($\\sim 50-100$ objects) compared to IRAS, but at different wavelengths, almost unexplored for this class of Solar System objects. This will improve our understanding of the thermal emission and the related surface properties of asteroids."
    },
    "0209/gr-qc0209102_arXiv.txt": {
        "abstract": "We solve Einstein's field equations coupled to relativistic hydrodynamics in full 3+1 general relativity to evolve astrophysical systems characterized by strong gravitational fields.  We model rotating, collapsing and binary stars by idealized polytropic equations of state, with neutron stars as the main application. Our scheme is based on the BSSN formulation of the field equations.  We assume adiabatic flow, but allow for the formation of shocks. We determine the appearance of black holes by means of an apparent horizon finder. We introduce several new techniques for integrating the coupled Einstein-hydrodynamics system.  For example, we choose our fluid variables so that they can be evolved without employing an artificial atmosphere. We also demonstrate the utility of working in a rotating coordinate system for some problems.  We use rotating stars to experiment with several gauge choices for the lapse function and shift vector, and find some choices to be superior to others.  We demonstrate the ability of our code to follow a rotating star that collapses from large radius to a black hole.  Finally, we exploit rotating coordinates to evolve a corotating binary neutron star system in a quasi-equilibrium circular orbit for more than two orbital periods. ",
        "introduction": "\\label{intro}  With the availability of unprecedented observational data, the physics of compact object is entering a particularly exciting phase.  New instruments, including X-ray and $\\gamma$-ray satellites and neutrino observatories, are detecting signals from highly relativistic events in regions of strong gravitational fields around neutron stars and black holes.  A new generation of gravitational wave interferometers is promising to open a completely new window for the observation of compact objects.  The ground-based gravity wave observatories LIGO and TAMA are already operational and are collecting data, GEO and VIRGO will be completed soon, and a space-based interferometer LISA is currently under design.  Given the small signal-to-noise ratio in these new gravitational wave detectors, theoretical models of likely sources are needed for the positive identification of the signal as well as for its physical interpretation \\cite{c93}.  One promising technique for the identification of signals in the noise output of the detector is matched filtering, which requires accurate theoretical gravitational wave templates \\cite{fh98}.  The need for such templates has driven a surge of interest in developing reliable techniques capable of their construction.  Compact binaries, i.e.~binaries consisting of either black holes or neutron stars, are among the most promising sources of gravitational radiation.  Much progress has been made in refining post-Newtonian point-mass approximations.  These are suitable for large binary separations for which relativistic effects are sufficiently small and any internal structure can be neglected~\\cite{bd00}.  At small binary separations, the most promising technique for modeling the inspiral, coalescence and merger is numerical relativity.  Several other observed phenomena involving compact objects require numerical relativity for their modeling.  One such example is Gamma Ray Bursts (GRBs).  While it is not yet known what the origin of GRBs is, the central source is almost certainly a compact object~\\cite{npp92}.  Most scenarios involve a rotating black hole surrounded by a massive magnetized disk, formed by a supernova, or the coalescence of binary neutron stars \\cite{Piran:2002kw}.  To confirm or refute any GRB scenario requires numerical studies in full 3+1 relativistic magnetohydrodynamics.  Another astrophysical scenario requiring numerical treatment is the formation of supermassive black holes (SMBHs).  Among the scenarios proposed to explain SMBH formation are the collapse of a relativistic cluster of collisionless matter, like a relativistic star cluster \\cite{rsc} or self-interacting dark matter halo \\cite{sidmh}, or the collapse of a supermassive star~\\cite{r01}. Depending on the details of the collapse, SMBH formation may generate a strong gravitational wave signal in the frequency band of the proposed space-based laser interferometer LISA.  Understanding the SMBH formation route may shed key insight into structure and galaxy formation in the early universe.  Solving the coupled Einstein field and hydrodynamics equations is a challenging computational task, requiring the simultaneous solution of a large number of coupled nonlinear partial differential equations. In addition to all of the usual problems of numerical hydrodynamics -- handling advection, shock discontinuities, etc -- one encounters the problems inherent to numerical relativity.  The latter include identifying a suitable formulation of Einstein's field equations, enforcing a well-behaved coordinate system, and, if black holes are formed, dealing with spacetime singularities.  The construction of self-consistent numerical solutions to the coupled equations of relativistic hydrodynamics and gravitation dates back to the pioneering work of May and White in spherical symmetry \\cite{May:1966} (see also \\cite{Font:2000} for a review).  In one of the first attempts to perform numerical integrations in three spatial dimensions, Wilson, Mathews, and Marronetti~\\cite{Wilson:1989,Wilson:1995ty,Wilson:1996ty} (see \\cite{Flanagan:1999,Mathews:2000} for later corrections) tackled the binary neutron star problem.  They simplified Einstein's field equations by assuming that the spatial metric remains conformally flat at all times.  Their implementation of relativistic hydrodynamics was based on earlier work by Wilson \\cite{Wilson:1972} and used upwind differencing to handle advection and artificial viscosity to capture shocks.  The first fully self-consistent relativistic hydrodynamics code, which treats the gravitational fields without approximation, was developed by Shibata \\cite{Shibata:1999aa}. This code, based on earlier work by Shibata and Nakamura~\\cite{sn95}, adopts a Van Leer hydrodynamics scheme~\\cite{vl77,on89} and also employs artificial viscosity for shocks.  This code has been used in various astrophysical applications, including the coalescence and merger of binary neutron stars \\cite{Shibata:1999wm,Shibata:2002jb} and the stability of single, rotating neutron stars \\cite{sbs00a,sbs00b,bss00}.  In an alternative approach, Font {\\it et al}~\\cite{fmst00} implemented a more accurate high-resolution shock-capturing technique to solve the equations of relativistic hydrodynamics.  This code has been used to study pulsations of relativistic stars \\cite{fgimrssst02}.  In this paper, we report on the status and some astrophysical applications of our new 3+1 general relativistic hydrodynamics code. Our code, based on the so-called Baumgarte-Shapiro-Shibata-Nakamura (BSSN) formulation of Einstein's equations \\cite{sn95,bs98b}, has several novel features, including an algorithm that does not require the addition of a tenuous, pervasive atmosphere that is commonly used in Eulerian hydrodynamical codes, both Newtonian and relativistic.  This ``no atmosphere'' algorithm proves to be very robust and eliminates many problems associated with the traditional atmospheric approach \\cite{Swesty:1999ke}.  We treat 1D shocks, spherical dust collapse to black holes, and relativistic spherical equilibrium stars to demonstrate the ability of our code to accurately evolve the coupled field and hydrodynamic equations in relativistic scenarios.  We then use the evolution of stable and unstable uniformly rotating polytropes as a testbed to determine which gauge conditions are best-behaved in the presence of strong-field matter sources with significant angular momentum.  We introduce rotating coordinate systems and show that these can yield more accurate simulations of rotating objects than inertial frames.  We demonstrate the ability of our code to hold accurately stable differentially rotating stars in equilibrium. We also show that our code can follow the collapse of rapidly differentially rotating stars reliably until an apparent horizon appears, by which time the equatorial radius has decreased from its initial value by more than a factor of ten.  We then turn to simulations of binary neutron stars.  We adopt initial data describing corotating $n=1$ polytropes in quasi-equilibrium circular orbit, and evolve these data for over two orbital periods.  In this paper we present results for one particular binary and discuss the effect of corotating frames as well as the outer boundaries.  An extended study, including binary sequences up to the dynamically identified innermost stable circular orbit (ISCO), will be presented in a forthcoming paper \\cite{Marronetti:2002}.  This paper is organized as follows.  Secs.~\\ref{fields} and \\ref{hydro} describe our method of evolving the field and hydrodynamic equations, respectively.  Sec.~\\ref{gauge_choices} summarizes the various gauge choices with which we experiment.  Sec.~\\ref{diag} lists the diagnostics used to gauge the reliability of our simulations. Sec.~\\ref{tests} describes several tests of our algorithm. Sec.~\\ref{iso_stars} applies our formalism to evolve non-rotating, uniformly rotating, and differentially rotating polytropes.  In Sec.~\\ref{Bin} sketches our binary neutron star calculations.  Our results are summarized in Sec. \\ref{summary}.  Some details of our hydrodynamic scheme and the rotating frame formalism are presented in the appendices. ",
        "conclusions": "\\label{summary}  We have tested our 3+1 relativistic hydrodynamics code on a variety of problems.  We find that our current algorithm, supplemented by driver gauge conditions, is rather robust.  The grid resources required for stable evolution and reasonable accuracy are modest. We accurately evolve shock tubes, spherical dust collapse, and relativistic spherical polytropes.  We also evolved uniformly and differentially rotating equilibrium polytropes, and maintained stable configurations stationary for several rotational periods.  Two applications carried out with our code are particularly significant. First, we examined the collapse from large radius of a star with significant spin to a Kerr black hole.  Second, we evolved stable binary neutron stars for several orbits, maintaining quasi-circular equilibrium.  The first application indicates that we can study the effects of angular momentum on gravitational collapse and on the resulting waveform, an effort already initiated in \\cite{n81,Shibata00}.  The second application indicates that we can identify and evolve dynamically stable quasi-circular neutron star binaries.  This ability can be used to locate the ISCO dynamically and to follow the transition from an inspiral to a plunge trajectory. In addition, dynamic simulation allows us to improve binary initial data, for example by allowing initial ``junk'' gravitational radiation to propagate away.  We also hope to compute detailed gravitational waveforms form these binaries, refining the wavetrains reported in \\cite{dbs01,dbssu02}.  We note that several challenges remain to be addressed before there exists a code capable of modeling all the gravitational wave sources of current astrophysical interest.  One problem is the need to maintain adequate grid coverage of the collapsing star or inspiralling binary while still keeping the outer boundaries sufficiently distant, i.e. the problem of dynamic range. Adaptive mesh techniques far more sophisticated than the crude rezoning used here may be necessary.  A related problem concerns gravitational wave extraction, as it currently is not possible to place outer boundaries in the wave zone.  Finally, the formation of black hole singularities in hydrodynamic collapse scenarios remains an additional challenge to determining the late-term behavior of such systems.  Special singularity-handling techniques, such as excision, need to be developed further."
    },
    "0209/astro-ph0209145_arXiv.txt": {
        "abstract": "Competing models for broad spectral features in the soft X-ray spectrum of the Seyfert I galaxy Mrk~766 are tested against data from a 130~ks XMM-Newton observation. A model including relativistically broadened Ly$\\alpha$ emission lines of \\ion{O}{8}, \\ion{N}{7} and \\ion{C}{6} is a better fit to 0.3-2 keV XMM-RGS data than a dusty warm absorber. Moreover, the measured depth of neutral iron absorption lines in the spectrum is inconsistent with the magnitude of the iron edge required to produce the continuum break at 17-18\\AA\\ in the dusty warm absorber model. The relativistic emission line model can reproduce the broad-band (0.1-12 keV) XMM-EPIC data with the addition of a fourth line to represent emission from ionized iron at 6.7 keV and an excess due to reflection at energies above the iron line. The profile of the 6.7 keV iron line is consistent with that measured for the low energy lines. There is evidence in the RGS data at the 3$\\sigma$ level for spectral features that vary with source flux. The covering fraction of warm absorber gas is estimated to be ~12\\%. Iron in the warm absorber is found to be overabundant with respect to CNO compared to solar values. ",
        "introduction": "The combination of the Reflection Grating Spectrometer (RGS) and the high throughput of XMM-Newton constitutes a powerful tool for probing the soft X-ray spectrum of bright AGN. The increase in spectral resolution has already caused a revision in thinking and has revealed a complex spectrum of narrow absorption lines arising in material with a range of ionisation states (e.g. IRAS 13349+2438; Sako et al. 2001). The XMM-RGS spectra of the Seyfert galaxies MGC-6-30-15 and Mrk~766 show evidence of broad spectral features (Branduardi-Raymont et al. 2001; hereafter BR01; see also Sako et al. 2002) and a similar spectrum of MCG-6-30-15 has been recorded using the grating spectrometers on the Chandra observatory (Lee et al. 2001). The nature of these features is controversial. Two competing models are:  (1) absorption edges imprinted on the underlying spectrum by a dusty, warm absorber (Lee et al. 2001), (2) emission lines of \\ion{O}{8}, \\ion{N}{7} and \\ion{C}{6} broadened by relativistic motion in an accretion disk around a massive spinning black hole (BR01).  In this paper we discuss the first results of a long (130 ksec) XMM-Newton observation of Mrk~766. This has better statistics than the shorter observation of BR01. We examine how well the competing models are able to reproduce the RGS data. The way in which the spectral features change with source flux is investigated by comparing the mean spectrum taken during flaring episodes with that from intervals between flares. We also discuss the nature of the warm absorber. The spectrum in the full X-ray energy range of XMM-Newton is investigated, combining the RGS data with those recorded at the same time by the EPIC CCD spectral imager. We compare the low energy spectral features to the $\\sim$ 6.7 keV iron line detected using EPIC. ",
        "conclusions": "The long observation of Mrk~766 reported in this paper, which sampled the source at a higher mean flux level than previously, provides the opportunity to test the rival models that have been put forward to explain the low-energy spectral features first reported by BR01. We have sufficient signal-to-noise to constrain the variability of these features as a function of source flux. We have also used the EPIC instrument on XMM-Newton to confirm the presence of a broad, asymmetric iron line feature at 6.7 keV, and use the combination of the RGS and EPIC instruments on XMM-Newton to compare the iron line profile with the low-energy spectrum.  A point of debate in the literature has been the discontinuity in the spectrum of Mrk~766 at about 18\\AA. A similar feature has also been seen in MCG-6-30-15 by BR01 and Lee et al. (2001). Conceivably it is an absorption edge. Its wavelength, however, is inconsistent with the O{\\sc VII} edge expected from a warm absorber unless the material producing it is redshifted by an implausible amount ($\\sim$ 16,000 km/s). Thus Lee et al. (2001) propose that the feature in MCG-6-30-15 is due to the L-edge of neutral iron, which matches the observed wavelength more closely than the O{\\sc VII} edge. When interpreted in this way, the depth of the feature in MCG-6-30-15 implies an equivalent hydrogen column of $N_H = 4 \\times 10^{21}$ cm$^{-2}$ assuming solar abundances. This cannot be a column of neutral gas, otherwise the soft X-ray spectrum would be highly absorbed by C{\\sc I} and O{\\sc I} edges, which is not seen. Besides, the presence of numerous narrow absorption lines from ionized species is indicative of a warm absorber. This leads Lee et al. to conclude that the neutral fraction of the absorber, particularly iron, is bound up in dust that is immersed in the ionized medium. Such dusty warm absorber (DWA) models have previously been invoked to explain the apparent discrepancy between the large reddening of certain AGN deduced from their optical-UV spectrum, and the lack of corresponding absorption of the X-ray spectrum (e.g. Reynolds et al. 1997; Walter \\& Fink 1993).  We have developed a DWA model and have fitted it to our RGS data on Mrk~766. The model struggles to reproduce the overall shape of the RGS spectrum. If we nevertheless interpret the sharp discontinuity in flux near 18\\AA\\ as being due to neutral iron, the column density implied is $1.7 \\times 10^{17}$ cm$^{-2}$. This is of the right order to be consistent with the estimated reddening of Mrk~766 (E(B-V)=0.4; Walter \\& Fink 1993) for reasonable abundances. Given the required depth of the Fe I L edge we would, though, expect to see much stronger Fe I L resonance lines than are observed. Furthermore, there is no compelling evidence for K edges of other neutral atoms such as \\ion{O}{1}, \\ion{Mg}{1} \\& \\ion{Si}{1} at the redshift of Mrk~766, and the composition of the putative dust grains is therefore unphysical. The DWA fit also requires a significant column of neutral carbon, forced by the slope of the long-wavelength spectrum in the RGS data. The carbon edge itself is not within the RGS range, but the MOS data extend to these energies and beyond. Even accounting for uncertainties in the MOS response matrix at these energies, we find no evidence for the expected increase in flux below 0.2 keV, which should amount to a factor of two if the edge is present at the level implied by the fits to the RGS data.  In contrast, the relativistic emission line (REL) model is more successful in reproducing the Mrk~766 RGS 0.3-2 keV data. The broad spectral features are consistent with Ly$\\alpha$ emission lines of \\ion{O}{8}, \\ion{N}{7} and \\ion{C}{6} at the rest wavelength of the galaxy, superimposed on an underlying power-law. An extension of this model including a fourth line to represent emission due to iron at 6.7~keV and reflection from a disk, also fits the 0.2-10 keV EPIC MOS data on Mrk~766. The profiles of the low energy emission lines in Mrk~766 are consistent with that measured for the iron emission at 6.7 keV. The REL model requires only a single underlying power law continuum (with the excess above 7 keV being accounted for by reflection), unlike the DWA model where a steep low-energy excess component below about 0.7 keV is also needed.  Thus, adding dust to a warm absorber is not a convincing description of the Mrk~766 X-ray spectrum while the REL model, suggested for Mrk~766 originally by BR01 to explain the earlier, shorter XMM-Newton observation, has become more compelling when confronted with the higher signal to noise data reported here. In many ways Mrk~766 is a more clear-cut example of the phenomenon than MCG-6-30-15 because the spectrum is less affected by the warm absorber.  The underlying physics that might produce such a (REL) spectrum is still controversial. Ballantyne \\& Fabian (2001) fit a model to the ASCA spectrum of MCG-6-30-15 that involves reflection of hard X-rays from ionized layers on the surface of an accretion disk. They optimise the model parameters to best fit the iron K line, but this falls short of reproducing the observed equivalent width of the low-energy lines by more than an order of magnitude. For example, their model predicts an equivalent width of only 6eV for the oxygen line assuming solar abundance. In the Ballantyne \\& Fabian  model, however, the iron K line is a mix of emission from an ionized inner disk, and a sharper neutral component from further out. It is not clear how well the ionisation state of the inner disk region is being constrained in the model.  In more recent work, Ballantyne, Ross \\& Fabian (2002) relax the requirement to fit the iron line and thus widen the scope of parameter space explored. In this case there are regimes in which their model does predict equivalent widths for the low energy lines comparable to those that we measure. Sako et al. (2002) also describe a thermal/ionization structure for the accretion disk surface layers in which Ly-$\\alpha$ recombination lines of appropriate equivalent width could plausibly be produced. The lines will be intrinsically broadened by Compton scattering in the hot disk atmosphere. However, the widths that Ballantyne et al. predict for the blue wing of the lines (a few tens of eV) are comparable to those that we measure for the underlying emission lines in our Mrk~766 data, when the effects of absorption line and edge features in the warm absorber are taken into account. The Ballantyne et al. models predict additional weaker line features due to \\ion{O}{7} and \\ion{N}{6} that are not required by our fits. However, the upper limit that we can set on an \\ion{O}{7} line in our data, measured with respect to the observed flux of \\ion{O}{8}, is only marginally inconsistent with the predictions of the Ballantyne et al. model for an ionizing parameter ($L/nr^2$) of 250 erg cm s$^{-1}$ and Solar oxygen abundance. Sako et al. (2002) in any case use a different ionisation structure for the hot layer, in which these lines are supressed. Sako et al. also argue that iron L-lines above the \\ion{O}{8} K edge at 14.22\\AA, predicted by Ballantyne et al., will be absorbed and the energy re-radiated at \\ion{O}{8} Ly-$\\alpha$.  We have searched for evidence of variability in the low-energy spectral features of Mrk~766 in response to the 1-2 hour timescale flares that are seen in the light curve. There are indications of a systematic change in the high-to-low flux ratio at the $\\pm$15\\% level particularly in the region between about 22\\AA\\ and 24\\AA. This signature is tantalising, since, in the context of the REL model, it may indicate that either the blue wing of the N line or the red wing of the O line are responding to the flares, which would be an important physical constraint on models of the emission region. However the current statistical significance of the effect is low. Confirmation of these changes with higher signal to noise data will be important.    Clearly more work is required before we fully understand the low energy X-ray spectrum of Mrk~766 and similar objects. On the observational side, there are a number of directions for future research. The variations in the profile of the lines, if that is what they are, with source flux are at the limits of detection in the current RGS data. Confirmation in other observations would provide an important physical constraint on potential models. We intend to examine the issue of spectral variability in Mrk~766 in more detail in a future publication, including consideration of the EPIC data. Similarly, a physical model of the warm absorber gas could be valuable in distinguishing which features come from the warm gas and which have other causes. It is encouraging, however, that the detail revealed by the current RGS data, coupled with the high throughput and energy coverage of the EPIC cameras, gives us real discriminating power. We have shown that the REL model is an elegant and relatively simple description of the 0.3-10 keV spectrum of Mrk~766. The physical processes that might produce such lines clearly demand further consideration as a mechanism to account for the spectrum of this and other Seyfert galaxies."
    },
    "0209/astro-ph0209623_arXiv.txt": {
        "abstract": "We present photometry and spectra of the type IIP supernova 1999em in NGC 1637 from several days after the outburst till day 642. The radioactive tail of the recovered bolometric light curve of SN 1999em indicates the amount of the ejected $^{56}$Ni to be $\\approx 0.02~M_{\\odot}$. The H$\\alpha$ and He I 10830~\\AA\\ lines at the nebular epoch show that the distribution of the bulk of $^{56}$Ni can be represented approximately by a sphere of $^{56}$Ni with a velocity of 1500 km s$^{-1}$, which is shifted towards the far hemisphere by about 400 km s$^{-1}$. The fine structure of the H$\\alpha$ at the photospheric epoch reminiscent of the ``Bochum event\" in SN~1987A is analysed in terms of two plausible models: bi-polar $^{56}$Ni jets and non-monotonic behaviour of the H$\\alpha$ optical depth combined with the one-sided $^{56}$Ni ejection. The late time spectra show a dramatic transformation of the [O I] 6300 \\AA\\  line profile between days 465 and 510, which we interpret as an effect of dust condensation during this period. Late time photometry supports the dust formation scenario after day 465. The [O I] line profile suggests that the dust occupies a sphere with velocity $\\approx800$ km s$^{-1}$ and optical depth $\\gg10$. The latter exceeds the optical depth of the dusty zone in SN~1987A by more than 10 times. Use is made of the Expanding Photosphere Method to estimate the distance and the explosion time, $D \\approx 7.83$ Mpc and  $t_{0}\\simeq$ 1999 October 24.5 UT, in accord with observational constraints on the explosion time and with other results of detailed studies of the method (Hamuy et al. 2001; Leonard et al. 2002). The plateau brightness and duration combined with the expansion velocity suggest a presupernova radius of $120-150~R_{\\odot}$, ejecta  mass of $10-11~M_{\\odot}$ and explosion energy of $(0.5-1)\\times10^{51}$ erg. The ejecta mass combined with the neutron star and a conservative assumption about mass loss implies the main sequence progenitor of $M_{\\rm ms}\\approx 12-14~M_{\\odot}$. The derived mass range is in agreement with the upper limit to the mass found using pre-supernova field images by Smartt et al. (2001). From the [O I] 6300,6364 \\AA\\ doublet luminosity we infer the oxygen mass to be a factor four lower than in SN~1987A which is consistent with the estimated SN~1999em progenitor mass according to nucleosynthesis and stellar evolution theory. We note a ``second-plateau\" behaviour of the light curve after the main plateau at the beginning of the radioactive tail. This feature seems to be common to SNe~IIP with low $^{56}$Ni mass. ",
        "introduction": "Type II (and Ib/Ic) SNe are generally associated with regions of recent star formation in spiral galaxies, suggesting that they represent the final episode in the life of massive stars ($M>8$$M_{\\odot}$) which explode owing to core collapse \\shortcite{fil2}. The study of core collapse events is important for understanding the range of progenitor masses which produce them, the consequent nucleosynthesis for its effect on galactic chemical evolution and the explosion energy which remains an ill-determined quantity for the vast majority of SNe, and which is also relevant for gas dynamical processes and ejection of material from galaxies. Detailed photometric and spectroscopic observations of SNe~II on a long time scale are still rare, especially for SNe~IIP (plateau). Meanwhile from the recent experience with SN~1987A we know, how valuable can be extended sets of photometric and spectroscopic data for understanding what really happens to the massive ($M>8M_{\\odot}$) star when its central iron core collapses. In this way one obtained information on the amount of ejected $^{56}$Ni and oxygen, and their mixing and clumpiness, possible asymmetry of $^{56}$Ni ejecta, ``Bochum event\", dust formation. The case of SN~1987A also provided us with the possibility for testing and revising the theory of stellar evolution, SN hydrodynamical models and models of spectra formation.  Supernova 1999em, discovered by the Lick observatory Supernova Search on Oct 29 UT at an unfiltered magnitude $\\sim 13.5$ mag in the nearby galaxy NGC 1637 ($D=7.8$ Mpc; IAUC 7294), has become another well observed type IIP event. Detected very soon after the explosion and followed for more that 600 days this event gives another boost to studies of SNe~II. SN 1999em is the first type IIP detected at both radio and X-ray wavelengths at early time \\shortcite{pool}. However it is the least radio luminous and among the least X-ray luminous SNe ($\\it Chandra$ X-ray and radio NRAO observations; IAUC 7318, 7336 ). This early and weak radio emission is consistent with a low mass loss shortly before the explosion for SNe IIP \\shortcite{baron}. The analysis of the spectra at the early photospheric epoch already emphasized the problem of He abundance in the hydrogen envelope of SN~IIP progenitors (Baron et al. 2000) and permitted one to check and upgrade the method of the expanding photosphere (EPM) for the distance determination (Hamuy et al. 2001; Leonard et al. 2002). The spectropolarimetry on the other hand has been studied at 5 different epochs (till day 163 after discovery; Leonard et al. 2000). The authors estimated the broadband polarization at day 7 to be $\\sim 0.2$ $\\%$, rising to $\\sim 0.5$ $\\%$ on day 161. The low polarization found for SN 1999em and hence the small implied asphericity especially at early photospheric phase is another encouraging reason for the validity of the EPM for this class of object (SNe IIP). The progenitor nature of SN 1999em was discussed by Smartt et al. (2002), who used pre-explosion $CFHT$ images to derive bolometric luminosity limits and thus constraints on the mass of the progenitor star of SN 1999em. They concluded that the main-sequence mass should be $< 12 \\pm1 M_{\\odot}$.  Here we present photometry and spectroscopy of SN~1999em from the phase of several days after the explosion till day 642. In what follows we describe the photometry and spectral evolution and provide estimates of the $^{56}$Ni and oxygen mass in SN~1999em; we will analyse fine structures of the line profiles at the photospheric and nebular epoch in an attempt to determine the effects of the $^{56}$Ni mixing and asymmetry. The last spectra are used as diagnostics of possible dust formation in the ejecta. The photometry and spectroscopy at the plateau phase will be used to recover global parameters of SN (presupernova radius, mass and energy of SN). We then discuss the main sequence mass of the progenitor and some important correlations. \\begin{figure} \\centerline{\\psfig{file=sn99em2.ps,width=9truecm,height=9truecm}} \\caption{ SN 1999em in NGC 1637. Image taken at ESO on Mar 27. 2000, about 150 days after discovery (in the R band). The SN position is indicated, also shown are the reference sequence stars. North is up, East is to the left.} \\end{figure} ",
        "conclusions": "We have presented  photometric and spectroscopic data for SN 1999em, from $\\sim9~\\rm d$ until $\\sim642~\\rm d$ after the explosion. The shape of the light curve ($t_{\\rm p}\\sim$ 80 days) as well as spectral features show that it is a type IIP supernova. The problem of reddening has been discussed and we conclude that an optimum choice is close to that determined by Hamuy et al. (2001). The analysis of late phase photometry, up to $\\sim510~\\rm d$, shows that the exponential tail decay rate is close to the one of the radioactive decay $^{56}$Co to $^{56}$Fe, indicating that this is the main source of energy powering the light curve.  A photometric comparison of SN 1999em with SN 1987A, especially in the later phases, provides constraints on the radioactive $^{56}$Ni mass. We have constructed the ``UBVRI'' bolometric light curve, and comparing it with that of SN 1987A we obtain an estimate of the amount of  $^{56}$Ni produced by the explosion of 0.02 $M_{\\odot}$, a smaller value than that derived for typical type IIP SNe such as SN 1969L and SN 1988A which have $M(\\rm ^{56}Ni)\\sim 0.07~\\it M_{\\odot}$.  We have noticed some flattening in the light curves, just after the steep decline from the plateau phase, and clearer for the blue bands. This behaviour is also seen in the light curves of two other objects, namely the peculiar SN IIP 1997D and the SN IIP 1991G. All three events have some common features, being type IIP and all having a lower ejected $^{56}$Ni mass: a similar amount for both SN 1999em and SN 1991G,($\\sim$ 0.02 $M_{\\odot}$) while SN 1997D ejected an even lower mass ($\\sim$ 0.002 $M_{\\odot}$). In addition, the duration of this $``$second plateau'' on the tail seems greater for SN 1997D than for SNe 1991G and 1999em. The decline rates and the observed duration of the flattening period are reported in Table 4. These measurements suggest that the $``$second plateau'' feature is a common feature for this low $^{56}$Ni mass type IIP supernovae, and that its duration is correlated with the amount of ejected $^{56}$Ni. Further support for this possible correlation comes from the case of SN IIP 1999eu (Pastorello et al. in preparation) which shows a very clear second plateau feature of duration $\\sim 200~\\rm d$ (clear in V and B bands). Moreover it seems that this SN has a very low ejected $^{56}$Ni mass similar to SN 1997D. Note that the prototype type IIP supernovae SN 1969L and 1988A do not show clear evidence of this behaviour. On the other hand SN 1969L and SN 1988A are known to produce an amount of $^{56}$Ni similar to SN 1987A $\\sim$ 0.07 $M_{\\odot}$. Improved statistical samples and better sampled light curves are required in order to confirm or rule out this behaviour. Radiation diffusion effects, still important at the beginning of the radioactive tail, are a possible cause of this behaviour.  SN 1999em provided a test of the validity of the EPM since we have observational constraints on the explosion time and distance. In fact analysing the spectrophotometry, we derive an explosion time consistent with these constraints and in good agreement with what was found from more detailed broad band photometric studies of SN 1999em (Hamuy et al. 2001; Leonard et al. 2002).  We analysed the phenomenon of the fine structure of H$\\alpha$ at the photospheric epoch, which was reminiscent of the ``Bochum event\" in SN~1987A. Two possible explanations, a bi-polar jet proposed by Lucy (1988) for SN~1987A and underexcitation of hydrogen combined with $^{56}$Ni asymmetry, are discussed. This analysis does not permit one to discriminate between those models. Yet the one-sided $^{56}$Ni ejection seems to find support in the small red shift of the He I 10830 \\AA\\ profile and the larger red shift of the H$\\alpha$ line profile at the nebular epoch. We note that the He I 10830 \\AA\\ line should be more sensitive to non-thermal excitation resulting from $\\gamma$-ray deposition. These lines indicate that the $^{56}$Ni distribution could be imagined to be a filled sphere with a velocity of $\\sim 1500$ km s$^{-1}$ shifted to the far hemisphere by 400 km s$^{-1}$. A somewhat surprising coincidence is that in SN~1987A the $^{56}$Ni distribution also shows one-sidedness with a shift to the far hemisphere.  Analysing the [O I] 6300,6364 \\AA\\ line profile evolution we found a rapid change between days 465 and 510, which we interpret as an effect of the dust formation during this interval. Other support for dust formation comes from the deficit seen in optical radiation measured by late time photometry. It is the second SN~IIP (after SN~1987A) where convincing evidence of dust formation exists. In SN~1987A and SN~1999em we detected the rapid ($\\Delta t/t\\sim 0.1$) transformation of profiles during the nebular epoch. The dust phenomenon in SN~1999em has some distinctive characteristics compared to SN~1987A. The dust condensation happened earlier (between days 465 and 510) than in SN~1987A (after day 526), which is probably explained, by the lower $^{56}$Ni mass and, accordingly, lower temperature. The dust resides in the core with a velocity of $\\approx 800$ km s$^{-1}$, much lower than in SN~1987A. This greater confinement of dust in SN 1999em possibly results from the lower velocity creating a more confined metal-rich region where the condition for dust formation prevails. Another remarkable difference is the very large optical depth of the dusty zone ($\\tau\\gg10$) compared to SN~1987A ($\\tau \\sim 0.5$, Lucy et al. 1991). We interpret this difference as an indication that in SN~1999em the dust is distributed more homogeneously (or the number of the opaque dusty clumps is notably greater) than in SN~1987A.  These facts show clearly the importance of studying more samples of SNe IIP (SNe II in general), because of the diversity they provide in manifesting the same event (i.e. dust condensation) and thus the opportunity of understanding the physics behind such events.  We used relations by Litvinova \\& Nadyozhin (1985) and Popov (1993) to find SN parameters from observational characteristics. The estimated mass of the progenitor $M_{\\rm ms}\\approx 12-14~M_{\\odot}$ and presupernova radius ($120-150~R_{\\odot}$) are just consistent with the failure to detect a progenitor star with imaging of the pre-SN field. Moreover, our derived progenitor mass agrees with our finding that the oxygen mass in SN~1999em is about four times smaller compared to SN~1987A.  There is growing observational evidence of a significant decrease in the mass of Fe produced as the progenitor mass for type II SNe decreases when we include derivations for other objects. It is therefore encouraging to note that this behaviour is what is required to explain abundance patterns such as [O/Fe] and [Mg/Fe] in metal-poor halo stars modeled by Argast et al. \\shortcite{argast}.  The correlation of progenitor mass of SN~IIP and core collapse SN in general with other parameters is vital for testing explosion models and the theory of stellar evolution. In Table 5, we present parameters of SN~1999em along with other SNe~II. Note that the parameters reported in the table, except for SN 1999em, are not obtained directly by using the analytical models of Litvinova $\\&$ Nadozhin (1985) and Popov (1993). They are however the most reliably determined through modelling of observations.  Table 5 demonstrates that for type II SNe there begins to emerge a monotonic relation between progenitor mass on the one hand and both explosion energy and $^{56}$Ni mass on the other. Although the uncertainties in these parameters remain, of necessity, large, it seems that in the progenitor mass range $10 - 13$ $M_{\\odot}$ there is also a steep decrease in both energy and $^{56}$Ni mass. This suggests that these latter 2 parameters are not independent but physically linked. Results for other core collapse objects (type Ib,c) with higher progenitor masses tend to support this correlation even if the current scatter is unavoidably large. Future observations supported by modelling will surely elucidate this conclusion. It is clear from our analysis that high S/N spectra of SNe at late phases are an invaluable tool for understanding type II supernovae.  The fact that SN 1999em probably resulted from the explosion of a $G_{0}-G_{5}$ supergiant also indicates, as did SN 1987A, that the evolutionary stage at which massive stars can explode is not yet well delineated."
    },
    "0209/astro-ph0209415_arXiv.txt": {
        "abstract": "s{ Understanding the nature of Extremely~Red~Objects [EROs; ($I-K$)$\\ge$4] is one of the most challenging issues in observational cosmology. Here we report on the \\hbox{X-ray} constraints provided by the 2~Ms \\chandra\\ Deep Field-North Survey \\hbox{(CDF-N)}. X-ray emission has been detected from 11 out of 36 EROs ($\\approx$~30\\%). Five of these have hard X-ray emission and appear to be obscured AGNs, while non-AGN emission (star formation or normal elliptical galaxy emission) is likely to be the dominant source of X-rays from the soft \\hbox{X-ray} sources detected at the faintest X-ray flux levels. } ",
        "introduction": "EROs were serendipitously discovered more than a decade ago{\\,}\\cite{er88} and have optical/near-IR colors consistent with those expected for passively evolving elliptical galaxies and dust-enshrouded star-forming galaxies{\\,}\\cite{pm00} at $z$$\\simgt$1. Evidence for elliptical galaxies within the ERO population has been found from \\nolinebreak optical spectroscopic and morphological studies,{\\,}\\cite{c02,m00} while submm-IR observations \\nolinebreak have provided clear examples of dust-enshrouded galaxies.{\\,}\\cite{c98,a01} However, until \\nolinebreak recently the fraction of the ERO population hosting AGNs was unknown. Arguably, the clearest evidence for AGN activity is found from X-ray observations.{\\,}\\cite{h01} \\nolinebreak Here we \\nolinebreak provide an update on the 1~Ms CDF-N results{\\,}\\cite{a02} using the 2~Ms \\nolinebreak \\chandra\\ exposure, yielding the tightest X-ray constraints on the ERO population \\nolinebreak to \\nolinebreak date. ",
        "conclusions": ""
    },
    "0209/astro-ph0209553_arXiv.txt": {
        "abstract": "{ We suggest that the majority of the \"young\", so--called \"second parameter\" globular clusters (SPGCs) have originated in the outer Galactic halo due to a process other than a tidal disruption of the dwarf spheroidal (dSph) galaxies. Basic observational evidence regarding both the dSphs and the SPGCs, coupled with the latest data about a rather large relative number of such clusters among globulars in M33 and their low portion in M31, seems to be consistent with the suspected process. It might have taken place within the system of the most massive galaxies of the Local Group (LG) at the earliest stages of their formation and evolution. We argue that the origin and basic characteristics of the SPGCs can naturally be explained as a result of mass outflow from M31, during and due to formation of its Pop. II stars, and subsequent accretion of gas onto its massive companions, the Galaxy and M33. An amount of the gas accreted onto the Milky Way is expected to have been quite enough for the formation in the outer Galactic halo not only of the clusters under consideration but also a number of those dSph galaxies which are as young as the SPGCs. A less significant, but notable mass transfer from the starbursting protoGalaxy to the massive members of the LG might have occurred, too. ",
        "introduction": "Color--magnitude diagram (CMD) studies of the Galactic globular clusters (GCs) accumulated in 1960-ies have convincingly shown that metallicity is not the only parameter governing the clusters' horizontal branch (HB) morphology and that some other parameter(s) affect the distribution of stars on the HB. In other words, a \"second parameter\" effect takes place: at a given  metallicity, GCs with the second parameter phenomenon exhibit redder HB morphology. Thanks to numerous CMD studies, with limiting magnitude below the main sequence turnoff, of GCs in the Milky Way and in its nearest satellites, age has been revealed to play an important role in the second parameter effect. However, different investigators make contradictory conclusions regarding whether or not it is mainly responsible for the effect (for comparison, see Stetson et al. \\cite{stetson} and Sarajedini et al. \\cite{saraj97}; or VandenBerg \\cite{van}).  We stress the point that the second parameter problem itself is beyond the scope of the present paper. However, since the Galactic GCs under study have historically been called SPGCs, it is this term that is used hereafter to denote GCs younger than primeval ones, but with approximately the same mean metallicity, close to $[Fe/H] \\sim -1.6$.   It is now well recognized (for instance, Zinn \\cite{zinn}; van den Bergh \\cite{vdB93}; Lee et al. \\cite{lee}) that GCs in the outer halo of the Milky Way form two subpopulations, \"old\" and \"young\". The latter consists of the SPGCs displaying predominantly red HB morphology. Presently, there is little doubt about the younger ages ($\\sim2$ Gyr) of the SPGCs as compared to the old GCs in the Galactic halo (e.g., Lee et al. \\cite{lee}). In addition to such age characteristics, their predominant location in the halo at Galactocetric distance ${R_{GC}}>8$ kpc poses the fundamental problem of the origin of these clusters. It is a long--standing, poorly understood problem that arose more than three decades ago. Revealing the cause of these GCs in the outer Galactic halo would be beneficial in understanding their formation and the early history of the Milky Way, and other related questions.  By proceeding from the tentative inference that age is responsible for the second parameter, Searle \\& Zinn (\\cite{sz}), in their scenario of halo formation, hypothesized that the origin of SPGCs was in gaseous condensations, \"transient protogalactic fragments that continued to fall into dynamical equilibrium with the Galaxy for some time after the collapse of its central regions had been completed\". In this scenario the authors also considered the possibility of merging of low--mass galaxies with the Milky Way. Later, owning to development of the cold dark matter theories of galaxy formation, some investigators tentatively identified these hypothetical transient fragments with \"dark halos\" which, in turn, were found in the dSph galaxies, real objects with both the appropriate masses of visible matter and signs of a rather high dark matter content (see, among others, the review of various topics related to dSphs by Gallagher \\& Wyse \\cite{galwys}; review by Mateo \\cite{mateo} ). Moreover, a number of  important observations -- in particular, the discovery by Ibata et al. (\\cite{ibata}) of the Sagittarius dSph that shows obvious signs of tidal disruption; findings about the second parameter effect displayed by the old stellar populations in dSphs and by GCs belonging to them (e.g., among others, Buonanno et al. \\cite{buonan85}; Suntzeff et al. \\cite{suntz};  Demers \\& Irwin \\cite{demirw}  ); a highly ordered spatial distribution of the Galactic dSph galaxies, as noted by Kunkel \\& Demers (\\cite{kundem}) and Lynden-Bell (\\cite{lynden}), along with a close resemblance between three--dimentional distribution of SPGCs having the reddest HB morphology and some of the dSphs (Majewski \\cite{majew}) -- are evidence suggesting that the origin of the Galactic SPGCs is a consequence of tidal disruption/stripping (TD/S), during a Hubble time, of the formerly more numerous dSph--like systems or larger Galactic companion(s). Thus, the role of such a mechanism is at present widely believed to be crucial, and its reliability with regard to the formation of the Galactic halo is currently being investigated (for instance, Morrison et al. \\cite{morris}; Harding et al. \\cite{hard}).  Explaining the population of Galactic SPGCs by TD/S of the respective ancestral galaxies, dSph or some other satellites to the Milky Way appears to be a natural and well grounded approach to the problem. However, {\\it even if it was an absolute fact it would simply replace the problem by other one(s) related to the origin of the ancestors themselves}. Indeed, the Galactic SPGCs are, on the one hand, younger than the old halo globulars, but, on the other hand, they should have been among the oldest constituents of their hypothetical parent (dSph) systems. Therefore, we have to conclude that these tidally dissolved dSphs are younger (by $\\sim2$ Gyr or so) than the oldest objects populating the Galactic spheroid. {\\it Then, we need to explain the mysterious emergence of the \"young\" satellite galaxies in the outer Galactic halo, in about two Gyr after the beginning of its formation, as well as their relatively rapid subsequent dissolution by the Galactic tidal forces.} Moreover, different facts and their detailed analysis (see Section 2) can militate against such an attractive \"destructive\" mechanism as being the major one responsible for the Galactic SPGC system formation.  Instead of (or in addition to) this we have recently suggested (Kravtsov \\cite{krav}), and here we analyze in detail, a \"creative\" process which would perhaps be able to explain the origin of populations of both SPGCs and dSphs, belonging not only to the Galaxy but also to two other massive galaxies of the LG. Specifically, M31 includes six dSph systems (Grebel \\cite{grebela}), while M33 has no dwarf companions, but according to data of Sarajedini et al. (\\cite{saraj98}) it has appreciable population of SPGCs.  The role of the dSph--consuming process in the formation of the Galactic SPGCs population and populations of sush GCs in M33 and M31 is analyzed in the next section, taking into consideration diverse observational data obtained to date. In Section 3, we summarize the basic characteristics of both the population of Galactic SPGCs and the dSph galaxies, which may be consistent with and explained by the processes described in Section 4. ",
        "conclusions": "The present paper discusses long--standing problem of the origin of the Galactic SPGCs. We show that there arise questions about the currently popular and conventional view on the origin of a population of these globulars as a result of TD/S of dSph galaxies, especially taking into account recent relevant observational data including those on the populations of such globulars in the two other most massive galaxies of the LG, M31 and M33. On the one hand, answering these questions requires a number of a priori assumptions, some of which seem to be rather artificial and contradictory, about hypothetical events and objects. On the other hand, we argue that even if this view is correct, the problem is merely replaced by other one(s) related to the origin of dSphs themselves. In order to interpret and reconcile the various observational data on SPGCs (and probably dSphs) belonging to the three spirals of the LG we propose here other scenario. It is based on the realistic suggestion about natural consequences of mass loss from M31 in its early history, namely, subsequent partial accretion of the gaseous material onto the early Galaxy and M33 should have contributed to the formation of their (outer) halos.  In addition, we draw attention to the possible interaction between massive galaxies at the early stage of their formation and evolution in pairs, groups, and perhaps in small clusters. In contrast to the relatively well--known and recognized consequences of a tidal interaction between galaxies, the suspected role of the interaction we deal with seems to be presently missing or underestimated. We suggest that if some starbursting massive protogalaxy, like M31 or our Galaxy, really loses a substantial amount of its gas, then the partially enriched gas expanding out and escaping from the protogalaxy not only contributes to the intergalactic medium but also should partially be trasferred to massive companion(s) of the protogalaxy. Because of the sizable amount of the gaseous material (probably, around a few per cent of mass lost by the starbursting massive protogalaxy) trasferred to and accreted onto the massive companions, they may experience additional large--scale starbursts leading to the formation, out of this material, of supplementary objects (stars, star clusters, and even dwarf galaxies) with particular appropriate characteristics. Therefore, strictly speaking, such galaxies do not form and evolve as completely isolated entities, and they may contain populations of SPGCs (and probably dSphs) as tracers of the interaction under study.  We suppose that just this kind of interaction occurred between the most massive galaxies of the LG and lasted approximately 2 -- 3 Gyr after the beginning epoch of their formation, and it could primarily be responsible for the origin of both their SPGC populations and some of the dSph galaxies, as well as for the basic observables of these objects."
    },
    "0209/astro-ph0209079_arXiv.txt": {
        "abstract": "We have completed a systematic survey for disks around young brown dwarfs and very low mass stars.  By choosing a well-defined sample and by obtaining sensitive thermal IR observations, we can make an unbiased measurement of the disk fraction around such low mass objects.  We find that $\\approx$75\\% of our sample show intrinsic IR excesses, indicative of circum(sub)stellar disks.  We discuss the physical properties of these disks and their relation to the much better studied disks around solar-mass stars.  The high incidence of disks around substellar objects also raises the possibility of planetary systems around brown dwarfs. ",
        "introduction": "The existence of circumstellar disks and their role in planet formation are well established for young solar-type stars.  However, little is known about disks around young substellar objects.  Such circum(sub)stellar disks might provide a laboratory for studying the physical processes of disks over a wide range of (central object and disk) mass.  In addition, the presence of disks around young substellar objects may be an important clue to the origin of brown dwarfs.  We have recently completed a large thermal IR ($L^\\prime$-band; 3.8~\\micron) survey to study the frequency and properties of disks around young brown dwarfs and very low mass stars (Liu et~al.\\ 2002; see also Liu~2002).  Our sample comprises young ($\\sim$1--3~Myr) objects in nearby star-forming regions which have been spectroscopically classified to be very cool, corresponding to masses of $\\sim$15 to $\\sim$100~$M_{\\rm Jup}$ based on current models.  As described in our paper, the objects constitute a well-defined sample and are largely free of selection biases.  A priori, brown dwarf disks are expected to be harder to detect than disks around stars because of lower contrast.  Substellar objects are less luminous and have shallower gravitational potentials. Hence, their disks should be relatively cool and may have negligible excesses in the commonly used $JHK$ (1.1$-$2.4~\\micron) bands.  In fact, we find that thermal IR data are {\\em required} to detect most disks around young brown dwarfs. Our survey is also sensitive enough to detect brown dwarf photospheres --- hence the absence of a disk can be discerned, and the disk frequency of young brown dwarfs can be measured for the first time. ",
        "conclusions": ""
    },
    "0209/astro-ph0209590_arXiv.txt": {
        "abstract": "In this pedagogical lecture, I introduce some of the basic terminology and description of fluctuating fields as they occur on cosmology. I define various statistical, cosmological and sample homogeneity and explain what is meant by the fair sample hypothesis and cosmic variance. I illustrate these concepts using the simplest second-order statistics, i.e. the two--point correlation function and its Fourier transform the power-spectrum. I then give a brief overview of the properties of information relating to the properties of the phases of the Fourier modes of cosmological fluctuations which is not contained in these simpler statistics. Specifically, I explain how phase information of a particular form (called quadratic phase coupling) is encoded in the three--point correlation function (or, equivalently, the bispectrum). ",
        "introduction": "In most popular versions of the gravitational instability model for the origin of cosmic structure, particularly those involving cosmic inflation (Guth 1981; Guth \\& Pi 1982), the initial fluctuations that seeded the structure formation process form a Gaussian random field (Adler 1981; Bardeen et al. 1986). Gaussian random fields are the simplest fully-defined stochastic processes, which makes analysis of them relatively straightforward. Robust and powerful statistical descriptors can be constructed that have a firm mathematical underpinning and are relatively simple to implement. Second-order statistics such as the ubiquitous power-spectrum (e.g. Peacock \\& Dodds 1996) furnish a complete description of Gaussian fields. They have consequently yielded invaluable insights into the behaviour of large-scale structure in the latest generation of redshift surveys, such as the 2dFGRS (Percival et al. 2001). Important though these methods undoubtedly are, the era of precision cosmology we are now entering requires more thought to be given to methods for both detecting and exploiting departures from Gaussian behaviour.  The pressing need for statistics appropriate to the analysis of non-linear stochastic processes also suggests a need to revisit some of the fundamental properties cosmologists usually assume when studying samples of the Universe. Gaussian random fields have many useful properties. It is straightforward to impose constraints that result in statistically homogeneous fields, for example. Perhaps more relevantly one can understand the conditions under which averages over a single spatial domain are well-defined, the constraint of sample-homogeneity. The conditions under which such fields can be ergodic are also well established. It is known that smoothing Gaussian fields preserves Gaussianity, and so on. These properties are all somewhat related, but not identical. Indeed,  looking at the corresponding properties of non-linear fields turns up some interesting results and delivers warnings to be careful. Exploring these properties is the first aim of this lecture.  Even if the primordial density fluctuations were indeed Gaussian, the later stages of gravitational clustering must induce some form of non-linearity. One particular way of looking at this issue is to study the behaviour of Fourier modes of the cosmological density field. If the hypothesis of primordial Gaussianity is correct then these modes began with random spatial phases. In the early stages of evolution, the plane-wave components of the density evolve independently like linear waves on the surface of deep water. As the structures grow in mass, they interact with other in non-linear ways, more like waves breaking in shallow water. These mode-mode interactions lead to the generation of coupled phases. While the Fourier phases of a Gaussian field contain no information (they are random), non-linearity generates non-random phases that contain much information about the spatial pattern of the fluctuations. Although the  significance of phase information in cosmology is still not fully understood, there have been a number of attempts to gain quantitative insight into the behaviour of phases in gravitational systems. Ryden \\& Gramann (1991), Soda \\& Suto (1992) and Jain \\& Bertschinger (1998) concentrated on the evolution of phase shifts for individual modes using perturbation theory and numerical simulations. An alternative approach was adopted by Scherrer, Melott \\& Shandarin (1991), who developed a practical method for measuring the phase coupling in random fields that could be applied to real data. Most recently Chiang \\& Coles (2000), Coles \\& Chiang (2000), Chiang (2001) and Chiang, Naselsky \\& Coles (2002) have explored the evolution of phase information in some detail.   Despite this recent progress, there is still no clear understanding of how the behaviour of the Fourier phases manifests itself in more orthodox statistical descriptors.  In particular there is much interest in the usefulness of the simplest possible generalisation of the (second-order) power-spectrum, i.e. the (third-order) bispectrum (Peebles 1980; Scoccimarro et al. 1998; Scoccimarro, Couchman \\& Frieman 1999; Verde et al. 2000; Verde et al. 2001; Verde et al. 2002). Since the bispectrum is identically zero for a Gaussian random field, it is generally accepted that the bispectrum encodes some form of phase information but it has never been elucidated exactly what form of correlation it measures. Further possible generalisations of the bispectrum are usually called polyspectra; they include the (fourth-order) trispectrum (Verde \\& Heavens 2001) or a related but simpler statistic called the second-spectrum (Stirling \\& Peacock 1996). Exploring the connection between polyspectra and non-linearly induced phase association is the second aim of this lecture.    The plan is as follows. In the following section I introduce some fundamental concepts underlying statistical cosmology, more-or-less from first principles. I do this in order to allow the reader to see explicitly what assumptions underlie standard statistical practise. In Section 3 I look at some of the contexts in which quadratic non-linearity may arise, either primordially or during the non-linear growth of structure from Gaussian fields. In Section 4 I revisit some of the basic properties used in Section 2 from the viewpoint of  a particularly simple form of non-linearity, known as quadratic non-linearity, and show how some basic implicit assumptions may be violated. I then, in Section 5, explore how phase correlations arise in quadratic fields and relate these to higher-order statistics of quadratic fields. ",
        "conclusions": "In this lecture I addressed two main issues, using the quadratic model as an illustrative example. First I showed explicitly how this non-Gaussian model has properties that contradict standard folklore based on the assumption of Gaussian fluctuations. We used this model to distinguish carefully between various inter-related concepts such as sample homogeneity, statistical homogeneity, asymptotic independence, ergodicity, and so on. I showed the conditions under which each of these is relevant and deployed the quadratic model for particular examples in which they are violated. I then used the quadratic model to show how phase association arises in non-linear processes which has exactly the correct form to generate non-zero bispectra and three--point covariance functions. The magnitude of these statistical descriptors is of course related to the magnitude of the Fourier modes, but the factor that determines whether they are zero or non-zero is the arrangement of the phases of these modes.  The connection between polyspectra and phase information is an important one and it opens up many lines of future research, such as how phase correlations relate to redshift distortion and bias. Also, I assumed throughout this study that we could straightforwardly take averages over a large spatial domain to be equal to ensemble averages. Using small volumes of course leads to sampling uncertainties which are  quite straightforward to deal with in the case of the power-spectra but more problematic for higher-order spectra like the bispectrum. Understanding the fluctuations about ensemble averages in terms of phases could also lead to important insights."
    },
    "0209/astro-ph0209073_arXiv.txt": {
        "abstract": "Unsupervised pattern recognition algorithms support the existence of three gamma-ray burst classes; Class I (long, large fluence bursts of intermediate spectral hardness), Class II (short, small fluence, hard bursts), and Class III (soft bursts of intermediate durations and fluences). The algorithms surprisingly assign larger membership to Class III than to either of the other two classes. A known systematic bias has been previously used to explain the existence of Class III in terms of Class I; this bias allows the fluences and durations of some bursts to be underestimated (Hakkila et al., ApJ 538, 165, 2000). We show that this bias primarily affects only the longest bursts and cannot explain the bulk of the Class III properties. We resolve the question of Class III existence by demonstrating how samples obtained using standard trigger mechanisms fail to preserve the duration characteristics of small peak flux bursts. {\\em Sample incompleteness is thus primarily responsible for the existence of Class III.} In order to avoid this incompleteness, we show how a new dual timescale peak flux can be defined in terms of peak flux and fluence. The dual timescale peak flux preserves the duration distribution of faint bursts and correlates better with spectral hardness (and presumably redshift) than either peak flux or fluence. The techniques presented here are generic and have applicability to the studies of other transient events. The results also indicate that pattern recognition algorithms are sensitive to sample completeness; this can influence the study of large astronomical databases such as those found in a Virtual Observatory. ",
        "introduction": "In recent years, data mining algorithms have been used to aid the process of scientific classification. {\\em Data mining} is the extraction of potentially useful information from data using machine learning, statistical, and visualization techniques. Pattern recognition algorithms (or classifiers) are data mining tools that search for clusters in complex, multi-dimensional spaces of {\\em attributes} (observed and/or measured properties). These algorithms typically operate in one of two modes: supervised (in which the classifier is trained with known classification instances) and unsupervised (in which classification occurs without training examples).  Algorithms are designed to identify data patterns such as clustering and/or correlations, but their limitations must also be understood: it is up to the scientist to interpret physical mechanisms responsible for producing identified clusters.  Clusters found by classifiers can represent source populations; this happens when the class properties are produced by physical mechanisms pertaining to the sources. Clusters can also result from the way in which source properties are measured; sampling biases, systemic instrumentation errors, and correlated properties can all force data to cluster and thus give the appearance of class structure when there is none.  Data mining algorithms can be applied to gamma-ray burst classification. Two gamma-ray burst classes have been recognized for years \\citep{maz81,nor84,kle92,hur92,kou93} on the basis of duration and spectral hardness. Class I (Long) bursts are longer, spectrally softer, and have larger fluences than Class II (Short) bursts. Recent classification schemes have used data collected by BATSE (the Burst And Transient Source Experiment on NASA's Compton Gamma-Ray Observatory; CGRO) \\citep{mee92} because this large database (2704 bursts observed between 1991 and 2000) was collected by a single instrument with known instrumental characteristics. Three attributes of BATSE gamma-ray burst data have been identified as being significant (using techniques such as principal component analysis) in delineating gamma-ray burst classes \\citep{muk98,bag98,hak00}: duration T90 (the time interval during which 90\\% of a burst's emission is detected), fluence S (time integrated flux in the 50 to 300 keV spectral range), and spectral hardness HR321 (the 50 to 300 keV fluence divided by the 25 to 50 keV fluence). Logarithmic measures of these values are typically used because classes are more clearly delineated when attributes are defined logarithmically. Historically, bursts with durations $T90 < 2$ seconds have been typically considered to belong to Class II.  Data mining techniques allow a third gamma-ray burst class to be identified in BATSE data. Three classes are preferably recovered instead of two by both statistical clustering techniques \\citep{muk98,hor98} and unsupervised pattern recognition algorithms \\citep{roi00,bal01,raj02}. The third class forms at the boundary between Class I and Class II, and primarily contains the softest and smallest fluence bursts from Class I. Since Class II appears to be relatively unchanged by the detection of the third class, the three classes are called Class II (short, small fluence, hard bursts), Class I (long, large fluence bursts of intermediate hardness), and Class III (intermediate duration, intermediate fluence, soft bursts; also referred to as Intermediate bursts).  The boundaries between classes are fuzzy, as some bursts are not easily assigned to a specific class. Different data mining algorithms do not necessarily assign individual bursts to the same classes because each classifier operates under different assumptions concerning correlations between data attributes and how these relate to clustering criteria. Some classifiers are designed to work with nominal data while others are not; some employ Bayesian while others employ frequentist statistics; some assume {\\em a priori} distributions of attribute values while others do not. The results of any classifier can change if the size and makeup of the data set is altered. Data errors can influence the results since few classifiers currently include measurement error information in their analyses. However, irreproducibility is not necessarily a fault of machine learning methodology. Each classifier provides different insights into the way the data are structured.  For any given data set, there is a good possibility that some critical experiment or observation has not been performed, or that some key measurements have yet to be made, or that the relative importance of some attribute has been underestimated or overestimated. {\\em There is no correct way of classifying a dataset because the usefulness of the classification depends on the insights gained from it by the user.}  In a previous application of supervised classification \\citep{hak00} to gamma-ray burst data we hypothesized that Class III does not necessarily represent a separate source population. Instead, instrumental and sampling biases have been proposed as a way in which some Class I bursts can take on Class III characteristics. Due to a known correlation between hardness and intensity \\citep{pac92,mit92,nem94,att94,dez97,qin01}, small fluence Class I bursts are typically softer than bright Class I bursts; this is supported by principal component analysis \\citep{bag98}. Since the correlation results from a shift to smaller average peak spectral energy $\\langle E_p \\rangle$ at lower peak flux but not from changes in the average low-energy spectral index $\\langle \\alpha \\rangle$ or the average high-energy spectral index $\\langle \\beta \\rangle$ \\citep{mal95,hak00,pac02}, this correlation has been attributed to the softer bursts being generally at larger cosmological redshift. (This conclusion may not necessarily be correct because a broad range of gamma-ray burst luminosities is suggested from redshifts of gamma-ray burst afterglows \\cite{van00}; however, it should be noted that only a small afterglow sample is available.) Additionally, fluences and durations of some Class I bursts can systematically be underestimated \\citep{kos96,bon97,hak00}; we refer to this as the {\\em fluence duration bias} \\citep{hak00}. Simply put, fluences and durations of some Class I bursts (particularly those with the smallest peak fluxes) can be underestimated due to the unrecognizability of low signal-to-noise emission; combined with their spectral softness, this gives them characteristics consistent with Class III.  Unfortunately, the fluence duration bias has been difficult to quantify. The amounts by which the fluence and duration of an individual burst are affected depend on the fitted background rates and estimated burst durations at all energies; to remove the background properly assumes {\\em a priori} knowledge of the burst's intrinsic temporal and spectral structure. Such {\\em a priori} knowledge can only be acquired in the absence of background, and gamma-ray burst observations are inherently noisy. Very high signal-to-noise estimates of a burst's temporal and spectral structure can only be obtained for a small number of the bursts with the largest fluences. These well-measured quantities are not entirely intrinsic; it appears that even the brightest bursts require systematic correction because they are at large redshift ($z \\approx 1$).  Our objective is to determine whether or not the fluence duration bias can account for the number of bursts with Class III characteristics. In order to do this, we determine the total number of bursts that exhibit Class III characteristics using several different unsupervised classifiers. Then, we statistically model the suspected bias and determine whether it is strong enough to produce the Class III bursts.  A number of pertinent questions will have to be addressed in pursuing this objective: Do theoreticians need to develop models for one, two, or three gamma-ray burst classes? How can data mining techniques be used to aid scientific classification? Are systematic effects present in data collected by BATSE or other gamma-ray burst experiments that alter classification structures? Can these effects be understood? Can information on intrinsic properties of the source population be extracted if these effects are present? Can future instruments be designed to minimize or eliminate these effects? ",
        "conclusions": "We have demonstrated that  \\begin{enumerate} \\item Gamma-ray burst Class III does not have to represent a separate source population; it can be produced by the integration time of the instrumental trigger, \\item the fluence duration bias by itself, as modeled from a sample of high signal-to-noise bursts, is unlikely to be responsible for the existence of Class III. \\item Class III is likely produced by an excess of short, low fluence bursts detected by BATSE's short trigger temporal window. \\item The excess bursts can be eliminated via a selection process that is dual timescale peak flux-limited, rather than peak flux-limited or fluence-limited. \\item The dual timescale peak flux measure resulting from this selection process appears to correlate better with hardness (and therefore with $E_{\\rm peak}$ and redshift) than either peak flux or fluence. This adds support to the argument that dual timescale peak fluxes correct the temporal limitations introduced by using single timescale peak fluxes. Dual timescale peak fluxes can be established for many combinations of temporal measurements. \\end{enumerate}  The results found here are important to gamma-ray burst astrophysics as well as to the general problem of scientific classification. Data mining tools can help identify complex clusters in multi-dimensional attribute spaces. The tools are sensitive to clusters and data patterns, as evidenced here because they have allowed us to discover clusters produced artificially as a result of sample incompleteness. This sensitivity is advantageous, because a better understanding of instrumental response and sampling biases can be used to improve the design of future instruments.  We note that sample incompleteness is generic and applies to the detection of any transient sources identified as the result of a temporal trigger. Examples of transient event statistics that might be biased by a temporal trigger include flare stars, soft gamma repeaters, x-ray bursts, and earthquakes.  However, the sensitivity of data mining tools can also cause problems. Data mining is central to the operation of planned Virtual Observatories, which will electronically combine data collected from a variety of instruments with a range of temporal, spectral, and intensity responses. Since sample incompleteness can cause a single instrument with one set of characteristics to find phantom classes, classes identified using multiple instruments should be interpreted cautiously. The instrumental responses of Virtual Observatory components will have to be accurately known in order for newly-identified classes to be recognized as separate source populations.  It is important to recognize that data mining techniques have their limitations. Principal component analysis has identified fluence, duration, and hardness as being critical gamma-ray burst classification attributes, while the trigger attribute of peak flux was not chosen. Data mining classifiers failed to recognize that attribute selection had removed the attribute that could have provided the most insight into the gamma-ray burst clustering structure."
    },
    "0209/astro-ph0209559_arXiv.txt": {
        "abstract": "In an attempt to reconcile recent spectral data with predictions of the standard cooling flow model, it has been suggested that the metals in the intracluster medium (ICM) might be distributed inhomogeneously on small scales. We investigate the possible consequences of such a situation within the framework of the cooling flow scenario. Using the standard isobaric cooling flow model, we study the ability of such metallicity variations to preferentially suppress low-temperature line emission in cooling flow spectra. We then use simple numerical simulations to investigate the temporal and spatial evolution of the ICM when the metals are distributed in such a fashion. Simulated observations are used to study the constraints real data can place on conditions in the ICM\\@. The difficulty of ruling out abundance variations on small spatial scales with current observational limits is emphasized. We find that a bimodal distribution of metals may give rise to interesting effects in the observed abundance profile, in that apparent abundance gradients with central abundance drops and off-centre peaks, similar to those seen recently in some clusters, are produced. Different elements behave in different fashion as governed by the temperature dependence of their equivalent widths. Our overall conclusion is that, whilst this process alone seems unlikely to be able to account for the sharp reduction in low temperature emission lines seen in current spectral data, a contribution at some level is possible and difficult to rule out. The possibility of small-scale metallicity variations should be considered when analysing high resolution cluster X-ray spectra. ",
        "introduction": "Data from the latest generation of X-ray satellites, \\chandra{} and \\xmm, are forcing us to re-examine some of the basic tenets of the traditional cooling flow model \\citep[e.g.,][]{fabi94}. Various authors \\citep[e.g.,][]{kaas01,pete01,tamu01} have drawn attention to the discrepancy between the predictions of the standard cooling flow model and observed spectra of cluster central regions. The expected emission lines for several important species (e.g.\\ the \\ion{Fe}{xvii}{} 15 and 17\\,\\AA{} lines) do not seem to be present at the levels which simple models would expect. This is a trend that appears to be repeated in many clusters that have traditionally been thought to harbour cooling flows \\citep{pete02}. Lines such as these are important because they are strong indicators of low ($\\la 1 \\unit{keV}$) temperature gas (see Section~\\ref{sec:icm_thermo}). In a standard cooling flow, in which gas is cooling down to essentially zero, we expect a significant flux in such lines.  Several ideas \\citep[e.g.,][]{fabi01a,pete01} have been put forward to explain this discrepancy. Here we focus on just one of these, the suggestion that the intracluster medium (ICM) metals might be distributed inhomogeneously on small-scales. We examine in detail the consequences of such a scenario, investigating the effects of an ICM metallicity which varies on small, unresolved scales (sub-kpc, say). This idea is no more than a minor extension of the multiphase cooling flow model \\citep{nuls86}, which has always relied upon the coexistence of phases with different densities and temperatures at the same radius in the ICM\\@. We merely allow the chemical composition of these phases to vary as well. This idea is attractive in the context of cooling flows since it leaves the spectra of high ($\\ga 1\\unit{keV}$) temperature gas essentially unchanged, whilst reducing the line emission from low temperature gas. It (potentially) enables us to reconcile the data and models whilst making the smallest conceptual changes to the standard model.  Indeed, given that a suppression of thermal conduction (electron motion) implies an even more severe reduction in the freedom of heavy ions to diffuse through the ICM, we might expect small-scale metallicity variations to be a natural consequence of the multiphase cooling flow model, in which conduction is by necessity heavily reduced. Conversely, if ICM thermal conduction is relatively efficient (as has been suggested recently by several authors, e.g., \\citealt{nara01,voig02}) then this does not necessarily imply that the same is true of ion motion.  In Section~2 we examine the effects of an inhomogeneous metallicity within the framework of the standard isobaric cooling flow model \\citep[e.g.,][]{john92}, quantifying the reduction in equivalent width that can be expected for several important ICM species as the degree of inhomogeneity is varied. In Section~3 we outline an improved numerical model that allows us to carry out simulations with spatial and temporal evolution, and to produce simulated observations of \\chandra{} spectral data. Section~4 describes some of the results of this model for a simple case of metallicity inhomogeneities. We find that small-scale abundance variations may give rise to the appearance of abundance gradients where none in fact exist, with the results for individual species being controlled by the temperature dependence of their equivalent widths. We stress the difficulty of resolving such variations without a combination of good spatial and spectral resolution. In the final Section we make some discussion concerning the likelihood of such small-scale metallicity variations in the ICM\\@. Given that direct detection of such structures is currently unfeasible, we put forward some suggestive circumstantial evidence.  Throughout this paper we assume: $H_{0} = 50$\\kmpspMpc; $\\omegam = 1.0$, $\\omegal = 0.0$. We make use of \\xspec{} version 11. ",
        "conclusions": "\\label{sec:disc}  To address the question of how small-scale ICM abundance variations might come about, supernovae can be considered as essentially point sources (on ICM scales) of extremely high metallicity gas. In of themselves, these have structures that are far from simple \\citep[e.g.,][]{hugh00}. One way in which escape of the enriched gas from the potential well of the host galaxy into the ICM may occur is via superwinds \\citep[e.g.,][]{heck01} (although other processes such as ram-pressure stripping and pregalactic winds are doubtless significant). Superwinds have a complex, multiphase structure in which it is difficult to probe directly the energetic, enriched gas driving the wind. There is, however, clear evidence for complicated structure in the wind on very small scales in the optical and soft X-ray \\citep[e.g.,][]{stri01}. In our opinion, the question ought to be posed the other way around: how might the ICM become uniformly enriched?  There are essentially two issues that determine whether or not the situation as we have chosen to model it is physically realistic: i) how are the metals injected into the ICM, and ii) once in the ICM, do the metals move to spread out over a wide area or remain confined? The answers to both these questions remain uncertain at present. To address the second question first, following \\citet{spit62}, we may express the deflection time (average time for a cumulative deflection of $90^{\\circ}$) for particles of mass $m$, charge $Ze$ diffusing through field particles of mass $\\mysub{m}{f}$, charge $\\mysub{Z}{f}e$, density $\\mysub{n}{f}$ as \\begin{equation} \\tau = \\frac{6 \\sqrt{3} \\pi \\epsilon_{0}^{2} \\mysub{k}{B}^{3/2}}{e^{4}} \\frac{m^{1/2} T^{3/2}}{Z^{2} \\mysub{Z}{f}^{2} \\mysub{n}{f} \\mathcal{F}(x) \\ln{\\Lambda}} \\end{equation} where \\begin{eqnarray} x & = & \\sqrt{\\frac{3 \\mysub{m}{f}} {2 m}}\\\\ \\mathcal{F}(x) & \\equiv & \\erf{x} + \\frac{\\diff}{\\diff\\,x} \\left( \\frac{\\erf{x}}{2 x} \\right) \\end{eqnarray} with $\\erf$ the standard error function, $\\ln{\\Lambda}$ the Coulomb logarithm, and assuming that all species are in thermal equilibrium at temperature $T$. The factor $\\mathcal{F} \\le 1$ is the Chandrasekhar correction for the finite mass of the field particles.  Numerically \\begin{equation} \\tau \\approx 0.26 \\unit{Myr}\\quad A^{1/2} \\left( \\frac{T}{10^{8}\\unit{K}} \\right)^{3/2} \\left( \\frac{\\mysub{n}{f}}{\\unit{cm}^{-3}} \\right)^{-1} \\frac{1}{Z^{2} \\mysub{Z}{f}^{2} \\mathcal{F} \\ln{\\Lambda}} \\end{equation} with $A$ the test particle mass in units of the proton mass.  As was shown by \\citet{reph78} for the case of sedimentation, in the ICM the contributions from field particles other than protons (particularly helium nuclei) are significant. Summing over the elements in a $0.3\\Zsun$ plasma with iron nuclei as test particles, we find that $\\tau$ is reduced by around a factor of two from the value for a pure proton plasma. At $10^{8}\\unit{K}$, and with $\\mysub{n}{H} = 10^{-3}\\unit{cm}^{-3}$, we find the corrected deflection times for iron and helium nuclei are $\\mysub{\\tau}{Fe} \\approx 0.3 \\unit{Myr}$, $\\mysub{\\tau}{He} \\approx 4 \\unit{Myr}$.  From standard kinetic theory, the root-mean-square three-dimensional distance a particle will diffuse in a time t is given by \\begin{equation} \\mysub{r}{rms} \\; = \\; \\sqrt{2 \\lambda \\mysub{v}{rms} t} \\; = \\; \\mysub{v}{rms} \\sqrt{2 \\tau t} \\; = \\; \\sqrt{\\frac{6 \\mysub{k}{B} T \\tau t}{m}}, \\end{equation} where $\\lambda = \\tau \\mysub{v}{rms}$ is the mean free path. Hence we can estimate that iron nuclei may diffuse about 20\\unit{kpc} in 1\\unit{Gyr}. To maintain small-scale abundance variations over significant timescales in the ICM, we therefore require a strong suppression of diffusion. This result is to be expected, as it is equivalent to the strong suppression of conduction that has traditionally been invoked for the multiphase cooling flow picture. Magnetic fields have normally been appealed to as the causative agent to dampen transport properties in an ionized gas. Early calculations \\citep[e.g.,][]{trib89} appeared to show that a tangled magnetic field would indeed produce a strong reduction in transport. More recent calculations considering a chaotic field with turbulence extending over a wide range of length scales indicate that the suppression may in fact be minimal \\citep{nara01}.  In practice, simple diffusion is probably unlikely to be the limiting factor for the spread of ICM metals. Convection, turbulence due to radio sources, galactic wakes, etc., will also play roles to varying degrees. If, however, the abundance drops observed with \\chandra{} in the central regions of the ICM for several clusters are genuine, then this will imply limits on the amount of convection or mixing that can have taken place (else these features would have been smoothed out). It does not seem unreasonable to explore some of the consequences of small-scale metallicity variations, and to keep the possibility in mind during spectral analyses.  The question as to whether or not the metals in the intracluster and intergalactic media are homogeneously distributed remains an open one. There is a non-negligible scatter in Galactic stellar metallicities for stars of all ages (the scatter increases for low metallicity stars), which has been taken to suggest that that Galactic disc has been chemically inhomogeneous throughout its development \\citep[e.g.,][]{mcwi97}. Classical galactic chemical evolution models have tended to assume instantaneous dispersion of synthesized elements.  The average metallicity of the intracluster medium on large scales has for some time been reasonably well established at roughly 1/3 solar in most cases, both for nearby \\citep{edge91a} and distant \\citep{mush97} clusters. Until comparatively recently, however, there has been little information on how the ICM metals might be distributed on finer scales. There is, in our opinion, no particular reason why the whole ICM should be uniformly enriched to the same metallicity, although this might be one's natural assumption. As we discuss in Section~\\ref{subsec:reso}, direct detection of small-scale abundance variations in the ICM will be very difficult. Consequently, we feel it is impossible to rule out such variations at the present time.  Recently, a deal of support has been given to the idea that thermal conduction might be operating in cluster cores at significant levels \\citep[e.g.,][]{nara01,voig02,fabi02}. Whilst this runs contrary to the established multiphase cooling flow picture \\citep{nuls86}, it seems to have a degree of success in matching the observed temperature profiles. It is certainly the case, though, that at least in some regions of the ICM, conduction is highly suppressed. This is revealed by the `cold-fronts' seen in several cluster cores, e.g.\\ A2142 \\citep{mark00}. It was shown by \\citet{etto00} that such features, which correspond to a sharp jump in the surface brightness profile, indicate a strong (factors of several hundred) \\textit{local} suppression of the thermal conductivity, else the temperature discontinuity that is responsible for the brightness jump would be quickly washed out. If thermal conduction is suppressed, then ion movement must be even more restricted, since net heat conduction may still take place without individual electrons travelling a significant distance. If thermal conduction is operating with a degree of efficiency, however, then the same need not necessarily apply to ion motion. If this were the case, though, there would be the possibility of sedimentation of the heavy elements \\citep{fabi77,reph78,fabi02}. Cold-fronts of course only provide direct evidence for inhibited thermal conduction across the fronts themselves. It is possible (though as yet far from proven) that conduction may reach the \\citet{spit62} value elsewhere \\citep[e.g.,][]{voig02}. If conduction is operating efficiently, it does not necessarily rule out small-scale metallicity variations, but it does weaken the case, both from the point of view of reducing/removing the motivation for them (suppressing the low-temperature cooling flow line emission by suppressing the low temperature gas), and by making it less likely that such conditions can persist.  Our modelling is simplistic, and it may be argued that the conditions as we have chosen to represent them are not physically relevant. For example, there is no possibility of segregation of elements based on weight (that is, it is not possible for the metal-rich gas to sink to the centre). This is a consequence of our adopting the theoretical framework of \\citet{nuls86}, which requires that in a multiphase flow the various phases co-move, i.e.\\ there is a single velocity profile. The arguments for this idea are set out in detail in \\citet{nuls86}. Also, the timescale over which the changes in the abundance profile occur is somewhat short. This is a result of the rather unphysical initial condition; starting out with regions enriched to 3\\Zsun{} and not allowing any replacement of cooling metals. A more realistic treatment would allow the metallicity to build up initially with time and then allow for some replenishment, and possibly also for different distributions of SNe Ia and II products. However, we are not concerned here with producing detailed models of the evolution, but rather in seeing what the general trends of behaviour might be. Note that because we have restricted our attention to single-phase models, there is no real transport of material by the cooling flow, since all the mass loss (except for the metal-rich zones) takes place in the centre. \\citet{reis96} demonstrated the ability of multiphase cooling flows to create (genuine) abundance gradients through transport of injected materials. Thus we may expect that a more sophisticated multiphase treatment of this process might reveal more complex effects.  Given that the abundance gradients observed with \\asca{} and \\chandra{} are seen clearly in the iron K line, the results of Section~\\ref{sec:kgradient} imply that this cannot be solely due to a bimodal distribution of metallicities. The results are not, however, inconsistent with such a metal distribution, at least in the simple scenario outlined here. At the very least, we have presented another reason to be wary of the iron L complex when fitting to X-ray spectra \\citep[e.g.,][]{fino00}. In our case, it is not due to the complex atomic physics of the L shell (the code we use to generate the spectra is the same as that we use to fit them), but rather to the temperature dependence of the equivalent width. Recall that iron is really the only ICM element for which one has two strong spectral indicators (K and L); for other elements only the K lines are useful. Playing devil's advocate, one could therefore imagine a situation in which the observed gradients in the \\textit{iron} profile are genuine, but those in the other elements are due to a process such as the one outlined here. This would, for example, severely impact on estimates of the SNe Ia:II importance ratio. Of course, this simple model would not explain the matching radial scales of the variations for iron and the other elements in this case. Nor could it explain any correlation between the iron profile and the visible light in the central regions of the cluster.  Another process which may give rise to the same sort of radial abundance profile as those seen here (namely a peak in the abundance at an off-centre position) is resonant scattering \\citep{gilf87}. This acts to redistribute photons from the central regions of the cluster (where the optical depth is highest) to a surrounding ring. See the work of \\citet{math01} for an application of these ideas to M87. Computation of detailed optical depths requires knowledge of the velocity structure of the gas along the line of sight. We will not comment on the resonant scattering issue here, except to say that to some extent the ideas of this paper and those of resonant scattering are in conflict. As was pointed out by \\citet{wise92}, clumping of the ICM reduces the amount of scattering that takes place. X-ray surface brightness profiles depend on the rms density $\\sqrt{\\langle n^{2} \\rangle}$, whereas electron scattering depends on the mean density $\\langle n \\rangle$. The rms of a set of numbers is always greater than their mean (unless the numbers are all equal). Thus increasing the degree of clumping in the plasma reduces the mean density relative to the rms density and so reduces the relative effect of scattering.  Another possibility is that the `extra' metals in the central regions of many clusters are to be associated with the central cD galaxy. \\citet{maki01} have suggested an alternative explanation for the enhanced emission seen at the centre of many clusters, which has traditionally been attributed to the cooling flow phenomenon by many researchers. Instead, these authors suggest the excess may be associated with the inter\\textit{stellar} medium of the central cD galaxy. One argument invoked against the cooling flow interpretation is the fairly frequent presence of metallicity increases near cluster centres. Our present work suggests a mechanism by which such effects may indeed be produced by cooling flows. The reality or otherwise of any central dips in abundance would be an important discriminant for these two interpretations.  \\citet{bohr02} also give some consideration to the possible effects of inhomogeneous metallicities on the observed abundances (section 2.2 of their paper). Unlike our consideration of individual emission lines, they look at the overall observed metallicity of the spectrum, as inferred from the global ratio of power in all emission lines to power in the continuum. These authors also discuss some comparisons between the shape of the spectrum around the 1\\unit{keV} region for M87 and the predictions of the bimodal model. All hypotheses live or die through comparison with data, so such investigations are highly necessary, but there is a large parameter space to investigate. And as they point out, such checks must be made on a source-by-source basis. They find a poor fit between the actual spectral shape and the predictions of the bimodal hypothesis (e.g.\\ their fig.~6). Note, however, that this is for the case where the normalizations of the metal-rich and metal-poor phases are `roughly equal'. We would not expect such a division to be successful in reducing the EW of the low temperature lines by an appreciable amount.  Our main result is that the observed equivalent width suppression for low temperature emission lines in a cooling flow spectrum due to a metal distribution which is inhomogeneous on small-scales is not as great as one might expect. For example, if all the metals are concentrated in 10\\% of the gas, the suppression of the low-temperature lines relative to those from high temperature gas is only about a factor of 3 (Section~\\ref{sec:cflow_ew}). It seems unlikely, therefore, that this method in isolation could produce a reduction in equivalent width equal to that seen in data, without pushing the bimodality of the metal distribution to extreme levels. There is an effect, but it is not large enough. We have also shown that small-scale metallicity variations can give rise to interesting effects in the observed abundance profiles (Section~\\ref{sec:grads}) as compared to the true profiles. Such effects would give rise to serious difficulties in terms of interpreting abundances in cluster central regions. The possibility of small-scale metallicity variations ought to be borne in mind when analysing high resolution X-ray spectra of cluster central regions."
    },
    "0209/astro-ph0209245_arXiv.txt": {
        "abstract": "A short Chandra ACIS-S observation of the Seyfert 2 galaxy IC 2560, which hosts a luminous nuclear water megamaser, shows: 1) the X-ray emission is extended; 2) the X-ray spectrum shows emission features in the soft ($E < 2$ keV) X-ray band; this is the major component of the extended emission; and 3) a very strong (EW$ \\sim 3.6$ keV) iron K$\\alpha$ line at 6.4 keV on a flat continuum. This last feature clearly indicates that the X-ray source is hidden behind Compton-thick obscuration, so that the intrinsic hard X-ray luminosity must be much higher than the observed, probably close to $\\sim 3\\times 10^{42}$\\ergps. We briefly discuss the implications for powering of the maser emission and the central source. ",
        "introduction": "IC 2560 is a relatively nearby ($D=26$ Mpc) barred spiral galaxy, classified as a Seyfert 2 (Fairall 1986; Kewley et al 2001). It is notable for exhibiting luminous \\htwoo\\ maser emission from its nucleus (Braatz, Wilson, \\& Henkel 1996). This maser emission resembles that from the archetypal water megamaser source NGC 4258 in that high-velocity emission is seen up to $\\Delta V\\approx 400$ \\kmps\\ away from the systemic velocity, the systemic emission is much stronger than the high-velocity emission, and centripetal acceleration of systemic velocity features has been reported (Ishihara \\etal 2001). The high-velocity emission has not yet been imaged; if the data are interpreted in the framework of a Keplerian disk, as in NGC 4258, the implied central mass is $M_c\\approx 2.8\\times 10^6$ \\Ms. Ishihara \\etal also analyzed an ASCA observation of IC 2560 and concluded that it possesses a fairly heavily obscured ($N_H\\sim 3\\times 10^{23}$ \\psqcm) but low luminosity ($L_{2-10 \\rm keV}\\sim 10^{41}$ \\ergps) X-ray source.  In this paper we report on a short Chandra observation of IC 2560, which reveals spatially extended soft X-ray emission and a substantially different nature to the hard X-ray source than was inferred by Ishihara \\etal (2001) from the ASCA data. Throughout we assume the distance to the galaxy to be 26 Mpc, giving an angular scale of 120 pc arcsec$^{-1}$. ",
        "conclusions": "\\subsection{The true luminosity of the hidden active nucleus}  The hard X-ray spectrum dominated by an iron K line indicates the absence of direct continuum emission from a central source in the Chandra band, meaning that we are seeing only reflected light from a hidden nucleus. The absorption column density must be larger than \\nH $\\sim 1\\times 10^{24}$ \\psqcm, i.e., the X-ray source is Compton-thick. How much of the intrinsic luminosity emitted by the hidden nucleus is seen in reflection depends on the obscuration/reflection geometry. Reflection from cold material, as inferred from the iron line energy and the flat hard X-ray continuum, is not an efficient process, since photoelectric absorption within the reflector suppresses reflected light significantly, in particular at low energies. With the observed continuum luminosity of $1.3\\times 10^{40}$ \\ergps\\ in the 2--10 keV band, the minimum incident luminosity to yield the reflection is about $2.6\\times 10^{41}$ \\ergps. However, in a realistic toroidal geometry in which the incident source is hidden from our direct view, the true value will be much larger.   We estimate the upper limit to the 2--10 keV luminosity to be $\\sim 3\\times 10^{42}$\\ergps, 10 per cent of the infrared ($\\approx$ bolometric) luminosity, $L_{\\rm 8-1000\\mu m}\\simeq 3\\times 10^{43}$ \\ergps\\ (obtained from the Infra-Red Astronomical Satellite (IRAS) measurements using the formula given in Sanders \\& Mirabel (1996)), assuming the typical 2-10 keV X-ray to bolometric luminosity ratio for Seyfert galaxies (e.g., Mushotzky, Done \\& Pounds 1993). The AGN luminosity is likely to be close to the above upper limit, given the warm IRAS colour ($S_{60\\mu {\\rm m}}/S_{25\\mu {\\rm m}}=3.4$) and the AGN-dominated nuclear optical spectrum, suggesting that the infrared emission is predominantly powered by a hidden active nucleus.   \\subsection{Large EW of Fe K line}   The equivalent width for the iron K$\\alpha $ line, in excess of 3 keV, is one of the largest measured among the reflection-dominated Seyfert 2 galaxies (see Matt et al 2001 and Levenson et al 2002 for recent compilations).  Iron overabundance (2--3 solar) could be a possible reason (Ballantyne, Fabian \\& Ross 2002), but even with solar metallicity, an optically thick torus can produce a very large EW. In models assuming obscuring matter in a toroidal form, the Fe K line EW depends on the optical depth and geometry of the torus (Leahy \\& Creighton 1993; Ghisellini, Haardt \\& Matt 1994; Krolik, Madau \\& \\.Zycki 1994; Levenson et al 2002). All these models indicate that, in order to produce an EW as large as 3.6 keV, a torus needs to have a column density \\nH $\\geq 3\\times 10^{24}$ \\psqcm (or Thomson optical depth, $\\tau_{\\rm T}\\geq 2$) and a small half-opening angle, $\\theta\\leq 20^{\\circ}$, and to be viewed nearly edge-on.  The required column density is consistent with the lack of transmitted primary source emission in the Chandra spectrum, from which the lower limit of the line of sight absorption column density, \\nH $\\geq 1\\times 10^{24}$\\psqcm, has been derived (Section 4.1). The requirement of a small opening angle means that the X-ray absorbing matter should be located close to the central source. The inner radius of the water maser disk was inferred to be 0.07 pc (Ishihara et al 2001), but the accretion disk could extend inwards, with the inner part of the disk too cold to induce masing (Neufeld \\& Maloney 1995, hereafter NM95). Perhaps this dense ($n>10^{10}$ cm$^{-3}$) inner part of the accretion disk may be the region where the X-ray absorption primarily occurs. However, in this case, the water maser-emitting part of the disk must be warped in order to see the X-ray flux directly, otherwise the X-rays from the central source will not impinge on the disk.   \\subsection{Water maser disk and accretion flow}  If the geometry of the maser emission in IC 2560 is similar to that in NGC 4258, which has not yet been determined by imaging, then we can use the X-ray emission and the kinematic data to place constraints on fueling of the central massive black hole. The observed velocities of maser emission in IC 2560 imply that the outer disk radius is at $R_{out}\\approx 0.26$ pc. Assuming that the maser emission is powered by the X-ray flux from the central source (Neufeld, Maloney \\& Conger 1994; NM95), as appears likely for the majority of \\htwoo\\ megamaser sources (see Maloney 2002 for a recent review), we can identify the outer radius of the disk with the location of the molecular to atomic phase transition (see equation 4 of NM95). For a hard X-ray luminosity of $L_{2-10}=10^{42}L_{42}$ \\ergps\\ and black hole mass $M_c=10^6 M_6$ \\Ms, the ratio of the mass accretion rate to the viscosity parameter $\\alpha$ is then \\begin{equation} {\\dot M\\over \\alpha}\\approx 6.2\\times 10^{-3} \\left({R_{out}\\over 0.26\\;{\\rm pc}}\\right)^{1.23} L_{42}^{0.53}M_6^{-0.77}\\ M_\\odot\\;{\\rm yr^{-1}} \\end{equation} which, for the inferred mass of the central black hole $M_c\\approx 2.8\\times 10^6$ \\Ms, is $\\dot M/\\alpha\\approx 2.8\\times 10^{-3}$ \\Ms~\\pyr. Assuming that the bolometric luminosity is approximately ten times the 2--10 keV X-ray luminosity, the product of $\\alpha$ and the radiative efficiency factor $\\epsilon$ is then $\\alpha\\epsilon\\sim 0.06$, similar to the value inferred for NGC 4258 (NM95). (In making this estimate we have assumed that flow through the disk is steady over the accretion timescale from the disk outer edge, and that the disk sees the true X-ray luminosity.) Although the fractional Eddington luminosity ($\\sim 0.03$) and mass accretion rate are much higher than in NGC 4258, the efficiency of radiation appears to be similar. We also note that the Compton-thick obscuration of the hard X-ray source is consistent with the accretion disk itself acting as the obscurer, as in NGC 4258, given the larger derived value of $\\dot M/\\alpha$."
    },
    "0209/astro-ph0209135_arXiv.txt": {
        "abstract": "A new mechanism for acceleration and enrichment of $^3\\mathrm{He}$ during impulsive solar flares is presented. Low-frequency electromagnetic plasma waves excited by the Electron Firehose Instability (EFI) can account for the acceleration of ions up to \\mev energies as a single stage process. The EFI arises as a direct consequence of the free energy stored in a temperature anisotropy ($T^e_\\parallel>T^e_\\perp$) of the bulk energized electron population during the acceleration process. In contrast to other mechanisms which require special plasma properties, the EFI is an intrinsic feature of the acceleration process of the bulk electrons. Being present as a side effect in the flaring plasma, these waves can account for the acceleration of $^3\\mathrm{He}$ and $^4\\mathrm{He}$ while selectively enhancing $^3\\mathrm{He}$ due to the spectral energy density built up from linear growth. Linearized kinetic theory, analytic models and test-particle simulations have been applied to investigate the ability of the waves to accelerate and fractionate. As waves grow in both directions parallel to the magnetic field, they can trap resonant ions and efficiently accelerate them to the highest energies. Plausible models have been found that can explain the observed energies, spectra and abundances of $^3\\mathrm{He}$ and $^4\\mathrm{He}$. ",
        "introduction": "Solar flares are commonly divided into two different classes: impulsive and gradual \\citep*{caneetal1986}. The division into these two categories can be done on the basis of the duration of their soft X-ray emission \\citep*{pallavicinietal1977}. But it is not only the timescale of the events that justifies the distinction: the energetic particles observed in space from impulsive flares exhibit strong abundance enhancements over coronal values \\citep[and references therein]{lin1987,reames1990}. Impulsive flares are usually dominated by energetic electrons and are characterized by $^3\\mathrm{He}/ ^4\\mathrm{He}$ ratios at \\mev energies that are frequently 3 to 4 orders of magnitude larger than the corresponding value in the solar corona and solar wind where $^3\\mathrm{He}/^4\\mathrm{He} \\sim 5\\cdot10^{-4}$. They also exhibit enhanced $^4\\mathrm{He}/\\mathrm{H}$ and $\\mathrm{Fe}/\\mathrm{C}$ ratios. Although the occurrence of \\hedrei and \\fe enrichments are correlated in impulsive flares, the ratio $^3\\mathrm{He}/\\mathrm{Fe}$ shows huge variations as observed by \\citet*{masonetal2000}. This suggests, that different mechanisms are responsible for the acceleration of the two species. Gradual flares usually have large energetic proton fluxes, small $^4\\mathrm{He}/ \\mathrm{H}$ and do not show large $^3\\mathrm{He}/ ^4\\mathrm{He}$ or $\\mathrm{Fe}/\\mathrm{C}$ enhancements in the energetic particles, although approximately $5\\%$ admixture of suprathermal remnant particles from impulsive flares have been observed in gradual events by \\citet{tylkaetal2001}. The standard interpretation for these observations is that the energetic particles in impulsive events origin in the energy release region on the sun while the energetic particles in gradual events are accelerated via shocks, either coronal or interplanetary \\citep{lin1987,luhnetal1987}.  Abundance ratios therefore are a valuable diagnostics for the flaring plasma itself and in particular the specific acceleration mechanism for the energetic particles. The selectivity of the mechanism, especially for \\hedrei and \\hevier indicates resonant processes such as gyroresonant interaction of plasma waves with the ions. Theoretical ideas therefore focus on the unique charge-to-mass ratio of \\hedrei which allows it to be selectively pre-heated or accelerated via gyroresonance.  A well-known theory of the initial set among theories for \\hedrei enhancement was published by \\citet{fisk1978}, explaining the preferential acceleration of the ions by electrostatic ion cyclotron (EIC) waves at a frequency in the vicinity of the gyrofrequency of \\hedrei. The waves are excited by an electron current and interact with \\hedrei via cyclotron resonance. A large enhancement of $^4\\mathrm{He}/\\mathrm{H}$ is required in the ambient plasma for this instability to excite waves above the \\hevier gyrofrequency.  More recently a theory was suggested by \\citet{temerinroth1992} who accounted for the preferential \\hedrei acceleration proton electromagnetic ion cyclotron ($\\mathrm{H}^+$ EMIC) waves. These waves are driven unstable by non-relativistic (keV range) electron beams and their frequencies lie at around the \\hedrei gyrofrequency at almost perpendicular propagation. In \\citet{temerinroth1992}, auroral observations of keV electron beams and $\\mathrm{H}^+$~EMIC waves are taken as experimental evidence that $\\mathrm{H}^+$ EMIC waves also may acquire a substantial fraction (order of few percent) of the electron beam energy under coronal conditions. At the Sun, plasma emission from tens of keV electron beams on open magnetic field lines is the explanation for type III radio emission with its equivalent, the U-bursts, on closed field lines (in loops) at beam energies of order keV. In order to excite $\\mathrm{H}^+$~EMIC waves fulfilling the requirements of this model, electron beams of much higher density than the observed ones have to be postulated \\citep{millervinas1993} with no direct observational evidence. Moreover, it is difficult to explain from a theoretical point of view that the free energy in the electron beam is transferred to the $\\mathrm{H}^+$~EMIC waves and not to the much faster growing ($\\sim 4$ orders of magnitude in growth rate) electron plasma waves.  In this work an alternative model for acceleration of \\hedrei and its enhancement over \\hevier is presented. The approach is different from the models described above. While the models mentioned above postulate rather special plasma properties ($^4\\mathrm{He}/\\mathrm{H}$ overabundance in the pre-flaring plasma, dense low-energy beams) in order to produce the required plasma waves, the model presented herein explains the unique overabundance of \\hedrei by plasma waves excited as an intrinsic feature of the electron acceleration process itself. Parallel propagating, lefthand polarized electromagnetic waves driven by the Electron Firehose instability (EFI) can account for \\hedrei acceleration via gyroresonant interaction. As suggested in \\citet{paesoldbenz1999} such Electron Firehose (EF) waves are excited by anisotropic electron distribution functions ($T_\\parallel>T_\\perp$) that occur in the course of the acceleration process of bulk electrons in solar flares.  Although the EF waves are not narrowbanded around the gyrofrequency of \\hedrei as the $\\mathrm{H}^+$~EMIC proposed by~\\citet{temerinroth1992}, selectivity of the process is achieved by the natural profile of the spectral wave energy. No additional assumptions besides the electron anisotropy are needed, and the model can be embedded as an intrinsic feature in acceleration scenarios such as transit-time damping, the currently most popular stochastic acceleration model.  In \\S~\\ref{idea} the basics of the new acceleration scenario are described. The properties of the EF waves under coronal conditions are presented in \\S~\\ref{props}. Analytical and numerical results on the heating rates are presented in \\S~\\ref{heating} and \\S~\\ref{sim}. The mechanism for enhancement of \\hedrei over \\hevier is described in \\S~\\ref{enhancement}, and heavier ions are discussed in \\S~\\ref{heavy}. \\S~\\ref{conc} concludes this work. ",
        "conclusions": "\\label{conc} A model for the enrichment of \\hedrei during impulsive solar flares is presented. The acceleration of \\hedrei and \\hevier can be understood as consequence of Electron Firehose (EF) waves, resulting from an unstable anisotropic electron distribution with $T^e_\\parallel>T^e_\\perp$. The model does not need additional sources of free energy or plasma properties than those resulting directly from the acceleration of the bulk of electrons. The essential result of the investigation is that EF waves, excited by an anisotropic electron distribution function, can accelerate \\hedrei and \\hevier via cyclotron resonance to $\\mathrm{MeV\\;amu}^{-1}$ energies on timescales of $\\sim 1\\;\\mathrm{s}$. In support of this conclusion, the following specific results are found:  1. The EF waves accelerate \\hedrei ions via cyclotron resonance. The symmetry of EF waves in $k_\\parallel$ greatly enhances the efficiency of the acceleration process with respect to the standard cyclotron resonant wave-particle interaction of a unidirectional propagating wavefield. The parallel force driving the ion out of resonance, and therefore limiting the acceleration time in the unidirectional case, cancels out in the time average for the counterpropagating wavefield. The total resonant interaction time therefore is strongly increased and the particle can reach high energies.  2. The linear growth of the EF waves self-consistently generates a spectral energy distribution, causing enhanced acceleration of \\hedrei with respect to \\hevier. It is found from test-particle simulations that the heating rate of \\hedrei exceeds the values for \\hevier by about a factor of 2--3.  3. It is shown that this small enhancement in the heating rate of \\hedrei above \\hevier already can account for the observed enhancement in \\heratio from the coronal value of $\\sim5\\times10^{-4}$ up to $0.1-1$ during impulsive solar flares. Due to the larger heating rates, the high energy tail of \\hedrei populates energies above \\mev faster. It reaches the required abundance ratio of \\heratio for an acceleration timescale of $\\sim0.76\\;\\mathrm{s}$.  4. The fraction of \\hedrei accelerated by the EF waves is large enough to account for the observed particle fluxes. Between $4-11\\%$ of the coronal \\hedrei population in the flare is accelerated above \\mev.  5. The acceleration model reproduces the qualitative behavior of the \\hedrei energy spectrum with respect to the \\hevier energy spectrum. As observed by \\citet{moebiusetal1982} the \\hedrei spectrum is generally harder than the \\hevier spectrum. Moreover, a turnover in the \\hedrei energy spectrum around \\kev is obtained, reproducing the observations by \\citet{masonetal2000}.  \\smallskip  EF waves are an inherent property of stochastic electron acceleration by transit-time damping of cascading fast-mode waves. Thus, their ability of accelerating \\hedrei and \\hevier supports the scenario of stochastic acceleration of the bulk electrons in impulsive flares. Future work will have to extend these results to the oblique EF mode and investigate its role in the enhancement of heavy ions and in the flare acceleration process in general."
    },
    "0209/astro-ph0209303_arXiv.txt": {
        "abstract": "s{ After a short historical introduction to the field of \\ga-ray line astronomy with radioactivities, I present an overview of recent results concerning the massive star yields of those radioactivities. I comment on the implications of  those results (concerning long-lived radioactivities, like \\Al \\ and \\Fh) for \\ga-ray line astronomy, in the light of past ({\\it COMPTEL} \\ and GRIS) and forthcoming ({\\it INTEGRAL}) observations. } ",
        "introduction": "Shortly after the discovery of the phenomenon of radioactivity, radionuclides revealed to be unique '''probes'' in our study of the cosmos and important agents in its evolution (radioactive dating of the Earth, meteorites and stars; radioactive heating of planetary and supernova interiors; radioactive origin of abundant stable nuclei, like \\Fe, and of isotopic anomalies in meteorites, etc).  As most other stable nuclei, radionuclides are produced in stellar interiors and ejected in the interstellar medium through stellar winds and explosions (nova or supernova). In a few cases, concerning extra-solar objects, the characteristic \\ga-ray line signature of their radioactive decay has been detected and used as a probe of a large variety of astrophysical sites; indeed, \\ga-ray line astronomy with cosmic radioactivities has grown to a mature astrophysical discipline in the last decade. See. e.g. Diehl and Timmes 1998, Arnould and Prantzos 1999, Kn\\\"odlseder and Vedrenne 2001, for recent reviews; also, the proceedings of the {\\it Astronomy with Radioactivities} Conference, organised every two years, nicely reflects the status of that discipline (web site: {\\tt http://www.mpe.mpg.de/gamma/science/lines/workshops/radioactivity.htm } ).  In this review I shall focus on radioactivities produced by massive stars (SNII and WR stars); radioactivities produced by exploding white dwarfs (novae and SNIa) are  reviewed by Hernanz (this volume). ",
        "conclusions": "The aim of {\\it Gamma-Ray Astronomy with Radioactivities}, as explicitly defined by the ``founding fathers'' of the field in the 60ies (see Sec. 2) was to probe stellar nucleosynthesis as well as supernova structure and energetics. This original aim was reached in a spectacular way in the case of SN1987A (which, however, remains today - and, probably, for sometime in the future -  a unique object in that respect).  On the other hand, the legacy of HEAO-3 and {\\it COMPTEL} \\ set new aims to the field of {\\it Gamma-Ray Astronomy with long-lived Radioactivities}: to probe the large-scale distribution of active nucleosynthesis sites in the Galaxy and the properties/history of any clusterings in that distribution (young stellar associations, individual objects). {\\it INTEGRAL} is expected to perform this next step."
    },
    "0209/astro-ph0209629_arXiv.txt": {
        "abstract": "The orbital period of Scorpius X-1 has been accepted as 0.787313 d since its discovery in archival optical photometric data by Gottlieb, Wright \\& Liller (1975). This period has apparently been confirmed multiple times in the years since in both photometric and spectroscopic optical observations, though to date only marginal evidence has been reported for modulation of the X-ray intensity at that period. We have used data taken with the All Sky Monitor on board the {\\it Rossi X-Ray Timing Explorer } over the past 6 years to search for such a modulation.  A major difficulty in detecting the orbit in X-ray data is presented by the flaring behavior in this source, wherein the (1.5-12 keV) X-ray intensity can change by up to a factor of two within a fraction of a day.  These flares contribute nearly white noise to Fourier transforms of the intensity time series, and thereby tend to obscure weak modulations, i.e., of a few percent or less.  We present herein a new technique for substantially reducing the effects of the flaring behavior while, at the same time, retaining much of any periodic orbital modulation, provided only that the two temporal behaviors exhibit different spectral signatures. Through such a search, we have found evidence for orbital modulation at the $\\sim 1\\%$ level with a period of 0.78893 d, equal within our accuracy to a period which differs by 1 cycle per year from the accepted value. If we compare our results with the period of the 1 year sideband cited by Gottlieb et al. we conclude that the actual period may in fact be 0.78901 d.  Finally, we note that many of the reported optical observations of Sco X-1 have been made within one or two months of early June, when Sco X-1 transits the meridian at midnight.  All periodicity searches based only on such observations would have been subject to the same 1 cycle per year alias that affected the search of \\citet{GWL75}. ",
        "introduction": "Scorpius X-1 is the most prominent of the bright Galactic bulge X-ray sources. These are persistent, high-luminosity X-ray sources located in directions more or less towards the Galactic center, and are typified by, in addition to Sco X-1, GX340+0, GX349+2, GX3+1, GX5--1, GX9+1, GX9+9, GX13+1, and GX17+2.  The 2 to 10 keV intensities of these sources are often seen to vary within a range limited to about a factor of two or so; much larger changes in strength are rare. Neither eclipses nor persistent periodic pulsations have been detected in any of them.  These sources are believed to be low-mass X-ray binaries (LMXBs), binaries with short orbital periods in which accretion onto a neutron star primary is fed by Roche lobe overflow from a companion less massive than the Sun.  Two classes of LMXBs, the ``Z'' and ``atoll'' types, are distinguished by distinctive tracks seen in X-ray color-color diagrams.  In both types of source, the overall intensity and features of Fourier power density spectra, including quasiperiodic oscillations, also change in a characteristic way with location along the color-color track \\citep{Has89,vdK00}. The brightest bulge sources are of the Z-type, which may be emitting close to the Eddington limit for a neutron star of mass $\\sim 1.4~M_\\odot$. Although Sco X-1 is by far the brightest persistent source, it is not likely to be intrinsically more luminous than the other Z sources; rather it is thought to be $\\sim 3$ times closer than most of the bulge sources \\citep{BFG99}.  Sco X-1 exhibits only the so-called normal and flaring branches of the Z-source X-ray color-color diagram; it does not exhibit a ``horizontal branch'' \\citep{Has89}.  During frequent excursions onto the flaring branch, the source intensity increases by up to a factor of $\\sim 2$. These flares often come and go on time scales of hours \\citep[and references therein]{Bra75}.  The high level of activity that Sco X-1 displays in X-rays also appears in the optical and radio bands where the variability is correlated with the variability in the X-ray band \\citep{San66,Bra75,Can75,WGL75}.  In the optical, it ranges over blue magnitudes of 11.8 to 13.6, but on time scales of 1 day or less it rarely changes by more than 1 magnitude.  Apart from strong X-ray emission and occasional weak radio emission, the binary nature of the bright Galactic bulge sources is only subtly, if at all, manifested observationally.  Of the 9 sources listed above, orbital periods have been determined for only Sco X-1, GX 9+9, and, possibly, GX349+2 (Gottlieb, Wright \\& Liller 1975, hereafter GWL; Hertz \\& Wood 1988; Schaefer 1990; Wachter \\& Margon 1996).  The orbital period of Sco X-1 was first found by GWL in a search for periodicities in 85 years of photometric optical data obtained from plates in the Harvard collection. GWL found four distinct candidates for the orbital period, viz., 0.787313 d, 0.78901 d, 0.81069 d, and 3.74001 d. They identified 0.787313 d as the likely orbital period, while the other periods were explained as one year, one month, and one day sidebands, respectively, i.e., as artifacts of the limited coverage of the observations.  Use of a large number of measurements was critical to GWL's success in elucidating these periodicities against the background of high frequency noise from random variations in brightness; in the optical the orbital modulation has a full amplitude of only $\\sim 0.2$ magnitude \\citep{GWL75,WGL75}.  \\citet{CC75} carried out spectroscopic observations of Sco X-1 and found radial velocity variations in the He II and H emission lines with a period of $0.787 \\pm 0.006$ d and thereby conclusively demonstrated the orbital nature of the 0.787 day variations.  If the emission lines are formed, e.g., in an accretion disk around the X-ray emitting star, then the epoch of minimum brightness found by GWL corresponds to superior conjunction of the X-ray source \\citep{CC75,Cra76}.  The presence of a 0.787 d period has been confirmed in a number of other studies based on optical photometry (e.g., van Genderen 1977; see Augusteijn et al. 1992) or on radial velocities obtained from optical spectroscopy (e.g., Bord et al. 1976; LaSala \\& Thorstensen 1985).  Further information on the binary system has recently been obtained by \\citet{SC02}, who used optical spectroscopy to uncover evidence of the donor star in the binary system and to obtain estimates of its radial velocity amplitude.  No significant periodicity at 0.787313 d or any nearby period in the X-ray flux from Sco X-1 has been reported.  \\citet{PH87} searched for a signal at the GWL period in 5 years of 3-6 keV intensity measurements from the All-Sky Monitor on {\\it Ariel V } and found only a weak modulation with an amplitude of $\\sim 0.4$\\%. Priedhorsky \\& Holt did not report any investigations of variability at other periods.  We have analyzed X-ray intensities of Sco X-1 recorded over more than six years by the All Sky Monitor (ASM) on board the {\\it Rossi X-ray Timing Explorer}, and have found evidence of a periodicity with a period of 0.78893 d. This period is consistent with GWL's 0.78901 d period, and we therefore tentatively propose that the latter value is the orbital period of Sco X-1, while the 0.787313 d period is a one-year sideband produced as a consequence of the observational window function. ",
        "conclusions": "We find evidence for a periodicity in the X-ray flux from Sco X-1 with a period of $0.78893 \\pm 0.00010$ d, and identify this as likely caused by some physical effect producing modulations of the flux at the orbital period.  We do not find good evidence for variability at the 0.787313 d period.  In fact, GWL found a periodicity at 0.78901 d that is consistent with the period we find in the ASM data, but since they found that periodicity to be $\\sim 20$\\% weaker than that at 0.787313 d, they concluded that the latter was the orbital period while the former was the 1-year sideband produced by the observational window function.  We suggest rather that the true orbital period is 0.78901 d and that the 0.787313 d period is the 1-year sideband. Unfortunately, we are unable to evaluate quantitatively the likelihood of the sideband appearing stronger in the photometric data than the true modulation frequency since we do not have access to the plate stack data used by GWL.  We note, however, that ASM observations of Sco X-1 are generally available for more than 10 months each year while most ground-based optical observations of Sco X-1 were likely taken within a couple of months of early June when Sco X-1 is nearly opposite from the Sun. Thus, it is likely that the window function of the optical observations analyzed by GWL would have strong components at a 1-year period.  We further note that a modulation with a period of 0.78901 d that is in phase with a modulation at 0.787313 d as described by GWL in early June would be difficult for optical observers to distinguish without observations at other times of the year.  For example, the two oscillations would be out of phase by only 0.08 cycles in early May or early July.  At even earlier or later dates, the phases of the two modulations would gradually drift apart until they came back together (with a difference of 1 cycle) after one year.  We can therefore speculate that there were a limited number of optical observations in the set used by GWL that could have differentiated between the two periods.  The difference between the two periods could well have been missed by independent observers who confirmed the GWL period. Although the optical spectroscopic results of Cowley \\& Crampton (1975) do support the presence of a periodicity, they did not have the frequency resolution to distinguish between the GWL period and a one year sideband.  We also expect that the independent measurements of LaSala \\& Thorstensen (1985) cannot be used to distinguish the GWL period from one which differs by 1 cycle yr$^{-1}$.  Due to the small observed amplitude of the X-ray orbital modulation in Sco X-1 we are unable to investigate either the detailed shape of the light curve or the spectral character of the modulations.  However, it is clear from the lack of higher harmonics in the FFTs and the similarity in amplitudes of the peaks in Fig. 5 that the light curve has no prominent sharp structure nor is it highly energy dependent. Specifically, the modulation cannot be due to photoelectric absorption in unionized matter which would have a much larger effect in the 1.5-3 keV channel than in the 5-12 keV band.  As noted in the previous section, the phase of X-ray minimum appears to be 0.42 cycles earlier than that in the optical.  Based on any simple picture for scattering or absorption in the system, one would expect the optical and X-ray minima to be roughly coincident, unless the outgoing radiation is affected by structure on the rim of the accretion disk which is fixed asymmetrically in the rotating frame of the binary.  At the low inclination angle that we are likely to be viewing Sco X-1 (e.g., $i \\simeq 30^\\circ$, Steeghs \\& Casares 2002) we would not expect this type of effect.  Thus, if our orbital period is correct, we have no ready explanation for the phase of X-ray minimum.  In general, the orbital modulation could be due to attenuation of the beam, i.e., scattering or absorption along the line of sight, or to scattering into the line of sight.  Possible sources of the scattering and/or absorption could include a stellar wind of the companion, the surface of the companion, a wind from the accretion disk, or an accretion disk corona (see, e.g., Priedhorsky \\& Holt 1987). If the modulation takes place near the accretion disk, then some asymmetry around the azimuth of the disk, fixed in the rotating frame of the binary, must be present.  If the donor star serves as a source of scattered radiation (see Priedhorsky \\& Holt 1987), then we can estimate the maximum effect by computing the solid angle subtended by the donor star.  The fractional solid angle is given by $\\sim 0.053 (1+q)^{-2/3}$, where $q$ is the ratio of the mass of the neutron star to that of the donor star.  For plausible masses (see Steeghs \\& Casares 2002) this fraction is $\\sim 2$\\%.  In order for a $\\sim 1$\\% orbital modulation to be due to X-rays scattered into the line of sight from the companion star, this would require an implausibly high X-ray albedo of $\\sim 0.5$.  The accretion disk might afford a larger solid angle for intercepting X-rays and scattering them into the line of sight.  For example, an accretion disk with a half opening angle of $10^\\circ$ would subtend at the neutron star a fractional solid angle of $\\sim$17\\%.  For reasonable X-ray albedos of $\\sim 10$\\% for glancing incidence (see below), this could result in some 1.7\\% of the X-rays emanating from the Sco X-1 system being scattered.  If there is some asymmetry around the disk that yields about a factor of two more scattering at some azimuths than others, this could produce a net orbital modulation of $\\sim 1$\\%.  The X-ray albedo is $\\sim 0.5 \\sigma_e/\\sigma_{pe}$, where $\\sigma_e$ is the angle-averaged electron scattering cross section, and $\\sigma_{pe}$ is the energy-dependent photoelectric absorption cross section appropriate for the composition and ionization state of the scattering medium.  The exact coefficient depends on the geometry experienced by the incident and emerging X-rays.  For material that is of solar composition and nearly unionized, the albedo should scale roughly as $E^{8/3}$; it should be a strongly increasing function of photon energy, $E$, that reaches $\\sim 6$\\% in the middle of the ASM C band.  For more highly ionized plasmas, the albedo can be substantially higher and less energy dependent. The latter property would be relevant in trying to model the nearly energy-independent amplitude we find for the orbital modulation.  A small number of X-ray sources that exhibit orbital modulation have been classified as ``accretion disk corona'' (ADC) sources, e.g., 4U 1822--37 and 4U 2129+47 (see, e.g., White \\& Holt 1982).  The ADC sources are supposedly viewed nearly edge on, with the direct path of X-rays from the neutron star blocked by an accretion disk.  In this case, most of the X-rays detected at the Earth may be scattered by a hot corona located above and below the inner parts of the accretion disk.  The orbital modulation of the observed X-ray flux may then be produced by a thickened disk rim that occults the corona to differing degrees as a function of orbital phase \\citep{WH82}.  The modulation is typically large enough in amplitude, e.g., $\\sim$20\\% of the average flux, to be easily detectable even outside of the relatively narrow partial eclipse.  However, we cannot say whether similar systems observed at much lower inclinations would show modulations at the $\\sim$ 1\\% level, because (1) the neutron star would be visible and (2) the disk rim would not occult the corona.  Finally, to help establish the correct orbital period in Sco X-1 we urge optical astronomers to search for modulations in the light from or radial velocity of this system as far from June as practicable.  Acknowledgements  The authors thank Bill Liller, Ron Remillard, and Ned Wright for very helpful discussions, and Arnout van Genderen and Jan Lub for providing some of their archival optical data from Sco X-1. Partial support for this research was provided by NASA Contract NAS5-30612 and NASA Grant NAG5-9189.  \\begin{deluxetable}{lcccc} \\tablewidth{0pt} \\tablecaption{Intensity vs. Hardness Ratio Parameters for SSC 2\\label{table1}} \\tablehead{ \\colhead{Energy Band} & \\colhead{$\\alpha_k$} & \\colhead{$\\beta_k$} & \\colhead{$\\gamma_k$} & \\colhead{$d\\ln F_k/d\\ln R_{\\rm C,B}$\\tablenotemark{a}} } \\startdata 1.5-3 keV (A) & 140 & 100 & 2.0 & 1.0 \\\\ 3-5 keV (B) & 113 & 141 & 2.0 & 1.3 \\\\ 5-12 keV (C) & 50 & 210 & 2.5 & 2.2 \\\\ 1.5-12 keV (sum) & 500 & 280 & 3.0 & 1.5 \\\\ \\enddata \\tablenotetext{a}{$R_{\\rm C,B}$ is the 5-12 keV band to 3-5 keV band hardness ratio, and $d\\ln F_k/d\\ln R_{\\rm C,B}$ is given for $R_{\\rm C,B} = 1.2$.} \\end{deluxetable}"
    },
    "0209/astro-ph0209379_arXiv.txt": {
        "abstract": "A combination of recent observational results has given rise to what is currently known as the dark energy problem. Although several possible candidates have been extensively discussed in the literature to date the nature of this dark energy component is not well understood at present. In this paper we investigate some cosmological implications of another dark energy candidate: an exotic fluid known as the Chaplygin gas, which is characterized by an equation of state $p = -A/\\rho$, where $A$ is a positive constant. By assuming a flat scenario driven by non-relativistic matter plus a Chaplygin gas dark energy we study the influence of such a component on the statistical properties of gravitational lenses. A comparison between the predicted age of the universe and the latest age estimates of globular clusters is also included and the results briefly discussed. In general, we find that the behavior of this class of models may be interpreted as an intermediary case between the standard and $\\Lambda$CDM scenarios. ",
        "introduction": "From a large number of observational evidence, the currently favoured cosmological model is flat, accelerated and composed of $\\sim 1/3$ of matter (barionic + dark) and $\\sim 2/3$ of a negative-pressure dark component, usually named dark energy or ``quintessence\". The nature of such an unclustered dark energy component, however, is not very well understood at present, giving rise to many theoretical speculations.  Certainly, the most extensively studied explanation for this dark energy problem is the vaccum energy density or cosmological constant ($\\Lambda$) although other interesting possibilities are also alive in the current literature. Some examples are: a very light scalar field $\\phi$, whose effective potential $V(\\phi)$ leads to an accelerated phase at the late stages of the Universe \\cite{peebles}, a X-matter component \\cite{turner}, which is simply characterized by an equation of state $p_x = \\omega_x\\rho_x$, where $-1 \\leq \\omega_x < 0$ and that includes, as a particular case, models with a cosmological constant ($\\Lambda$CDM), a vaccum decaying energy density or a time varying $\\Lambda$-term whose the present value of the cosmological constant ($\\Lambda_o$) is a remnant of the primordial inflationary/deflationary stage \\cite{ozer}, geometrical effects from extra dimensions \\cite{dvali} or still an exotic fluid, the so-called Chaplygin gas, whose equation of state is given by \\begin{equation} p = -A/\\rho, \\end{equation} where $A$ is a positive constant \\cite{kamen}.   All the above mentioned candidates for quintessence have interesting features that make them at some level compatible with the recent obervational facts (see, for example, \\cite{tur,cald,cunha,jailson}). Although most of these scenarios have been extensively explored in the recent literature, in the case of a Chaplygin gas-type dark energy, however, only few analysis have focused attention on its cosmological consequences. From a theoretical viewpoint, an interesting connection between the Chaplygin gas equation of state and String theory has been identified \\cite{bord,jac,bilic}. As explained in \\cite{fabris,bento}, a Chaplygin gas-type equation of state is associated with the parametrization invariant Nambu-Goto $d$-brane action in a $d + 2$ spacetime. In the light-cone parametrization, such an action reduces itself to the action of a Newtonian fluid which obeys Eq. (1) so that the Chaplygin gas corresponds effectively to a gas of $d$-branes in a $d + 2$ spacetime. Moreover, the Chaplygin gas is the only gas known to admit supersymmetric generalization \\cite{jac}.  From the observational viewpoint, it has been argued that the Chaplygin gas may unify the cold dark matter and the dark energy scenarios \\cite{bilic}. The reason for such a belief is the general behaviour of the Chaplygin gas equation of state: it can behave as cold dark matter at small scales and as a negative-pressure dark energy component at large scales. Recently, Fabris {\\it et al.} \\cite{fabris1} analysed a cold dark matter plus a Chaplygin gas scenario in the light of type Ia supernovae data (SNe Ia). As a general result, they found a universe completely dominated by the Chaplygin gas as the best fit model. More recently, Avelino {\\it et al.} \\cite{avelino} used a larger sample of SNe Ia and the shape of the matter power spectrum to show that such data restrict the model to a behaviour that closely matches that of a $\\Lambda$CDM models while Bento {\\it et al.} \\cite{bento1} showed that the location of the CMB peaks imposes tight constraints on the free parameters of the model.  The aim of this paper is to explore some other observational consequences of a Chaplygin gas dark energy. We mainly focus our attention on the constraints from statistical properties of gravitationally lensed quasars (QSOs) on the Eq. (1). We also investigate other observational quantities like the deceleration parameter, the acceleration redshift and the expanding age of the the universe. To obtain such results we assume a flat model driven by non-relativistic matter plus a Chaplygin gas dark energy component (from now on CgCDM).  This paper is organized in the following way. In Sec. II the field equations and distance formulas are presented. We also derive the expression for the deceleration parameter and discuss the redshift at which the accelerated expansion begins. The predicted age of the Universe in the context of CgCDM models is briefly discussed in Sec. III. We then proceed to analyse the constraints from lensing statistics on these scenarios in  In Sec. IV. We end the paper by summarizing the main results in the conclusion section. ",
        "conclusions": "The search for alternative cosmologies is presently in vogue and the leitmotiv is the observational support for an accelerated universe provided by the SNe Ia results. In general, such alternative scenarios contain an unkown negative-pressure dark component that explains the SNe Ia results and reconciles the inflationary flatness prediction ($\\Omega_{\\rm{T}} = 1$) with the dynamical estimates of the quantity of matter in the Universe ($\\Omega_{\\rm{m}} \\simeq 0.3 \\pm 0.1$). In this paper we have focused our attention on another dark energy candidate: the Chaplygin gas. We showed that the predicted age of the Universe in the context of CgCDM models is compatible with the most recent age estimates of globular clusters for values of $\\Omega_{\\rm{m}} \\simeq 0.2$ and $A_s \\geq 0.96$. We also studied the influence of such a component on the statistical properties of gravitational lensing. At $1\\sigma$ level we found that a large class of these scenarios is in agreement with the current lensing data with the maximum of the likelihood function (Eq. \\ref{LLF}) located at $\\Omega_{\\rm{m}} = 0.4$ and $A_s = 1.0$. As a general result, the predicted number of lensed quasars requires $\\Omega_{\\rm{m}} \\leq 0.45$ and $A_s \\geq 0.72$."
    },
    "0209/astro-ph0209009_arXiv.txt": {
        "abstract": "{ We detected four outbursts of V359 Cen (possible nova discovered in 1939) between 1999 and 2002.  Time-resolved CCD photometry during two outbursts (1999 and 2002) revealed that V359 Cen is actually a long-period SU UMa-type dwarf nova with a mean superhump period of 0.08092(1) d.  We identified its supercycle length as 307--397 d.  This secure identification of the superhump period precludes the previously supposed possibility that V359 Cen could be related to a WZ Sge-type system with a long persistence of late superhumps.  The outburst characteristics of V359 Cen are, however, rather unusual in its low occurrence of normal outbursts. ",
        "introduction": "Cataclysmic variables (CVs) are close binary systems consisting of a white dwarf and a red dwarf secondary transferring matter via the Roche lobe overflow (for a review of CVs, see \\cite{war95book}).  CVs are subdivided into several categories, including dwarf novae (DNe) and novae.  Both DNe and novae are characterized by the presence of a sudden increase of brightness (outburst).  Although the mechanisms of DN-type outbursts (cf. \\cite{osa96review}) and nova outbursts (cf. \\cite{sta87novareview,sta99novareview,sta00novareview}) are different, observational discrimination between rarely outbursting DNe and novae can be sometimes difficult (see \\cite{dow81wzsge} and \\cite{kat01hvvir} for classical and recent examples, respectively). Since rarely outbursting DNe can be easily confused with very fast novae, these confusions may have skewed our statistical view of classical novae \\citep{dow86novadensity,lil87novarate,sha97novarate}.  A large fraction of such confusions turned out to be SU UMa-type dwarf novae or WZ Sge-type dwarf novae \\citep{kat01hvvir}.  SU UMa-type dwarf novae are a subclass of DNe.  WZ Sge-type dwarf novae are still enigmatic, both in theory and to observations, SU UMa-type dwarf novae, which very infrequently (once in $\\sim$10 yr) show large-amplitude ($\\sim$8 mag) outbursts \\citep{bai79wzsge,dow81wzsge,pat81wzsge,odo91wzsge}. All SU UMa-type dwarf novae, including WZ Sge-type dwarf novae, show superhumps during their long, bright outbursts (superoutbursts). [For a recent review of dwarf novae and SU UMa-type dwarf novae, see \\citet{osa96review} and \\citet{war95suuma}, respectively.]  Superhumps have periods (superhump period: $P_{\\rm SH}$) a few percent longer than the orbital periods ($P_{\\rm orb}$) \\citep{vog80suumastars,war85suuma}, which is believed to be a consequence of the apsidal motion \\citep{osa85SHexcess,mol92SHexcess} of a tidally induced eccentric accretion disk \\citep{whi88tidal,hir90SHexcess,lub91SHa}.  WZ Sge-type dwarf novae are known to show a different kind of (super)humps during the earliest stage of superoutbursts \\citep{kat96alcom,mat98egcnc,ish02wzsgeletter,osa02wzsgehump,kat02wzsgeESH}. These (super)humps in WZ Sge-type dwarf novae have periods close to $P_{\\rm orb}$, which can be easily distinguished from usual SU UMa-type superhumps. The presence of superhumps thus provides a powerful photometric tool in discriminating novae and SU UMa-type/WZ Sge-type dwarf novae once an object undergoes another outburst.  V359 Cen was originally discovered as a possible nova by A. Opolski (see \\cite{due87novaatlas}).  The object was visible on 19 plates taken between 1939 April 20 and 27, and the recorded maximum was m$_{pg}$ = 13.8 \\citep{due87novaatlas}.  After an examination of Harvard plates of the corresponding epoch and Opolski's finding chart, \\citet{due87novaatlas} suggested a 21.0 mag quiescent counterpart. The true nature of the object, however, remained uncertain.  The object was even proposed to be a nova in the Galactic halo.  From distant nova candidates, \\citet{kat01hvvir} selected V359 Cen as a candidate for a rarely outbursting dwarf nova.  A finding chart of the proposed quiescent counterpart was presented in \\citet{due87novaatlas}.  \\citet{mun98CVspec5} tried to study the proposed quiescent counterpart spectroscopically, but the attempt failed because of its faintness ($V$ fainter than 20.5).  \\citet{gil98v359cen} obtained a deep image around V359 Cen, and showed that the profile is indistinguishable from that of a normal star; there was no evidence of a nova shell.  The situation dramatically changed when one of the authors (Rod Stubbings) detected the second historical outburst on 1999 July 13 (vsnet-alert 3216).\\footnote{ http://www.kusastro.kyoto-u.ac.jp/vsnet/Mail/alert3000/\\\\msg00216.html. }  The object further underwent outbursts in 2000 May, 2001 April and 2002 June.  We photometrically observed two outbursts (1999 July and 2002 June) and revealed that V359 Cen is an SU UMa-type dwarf nova. \\citet{wou01v359cenxzeriyytel} obtained time-resolved CCD photometry following the 1999 July outburst and detected a periodicity of 0.0779 d (112 min), but interpretation of this period remained rather uncertain. ",
        "conclusions": "We detected four outbursts of V359 Cen (possible nova discovered in 1939) between 1999 and 2002.  Time-resolved CCD photometry during two outbursts (1999 and 2002) revealed that V359 Cen is actually a long-period SU UMa-type dwarf nova with a mean superhump period of 0.08092(1) d.  We identified its supercycle length as 307--397 d.  This secure identification of the superhump period precludes the previously supposed possibility that V359 Cen could be related to a WZ Sge-type system with a long persistence of late superhumps.  The outburst characteristics of V359 Cen are, however, rather unusual in its low occurrence of normal outbursts.  The fractional superhump excess is 3.9--4.1\\%, which suggests that V359 Cen should have a normal binary mass ratio in spite of its rather unusual outburst characteristics.  We also obtained a secure identification of the quiescent counterpart and discussed on the possibility of high/low state changes. The evolution of superhumps and their period change was closely followed. We also detected super-QPO-type variation (period $\\sim$0.02 d) during the earliest stage of the 2002 superoutburst.  \\vskip 3mm  The authors are grateful to Andrew Pearce who reported visual observations of V359 Cen to VSNET. This work is partly supported by a grant-in aid (13640239) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. Part of this work is supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists (MU). The CCD operation of the Bronberg Observatory is partly sponsored by the Center for Backyard Astrophysics. The CCD operation by Peter Nelson is on loan from the AAVSO, funded by the Curry Foundation. This research has made use of the Digitized Sky Survey producted by STScI, the ESO Skycat tool, the VizieR catalogue access tool, and the USNOFS Image and Catalogue Archive operated by the United States Naval Observatory, Flagstaff Station (http://www.nofs.navy.mil/data/fchpix/)."
    },
    "0209/astro-ph0209523_arXiv.txt": {
        "abstract": "The Mopra 22m and SEST 15m telescopes have been used to detect and partially map a region of \\12co(1-0) line emission within the Magellanic Bridge, a region lying between the Large and Small Magellanic Clouds.  The emission appears to be embedded in a cloud of neutral hydrogen, and is in the vicinity of an IRAS source.  The CO emission region is found to have a 60\\mum/100\\mum\\ flux density ratio typical for \\12co(1-0) detections within the SMC, although it has a significantly lower \\12co brightness and velocity width. These suggest that the observed region is of a low metallicity, supporting earlier findings that the Magellanic Bridge is not as evolved as the SMC and Magellanic Stream, which are themselves of a lower metallicity than the Galaxy. Our observations, along with empirical models based on SMC observations, indicate that the radius of the detected CO region has an upper limit of \\sm16 pc. This detection is, to our knowledge, the first detection of CO emission from the Magellanic Bridge and is the only direct evidence of star formation through molecular cloud collapse in this region. ",
        "introduction": "The Magellanic Bridge forms a link of primarily neutral hydrogen between the Small and Large Magellanic Clouds (SMC and LMC). Recent high-resolution observations of the \\hi fraction of the Bridge, using the Australia Telescope Compact Array (ATCA) and the Parkes Telescope by Muller et al. (MNRAS, in press), show an intricate morphology, comprising numerous clumps and filaments across all observed spatial scales.  The mechanism responsible for the formation of the Bridge is widely considered to be the gravitational influence of the LMC, and numerical simulations have shown that its formation was possibly triggered during a close pass of the Clouds to each other around 200Myr ago (eg Gardiner \\& Noguchi 1996).   The stellar population of the Bridge has been shown to be relatively young, with an age range of \\sm10 to 25 Myr (Grondin, Demers \\& Kunkel, 1992 and Demers \\& Battinelli, 1998). The young ages imply that star-formation within the Bridge is an active process, yet no previous evidence of star-forming regions through CO emission has been found.  The Magellanic Clouds, as well as the Magellanic Bridge, have been the focus for a number of searches of molecular transition lines. From these searches, a number of different carbon isotopes, as well as other molecular species have been identified.  It appears that the Clouds are rather metal deficient, the SMC particularly so (Israel et al. 1993, and papers of that series). The low CO luminosity of the SMC, being weak and cold in comparison to Galactic Molecular Clouds, has been ascribed to a smaller molecular cloud size which may be, in part, due to a higher UV radiation field causing more thorough photo-disassociation (Rubio et al., 1991, Israel et al., 1993).  These authors suggest that the higher UV field of the SMC is the product of a more active star formation per unit mass than the galaxy, and lower absorptive dust fraction.  Rubio et al. (1993) note that the luminosity and metallicity of CO regions within the SMC are similarly less than those of the Galaxy, although they do not speculate on a reliable and direct correlation between these two parameters. Spectral studies of young, early-type stars throughout the SMC (Rolleston et al., 1993) have shown that the heavy-metal abundance within this SMC is not homogeneous, and is lower than Galactic values.  A few stars within the Bridge and SMC wing have been found to be more metal deficient than star-forming regions in the SMC by \\sm0.5 dex (Rolleston et al., 1999).  More recent spectral studies towards the centre of the Bridge by Lehner et al. (2001) confirm the mismatch of abundances between the SMC and the Magellanic Bridge, and these authors suggest that the Bridge comprises a mix of gas from SMC gas, and from its relatively un-enriched halo.  Israel et al. (1993) made extensive searches for CO emission regions within the SMC towards locations selected from IRAS maps and from known H$\\sc ii$ emission regions.  They found that strong sources of CO emission were generally associated with regions where the ratio of 60\\mum\\ and 100\\mum\\ flux densities (\\mfr) were \\mfr\\lesim1.0, and that sites of CO emission where \\mfr\\gesim1.0 were relatively weak. In general, SMC CO line emissions were found to be weaker and narrower than those of Galactic CO emission.  At this stage, CO emission of any kind has not previously been reported in the Magellanic Bridge further east than the molecular cloud N88, which occupies the north eastern corner of the SMC.  N88 has also been observed in a number of other molecular lines (Testor et al., 1999). Smoker et al. (2000) have conducted a search for \\12co(1-0) near the centre of the Magellanic Bridge, towards a region of cold atomic hydrogen (Kolbunicky \\& Dickey 1999).  This target was considered a likely candidate based on studies by Garwood \\& Dickey (1989), who had found that CO emission regions were occasionally associated with cold atomic gas.  These SEST observations by Smoker et al.  however, showed no \\12co(1-0) emission down to an RMS of 60 mK.  Maps of CO emission from the tidally affected M81 galaxy have shown that some CO regions can be associated with tidally extruded \\hi (Taylor, Walter \\& Yun, 2001), although the correlation is not particularly outstanding and the CO appears to only loosely trace the \\hi mass.  CO emission in Tidal Dwarf Galaxies (TDGs) has been found to generally correlate with regions of high \\hi column density by Braine et al. (2001), although only single pointings were made during this study. TDGs are thought to condense from the remnant material exported from a host galaxy during a tidal stripping event (eg Braine et al. 2001). They may represent a class of objects where star-formation has preceded significant tidal perturbation. This is the reverse of the sequence of processes thought to be active in the Magellanic Bridge and is a relevant benchmark to bear in mind.  This paper presents results from a CO survey of a region in the Magellanic Bridge using selection criteria based on findings by Israel et al. (1993), Taylor, Walter \\& Yun (2001) and Braine et al. (2000, 2001).  We discuss the selection procedure of candidate CO emission sites in Section \\ref{sec:selection}. Observation techniques are outlined in Section \\ref{sec:obs}. In Section \\ref{sec:analysis}, we present the results and comparisons with \\hi data. We discuss the results in Section \\ref{sec:discussion}. \\begin{center} \\begin{table*} \\begin{tabular}{cllccccc}\\hline\\hline Pointing &\\multicolumn{2}{c}{Offset}& \\hi Col. dens.  &\\mfr &Obs.time&RMS (after reduction)&S/N\\\\ &RA&Dec.&cm$^{-2}$&&(min)&(mK)&\\\\\\hline 1 &0.0 & 0.0 &2.7\\exp21&0.18 &157&21&6.5\\\\ 2 &0.0 & $+$45\\asec &2.7\\exp21&0.24&48 &34&6.1\\\\ 3 &$+$45\\asec & $+$45\\asec &2.6\\exp21&0.23&48 &39&-\\\\ 4 &0.0 & +90\\asec &2.5\\exp21&0.26&48 &39&3.1\\\\ \\hline \\hline \\end{tabular} \\caption{Positions, \\hi column densities and \\mfr\\ values for the detected CO cloud, and adjacent positions (See Figure~\\ref{fig:intintmap}). The reference position is at RA (J2000) = 01\\hr56\\min47\\sec, Dec. (J2000) =$-$74\\deg17\\amin41\\asec} \\label{tab:pointstats} \\end{table*} \\end{center} ",
        "conclusions": "\\label{sec:conclusions} We have detected a $^{12}$CO(1-0) emission region within the Magellanic Bridge.  The line width and integrated intensity is compatible with other CO detections within the SMC, although it is somewhat weaker and cooler, implying a lower metallicity.  The findings at this point appear to be consistent with studies of CO emission regions within the SMC where CO has been found to associated with high \\hi integrated intensity and with \\mfr$<$0.2. Estimates made from log-log model fits to SMC CO observations indicate that the Magellanic Bridge CO emission region is spatially small, and has a radius of $<$16 pc.  The CO emission region has a narrow line-width compared with observations of the SMC and with other TDGs, lending support to findings that suggest the Bridge has a lower metallicity than the SMC. At this stage, a reliable estimation of H$_2$ mass is not really possible from the small amount of spatial information, although these issues will be addressed in a future paper.  This finding confirms for the first time that star formation through molecular cloud collapse is an active and current process within the Bridge."
    },
    "0209/astro-ph0209471_arXiv.txt": {
        "abstract": "{We present $JHK$ and 3.8\\micron\\ (\\Lp) photometry of 26 galaxies in the {\\it Infrared Space Observatory} (ISO) Normal Galaxy Key Project (KP) sample and of seven normal ellipticals with the aim of investigating the origin of the 4\\micron\\ emission. The majority of the KP galaxies, and all the ellipticals, have $K-L\\,\\ltrsim\\,1.0$, consistent with stellar photospheres plus moderate dust extinction. Ten of the 26 KP galaxies have $K-L\\,\\gtrsim\\,1.0$, corresponding to a flat or rising 4\\micron\\ continuum, consistent with significant emission from hot dust at 600--1000\\,K. $K-L$ is anticorrelated with ISO flux ratio $F_{6.75}/F_{15}$, weakly correlated with line ratio \\oi/\\cii, but not with \\cii/FIR or IRAS ratio $F_{60}/F_{100}$. Photodissociation-region models for these galaxies show that the hot dust responsible for red $K-L$ resides in regions of high pressure and intense far-ultraviolet radiation field. Taken together, these results suggest that star formation in normal star-forming galaxies can assume two basic forms: an ``active'', relatively rare, mode characterized by hot dust, suppressed Aromatic Features in Emission (AFEs), high pressure, and intense radiation field; and the more common ``passive'' mode that occurs under more quiescent physical conditions, {\\it with} AFEs, and {\\it without} hot dust. The occurrence of these modes appears to only weakly depend on the star-formation rate per unit area. Passive star formation over large scales makes up the bulk of star-forming activity locally, while the ``active'' regime may dominate at high redshifts. ",
        "introduction": "The {\\it Infrared Space Observatory} (ISO) mission has provided an unprecedented view of the interstellar medium (ISM) in galaxies from the near-infrared (NIR) continuum between 3 and 5\\micron\\  to the C$^+$ fine-structure transition at 158\\micron\\ and beyond. Infrared spectra (3 to 12\\micron) obtained by ISO-PHOT reveal the mid-infrared (MIR) emission from {\\it normal galaxies} to be characterized by the Aromatic Features in Emission (AFEs) at 3.3, 6.2, 7.7, 8.6, and 11.3\\micron, and an underlying continuum which contributes about half the luminosity in the 3 to 12\\micron\\ range (Helou et al. \\cite{h2000}).  However, the ISO data have also raised many questions about the ISM constituents. In particular, the nature of the dust responsible for the AFEs is not yet clear, and the debate over three-dimensional Very Small Grains (VSGs), two-dimensional Polycyclic Aromatic Hydrocarbon molecules (PAHs), or Amorphous Hydrogenated Carbon particles (HACs), and their relative contribution in different environments is still underway (Jenniskens \\& D\\'esert  \\cite{jenniskens}; Lu \\cite{lu}; Cesarsky et al. \\cite{cesarsky98}). Moreover, the relationship between the carriers of the AFEs and the underlying continuum remains obscure (Boulanger et al. \\cite{boulanger}, Helou et al. \\cite{h2000}).  A related question is the relative contributions of dust in the ISM, stellar photospheres, and circumstellar dust to the MIR radiation in spirals and ellipticals. IRAS data already convincingly showed that the MIR emission in spirals is mainly due to a continuum from small grains transiently heated to high temperatures and AFEs. On the other hand, in elliptical galaxies the ISM-to-stellar ratio is generally low, so that photospheric and circumstellar emission from evolved red giants may contribute significantly to the MIR (Knapp et al. \\cite{knapp}, Mazzei \\& de Zotti \\cite{mazzei}). It has been argued, though, that the same IRAS color-color relation (e.g., Helou \\cite{h86}) holds for ellipticals and spirals, which implies a similar ISM origin (Sauvage \\& Thuan \\cite{st}).  The continuum component between 3 and 5\\micron\\ can be extrapolated fairly well to the continuum level at 9--10\\micron\\ (Helou et al. \\cite{h2000}), which suggests that some of the 3--5\\micron\\ continuum must be produced by the ISM. Nevertheless, a portion of the 3--5\\micron\\ continuum must arise from stellar photospheres since they dominate the emission at 1--2\\micron . Although the signal-to-noise in the 3--5\\micron\\ ISO spectra is relatively low, that is the spectral region where the MIR spectra of the 43 galaxies observed by Helou et al. (\\cite{h2000}) show the most significant galaxy-to-galaxy variation. To understand the physical origin of this, and assess the relationship of the MIR continuum to the ubiquitous AFEs, it is necessary to separate the ISM contribution from that of stellar photospheres and circumstellar dust. To this end, we have acquired $JHK$\\Lp\\ photometry of the galaxies studied in Helou et al. (\\cite{h2000}), which constitutes one of the ISO Normal Galaxy Key Projects. A variety of ISO observations have been secured for this sample, including mid-infrared spectra (Helou et al. \\cite{h2000}), ionic and atomic fine-structure line fluxes (Malhotra et al. \\cite{m1997}, Malhotra et al. \\cite{m2001}), mid-infrared maps at 7 and 15\\micron, and a small number of 4.5\\micron\\ images (Dale et al. \\cite{d2000}).  The rest of the paper is organized as follows: Sect. \\ref{obs} describes the observations, reduction, and photometric calibration. We compare the $JHKL$ colors of the sample galaxies with their ISO properties in Sect. \\ref{iso}, and with photodissociation-region models in Sect. \\ref{pdr}. Section \\ref{nature} discusses the nature of the 4\\,\\micron\\ continuum, and our interpretation of these results in terms of ``active'' and ``passive'' star formation. \\begin{figure*} {\\rotatebox{180}{\\includegraphics*[100,100][570,820]{MS3010tab1.eps}}} \\label{tbl_phot} \\end{figure*} ",
        "conclusions": "} With our $JHK$\\Lp\\ photometry, we have analyzed the 4\\micron\\ continuum and its relation with the MIR spectrum and FIR emission lines. We find the following: \\begin{enumerate} \\item The majority of the 26 KP galaxies have a falling 4\\micron\\ continuum, $K-L\\,<\\,1.0$, consistent with stellar photospheres and moderate dust extinction. 10 of them have a flat or rising 4\\micron\\ continuum, $K-L\\,\\gtrsim\\,1.0$, consistent with a measurable fraction of 600--1000\\,K hot dust. \\item $K-L$ is anticorrelated with ISO ratios $F_{6.75}/F_{15}$ and IRAS ratio $F_{12}/F_{25}$, but only weakly with \\cii/\\oi, and not at all with \\cii/FIR/ or IRAS ratio $F_{60}/F_{100}$. \\item PDR models for these galaxies show that the hot dust measured by red $K-L$ is associated with high pressures and intense far-ultraviolet radiation fields in compact ($\\ltrsim\\,$100\\,pc) regions. \\item These results taken together suggest that star formation in these galaxies occurs in two ``extreme forms'': \\begin{enumerate} \\item a relatively rare ``active'' mode characterized by hot dust, suppressed AFEs, high pressure, intense ultraviolet radiation field, and compact size; \\item a more common ``passive'' mode characterized by photospheric $K-L$ colors, with moderate extinction, and less extreme physical conditions. \\end{enumerate} \\item The physical conditions we infer for the star-forming regions containing hot dust are similar to those created by interactions and mergers. We speculate that such intense episodes may have been more common in the past, so that the ``active'' regime could dominate star formation at high redshift. \\end{enumerate}"
    },
    "0209/astro-ph0209192_arXiv.txt": {
        "abstract": "We extend existing work on the propagation of ultra-high energy cosmic rays in extragalactic magnetic fields to a possible component of heavy nuclei, taking into account photodisintegration, pion production, and creation of $e^{\\pm}$ pairs. We focus on the influence of the magnetic field on the spectrum and chemical composition of observed ultra-high energy cosmic rays. We apply our simulations to the scenarios proposed by Anchordoqui {\\it et al.}, in which Iron nuclei are accelerated in nearby starburst galaxies, and show that it is in marginal agreement with the data. We also show that it is highly unlikely to detect He nuclei from M87 at the highest energies observed $\\sim3\\,10^{20}\\,$eV as required for the scenario of Ahn {\\it et al.} in which the highest energy cosmic rays originate from M87 and are deflected in a Parker spiral Galactic magnetic field. ",
        "introduction": "The origin of ultra-high energy cosmic rays (UHECR) is one of the major open questions in astro-particle physics. Data from the Fly's Eye experiment \\cite{Fly} suggest that the chemical composition is dominated by heavy nuclei up to the ankle ($E\\simeq10^{18.5}\\,$eV) and then progressively by protons beyond, while other data \\cite{Hayashida} may suggest a mixed composition of both protons and heavier nuclei.  The fact that present experiments do not give a clear answer to the question of chemical composition of primary particles motivates to test scenarios with a heavy component.  Nucleons cannot be confined in our galaxy at energies above the ankle; together with the absence of a correlation between their arrival directions and the galactic plane, this suggests that if nucleons are primary particles they should have an extragalactic origin.  At the same time, nucleons at energies above $\\simeq4\\times 10^{19}$ eV interact with the photons of the Cosmic Microwave Background (CMB) by photopion production; this would predict a break in the cosmic ray flux, the so-called GZK cut-off~\\cite{gzk}, and the sources of UHECR above the GZK cut-off should be nearer than about $50\\,$Mpc.  The GZK cut-off has not been observed by the experiments such as Fly's Eye~\\cite{Fly}, Haverah Park~\\cite{Haverah}, Yakutsk~\\cite{Yakutsk}, and AGASA~\\cite{AGASA}. However, currently there seems to be a disagreement specifically between the AGASA ground array~\\cite{AGASA} which detected about 10 events above $10^{20}\\,$eV, as opposed to about 2 expected from the GZK cut-off, and the HiRes fluorescence detector~\\cite{hires} which seems consistent with a cut-off~\\cite{wb}. The resolution of this problem may have to await the completion of the Pierre Auger project~\\cite{auger} which will combine the two existing complementary detection techniques.  In the acceleration scenario, UHECR can achieve these extreme high energies by acceleration in shocked magnetized plasmas in powerful astrophysical sources, such as hot spots of radio galaxies and active galactic nuclei\\cite{biermann}.  Attributing sources to the highest energy events is complicated by the lack of observed counterparts~\\cite{ssb,ES95}. A possible explanation is the existence of large scale intervening magnetic fields with intensities $B\\sim0.1-1\\,\\mu$G~\\cite{ES95}, which would provide sufficient angular deflection even for high energies and could explain the large scale isotropy of arrival directions observed by the AGASA experiment~\\cite{AGASA} as due to diffusion. In this framework, the clusters of events seen by the AGASA and Yakutsk experiments~\\cite{AGASA,TT} are interpreted as due to focussing of the highest energy cosmic rays in caustics of the extra-galactic magnetic fields, as originally suggested in Ref.~\\cite{LSB99} (see also Ref.~\\cite{HMR99} for nuclei propagating in the Galactic magnetic field and Ref.~\\cite{HMR02} for recent detailed analytical studies). Indeed it has been realized recently that magnetic fields as strong as $\\simeq 1 \\mu G$ in sheets and filaments of large scale structures, such as our Local Supercluster, are compatible with existing upper limits on Faraday rotation~\\cite{vallee,ryu,blasi}.  Heavy nuclei as UHECR primaries are interesting in two ways in this context : they can be accelerated more easily to high energies, as the maximal acceleration energy a particle can achieve depends linearly on its charge ${\\rm Ze}$, and, in addition, the increased deflection (also proportional to ${\\rm Ze}$), could explain more easily the absence of correlation between the arrival direction of the events and the nearest powerful astrophysical objects. However, even in this case there is a limit on the distance to the source because of photodisintegration processes due to the interaction with infra-red and CMB.  The study of the propagation of heavy nuclei in the absence of magnetic deflection has been treated in some detail in the literature. The pioneering work of Puget, Stecker and Bredekamp (PSB in the following)~\\cite{PSB} which included all energy loss mechanisms, has been recently updated~\\cite{Epele,salamon} to take into account new empirical estimates of the infrared background density of photons~\\cite{Malkan} which are about one order of magnitude lower than used by PSB.  In this paper we study the propagation of a distribution of heavy nuclei in a stochastic magnetic field, including all relevant energy loss processes. Our numerical simulations allow to treat in a consistent way the interplay between magnetic deflection and photodisintegration losses. We also keep track of the propagation of all nucleon secondaries produced in photodisintegration events, and propagate these secondaries in the magnetic field. These effects had not been considered in previous studies of UHE nuclei propagation. In particular, we focus here on the influence of the magnetic field on the observable UHECR spectrum and its chemical composition.  In contrast to the sky distribution, these quantities are not significantly influenced by Galactic magnetic fields which we therefore neglect. As will be seen in what follows, a relatively strong magnetic field ($B\\gtrsim 10^{-8}\\,$G), {\\it i.e.} such that UHECR of low energies diffuse, modify by its presence the chemical composition and the energy spectrum recorded at a given distance. This is due to the effect of diffusion, which increases the local residence time differentially with energy, as well as the effective length traveled hence the photodisintegration probability. The interplay between these effects is rather complex, and the output spectrum and chemical composition thus depend on several parameters such as the maximum injection energy, injection spectral index, linear distance and initial chemical composition.  Due to the rather high dimensionality of the parameter space, we will show results for fixed values of the maximum injection energy $E_{\\rm max}=10^{22}\\,$eV and spectral index ${\\rm d}n/{\\rm d}E\\propto E^{-2}$, at the expense of generality, and discuss how the conclusions would be modified for other values of these parameters.  The paper is organized as follows: in section II we describe the propagation of UHE heavy nuclei, in section III we describe our numerical simulation, in section IV we present our results, in section V we apply our results to test the validity of some recent models, and in section VI we conclude. ",
        "conclusions": "In this paper we studied the propagation of a distribution of heavy nuclei in a stochastic magnetic field, including all relevant energy loss processes.  For the propagation in the magnetic field we used the same numerical approach as in Ref.~\\cite{SLB99,LSB99,isola}. This approach was here generalized to heavy nuclei and their photodisintegration processes.  One main conclusion of this paper is that a strong magnetic field, i.e. such that some UHECRs experience a diffusive propagation regime, can strongly modify the chemical composition and energy spectrum at a given energy with respect to what would be seen in the absence of a magnetic field. Rather generically, an increased magnetic field implies a larger effective length travelled, hence a larger photodisintegration probability, hence a chemical composition shifted to ligther species. As we have argued, the extent of this effect also depends on the injection spectrum spectral index, and on the maximal injection energy. If the injection spectrum ${\\rm d}n/{\\rm d}E \\propto E^{-\\alpha}$, then if $\\alpha>2$ the secondary protons produced in photodisintegration interactions do not give a dominant contribution in the low energy observed flux. The converse is not generally true in the case of a strong magnetic field, as the injection spectrum is softened by diffusion.   We applied our results to the discussion of two models recently proposed to explain the origin of UHECR. Our simulations suggest that the model proposed by Anchordoqui et al.~\\cite{anchor}, in which UHECR are iron nuclei accelerated in nearby starburst galaxies, is in relatively good agreement with the data as far as the energy spectrum is concerned. However, it requires a relatively hard injection spectrum ($\\alpha\\simeq1.6$), and the three highest energy events from AGASA and Fly's Eye are $\\gtrsim40^\\circ$ away from the galaxies proposed as sources, in marginal agreement with the expected deflection.  We also showed that for an injection spectrum dominated by Helium nuclei, the relative abundance of Helium compared to nucleons turns out to be smaller than $0.01$ at distances $\\sim20\\,$Mpc from the source. This implies that the scenario of Ahn {\\it et al.}~\\cite{Ahn}, which suggests that the UHECR originate from M87 and are deflected in a powerful Parker spiral Galactic magnetic field, and which requires that the two highest energy cosmic rays (out of 13 above $10^{20}\\,$eV) are He nuclei, is highly fine-tuned."
    },
    "0209/astro-ph0209317_arXiv.txt": {
        "abstract": "Galaxy-galaxy interactions rearrange the baryons in galaxies and trigger substantial star formation; the aggregate effects of these interactions on the evolutionary histories of galaxies in the Universe are poorly understood. We combine $B$ and $R$-band photometry and optical spectroscopy to estimate the strengths and timescales of bursts of triggered star formation in the centers of 190 galaxies in pairs and compact groups.  Based on an analysis of the measured colors and \\ew, we characterize the pre-existing and triggered populations separately.  The best-fitting burst scenarios assume stronger reddening corrections for line emission than for the continuum and continuous star formation lasting for $\\gtrsim$ a hundred Myr.  The most realistic scenarios require an initial mass function that is deficient in the highest-mass stars. The color of the pre-existing stellar population is the most significant source of uncertainty.  Triggered star formation contributes substantially (probably $\\gtrsim 50$\\%) to the $R$-band flux in the central regions of several galaxies; tidal tails do not necessarily accompany this star formation. Many of the galaxies in our sample have bluer centers than outskirts, suggesting that pre- or non-merger interactions may lead to evolution along the Hubble sequence. These objects would appear blue and compact at higher redshifts; the older, redder outskirts of the disks would be difficult to detect. Our data indicate that galaxies with larger separations on the sky contain weaker, and probably older, bursts of star formation on average. However, confirmation of these trends requires further constraints on the colors of the older stellar populations and on the reddening for individual galaxies. ",
        "introduction": "\\setcounter{footnote}{1} \\footnotetext{Hubble fellow.}  \\citet{LT78} showed that galaxies in pairs exhibit a broader scatter in the $U-B$/$B-V$ plane than ``field'' galaxies, demonstrating that interactions trigger bursts of star formation.  Since then, numerous studies, both of statistical samples and of individual interacting systems, have confirmed their results \\citep[e.g.,][]{K84,m86,K87,js89,sw92, Ke93,lk95a,lk95b,Ke96,dz97,BGK00}. These studies have conclusively established the link between interactions and star formation.  However, many questions remain about the true role of interactions in the evolutionary history of galaxies in the Universe.  Because interaction timescales are much shorter than a Hubble time, galaxies may participate in many interactions and/or mergers during their lifetimes. The strengths and durations of typical triggered bursts of star formation are therefore important cosmological parameters. An understanding of interactions is a necessary basis for including interactions and mergers in models of galaxy formation \\citep[e.g.,][]{SP99,Di99,Ka99a,Ka99b}. Studies of local interactions also aid interpretation of the unusual morphologies and structural parameters of intermediate redshift galaxies \\citep[e.g.,][]{K94,A96,BvZ01}.  Finally, a measure of the total effects of interaction may yield constraints on the causes of evolution along the Hubble sequence; large amounts of gas funneled into the centers of the galaxies could result in interaction-induced forms of the processes in secular evolution \\citep{PN90} that cause the formation of ``exponential bulges'' \\citep{AS94,C99}.  \\citet[BGK hereafter]{BGK00} study 502 galaxies in pairs or compact groups selected from the CfA2 redshift survey.  They find a significant correlation between pair separation on the sky, $\\Delta D$, and H$\\alpha$ equivalent width, \\ew.  The correlation also extends to the $\\Delta V$/\\ew\\ plane, where $\\Delta V$ is the pair separation in velocity. BGK argue that the $\\Delta D$ -- \\ew\\ correlation results from the aging of a continuing burst of star formation. If their interpretation is correct, the correlation provides a method of measuring the durations and initial mass functions (IMFs) of the bursts by comparing dynamical timescales with star-formation timescales.  Here we investigate the origin of this correlation by using $B$ and $R$ photometry to constrain the old stellar population and the new burst of star formation separately.  To measure the amount and current age of the new burst of star formation, we explore a method of using the measured colors and \\ew\\ to characterize recent star formation superposed on a pre-existing stellar population. We relate these quantities to the orbital parameters of the pairs, thus exploring the origin of the BGK correlation. In Sec. 2 we describe the pair sample and the data.  Sec. 3 contains a description of the ``two-population'' model we apply to characterize the burst of star formation independent of the underlying older stellar population.  We include a brief discussion of the results of applying the model to the data.   We discuss the origin of the BGK $\\Delta D$ -- \\ew\\ correlation in Sec. 4.  In Sec. 5 we explore the dependence of burst strength and age on the R-band luminosities and rotation speeds of the galaxies.  We describe the galaxies with strong bursts of central star formation in Sec. 6 and conclude in Sec. 7. ",
        "conclusions": "We combine $B$ and $R$ photometry with longslit and central spectroscopy to explore the strengths and ages of tidally-triggered bursts of star formation in galaxies in pairs.  We construct a two-population description of the pre-existing and triggered stellar populations near the centers of the galaxies.  Our primary conclusions are:  \\begin{enumerate}  \\item The most realistic starburst models that describe the data statistically include continuous star formation, a steep (or truncated Salpeter) IMF, and the reddening correction of \\citet{CKS94}. Consistency with the Salpeter IMF requires a very blue distribution of progenitor-galaxy colors.  \\item In our picture, the models require strong triggered bursts in some of the pair galaxies; these bursts constitute $\\gtrsim$50\\% of the central R-band light. This star formation leads to a high incidence of galaxies with blue centers; tidal tails do not always accompany these central bursts of star formation.  \\item  Our results suggest that the strengths, and most likely the ages, of triggered bursts of star formation depend on the galaxy separation on the sky. Thus, although they do not conclusively verify the model, our data are consistent with the picture of a burst of star formation triggered by a close galaxy-galaxy pass that continues and ages as the galaxies move apart. In this picture, the strongest bursts of star formation occur only in the tightest orbits, giving rise to the strength-separation correlations. The data also suggest a possible correlation between the strength of a triggered burst and the rotation speed (mass) of the progenitor galaxy, in the sense that low-mass galaxies experience stronger bursts of triggered star formation. This result supports the hypothesis that the evolution of galaxies is mass-dependent. Verifying these conclusions requires additional constraints on the colors and dust content of the pre-existing stellar populations.   \\end{enumerate}  The colors of the pre-existing stellar populations and the reddening corrections are the dominant uncertainties affecting conclusion (3). More accurate characterization of the correlations between orbit parameters and star formation properties in tidally interacting galaxies requires independent constraints on the colors of the old stellar populations. Near-IR observations can provide more direct measurements of properties of the older stellar population; we plan to report soon on a comprehensive UBVRJHK imaging study of a subset of these galaxies in pairs."
    },
    "0209/astro-ph0209121_arXiv.txt": {
        "abstract": "The GLAST Large Area Telescope~\\cite{GLAST}, scheduled for launch in 2006, is a next generation space based gamma ray telescope which will improve in point source sensitivity by a factor of 30 over that of EGRET~\\cite{EGRET} below 10~GeV, and extend beyond EGRET up to 300~GeV.  Thus GLAST offers a unique opportunity to discover WIMP dark matter through precision studies of gamma rays produced in pair annihilations.  The most dense region of dark matter in our galaxy is currently thought to occur at the center; in particular, dark matter should concentrate within 3~pc of the putative supermassive black hole located at the SgrA* radio source~\\cite{GS}.  In fact, the 2nd and 3rd EGRET catalogs contain a significant point source coincident with the Milky Way galactic center within a resolution of 12~arcminutes~\\cite{MH}.  The EGRET team has determined that the spectral and temporal characteristics of this point source are consistent with dark matter WIMP annihilations.  More detailed analysis \\cite{Silk2} has determined that the magnitude and spectrum of the EGRET source is consistent with relic WIMPs concentrated within 3~pc of the central supermassive black hole.  Furthermore, the SgrA* radio emission is consistent with the synchrotron radiation expected from electrons and positrons produced in WIMP annihilations.  If true, then GLAST should be able to constrain the particle properties of the postulated WIMP with 1 month of data. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209251_arXiv.txt": {
        "abstract": "The collapse of a uniformaly rotating, supermassive star (SMS) to a supermassive black hole (SMBH) has been followed recently by means of hydrodynamic simulations in full general relativity. The initial SMS of arbitrary mass $M$ in these simulations rotates uniformly at the mass--shedding limit and is marginally unstable to radial collapse. The final black hole mass and spin have been determined to be  $M_h/M \\approx 0.9$ and $J_h/M_h^2 \\approx 0.75$. The remaining mass goes into a disk of mass $M_{\\rm disk}/M \\approx 0.1$. Here we show that these black hole and disk parameters can be calculated {\\it analytically} from the initial stellar density and angular momentum distribution. The analytic calculation thereby corroborates and provides a simple physical explanation for the computational discovery that SMS collapse inevitably terminates in the simultaneous formation of a SMBH {\\it and} a rather substantial ambient disk. This disk arises even though the total spin of the progenitor star, $J/M^2 = 0.97$, is safely below the Kerr limit. The calculation performed here applies to {\\it any} marginally unstable $n = 3$ polytrope uniformly rotating at the break--up speed, independent of stellar mass or the source of internal pressure. It illustrates how the black hole and disk parameters can be determined for the collapse of other types of stars with different initial density and rotation profiles. ",
        "introduction": "Recent observations provide increasingly strong evidence that supermassive black holes (SMBHs) of mass $\\sim 10^6 - 10^{10}M_{\\odot}$ exist and that they are the central engines that power active galactic nuclei (AGNs) and quasars (Rees 1998, 2001). The dynamical formation of SMBHs, as well as the inspiral, collision and merger of binary SMBHs, are promising sources of long-wavelength gravitational waves for the proposed Laser Interferometer Space Antenna (LISA) (Thorne 1995; Schutz 2001).  However, the scenario by which SMBHs form is still very uncertain (see, e.g., Rees 1984, for an overview). One promising route is the collapse of a supermassive star (SMS). Once they form out of primordial gas, sufficiently massive stars will evolve in a quasistationary manner via radiative cooling, slowly contracting until reaching the point of onset of relativistic radial instability. At this point, such stars undergo catastrophic collapse on a dynamical timescale,  leading to the formation of a SMBH (Bisnovatyi-Kogan, Zel'dovich \\& Novikov 1966; Zel'dovich \\& Novikov 1971; Shapiro \\& Teukolsky 1983).   Because most objects formed in nature have some angular momentum, rotation is likely to play a significant role in the quasistationary evolution, as well as the final collapse of a SMS. The slow contraction of even a slowly rotating SMS will likely spin it up to the mass--shedding limit, because such stars are so centrally condensed. At the mass--shedding limit, matter on the equator moves in a Keplerian orbit about the star, supported against gravity by centrifugal force and not by an outward pressure gradient. Baumgarte \\& Shapiro (1999) recently performed a detailed numerical analysis of the structure and stability of a rapidly rotating SMS in equilibrium. Assuming the viscous or magnetic braking timescale for angular momentum transfer is shorter than the evolution timescale of a typical SMS (Zel'dovich \\& Novikov 1971; Shapiro 2000), the star will settle into rigid rotation and evolve to the mass--shedding limit following cooling and contraction. They found that all stars at the onset of quasi-radial collapse have an equatorial radius $R \\approx 640GM/c^2$ and a nondimensional spin parameter $a/M \\equiv c J/GM^2 \\approx 0.97$. Here $J$, $M$, $c$, and $G$ are spin, gravitational mass, light velocity and gravitational constant. (Hereafter we adopt gravitational units and set $c=G=1$). Because of the large value of $a/M$, it was uncertain whether the rotating SMS would collapse directly to a black hole and/or form a disk. Because of the growth of $T/|W| \\propto 1/R$ during collapse (where $T$ is the rotational kinetic energy and $W$ is the gravitational potential energy), it was unclear whether the collapse would trigger the growth of bars or other nonaxisymmetric instabilities. We therefore investigated the collapse of a rotating SMS with a $3+1$ post-Newtonian (PN) hydrodynamic simulation (Saijo et al. 2002). We found that the collapse proceeds nearly homologously and axisymmetrically and inferred that a SMBH is likely to be formed, with some gas remaining outside the hole. Shibata \\& Shapiro (2002) then performed simulations of the collapse using an axisymmetric code in full general relativity. We showed conclusively that a black hole forms at the center and determined that the mass of the hole contains about $90\\%$ of the total mass of the system and has a spin parameter $J/M^2 \\sim 0.75$. The remaining gas forms a rotating disk about the nascent hole.  Guided by our numerical simulations,we show here that the final black hole mass and spin can be calculated {\\it analytically} (up to quadrature) from the stellar density and angular momentum profiles in the progenitor SMS. The calculation provides a simple physical explanation for the important finding that a SMBH formed from the collapse of a rotating SMS is always born with a ``ready--made'' disk that can provide fuel for accretion to power an energy source. ",
        "conclusions": "We have considered a SMBH formed by the collapse of a radially unstable, uniformly rotating SMS of mass $M$ spinning at the mass--shedding limit. We calculated that the black hole has a mass $M_h/M \\approx 0.87$ and a spin $J_h/M_h^2 \\approx 0.71$. Most significant, we found that at birth such a  black hole will be  automatically embedded in a substantial disk of mass $M_{\\rm disk}/M \\approx 0.13$. A disk forms even though the total angular momentum of the star at the onset of collapse is $J/M^2 \\approx 0.97$, below the Kerr limit.  We have also calculated the black hole and disk parameters for unstable SMSs which, at the onset of collapse, are spinning more slowly than the angular frequency at break--up (i.e. mass--shedding). We find that the parameters are very insensitive to the ratio $\\alpha = \\Omega/\\Omega_{\\rm shedd}$, where $\\Omega_{\\rm shedd}$ is given by eq.~(\\ref{shedd}), over the range $ 0.1 \\leq \\alpha \\leq 1$. The reason is that for small $\\Omega$, a rotating SMS destabilizes at large $R_p/M$ ($R_p/M \\approx 427/\\alpha^2$) with a value of $J/M^2 \\approx 0.97$ nearly independent of $\\alpha$. The corresponding value of $\\varpi_{\\rm ISCO}/R_p \\approx 0.43$ is nearly unchanged, and as a result, the final black hole and disk parameters do not vary significantly as $\\alpha$ decreases. (For sufficiently low spin rates, the effects of thermal gas pressure compete with rotation to determine the onset of stability and must be included). Nevertheless, we believe, on evolutionary grounds, that it is most likely that a SMS will be spinning near the break--up value at the onset of collapse (Baumgarte \\& Shapiro 1999).  The nascent disk is fairly massive. Such a massive, rotating disk may be dynamically unstable to nonaxisymmetric instabilities. If the disk deforms, it could emit long wavelength, quasi-periodic gravitational waves with a frequency of about $10^{-3} (M/10^6 M_\\odot)^{-1}$ Hz, which might be detectable by LISA.  At birth, the disk will be hot and thick, radiation dominated, and will extend from radius $ \\sim 640 M$ at the outer edge in the equatorial plane all the way down to the horizon. Subsequently, the matter in the disk will likely cool by both photon and neutrino emission and flatten. Under the influence of viscosity, the gas will diffuse outwards in the outermost region, transporting  angular momentum away from the inner region.  At the same time, gas in the inner region will accrete onto the central black hole, increasing its mass and powering a source of luminous energy. The lifetime of the disk $t_{\\rm disk}$ may be estimated crudely by speculating that the accretion eventually will settle into a steady state, generating photons at the Eddington luminosity at $\\sim 10\\%$ efficiency: $L = L_{\\rm Edd} = 1.3 \\times 10^{38} (M_h/M_\\odot)$ and $\\dot{M} = 10L/c^2$. Accordingly, we have $t_{\\rm disk} = M_{\\rm disk}/ \\dot{M}$, or \\begin{equation} t_{\\rm disk} = \\frac{M_{\\rm disk} c^2}{10 L_{\\rm Edd}} \\approx 5 \\times 10^6 \\, {\\rm yrs}. \\end{equation} The above estimate for $t_{\\rm disk}$ is independent of $M$. However, it may be off by a considerable factor: a lower luminosity will increase the disk lifetime, while a lower efficiency (e.g, an efficiency characterizing an ADAF disk; Narayan \\& Yi 1994) or an enhancement in the total luminosity via the addition of neutrino cooling, could decrease the lifetime. Moreover, the disk at black hole birth may comprise far more material than the gas which escapes capture during the final implosion. Prior to undergoing collapse, a rotating SMS will likely evolve along a mass--shedding sequence as it cools and contracts, provided the stellar viscosity and/or magnetic braking proves sufficient to enforce uniform rotation (Baumgarte \\& Shapiro 1999). In this case, the secular cooling and contraction phase will be accompanied by mass loss, and the gas which is ejected then might already form an extensive, ambient disk prior to collapse. Finally we note the possibility that such a massive disk around the hole may suffer a global ``runaway'' instability that may cause a more rapid destruction of the disk than implied by steady-state accretion (Abramowicz et al. 1983; Nishida et al. 1996; Nishida \\& Eriguchi 1996). These are important issues which are ripe for further analysis.  The  calculation performed here applies to {\\it any} marginally unstable $n=3$ polytrope rotating at the mass--shedding limit. The results are independent of stellar mass or the source of internal pressure. The method should be applicable for estimating the  black hole and disk parameters for the collapse of other types of stars with different equations of state, initial density and rotation profiles. For example, a preliminary calculation indicates that for a marginally unstable star uniformly rotating at break--up speed, the mass fraction that avoids collapse and forms a disk falls sharply as the polytropic index of the star drops below $n=3$. The reason is that such stars are more compact at the onset of collapse, with smaller values of $R_p/M$ and, hence, less extended envelopes. This may provide a qualitative explanation for the results of previous simulations in full general relativity for collapse of rotating neutron stars modeled as $n=1$ polytropes (Shibata, Baumgarte \\& Shapiro, 2000; Shibata, 2000). We hope to provide further examples in the future."
    },
    "0209/astro-ph0209298_arXiv.txt": {
        "abstract": "s{Little is presently known about the nature of ultraluminous X--ray sources (ULX). Different hypotheses have been proposed to explain their properties: intermediate--mass BHs, Kerr BHs, young SNR, or background AGN. Some of the current problems and open questions in this research field are here reviewed.} ",
        "introduction": "The study of the X--ray emission from discrete sources in nearby galaxies began with the Einstein satellite~\\cite{GF1}. Soon after, ROSAT contributed with other important steps and drew the attention to those pointlike X--ray sources with high luminosities ($10^{39}-10^{40}$~erg/s)~\\cite{RE,CO}, while ASCA observations produced the first spectra and lightcurves~\\cite{MA}, triggering a number of hypotheses about the nature of these ultraluminous X--ray sources (ULX). The most intriguing hypothesis is that these ULX are powered by accretion around a black hole with masses up to $10^2-10^4$ solar masses. Chandra, with its sub--arcsecond resolution produced a great advancement in this study: in the observation of the Antennae galaxies, 14 ULX were found~\\cite{GF2}. Among these 14 ULX a few could well be background AGN, but certainly not all. One of these ULX has $L_{X}\\approx 10^{40}$~erg/s, similar to low luminosity AGN. It is however located far from the centre of the host galaxy, posing several problems about the evolution of the off--nuclear black holes in nearby galaxies. During the last few years, with the advent of Chandra and XMM--Newton, a lot of efforts have been done to observe and to understand these sources and how they are linked to the host galaxy. Here we review some of the most important problems in this type of research field. ",
        "conclusions": ""
    },
    "0209/astro-ph0209067_arXiv.txt": {
        "abstract": "The objective of this work is twofold: First, we seek evidence for or against the depletion of massive stars in metal-rich starbursts. A second, equally important goal is to perform a consistency test of the latest generation of starburst models in such a high-metallicity environment. We have obtained high-spatial resolution ultraviolet and optical STIS spectroscopy and imaging of the metal-rich nuclear starburst in NGC~3049. The stellar continuum and the absorption line spectrum in the ultraviolet are used to constrain the massive stellar population. The strong, blueshifted stellar lines of CIV and SiIV detected in the UV spectra indicate a metal-rich, compact, massive ($\\sim$10$^6$~M$_\\odot$) cluster of age 3~--~4~Myr emitting the UV-optical continuum. We find strong evidence against a depletion of massive stars in this metal-rich cluster. The derived age and the upper mass-limit cut-off of the initial mass function are also consistent with the detection of Wolf-Rayet (WR) features at optical wavelengths. As a second independent constraint on the massive stellar content, the nebular emission-line spectrum is modeled with photoionization codes using stellar spectra from evolutionary synthesis models. The morphology of the nuclear starburst of NGC~3049 from the STIS images indicates a simple geometry for the nebular emission-line region. However, the nebular lines are badly reproduced by 3~--~4~Myr instantaneous bursts, as required by the UV line spectrum, when unblanketed WR and/or Kurucz stellar atmospheres are used. The corresponding number of photons above 24 and 54~eV in the synthetic models is too high in comparison with values suggested by the observed line ratios. Since the ionizing spectrum in this regime is dominated by emission from WR stars, this discrepancy between observations and models is most likely the result of incorrect assumptions about the WR stars. Thus we conclude that the nebular spectrum of high-metallicity starbursts is poorly reproduced by models for WR dominated populations. However, the new model set of Smith et al. (2002) with blanketed WR and O atmospheres and adjusted WR temperatures predicts a softer far-UV radiation field, providing a better match to the data. ",
        "introduction": "Starbursts are the site where most of the high-mass star formation occurs. They appear in disks, in bulges and in the nuclei of different types (e.g., spirals, irregulars, or dwarfs) of nearby and distant galaxies. Terlevich (1997) defined starburst galaxies as objects in which the energy output of the starburst dominates that of the host galaxy. Thus, starbursts play a major role in the processes of evolution and formation of galaxies due to their high star-formation activity. They constitute ideal laboratories to investigate some key issues: the formation and evolution of massive stars; the feedback between the interstellar medium and the star formation processes; or the star formation and chemical evolution of the universe.  One important question is which stars form in starbursts. The amount of mass transformed into stars and the mass range of the newly formed stars determine the period of time over which a galaxy can support a starburst phase, and therefore the effect of starbursts on the evolution of galaxies. Thus, it is crucial to know if starbursts have an extreme initial mass function (IMF) with respect to more quiescent systems.  Numerous studies performed in the last few years suggest that the IMF has an universal nature, having a slope close to Salpeter for a mass range between 5~M$_\\odot$ and 60~M$_\\odot$ (e.g., references in Gilmore \\& Howell 1998). However, the IMF in starbursts is still not well known, in particular in high-metallicity environments. Contradictory results have been presented. An extreme IMF for starbursts was suggested by Rieke et al. (1980) in a pioneering work on M82. They proposed a {\\it top heavy} IMF (with a deficit of stars below 3~--~8~M$_\\odot$) to explain the mass-luminosity ratio estimated from K-band photometry and a dynamical mass measurement. This IMF includes a higher fraction of red supergiant stars over red giants and dwarfs than expected for a normal IMF. Their results have been confirmed by Rieke et al. (1993). Satyapal et al. (1997) and Foerster-Schreiber (2000) accounted for the K-band luminosity with a Salpeter IMF and ascribe this apparently discrepant result to the complex morphology of M82 and to dust obscuration in the starburst.  The analysis of the nebular optical+near-infrared (IR) lines indicates the suppression of stars more massive than 30~M$_\\odot$ (Goldader et al. 1997; Bresolin, Kennicutt, \\& Garnett 1999;  Coziol, Doyon, \\& Demers 2001). Thornley et al. (2000) used the mid-infrared line ratio of [NeIII]$\\lambda$15.6/[NeII]$\\lambda$12.8 to measure the hardness of the stellar ionizing radiation. They found that, on average, stars with masses above about 40~M$_\\odot$ are not present in starburst galaxies, either because they were never formed or because they already have disappeared as a result of aging effects. A complication of the interpretation of [NeIII]/[NeII] are the uncertain stellar evolution models at high metallicity. As pointed out by Thornley et al., observations of gas and stars in the Galactic Center suggest a  disagreement between the tracks and the observations, which  may be related to the difficulty of defining mass loss, atmospheres, and effective temperatures for metal-rich stars.  On the other hand, the UV-optical continuum from the integrated spectrum of starbursts suggests the formation of such massive stars (Leitherer 1996; Schaerer 2000; Gonz\\'alez Delgado 2001). This technique has the advantage of probing the stellar light from massive O stars directly, as opposed to a nebular analysis which relies on an indirect measure of the stellar radiation. Using WR-star features, Schaerer et al. (2000) find that a Salpeter IMF extending to masses \\mup~$\\geq$~40~M$_\\odot$ is compatible with the WR-star census  observed in metal-rich starbursts. A similar conclusion has been obtained by Bresolin \\& Kennicutt (2002) and by Pindao et al. (2002) based on the detection of WR features in high-metallicity HII regions.   Most of the results on the IMF have been obtained from a limited number of constraints, through the modeling of a few starburst properties at a specific wavelength. These results can depend strongly on the model ingredients. In particular, the evolutionary synthesis models used to analyze the UV+optical+near-IR properties of the starbursts depend on the stellar tracks, atmospheres, and libraries (see the discussion in Schaerer 2000). In addition, the modeling of the nebular lines depends on the assumptions of the gas geometry, the electron density structure, and the chemical composition of the gas. Thus, inclusion of as many different starburst properties as possible is required for a determination of the IMF.  We have obtained UV and optical spectra and images of the starburst galaxy NGC~3049 with Space Telescope Imaging Spectrograph (STIS) on board of the Hubble Space Telescope (HST) to investigate the possible evidence of depletion of massive stars in metal-rich starbursts. NGC~3049 is a barred spiral galaxy, SB(rs)ab, in the Virgo cluster, known as Mrk~710 in the Markarian \\& Lipovetskii (1976) catalog. It is also classified as a nuclear starburst (Balzano 1983), a WR galaxy (Conti 1991) and an HII galaxy (Terlevich et al. 1991). Kunth \\& Schild (1986) first reported the detection of broad emission features at NIII $\\lambda$4640 and HeII $\\lambda$4686 produced by WR stars. More recently, Schaerer, Contini, \\& Kunth (1999) reported the detection of broad CIV $\\lambda$5808 indicating the presence of WC in addition to late WN stars in the starburst. Due to its nebular emission lines and WR features, NGC~3049 has been observed intensively in the past (e.g. Masegosa, Moles, \\& del Olmo 1991; Vacca \\& Conti 1992; Storchi-Bergmann et al. 1994, 1995; Contini 1996; Guseva, Izotov, \\& Thuan 2000). NGC~3049 is  very bright in the far-IR ($L_{\\rm{IR}} = 9 \\times 10^{42}$~erg~s$^{-1}$; Heckman et al. 1998) and at UV wavelengths (Kinney et al. 1993). Its IUE spectrum suggests a young stellar population dominating the central (20$\\times$10 arcsec) emission (Mas-Hesse \\& Kunth 1999). Optical images show extended recent star formation  along the bar (Mazzarella \\& Boronson 1993; Contini et al. 1997; Schaerer et al. 1999). Molecular gas has been detected at the center of the galaxy, but no dense gas as traced by the HCN or CS molecules was found (Contini et al. 1997). The oxygen abundance of the nucleus has been estimated by the {\\it strong line} semi-empirical methods because auroral lines (e.g., [OIII] $\\lambda$4363, [NII] $\\lambda$5755, [SIII] $\\lambda$6312) that are used as electron temperature diagnostics have not been detected in the spectrum. These methods rely, e.g., on $R_{23}$=([OII]$\\lambda$3727+[OIII]$\\lambda$5007)/H$\\beta$ or [NII]$\\lambda$6584/H$\\alpha$ to derive a supersolar oxygen abundance. Vacca \\& Conti (1992) give $12+\\log$(O/H)~=~9.08, Storchi-Bergmann et al. (1994) give 8.87, Guseva et al. (2000) give 9.03, and Contini (1996) gives 9.05\\footnote{These abundances have been estimated using the $R_{23}$ calibration of Edmund \\& Pagel (1984) or the relation between the oxygen abundance and the [NII]$\\lambda$6584/H$\\alpha$ ratio derived by van Zee et al (1998), that was obtained also using the Edmund \\& Pagel calibration}. Even considering the uncertainties associated with these methods, we are confident that the metallicity in the nucleus of NGC~3049 is supersolar ($12+\\log$(O/H)~$ \\geq 8.9$).  NGC~3049 has been chosen for this project  because previous WFPC (pre-Costar) images at UV wavelengths indicate a relatively simple morphology, with the UV flux dominated by one extremely bright star cluster. This morphology minimizes geometry effect and it makes it easier to study the relation between the stellar and the nebular spectrum, and to search for evidence of an extreme IMF in NGC~3049. In addition, our goal is to perform a critical test of the most sophisticated hot-star models currently available to predict the UV stellar properties and the optical nebular lines in metal-rich starbursts. Our work is organized as follows: the observations and data reduction are in Section~2; in Section~3, we describe the morphology of the UV and optical emission; Sections~4 to 6 deal with the analysis and interpretation of the UV light, UV-optical continuum, and nebular optical lines, respectively. The summary and conclusions are in Section~7. ",
        "conclusions": "We have obtained HST ultraviolet and optical STIS spectroscopic and imaging observations of the metal-rich starburst NGC~3049. These data are interpreted using evolutionary synthesis models optimized for star forming regions, which allow us to constrain the stellar content of the nuclear starburst. From the analysis of the UV-optical continuum and the optical emission lines we have obtained the following results:    \\begin{itemize}    \\item{The nuclear starburst in NGC~3049 is very compact. The UV-optical continuum is dominated by a central cluster that is unresolved in the UV and optical STIS images. The FWHM of the spatial profiles is only $\\sim$0.1~arcsec, corresponding to 9~pc for a distance of 18 Mpc.}    \\item{The central UV flux shows strong wind (CIV, SiIV, NV) and photospheric lines (OV $\\lambda$1371, FeV $\\lambda$1360-1380, SiIII $\\lambda$1417, CIII $\\lambda$1427, and SV $\\lambda$1501). These lines suggest a powerful starburst. The modeling of the continuum and stellar absorption lines indicates that a high-metallicity ($Z=0.02-0.04$) 3~--~4~Myr old instantaneous burst with a Salpeter IMF and \\mup~=~100~M$_\\odot$ fits  the central 0.5~arcsec spectrum observed in the UV. The mass estimated for the cluster is $\\sim$10$^6$~M$_\\odot$, if the extinction toward the cluster is $E(B-V)=0.2$. The reddening-corrected luminosity of this cluster accounts for almost half of the far-IR luminosity of NGC~3049. The massive stars in this cluster provide enough photons ($\\log Q= 52.30$ (ph~s$^{-1}$)) to explain the observed nuclear H$\\alpha$ luminosity. Models with very few massive stars (e.g., with \\mup~$\\leq40$~M$_\\odot$) are not able to fit the UV stellar lines. The analysis of the central 1.2~arcsec UV flux also indicates that there is no significant change of the extinction and a very small age spread.   }    \\item{The central 0.3~arcsec in the optical shows WR features at 4660~\\AA. These observations confirm previous results (Schaerer et al 1999; Guseva et al 2000) that late WN and early and late WC stars are the dominant WR stellar population. The total luminosity of the bump is $9 \\times 10^{38}$~erg~s$^{-1}$ if the stellar reddening is $E(B-V)=0.2$, and the Ew of the bump is 14~\\AA. This luminosity is provided by $\\sim$275 WR stars. A  WR/O ratio equal to 0.065 is estimated for the cluster. This ratio and the Ew of the bump is in very good agreement with the predictions of the evolutionary models for a 3~--~4~Myr old high-metallicity starburst. Therefore the number of WR stars is correctly predicted by the stellar evolution models.}    \\item{We have not detected any absorption lines that could indicate the presence of red supergiants, or an intermediate or old stellar population. The blue optical continuum of the central 0.5$\\times$0.1~arcsec is well fitted by a 3~Myr old instantaneous burst with $E(B-V)=0.2$, as predicted by the UV continuum. Thus, no additional older stellar population contributes significantly to the nuclear optical continuum.}    \\item{The Ew of the H$\\beta$ lines is only 7~\\AA. This is lower than the values of 200~--~60~\\AA\\ predicted by a 3~--~4~Myr instantaneous burst model. However, this discrepancy is naturally explained if the ionized gas is more extended than the compact cluster.}    \\item{Two kinematic components are resolved in the H$\\alpha$ emission line. These two components are extended by $\\sim$30~pc south-west of the maximum of the continuum, and by $\\sim$8~pc at $\\sim$50~pc north-east. The second component is blueshifted $\\sim$60~\\kms\\ with respect to brightest main component. It may represent a bubble powered by the winds from massive stars. Additional evidence for an outflow of the interstellar medium comes from the blueshift of 200~\\kms\\ of the UV interstellar lines with respect to the radial velocity reported by NED of 300~\\kms\\ with respect to the nebular optical radial velocity.}   \\item{The modeling of the nebular emission lines failed to constrain the hot stellar content of the nuclear starburst of NGC~3049. The models which predict emission-line ratios in agreement with the observations contain very few massive stars. This result, which is clearly in contradiction with the UV spectrum, would be in line with previous suggestions for a depletion of massive stars in the IMF of metal-rich starbursts. However, this conclusion is an artifact of the failure of the population synthesis models to predict the ionizing radiation field. Most likely, the overly hard radiation field results from the failure of the currently available WR atmosphere grid. The new model set of Smith et al. (2002) includes blanketing and uses different temperature and density structures. These models applied to NGC~3049 predict a softer far-UV radiation field and  provide a better match to the data.   The CoStar and SB99 models predict HeI $\\lambda$5876/H$\\beta$ and [OIII]/[OII] ratios that are higher than observed if the ionizing radiation is provided by a 3~--~4~Myr old cluster formed with a Salpeter IMF and \\mup~=~100~M$_\\odot$. SEDs using only the Kurucz atmospheres can adequately predict the collisional line ratios but they underpredict $Q$(He)/$Q$(H). Thus extended atmospheres are required but the currently available pure He WR models are inadequate to reproduce the nebular spectrum of high-metallicity starbursts in the WR phase.}    \\end{itemize}    {\\bf Acknowledgments}    We thank Gary Ferland for kindly making his code available, Enrique P\\'erez and Miguel Cervi\\~no for their comments from a thorough reading of the paper, Daniel Schaerer, the referee, for his detailed and useful report, and Linda Smith for sending us their paper in advance of publication, and making their new version of SB99 available. This work was supported by Spanish projects AYA-2001-3939-C03-01 and by HST grants GO-7513.01-96A from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555.    \\clearpage"
    },
    "0209/astro-ph0209584_arXiv.txt": {
        "abstract": "The fact that the $\\Lambda$CDM model fits the observations does not necessarily imply the physical existence of `dark energy'. Dropping the assumption that cold dark matter (CDM) is a perfect fluid opens the possibility to fit the data without dark energy. For imperfect CDM, negative bulk pressure is favoured by thermodynamical arguments and might drive the cosmic acceleration. The coincidence between the onset of accelerated expansion and the epoch of structure formation at large scales might suggest that the two phenomena are linked. A specific example is considered in which effective (anti-frictional) forces, which may be due to dissipative processes during the formation of inhomogeneities, give rise to accelerated expansion of a CDM universe. ",
        "introduction": "It might appear that overwhelming evidence for the existence of dark energy (be it a cosmological constant or some other form of energy) has been collected. Observations of the cosmic microwave background (CMB) \\cite{CMB}, high redshift supernovae \\cite{SN}, the large scale structure \\cite{LSS} and clusters \\cite{C} give a self-consistent picture, which can be summarized by the dimensionless energy densities $\\Omega \\approx 1, \\Omega_\\Lambda \\approx 0.7, \\Omega_{\\rm b} h^2 \\approx 0.02, \\Omega_{\\rm cdm} h^2 \\approx 0.12$, implying $h \\approx 0.7$ and an age of the Universe of $14$ Gyr. But this evidence for dark energy is obtained from a fit to the $\\Lambda$CDM model, not by a direct observation of dark energy itself.  In this work we point out that the $\\Lambda$CDM model makes a very important, yet untested assumption: CDM\\footnote{ Any dark matter with $|P| \\ll \\rho$ during the epoch of structure formation is called CDM.} is described by a perfect fluid (at the largest scales). We show below that if this assumption were wrong, the evolution of the Hubble rate would have to be modified.  Let us first check whether there are any theoretical or observational reasons for believing that CDM behaves like a perfect fluid. The observed isotropy of the CMB together with the Copernican principle gives sound evidence that the Universe is homogeneous and isotropic at very large scales (cosmological principle). This implies that the energy-stress tensor of the Universe is of the form \\begin{equation} \\label{emt} T^a_b = (\\rho + P)u^a u_b + P \\delta^a_b, \\end{equation} where $u^a$ is the velocity of an observer comoving with the CMB heat bath, $\\rho$ and $P$ are the energy density and hydrodynamic pressure as measured by this observer. Note that this form does not imply that the fluid is perfect, i.e.~dissipationless. However, the observed CMB spectrum proves that the radiation component is in thermal equilibrium. Assume for the moment that CDM consists of weakly interacting massive particles (WIMPs) that were in thermal equilibrium with the radiation in the early Universe. For the most popular WIMP, the neutralino, kinetic decoupling from the radiation fluid happens at a temperature of about $10$ MeV \\cite{HSS}, long before structure formation starts. After kinetic decoupling any perturbation can easily disturb the CDM equilibrium distribution. A general result from kinetic theory within general relativity is that it is very hard to maintain kinetic equilibrium for freely streaming particles. This is possible only if a conformal time-like Killing vector exists and the particles are either massless or highly non-relativistic \\cite{Ehlers}. For CDM these conditions are approximately true until density perturbations become non-linear, and it is therefore well justified (at least for WIMPs that once were in thermal contact) to use a model with radiation plus CDM at the beginning of structure formation and during photon decoupling.  But today's Universe is very inhomogeneous on smaller scales, and we certainly cannot describe CDM by a perfect fluid on those scales. All kinds of dissipative effects might take place. Now the question arises whether looking at larger scales only (this is nothing but averaging over the smaller scales) can justify the assumption of a perfect fluid. A simple argument shows why this cannot be true in general. Assume that entropy is created by the dissipative effects in every small volume. Averaging over many small volumes will only add up the produced entropy, the net result being that entropy is produced at large scales as well, although no physical processes act at the large scales themselves.  To take into account dissipative effects in the formation of small scale inhomogeneities on the cosmic evolution on larger scales, one can start from the energy-stress tensor of an imperfect fluid (\\ref{emt}). If dissipative processes are taking place the hydrodynamic pressure $P$ no longer equals the kinetic pressure $p$, and we write $P = p + \\Pi$. For CDM $p \\approx 0$. We argue below that, if $\\Pi \\neq 0$, the observational evidence for two different dark components of the Universe breaks down. This is consistent with Pav\\'on's contribution to this conference, in which he shows that in order to have late-time cosmic acceleration and to solve the coincidence problem at the same time, matter must be dissipative \\cite{Diego}, although a quintessence component is admitted in that work.  What can be said about the sign of $\\Pi$? Let us assume for simplicity that the energy density of baryons is much less than that of dark matter, so all dynamically important energy density is in CDM. As long as the baryon number is conserved, we can define the entropy per baryon (specific entropy) $\\sigma$, and the relation \\begin{equation} T {\\rm d}\\sigma = {\\rm d}(\\rho/n_{\\rm B}) + p {\\rm d}(1/n_{\\rm B}) \\end{equation} holds true for any change of the thermodynamic state. Note that it is the kinetic pressure that enters in this expression. Using the covariant conservation of energy density and the baryon number conservation, we find the change of the heat density with time \\begin{equation} n_{\\rm B} T \\dot{\\sigma} = - 3 H \\Pi, \\end{equation} where $H$ denotes the Hubble rate. From the second law of thermodynamics $\\Pi \\leq 0$ follows for an expanding Universe, and in the case of CDM ($p \\approx 0$) we find $P \\approx \\Pi \\leq 0$! Thus a negative CDM bulk pressure seems possible from non-equilibrium effects during structure formation. A classic example in which the non-equilibrium pressure reduces the kinetic pressure is the bulk viscosity of cosmic fluids in the linear approximation, $\\Pi = - 3 H \\zeta < 0$ \\cite{Weinberg,PZ}.  Let us now consider the hypothesis that the Universe contains just one CDM component, which is described as an imperfect fluid at late times (redshifts of a few). Such a solution to the dark energy problem would be most elegant because no new particles or fields should be introduced and no new physics, such as extra dimensions or new forces, are needed. Under this hypothesis the cosmic coincidence problem turns into the question: Why do cosmic acceleration and the formation of large structures happen at about the same time? If the non-linear evolution of inhomogeneities could be identified as the driver for cosmic acceleration, the coincidence would be explained naturally \\cite{DJS}. ",
        "conclusions": "It seems that a careful investigation should be done to sort out the number of dark components of the Universe. We have shown that a fit to the $\\Lambda$CDM model cannot be sufficient evidence, as long as the assumption that CDM is a perfect fluid at large scales remains untested.  A possible argument against the proposal of this paper might be that observations of galaxy clusters indicate $\\Omega_{\\rm m} \\sim 0.3 < 1$! The discrepancy can be resolved by taking into account that the gravitating mass density is $\\rho + 3P$, and therefore cluster mass estimates probe $\\Omega_{\\rm m}(1 + 3P/\\rho)$ rather than $\\Omega_{\\rm m}$. Only geometrical mass estimates probe $\\Omega$ directly (as in the case of the CMB). On the largest scales (CMB, large scale structures and supernovae) $\\Omega_{\\rm m} =1$ would imply that $P \\sim  - 0.7 \\rho$, whereas on cluster scales the observed low mass density would be consistent with $P \\sim - 0.2 \\rho$. Thus an observational signature of the present scenario is a time- and scale-dependent effective equation of state of CDM \\cite{DJS}. These and other questions should be addressed in detail, before the energy budget of the Universe is understood."
    },
    "0209/astro-ph0209417.txt": {
        "abstract": "\\noindent The three-dimensional Monte Carlo photoionization code Mocassin has been applied to construct a realistic model of the planetary nebula NGC~3918. Three different geometric models were tried, the first being the biconical density distribution already used by \\citet{clegg87}. In this model the nebula is approximated by a biconical structure of enhanced density, embedded in a lower density spherical region. Spindle-like density distributions were used for the other two models (models~A and B). Model~A used a mass distribution slightly modified from one of Mellema's (1996) hydrodynamical models that had already been adopted by \\citet{corradi99} for their observational analysis  of NGC~3918. Our spindle-like model~B instead used an analytical expression to describe the shape of the inner shell of this object as consisting of an ellipsoid embedded in a sphere.  The effects of the interaction of the diffuse fields coming from two adjacent regions of different densities were investigated. These are found to be non-negligible, even for the relatively uncomplicated case of a biconical geometry. We found that the ionization structure of low ionization species near the boundaries is particularly affected.  It is found that all three models provided acceptable matches to the integrated nebular optical and ultraviolet spectrum. Large discrepancies were found between all of the model predictions of infrared fine-structure line fluxes and {\\it ISO~SWS} measurements. This was found to be largely due to an offset of $\\approx$14~arcsec from the centre of the nebula that affected all of the {\\it ISO} observations of NGC~3918.  For each model, we also produced projected emission-line maps and position-velocity diagrams from synthetic long-slit spectra, which could be compared to recent {\\it HST} images and ground-based long-slit echelle spectra. This comparison showed that spindle-like model~B provided the best match to the observations. Although the integrated emission line spectrum of NGC~3918 can be reproduced by all three of the three-dimensional models investigated in this work, the capability of creating projected emission-line maps and position-velocity diagrams from synthetic long-slit spectra was found to be crucial in allowing us to constrain the structure of this object. ",
        "introduction": "\\label{sec:ngc3918intro}  The southern planetary nebula NGC~3918 (PN G294.6+04.7) is a very well known and widely studied object. A detailed study, based on UV, optical and radio observations, was presented by \\citet[][from now on C87]{clegg87}. The morphological and  kinematical information available at the time of their work, was, however, very limited, and, based on this, they constructed a photoionization model, using the Harrington code \\citep[e.g.][]{harrington82}, assuming a biconical geometry for a nebula seen almost {\\it pole-on}. A two-dimensional representation of their model is shown in Figure~10 of C87. They had first tried a spherical model, deriving the radial hydrogen density distribution from the average radial intensity profile in the H$\\beta$ map they used, which had been obtained at the Boyden Observatory, South Africa, in 1973 by Drs.~K.~Reay and S.~P.~Worswick. The Zanstra temperature they derived for the central star was $117,000$\\,K, corresponding to a luminosity of $4900\\,L_{\\odot}$, for an adopted distance of 1.5~kpc. This model did not succeed in reproducing the line strengths from some of the high ionization species observed, such as, for example,  Ne~{\\sc v}, O~{\\sc v} and O~{\\sc iv}. They interpreted this as an indication that the nebula could be optically thin in some directions as seen from the star; this led to the formulation of a biconical model. The presence of an optically thin phase required an upward correction to the original Zanstra temperature and, in the final model, they adopted the ionizing spectrum described by a non-LTE model atmosphere for a central star having an effective temperature of 140,000~K, a surface gravity of ${\\rm log}~g~=~6.5$ and and a photospheric ${\\rm He}/{\\rm H}~=~0.10$ by number. This model atmosphere was calculated by C87 using the program of \\citet{mihalas75}. The resulting nebular model seemed to reproduce the main spectroscopic features observed.  Present observational data, such as for example images of NGC~3918 taken by the {\\it HST} (see Figure~\\ref{fig:hstimage}) and the echellograms obtained in several emission lines by \\citet[][from now on C99]{corradi99} (Figure~\\ref{fig:longslit}), show, however, that a biconical model is inconsistent with the spatio-kinematical structure of this planetary nebula. C99 presented an analysis of optical images and high resolution long slit spectra of three planetary nebulae, including NGC~3918. They concluded that the large scale structure of this object consists of a bright inner shell of roughly elliptical shape, from which two fainter protrusions extend in the polar directions, giving what was described in their paper as an overall spindle-like appearance. This inner shell, which has a size of $12'' \\times\\,20''$, measured along its major and minor axes, is surrounded by an elliptical shell with a diameter of $16''$. From the images and from the long slit spectra they obtained the basic kinematical, geometrical and orientation parameters of the inner shell. They adopted a hydrodynamical model  by G.~Mellema (1996) to reproduce, at least qualitatively, the observations. The model which gave the best fit was one, from Mellema's set, which posits an initial mass distribution strongly enhanced in the equatorial region, with a density contrast between the equatorial and the polar regions as large as 10 (see Figure~\\ref{fig:mellemadist}). The effects of an expanding shock driven by a strong wind would give a spindle-like structure similar to the one observed in the inner shell of NGC~3918. C99 derived the inclination and the kinematical parameters of the inner shell by using the spatio-kinematical model of \\citet{solf85} (which was also used to obtain $H\\alpha$ position-velocity diagrams, as well as the shape of the inner shell) to match the observational data. Their final model still showed some deviation from the observations, particularly in the long-axis position-velocity plot, which they attributed to simplified assumptions in the spatio-kinematical model. Photoionization calculations, however, were not carried out in their work, and therefore no comparison with the observed spectrum was available to them.  \\begin{figure} \\begin{center} \\psfig{file=mellemadist.ps, height=40mm, width=80mm} \\caption[Hydrodynamic model by \\citet{mellema96} used for NGC~3918] {Model image (left) and long axis synthetic slit spectrum (right) from the hydrodynamical models of \\citet{mellema96}. Figure adapted by C99 from Mellema's original thesis work.} \\label{fig:mellemadist} \\end{center} \\end{figure}  Given the large amount of observational data available for this object and the existence of the two different models described above, NGC~3918 seemed  an excellent candidate for a detailed three-dimensional photoionization study using the Mocassin code described by \\citet{ercolano02} and \\citet{ercolanoI}. Three photoionization models were constructed, using different density distribution descriptions, in order to try to reproduce the main spectroscopic features, as well as the projected maps published by C99. The models used a 23$\\times$23$\\times$23 Cartesian grid, with the ionizing source being placed in a corner in order to utilize the symmetry of the geometries used. The complete model nebulae were, therefore, contained in 45$\\times$45$\\times$45 cubic grids. All the grid cells have the same size. Velocity fields were then applied to the final converged grids in order to produce position-velocity diagrams to compare with the observations. In Section~2 the biconical density distribution model of C87 is described and the results obtained for it with Mocassin are presented and compared with those of C87, who used the one-dimensional Harrington code. A consistency test for the diffuse radiation field is also carried out in this section, in order to investigate the effects of discontinuities in the diffuse field transport in one-dimensional codes. The results obtained from the spindle-like models of NGC~3918 are presented in Section~3. Discrepancies were found between the predictions of all the models and the {\\it ISO~SWS} measurements of the infrared fine-structure lines; the possible reasons of these discrepancies are discussed in Section~4. ",
        "conclusions": "In this paper three photoionization models were constructed for the planetary nebula NGC~3918, a biconical model and two spindle-like models. The first spindle-like model (A) used the density distribution of the \\citet{mellema96} model already applied to this object by C99. The second spindle-like model (B) instead used an analytical expression for the density distribution, which aimed to mimic, by means of an ellipsoid embedded in a sphere, the shape of the inner shell of NGC~3918. The aim was to find a model which could not only reproduce the main spectroscopic features observed for this object, but also the spatio-kinematical properties of the nebula which recent observations (C99) have uncovered.  The integrated emission line spectra obtained from the three different models were all in fair agreement with the observations. Discrepancies exist between the observed line fluxes of the C~{\\sc iv}~$\\lambda\\lambda$1548,1550 and Mg~{\\sc ii}$\\lambda\\lambda$2796,2803 resonance doublets, which can largely be explained by dust absorption in the nebula \\citep{harrington88}. Large discrepancies were also found between the models' predictions for the fluxes of the infrared fine-structure lines and the measurements of the {\\it ISO~SWS} spectra. The main cause for this discrepancy is a pointing error which affected {\\it ISO} observations of NGC~3918, causing an offset of approximately 14~arcsec from the centre of the nebula. The corrected predicted line fluxes, obtained by convolving the {\\it ISO~SWS} aperture profiles with projected nebular maps in the relevant emission lines, confirmed that most of the discrepancy between the observed and model fluxes could be attributed to the pointing error, although there is still a factor of three discrepancy for the [S~{\\sc iv}]~10.5~$\\mu$m line.  A diffuse radiation field consistency test was also carried out in this work, which showed that the interaction of the diffuse radiation fields from two adjacent regions of different densities is not negligible, even in the relatively uncomplicated case of the biconical density distibution used by C87. We found that the low ionization species in the optically thick cones were particularly effected by the diffuse radiation field coming from the optically thin sector.  Although the volume-integrated emission line spectra obtained by the three models of NGC~3918 were in agreement with each other, the projected maps and the synthetic long-slit spectra obtained in several emission lines were, however, very different from one model to the other. Spindle-like model~B produced the best fits to the observations, although some discrepancies still exist, particularly in the [N~{\\sc ii}] maps and long slit spectra, as discussed in Section~\\ref{sub:PV}.  Confirming the conclusions of \\citet{monteiro00}, from their three-dimensional modelling of NGC~3132, this work has indicated that a detailed model of a nebula cannot be verified just by comparison of the observed integrated spectrum with model predictions. In fact, in the case of NGC~3918 approximately the same spectrum can be obtained with a number of different geometries and density distributions. For this reason, three-dimensional models are necessary in order to allow a spatio-kinematical analysis to be carried out by comparing predicted images and position-velocity diagrams in several lines to available observational data. Mocassin provides all the tools needed for such simulations and for the visualization of the final results.  \\vspace{7mm} \\noindent {\\bf Acknowledgments} We are most grateful to Dr. G. Mellema for providing the density distribution file used by C99 and in our spindle-like model~A of NGC3918. We thank the anonymous referee for useful comments. BE aknowledges support from PPARC Grant PPA/G/S/1997/00728 and the award of a University of London Jubber Studentship."
    },
    "0209/astro-ph0209516_arXiv.txt": {
        "abstract": "\\rightskip 0pt \\pretolerance=100 \\noindent We present high-resolution X-ray spectroscopy of GRB 020405 obtained with the Low Energy Transmission Grating Spectrometer (LETGS) on board the \\cha\\ {\\em X-ray Observatory\\/} starting 1.68 days after the burst. The spectrum appears featureless, with no evidence for emission lines, absorption edges, or narrow radiative recombination continua. The continuum can be fitted by a power law of photon index $\\Gamma = 1.72 \\pm 0.21$ and temporal decay index $\\alpha = 1.87 \\pm 0.1$, with a marginally significant excess column density of cold gas $N_{H}$=$(4.7 \\pm 3.7) \\times 10^{21}$ cm$^{-2}$ at the redshift of the host galaxy. The absence of iron lines indicates that the density of nearby surrounding material was unlikely to be very dense ($n \\lax 5 \\times 10^{12}$ cm$^{-3}$) at the time of the \\cha~observation. In the case of recombination following photoionization in an optically thin medium, most ionic species would be completely stripped at lower gas densities than this.  In the case of a power-law spectrum reflecting off a ``cold'', opaque medium of low density, negligible emission features would be produced. Alternative to these possible explanations for the lack of emission features, any X-ray line emission taking place in a dense medium in a ``nearby reprocessor'' scenario might have been overwhelmed by the bright afterglow continuum. Although the absence of discrete features does not unambiguously test for a connection between GRB 020405 and nucleosynthesis, it emphasizes the need for high-resolution X-ray spectroscopy to determine the exact emission mechanism responsible for the reported discrete lines in other GRB afterglows. ",
        "introduction": "The nature of gamma-ray burst (GRB) progenitors remains a mystery despite the localization of $\\approx$ 40 X-ray afterglows\\footnote{See http://www.aip.de/$\\sim$jcg/grbgen.html}. An increasing number of X-ray observations reporting discrete spectral features in GRB afterglows (Piro et al. 1999; Antonelli et al. 2000; Piro et al. 2000) has motivated attempts to find an explanation in terms of GRB progenitors. Thus far the claimed features are consistent with n = 2 $\\rightarrow$ 1 transitions in all possible charge states of Fe, i.e., from Fe K-shell fluorescence at 6.4 keV to H-like Fe Ly $\\alpha$ at 6.95 keV. More recently Reeves et al. (2002)  have reported the detection of H-like emission from multiple $\\alpha$-elements including Si and S in the spectrum of GRB 011211; however, the statistical significance of those features is debatable (Rutledge \\& Sako 2002). The possible presence of Fe and/or $\\alpha$-elements in the X-ray spectra of GRB afterglows could provide valuable information about the physical conditions in the vicinity of the progenitors.  Determining the relevant X-ray reprocessing mechanism in GRBs will involve the physical properties and geometry of the surrounding medium (see for instance Lazzati, Campana, \\& Ghisellini 1999; Paerels et al. 2000). Discrete spectral features are expected from hot gas when electron collisions cause excitations from the ground state. Emission lines can also arise via recombination if continuum photons (either from the burst itself or its afterglow) create a highly photoionized plasma. In certain cases fluorescence may be important if ions retain sufficient numbers of bound electrons. An important diagnostic that might discriminate between the different alternatives is the appearance of radiative recombination continua (RRC) produced by the return of unbound electrons to the ground state.  RRC from a photoionized plasma resemble narrow emission lines, and could be used to verify that recombination is occurring. A 10 ks grating observation of GRB 991216 (Piro et al. 2000) showed a possible iron RRC that could represent the first evidence of a recombining plasma near a GRB.  Nevertheless, and despite these efforts, the X-ray mechanism responsible for discrete features in GRBs is still uncertain due to limited spectral resolution and sensitivity of the existing observations (Paerels et al. 2000). In this letter we present results of the LETGS/ACIS-S spectroscopy of GRB 020405, and we show how density and geometry regulate the feasibility of detecting discrete lines in X-ray afterglows. ",
        "conclusions": "Our analysis of various discrete X-ray emission models show that the pure power-law spectrum of GRB 020405 is adequately explained by a low-density medium with $n \\lax 5 \\times 10^{12}$ cm$^{-3}$, or alternatively by a ``photon curtain'' effect produced by the afterglow continuum in a dense ``nearby reprocessor'' scenario. In the future, as the rapidity of localization improves, high-resolution spectroscopy could play a decisive role in resolving the emission mechanism (Paerels et al. 2000).  Figure 4 shows a simulated Gaussian line profile for Fe K$\\alpha$ with line width and flux identical to those reported for GRB 991216 (Piro et al. 2000), added to the observed power-law spectrum of GRB 020405. A Fe K$\\alpha$ line with similar properties would have been clearly detected in our LETGS spectrum. But quite possibly the major accomplishment of the \\cha~LETGS in grating spectroscopy could come in the detection of medium-$Z$ $\\alpha$ elements, $Z < 14$, where its higher effective area and sensitivity are superior to other instruments on board \\cha. Continuous high-resolution spectroscopy might also provide an avenue to detect a change in the emission mechanism as a function of time \\ie, from recombination to fluorescence. In summary, our results confirm the need for further high-resolution observations to improve the significance of previous claims of detections for discrete features in GRB afterglows."
    },
    "0209/astro-ph0209450_arXiv.txt": {
        "abstract": "The spectral energy distributions (SEDs) of dusty high-redshift galaxies are poorly sampled in frequency and spatially unresolved. Their form is crucially important for estimating the large luminosities of these galaxies accurately, for providing circumstantial evidence concerning their power sources, and for estimating their redshifts in the absence of spectroscopic information. We discuss the suite of parameters necessary to describe their SEDs adequately without introducing unnecessary complexity. We compare directly four popular descriptions, explain the key degeneracies between the parameters in each when confronted with data, and highlight the differences in their best-fitting values. Using one representative SED model, we show that fitting to even a large number of radio, submillimetre and far-infrared (far-IR) continuum colours provides almost no power to discriminate between the redshift and dust temperature of an observed galaxy, unless an accurate relationship with a tight scatter exists between luminosity and temperature for the whole galaxy population. We review our knowledge of this luminosity--dust temperature relation derived from three galaxy samples, to better understand the size of these uncertainties. Contrary to recent claims, we stress that far-IR-based photometric redshifts are unlikely to be sufficiently accurate to impose useful constraints on models of galaxy evolution: finding spectroscopic redshifts for distant dusty galaxies will remain essential. ",
        "introduction": "The rest-frame far-infrared (far-IR) thermal emission from dust grains heated by various sources -- the diffuse interstellar radiation field (ISRF) in galaxies, sites of active star formation, and a central active galactic nucleus (AGN) -- can dominate the spectral energy distribution (SED) of galaxies (Soifer \\& Neugebauer 1991; Sanders \\& Mirabel 1996). The most luminous galaxy apparent in the Universe (APM\\,08279+5255; Irwin et al.\\ 1998) emits approximately 60\\,per cent of its bolometric luminosity in the far-IR waveband, while low-redshift galaxies with blue optical colours that were detected by the {\\it IRAS} satellite also release about 60\\,per cent of their total bolometric luminosity as thermal radiation from dust (Mazzarella \\& Balzano 1986). Even the most quiescent spiral galaxies such as the Milky Way emit of order 30\\,per cent of their total luminosity from dust (Reach et al.\\ 1995; Alton et al.\\ 1998; Dale et al.\\ 2001; Dale \\& Helou 2002). Dust emission remains important at high redshifts. The most distant quasi-stellar objects (QSOs) (Benford et al.\\ 1999; Carilli et al.\\ 2001; Isaak et al.\\ 2002) and more typical, but still very luminous galaxies detected in submillimetre(submm) wave surveys (Blain et al.\\ 2002; Smail et al.\\ 2002) emit strongly at rest-frame far-IR wavelengths.  As compared with the rich variety of features in the SEDs of galaxies at near-IR, optical and ultraviolet wavelengths, the far-IR SED is simple, dominated by a smooth pseudo-thermal continuum emission spectrum. At most about 1\\,per cent of the emitted energy is associated with spectral lines from atomic fine-structure and molecular rotational transitions (Malhotra et al.\\ 1997; Luhman et al.\\ 1998; Combes, Maoli \\& Omont 1999; Blain et al.\\ 2000). The mid-IR spectra of galaxies from 10 to 30\\,$\\mu$m are expected to be significantly more complex, especially because of broad line emission from polycyclic aromatic hydrocarbon (PAH) molecules (Dale et al.\\ 2001).  A variety of models have been used to describe the far-IR SEDs of dusty galaxies. We compare four well-constrained descriptions with data for a variety of types of galaxy, and highlight the importance both of degeneracies between the parameters and the need to avoid baroque descriptions that require a greater number of parameters than can be justified and fixed by existing data. Using one uniform, self-consistent description of the SED we discuss the accuracy of photometric redshifts that can be derived for high-redshift galaxies based on their observed colours, making assumptions concerning their SEDs. We describe in detail the degeneracy between redshift and dust temperature when fitting photometric data for high-redshift galaxies (Blain 1999b; Blain et al.\\ 2002), and discuss the prospects for breaking this degeneracy using information about absolute luminosity, obtained from a luminosity--temperature ({\\it LT}) relation for dusty galaxies. A narrow range of SEDs was included implicitly in recent discussions of the prospects for determining mm-wave photometric redshifts (Hughes et al.\\ 2002; Aretxaga et al.\\ 2003; Dunlop et al.\\ 2003), which leads to encouraging results. We discuss existing data on the {\\it LT} relation (Dunne et al.\\ 2000; Stanford et al.\\ 2000; Dale et al. 2001; Dale \\& Helou 2002; Garrett\\ 2002; Barnard \\& Blain 2003; Chapman et al. 2003), which leads to a much less optimistic outlook for far-IR/submm photometric redshifts. The observed dispersion in the {\\it LT} relation is the key quantity that limits the effectiveness of the technique.  In Section 2 we describe four SED models, and compare them with a range of observed galaxy SEDs. We highlight the consequences of errors in the fitted SEDs and the {\\it LT} relation for determining photometric redshifts in Section 3. Finally, in Section 4, we describe the requirements for spectroscopic observations that will remove this uncertainty, and describe the opportunities that much more detailed far-IR SEDs measured using {\\it SIRTF}\\footnote{See http://sirtf.caltech.edu} from 2003 will provide for better understanding the {\\it LT} relation and for determining far-IR-based photometric redshifts. ",
        "conclusions": "We have discussed the description of the SEDs of dusty galaxies using four different models that are appropriate to describe data that is available at present and is likely to be generated by forthcoming space missions. One of the parameters in each model always describes the peak frequency of the thermal dust SED (`temperature'), while two spectral indices describe the fraction of hot and cold dust: an `$\\alpha$' and `$\\beta$' parameter respectively. Observational data constrains these SED descriptions to within 10\\,per cent accuracy across the full range of interesting wavelengths longer from about 20\\,$\\mu$m to deep in the radio waveband. It is important to be careful in interpreting the values of dust temperatures, emissivities and masses that are inferred with the values of real physical parameters.  There is a huge degeneracy between temperature and redshift when fitting the SED of a distant galaxy. Assuming some link between the luminosity and SED allows this degeneracy to be broken, but a range of available information indicates that this relationship has a very considerable scatter, by up to a factor of 2. Without the knowledge that this {\\it LT} relation has a scatter as narrow as the required accuracy of the photometric redshift, continuum far-IR/submm/radio photometric redshifts are almost useless for providing constraints from the data for an individual galaxy.  In order to finally assess their usefulness it is essential to quantify the {\\it LT} relationship accurately, based on a large number of galaxies with known redshifts and well-sampled SEDs. This information is not available at present, but will be generated by {\\it SIRTF}. Only if the true scatter in the {\\it LT} relation turns out to be less than about 20\\,per cent will the photometric redshift technique be useful. Spectroscopic observations to fix the redshifts and SEDs of dusty galaxies remain essential to understand the population, and it is important to develop new types of spectroscopic instruments that can address these questions, for example wide-band detectors for multiple CO emission lines (Bradford et al. in preparation)."
    },
    "0209/astro-ph0209499_arXiv.txt": {
        "abstract": "{ We present the description and the first results\\thanks{Tables 2 and 3 are only available in electronic form at CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsweb.u-strasbg.fr/Abstract.html. Figures A1--A6 are only  available in electronic form at Edpsciences (http://www.edpsciences.org)} of a new project devoted to the search for extremely metal-deficient blue compact/\\ion{H}{ii}-galaxies (BCGs) and to the creation of a well selected large BCG sample with strong emission lines. Such galaxies should be suitable for reliable determination of their oxygen abundance through the measurement of the faint [\\ion{O}{iii}]\\,$\\lambda$\\,4363~\\AA\\ line. The goals of the project are two-fold: a) to discover a significant number of new extremely metal-poor galaxies ($Z \\lesssim$ 1/20~$Z$\\sunn), and b) to study the metallicity distribution of local BCGs. Selection of candidates for follow-up slit spectroscopy is performed on the database of objective prism spectra of the Hamburg Quasar Survey. The sky region is limited by $\\delta \\geq 0$\\degr\\ and $b^{II} \\leq -30$\\degr. In this paper we present the results of the follow-up spectroscopy conducted with the Russian 6\\,m telescope. The list of observed candidates contained 52 objects, of which 46 were confirmed as strong-lined  BCGs ($EW$([\\ion{O}{iii}]\\,$\\lambda$\\,5007) $\\ge$ 100~\\AA). The remaining five lower excitation ELGs include three BCGs, and two galaxies classified as SBN (Starburst Nucleus) and DANS (Dwarf Amorphous Nucleus Starburst). One object is identified as a quasar with a strong Ly$\\alpha$ emission line near $\\lambda$\\,5000~\\AA\\ (z~$\\sim 3$). We provide a list with coordinates, measured radial velocities, $B$-magnitudes, equivalent widths $EW$([\\ion{O}{iii}]\\,$\\lambda$\\,5007) and $EW$(H$\\beta$) and for the 46 strong-lined BCGs the derived oxygen abundances 12+$\\log$(O/H). The abundances range between 7.42 and 8.4 (corresponding to metallicities between 1/30 and 1/3~$Z$\\sunn). The sample contains four galaxies with $Z \\lesssim$ 1/20~$Z$\\sunn, of which three are new discoveries. This demonstrates the high efficiency of the new project to find extremely metal-deficient galaxies. The radial velocities of the strong--lined ELGs range between 500 and 19000~\\kms\\ with a median value of $\\sim$~6400~\\kms. The typical $B$-magnitudes of the galaxies presented are $17\\fm0-18\\fm0$. ",
        "introduction": "Blue compact/\\ion{H}{ii} galaxies (BCGs) are low-mass gas-rich objects which are currently undergoing an episode of enhanced star formation. Such episodes are usually recognized by a strong emission-line spectrum of \\ion{H}{ii} type. Their duration is relatively short: strong emission lines are detectable on timescales from ten to a hundred Myr. These episodes of intense star formation are often called starbursts. BCGs have been intensively studied since the seminal work by Sargent \\& Searle (\\cite{Sargent70}), and has been especially actively during the last decade. The list of publications on this subject is very long, and we list below only those studies dealing with observations and/or analysis of sufficiently large samples. These papers, among others, are Campos-Aguilar et al. (\\cite{Campos93}), Masegosa et al. (\\cite{Masegosa94}), Izotov et al. (\\cite{Izotov94}), Thuan et al. (\\cite{Thuan95}),  Salzer et al. (\\cite{Salzer95}), Pustilnik et al. (\\cite{PULTG}), Taylor et al. (\\cite{Taylor95}), Vilchez (\\cite{Vilchez95}), Stasinska \\& Leitherer (\\cite{SL96}), Telles \\& Terlevich (\\cite{Telles97}), van Zee et al. (\\cite{Zee98a}), Schaerer et al. (\\cite{Schaerer97}, \\cite{Schaerer99}), Izotov \\& Thuan~(\\cite{Izotov99}), Bergvall et al. (\\cite{Bergvall99}), Thuan et al. (\\cite{TLMP99}), Kunth \\& \\\"Ostlin (\\cite{KO}), Pustilnik et al. (\\cite{BCG_ENV}), and Ugryumov et al.~(\\cite{Ugryumov01}, and references therein).  To date more than thousand galaxies of this type are known. Some of them had been found in early studies as Zwicky compact galaxies (Zwicky \\cite{Zwicky}) or Haro blue galaxies (Haro \\cite{Haro56}). However, the great majority of BCGs were picked up by objective prism surveys such as the First and Second Byurakan (FBS, SBS) (Markarian~\\cite{Markarian67}; Markarian et al.~\\cite{Markarian83}; Stepanian~\\cite{Stepanian94}), the University of Michigan (Salzer et al. \\cite{Salzer89b}), the Tololo (Terlevich et al.~\\cite{Terlevich91}), the Case (Pesch et al.~\\cite{Pesch95}; Salzer et al.~\\cite{Salzer95}; Ugryumov et al. \\cite{Ugryumov98}), the Heidelberg void (Popescu et al. \\cite{Popescu98}) and the Hamburg/SAO survey (Ugryumov et al.~\\cite{Ugryumov01}, and references therein).  \\begin{figure} \\centering \\psfig{figure=figure01.ps,width=9cm,angle=-90} \\caption[]{\\label{Fig1} The sky area covered by the HSS--LM is determined by $b^{II}~\\le~-30$\\degr\\ and $\\delta \\ge$ 0\\degr. The positions of ELGs presented in this paper are shown with  filled circles.  The zones of three other surveys: University of Michigan (UM), Heidelberg void survey (HVS) and Sloan Digital Sky Survey (SDSS) are shown superimposed on the HSS--LM region (see references in the text). } \\end{figure}  Like other low-mass galaxies, BCGs have metallicities $Z$ many times lower than those of normal bright galaxies. Their characteristic range of metallicity $Z$ is $\\sim$~($1/15-1/3$)~$Z$\\sunn\\ (e.g., Terlevich et al.~\\cite{Terlevich91}, Izotov et al. \\cite{Izotov93}), while very few of them show $Z$ as low as ($1/50-1/20$)~$Z$\\sunn. All well-studied BCGs with characteristic metallicities were found to be old galaxies, which evolve slower than their massive counterparts. Instead of having a lower gas consumption rate, it was discussed many years galactic winds were responsible for a significant loss of heavy elements (Mac Low \\& Ferrara \\cite{MLF99}); however more recent models (Silich \\& Tenorio-Tagle \\cite{STT01}, Legrand et al. \\cite{Legrand02}) indicate that for the majority of BCGs a significant loss of metals is rather improbable. In contrast to all typical BCGs, a few well studied BCGs with an extreme deficit of heavy elements show properties of young galaxies, undergoing their first SF episode. In particular, they have gas-mass fractions (that is, gas mass relative to all visible baryon mass) of the order $0.9-0.99$ (e.g., Salzer et al. \\cite{Salzer91}, Kniazev et al. \\cite{Kniazev00b}, Pustilnik et al. \\cite{PBTLI}) and in the outermost parts of their low-surface brightness disks show very blue colours, consistent with ages of the ``old'' stellar population less than $100-200$~Myr.  One of the most often discussed candidate objects for low age is I~Zw~18 with $Z = 1/50~Z$\\sunn, discovered by Searle \\& Sargent (\\cite{Searle72}). The most recent data are summarized by Izotov \\& Thuan (\\cite{Izotov99}), Izotov et al. (\\cite{Izotov01}), van Zee et al. (\\cite{Zee98b}), Aloisi et al. (\\cite{Aloisi99}), \\\"Ostlin (\\cite{Ostlin00}), and Papaderos et al. (\\cite{Papaderos02}).  While the first two papers forward arguments for a maximum age of less than 200~Myr, Aloisi et al. (\\cite{Aloisi99}) and \\\"Ostlin (\\cite{Ostlin00}) suggest an age of at least 1~Gyr. The age of the extremely metal-poor galaxies remains therefore a matter of debate.  Several other well-known objects are SBS~0335--052 (E and W) with $Z = 1/42~Z$\\sunn\\ and 1/50~$Z$\\sunn\\ (e.g., Izotov et al. \\cite{Izotov97}; Papaderos et al. \\cite{Papaderos98}; Lipovetsky et al. \\cite{Lipovetsky99}; Pustilnik et al. \\cite{PBTLI}, \\\"Ostlin \\& Kunth \\cite{OK01}), HS~0822+3542 with $Z = 1/30~Z$\\sunn\\ (e.g., Kniazev et al. \\cite{Kniazev00b}; Pustilnik et al. \\cite{LSBD}; Izotov et al. \\cite{Izotov02}), Tol~1227--273 with $Z = 1/28~Z$\\sunn\\ (e.g., Fricke et al. \\cite{Fricke01}) and CG~389 (1415+437) with $Z = 1/22~Z$\\sunn\\  (e.g., Thuan et al. \\cite{Thuan99}).  Most of the evidences is still indirect and in many cases controversial, but the study of these rare galaxies clearly stimulates our understanding of the earliest periods of galaxy evolution. Probably, these extremely metal-deficient BCGs are the best approximation of low-mass young galaxies expected to form commonly at redshifts $3 < z < 5$.  Despite 30 years of observations and more than a thousand known BCG galaxies, another important aspect of BCG studies remains unclear. What is the true metallicity distribution of BCGs and their progenitors? This has important cosmological implication, because the knowledge of the contemporary  $Z$ distribution allows one to pose question regarding their chemical evolution during the last few Gyr.  The results of BCG morphological studies indicate that they do not comprise a homogeneous group, but rather are a mixture of various types (e.g. Kunth et al. \\cite{Kunth88}; Doublier et al. \\cite{Doublier97}). This implies that BCG progenitors probably are not a rare type of gas-rich low-mass galaxy (see, however, Salzer \\& Norton \\cite{SN99}). On the other hand, the recent results by Mouri \\& Taniguchi (\\cite{Mouri00}), based on far--IR tracers of recent SF, indicate that up to $\\sim(30-40)$\\% of the galaxies in their magnitude-limited sample within the galactic neighborhood have experienced SF activity during the last 100 Myr. The evidence that such a large fraction of galaxies experience episodes of significantly enhanced SF emphasizes the importance of BCG studies. In particular, the study of galaxies in the BCG phase allows us to more easily understand some of their progenitor parameters (such as $Z$), which are difficult to determine in more quiet periods of galaxy evolution.  Based on a large well-selected sample of BCGs and using the model predictions for the evolution of observable parameters, such as $EW$(H$\\beta$), we hope to advance in establishing the true metallicity distribution of BCGs. This would be a major step to obtain the true $Z$-distribution of the BCG parent population -- gas-rich low-mass galaxies.  These ideas are the motivation of our new project, ``The Hamburg/SAO survey for low metallicity BCG/\\ion{H}{ii} galaxies''. The sky region covered is limited by the southern galactic hemisphere north of $\\delta=0$\\degr\\ and $b^{II}\\leq -$30\\degr. This large section of the sky is not well covered by spectroscopic surveys of sufficient depth. Besides the First Byurakan Survey (Markarian~\\cite{Markarian67}) dealing with relatively bright galaxies, a 9.5\\degr wide strip of the University of Michigan (UM) Survey (MacAlpine et al.~\\cite{McAlpine77}; Salzer et al.~\\cite{Salzer89b} and references therein) and of several fields of the Heidelberg void survey (HVS) (Hopp \\& Kuhn \\cite{Hopp95}; Popescu et al. \\cite{Popescu96}) there are no large deep surveys in this region. It will be covered in the near future by two 1.5\\degr\\ wide strips of the SDSS, having good spectroscopy for all galaxies brighter than $B \\sim 18\\fm5$ (York et al. ~\\cite{York2000}), and in the more distant future this region will be covered by the 6dF-z project (Mamon \\cite{Mamon99}). The latter will obtain spectra only for relatively bright galaxies selected in the near-IR corresponding to $B \\lesssim 16\\fm0$ for blue galaxies and hence will not compete with our significantly deeper survey. Therefore, most of the results obtained in the frame of this new survey will be valuable material for studies of strong-lined BCGs/\\ion{H}{ii} galaxies in this region.  This project has several features in common with the earlier ``Hamburg/SAO survey for emission-line galaxies'' (HSS for ELGs), presented in a series of papers by Ugryumov et al. (\\cite{Ugryumov99})~(I), Pustilnik et al. (\\cite{Pustilnik99})~(II), Hopp et al. (\\cite{Hopp00})~(III), Kniazev et al. (\\cite{HSS_4})~(IV), Ugryumov et al. (\\cite{Ugryumov01})~(V) and Pustilnik et al. (\\cite{PEM02})~(VI).  The new project we will abbreviate as ``HSS--LM'' (LM --- for low metallicity) to distinguish it from the ``HSS'' (for ELGs). The selection procedure used in the HSS is described in detail by Ugryumov et al. (\\cite{Ugryumov99}). The main difference introduced in the HSS--LM is a more strict criterion on the strength of emission-line features in the full-resolution objective prism spectra. We select for follow-up long-slit spectroscopy only candidate galaxies with the strongest emission feature near $\\lambda$\\,5000\\,\\AA, which, according to our previous experience, corresponds to the ([\\ion{O}{iii}]\\,$\\lambda$\\,5007) line with equivalent widths $EW > 150-200$~\\AA. Such galaxies are quite rare and their prism spectra are prominent enough not to miss them. Our slit spectroscopy justifies the correctness of the elaborated criterion. The great majority of the galaxies observed indeed have $EW$([\\ion{O}{iii}]\\,$\\lambda$\\,5007) $>$ 150~\\AA, and the fraction of galaxies with $EW$([\\ion{O}{iii}]\\,$\\lambda$\\,5007) $<$ 100~\\AA\\ is low (see Fig.~\\ref{Fig3}).  The main advantage of this sample is that for most of these faint candidates the 6\\,m telescope spectra with moderate integration times ($\\sim20$ min.) give a well detected [O{\\sc iii}]\\,$\\lambda$\\,4363\\,\\AA\\ line, allowing the reliable determination of oxygen abundance by the standard method (Pagel et al. \\cite{Pagel92}).  In the context of an instantaneous SF burst the imposed selection criterion selects galaxies with very young SF bursts: typically younger than $5-10$ Myr (Schaerer \\& Vacca \\cite{SV98}; Leitherer et al. \\cite{Starburst99}). However, recent results indicate that quite often SF in BCGs proceeds in a more complicated mode, similar to a propagating wave of SF along the galaxy body (e.g., Zenina et al.~\\cite{Zenina97}; Thuan et al.~\\cite{Thuan99} among others). Probably in this case, lines with large $EW$([\\ion{O}{iii}]\\,$\\lambda$\\,5007) come from regions close to the front of such a propagating SF wave.  In this paper we present the general outline of the project, give the details of the selection procedure and present the first list of galaxies observed spectroscopically. In Section~2 the selection criteria are discussed. In Section~3 the follow-up spectral observations and the data reduction are described. We present some results of the spectrophotometric observations and a brief summary in Section~4. The efficiency of this survey in detecting strong-lined ELGs and its perspectives are discussed and the respective conclusions are drawn in Section~5. Throughout the paper a Hubble constant of H$_{0} = 75$~\\kms\\mbox{Mpc}$^{-1}$ is used. ",
        "conclusions": "  \\begin{enumerate}  \\item The HSS--LM is a new HQS based project, which very efficiently samples the strong-lined ELGs of \\ion{H}{ii}/BCG type. After the 6\\,m telescope follow-up spectroscopy 71~\\% of the candidates (37 galaxies) were found to be BCG/\\ion{H}{ii}-galaxies with $EW$([\\ion{O}{iii}]\\,$\\lambda$\\,5007) in the range of 200 to 2100~\\AA.  \\item Among these galaxies 3 new very metal-deficient BCGs were found with $Z$ in the range 1/28 to 1/20~$Z$\\sunn. One more such galaxy -- UGCA~20 -- with $Z =$ 1/32~$Z$\\sunn\\ was rediscovered as one of the HSS--LM targets.  \\item For 43 strong-lined BCGs the oxygen abundance was determined with the use of the temperature sensitive faint line [\\ion{O}{iii}]\\,$\\lambda$~4363~\\AA. Its full range corresponds to ionized gas metallicities between 1/28 and 1/3~$Z$\\sunn.  \\end{enumerate}  This paper presents about 1/3 of the galaxies from the final sample of strong-lined BCGs of this survey. The observations of the remaining objects are expected to be completed during the years $2002-2003$."
    },
    "0209/astro-ph0209385_arXiv.txt": {
        "abstract": "It is sometimes asserted that the rapidity of biogenesis on Earth suggests that life is common in the Universe.% We critically examine the assumptions inherent in this argument. Using a lottery model for biogenesis in the Universe, we convert the observational constraints on the rapidity of biogenesis on Earth into constraints on the probability of biogenesis on other terrestrial planets. For example, if terrestrial biogenesis took less than 200 Myr (and we assume that it could have taken 1 billion years) then we find the probability of biogenesis on terrestrial planets older than $\\sim 1$ Gyr,  is $> 36\\%$ at the $95\\%$ confidence level. However, there are assumptions and selection effects that complicate this result: although we correct the analysis for the fact that biogenesis is a prerequisite for our existence, our result depends on the plausible assumption that {\\bf{\\it rapid}} biogenesis is not such a prerequisite. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209100_arXiv.txt": {
        "abstract": "Bose-Einstein condensation of antikaons in cold and dense beta-equilibrated matter under the influence of strong magnetic fields is studied within a relativistic mean field model. For magnetic fields $> 5 \\times 10^{18}$G, the phase spaces of charged particles are modified resulting in compositional changes in the system. The threshold density of $K^-$ condensation is shifted to higher density compared with the field free case. In the presence of strong fields, the equation of state becomes stiffer than that of the zero field case. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209336_arXiv.txt": {
        "abstract": "We study the influence of converging cooling flow bulk motions on the Sunyaev-Zel'dovich (SZ) effect. To that purpose we derive a modified Kompaneets equation which takes into account the contribution of the accelerated electron media of the cooling flow inside the cluster frame. The additional term is different from the usual kinematic SZ-effect, which depends linearly on the velocity, whereas the contribution described here is quadratic in the macroscopic electron fluid velocity, as measured in the cluster frame. For clusters with a large cooling flow mass deposition rate and/or a small central electron density, it turns out that this effect becomes relevant. ",
        "introduction": "\\label{intro}  The SZ-effect is becoming more and more an important astrophysical tool thanks to the rapid progress of the observational techniques, which allow increasingly precise measurements. It has thus become relevant to study further corrections to it, such as relativistic effects \\citep{re95}, the shape of the galaxy cluster and its finite extension or a polytropic temperature profile (see e.g. \\citet{pu00}), the presence of cooling flows \\citep{sc91,ma00}, corrections induced by halo rotation \\citep{co01} or even by Brillouin scattering \\citep{sm02} and the influence of early galactic winds \\citep{ma01}. These additional effects are of different relevance and often depend on the specific values of the parameters which describe a given cluster. Generally, they range from the percent level up to even 20-30\\% and accordingly in the subsequent determination of the Hubble constant. Taking into account such corrections is on one side relevant when determining the Hubble constant via the SZ-effect, and on the other hand they could be interesting as a tool to study the detailed structure of the cluster itself.\\\\  Following these lines we study here another effect, which has not yet been taken into account: The possible influence on the SZ-effect of the cooling flow bulk motion of the electron media inside the cluster frame. Indeed, one expects such a motion to be present in cooling flows, for which in some clusters there is evidence in the central regions \\citep{al93,fa91,fa94,fa97}. Recently, \\citet{db02} investigated the prospects to detect gas bulk velocities in clusters of galaxies through the kinematic SZ-effect, which depends linearly on the velocity component along the line of sight. Bulk motions along a given line of sight will contribute to the kinematic SZ term as long as their averaged velocity, in the observer frame, does not vanish. In the case of cooling flow bulk motion the averaged velocity, in the cluster frame, along a given line of sight vanishes in good approximation, since we assume spherical symmetric infall. Thus, the cooling flow bulk motion will not contribute as such to the kinematic SZ-effect. Indeed, the effect we  consider depends quadratically on the velocity and clearly, the averaged quadratic velocity does not vanish along a line of sight in the cluster frame. As we will see, the considered effect usually turns out to be small, since the cooling flow bulk motion velocities are rather small, unless in the very central regions of a cooling flow clusters, where the cooling flow approaches a sonic radius and changes from the subsonic to the supersonic regime. Nonetheless, in some favourable cases it might be of the order of some percent of the thermal SZ-effect.\\\\  The aim of this paper is to examine the influence on the thermal SZ-effect of the moving electron media inside a cooling flow region of a galaxy cluster. As it will be seen in more detail later, this effect has clearly to be distinguished from the kinematic SZ-effect, which takes into account the motion of the whole cluster or a fraction of it along a given line of sight. In fact, the standard thermal SZ-effect describes the frequency dependent intensity change of the Cosmic Microwave Background (CMB) photons by inverse Compton scattering off the hot intracluster plasma (electrons), where the electrons are supposed to be static scatterers.\\\\  The paper is organised as follows: In section 2 we briefly outline the dynamics of a homogeneous steady-state cooling flow model in order to get the velocity profile of the bulk motion as well as the corresponding electron density distribution. In section 3 we derive the modification to the standard Kompaneets equation due to the inclusion of bulk motion of the scatterers inside the cluster frame and in section 4 we present and discuss our results. In section 5 we give a summary and an outlook. ",
        "conclusions": "We showed how the Kompaneets equation has to be modified to include the effect of the bulk motion due to a homogeneous steady-state cooling flow. This effect is quadratic in the cooling flow velocity, as it is measured in the cluster frame, and thus different from the usual kinematic SZ-effect, which is linear in the velocity. \\\\ The contribution strongly depends on the specific dynamics of the cooling flow and the cluster properties. Generally, it turns out to be rather small. However, for some clusters which have more extreme parameters, though not irrealistic, one finds a contribution at the percent level. One might speculate that this effect could be relevant when observing young clusters which are not yet virialized, still in their formation phase and for which one might expect the presence of rather large and extended regions with bulk motions, or perhaps in superclusters which are still in the formation phase. Such young clusters might be observable in future with the Planck satellite. \\\\ Another interesting situation might arise from shocks in merging processes of clusters of galaxies, in situations where the average velocity along a line of sight approximately vanishes. If so, the merging process will not be detected through the kinematic SZ-effect which is linear in the velocity, but could be instead observed through the above discussed effect which is quadratic in the velocity.    \\noindent{\\bf Acknowledgments}  We would like to thank S. Majumdar and S. Schindler for useful discussions. This work was supported by the Swiss National Science Foundation. Part of the work of D.P. has been conducted under the auspices of the {\\it D\\raisebox{0.5ex}{r} Tomalla Foundation}."
    },
    "0209/astro-ph0209046_arXiv.txt": {
        "abstract": "As a follow up to our recent study of a large sample of LMC clusters \\cite{me}, we have conducted a similar study of the structures of ten SMC clusters, using archival {\\em Hubble Space Telescope} snapshot data. We present surface brightness profiles for each cluster, and derive structural parameters, including core radii and luminosity and mass estimates, using exactly the same procedure as for the LMC sample.  Because of the small sample size, the SMC results are not as detailed as for the larger LMC sample. We do not observe any post core-collapse clusters (although we did not expect to), and there is little evidence for any double clusters in our sample. Nonetheless, despite the small sample size, we show for the first time that the SMC clusters follow almost exactly the trend in core radius with age observed for the LMC system, including the apparent bifurcation at several hundred Myr. This further strengthens our argument that this relationship represents true physical evolution in these clusters, with some developing significantly expanded cores due to an as yet unidentified physical process. Additional data, both observational and from $N$-body simulations, is still required to clarify many issues. ",
        "introduction": "We recently conducted a study of the structures of a large sample of rich star clusters in the Large Magellanic Cloud (LMC) (Mackey \\& Gilmore 2002; hereafter Paper I), in which we compiled a pseudo-snapshot data set from the {\\em Hubble Space Telescope} ({\\em HST}) archive and used these observations to construct high resolution surface brightness profiles. From these profiles, we were able to obtain measurements of the structural parameters of each cluster, including their core radii, and total luminosities and masses. We also demonstrated that these clusters followed a trend in core radius with age -- namely that the spread in core radius increases significantly as the clusters grow older, a result previously discussed by Elson and collaborators \\cite{efl,elson,ellson}. It seems likely that this trend reflects real physical evolution of these clusters, as argued in Paper I, although the mechanism by which the cores of some clusters expand during their lifetimes while the cores of others do not, is as yet unidentified. We are currently carrying out $N$-body simulations to investigate in detail several physical processes which might drive core expansion (e.g., Wilkinson et al., in prep.).  Having exhausted the suitable archival LMC data, we now turn our attention to the cluster system in the Small Magellanic Cloud (SMC). The SMC cluster system is similar to that of the LMC in that it contains rich star clusters of masses comparable to Galactic globular clusters, and with ages spanning the range $10^{6}-10^{10}$ yr. The two systems however, have dissimilar cluster formation histories and age-metallicity relationships (see e.g., Rich et al. \\shortcite{rsz}, for a brief review). The number of SMC clusters is also far fewer than the number of clusters in the LMC, and the SMC system has been far less extensively studied. It is not surprising then that there are few available surface brightness profiles for SMC clusters, and (prior to the present study) it was not known whether the SMC cluster system followed the core radius vs. age relationship observed for LMC clusters. In fact, we were only able to locate in the literature three low resolution density profile studies from photographic plates -- those of Kontizas, Danezis \\& Kontizas \\shortcite{kdk}, Kontizas \\& Kontizas \\shortcite{kk}, and Kontizas, Theodossiou \\& Kontizas \\shortcite{ktk}.  It therefore seemed very worthwhile to extract what archival {\\em HST} frames we could locate, and reduce these using the same procedure we had applied to the LMC sample, so that we now have two directly comparable and entirely homogeneous data sets. We describe these data in Section \\ref{sample} and briefly reiterate the reduction process in Section \\ref{reduction} -- this process and the problems and uncertainties associated with it have been described in great detail in Paper I. We present the surface brightness profiles and the derivation of key structural parameters in Section \\ref{results}, and a discussion of these results in the context of the core radius vs. age trend in Section \\ref{discussion}. The results from the present study (Tables \\ref{data}, \\ref{ages}, \\ref{params}, and \\ref{luminmass}) together with the surface brightness profiles, are available on-line at {\\em http://www.ast.cam.ac.uk/STELLARPOPS/SMC\\_clusters/}. ",
        "conclusions": "In a follow-up to our recent work detailing the structures of a large sample of LMC clusters (Paper I), we have obtained a similar archival {\\em HST}  snapshot data set for a sample of ten SMC clusters. We have constructed surface brightness profiles for these clusters, and obtained measurements of their structural parameters, following a method exactly similar to that applied to the LMC sample. Luminosities, masses and central densities have also been estimated for the SMC sample. These data, along with the surface brightness profiles, are available on-line at {\\em http://www.ast.cam.ac.uk/STELLARPOPS/SMC\\_clusters/}. Unlike for the LMC sample, we do not see any evidence for post core-collapse clusters in our sample, but this is not unexpected. Similarly, we do not see any compelling evidence from the surface brightness profiles for double clusters in our sample, although a couple of profiles show bumps similar to those observed for several LMC clusters.  We have used our core radius measurements for the SMC sample to investigate further the core radius vs. age relationship, which was described in detail in Paper I. Although compromised somewhat by the small sample size, our analysis shows that the SMC clusters apparently follow a very similar relationship to the LMC clusters, with some clusters maintaining small cores throughout their lives, but with others developing much enlarged cores. It is possible that a higher percentage of rich SMC clusters than rich LMC clusters develop such expanded cores. Additional data, both observational and computational, is required to further explore this relationship and the physical processes involved."
    },
    "0209/astro-ph0209264_arXiv.txt": {
        "abstract": "s{ The idea that the vacuum polarization process occurring during gravitational collapse to a black hole endowed with electromagnetic structure (EMBH) could be the origin of gamma ray bursts (GRBs) is further developed. EMBHs in the range 3.2 -- 10$^6$ solar masses are considered. The formation of such an EMBH, the extraction of its mass-energy by reversible transformations and the expansion of the pair-electromagnetic pulse (PEM pulse) are all examined within general relativity. The PEM pulse is shown to accelerate particles to speeds with Lorentz gamma factors way beyond any existing experiment on Earth. Details of the expected burst structures and other observable properties are examined. }  The Kerr-Newman mathematical solution of the Einstein-Maxwell equations, whose metric is \\cite{nccept65} \\begin{equation} ds^2=\\Sigma\\Delta^{-1}dr^2+\\Sigma d\\vartheta^2+\\Sigma^{-1}\\sin^2\\vartheta\\left[adt-\\left(r^2+a^2\\right)d\\varphi\\right]^2-\\Sigma^{-1}\\Delta\\left[dt-a\\sin^2\\vartheta d\\varphi\\right]^2 , \\label{KNmetric} \\end{equation} where \\begin{equation} \\Delta=r^2-2Mr+Q^2+a^2\\, ,\\quad \\Sigma=r^2+a^2\\cos^2\\vartheta\\, , \\label{KNmetric2} \\end{equation} has been one of the most powerful theoretical tools for probing spacetime structure by relating the concepts of mass $M$, charge $Q$ and angular momentum $L$, with $a=L/M$, within a relativistic field theory. This quantities have been expressed in geometrical units (see e.g. Ruffini in Ref.~\\citen{gr78}). The aim of this talk is to show how this metric is currently becoming the fundamental theoretical tool for explaining the most energetic events ever discovered in our universe: the gamma ray bursts (GRBS). Essential for doing this is to consider the quantum vacuum polarization process in a Kerr-Newmann geometry and relate it to the reversible transformations of the black hole and to its  mass-energy formula. This leads us to believe that in these events we are witnessing energy extraction from a black hole for the first time. We will take a somewhat historical approach to recover the ``crescendo\" of this field of research and recollect some of the most crucial moments which are making this discovery possible. ",
        "introduction": " ",
        "conclusions": "GRBs offer an unprecedented opportunity to probe entire new domains of physics and astrophysics, ranging from high energy particles to the fundamental physical laws and field theories in the extreme spacetimes of black holes and for the first time to give evidence of an energy extraction process for black holes.  \\begin{itemize} \\item GRBs are giving the first evidence for the vacuum polarization process in a strong electromagnetic field studied by Sauter-Heisenberg-Schwinger which for many years has been searched for without success in Earth bound accelerators (see e.g. \\cite{ga96,la97,la98,ha98}). In order to familiarize the larger scientific community of particle physicists with the extraordinary aspects of this EMBH model for GRBs, we have suggested an analogy with a high-energy accelerator. All the processes described in the previous sections can be visualized as the injector-accelerator phase of an enormous accelerator. The injector phase is characterized by the vacuum polarization process leading to the electron-positron pair creation. The accelerator phase corresponds to the PEM and PEMB pulse phases. Unlike the accelerators on the Earth (CERN, Fermilab, Dubna, etc.), where the acceleration process is due to electromagnetic field, in the cosmic GRB accelerators the acceleration originates in the positron-electron annihilation process and in the mean free path of the photons in such a plasma increasing during the optically thick PEM and PEMB pulse phases. The baryons are carried along by this $e^+e^-$ and photon plasma. This first phase, the injector-accelerator one, terminates with the reaching of the condition of transparency at which point the P-GRB and an accelerated baryonic matter (ABM) pulse are emitted. Clearly both the Lorentz gamma factor of this ABM pulse and the flux are larger then the ones already reached in Earth bound accelerators. If we compare with Fig.~\\ref{gf1} where the Lorentz gamma factors of protons at CERN are given, we realize that the Lorentz gamma factor of baryonic matter accelerated by GRBs (see Fig.~\\ref{gammab}) can be larger then the ones achievable in the forthcoming years in the Earth bound accelerators. But the enormity of the GRBs accelerators compared to the Earth bound ones is in the differences in fluxes, which can be $10^{38}$ times larger in the astrophysical setting than the ones on the surface of the Earth. The target of the P-GRB and ABM pulse generated in the injector-accelerator phases is represented by the interstellar medium (ISM) and the measuring devices are, in the astrophysical setting, at billions light years of distance, on the surface of the planet Earth. It is no surprise that a wealth of fundamental observations also in particle physics will be obtained from GRBs, in due course. \\begin{figure} \\vspace{-.5cm} \\epsfxsize=\\hsize \\begin{center} \\mbox{\\epsfbox{gf1.eps}} \\end{center} \\vspace{-0.2cm} \\caption[]{Lorentz gamma factor at CERN over the years.} \\label{gf1} \\end{figure} \\item All the computations on the ``dyadosphere'' we have carried out until now refer to an already formed Reissner-Nordstr\\\"om black hole. What we need to develop in the near future is the theoretical framework for the time varying process of the dynamical formation of the ``dyadosphere'' and the gradual approach to the horizon (see Fig.~\\ref{dyaform}). \\begin{figure} \\vspace{-.5cm} \\epsfxsize=\\hsize \\begin{center} \\mbox{\\epsfbox{dyaform.eps}} \\end{center} \\vspace{-0.2cm} \\caption[]{Spacetime diagram of the collapse process leading to the formation of the ``dyadosphere''. As the collapsing core crosses the ``dyadosphere'' radius the pair creation process starts, and the pairs thermalize into a neutral plasma configuration. Then the horizon is  also crossed and the singularity is formed.} \\label{dyaform} \\end{figure} This time varying evolution can be followed by a sequence of PEM/PEMB pulses emitted outside the collapsing core as the radius of the ``dyadosphere'' $r_{ds}$ is crossed and the horizon radius $r_+$ is approached. From a preliminary analysis we have evidence that there exists a minimum radius $r_{min}>r_+$ such that for $r_+<r<r_{min}$ the PEM/PEMB pulses will be captured by the black hole, while for $r>r_{min}$ the PEM/PEMB pulses will propagate out and will carry their information outwardly, encoded in the structure of the P-GRB. The analysis of the  P-GRB is therefore an essential tool for retracing all the general relativistic effects, including the gravitational redshift and all the time dilation effects which occur in the approach to the horizon of the EMBH. \\item There are enormous differences in the energetics, in the spectra and in the time scale of the radiation processes expected from black holes due to the Damour-Ruffini process and the Beckenstein-Hawking process. In addition to these enormous observational differences, there are also additional conceptual differences between these two processes. While the role of the horizon appear to be predominant in the Beckenstein-Hawking process, it is clear from the computation of the energy extraction process from the ``dyadosphere'' that in the Damour-Ruffini process the role of the horizon in marginal and the energy emission occurs in an extended region between the horizon $r_+$ and the radius of the ``dyadosphere'' $r_{ds}$ (see Fig.~\\ref{dyaon}). In addition, the energetics of the energy extraction from black holes is essentially made on the basis of the Christodoulou-Ruffini mass-energy formula (see Eqs.(\\ref{em}, \\ref{sa}, \\ref{s1})). There are a variety of issues still to be addressed. Among these: \\begin{enumerate} \\item What is the physical nature of the irreducible mass and of the Coulomb term in the mass-energy formula? \\item Why 50\\% efficiency in the energy extraction process can be reached in an EMBH under the condition of total reversibility\\cite{ruffc}? \\item How the reversibility condition in the Damour-Ruffini process, which occurs outside the horizon and in the entire ``dyadosphere'' region, differs from the reversibility condition considered by Christodoulou and Ruffini in the energy extraction, which is instead essentially related to the properties of horizon? \\end{enumerate} In order to answer these basic physics questions related to the energetics of black holes we need to find a ``gedanken'' process which allows us to describe continuously the transition from a collapsing core to the final asymptotic formation of the black hole. In this sense it appears particularly attractive to use the mathematical solution of the Einstein-Maxwell equations obtained by Werner Israel and his collaborators\\cite{i66,dci67} for a charged shell collapsing either on itself or on an already formed EMBH. It seems very probable that, just like the work of Brandon Carter\\cite{bc} on the geodesics of Kerr-Newman black holes has offered the essential mathematical and ``gedanken'' transformation tool which led us to the Christodoulou-Ruffini\\cite{ruffc} mass formula of black hole, the Israel collapsing shell treatment appears to be the mathematical ``gedanken'' transformation tool essential in probing the energy extraction process for an EMBH (see Fig.~\\ref{horform}). \\begin{figure} \\vspace{-.5cm} \\epsfxsize=\\hsize \\begin{center} \\mbox{\\epsfbox{horform.eps}} \\end{center} \\vspace{-0.2cm} \\caption[]{The collapse of two successive charged shells is qualitatively represented. The first one gives rise to an EMBH with a given horizon which is further increased by the accretion of the second shell. This phenomenon can be described exactly with the analytic equations given in Ref.~\\citen{crv02}.} \\label{horform} \\end{figure} The Israel formalism also allows us to evaluate the velocity of the charged shell as it crosses the ``dyadosphere'' radius and finally closes in on the horizon (see Fig.\\ref{shelvel}). \\begin{figure} \\vspace{-.5cm} \\epsfxsize=\\hsize \\begin{center} \\mbox{\\epsfbox{shelvel.eps}} \\end{center} \\vspace{-0.2cm} \\caption[]{The qualitative behavior of the velocity of a collapsing shell for selected values of the mass of the EMBH and of $\\xi$.} \\label{shelvel} \\end{figure} From the fact that the collapsing core moves inward through the ``dyadosphere'' radius at almost the speed of light, it is then clear that the collapse to a black hole, compared to all the other process of gravitational collapse, is the only one which can guarantee the state of baryonic noncontamination in the ``dyadosphere'' essential to reach the critical value of the electromagnetic field. \\end{itemize}"
    },
    "0209/astro-ph0209578_arXiv.txt": {
        "abstract": "We study the synchrotron and synchrotron self-Compton (SSC) emission from internal shocks that are responsible for the prompt $\\gamma$-ray emission in Gamma-Ray Bursts (GRBs), and consider the relation between these two components, taking into account the high energy cutoff due to pair production and Thomson scattering. We find that in order for the peak energy of the synchrotron to be $E_p\\sim 300\\;{\\rm keV}$ with a variability time $t_v\\gtrsim 1\\;{\\rm ms}$, a Lorentz factor of $\\Gamma\\lesssim 350$ is needed, implying no high energy emission above $\\sim 30\\;{\\rm MeV}$ and the synchrotron component would dominate at all energies. If we want both $E_p\\sim 300\\;{\\rm keV}$ and prompt high energy emission up to $\\sim 2\\;{\\rm GeV}$, as detected by EGRET for GRB 940217, we need $\\Gamma\\sim 600$ and $t_v\\sim 0.1\\;{\\rm ms}$, which might be resolved by super AGILE. If such prompt high energy emission is common in GRBs, as may be tested by GLAST, then for $t_v\\gtrsim 1\\;{\\rm ms}$ we need $\\Gamma\\gtrsim 350$, which implies $E_p\\lesssim 100\\;{\\rm keV}$. Therefore if X-ray flashes are GRBs with high values of $t_v$ and $\\Gamma$, they should produce $\\gtrsim 1\\;$GeV emission. For an electron power law index $p>2$, the SSC component dominates the emission above $\\sim 100\\;{\\rm MeV}$. Future observations by GLAST may help determine the value of $p$ and whether the high energy emission is consistent with a single power law (implying one component--the synchrotron, is dominant) or has a break where the $\\nu F_\\nu$ slope turns from negative to positive, which implies that the SSC component becomes dominant above $\\sim 100\\;$MeV. The high energy emission is expected to show  similar variability and time structure to that of the soft $\\gamma$-ray emission. Finally, we find that in order to see delayed high energy emission from the prompt GRB due to pair production with the cosmic IR background, extremely small inter-galactic magnetic fields ($\\lesssim 10^{-22}\\;{\\rm G}$) are required. ",
        "introduction": "\\label{sec:intro}  The leading models of Gamma-Ray Bursts (GRBs) involve a relativistic flow emanating from a compact central source, where the prompt gamma-ray emission is attributed to internal shocks within the outflow itself, that arise from variability in its Lorentz factor, while the afterglow results from an external shock that is driven into the ambient medium, as it decelerates the original ejecta (Rees \\& M\\'esz\\'aros 1994; Sari \\& Piran 1997). In this so called `internal-external' shock model, the duration of the prompt GRB is directly related to the time during which the central source is active. The most popular emission mechanism is synchrotron radiation from relativistic electrons accelerated in the shocks, that radiate in the strong magnetic fields (close to equipartition values) within the shocked plasma. An additional radiation mechanism that may also play an important role is synchrotron self-Compton (SSC), which is the upscattering of the synchrotron photons by the relativistic electrons, to much higher energies.  The synchrotron and SSC components from internal shocks have been considered in previous works in various contexts. Papathanassiou \\& M\\,esz\\,aros (1996) studied the emission from internal shocks, focusing on the comparison between internal and external shocks. Pilla \\& Loeb (1998) calculated the spectrum from internal shocks taking into account multiple Compton scattering and pair production. They show the broad band spectrum for a fixed radius of collision, $R$, and varying Lorentz factor, $\\Gamma$, of the outflow, and for a fixed $\\Gamma$ and a varying $R$. Our treatment differs in that we assume different free parameters, namely the Lorentz factor $\\Gamma$ and the variability time, $t_v$ of the central engine that emits the outflow, rather than the $\\Gamma$ and $R$. Under our assumptions, the radius of collision scales as $R\\sim 2\\Gamma^2ct_v\\propto\\Gamma^2$, and is not independent of $\\Gamma$. This results in different conclusions as to the relation between the prompt gamma-ray emission in the BATSE range and the emission at higher energies, which is the main subject of our work. Panaitescu \\& M\\,esz\\,aros (2000) explored the possibility that the prompt gamma-ray emission in the BATSE range arises from the SSC component rather than the synchrotron, where the latter is in the optical or UV range. Recently, Dai \\& Lu studied the SSC emission from internal shocks, concentrating on the possible interaction of high energy photons with the IR background (see section \\S \\ref{limits}).  In this {\\it Letter} we calculate the high energy emission during the prompt GRB from internal shocks, for both the synchrotron and SSC components, and consider the relations between these two components. We estimate the high energy cutoff and study the constraints on the model parameters that arise from the requirement that $E_p$ (the typical photon energy of the synchrotron component) will be in the BATSE range.  The synchrotron and SSC spectra are calculated in \\S \\ref{IS}, and expressions are provided for the break frequencies and flux normalization. In \\S \\ref{limits} we derive the constraints on the model that arise from the optical depth to pair production and to Thomson scattering. We consider the recent claim for a possible delayed emission due to the pair production of high energy photons with the IR background (Dai \\& Lu 2002) and we show that in order for this radiation to be detectable, a very small ($\\lesssim 10^{-22}\\;{\\rm G}$) inter-galactic magnetic field is needed. Our results are discussed in \\S \\ref{sec:dis}. ",
        "conclusions": "\\label{sec:dis}  We have calculated the synchrotron and SSC emission during the prompt GRB from internal shocks, and studied the relation between these two components and its dependence on the model parameters. Our analysis takes into account the high energy cutoffs due to the Klein-Nishina effect, pair production with low energy photons and with the cosmic IR background, and Thomson scattering.  For $p>2$ the emission above $\\sim 100\\;{\\rm MeV}$, is typically dominated by the SSC emission, while the synchrotron component is dominant at lower energies. If the variability time is $t_v\\gtrsim 1\\;{\\rm ms}$, then $E_p\\sim 300\\;{\\rm keV}$ would imply a cutoff at $\\sim 30\\;{\\rm MeV}$, and the synchrotron emission would be dominant at all energies. Future observations by GLAST may help determine the value of $p$ and whether the high energy emission is consistent with a single power law or has a break where the $\\nu F_\\nu$ slope turns from negative to positive. The former would imply that the high energy emission is dominated by a single component-- the synchrotron emission, while the latter implies that the SSC component becomes dominant above a certain energy ($\\sim 100\\;$MeV).  The SSC high energy emission should show a similar variability to that observed in the BATSE range. An additional emission mechanism that might contribute to the high energy emission, is external Compton, which may be relevant if GRBs occur inside pulsar wind bubbles (Guetta \\& Granot 2002).  In addition, there might be delayed high energy emission from the internal shocks, due to upscattering, aas was suggested by  of the CMB by $e^\\pm$ pairs produced by the interaction between $\\gtrsim 300\\;$GeV photons from the prompt GRB with the IR background photons (Dai \\& Lu 2002). The detection of such emission would be possible only for inter galactic magnetic fields (IGMF) $\\lesssim 10^{-21}\\;$G at a distance of a few Mpc from the site of the GRB, so that it may serve a probe for the strength of the IGMF in voids.  As can be seen from Eqs. (\\ref{nu_c}), (\\ref{e_max}) and Figure \\ref{fig1}, larger values of $\\Gamma$ or $t_v$ shift the cutoff at $h\\nu_{\\gamma\\gamma}$ to larger energies, while at the same time, it implies lower values of $E_p$. For example, in order to have $\\sim 1\\;{\\rm GeV}$ photons for $t_v\\sim 1\\;{\\rm ms}$ we need $\\Gamma\\gtrsim 350$, which in turn, imply $E_p\\lesssim 100\\;{\\rm keV}$. If X-ray flashes are GRBs with such parameters, as suggested by Guetta, Spada \\& Waxman (2001), then we can expect GeV emission from X-ray flashes.  In order to explain the prompt high energy photons, of up to $\\sim 3\\;{\\rm GeV}$, that were observed in GRB 940217 (Hurley et al. 1994), together with the value of $E_p\\approx 200\\;{\\rm keV}$ that was measured for this burst, we need a very small variability time $t_v\\sim 0.1\\;{\\rm ms}$, and $\\Gamma\\sim 600$ (see the solid curve in the lower panel of Figure \\ref{fig1}). If indeed such high energy emission is typical for GRBs with $E_p$ in the BATSE range, as can be tested by the future mission GLAST, this might suggest very low variability times ($t_v\\lesssim 0.1\\;{\\rm ms}$). This possibility is consistent with the fact that in many GRBs the shortest measured variability time is limited by the temporal resolution of the instrument, and there is no observational lower limit on $t_v$. On the other hand, $t_v\\sim 0.1\\;{\\rm ms}$ implies a source size $\\lesssim ct_v\\sim 30\\,t_{v,-4}\\;{\\rm Km}$, so that it is unlikely that $t_v$ can be much smaller that $0.1\\;{\\rm ms}$. Therefore, this might imply a typical variability time of $t_v\\sim 0.1\\;{\\rm ms}$, which may be resolved by super AGILE."
    },
    "0209/astro-ph0209608_arXiv.txt": {
        "abstract": "We report the results of a spectroscopic survey of 20 close T~Tauri binaries in the Taurus-Auriga dark cloud where the separations between primaries and their secondaries are less than the typical size of a circumstellar disk around a young star. Analysis of low-resolution and medium-resolution STIS spectra yields the stellar luminosities, reddenings, ages, masses, mass accretion rates, IR excesses, and emission line luminosities for each star in each pair. We examine the ability of IR color excesses, H$\\alpha$ equivalent widths, [O~I] emission, and veiling to distinguish between weak emission and classical T Tauri stars.  Four pairs have one cTTs and one wTTs; the cTTs is the primary in three of these systems. This frequency of mixed pairs among the close T~Tauri binaries is similar to the frequency of mixed pairs in wider young binaries. Extinctions within pairs are usually similar; however, the secondary is more heavily reddened than the primary in some systems, where it may be viewed through the primary's disk. Mass accretion rates of primaries and secondaries are strongly correlated, and H$\\alpha$ luminosities, IR excesses, and ages also correlate within pairs. Primaries tend to have somewhat larger accretion rates than their secondaries do, and are typically slightly older than their secondaries according to three different sets of modern pre-main-sequence evolutionary tracks.  Age differences for XZ~Tau and FS~Tau, systems embedded in reflection nebulae, are striking; the secondary in each pair is less massive but more luminous than the primary.  The stellar masses of the UY~Aur and GG~Tau binaries measured from their rotating molecular disks are about 30\\%\\ larger than the masses inferred from the spectra and evolutionary tracks.  This discrepancy can be resolved in several ways, among them a 10\\%\\ closer distance for the Taurus-Auriga dark cloud. ",
        "introduction": "A large body of work in the last two decades has established that binaries with active accretion disks are the most common way that stars form. Surveys of regions of star formation using speckle images \\citep{ghez93,lein93} and lunar occultations \\citep{simon99} have established that binaries are ubiquitous among the youngest stars. Direct imaging \\citep{mcc96,padgett99} and surveys of infrared excesses \\citep{haisch01} have discovered circumstellar disks around many newly-formed stars.  Most young circumstellar disks accrete onto their central stars, as evidenced by inverse P-Cygni line profiles \\citep{edw94} and excess emission from the accretion hot spot at optical and ultraviolet wavelengths \\citep[e.g.][]{hartigan95,gullbring98}.  To understand how stars form we must address several fundamental questions concerning binaries and their disks. One of the most basic of these is whether or not the components of a young binary have the same age. If capture plays a role in binary formation then a range of ages should occur within pairs, while if the secondary forms out of a disk around the primary then the secondary should be younger than the primary. If the relative location of primaries and their secondaries is consistent throughout the HR-diagram, it may be possible to measure the location of isochrones that must be reproduced by theoretical models of young stars.  While most pairs generally fall along isochrones derived from theoretical pre-main-sequence evolutionary tracks \\citep{hartigan94,brandner97,white99}, uncertainties in the effective temperature scale at the lowest masses and difficulties in estimating stellar luminosities, especially when individual spectra of the two stars are lacking, have made it difficult to compare ages within pairs. In some pairs infrared companions exist that are much more obscured, and therefore possibly significantly younger, than their optically visible primaries. However, extinction does not necessarily correlate inversely with age -- highly embedded companions may be obscured only temporarily by an episode of infalling envelope material. Hence, these infrared companions are not necessarily younger simply because they are more reddened \\citep{koresko97}.  Any secondary that is more luminous than its primary must also be younger, because isochrones always slope down and to the right in an HR diagram for pre-main-sequence stars.  Other important issues are whether the masses, spatial extents, lifetimes, and accretion histories of primary circumstellar disks differ from those that surround the secondaries.  Recent studies of the frequency of disks in young binaries have concluded that mixed pairs -- those where one star is a classical T~Tauri star (cTTs; stars with circumstellar disks) and the other a weak-lined T~Tauri star (wTTs; stars without detectable disks) -- are rare \\citep{prato97,duchene99}. Stars with the highest accretion rates tend to be more massive (i.e., primaries, on average, accrete more rapidly than secondaries do), but there is no obvious correlation between mass accretion rates and separations \\citep[][hereafter WG01]{wg01}.  How disks interact in binary systems when the distance between the stars is smaller than the typical size of a disk, $\\sim$ 100~AU, is a topic of much current research. Models of these close binaries typically surround the stars with a circumbinary disk, as has been imaged with adaptive optics and with HST around some systems \\citep{close98,silber00}. In DQ~Tau, it appears that the eccentric orbit of the binary leads to cyclic pulses of accretion from the circumbinary disk onto the stars \\citep{mathieu97,basri97}. In such systems the less massive star may accrete faster than the more massive star does \\citep{arty96}, a prediction that can in principle be tested observationally.  To make progress in these areas one must measure the spectral types, accretion rates, luminosities, emission line fluxes, and reddenings of each primary and secondary independently, which requires individual spectra and photometry for each star. The stellar luminosities and effective temperatures, when compared with predictions of published pre-main-sequence evolutionary tracks, provide the masses and ages of each star. If the primaries are, on average, coeval with their secondaries, it should be possible to test the accuracy of the theoretical tracks, which have been undergoing significant revisions as the assumption of grey photospheres has been relaxed \\citep{siess00,baraffe98,palla99}. It is also possible to test the evolutionary tracks if there is an independent measurement of the stellar masses from orbital motion \\citep{ghez95,stef01} or from Keplerian rotation of the molecular disk \\citep{guill99,simon00}.  In this study we focus on 20 close pairs where the projected separations are $\\lesssim$ 100~AU. Even in the Taurus-Auriga complex, the closest significant population of cTTs, these size scales are all subarcsecond and most cannot be resolved from the ground. Our spectroscopic study extends the work of WG01, who mainly used photometric HST observations to investigate close binaries. By combining both low-resolution and medium-resolution spectra together we can measure spectral types, reddenings, accretion and emission line properties of each component of each pair accurately. Together with the WG01 photometry, the new spectra give the best overview to date of how young close binary systems interact with their disks. ",
        "conclusions": "This paper reports the results of a spectroscopic survey of 20 close (subarcsecond) binary T Tauri stars in the Taurus-Auriga dark cloud done with STIS on the Hubble Space Telescope. By obtaining spectra of each component it is possible to determine the reddening, spectral types, and stellar luminosities of the primaries and secondaries in all the pairs. This information suffices to place each star in the HR diagram, which determines its mass and age. When combined with disk characteristics such as mass accretion rates measured from veiling emission, and including emission line luminosities and infrared color excesses, it is possible to address many of the outstanding questions regarding binary formation in systems where the stellar separations are less than the extent of the circumstellar disks.  Such systems are not only interesting from the standpoint of accretion dynamics, but also make up the most common mode of star formation in our galaxy.  Components of close binaries share common characteristics in a number of ways. Correlations in mass accretion rates particularly stand out -- the best place to look for a rapidly accreting secondary is next to a rapidly accreting primary. The actual values of the accretion rates are remarkably similar between primaries and secondaries. Because secondaries are less luminous than their primaries in most cases, it is often easier to see accretion signatures such as emission lines and veiling around secondaries than it is around primaries. On average, secondaries actually accrete more per unit mass than their primaries do. Other accretion signatures such as H$\\alpha$ emission and color excesses also correlate within pairs, probably as a result of the correlation of these quantities with mass accretion rates. Extinctions of primaries correlate well with those of their secondaries, as expected.  We have examined four independent signatures of disks: H$\\alpha$ emission, K$-$L color excesses, veiling and [O~I] in each of our objects, and have devised rules each of these signatures must satisfy to classify a star as a cTTs. Most stars are clearly a cTTs or a wTTs, but several have at least one misleading indicator, most often [O~I] although each of the four signatures is inaccurate in at least one case. Overall, the frequency of mixed pairs, where one star is a cTTs and one a wTTs, comprise about the same fraction among close pairs as they do among wider pairs.  Several subtle trends have emerged from the data set. When extinctions differ within a pair it is usually the secondary that is more heavily obscured. One can explain this result geometrically by having a larger, more opaque disk around the primary than exists around the secondary. Ages of primaries and secondaries also clearly correlate, and are equal to within the errors in most cases. However, there is a clear tendency for secondaries to be a bit younger than their primaries. Slightly flatter isochrones in the pre-main-sequence tracks, perhaps caused by ongoing accretion, would remove this bias. However, pairs such as FS~Tau and XZ~Tau have more luminous secondaries that will be very difficult to make coeval with their primaries for any set of tracks.  Masses of the pairs inferred from their locations in the HR diagram are $\\sim$ 30\\% less than those measured from rotating Keplerian disks or from the orbital motion within the binary. These discrepancies are most easily removed by reducing the distance to Taurus by about 10\\%\\ to $\\sim$ 126~pc, but can also be accomplished in other ways, such as altering the effective temperature scale for cool photospheres or by generating a substantially different set of pre-main-sequence tracks."
    },
    "0209/astro-ph0209322_arXiv.txt": {
        "abstract": "We observed the 2001 November superoutburst of CC Cnc.  This observation makes the first detailed coverage of a superoutburst of this object. The best-determined mean superhump period is 0.075518$\\pm$0.000018 d, which is 2.7\\% longer than the reported orbital period.  This fractional superhump excess is a quite typical value for a normal SU UMa-type dwarf nova, excluding the previously raised possibility that CC Cnc may have an anomalously large fractional superhump excess. During the superoutburst plateau, the object showed a decrease of the superhump period at $\\dot{P}/P$ = $-$10.2$\\pm$1.3 $\\times$ 10$^{-5}$, which is one of the largest negative period derivative known in all SU UMa-type dwarf novae. ",
        "introduction": "Dwarf novae are a class of cataclysmic variables (CVs), which are close binary systems consisting of a white dwarf and a red dwarf secondary transferring matter via the Roche lobe overflow.  A class of dwarf novae, called SU UMa-type dwarf novae, show superhumps during their long, bright outbursts (superoutbursts).  [For a recent review of dwarf novae and SU UMa-type dward novae, see \\citet{osa96review} and \\citet{war95suuma}, respectively.]  Superhumps have periods a few percent longer than the orbital periods (\\cite{vog80suumastars}; \\cite{war85suuma}), which is believed to be a consequence of the apsidal motion of an eccentric accretion disk \\citep{osa85SHexcess}. The fractional superhump excess ($\\epsilon=P_{\\rm SH}/P_{\\rm orb}-1$, where $P_{\\rm SH}$ and $P_{\\rm orb}$ are superhump and orbital periods, respectively) is widely believed to be an excellent measure of the mass ratio ($q=M_2/M_1$) of the binary system both from theoretical calculations (\\cite{osa85SHexcess}; \\cite{hir90SHexcess}; \\cite{lub91SHa}; \\cite{lub91SHb}; \\cite{mur98SH}; \\cite{mur00SHprecession}; \\cite{woo00SH}; \\cite{mon01SH}) and observations (\\cite{mol92SHexcess}; \\cite{min92BHXNSH}; \\cite{pat98evolution}; \\cite{odo00SH}).  Most of SU UMa-type systems are on a tight relation (originally discovered by \\citet{StolzSchoembs} and extended by various authors, e.g. \\cite{tho96Porb}) between $P_{\\rm SH}$ and $\\epsilon$, which is considered to be a natural consequence that most of CVs have non-evolved low-mass secondary stars (cf. \\cite{pat84CVevolution}), i.e. $M_2$ is a strong function of $P_{\\rm orb}$, which mostly determines $q$.  Most recently, an SU UMa-type dwarf nova (1RXS J232953.9+062814: \\cite{uem01j2329iauc}) is found to conspicuously violate this relation \\citep{uem02j2329letter}. Subsequent spectroscopy revealed that this object has a secondary star more massive and evolved than what is expected for the orbital period \\citep{tho02j2329}.  Departures from this $P_{\\rm SH}$ vs. $\\epsilon$ relation are thus candidate systems with unusual stellar parameters.  CC Cnc [see \\citet{kat97cccnc} for a historical review of this object] is one of such candidates which was reported to have a significantly large $\\epsilon$=4.9$\\pm$0.5 \\% \\citep{tho97vzpyxcccncaheri}, who reported $P_{\\rm orb}$ = 0.07352(5) d. Since accurate determination of the superhump period of CC Cnc was difficult owing to unfavorable seasonal occurrences of the past superoutbursts \\citep{kat97cccnc}, a further check of the superhump period throughout a superoutburst under favorable condition has been absolutely needed \\citep{tho97vzpyxcccncaheri}.  An excellent opportunity arrived when the system underwent a superoutburst in 2001 November.  This outburst enabled us to for the first time follow the entire superoutburst.  The observation started within 2.5 d of the outburst detection by Mike Simonsen (visual magnitude 13.2 on November 10). ",
        "conclusions": "\\subsection{Mean Superhump Period and Profile}\\label{sec:psh}  We performed period analysis using Phase Dispersion Minimization \\citep{PDM} to all the data between 2001 November 12 and 19, after removing the systematic trend of decline.  A correction of 0.220 mag has been added for the 2001 November 12 data in order to correct the systematic offset from the linear fit.  This offset was most likely a result from a systematic difference caused by a different telescope only on this night. The resultant $\\theta$-diagram and the phase averaged profile of superhumps are shown in figures \\ref{fig:pdm} and \\ref{fig:avesh}, respectively. The best-determined superhump period is 0.075518$\\pm$0.000018 d.  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig2.eps} \\end{center} \\caption{Period analysis of superhumps in CC Cnc.  The analysis was done for the data between 2001 November 12 and 19 (during the superoutburst plateau).} \\label{fig:pdm} \\end{figure}  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig3.eps} \\end{center} \\caption{Phase-averaged light curve of CC Cnc superhumps.} \\label{fig:avesh} \\end{figure}  \\subsection{Development of Superhumps}\\label{sec:shprof}  Figure \\ref{fig:nightave} shows nightly averaged profiles of superhumps during the plateau stage of the superoutburst. The amplitude of superhumps reached a maximum (0.21--0.24 mag) around November 15--16, five days after the start of outburst.  This development of superhumps is relatively slow compared to other SU UMa-type dwarf novae [one of the best examples can be found in \\citet{sem80v436cen}; see also \\citet{vog80suumastars} and \\citet{war85suuma} for general descriptions; this delay is theoretically explained as a growth time of the tidal instability \\citep{lub91SHa}]. Although the phase coverage was not complete because of unfavorable sky condition, the amplitude of superhumps seems to have once decayed on November 17, and again grew on November 18. Such a regrowth of the superhump amplitude may be related to a phenomenon observed during the late stage of a superoutburst in V1028 Cyg \\citep{bab00v1028cyg}\\footnote{ We must note that ER UMa stars (a small, peculiar subgroup of SU UMa-type dwarf novae with extremely short supercycles; presently known members being ER UMa, V1159 Ori, RZ LMi, DI UMa and IX Dra (\\cite{kat95eruma}; \\cite{rob95eruma}; \\cite{nog95rzlmi}; \\cite{kat96diuma}; \\cite{ish01ixdra}) show a similar pattern of decay and regrowth of superhumps \\citep{kat96erumaSH}.  CC Cnc, however, has a much longer supercycle ($\\sim$400 d) than those of ER UMa stars (19--45 d), indicating that CC Cnc has a much lower mass-transfer rate than in ER UMa stars.  Although detailed mechanisms of regrowth is not yet identified, we consider that different mechanisms of superhump regrowth may be naturally taking place between ER UMa stars and other SU UMa-type dwarf novae. }. Alternately, this phenomenon seen in CC Cnc may be also interpreted as a result of the beat phenomenon between the superhump and orbital period (most evidently seen in eclipsing systems; e.g. \\cite{vog82zcha}; \\cite{krz85oycarsuper}), as was prominently seen even in a non-eclipsing system RZ Leo \\citep{ish01rzleo}. The calculated beat period  \\begin{equation} P_{\\rm beat}={1 \\over {1/P_{\\rm orb}-1/P_{\\rm SH}}} = 2.8~{\\rm d} \\end{equation}  close to the observed time-scale of the regrowth may suggest a stronger possibility of the second interpretation.  In this case, the orbital inclination of CC Cnc is expected to be high, which would provide an excellent opportunity in spectroscopically determining the component masses and other orbital parameters.  \\begin{figure} \\begin{center} \\FigureFile(88mm,110mm){fig4.eps} \\end{center} \\caption{Evolution of CC Cnc superhumps during the plateau stage of the superoutburst.  Each point represents an average of a 0.02 phase bin, except for November 12 data which used 0.04 phase bin. The phase zero corresponds to the zero-phase epch of equation \\ref{equ:reg1}.  The mean superhump period (0.075518 d) was used to calculate the phases. } \\label{fig:nightave} \\end{figure}  \\subsection{$O-C$ Changes}  We determined the maximum times of superhumps from the light curve by eye. The averaged times of a few to several points close to the maximum were used as representatives of the maximum times.  The errors of the maximum times are usually less than $\\sim$0.004 d, which corresponds to the maximum lengths of the data bins (i.e. a few to several points) to deduce the maximum times. We did not use cross-correlation method to obtain individual maxima because the profile of superhumps was rather strongly variable (subsection \\ref{sec:shprof}).  The resultant superhump maxima are given in table \\ref{tab:shmax}.  The values are given to 0.0001 d in order to avoid the loss of significant digits in a later analysis.  The cycle count ($E$) is defined as the cycle number since Barycentric Julian Date (BJD) 2452226.322 (2001 November 12.822 UT).  A linear regression to the observed superhump times gives the following ephemeris:  \\begin{equation} {\\rm BJD (max)} = 2452226.3315 + 0.0755135 E. \\label{equ:reg1} \\end{equation}  \\begin{table} \\caption{Times of superhump maxima.}\\label{tab:shmax} \\begin{center} \\begin{tabular}{ccc} \\hline\\hline $E$$^*$ & BJD$-$2400000 & $O-C$$^\\dagger$ \\\\ \\hline 0 & 52226.3223 & $-$0.0092 \\\\ 11 & 52227.1548 & $-$0.0073 \\\\ 12 & 52227.2362 & $-$0.0015 \\\\ 23 & 52228.0679 & $-$0.0004 \\\\ 24 & 52228.1414 & $-$0.0024 \\\\ 25 & 52228.2222 &   0.0029  \\\\ 26 & 52228.2972 &   0.0023  \\\\ 29 & 52228.5276 &   0.0062  \\\\ 37 & 52229.1284 &   0.0029  \\\\ 38 & 52229.2048 &   0.0038  \\\\ 39 & 52229.2802 &   0.0037  \\\\ 42 & 52229.5049 &   0.0018  \\\\ 51 & 52230.1848 &   0.0021  \\\\ 64 & 52231.1680 &   0.0036  \\\\ 79 & 52232.2979 &   0.0008  \\\\ 80 & 52232.3738 &   0.0012  \\\\ 90 & 52233.1287 &   0.0010  \\\\ 91 & 52233.2027 & $-$0.0005 \\\\ 92 & 52233.2810 &   0.0023  \\\\ 104 & 52234.1817 & $-$0.0032 \\\\ 105 & 52234.2563 & $-$0.0041 \\\\ 106 & 52234.3305 & $-$0.0054 \\\\ \\hline \\multicolumn{3}{l}{$^*$ Cycle count since BJD 2452226.322.} \\\\ \\multicolumn{3}{l}{$^\\dagger$ $O-C$ calculated against equation \\ref{equ:reg1}.} \\\\ \\end{tabular} \\end{center} \\end{table}  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig5.eps} \\end{center} \\caption{$O-C$ diagram of superhump maxima. The parabolic fit corresponds to equation \\ref{equ:reg2}.} \\label{fig:oc} \\end{figure}  Figure \\ref{fig:oc} shows the ($O-C$)'s against the mean superhump period (0.0755135 d).  The diagram clearly shows the decrease in the superhump period throughout the superoutburst plateau. The times of the superhump maxima in this interval can be well represented by the following quadratic equation (the quoted errors represent 1-$\\sigma$ errors):  \\begin{eqnarray} {\\rm BJD} & {\\rm (max)} = 2452230.4892(7) + 0.075531(13) (E-55) \\nonumber \\\\ & -3.86(50) \\times 10^{-6} E^2. \\label{equ:reg2} \\end{eqnarray}  The quadratic term corresponds to $\\dot{P}$ = $-$7.7$\\pm$1.0 $\\times$ 10$^{-6}$ d cycle$^{-1}$, or $\\dot{P}/P$ = $-$10.2$\\pm$1.3 $\\times$ 10$^{-5}$. \\citet{kat01hvvir} noted that short-period systems or infrequently outbursting SU UMa-type systems predominantly show an increase in the superhump periods in contrast to a ``textbook\" decrease of the superhump periods in usual SU UMa-type dwarf novae.  However, observations of period changes in long $P_{\\rm orb}$ systems are relatively lacking in the literature.  Considering that the longer $P_{\\rm orb}$ systems have larger (i.e. closer to zero) $\\dot{P}/P$ \\citep{kat01hvvir}, or even virtually zero (e.g. V725 Aql: \\cite{uem01v725aql}; EF Peg: K. Matsumoto in preparation, see also \\citet{kat02efpeg}), there may be a possibility that $\\dot{P}/P$ makes a minimum around the period of CC Cnc. From a theoretical viewpoint, this decrease of superhump period is generally attributed to decreasing apsidal motion due to a decreasing disk radius \\citep{osa85SHexcess}, or inward propagation of the eccentricity wave \\citep{lub92SH}.  It may be possible these ``intermediate period\" systems like CC Cnc enable effective propagation of the eccentricity wave, although the possibility needs to be tested by future detailed fluid calculations.  \\subsection{Superhumps during the Rapid Decline Phase}  In some SU UMa-type dwarf novae, what are called {\\it late superhumps} appear during the final stage of superoutbursts.  Late superhumps have similar periods with ordinary superhumps (i.e. superhumps observed during the plateau stage, subsections \\ref{sec:psh}, \\ref{sec:shprof}), but have phases of $\\sim$0.5 different from those of ordinary superhumps (\\cite{hae79lateSH}; \\cite{vog83lateSH}; \\cite{vanderwoe88lateSH}; \\cite{hes92lateSH}).  Figure \\ref{fig:late} shows the late-stage evolution of superhumps in CC Cnc.  On November 20, the system started to decline rapidly.  Ordinary superhumps were clearly present, without a hint of $\\sim$0.5 phase jump.  On November 21, the system further faded by $\\sim$1.0 mag.  Although the profile of variation became more irregular, the maximum phase remained close to zero, suggesting that late superhumps were weak in this system.  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig6.eps} \\end{center} \\caption{Superhumps during the rapid decline phase (November 20--21). The upper panels show raw light curves.  The lower panels show averaged superhump profiles.  The phase-averaging follow the same prescription in figure \\ref{fig:nightave}, after subtracting the linear decline trend from the raw data. } \\label{fig:late} \\end{figure}"
    },
    "0209/astro-ph0209114_arXiv.txt": {
        "abstract": "High-resolution imaging and long-slit spectroscopy obtained with HST, combined with ground-based integral-field spectroscopy, provides the kinematics of stars and gas in nearby galactic nuclei with sufficient accuracy to derive the intrinsic dynamical structure, and to measure the mass of the central black hole. This has revealed that many nuclei contain decoupled kinematic components and asymmetric structures, and that nuclear and global properties of galaxies are correlated.  Higher spatial resolution and significantly increased sensitivity are required to cover the full range of galaxy properties and types, including the nearest powerful active radio galaxies, and to study the evolution of galactic nuclei as a function of redshift.  The prospects in this area are discussed. ",
        "introduction": "The past decade has seen a revolution in instrumentation for spatially resolved spectroscopy of galaxies and their nuclei. At many observatories, traditional long-slit spectrographs have been replaced by integral-field devices which produce spectra over an area, fully spatially sampled, in many cases taking advantage of adaptive optics capabilities (see, e.g., Emsellem \\& Bland--Hawthorn 2002 for a recent summary).  These instruments are very efficient in use of telescope time, and allow complete reconstruction of the intensity distribution from the spectra, so that errors in the registration of the spectra relative to the galaxy image, familiar from aperture and long-slit spectroscopy, are avoided. ",
        "conclusions": "The combination of HST imaging and ground-based integral-field spectroscopy is very powerful for the study of the nuclei of normal and active galaxies, and provides a compelling argument for integral-field capabilities in space.  Much progress in our understanding of galactic nuclei is expected on the ground in the next decade, and this may continue {\\em if} near-diffraction limited observations become possible on telescopes with apertures of 30m or larger. While the study of galactic nuclei is not the main science driver for a 6-8m class space telescope, even a smaller aperture equipped with an integral-field spectrograph would yield a rich scientific harvest."
    },
    "0209/astro-ph0209391_arXiv.txt": {
        "abstract": "We present  $(V,I)$ photometry of two wide ($\\simeq 25 \\times 25$ arcmin$^2$) fields centered on the low surface brightness dwarf spheroidal galaxies Draco and Ursa Minor. New estimates of the distance to these galaxies are provided ($(m-M)_0(UMi)=19.41 \\pm 0.12$ and $(m-M)_0(Dra)=19.84 \\pm 0.14$) and a comparative study of their evolved stellar population is presented. We detect for the first time the RGB-bump in the Luminosity Function of UMi ($V_{RGB}^{Bump}=19.40\\pm 0.06$) while the feature is not detected in Draco. Photometric metallicity distributions are obtained for the two galaxies and an accurate analysis to determine the intrinsic metallicity spread is performed by means of artificial stars experiments. The adopted method is insensitive to stars more metal poor than $[Fe/H]\\sim -2.5$ and it rests on the assumpion that the age spread in the considered populations is small (i.e. the impact of the actual age spread on the colors of the RGB stars is negligible). We find that, while the average metallicity of the two galaxies is similar ($<[Fe/H]>_{UMi} = -1.8$ and $<[Fe/H]>_{Dra} = -1.7$) the metallicity distributions are significantly different, having different peak values ($[Fe/H]_{UMi}^{mod} = -1.9$ and $[Fe/H]_{Dra}^{mod} = -1.6$) and different maximum metallicities. We suggest that such differences may be partly responsible for the difference in HB morphology between the two galaxies. The intrinsic metallicity $1-\\sigma$ spread is $\\sigma_i=0.10$ in UMi and $\\sigma_i=0.13$ in Draco. We demonstrate that the inner region of UMi is significantly structured, at odds with what expected for a system in dynamical equilibrium. In particular we show that the main density peak of UMi is off-centered with respect to the center of symmetry of the whole galaxy and it shows a much lower ellipticity with respect to the rest of the galaxy. Moreover, UMi stars are shown to be clustered according to two different characteristic clustering scales, as opposite to Draco, which instead has a very symmetric and smooth density profile. The possible consequences of this striking structural difference on our ideas about galaxy formation are briefly discussed. Combining our distance modulus with the more recent estimates of the total luminosity of UMi, we find that the mass to light (M/L) ratio of this galaxy may be as low as $M/L\\sim 7$, a factor 5-10 lower than current estimates. ",
        "introduction": "The Draco ($\\alpha_{2000}= 17^h 20^m 19^s$, $\\delta_{2000} = 57^{o} 54.8^{'}$) and Ursa Minor ($\\alpha_{2000}= 15^h 09^m 11^s$, $\\delta_{2000} = 67^{o} 12.9^{'}$) dwarf spheroidal (dSph) galaxies are the faintest known members of the Local Group of galaxies and they are among the lowest surface brightness members of the group \\citep{m98}.  They appear to be  dominated by very old (age $> 8-10$ Gyr) and metal deficient ($[Fe/H]\\sim -2$) stellar populations \\cite[see][]{m98,dradeep,apadra,grill,ken,dol7}, thus they may represent a very early and elementary stage of the evolution of the building blocks that may have had a primary role in the assembly of the Milky Way galaxy \\cite[see][]{bfpsex}. Furthermore, they are reported to have the highest Mass to Light (M/L) ratio of any other known galaxy \\cite[e.g., up to $M/L=300 - 1000$ for Draco, and $M/L\\sim 50-100$, for UMi, see] []{kle01a,taft}. Hence their stellar content may just represent the handful of baryons trapped in a system whose true nature is that of a huge dark halo. All these exceptional properties have made these galaxies a classical case of study. Nevertheless, serious observational problems (e.g., the very low surface brightness that requires very wide field photometry to obtain statistically significant samples of members stars) have hampered our knowledge of these intriguing stellar systems.  Draco and UMi are twin galaxies under many aspects: they have a similar distance from the center of the Galaxy, similar masses and luminosities, similar metal content and are both devoid of gas (see Table~3 for a summary of their physical parameters). In this context, a comparative study performed with strictly homogeneous observational material and with the same data reduction / data analysis techniques may reveal interesting features. Here we present the results of a comparative analysis, performed by obtaining well calibrated (V,I) photometry of the evolved stars of Umi and Draco over a field of view $\\sim 25 \\times 25$ arcmin$^2$, under strictly homogeneous conditions \\cite[see][for a discussion of the possible problems associated with the comparison of non-homogeneous photometries]{mb01}.  The plan of the paper is as follows: in \\S2 we describe the observational material, the data reduction process and the photometric calibration; in \\S3 we present the Color Magnitude Diagrams (CMD) and the results of artificial stars experiments; \\S4 is devoted to the estimate of the distance to Draco and Umi; in \\S5 we study the properties of the Red Giant Branch of the two galaxies. In \\S6 we compare the structures of Dra and UMi and we demonstrate that the inner region of UMi is significantly structured and that its stars are clustered according to two different characteristic scale-lenghts. \\S7 is dedicated to the discussion of our results. ",
        "conclusions": "The distance estimates to Draco and UMi derived in \\S4 are tied to the distance scale introduced by F99  and compatible with the TRGB scale introduced by \\citet{bfptip}. The F99 scale is fully consistent with the scale based on the revised Hipparcos parallaxes by \\citet{car00}. F99 showed that their distance moduli of globular clusters are tipically $\\sim 0.2$ mag larger than those tied to traditional $M_V(RR) - [Fe/H]$ calibrating relation. This is probably the main reason why we find distance moduli for UMi and Draco that are larger by $0.2 - 0.3$ mag than what generally found in literature \\cite[see, e.g.][and references therein]{m98}. We note also that, independently of the distance scale,  our procedure of HB matching leads to results that are in excellent agreement with works based on the RR Lyrae \\cite[e.g., N88][]{nemdra} while a sensible mismatch is noticed with respect to the Main Sequence fitting technique adopted by \\citet{ken}, using HST data. We argue that the problems with the absolute photometric calibration of HST-WFPC2 data may be at the origin of the observed differences.  The comparison with the recent distance estimates by \\citet{apadra} (for Draco) and \\citet{car02} (for UMi) is of particular interest. \\citet{apadra} derived a distance modulus to Draco by (1) estimating $V_{RR}$ from the observed HB and (2) by adopting the classical $M_V(RR)-[Fe/H]$ relation by \\citet{lee}. They obtain $V_{RR}=20.14 \\pm 0.12$, compatible with our estimate within the uncertainties, but their final distance modulus ($(m-M)_0=19.5 \\pm 0.2$) is 0.34 mag smaller than ours. However, if the above quoted systematic difference of $\\sim 0.2$ mag between the F99 distance scale and the scale by \\citet{lee} is taken into account the consistency between the two estimates is clearly recovered. On the other hand, \\citet{car02} derived the mean level of RR Lyrae in UMi by a comparison with template globular cluster of similar metallicity, i.e. the same method adopted here. Moreover, the distance moduli of the template clusters were taken from \\citet{reid} who derived them from Main Sequence fitting to Hipparcos subdwarfs. With this approach \\citet{car02} obtained $V_RR = 19.84\\pm 0.07$ and $(m-M)_0=19.40 \\pm 0.10$ in {\\em excellent} agreement with our results.   Independently of the above considerations, it is important to remark that the actual uncertainty on state-of-the-art distance estimates to Draco and UMi remains $\\sim \\pm 25$ \\%, in spite of all efforts. It is clear that classical distance indicators are unefficient in this context and valid alternatives  are certainly needed. The search for detached double-lined eclipsing binaries (DD-EB) is now possible with new generation instruments and may be very rewarding, given the high binary fraction that has been reported for these two galaxies \\citep{olsbin}.  \\subsection{Evolutionary history}  The metallicity distributions presented here confirm (with larger samples, with respect to existing spectroscopic surveys) that the stellar populations of both galaxies have a sizeable spread in metal content. Since they are dominated by very old stars, we shall conclude that significant self-enrichment took place in these systems in a quite short time scale (i.e., few Gyr) at very early epochs. The same conclusions have been previously reached for Sculptor \\citep{sculp} and Sextans \\citep{bfpsex}. \\citet{shet2} have shown that some of the stars of UMi, Dra and Sex show abundance patterns significantly different from those typical of Galactic halo stars. In particular it has been suggested \\citep{ikuta} that the transition from an $\\alpha$ -enhanced regime to nearly solar $[\\alpha/Fe]$ ratios occurred at a much lower [Fe/H] value than in our own Galaxy. This evidence led  \\citet{shet2} to conclude that it is unlikely that systems like UMi and Draco gave a significant contribution to the assembly of the Galactic halo. We remark here that this conclusion is correct only for satellites that would have been accreted by the Milky Way in an advanced stage of their chemical evolution. A very early accretion (i.e., within $\\sim 1$ Gyr from the onset of star formation) or simply a very early stripping of the gaseous component by the Galaxy wouldn't leave any peculiar chemical signature in a Galactic halo largely composed by building blocks like Draco, UMi or Sextans (as they were in their first Gyr of ``stellar'' life).  Our results show also that, while the average metallicity of the two galaxy is similar, the difference in the MDs is significant. The MD of UMi peaks at $[Fe/H]\\simeq -1.9$ and barely reaches $[Fe/H]\\simeq -1.6$ while that of Draco peaks at $[Fe/H]\\simeq -1.6$ and reaches $[Fe/H]\\simeq -1.3$. This fact has to be taken into account, in particular in the interpretation of the great difference in HB morphology between the two galaxies. Dra and UMi {\\em are not} a good ``second parameter pair'' \\cite[see][and references therein]{mb01}: a sizeable part of the HB morphology difference may be due to the differences in the actual MD. The large population of binaries (and Blue Stragglers) hosted by the galaxies \\citep{olsbin} may also be have some impact on the observed HB morphology, especially in the case of Draco \\cite[see][and references therein]{ffp92,mb01,mb02a,carney}.  \\subsection{A striking structural difference: not so twins after all?}  Draco and UMi have similar distance from the center of the Galaxy, similar luminous mass and similar star formation histories. Both are thought to have high M/L ratios and probably have also similar orbits, since they have similar radial velocities and (possibly) proper motions \\cite[see][]{m98,SH94,cud}. In spite of that, while Draco shows a very symmetrical and smooth profile out to a very large distance from the center \\citep{sdss,piatek}, we have demonstrated that {\\em even the innermost regions} of UMi are strongly structured and asymmetrical. The presence of a massive and extended CDM halo should have inhibited the formation of any significant substructure in the stellar component of UMi. The star formation activity cannot be responsible of  the disturbed structure of this galaxy since significant star formation episodes have ceased many Gyr ago. Furthermore we accurately checked for possible differences in the stellar populations between the off-centered (and nearly round) density peak of UMi and its outer regions and we found none.  \\citet{omega} showed that a small self-gravitating stellar system is embedded into the main body of the peculiar globular cluster $\\omega$ Cen. This stellar system has likely managed to preserve its individuality over many Gyr, while orbiting into the stellar halo of the cluster. A similar possibility could be envisaged also to explain the origin of the peculiar density peak of UMi. The process of hyerarchical merging is expected to be scale-free, thus the records of the formation of a galaxy like UMi from smaller fragments may be still detectable today \\citep[see][]{kroupa}, possibly favoured by the fact that the putative fragment we are considering is much denser than the stellar medium it is embedded in. The main argument against this hypothesis is the already reported observational evidence that the stellar population in the density peak does not differ from the average UMi population \\cite[as opposite to the $\\omega$ Cen case, see][and references therein]{omega}. Large high-resolution spectroscopic surveys may possibly provide more stringent indications on the viability of this scenario.  While we are writing, a new preprint was posted \\citep{palma} in which our results about the structure of UMi are fully confirmed\\footnote{However the clustering properties of UMi stars are not considered in that work.}. Furthermore \\citet{palma} were able to follow the structure of UMi out to a distance of $\\sim 1.7$ deg from the center and demonstrated that (1) the galaxy is elongated along the direction of its proper motion vector and (2) the outer isodensity contours of UMi have the typical S-shape of tidally disturbed stellar systems. If it will be spectroscopically confirmed that the stars in the S-shaped structures are unbound, or the signature of apparent rotation will be found, this would directly disprove a specific prediction of the CDM simulations \\citep{haya}. In any case, the strongly disturbed structure of UMi provides a severe challenge to standard CDM scenarios, above all if the comparison with Draco is considered\\footnote{In fact, the presently available observations of Draco are very well fitted by the predictions of CDM models \\citep{haya}}. Moreover, \\citet{palma} showed that previous estimates of the total luminosity of UMi had missed a significant amount of light that is found at large distances from the center of the galaxy. Their estimate of total luminosity is 2.7 times larger than previous ones. Taking this into account and considering the (possible) effects of anisotropic velocity dispersions they were able to reduce the mass to light ratio of UMi down to $M/L\\simeq 16$. If we include in this computation the effects of our larger distance modulus \\cite[and assuming $M_{tot}(UMi)=2.3\\times 10^7 M_{\\odot}$, after][]{m98} we obtain $M/L\\simeq 7$, just a factor $\\sim$ 4 larger  than the typical M/L of ordinary globular clusters.  From the above discussion it appear that further observational and theoretical efforts are still needed to fully understand the nature of UMi and Draco. It is quite possible that we have some fundamental lesson to learn  from these faint and unassuming stellar systems."
    },
    "0209/astro-ph0209358_arXiv.txt": {
        "abstract": "Power spectra of the light curves of semiregular variables, based on visual magnitude estimates spanning many decades, show clear evidence for stochastic excitation.  This supports the suggestion by Christensen-Dalsgaard et al.\\ (2001) that oscillations in these stars are solar-like, i.e., stochastically excited by convection, with mode lifetimes ranging from years to decades. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209444_arXiv.txt": {
        "abstract": "{ We present WSRT observations of high sensitivity and resolution of the neutral hydrogen in the starburst dwarf galaxy NGC~1569.  Assuming a distance of 2.2 Mpc, we find a total HI mass of $1.3 \\times 10^8\\ \\rm M_\\odot$ to be distributed in the form of a dense, clumpy ridge surrounded by more extended diffuse HI containing a few additional discrete features, such as a Western HI Arm and an HI bridge reaching out to a small counterrotating companion cloud. About 10\\% by mass of all HI in NGC~1569 is at unusually high velocities. Some of this HI may be associated with the mass outflow evident from H$\\alpha$ measurements, but some may also be associated with NGC~1569's HI companion and intervening HI bridge, in which case, infall rather than outflow might be the cause of the discrepant velocities. No indication of a large bubble structure was found in position-velocity maps of the high-velocity HI. The galaxy as a whole is in modest overall rotation, but the HI gas lacks any sign of rotation within $60''$ (0.6 kpc) from the center, i.e.  over most of the optical galaxy. Here, turbulent motions resulting from the starburst appear to dominate over rotation. In the outer disk, the rotational velocities reach a maximum of $35 \\pm 6\\ \\rm km\\ s^{-1}$, but turbulent motion remains significant. Thus, starburst effects are still noticeable in the outer HI disk, although they are no longer dominant beyond 0.6 kpc. Even excluding the most extreme high-velocity HI clouds, NGC~1569 still has an unusually high mean HI velocity dispersion of $\\sigma_v=21.3\\rm\\ km\\ s^{-1}$, more than double that of other dwarf galaxies.  \\\\ ",
        "introduction": "NGC~1569 is a small Im type galaxy at a distance of $\\rm\\ 2.2 \\pm 0.6\\ Mpc$ (Israel 1988), and a probable member of the IC 342 group (Huchtmeier et al. 2000). With a maximum optical size of 2.9$'$ (1.85 kpc), NGC~1569 is dominated by the aftermath of a burst of star formation. Assuming  a Salpeter IMF with slope 2.35 and lower and upper mass cut-offs at 0.1 and 120 $\\rm M_\\odot$ respectively, Greggio et al. (1998) found a star formation rate of 0.5 $\\rm M_\\odot\\ yr^{-1}$ with little change over the past 150 Myr. Taking into account the limited size and mass of NGC 1569, this is a very high rate (Israel 1980). Although the present star formation rate is still high, the most intense starburst phase occurred between 5 and 10 Myr ago (Israel \\& de Bruyn 1988; Vallenari \\& Bomans 1996; Greggio et al. 1998). The high star formation rates imply a type II supernova production corresponding to a total of $2-3 \\times 10^5$ over the past 100 Myr in this small volume of space.  Outflow of gas from NGC~1569 is inferred from the kinematics of an extended system of H$\\alpha$ filaments (De Vaucouleurs et al. 1974, Waller 1991, Heckman et al. 1995, Martin 1998) and from extended X-ray emission (Heckman et al. 1995). NGC~1569 is surrounded by an extended halo of relativistic electrons emitting synchrotron radiation (Israel \\& De Bruyn 1988).  NGC~1569 contains three very luminous, compact star clusters designated NGC~1569 A, B and C.  NGC~1569 A may be a double cluster; the mass of its brightest component is about $3 \\times 10^{5}\\ \\rm M_\\odot$ (Ho \\& Filippenko 1996, De Marchi et al. 1997). Given its mass, luminosity and colour, NGC~1569 A will closely resemble a globular cluster in the Milky Way after an elapsed time of 15 Gyr.  In a previous paper, we have presented evidence for interaction with a massive nearby HI cloud, NGC~1569-HI. This cloud is probably related to the recent starburst in NGC~1569, whereas the nearby dwarf companion UGCA~92 almost certainly is not (Stil \\& Israel 1998).  Observations of NGC~1569 in the 21-cm line of atomic hydrogen (HI) have been presented by Reakes (1980) and by Israel \\& Van Driel (1990). Their maps show the HI structure to consist mainly of a high column-density ridge with three local maxima and an arm-like feature extending approximately $3'$ to the west. The maximum rotational velocity of NGC~1569 is about $30\\ \\rm km\\ s^{-1}$. Israel \\& Van Driel (1990) also found a region with a local velocity dispersion comparable to the rotational velocity. ",
        "conclusions": "\\noindent 1. The 1.4 GHz continuum flux of NGC~1569 is $438 \\pm 5\\ \\rm mJy$. Peaks in the radio continuum distribution coincide with the two brightest HII regions.  The continuum emission largely follows the H$\\alpha$ distribution, including the so-called H$\\alpha$ arm.  \\noindent 2. The HI flux of NGC~1569 measured in the interferometer maps is $116\\ \\rm Jy\\ km\\ s^{-1}$.  This corresponds to an HI mass of $1.3 \\times 10^8\\ \\rm M_\\odot$ at 2.2 Mpc. The HI structure of the galaxy is that of extended diffuse HI emission, centered on a clumpy ridge of dense gas associated with the small optical galaxy. Much further structure is evident, in the form of a counterrotating companion, and intervening bridge and other features of possible tidal origin.  \\noindent 3. There is no detectable rotation of the HI gas within radii of $60''$ (0.6 kpc) from the center, confirming earlier results obtained for the ionized gas by other authors. Beyond 0.6 kpc from the center, the rotation velocity increases to $35 \\pm 6\\ \\rm km\\ s^{-1}$. An apparent turnover in the rotation curve is an artifact introduced by the presence of significant HI with peculiar velocities. Even excluding areas with high-velocity HI, NGC~1569 has a high average HI velocity dispersion of $\\sigma_v=21.3\\rm\\ km\\ s^{-1}$.  \\noindent 4.  The starburst has deposited significant amounts of energy into the disk, creating high velocity features and turbulence such that the line-of-sight velocity dispersion is much larger than the underlying rotational velocity due to the mass distribution of the galaxy.  The starburst has had a more limited effect on the outer disk where rotational velocities dominate otherwise still significant turbulent motions.  \\noindent 5. In addition, NGC~1569 contains a significant amount of HI gas at discrepant, high velocities. Components that can be separated from the regularly rotating gas already provide 10\\% of the total HI mass of the galaxy. At least some of the high-velocity HI appears to be associated with the H$\\alpha$ filaments, but it is unclear whether it respresents the neutral component of the outflow.  No evidence was found for a large bubble associated with the high-velocity HI. It is equally likely that at least part of the high-velocity HI clouds is associated with the HI companion and bridge linking this companion with NGC~1569. In that case, infall rather than outflow might be the cause of the discrepant velocities."
    },
    "0209/astro-ph0209502_arXiv.txt": {
        "abstract": "{We investigate the statistical properties of the polarized emission of extragalactic radio sources and estimate their contribution to the power spectrum of polarization fluctuations in the microwave region. The basic ingredients of our analysis are the NVSS polarization data, the multifrequency study of polarization properties of the B3-VLA sample (Mack et al. 2002) which has allowed us to quantify Faraday depolarization effects, and the 15 GHz survey by Taylor et al. (2001), which has provided strong constraints on the high-frequency spectral indices of sources. The polarization degree of both steep- and flat-spectrum at 1.4 GHz is found to be anti-correlated with the flux density. The median polarization degree at 1.4 GHz of both steep- and flat-spectrum sources brighter than $S(1.4\\,\\hbox{GHz})=80\\,$mJy is $\\simeq 2.2\\%$. The data by Mack et al. (2002) indicate a substantial mean Faraday depolarization at 1.4 GHz for steep spectrum sources, while the depolarization is undetermined for most flat/inverted-spectrum sources. Exploiting this complex of information we have estimated the power spectrum of polarization fluctuations due to extragalactic radio sources at microwave frequencies. We confirm that extragalactic sources are expected to be the main contaminant of Cosmic Microwave Background (CMB) polarization maps on small angular scales. At frequencies $< 30\\,$GHz the amplitude of their power spectrum is expected to be comparable to that of the $E$-mode of the CMB. At higher frequencies, however, the CMB dominates. ",
        "introduction": "Polarization measurements provide crucial information on the physics of radio sources. At high enough frequencies for Faraday rotation to be negligible we can reliably assume that the magnetic field direction lies perpendicular to the observed polarization position angle. On the other hand, determinations of the Faraday rotation measures (RMs) are informative on the magneto-ionic properties of the medium embedding the emitting region or along its line-of sight.  Another very important use of polarization measurements of large samples of radio sources is to quantify the contamination by these sources of polarization maps of the Cosmic Microwave Background (CMB). The astonishing advances in our understanding of the basic properties of the Universe and in precision determinations of its fundamental parameters made possible by the recent accurate measurements of acoustic peaks of the Cosmic Microwave Background (CMB) anisotropy power spectrum by the TOCO (Miller et al. 1999), BOOMERanG (de Bernardis et al. 2002), MAXIMA (Lee et al. 2001), DASI (Halverson et al. 2002), and CBI (Pearson et al. 2002) experiments, have put further impetus in experimental efforts to exploit the extraordinary wealth of cosmological information carried by the CMB.  A Gaussian CMB fluctuation field is fully characterized by four power spectra, $C_\\ell^{TT}$, $C_\\ell^{EE}$, $C_\\ell^{BB}$, $C_\\ell^{TE}$, where $T$ stands for ``temperature'', while $E$ and $B$ are rotationally invariant fields, which are linear, but non-local, combinations of the Stokes parameters Q and U (Seljak 1997; Kamionkowski et al. 1997; Zaldarriaga \\& Seljak 1997). It is then clear that polarization measurements are crucial to fully exploit the CMB information content. In particular, detection of CMB polarization is critical for tests of Planck-scale physics (Hu \\& White 1997; Kamionkowski \\& Kosowski 1999; Hu \\& Dodelson 2002). This has motivated experimental efforts by many groups (see Staggs et al. 1999; Cecchini et al. 2002; De Zotti 2002). Although these measurements are very challenging because of the weakness of the expected signal (at, or below, several $\\mu$K level), recent upper limits (Hedman et al. 2001; Keating et al. 2001) have already got close to it, and a detection may be achieved in the next few years.  It is not yet clear, however, whether the ultimate limit to our ability of measuring the CMB polarization will be set by detector sensitivity or by foregrounds, because the latter are still very poorly understood. On small angular scales, foreground intensity fluctuations at cm and mm wavelengths are dominated by extragalactic radio sources (Toffolatti et al. 1998, 1999), which are significantly polarized and are therefore expected to dominate also foreground polarization fluctuations up to at least $\\sim 100\\,$GHz. Preliminary investigations have been carried out by Sahzin \\& Kor\\\"olev (1985) and De Zotti et al. (1999). These works assumed a constant mean polarization degree for all classes of radio sources (estimated from rather small samples), a Poisson space distribution, and adopted mean values of the spectral indices to extrapolate to high frequencies.  The NRAO VLA Sky Survey (NVSS, Condon et al. 1998), covering $\\simeq 82\\%$ of the sky to a flux density limit of $\\simeq 2.5\\,$mJy at 1.4 GHz and containing data on Stokes $I$, $U$, and $Q$ parameters for almost $2\\times 10^6$ sources, has provided an extensive data base for a statistical investigation of the polarization properties of extragalactic sources. Important complementary information comes from the multifrequency study of over 100 sources drawn from the B3-VLA sample (Mack et al. 2002), which allows us to get insight into the effect of Faraday depolarization, which strongly affects polarization measurements at 1.4 GHz. Our analysis is presented in Sect.$\\,2$. The data by Condon et al. (1998) and Mack et al. (2002) also allow us to considerably improve on the available estimates of polarization fluctuations due to extragalactic sources in the microwave region. The new analysis, presented in Sect.$\\,3$, exploits the real space distribution of NVSS sources, which covers the flux density range relevant for CMB experiments, the true 1.4--5.85$\\,$GHz spectral index distribution of sources, obtained combining NVSS and GB6 (Gregory et al. 1996) data, the polarization degree distribution at 1.4 GHz, and the correction for Faraday depolarization, based on the data by Mack et al. (2002). The new constraints on the high frequency spectral indices set by the 15 GHz survey by Taylor et al. (2001) are also taken into account. Our main conclusions are summarized and discussed in Sect.$\\,4$. ",
        "conclusions": "Until adequate high frequency ``blind'' polarization surveys will become available, estimates of the contamination of CMB polarization maps by extragalactic sources will require delicate extrapolations of data from low-frequency surveys.  The main strength of the present analysis, compared with previous ones, is the use of a far more extended data-base. The NVSS survey has yielded polarization data for a very large, complete sample of extragalactic sources, allowing a direct observational determination of the power spectrum of polarization fluctuations due to them at $1.4\\,$GHz. The data also show a previously unnoticed anti-correlation of the polarization degree with flux density, which deserves further investigation.  Furthermore, coupling NVSS with GB6 data made possible to separately determine the contributions of steep and ``flat''-spectrum sources. The median polarization degree of the two populations for $S(1.4\\,\\hbox{GHz}) \\ge 80\\,$mJy turns out to be essentially equal (at the level of 2.2\\%) and shows a similar decrease with increasing flux density.  The data by Mack et al. (2002) have allowed us to investigate how the polarization properties of sources vary with frequency. For steep-spectrum sources we found clear evidence of Faraday depolarization corresponding to an effective value of the rotation measure $\\hbox{RM} \\simeq 260 \\hbox{rad}\\,\\hbox{m}^{-2}$. The average intrinsic polarization degree of these sources, observed at $\\nu \\gsim 10\\,$GHz, is estimated to be, on average, 3 times higher than at 1.4 GHz.  On the other hand, the mean polarization degree of flat-/inverted-spectrum sources increases only weakly from 1.4 to 10.6 GHz, suggesting that these sources have either small or really extreme RMs. Of course, polarization properties of radio sources may change with frequency not only by Faraday rotation but also for other reasons, such as the appearance of new emission components with different polarization properties. Only high-frequency polarization surveys may resolve this issue. We have considered two possibilities: either the mean polarization degree is frequency-independent, as indicated by the multifrequency studies by Jones et al. (1985) and Rudnick et al. (1985), or it increases, at high frequencies, by a factor $\\simeq 3$ compared to 1.4 GHz, as is the case for the sample of Nartallo et al. (1998).  \\begin{figure} \\centering \\includegraphics[width=3in,height=3in]{H3815F7.ps} \\caption{Rotation measures as a function of the 1.4 GHz luminosity for steep-spectrum sources in the sample by Mack et al. (2002) with spectroscopic (filled circles) or photometric redshift (crosses). An Einstein-de Sitter universe with $H_0=50$ has been adopted. } \\label{RMvsLumVigotti} \\end{figure}  Another critical issue is the extrapolation in frequency of the observed flux densities. Adoption of the observed 1.4--4.85 GHz spectral indices leads to over-predicting the 15 GHz counts by a factor $\\simeq 3$, compared with results of the survey by Taylor et al. (2001). Predictions of the most commonly used evolutionary models (Dunlop \\& Peacock 1990; Toffolatti et al. 1998), accounting for existing source counts up to 8.4 GHz and for the associated redshift/luminosity distributions, yield 15 GHz counts in excess by similar factors. Therefore, to extrapolate the 1.4 GHz flux densities we have exploited the observed spectral indices only up to 5 GHz. Above this frequency we have adopted the average spectral steepenings necessary to ensure, for each population (steep-, flat- and inverted-spectrum sources), consistency with the results by Taylor et al. (2001).   \\begin{figure*} \\centering \\includegraphics[width=7in,height=5in]{H3815F8.ps} \\caption{Power spectrum of polarization fluctuations due to extragalactic radio sources (irregular lines; $E$- and $B$-modes are indistinguishable) at {\\sc Planck}-LFI frequencies: 30 GHz (upper left-hand panel), 44 GHz (upper right-hand panel), 70 GHz (lower right-hand panel), and 100 GHz (lower right-hand panel). The lower irregular lines correspond to the case of a frequency-independent polarization degree for flat- and inverted spectrum sources, while the upper ones correspond to a factor of 3 increase of the mean polarization degree at Planck frequencies, compared to NVSS measurements. In both cases the mean polarization degree of steep-spectrum sources measured by the NVSS survey has been corrected upwards by a factor of 3. The dot-dashed straight line shows the preliminary estimate by De Zotti et al. (1999). Also shown, for comparison, are the CMB $E$- (dot-dashed curve) and $B$-mode (long-short dashes) power spectra for a flat CDM cosmological model with $\\Omega_\\Lambda =0.7$, $\\Omega_{\\rm dark matter}=0.25$, $\\Omega_{\\rm baryon}=0.05$, $H_0=70\\,\\hbox{km}\\,\\hbox{s}^{-1}\\,\\hbox{Mpc}^{-1}$, and a tensor contribution to the temperature quadrupole equal to 30\\% of that of scalar perturbations. The CMB power spectra were computed with CMBFAST (Seljak \\& Zaldarriaga 1996). } \\label{Cl} \\end{figure*}   An estimate of the power spectrum of polarization fluctuations at {\\sc Planck} frequencies obtained extrapolating the $1.4\\,$GHz fluxes as described in Sect.$\\,$2, with an upper flux-density cut-off corresponding to the source detection limit in each {\\sc Planck} channel as estimated by Toffolatti et al. (1998) and applying the average correction for Faraday depolarization derived in Sect.$\\,$3, is shown in Fig.~\\ref{Cl}. The present estimates are significantly below those by De Zotti et al. (1999), mostly due to the substantial steepening of the source spectra between 5 and $15\\,$GHz implied by the results of the survey by Taylor et al. (2001). As illustrated by Fig.~\\ref{Cl}, extragalactic radio sources are not expected to be a serious hindrance for measurements of the CMB $E$-mode power spectrum at $\\nu \\geq 30\\,$GHz. They are even less of a problem for measurements of the temperature-$E$-mode correlation since their $TE$ power spectrum vanishes, owing to the random distribution of their polarization position angles."
    },
    "0209/astro-ph0209028_arXiv.txt": {
        "abstract": "{ $\\omega$ Cen contains the largest population of very hot horizontal branch (HB) stars known in a globular cluster. Recent UV observations (Whitney et al.\\ \\cite{whro98}; D'Cruz et al.\\ \\cite{dcoc00}) show a significant population of hot stars below the zero-age horizontal branch (``blue hook'' stars), which cannot be explained by canonical stellar evolution.  Stars which suffer unusually large mass loss on the red giant branch and thus experience the helium core flash while descending the white dwarf cooling curve could populate this region.  Theory predicts that these ``late hot flashers'' should show higher temperatures than the hottest canonical HB stars and should have helium- and carbon-rich atmospheres.  We obtained and analysed medium resolution spectra of a sample of blue hook stars to derive their atmospheric parameters. The blue hook stars are indeed both hotter (\\teff $\\ge$35,000~K) and more helium-rich than classical extreme HB stars. In addition we find indications for a large enhancement of the carbon abundance relative to the cluster abundance. ",
        "introduction": "Horizontal-branch (HB) stars consist of a helium-burning core of about 0.5~M$_\\odot$ surrounded by a hydrogen-burning shell and a hydrogen-rich envelope of varying mass.  The temperature of an HB star (at a given metallicity) is determined by the mass of its hydrogen envelope, with the envelopes of the cooler HB stars being more massive. The increase in the bolometric correction with increasing temperature turns the blue HB into a vertical blue tail in optical colour-magnitude diagrams (CMDs, cf. Fig.~\\ref{ocen_kalu}) with the faintest blue tail stars being the hottest and least massive. The hottest HB stars (so-called extreme HB or EHB stars) with $T_{\\rm eff}$ $>$ 20,000 K have so little envelope mass that they are unable to sustain hydrogen-shell burning.  Nearly all of their surface luminosity comes from helium burning in the core.  Such EHB stars can be identified with the subdwarf B (sdB) stars in the field of the Milky Way and are believed to be mainly responsible for the UV excess observed in the spectra of elliptical galaxies. The globular cluster $\\omega$~Cen possesses an especially long blue tail containing the largest known population of EHB stars in a globular cluster.  Observations of $\\omega$~Cen in the far-UV (Whitney et al. \\cite{whro98}; D'Cruz et al. \\cite{dcoc00}) revealed a puzzling feature: the very hot end of the HB shows a surprisingly large spread in UV brightness, including a substantial population of subluminous stars lying up to \\magpt{0}{7} below the zero-age HB (ZAHB).  Such subluminous EHB stars are so far known to exist only in one other globular cluster (NGC~2808; Brown et al. \\cite{brsw01}; Sweigart et al. \\cite{swbr02}).  While stars brighter than the ZAHB can be produced by evolution away from the ZAHB, the stars fainter than the ZAHB cannot be explained by canonical HB evolution.  Within the framework of canonical HB theory there is no way to populate this region of the UV CMD without requiring an implausibly large decrease in the helium-core mass.  The subluminous EHB stars appear to form a hook-like feature in the UV CMD and are therefore called ``blue hook'' stars.  In optical CMDs these stars show up at the very faint end of the blue tail (cf. Fig.~\\ref{ocen_kalu}), in agreement with the high temperatures suggested by their UV photometry.  \\begin{figure}[h] \\vspace*{9.cm} \\special{psfile=MS2756f1.eps hscale=45 vscale=45 hoffset=-20 voffset=-65 angle=0} \\caption{ Colour-magnitude diagram of the blue tail of $\\omega$ Cen (Kaluzny et al. \\cite{kaku97}) with the spectroscopic targets marked. To establish the absolute magnitude scale on the right side apparent distance moduli $(m-M)_V$ of \\magpt{13}{9} and \\magpt{13}{3} were used for $\\omega$ Cen and NGC~6752, respectively. \\label{ocen_kalu}} \\end{figure}   The blue hook stars in $\\omega$~Cen populate a range in absolute visual magnitude that extends beyond the faint limit of the long blue tail in NGC~6752, which has been studied extensively by Moehler et al. (\\cite{mosw00}).  That in itself would not be a problem, but the spectroscopic analyses of Moehler et al. (\\cite{mosw00}) show that the blue tail stars in NGC~6752 already populate the EHB to the hot end predicted by canonical HB models.  Thus canonical theory fails to explain both the faint UV luminosities and expected high temperatures of the blue hook stars.  One might suspect that hotter EHB stars could be produced by simply reducing the envelope mass even further.  However, Brown et al. (\\cite{brsw01}) have demonstrated that there is a lower limit to the envelope mass of canonical EHB stars.  Increasing the mass loss along the red-giant branch (RGB) will not reduce the envelope mass below this limit but instead will cause a star to die as a helium white dwarf without ever igniting helium in its core.  Thus the blue hook stars may represent a new evolutionary channel for populating the very hot end of the HB.  One possibility is that the blue hook stars have undergone a delayed helium-core flash.  Castellani \\& Castellani (\\cite{caca93}) were the first to suggest that - for very high mass loss on the RGB - the helium flash can occur at high effective temperatures after a star has left the RGB (the so-called ``hot flashers'').  Indeed, D'Cruz et al. (\\cite{dcdo96}, \\cite{dcoc00}) proposed that the blue hook stars could be the progeny of such hot flashers, but unfortunately the D'Cruz et al. models were, at most, only $\\approx$\\magpt{0}{1} fainter than the canonical ZAHB, much less than required by the observations.  More recently, Brown et al. (\\cite{brsw01}) have explored the evolution of the hot flashers through the helium flash to the EHB in more detail, especially in regard to the timing of the flash.  Their models show that under some circumstances the helium flash will induce substantial mixing between the hydrogen envelope and helium core, leading to helium-rich EHB stars that are much hotter than canonical ones.  Brown et al. (\\cite{brsw01}) suggest that this ``flash mixing'' may be the key for understanding the evolutionary status of the blue hook stars.  Such mixing may also be responsible for producing the helium-rich, high gravity field sdO stars (Lemke et al. \\cite{lehe97}), whose origin is otherwise obscure.  The purpose of this paper is to present a spectroscopic analysis of a sample of blue hook stars in $\\omega$~Cen in order to test the predictions of the flash-mixing scenario.  Following a brief description of this scenario in Sect.~2, we discuss our observational data and then derive the parameters of the blue hook stars (temperatures, gravities and helium abundances) in Sects. 3 and 4, respectively.  In Sect.~5 we compare our results with the predictions of the flash-mixing scenario. ",
        "conclusions": "The high temperatures and high helium and carbon abundances reported here for the blue hook stars in $\\omega$ Cen provide general support for the flash-mixing hypothesis of Brown et al. (\\cite{brsw01}).  However, several questions remain.  The CMD of $\\omega$ Cen of Kaluzny et al.\\ (\\cite{kaku97}) given in Fig.~\\ref{ocen_kalu} does not show clear evidence for a gap within the EHB such as was found in NGC 2808 at M$_V \\approx$ \\magpt{+4}{6} by Walker (\\cite{walk99}) and Bedin et al. (\\cite{bepi00}).  Brown et al. (\\cite{brsw01}) have shown that the EHB gap in NGC 2808 can be identified with the transition between the canonical and flash-mixed stars.  There is a gap at M$_V \\approx$ \\magn{+4} in Fig.~\\ref{ocen_kalu}, but this gap separates the canonical EHB from blue HB stars and is not related to the hot flasher scenario (D'Cruz et al. \\cite{dcoc00}).  A fuller discussion of the gap between the EHB and blue HB in a number of other globular clusters is given by Piotto et al. (\\cite{pizo99}). One possible reason for the absence of a clear EHB gap in the $\\omega$ Cen CMD may be the limited precision of the photometry of Kaluzny et al. (\\cite{kaku97}), who warn about possible problems at faint magnitudes.  Alternatively, the absence of a clear gap may be related to the metallicity spread  in $\\omega$ Cen, although the models of D'Cruz et al. (\\cite{dcdo96}; their Fig.~2) suggest that the temperature at the hot end of the EHB shows little dependence on metallicity.  Given the known radial metallicity gradient in $\\omega$ Cen (e.g.\\ Hilker \\& Richtler \\cite{hiri00}), it would be of interest to determine if there is a gradient in the fraction of EHB stars which are blue hook stars.  Another question is why flash-mixed stars appear in $\\omega$ Cen and NGC~2808 but not in other EHB clusters such as M~13 and NGC~6752. Both $\\omega$ Cen (M$_V$ = \\magpt{-10}{29}; Harris \\cite{harr96}) and NGC~2808 (M$_V$ = \\magpt{-9}{36}) are among the most massive globular clusters in the Galaxy, so that the observed large EHB population in these clusters is not unexpected. However, the question of why a larger fraction of EHB stars in these clusters should be blue hook stars remains unexplained, and can be considered as another twist in the general problem of understanding the origin of the HB morphologies in globular clusters."
    },
    "0209/astro-ph0209624_arXiv.txt": {
        "abstract": "We have studied twelve very young (1--5\\,Myr) bona fide and candidate brown dwarfs in the Cha\\,I star forming region in terms of their kinematic properties, the occurrence of multiple systems among them as well as their rotational characteristics. Based on high-resolution spectra taken with UVES at the VLT (8.2\\,m), radial and rotational velocities have been measured. % A kinematic study of the sample showed that their radial velocity dispersion is relatively small suggesting that they are not ejected during their formation as proposed in recent formation scenarios. By means of time-resolved UVES spectra, a radial velocity survey for close companions to the targets was conducted. The radial velocities of the targets turned out to be rather constant setting upper limits for the mass M$_2 \\sin i$ of possible companions to 0.1\\,M$_{\\mathrm{Jup}}$ -- 2\\,M$_{\\mathrm{Jup}}$. These findings hint at a rather low ($\\leq$\\,10\\%) multiplicity fraction of the studied brown dwarfs. Furthermore, a photometric monitoring campaign of the targets yielded the determination of rotational periods for three brown dwarf candidates in the range of 2.2 to 3.4 days. These are the first rotational periods for very young brown dwarfs and among the first for brown dwarfs at all. ",
        "introduction": "Although a large number of brown dwarfs have been detected up to date, fundamental parameters have been measured only for a small subset. Consequently, there are still a lot of open questions, for example, the mechanism that leads to the formation of brown dwarfs is still poorly constrained.  We have studied a population of twelve very young bona fide and candidate brown dwarfs in the Cha\\,I star forming region, Cha\\,H$\\alpha$\\,1 to 12 (Comer\\'on et al. 1999, 2000, Neuh\\\"auser \\& Comer\\'on 1998, 1999) by means of high-resolution spectroscopy ($\\rm \\lambda / \\Delta \\lambda=40\\,000$) with UVES at the 8.2\\,m Kueyen telescope of the VLT as well as by a photometric monitoring campaign carried out at the Danish 1.5\\,m telescope at ESO. These observations yielded radial and rotational velocities as well as rotational periods. The studied brown dwarfs are only a few million years old and their observation allows insights into the formation and early evolution of brown dwarfs. ",
        "conclusions": "The kinematic study of brown dwarfs in Cha\\,I based on the measurement of precise radial velocities from high-resolution UVES spectra showed that their radial velocities have a dispersion of only 2.2\\,km\\,s$^{-1}$ and span a total range of only 2.6\\,km\\,s$^{-1}$. We therefore conclude that the ejection-model for the formation of brown dwarfs (Reipurth \\& Clarke 2001) is not a likely formation mechanism for the studied brown dwarfs since first calculations (Bate et al. 2002, Sterzik \\& Durisen 2002, this volume) predict that 10\\% of the brown dwarfs should have a larger velocity than 5\\,km\\,s$^{-1}$ due to the ejection.  The small binary fraction ($\\leq$ 10\\%) found in the presented RV survey is in agreement with the result of a direct imaging survey for wide, low-mass companions to the same objects (Neuh\\\"auser et al. (2002), Joergens et al. (2001), Neuh\\\"auser et al. this volume). Combining both surveys, a wide range of possible companion separations has been covered. The exact separations depend on the companion masses. For example, for brown dwarf companions ($<$ 13\\,M$_{\\mathrm{Jup}}$) to the targets, separations $<$3\\,AU and between 50 and 1000\\,AU were covered. With more restricted separations ($<$0.1\\,AU and 300--1000\\,AU) the surveys were sensitive also to companion masses down to 1\\,M$_{\\mathrm{Jup}}$.  These results seem to be in contrast with the high multiplicity fraction observed for very young stars and may suggest that brown dwarfs form not by direct collapse of unstable cloud cores as stars. However, a significant comparison of the multiplicity fraction of stars and brown dwarfs at very young ages is still hampered by small-number statistics in the substellar regime. Such a study has to compare multiplicity fractions in a certain separation range, which has to agree for both samples. Furthermore, it is possible that the gravitational collapse of cloud cores of brown dwarf masses does not yield as much multiple systems as the collapse of clouds of solar masses. In addition, stars can have companions, which have only a tenth of the mass of the primary, whereas a companion of a brown dwarf with such a low mass ratio would be already a planet. However, there is still a lot to do for observers as well as for theorists in order to understand in which way objects of brown dwarf masses are formed.  Finally, we have determined rotational periods for the three brown dwarf candidates Cha\\,H$\\alpha$\\,2, Cha\\,H$\\alpha$\\,3 and Cha\\,H$\\alpha$\\,6 of 2.2 to 3.4 days by means of a photometric monitoring campaign. These are the first rotational periods for very young brown dwarfs and among the first for brown dwarfs at all. They provide valuable data points in an as yet almost unexplored region of the substellar period-age diagram. A comparison of the determined rotational periods at the age of a few million years with rotational properties of older brown dwarfs ($>$36\\,Myr, Eisl\\\"offel \\& Scholz 2001) shows that most of the acceleration of brown dwarfs during the contraction phase takes place within the first 30\\,Myr or less. It is known that Cha\\,H$\\alpha$\\,2 and 6 have optically thick disks (Comer\\'on et al. 2000), therefore magnetic braking due to interactions with the disk may play a role for them. This is suggested by the fact, that among the three brown dwarf candidates with determined periods, the one without a detected disk, Cha\\,H$\\alpha$\\,3, has the shortest period. If the interaction with the disk is responsible for the braking, the results from Eisl\\\"offel \\& Scholz (2001) indicate that brown dwarf inner disks have been dissipated at an age of 36\\,Myr. These limits for the time scale of disk dissipation for brown dwarfs are very similar to those for T~Tauri stars, which are known to dissolve their inner disks within about the first 10\\,Myr (e.g. Calvet et al. 2000).  Further measurements of rotational periods for brown dwarfs are much-needed in order to complete the picture of angular momentum evolution in the substellar regime as well as that of rotationally induced phenomena, like dynamo activity and meteorological processes."
    },
    "0209/astro-ph0209412_arXiv.txt": {
        "abstract": "We follow the nuclear reactions that occur in the accretion disks of stellar mass black holes that are accreting at a very high rate, 0.01 to 1 $\\Msunsec$, as is realized in many current models for gamma-ray bursts (GRBs). The degree of neutronization in the disk is a sensitive function of the accretion rate, black hole mass, Kerr parameter, and disk viscosity. For high accretion rates and low viscosity, material arriving at the black hole will consist predominantly of neutrons. This degree of neutronization will have important implications for the dynamics of the GRB producing jet and perhaps for the synthesis of the $r$-process. For lower accretion rates and high viscosity, as might be appropriate for the outer disk in the collapsar model, neutron-proton equality persists allowing the possible synthesis of $^{56}$Ni in the disk wind. $^{56}$Ni must be present to make any optically bright Type I supernova, and in particular those associated with GRBs. ",
        "introduction": "Growing evidence connects GRBs to the the birth of hyper-accreting black holes \\citep{fry99b}, that is stellar mass black holes accreting matter from a disk at rates from $\\sim$0.01 to 10 $\\Msunsec$. Such models include the collapsar \\citep{woo93,mac99}, merging neutron stars and black holes \\citep{eic89,jan99}, supranovae \\citep{vie98}, merging helium cores and black holes \\citep{zha01}, and merging white dwarfs and black holes \\citep{fry99a}. For such high accretion rates the disk is optically thick, except to neutrinos, and very hot, consisting in its inner regions of a (viscous) mixture of neutrons and protons. At its inner boundary the disk connects to the black hole. Whatever processes accelerate the putative GRB-producing jet might therefore be expected to act upon some mixture of disk material and other background medium (e.g., the collapsing star in the collapsar model). Farther out, material will be lost from the disk in a vigorous wind \\citep{mac99,nar02}. In fact, Narayan et al. suggest that {\\sl most} of the disk will be lost to a wind except for those models and in those regions where neutrino losses dominate the energy budget.  It is thus of some consequence to know the composition of such disks. In the least case, the nucleosynthesis may be novel and could account for rare species in nature, such as the $r$-process. At most, the presence of free neutrons may affect the dynamics of the GRB jet \\citep{der99,ful00}, the GRB neutrino signature \\citep{bah00}, light curve \\citep{pru02}, and afterglow (e.g. via a ``pre-acceleration'' mechanism similar to the one discussed by Beloborodov 2002). It is also of some consequence to know whether the disk wind consists of radioactive $^{56}$Ni, as is necessary if a visible supernova is to accompany the GRB. If the electron mole number, $Y_e = \\Sigma (Z_i X_i/A_i)$, is less than 0.485, the iron group will be dominated by $^{54,56}$Fe and other more neutron-rich species \\citep{har85} which will be incapable of illuminating the supernova. Since the jet itself is inefficient at heating sufficient matter to temperatures required for nuclear statistical equilibrium (T $\\gtaprx 5 \\times 10^9$ K), supernovae seen in conjunction with GRBs \\citep[for example]{gal98,blo02} would be difficult to understand.  We have thus undertaken a survey of the nucleosynthesis that happens in the disks of rapidly accreting black holes. The work is greatly facilitated by the existence of numerical and semi-analytic solutions that yield the temperature-density structure and drift velocity \\citep{pop99}. These solutions have been verified in the case of the collapsar model by direct numerical simulation \\citep{mac99}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209138_arXiv.txt": {
        "abstract": "Extra-solar planets can be efficiently detected in gravitational microlensing events of high magnification. High accuracy photometry is required over a short, well-defined time interval only, of order 10-30 hours. Most planets orbiting the lens star are evidenced by perturbations of the microlensing light curve in this time. Consequently, telescope resources need be concentrated during this period only. Here we discuss some aspects of planet detection in these events. ",
        "introduction": "The presence of a planet in the lens system of a high magnification microlensing event results in a small deviation from the single lens light curve. This deviation occurs near the peak of the microlensing event and is detectable in events of high magnification (Liebes 1964; Griest \\& Safizadeh 1998). Heavier mass planets induce a larger perturbation, and planets closer to the Einstein ring of the main lens also produce larger perturbations. The Einstein ring radius, $\\re$, and crossing time, $\\te$, gives the typical spatial and temporal scales for microlensing events:  \\[ \\re \\simeq 1.9 \\sqrt{\\frac{M_{\\rm L}}{0.3M_{\\sun}}}  \\:\\:\\rm AU \\:\\:\\: \\te \\simeq 16.6\\sqrt{\\frac{M_{\\rm L}}{0.3M_{\\sun}}} \\:\\: \\rm days. \\label{eq:bigre} \\]  The observer-lens and lens-source distances are assumed to be 6kpc and 8 kpc respectively. The projected transverse velocity of the source is assumed to be 220 $\\rm kms^{-1}$. The time of full-width at half maximum for a high magnification event is: $t_{\\rm FWHM} = \\frac{3.5\\te}{A_{\\rm max}} $, where \\Am is the maximum amplification of the event. A high magnification event occurs when the source and lens stars are well aligned, i.e. when $A_{\\rm max} \\simeq \\frac{\\re}{\\umin} \\gg 1$. ",
        "conclusions": "The observation of high magnification microlensing events is an effective method for the detection of extra-solar planets. The necessity of only observing during the critical time around the event peak enables efficient use of telescope scheduling and resources. Terrestrial mass planets are detectable using this technique, in regions bounding the Einstein ring of the lens system. For the most likely lens star mass, this distance is about $\\simeq 2$AU. Probing the distribution of light mass planets at these distances around the host star will complement the current understanding of planet formation and distribution. This technique requires continuous, high accuracy photometry of the event around the peak. A network of 1 - 2 m class telescopes situated around the world will be able to perform the required observations. Future events are expected to yield initial statistics on the abundance of terrestrial mass planets.  NJR thanks the Graduate Research Fund of the University of Auckland, the LOC and the MOA project for financial assistance.  \\vspace{-0.2cm}"
    },
    "0209/astro-ph0209281_arXiv.txt": {
        "abstract": "The influence of a dust grain mixture consisting of spherical dust grains with different radii and/or chemical composition on the resulting temperature structure and spectral energy distribution of a circumstellar shell is investigated. The comparison with the results based on an approximation of dust grain parameters representing the mean optical properties of the corresponding dust grain mixture reveal that (1) the temperature dispersion of a real dust grain mixture decreases substantially with increasing optical depth, converging towards the temperature distribution resulting from the approximation of mean dust grain parameters, and (2) the resulting spectral energy distributions do not differ by more than 10\\% if $\\ge 2^5$ grain sizes are considered which justifies the mean parameter approximation and the many results obtained under its assumption so far. Nevertheless, the dust grain temperature dispersion at the inner boundary of a dust shell may amount to $\\gg$100\\,K and has therefore to be considered in the correct simulation of, e.g., chemical networks.  In order to study the additional influence of geometrical effects, a two-dimensional configuration -- the HH\\,30 circumstellar disk -- was considered, using model parameters from Cotera et al.~(2001) and Wood et al.~(2002). A drastic inversion of the large to small grain temperature distribution was found within the inner $\\sim 1{\\rm AU}$ of the disk. ",
        "introduction": "The simulation of spectral energy distributions (SEDs), images, and polarization maps of young stellar objects has become a profound basis for the analysis and interpretation of observing results. Many techniques and approximations for the solution of the radiative transfer (RT) problem in different model geometries (1D--3D), considering more and more special physical processes, such as the stochastic and photo-electric heating of small grains (see, e.g, Draine \\& Li~2001; Bakes \\& Tielens~1994, Siebenmorgen, Kr\\\"ugel, \\& Mathis~1992), scattering by spheroidal grains (see, e.g., Wolf, Voshchinnikov, \\& Henning~2002; Gledhill \\& McCall~2000), or the  coupling of line and continuum RT (see, e.g., Rybicki \\& Hummer~1992), have been developed. The simulation of the temperature structure in simple-structured circumstellar shells or disks (see, e.g., Malbet, Lachaume, \\& Monin~2001; Chiang et al.~2001), the estimation of the properties (luminosity, temperature, mass) of heavily embedded stars (see, e.g., Kraemer et al.~2001), or the determination of the inclination of a circumstellar disk (see, e.g., Chiang \\& Goldreich~1999; Men'shchikov, Henning, \\& Fischer~1999; Wood et al.~1998) represented modest, first attempts of the application of the existing sophisticated numerical techniques. More recent efforts are directed to derive the dust grain size distribution in a circumstellar disk from its SED (Wood et al.~2002; D'Alessio, Calvet \\& Hartman~2001). Thus, it is clear that beside strong observational constrains, the model parameters and considered physical processes have to be questioned in depth in order to derive such detailed conclusions. However, looking behind the scenes, many of the RT models are based on simplifying assumptions of very basic processes such as isotropic instead of anisotropic scattering, mean dust parameters representing dust grain ensembles (different radii and chemical compositions), or the flux-limited diffusion approximation - approximations which are well-suited for handling the energy transfer in hydrodynamic simulations or a rough data analysis but which may not necessarily guarantee the desired accuracy for a detailed SED and image/polarization data analysis.  In the advent of (space) observatories such as SIRTF\\footnote{Space Infrared Telescope Facility}, which will be able to obtain SEDs of evolved debris disks around young stars, there exists a strong need for adequate numerical RT techniques in order to allow to trace dust grain growth (Meyer et al.~2001) and the influence of other physical effects and processes such as the Poynting-Robertson effect (see, e.g., Srikanth~1999) and dust settling (Dubrulle, Morfill, \\& Sterzik~1995, Miyake \\& Nakagawa~1995) on the dust grain size distribution in these disks. Therefore, we clearly have to understand which influence the different approximations (as long as they are required) in the RT simulations may have on the resulting observables. Based on two different grain size distributions consisting of astronomical silicate, the differences in the resulting dust grain temperature distributions and the resulting SEDs between simulations of the RT in (A) a ``real'' dust grain mixture and (B) under the assumption (approximation) of mean dust grain parameters will be discussed. This investigation is therefore mainly focused on two questions: (1) Of what order of magnitude are the differences (?) and (2) How many grain sizes have to be considered to represent the properties of a real grain size distribution? The spatial temperature distribution of the considered dust configurations is calculated on the basis of local thermal equilibrium. Stochastic heating processes which are expected in case of very small grains consisting of tens to hundreds of atoms (see, e.g., Draine \\& Li~2000 and references therein) are not subject of this investigation.  In Sect.~\\ref{rtmodel}, the RT and the dust grain model are briefly introduced. In Sect.~\\ref{mean}, the definition of the mean dust grain parameters is given and the expected deviations of the RT results -- once based on the mean dust parameters and once on a real grain size distribution -- are outlined. In Sect.~\\ref{rtem}, the RT in a spherical shell with variable optical depth and density distribution (see Sect.~\\ref{spsh}) is considered, while the temperature structure in a model of the HH\\,30 circumstellar disk is investigated in Sect.~\\ref{hh30}. The SED resulting from a models with a dust grain mixture is compared to the mean particle approximation in Sect.~\\ref{sed}. ",
        "conclusions": "In simulations of the radiative transfer in the circumstellar environment of young stellar objects it has been widely established to use mean values for those parameters which describe the interaction of the electromagnetic field with the dust grains (see, e.g., modelling efforts by Calvet et al.~2002: circumstellar disk around the young low-mass star TW Hya; Cotera et al.~2001 and Wood et al.~2002: circumstellar disk around the classical T~Tauri star HH~30; Fischer, Henning, \\& Yorke~1994, 1996: polarization maps of young stellar objects). A main reason for this is given by the fact that the consideration of a large number of single grain sizes and chemically different grain species results in a nearly linearly increasing amount of required computer memory in order to store the separate temperature distributions. Furthermore, the calculation of the spatial temperature distribution for all grains of different sizes and chemical composition requires significantly more computing power since (a) the heating by the primary sources (e.g., the star embedded in a circumstellar envelope) has to be performed independently for each grain species, and (b) the number of computing steps required to model the subsequent mutual heating of the different dust grain species due to dust re-emission scales even as $\\approx (n_{\\rm D} \\times n_{\\rm S})^2$, where $n_{\\rm D}$ is the number of chemically different components and $n_{\\rm S}$ is the number of separate dust grain radii considered. While these simulations are feasible in case of one-dimensional models (see, e.g., Chini, Kr\\\"ugel, \\& Kreysa~1986 and Efstathiou \\& Rowan-Robinson~1994: Dust emission from star forming region; Kr\\\"ugel \\& Siebenmorgen~1994 and Siebenmorgen, Kr\\\"ugel, \\& Zota~1999: Radiative transfer in galactic nulei) or simple-structured two-dimensional models (Men'shchikov \\& Henning~1997: Circumstellar disks; Efstathiou \\& Rowan-Robinson~1994: Disks in active galactic nuclei), it is hardly possible to handle two- and three-dimensional models with high density gradients and/or high optical depth which require high-resolution temperature grids.  In this study, the difference between the results of RT simulations based (a) on a mean dust grain parameter approximation and (b) real dust grain size distributions have been investigated. Based on a one-dimensional density distribution it was found that the temperature structure of a real grain size distribution shows a very complex behaviour in the inner, hot region of the shell depending on (1) the grain size distribution, (2) the effective temperature of the embedded star and the optical depth and therefore on the density distribution. However, the relative difference between the SED based on a real dust grain size distribution on the one hand and the approximation of mean dust grain parameters on the other hand was found to be smaller than about $\\approx$10\\% if a minimum number of $2^5$ to $2^6$ grain sizes have been considered.  As the temperature structure in a circumstellar disk -- based on the model for HH\\,30 -- shows, the geometry of the density distribution is a significant parameter for the resulting temperature differences between grains of different size, too. In the inner region of the disk with a diameter of a few AU a temperature inversion layer was found where the sign of the temperature difference of the largest and smallest grains is reverted. As this and the results obtained on the basis of the spherical shell (\\S\\ref{spsh}) show, the dust grain temperature structure is not sufficiently represented by dust grains with mean optical parameters. On the one hand, this is of tremendous importance for the simulation of chemical networks since the largest deviations from results based on the approximation of mean dust grain parameters have been found in the inner hot, dense region of the shell/disk where the chemical evolution takes place on its smallest timescale. On the other hand, the complex temperature structure may significantly change the hydrostatic properties of the considered gas/dust density distribution itself. However, these questions have to be investigated in future studies in order to find out the influence on observable quantities such as SEDs, images, polarization maps, and visibilities. Furthermore, the influence on processes taking place in more evolved circumstellar disks, such as the dust settling and dust grain growth, have to be considered taking into account the temperature structure of real grain size distributions."
    },
    "0209/astro-ph0209248_arXiv.txt": {
        "abstract": "We assess the impact of the trace element $\\neon$ on the cooling and seismology of a liquid C/O white dwarf (WD).  Due to this elements' neutron excess, it sinks towards the interior as the liquid WD cools. The gravitational energy released slows the WD's cooling by 0.5-1.6 Gyr. In addition, the $\\neon$ abundance gradient changes the periods of the high radial order $g$-modes at the 1\\% level. ",
        "introduction": "After $\\carb$ and $\\oxy$, the most abundant nucleus in a $M<\\msun$ white dwarf (WD) interior is $\\neon$.  As the slowest step in the CNO cycle is the $p+\\nitr$ reaction, almost all CNO nuclei end up as $\\nitr$ at the end of H burning.  During the He burning stage, the reaction $\\nitr(\\alpha\\,,\\gamma)\\flor(\\beta^+)\\oxyet(\\alpha,\\gamma)\\neon$ processes all $\\nitr$ into $\\neon$.  For recently formed WDs of $M<\\msun$ this results in a $\\neon$ mass fraction $\\xne \\approx Z_{\\mathrm{CNO}}\\approx 0.02$ (see, for example, \\cite{um99}).  $\\neon$ has a mass to charge ratio, $A/Z$ greater than that of $\\carb$ and $\\oxy$.    As pointed out by \\cite{brav92} and \\cite{bh01}, this affects the dynamics of $\\neon$ nuclei in the following manner.  In the degenerate WD interior, there exists an upward pointing electric field of magnitude $eE\\approx 2m_p g$ \\cite{bh01}, $g$ being the local gravitational acceleration.  The resulting net force on $\\neon$, $\\vec{F}=- 2m_pg \\rhat$, biases $\\neon$'s diffusion inward.  In contrast, the predominant $\\carb$ and $\\oxy$ ions experience no net force.  The resulting sedimentation of $\\neon$ will impact the cooling history and seismology of the WD.  The $\\neon$ evolution is governed by mass continuity \\begin{equation} \\label{eq:cont} \\pder{\\rho_{22}}{t} + \\nabla\\cdot {\\bf J}_{22} =0 \\;, \\end{equation} where $\\rho_{22}$ and ${\\bf J}_{22}$ are the local $\\neon$ density and flux respectively.  The flux is given by \\begin{equation} \\label{:flux} {\\bf J}_{22} = (- D \\pder{\\rho_{22}}{r} - v \\rho_{22}) \\rhat\\;, \\end{equation} where  $D$ is $^{22}\\mathrm{Ne}$'s diffusion coefficient in the background plasma and $v$ is the magnitude of its local drift velocity. $D$ and $v$ are related via the Stokes-Einstein equation: $v = 2 m_p g D/k T$.  As discussed in \\cite{bh01} and \\cite{del02}, $D$ is not well known for conditions in the WD interior.  So, for this work, we took as a starting point the self diffusion coefficient of the one-component plasma, $D_s$, as calculated in \\cite{hans75},$D_s\\approx 3 \\omega_p a^2 \\Gamma^{-4/3}$, where $\\omega_p^2=4\\pi n_i (Ze)^2/Am_p$, $\\Gamma = (Z e)^2/a k T$ and $a^3=3A m_p/4\\pi \\rho$ is the inter-ion spacing of the background ions. At $\\Gamma = 173$, the plasma crystallizes \\cite{far93}.  We quantify the effects of the uncertainty of $D$ on our results by performing calculations with varying $D$. ",
        "conclusions": ""
    },
    "0209/astro-ph0209554_arXiv.txt": {
        "abstract": " ",
        "introduction": " ",
        "conclusions": "\\begin{enumerate} \\item A theory of winds accelerated by the radiation pressure in lines with account of gravitational redshifting of photons is developed. A system of equations describing stationary, spherically-symmetric, isothermal flow is derived.  \\item A solution of these equations is obtained numerically for two cases: a standard line-driven wind (CAK theory), Gravitationally Exposed Flow (GEF)- a wind that is accelerated by the radiation pressure in lines if to take into account gravitational redshifting. It is shown that an increase of up to $35\\%$ in $v^\\infty$ can be obtained.  \\item To take approximately into account effects of general relativity, Paczynski - Wiita potential is adopted. A wind solution is calculated for this type of potential and comparative analysis with Newtonian case is presented.  \\item The developed theory can be used to explain fast outflows from AGN. \\end{enumerate}"
    },
    "0209/astro-ph0209004_arXiv.txt": {
        "abstract": "We have detected four Giant Molecular cloud Associations (GMAs) (sizes $\\leq 6.6'' \\approx 430\\,pc$) in the western and eastern region of the polar ring in NGC~2685 (the Helix galaxy) using the Owens Valley Radio Observatory (OVRO) millimeter interferometer. Emission from molecular gas is found close to the brightest H$\\alpha$ and HI peaks in the polar ring and is confirmed by new IRAM 30m single dish observations. The CO and HI line velocities are very similar, providing additional kinematic confirmation that the CO emission emerges from the polar ring. For the first time, the total molecular mass within a polar ring is determined ($M_{H_2}\\sim(8-11)\\times10^6\\,\\solm$, using the standard Galactic conversion factor). We detect about $M_{H_2}\\sim4.4\\times10^6\\,\\solm$ in the nuclear region with the single dish. Our upper limit derived from the interferometric data is lower ($M_{H_2}\\le0.7\\times10^6\\,\\solm$) suggesting that the molecular gas is distributed in an extended ($\\ge 1.3\\,kpc$) diffuse disk. These new values are an order of magnitude lower than in previous reports. The total amount of molecular gas and the atomic gas content of the polar ring are consistent with formation due to accretion of a small gas-rich object, such as a dwarf irregular. The properties of the NGC~2685 system suggest that the polar ring and the host galaxy have been in a stable configuration for a considerable time (few Gyr). The second (outer) HI ring within the disk of NGC~2685 is very likely at the outer Lindblad resonance (OLR) of the $\\sim 11\\,kpc$ long stellar bar. ",
        "introduction": "Polar ring galaxies (PRGs) represent an unusual, rare class of objects which show clear signs of galaxy interaction (Schweizer, Whitmore \\& Rubin 1983). Typically, an early-type (S0 or E) host galaxy is surrounded by a luminous ring (containing stars, gas and dust) of $\\sim$ 5 to 25~kpc diameter oriented almost perpendicular to the main stellar disk and rotating about the center of the main stellar body (see PRG atlas by Whitmore et al. 1990). To-date only about a dozen PRGs have been kinematically confirmed (e.g. Table 1 in Sparke \\& Cox 2000). In the generally accepted picture, the formation of polar rings is the result of a ''secondary event'': e.g., capture of a satellite galaxy or accretion of material between (tidally) interacting galaxies involving a pre-existing S0 galaxy (e.g. Toomre \\& Toomre 1972, Reshetnikov \\& Sotnikova 1997). Recently Bekki (1997, 1998) suggested a pole-on merger between two disk galaxies as an alternative formation mechanism. Observations suggest that polar rings are long-lived structures (Whitmore et al. 1987, Eskridge \\& Pogge 1997). Possible stabilizing mechanisms are self-gravitation in the ring (Sparke 1986), or a massive triaxial halo (Whitmore et al. 1987, Reshetnikov \\& Combes 1994)  The Helix galaxy, NGC~2685 ($D\\,\\sim\\,13.5\\,Mpc; 1''\\,\\sim\\,65\\,pc$; (R)SB0+ pec) is one of the kinematically confirmed PRGs. Two rings are detected in HI line emission which have orthogonal angular momentum vectors (Shane 1980).  Optical and NIR surface photometry (Peletier \\& Christodoulou 1993) suggests an age of 2-6 Gyr for the inner 'polar' ring and therefore a long-lived structure. The younger HII regions in the polar ring of NGC~2685 have solar abundances, making accretion of metal-poor material unlikely (Eskridge \\& Pogge 1997). ",
        "conclusions": "\\subsection{Probable formation mechanism of the polar ring}  The (detected) molecular gas mass of the polar ring is only about 4\\% of the atomic mass present there ($log(M_{H_2}/M_{HI})\\approx -1.40$), and the ratio between the blue luminosity and the molecular gas mass is also fairly low ($log(M_{H_2}/L_{B}\\approx-2.7$) using $L_{B}\\sim4.5\\times10^9\\,\\solar$ of Richter, Sackett \\& Sparke 1994). These values are comparable to those found for low-mass ($\\leq 10^{10}\\,\\solm$) very late-type galaxies (i.e. including dwarf irregulars, Casoli et al. 1998) and S0 galaxies with counter-rotating gas/stars (Bettoni et al. 2001).  The low far-infrared luminosity of $L_{FIR}\\sim 4\\times10^8\\solar$ (using the IRAS fluxes and the standard relation for $L_{FIR}$; Sanders \\& Mirabel 1996) indicates no recent {\\it massive} star formation as observed in starburst galaxies in the NGC~2685 system. This is in agreement with the old age ($\\sim 10\\,Gyr$) of the nuclear stellar population (Sil'chenko 1998). All this implies that there was no considerable gas inflow into the central kiloparsec in the recent past. This appears to contradict the models of the pole-on disk merger (Bekki 1998) and polar encounter with a spiral galaxy (Reshetnikov \\& Sotnikova 1997), which show an accumulation of a few $10^8\\,\\solm$ in the central kiloparsec. (Our derived molecular gas mass in the inner kiloparsec is only about $4\\times10^6\\,\\solm$.)  The host disk of NGC~2685 exhibits no obvious sign of an interaction besides the more extended diffuse HI emission at the north-eastern edge of the disk (Fig. \\ref{fig:hi}). We can use equation (29) of Lake \\& Norman (1983) which relates the density of the accreted dwarf to the mean density of the parent galaxy inside the radius of the orbit. For a dynamical mass of $M_{dyn} (r=34'') = 7.4 \\times 10^9 M_{\\odot}$ inside the ring radius for the host galaxy, and a total mass of $M_{dwarf}\\sim 5\\times M_{HI,ring}\\approx1.5\\times10^9\\,\\solm$ (assuming a typical scale factor of 5 between the HI mass and the dynamical mass for dwarf galaxies; e.g. Swaters 1999), we find a radius for the accreted dwarf galaxy of about 2.2\\,kpc. This radius and the HI mass of $M_{HI}\\sim3\\times10^8\\,\\solm$ are typical values for late-type dwarf galaxies (Mateo 1998, Swaters 1999).  Since the ring appears in its colors neither very young ($< 1\\,Gyr$) nor extremely ancient (Peletier \\& Christodoulou 1993), it seems likely that the polar ring was formed a few Gyr ago during an accretion event (of a dwarf/low-mass galaxy) similar to those proposed for S0 counter-rotators. The solar metallicity derived for the ring HII regions (Eskridge \\& Pogge 1997) implies that either (1) the accreted dwarf galaxy had an elevated intrinsic metallicity similar to those observed in a couple local dwarf galaxies (e.g. Mateo 1998), (2) low-level continuous star formation has enriched the polar ring ISM in the past few Gyr (see e.g. Legrand et al. 2001) and/or (3) the ISM of the accreted dwarf galaxy was mixed/enriched with more metal-rich material from the primary galaxy.  The outer HI ring can be explained as gas accumulated at the OLR of the 11\\,kpc diameter bar. The good spatial correspondence of the optical and HI rings suggests that there is enhanced star formation. We conclude that the suggested scenario by Peletier \\& Christodoulou (1993) seems unlikely, where the outer HI ring has been formed during the secondary accretion event which formed the inner polar ring as well.  \\subsection{Stability of the polar ring}  The OVRO map of the molecular gas provides the first unambiguous (kinematically confirmed) detection of molecular gas in a polar ring. This molecular gas is associated with about four GMAs which are located close to on-going star formation and peaks in the atomic gas which exceed $10^{21}\\,\\cms$. Note that the molecular and the atomic gas appear concentrated where the polar ring intersects the disk of the parent galaxy.  The polar ring exhibits at least two stellar populations. The HII regions present in the polar ring have an average age of 5\\,Myr (if instantaneous star formation is assumed) and about solar metallicity (Eskridge \\& Pogge 1997). However, Peletier \\& Christodoulou (1993) deduced an age of a few Gyr from the red colors of the polar ring. The contribution of the atomic gas in the polar ring to the total dynamical mass (enclosed out to its radius) is about 4\\%. This might already be enough to allow for self-gravitation of the polar ring, and explain its persistence (Sparke 1986). Alternatively, Mahon (1992) has found that the HI distribution and kinematics of NGC~2685 can be stable using models of prograde anomalous orbits in a triaxial gravitational potential.  Thus we conclude that: (a) the polar ring has been stable for a substantial time (few Gyr) and (b) the recent star formation in the polar ring has been triggered by another mechanism than the actual polar ring formation process."
    },
    "0209/astro-ph0209374_arXiv.txt": {
        "abstract": "s{ We present results concerning the occurrence of Seyfert galaxies in a new large sample of Compact Groups (Focardi \\& Kelm 2002). Seyfert galaxies turn out to be relatively rare ($<$ 3\\%), with a significant dominance of Sy2. Seyferts are preferentially associated to Compact Groups displaying relatively high velocity dispersion and a large number of neighbours. These characteristics, together with an excess of ellipticals among companions, suggest that Seyferts are to be found preferentially in rich-groups/poor-cluster like CGs. } ",
        "introduction": "Because of their high density (comparable to the galaxy density in clusters) and relatively low velocity dispersion ($\\approx$ 200-300 $km/s$) Compact Groups {\\bf (CGs)} are predicted to constitute the most probable sites for strong galaxy-galaxy interactions and mergers to occur. So far, this general expectation has been tested mainly on the Hickson Compact Group sample ({\\bf HCGs}, Hickson 1982, 1997). Indeed, several HCGs show evidence of ongoing interaction, but component usually remain distinct, with recognizable morphological type (Sulentic 1997). Zepf (1993) has estimated the fraction of currently merging galaxies in HCGs to be $\\approx$ 7\\%, and the fraction of blue ellipticals (which are plausible merger remnants) to be similarly low (4 in 55), and predominantly associated to faint members (Zepf {\\it et al.} 1991). Concerning the far infrared (FIR), Hickson {\\it et al.} (1989) found the FIR emission to be enhanced in HCGs, however Sulentic \\& de Mello (1993) and Verdes-Montenegro {\\it et al.} (1998) suggest there is no firm evidence for enhancement. A similar lack of FIR enhancement is found in the UZC-CG sample by Kelm {\\it et al.} (2002). These authors also state that data are compatible with IRAS galaxies in CGs being plausible candidates for accordant redshift projections, rather than interaction triggered starbursting galaxies.  Addressing the issue of AGNs in HCGs, Kelm {\\it et al.} (1998) find that only $\\approx$ 2\\% of the HCG member galaxies display a Seyfert spectrum. They find Sy to be hosted in luminous spirals, as is usually the case. However, computing also low-level AGN activity appears to dramatically increase the fraction of AGNs in CGs (Coziol {\\it et al.} 1998, 2000), with a significant preference for early type hosts. Coziol {\\it et al.} (2000) state that AGNs (including low luminosity/dwarf sources) are the most frequent (41\\%) activity type encountered in CGs. A similar high fraction of AGNs in HCGs is retrieved by Shimada {\\it et al.} (2000), who additionally compare AGN in HCGs with field sources and claim that the dense galaxy environment in HCGs does not affect the triggering of either AGNs or nuclear starbursts. ",
        "conclusions": "We retrieve only a marginal fraction of Sy galaxies in the UZC-CG galaxy sample, suggesting that either the interaction-activity connection does not hold for CGs, or, that most CGs are actually systems undergoing only mild interactions. Significant dynamical and environmental differences between CGs hosting/non-hosting a Sy member might indicate that Sy are typically associated to CGs already resembling rich-groups/poor-clusters. This interpretation is also supported by the fact that an excess of ellipticals is retrieved among companions to Sy. Our analysis clearly suggests that the low fraction of Sy among UZC-CGs galaxies results from a low fraction of physical systems in the sample. Accordingly, most CGs turn out to be accordant redshift projected groups, or groups at first approach in which no strong interaction has yet occurred."
    },
    "0209/astro-ph0209468_arXiv.txt": {
        "abstract": " ",
        "introduction": "There are at least four well known Very Long Baseline Interferometer (VLBI) networks that provide open access to astronomers around the world. These include: \\begin{itemize} \\item European VLBI Network, {\\tt www.evlbi.org} \\item Very Long Baseline Array, {\\tt www.aoc.nrao.edu/vlba/html/VLBA.html} \\item Coordinated Millimeter VLBI Array, {\\tt web.haystack.mit.edu/cmva} \\item Australian Long Baseline Array, {\\tt www.atnf.csiro.au/vlbi} \\& the Asia-Pacific Telescope, {\\tt www.atnf.csiro.au/apt}. \\end{itemize}  For a comprehensive guide to the various array characteristics, corresponding correlator capabilities, range of diverse observing modes, different proposal submission procedures, user support {\\it etc.},  I refer the reader to the on-line web pages highlighted above.  In this paper I have chosen to focus on a sub-set of these facilities, in particular the European VLBI Network (EVN) both in stand-alone mode but also as a major component of a Global VLBI Network, involving MERLIN (see {\\tt www.merlin.ac.uk}) and the Very Long Baseline Array (VLBA -- see Zensus, Diamond \\& Napier 1995, for a comprehensive review). ",
        "conclusions": ""
    },
    "0209/astro-ph0209142_arXiv.txt": {
        "abstract": "{ In Active Galactic Nuclei (AGNs) the presence of a star cluster around the central black hole can have several effects on the dynamics and the emission of the global system. In this paper we analyze the interaction of stellar atmospheres with a wind outflowing from the central region of the AGN nucleus. Even a small mass loss from stars, as well as possible star collisions, can give a non-negligible contribution in feeding matter into the AGN nuclear wind. Moreover, stellar mass loss can produce envelopes surrounding stars that turn out to be suitable for reproducing the observed emission from the Broad Line Region (BLR). In this framework, the envelope can be confined by the bow shock arising from the interaction between  the expanding  stellar atmosphere and the AGN nuclear wind. ",
        "introduction": "It is generally accepted that the broad emission lines present in AGN spectra originate in numerous, small and relatively cold gas concentrations in the neighborhood of the central black hole, the so-called Broad Line Region (BLR). The  observed line properties directly indicate the existence of such gas concentrations; however, the physical nature of these structures is still a matter of debate. Up to the present day, three possible classes of models have been envisaged: gas clouds, accretion disks, and stellar wind envelopes. Each scenario suffers from still unclarified problems and we refer to the recent review by Korista (\\cite{korista}) for a presentation of the problem.  In general,  broad emission line regions in AGNs turn out to be characterized by a relatively narrow range in ionization parameter $U$, {\\it i.e.}, $U\\equiv L/[4\\pi c r^2n_*<h\\nu>] \\simeq 0.01-1$ (Alexander \\& Netzer \\cite{alexander94}, Peterson \\cite{peterson97}, Netzer \\cite{netzer90}). In this expression $r$ is the distance from the central black hole, $n_*$ is a representative value of the number density of the line emitting gas, $L$ is the ionizing radiation field luminosity from a central source, illuminating the line emitting region, and $<h\\nu>$ is the average photon energy of the ionizing radiation field. A first estimate of the conditions of the line emitting gas, such as $n_*$ and $U$, can be obtained from an analysis of the emission spectrum; coupling this information with the evaluation of the luminosity and of the ionizing spectrum, the relation above, defining $U$, can be inverted to give an estimate of the characteristic size of the BLR, or, better, the characteristic distance of the line emitting gas from the central continuum source, $r_{\\ma {BLR}}$ (see Wandel et al. \\cite{wandel99}). In a first, simplistic use of photoionization arguments, under the assumption that the ionizing continuum shape is quite similar and both $n_*$ and $U$ are substantially the same for all the AGNs (see Kaspi et al. \\cite{kaspi00}), a relationship, between the characteristic BLR size and the AGN luminosity, of the type $r_{\\ma {BLR}}\\propto L^{0.5}$ would then be expected. Indeed, this would just be  an order of magnitude evaluation, since there are reliable indications that the BLR material density and the ionization parameter are not constant at all even across the BLR extension of a single AGN (see, e.g., Kaspi et al. \\cite{kaspi00}). Nonetheless, this was actually what reverberation mapping studies of several AGNs seemed to suggest until very recently, implying $r_{\\rm {BLR}} \\simeq 0.1~ L_{46}^{1/2}$ pc (Netzer \\& Peterson \\cite{netzer}, Kaspi \\cite{kaspi97}) (where $L_{46}$ is the AGN luminosity in units of $10^{46}$ erg/sec). In a reverberation technique study  of a sample of 17 quasars, combining their data with those available for Seyfert 1 galaxies, Kaspi et al. (\\cite{kaspi00}) propose a significantly different relationship between $r_{\\ma {BLR}}$ and the luminosity of the AGN, namely $r_{\\ma {BLR}}\\propto L^{0.7}$.  Although the exponent does not differ much from the previous one, it is remarkable that the ensemble of the data fit by the latter  relation, are not fit by the one with an 0.5 exponent.  In a previous paper (Pietrini \\& Torricelli \\cite {pietrini}, hereafter Paper I), motivated by recent observations supporting  radial outflows even in radio-quiet AGN's, such as Seyfert 1s, (Crenshaw et al. \\cite{crenshaw99b}, Weymann et al. \\cite{weymann97}), we have analysed the physical structure and characteristics of a global AGN outflow, presumably originating in the very central regions and expanding out to large distances as a kind of background/connection for various  observational components of the AGN structure [ BLR, UV absorbers, X-ray \"warm absorbers\"]. A solution for the wind equations, accounting for the physical requirements typical of central regions of Seyfert-like AGN, turned out to be possible only under rather specific conditions. One of the resulting constraints is that the outflowing wind may exist only if its density is  rather low. The only possibility to increase the wind density is to feed matter into the wind itself in a certain range of distances from the nucleus (see Sect.~8.4 of Paper I for details).   As far as the BLR cloud model is concerned,  the presence of a nuclear wind  generally augments problems of cloud survival, rather than solving them, unless the clouds are somehow comoving with the outflow itself. In fact, one problem for the cloud existence is the drag force at work between the moving clouds and a non-comoving confining medium. This drag force would rapidly disrupt the clouds. The observationally inferred information  on the cloud kinematics indicates that this motion is not radially directed and it must be characterized by a more complex velocity field. Also, the possibility of a Keplerian pattern has been suggested by Wandel et al. (\\cite{wandel99}). Since, for the reasons explained above, our choice is to work within the scenario of a radially directed nuclear outflow, the cloud survival problem would inevitably be worsened by the presence of a radial nuclear wind.  For the stellar wind envelopes, since they are continuously fed by the stellar wind,  disruption is not a problem; another velocity component in their relative motion with respect to the interstellar medium does not constitute a problem either. On the contrary, we will show in this paper that  this new component in the relative motion of the envelopes with respect to the surrounding medium can be a fundamental factor; in fact, it can contribute significantly to confine the plasma envelopes via the formation of bow-shock fronts.  The above considerations, together with the necessity of feeding matter in the AGN wind, induce us to invoke stellar envelopes as  a possible interpretation of the nature of BLR emitting plasma. In addition, the virial assumption of Keplerian motions used to interpret the emission line width (Wandel et al. \\cite{wandel99}) is in good agreement  with the stellar origin of BLR emission. Indeed, this possibility has been widely discussed in the literature.  In particular, after two pioneering works of Scoville \\& Norman (\\cite {scoville}) and Kazanas (\\cite{kazanas}), the star model has been reanalyzed by Alexander \\& Netzer (\\cite{alexander94}, hereafter AN1), and Alexander \\& Netzer (\\cite{alexander97}, hereafter AN2), who introduce ``bloated'' stars as stars characterized by a very extended envelope, but, due to the specific AGN environment, different from supergiant stars as known in the solar neighbourhood. We refer to AN1 for a wider presentation of the scenario and its problems.  In their papers, Alexander \\& Netzer (AN1,AN2) present different possibilities for confining ``bloated'' star envelopes and conclude that the more effective one is tidal disruption by the black hole.  However, AN1 and AN2 assume in their works that the stellar winds expand into vacuum, asserting that this does not constitute a limitation on the validity of the analysis. On the contrary, in this paper we analyze whether and how the presence of a nuclear AGN wind, such as the one described in   Paper~I, can have an influence on the BLR physical structure. In particular, by inserting mass losing stars in their specific environment and investigating the interaction of the stellar wind with the AGN wind, we introduce another confining mechanism for the star envelopes. We then compare it to  both the tidal one  and to the other possible confinement mechanisms investigated by AN1 and AN2.  Finally  we analyze the consequences for BLR parameter values.  In Sect.~2, we discuss confinement of the expanding stellar envelopes and introduce the bow-shock formation mechanism as a further means of defining an outer boundary for these envelopes. Section~3 is devoted to the connection and interaction between the AGN nuclear wind, as developed in Paper~I, and the central star cluster, with colliding and mass losing stars, in terms of the resulting mass deposition into the AGN nuclear wind. In nSect.~4, we describe the physical parameters  that turn out to be involved in the computation of the model, their significance in the outcome determination and the general properties of a stellar envelope necessary to produce the observed line emission.  We then identify a sort of ``observational test'' for the results of a BLR model, through the comparison with typical values inferred from observations of global BLR parameters, such as the covering factor, the characteristic BLR radius, the ionization parameter, and the fact that no broad forbidden lines are observed. Section~5 discusses  how the identified general requirements put physical limitations on our model parameters. The general results of our model are presented in Sects.~5.1, 5.2, 5.3 for three distinct cases,  that differ basically in the the choice of the mass loss rates for the various types of stars in the central cluster. Section 6 is devoted to  general discussion of the main results, including a comparison with the works of other authors applying different models to infer the general properties of the BLR of  a specific source, NGC~5548, which is well representative of the Seyfert~1 class of AGNs. Finally, in Sect.~7 we summarize our results and the general features that characterize our interpretation of the BLR, highlighting the qualities and the limitations of the model. ",
        "conclusions": "The  results presented in the previous section show that  our model is capable of reproducing the observational requirements listed in Sect.~4.2  (points 1-4). In fact,  from Fig.s \\ref{standfig}, \\ref{squarefig} and \\ref{intermfig}  it is evident that the ionization parameter $U$ is in the expected range of values, while $r_{\\ma{BRL}}$ is contained in the intervals [$r_1, r_2$] where the conditions a)-d) of Sect.~4.2 are satisfied. As far as the covering factor is concerned, it is evident that its value strongly depends on the choice adopted for the stellar mass loss rates (compare Fig. \\ref{cvstan}, \\ref{cvsquare} and \\ref{cvinterm}). In the  case of enhanced and intermediate mass loss rates, values in the range 0.05- 0.25 can be recovered by choosing suitable values of the stellar envelope expansion velocities, while for standard mass loss rates this is possible only for very low velocity values. Hence, our values of $C_{\\ma {tot}}$ tend to exclude stellar envelopes with standard mass loss rates as plausible emitting units for a BLR model. On the other hand,  regarding the covering factor for ``broad'' forbidden line emission, we recall that the star density distribution we have adopted has been chosen so that in all the presented cases  $C_{\\ma {forb}}$ is minimized (see Sect.~5); in fact, it turns out to be negligible in both the cases presented in Sects.~5.2 and 5.3. We recall here that in the ``standard'' mass loss rate case, $C_{\\ma{forb}}$ is not negligible and we consider this one of the reasons why that particular model is not viable for building up a realistic BLR model.  While the analysis of this paper  has produced  strong constraints to most of the free  model parameters, for some, namely the stellar distribution, the mass loss rates and the expansion velocity, the model cannot  operate a definite choice.  Any further test of the model and selection of these parameters requires more information coming from the comparison between the observed line profiles and relative intensities, and those predicted by the model. As we have already mentioned, we have planned to compute the detailed line profiles in a forthcoming paper, having now obtained a deeper insight in the role that different parameters play in the model.  However, at this stage we would like to compare our model with the results of some other authors (Goad \\& Koratkar \\cite{goad}, Kaspi \\& Netzer \\cite{kaspi}, Korista \\& Goad \\cite{korista00}) who made attempts to infer from line profiles and intensities some ``average'' physical conditions in BLR condensations.  In the last years much work has been devoted to infer the BLR physical configuration starting from the observed BLR properties. Since the BLR emitting gas has been shown to be stratified in density (see Peterson \\cite{peterson93} and Netzer \\& Peterson \\cite{netzer} for reviews)  the idea was to derive both the dependence of  the characteristic number density of the gas in the ``cloud-like'' emitting clumps ($\\hat n_*(r)$ in our notation)  and that  of the ``cloud-like'' structures' distribution ($\\rho _*(r)$ in our notation) on  the radial distance $r$ from the central black hole. This analysis has been applied by the authors mentioned above to the case of the well studied and representative Seyfert 1 NGC 5548.  Combining a photoionization code with an optimization routine, Goad \\& Koratkar (\\cite{goad}) have found  that line ratios and variability time scales can be reproduced assuming $\\hat n_*(r) \\propto r^{-2}$. Kaspi \\& Netzer (\\cite{kaspi}), through the ``direct'' method of guessing cloud properties and distribution and calculating the resulting emission lines, have also  found that  a density gradient is necessary but their best  fit is $\\hat n_*(r) \\propto r^{-\\rm s}$ with $1\\leq {\\rm s} <1.5$. In our results $\\hat n_*(r)$  slope is variable in the interval [$r_1, r_2$] where star envelopes contribute to build up the covering factor. For $\\hat n_*(r)$ profiles shown in Fig.s \\ref{standfig}, \\ref{squarefig} and \\ref{intermfig} a piece-wise power law approximation gives logarithmic slopes from 1.5 up to 2. These values cannot be used to exclude or to prefer one of the assumed mass loss profiles, since they do not contradict any of the two previously quoted analyses.  Recently Korista \\& Goad (\\cite{korista00}) have tested the ``locally optimally emitting clouds'' (LOC) model, proposed by Baldwin et al. (\\cite {bal95}), by comparing  the predicted spectrum with that of NGC 5548. One of the results they attain is that the power law index of the quantity they define as the radial cloud covering fraction must be in the range -1.6 to -0.5. In our notation this implies that the product  $R_{\\ma {ext}}^2(r) \\rho_*(r)$ must  be a function of $r$ which is representable by a power law whose exponent should be in the range $[-1.6, -0.5]$ mentioned above. Since the star distribution function $\\rho_*(r)$ that we have adopted has a logarithmic slope ranging from -0.7 around $r/r_{\\ma g} =10^3$ down to -2 around $r/r_{\\ma g} =10^4$, this comparison favours $R_{\\ma {ext}}(r)$ functions flatter than $r^{0.75}$. This fact again implies that ``enhanced'' and ``intermediate'' mass loss rates are preferable, since in these cases the functional form of $R_{\\ma {ext}}(r)$ does indeed fulfill the requirement above (in fact, it turns out that it is $\\propto r^{0.66}$ or flatter both for MS and RG stars).  Both Korista and Goad (\\cite{korista00}) and Kaspi and Netzer (\\cite{kaspi})  derive an estimate for the extension of the BLR. For the first ones the maximum BLR extension can be up to 200~light days, while for the second ones it can be up to 100~light days. In our model, the values for the derived external BLR radius, $r_2$, for the case of ``intermediate'' mass loss (see Sect.~5.3) are larger than those derived for the case of ``enhanced'' mass loss rates presented in Sect.~5.2; however, they are in any case consistent with the estimate obtained by Korista and Goad (\\cite{korista00}). Again, this comparison is encouraging for our model, but not definite enough to allow for a selection of most appropriate mass loss rate values. In any case, the general comparison of our model with the global inferred physical conditions for the BLR discussed above  favours the two configurations analyzed in Sects.~5.2 and 5.3, characterized by distance dependent stellar mass loss rates and especially by a substantial contribution to BLR emission coming from MS star envelopes, whose mass loss rate is strongly enhanced with respect to their ``standard'' value.  Besides the X-ray illumination mechanism that we have discussed and examined in Sect.~5.2, other mechanisms have been analyzed in literature, that support the physical plausibility of this type of induced and $r$-dependent stellar mass loss  rates. Here we just want to mention a couple of these works. For example, MacDonald, Stanev \\& Biermann (\\cite{mcdonald}) performed an analysis of the effects of neutrino and high energy particle flux from the central source on the stars of the central cluster and their winds; their main result is that these stellar winds can be affected in the sense of a mass loss enhancement. Also, Baldwin et al. (\\cite{bal96}) have computed the radiative acceleration on the  photospheric layers of stars at BLR distances for a number of high-luminosity AGNs; comparing this acceleration with the star surface gravity, they conclude that all the stars, including main sequence ones, can indeed be affected, thus producing an enhanced mass loss. We recall that for the case of $r$-dependent and enhanced mass loss rates that we have discussed in Sect.~5.3, we have not explicitly defined the physical mechanism inducing the enhancement, similarly to what is done in Scoville and Norman (\\cite{scoville}) (see Sect.~5.3), but we just  supposed one of such mechanisms to be at work.  Going back to what we have pointed out above, the two cases discussed in Sects.~5.2 and 5.3 fall in the same ``category'' of models, characterized by strongly enhanced mass loss rates and a significant contribution to the BLR emission coming from MS stars. This is interesting from two different, but connected, points of view. First, within this picture the large number of individual emitting units does not imply too large collision rates between the stars. In fact, the star collision rate (Eq. (\\ref{tau})) relative  to the star distribution, $\\rho_*(r)$, for which we have computed the models whose results are shown in Sect.~5 (the one labeled with ``1'' in Fig.~\\ref{star}) turns out to be $\\tau \\simeq 1$~collisions/yr. The total star collision rate is an estimate of the total number of stellar collisions that take place in the whole cluster in a year; however, to understand to what degree the cluster structure is affected by stellar collisions another quantity must be introduced, that is the time that it takes for a star to be destroyed by collisions with other stars belonging to the cluster. Following Begelman \\& Sikora (\\cite{begelman92}),  we introduce here the destruction time due to mutual collisions for a star in the cluster as \\beq t_{\\ma {coll}}(r) \\simeq {10 \\over \\rho_*(r) f_i~\\pi (R_*)_i^2 V(r)},$$ \\label{tcoll} \\eneq where the number 10 at numerator is the number of collisions that a star has to undergo to be finally disrupted; the value used above is estimated from the fraction of stellar mass lost per collision that Begelman \\& Sikora (\\cite{begelman92}) evaluate as $\\sim 0.1$. In our model the destruction time per star turns out to be the same for red giant and main sequence stars (since $(R_*)_{\\ma{RG}}=10~(R_*)_{\\ma {MS}}$ and $f_{\\ma {RG}}=0.01~f_{\\ma {MS}}$) and it varies with the distance $r$ from the central black hole. To derive an estimate of its value, we have computed it in the region of interest for this work, that is at a distance which is around a half of the estimated external radius ($r_2$)  of the BLR model, i.e. $\\sim 10^4r_{\\ma g}$, obtaining $t_{\\ma {coll}}(x=10^4)\\simeq 8\\times 10^8 $yr. A necessary condition for our picture to be consistent is that the MS star evolution time is shorter than the star destruction time. Taking into account the value of  $t_{\\ma {coll}}$ derived above, this condition is verified  for stars with masses larger than $2.6~M_{\\odot}$ which, in a star cluster of $3\\times 10^7$ stars, assuming the Salpeter initial mass function, constitute  $\\sim 3 \\%$ of the total star number.  This percentage of stars can account for the assumed quantity of evolved stars in our model. Obviously, going closer to the cluster center, $t_{\\ma {coll}}$ decreases  and this condition can no longer be fulfilled. However, there are a few arguments  that suggest that the destruction time for collisions computed above should be considered as a sort of lower limit evaluation. In fact, as Scoville \\& Norman ({\\cite{scoville}) argue, ordered stellar motions may result in a lower collision rate.  In addition, as discussed by these same authors,  stars orbiting on elongated elliptical orbits would spend only a short part of their life in the inner region of the cluster and this would result in a larger survival probability as well. Our present discussion is centered on survival conditions for the central stellar cluster as we have chosen to model it. Its relation to the AGN lifetime (possibly $< 10^8$yr) and to the time at which the nuclear activity switches on with respect to the evolutionary stage of the cluster itself would indeed deserve further insight. However, in the present context, it is probably just the case to mention the possible relevance of this issue and we leave its analysis to different work. Back to the original point, even accounting for these last considerations, the discussion above shows that the very large values of the star number density characterizing the stellar distribution $\\rho_*(r)$ that we have chosen for our models (the one labelled with 1 in Fig.(\\ref{star})) can be regarded as very close to the upper limit for the maintainance of the cluster integrity. Indeed, higher stellar densities would result in a cluster evolution much faster than what is required when the cluster star envelopes are believed to represent the structural components of an AGN Broad Line Region. As a matter of fact, this is indeed the case when the only stellar populations forming expanding envelopes suitable for Broad Line emission (i.e., contributing to the BLR) are those of RGs and SGs. In fact, both RG and SG populations amount to only a small fraction of the total stellar number in the cluster (i.e., $ \\sim 1\\%$, see Sect.~4.1); as a consequence, requiring a number density of RG and SG envelopes, as contributing emitting units, sufficiently large to justify the observationally inferred covering factors implies a total number density of stars around 100 times larger. This would lead to a destruction time (as estimated through Eq.~(\\ref{tcoll})) unacceptably  short. (see Begelman \\& Sikora \\cite{begelman92}).   In addition, a second point of interest regarding the significant contribution of MS star envelopes to the BLR emission, is that, owing to the large number of emitting structures,  our model can be easily set in accordance with the results of the studies of bright AGN spectra, performed with cross-correlation techniques, such as those of Arav et al. (\\cite{{arav97},{arav98}}), that  claim to determine a lower limit on the number of  individual emitting structures of the BLR, for models  based on discrete emitting units composing the region. In particular, for Mk~335, these same authors derive a  lower limit for the number of reprocessing cloud-like structures, which is expected to be around $3\\times 10^6$.   To compare our model to the results inferred by Arav et al.  (\\cite{{arav97},{arav98}}) for Mk 335, we have to take into account configurations with a more powerful central AGN nucleus. Therefore, we have analyzed models for $L= 10^{45}$~erg/s, that is the luminosity of this source. To model an AGN with this luminosity, we have chosen the same central black hole mass as that shown in Table~1 of Paper~I, namely $M_{\\ma{BH}}= 1.12\\times 10^8 ~M_{\\odot}$; also, for simplicity, we suppose that the central star cluster has just the same profile $\\rho_*(r)$ as the one adopted for the case of luminosity $L=10^{44}$~erg/s (see the curve labeled with ``1'' in Fig.~\\ref{star}). With these hypotheses, we can test the resulting total number of stars within a distance corresponding to the estimated external boundary of the BLR model, comparing it with Arav et al. (\\cite{{arav97},{arav98}}) lower limit evaluation of the number of individually emitting units in the source mentioned. Our result turns out to be $(N_{\\ma {*tot}})_{45}\\simeq 3.9\\times10^7$~stars, where we have computed this number using the estimate we obtain for $(r_2)_{\\ma {MS}}\\sim 2.3\\times 10^4 r_{\\ma g}$, since the total number of stars is essentially determined by the number of MS stars  in the cluster for scenarii of the type described in Sects.~5.2 and 5.3. Our evaluation of $(N_{\\ma{*tot}})_{45}$ turns out to be well above the lower limit found by Arav et al. (\\cite{{arav97},{arav98}}), and this is another encouraging outcome. As for other properties of broad line emitting stellar envelopes in this same case of ``high'' luminosity ($L=10^{45}$~erg/s) AGN, these turn out to show behaviours that are similar to those we have discussed in Sects.~5.2 and 5.3 for the case for $L=10^{44}$~erg/s. We note in passing that the bow shock mechanism for envelope confinement is in general still the most efficient, although in this higher luminosity case Comptonization confinement (see Sect.~2, Eq.~(\\ref{rcomp})) can be dominant in the inner ($r\\simlt 500r_{\\ma g}$) portion of the model broad line emitting zone (that is, closer to the central luminosity source, where the radiation flux is stronger).  As a final remark, we note that the fact that the model outcome in this scenario  can be set in accordance with the very large number of discrete broad line emitters in the region required by the analysis performed by Arav et al. (\\cite{{arav97},{arav98}}) could be of relevance with respect to the BLR structure problem, because it would show that discrete emitting units models of the BLR are not necessarily ruled out by the analysis of Arav et al. (\\cite{arav98}).  \\medskip \\medskip"
    },
    "0209/hep-th0209119_arXiv.txt": {
        "abstract": "We study the phenomenological implications of the classical limit of the ``stringy'' commutation relations $[\\hat{x}_i,\\hat{p}_j]=i\\hbar[(1+\\beta\\hat{p}^2)\\delta_{ij} + \\beta'\\hat{p}_i\\hat{p}_j]$. In particular, we investigate the ``deformation'' of Kepler's third law and apply our result to the rotation curves of gas and stars in spiral galaxies. ",
        "introduction": "In this note, we continue our investigation \\cite{Chang:2001kn,Benczik:2002tt} of the phenomenological implications of ``stringy'' commutation relations \\cite{Kempf:1995su} which embody the minimal length uncertainty relation $\\Delta x \\ge (\\hbar/2)( \\Delta p^{-1} + \\beta\\,\\Delta p)\\;$ of perturbative string theory \\cite{gross}. In particular, we study the classical limit of the ``deformed'' commutation relations \\begin{eqnarray} \\left[\\hat{x}_i,\\hat{p}_j\\right] & = & i\\hbar \\left\\{ ( 1 + \\beta \\hat{p}^2 )\\,\\delta_{ij} + \\beta' \\hat{p}_i \\hat{p}_j \\right\\} \\;,\\cr \\left[\\hat{p}_i,\\hat{p}_j\\right] & = & 0 \\;,\\cr \\left[\\hat{x}_i,\\hat{x}_j\\right] & = & i\\hbar\\,\\frac{(2\\beta-\\beta') + (2\\beta+\\beta')\\beta \\hat{p}^2} {(1+\\beta \\hat{p}^2) } \\left( \\hat{p}_i \\hat{x}_j - \\hat{p}_j \\hat{x}_i \\right)\\;, \\label{Stringy} \\end{eqnarray} leading to the following ``deformed'' Poisson brackets, \\begin{eqnarray} \\{x_i,p_j\\} & = & ( 1 + \\beta p^2 )\\,\\delta_{ij} + \\beta' p_i p_j\\;,\\cr \\{p_i,p_j\\} & = & 0 \\;,\\cr \\{x_i,x_j\\} & = & \\frac{(2\\beta-\\beta') + (2\\beta+\\beta')\\beta p^2} { (1+\\beta p^2) } \\left( p_i x_j - p_j x_i \\right)\\;. \\label{Eq:Poisson1} \\end{eqnarray} The classical Poisson bracket is required to possess the same properties as the quantum mechanical commutator, namely, it must be anti-symmetric, bilinear, and satisfy the Leibniz rules and the Jacobi Identity. These requirements allow us to derive the general form of our Poisson bracket for any functions of the coordinates and momenta as \\cite{Benczik:2002tt} \\begin{equation} \\{F,G\\} = \\left( \\frac{\\partial F}{\\partial x_i} \\frac{\\partial G}{\\partial p_j} - \\frac{\\partial F}{\\partial p_i} \\frac{\\partial G}{\\partial x_j} \\right) \\{ x_i, p_j \\} + \\frac{\\partial F}{\\partial x_i} \\frac{\\partial G}{\\partial x_j} \\{ x_i, x_j \\}\\;, \\label{Eq:Poisson2} \\end{equation} where repeated indices are summed. Thus, the time evolutions of the coordinates and momenta in our ``deformed'' classical mechanics are governed by \\begin{eqnarray} \\dot{x}_i & = & \\{x_i,H\\} \\;=\\; \\phantom{-}\\{x_i,p_j\\}\\,\\frac{\\partial H}{\\partial p_j} + \\{x_i,x_j\\}\\,\\frac{\\partial H}{\\partial x_j} \\;,\\cr \\dot{p}_i & = & \\{p_i,H\\} \\;=\\; -\\{x_i,p_j\\}\\,\\frac{\\partial H}{\\partial x_j} \\;. \\label{Eq:Poisson3} \\end{eqnarray}  In Ref.~\\cite{Benczik:2002tt}, we analyzed the motion of objects in central force potentials subject to these equations and found that orbits in $r^2$ and $1/r$ potentials no longer close on themselves when $\\beta$ and/or $\\beta'$ are non-zero. This allowed us to place a stringent limit on the value of the minimal length from the observed precession of the perihelion of Mercury, \\begin{equation} \\hbar\\sqrt{\\beta} < 2.3\\times 10^{-68}\\;\\mathrm{m}\\;, \\label{MercuryLimit} \\end{equation} which was 33 orders of magnitude below the Planck length.  The natural question to ask next is whether there exist other ``deformations'' of classical mechanics due to $\\beta$ and/or $\\beta'$ that are either 1) observable even for such a small value of $\\hbar\\sqrt{\\beta}$, or 2) lead to an even more stringent limit due to their absence. In the following, we will look at the ``deformation'' of Kepler's third law. We will find that while the deformation is not observable on the scale of the solar system, it may be observable at galactic scales. In fact, such an effect may have already been seen in the rotation curves of gas and stars in spiral galaxies. ",
        "conclusions": "In this note, we have investigated possible observable consequences of the ``deformed'' Kepler's third law as implied by the classical limit of the ``stringy'' commutation relations, Eq.~(\\ref{Stringy}), which were in turn based on the minimal length uncertainty relation of perturbative string theory \\cite{gross}.  Due to the stringent limit on $\\beta$, Eq.~(\\ref{MercuryLimit}), obtained from the precession of the perihelion of Mercury in a previous publication \\cite{Benczik:2002tt}, the predicted sizes of the deviations for the planets are too small to be observed. However, the effect of the deformation can be amplified at galactic scales due to the large momenta involved and lead to significant changes in the rotation curves of stars in spiral galaxies.  It would be interesting to contrast our deformed classical mechanics with another modification of Newton's theory, MOND (Modified Newtonian Dynamics)~\\cite{MOND}, which apparently succeeds in predicting a flat rotation curve by modifying Newton's second law to \\begin{equation} F \\sim m \\,\\frac{|\\ddot{x}|}{a_0}\\, \\ddot{x}\\;, \\end{equation} at very small accelerations ($\\ddot{x} \\ll a_0 \\sim 10^{-10}\\,\\mathrm{m/s^2}$). Our deformed dynamics predicts different rotation curves for different masses, so it is apparently quite different from MOND which does not violate the equivalence principle. Nevertheless, it is interesting to ask whether there exists a deformation of the usual Hamiltonian dynamics which incorporates some kind of UV/IR relation and is in the same universality class as MOND. We hope to address this question in the future."
    },
    "0209/astro-ph0209170.txt": {
        "abstract": "{ We present extensive new spectroscopy and imaging of \\png. We use these data as constraints to photoionization models to derive limits on the oxygen abundance. We find that \\png\\ has an oxygen abundance less than 1/50 of the solar value. Our models favour a value of $12 + \\log \\mathrm O/\\mathrm H$ between 5.8 and 6.5\\,dex, confirming that \\png\\ is the most oxygen-poor planetary nebula known (Tovmassian et al. \\protect\\cite{tovmassianetal2001}). We also derive $\\mathrm{Ne}/\\mathrm O = 0.5 \\pm 0.3$, $\\mathrm S/\\mathrm O < 0.094$, and $\\mathrm{Ar}/\\mathrm O < 0.23$. Although the value of Ne/O is nominally high, it need not imply that the progenitor of \\png\\ converted any of its initial oxygen abundance to neon. The helium abundance  appears to be very low, $\\mathrm{He}/\\mathrm H\\sim 0.08$, but a precise determination will require a much more detailed study.  We find that $\\mathrm H\\alpha/\\mathrm H\\beta$ is lower than expected and perhaps variable, a finding for which we have no clear explanation. ",
        "introduction": "Recently, \\sbs\\ has been recognized as a planetary nebula in the Galactic halo  by  Tovmassian et al. (\\cite{tovmassianetal2001}) and renamed \\png. The spectra then available for this object were quite unusual for a planetary nebula, presenting only the Balmer lines of hydrogen, He~{\\sc ii} $\\lambda\\lambda$4686,5411, and very weak \\Oiii\\ ($\\sim4 \\%$ of \\Hb). A photoionization model analysis showed that such a spectrum implies a strongly density bounded and extremely oxygen-poor nebula ionized by a very hot star. The oxygen abundance was estimated to be less than 1/50 of the solar value, and probably between 1/100 and 1/500 of solar assuming canonical properties for the central star, making of \\png\\ by far the most oxygen-poor planetary nebula known, with an oxygen abundance similar to the lowest measured to date in stars (Boesgaard et al \\cite{boesgaardetal1999}; Howard et al \\cite{howardetal1997}).  In this paper, we report detailed follow-up observations, aimed at providing more stringent constraints on the nature of this exceptional object. Section 2 presents the new spectroscopic data, while Section 3 deals with narrow-band imaging. In Section 4, we present an updated photoionization model analysis, taking full advantage of the constraints provided by our new observational data. This leads to a limit on the oxygen abundance which is now \\emph{independent of any assumption about the evolutionary status of the central star}. In Section 5, we estimate the abundances of the other elements. Section 6 presents a brief concluding discussion. ",
        "conclusions": "Our new, extensive observations of \\png\\ confirm its unusual nature.  Spectroscopy covering the wavelength interval 3400-9700\\AA\\ reveals lines of only \\ion{H}{i}, \\ion{He}{ii}, [\\ion{O}{iii}], and [\\ion{Ne}{iii}].  Deep H$\\alpha$ imaging was used to derive the radial density distribution.  Using these data as constraints, we constructed a new set of nebular models from which we derived the abundance of oxygen and the Ne/O, S/O, and Ar/O abundance ratios.  We confirm the extremely low value of the oxygen abundance, which we find to be less than 1/50 of the solar value: our models favour a value of $12 + \\log \\mathrm O/\\mathrm H$  between 5.8 and 6.5\\,dex. The distance implied by these models places \\png\\ in the Milky Way halo, in accordance with its radial velocity (Tovmassian et al \\cite{tovmassianetal2001}). The models also imply nebular masses in the range expected.  For the $\\alpha$-element ratios, we find $\\mathrm{Ne}/\\mathrm O = 0.5\\pm 0.3$, $\\mathrm S/\\mathrm O < 0.094$, and $\\mathrm{Ar}/\\mathrm O < 0.23$. The Ne/O ratio may be somewhat higher than is commonly found in planetary nebulae in the Milky Way disk (e.g., Henry \\cite{henry1989}; Kingsburgh \\& Barlow \\cite{kingsburghbarlow1994}). One possibility is that the progenitor of \\png\\ converted some of its O to Ne. It is also possible that the anomalous Ne/O ratio is the result of discrete chemical enrichment in the very early evolution of the galaxy (e.g., Burris et al. \\cite{burrisetal2000}). Regardless, of the cause, any conversion of O to Ne has been modest and does not affect our conclusion that \\png\\ is the progeny of an intrinsically very oxygen-poor star.  An unusual characteristic of \\png\\ is its low H$\\alpha/\\mathrm H\\beta$ ratio, for which we find no clear explanation.  Despite its low metallicity and the concomitant high electron temperature that should result in collisionally excited Balmer lines of \\ion{H}{i}, \\png\\ has an $\\mathrm H\\alpha/\\mathrm H\\beta$ ratio typically below 3. Furthermore, $\\mathrm H\\alpha/\\mathrm H\\beta$ appears to be variable between observing runs and even within a single night. One possible explanation for both the low $\\mathrm H\\alpha/\\mathrm H\\beta$ ratio and its variability is if \\png\\ contains an accretion disk, though the evidence is not convincing. At any rate, this issue does not appear to affect our conclusions regarding the chemical abundances.  We also measure a low $\\mathrm{He}/\\mathrm H$ ratio of $\\sim0.08$. This makes \\png\\ interesting as a probe of the pregalactic He abundance.  However, the derivation of a very precise He abundance will require the resolution of a number of outstanding issues, including the $\\mathrm H\\alpha/\\mathrm H\\beta$ problem, the foreground reddening, and the internal temperature structure."
    },
    "0209/astro-ph0209164.txt": {
        "abstract": "We present high resolution optical spectra obtained with the HIRES spectrograph on the W. M. Keck I telescope of seven low mass T Tauri stars and brown dwarfs (LMTTs) in Taurus-Auriga.  The observed Li I 6708 \\AA\\, absorption, low surface gravity signatures, and radial velocities confirm that all are members of the Taurus star forming region; no new spectroscopic binaries are identified.  Four of the seven targets observed appear to be T Tauri brown dwarfs.  Of particular interest is the previously classified \"continuum T Tauri star\" GM Tau, which has a spectral type of M6.5 and a mass just below the stellar/substellar boundary.  These spectra, in combination with previous high resolution spectra of LMTTs, are used to understand the formation and early evolution of objects in Taurus-Auriga with masses near and below the stellar/substellar boundary.  None of the LMTTs in Taurus are rapidly rotating (vsin$i$ $<$ 30 km/s), unlike low mass objects in Orion.  Many of the slowly rotating, non-accreting stars and brown dwarfs exhibit prominent H$\\alpha$ emission (equivalent widths of 3 - 36 \\AA), indicative of active chromospheres.  We demonstrate empirically that the full-width at 10\\% of the H$\\alpha$ emission profile peak is a more practical and possibly more accurate indicator of accretion than either the equivalent width of H$\\alpha$ or optical veiling: 10\\%-widths $> 270$ km/s are classical T Tauri stars (i.e. accreting), independent of stellar spectral type.  Although LMTTs can have accretion rates comparable to that of more typical, higher-mass T Tauri stars (e.g. K7-M0 spectral types), the average mass accretion rate appears to decrease with decreasing mass.  A functional form of $\\dot{M} \\propto M$ is consistent with the available data, but the dependence is difficult to establish because of both selection biases in observed samples, and the decreasing frequency of active accretion disks at low masses (M $<$ 0.2 M$_\\odot$).  The diminished frequency of accretion disks for LMTTs, in conjunction with their lower, on average, mass accretion rates, implies that they are formed with less massive disks than higher-mass T Tauri stars.  The radial velocities, circumstellar properties and known binaries do not support the suggestion that many of the lowest mass members of Taurus have been ejected from higher stellar density regions within the cloud.  Instead, LMTTs appear to have formed and are evolving in the same way as higher-mass T Tauri stars, but with smaller disks and shorter disk lifetimes. ",
        "introduction": "Theories of star and planet formation have suggested that stars (M $\\gtrsim$ 0.075 M$_\\odot$) form from the dynamical collapse of a cloud core, while planets (M $\\lesssim$ 0.013 M$_\\odot$) form via the accretional coagulation of material in a circumstellar disk \\citep[e.g.][]{boss89}.  The intermediate mass at which the dominant formation mechanism changes from collapse to coagulation within a disk, however, is not known.  Brown dwarfs are objects with masses intermediate between those of stars and planets. The formation process of brown dwarfs, therefore, offers an important mass-link between star and planet formation.  Unfortunately, little is known about the formation of brown dwarfs, or even of the the lowest mass stars (M $< 0.2$ M$_\\odot$), because of the difficulty in determining their basic properties (e.g. mass accretion rates, rotational velocities, spectroscopic binarity) at young ages.  Accurately determining these properties generally requires high resolution spectroscopy.  At the distances of the nearest regions of star formation, however, the lowest mass stars and brown dwarfs are too faint ($R_c > 14$ mag) to have been included in previous high spectral resolution surveys \\citep[e.g.][]{bb90, heg95}.  In order to ascertain the basic properties of young low mass stars and brown dwarfs, we have carried out a high resolution spectroscopic study of mid-M spectral type (and presumably low mass) T Tauri stars in the nearby Taurus-Auriga star forming region \\citep[D $\\sim$ 140 pc;][]{bertout99}. We use these spectra to confirm the T Tauri classification of the sample and to estimate effective temperatures from which mass estimates are derived.  We then investigate the mass dependence of rotation and mass accretion across the stellar/substellar boundary.  The results are used to understand the formation of the lowest mass stars and brown dwarfs in Taurus-Auriga. ",
        "conclusions": "The signatures of youth, location, and radial velocities of the LMTTs studied here compellingly support the claim that they are members of the Taurus-Auriga star forming region.  We use their properties to investigate the rotation rates, the mass accretion rates, signatures of circumstellar accretion, and possible formation scenarios for LMTTs in Taurus-Auriga.  \\subsection{Rotational Properties}  The accretion of high angular momentum material during the earliest stages of star formation is expected to produce very rapidly rotating stars \\citep[e.g.][]{durisen89}, with rotational velocities comparable to the break-up velocity ($v_{br} = \\sqrt{GM/R} \\sim 300$ km/s).  In contrast to this prediction, young stars in Taurus rotate slowly, with rotational velocities that are typically less than one-tenth $v_{br}$ \\citep[e.g.][]{hartmann86, bouvier90, bouvier95}.  Proposed explanations for the slow rotation rates usually involve ``magnetic braking'', a mechanism which involves the star being magnetically coupled to the accretion disk and transferring angular momentum to the slower rotating outer parts of the disk, or possibly a stellar wind \\citep{konigl91, shu94}.  The slow rotation of young stars in Taurus, however, is not common to all star forming regions.  Photometric monitoring studies of stars in Orion show that a large fraction (30\\%) have rotational velocities that are larger than 0.2 times their rotational break-up velocity \\citep{stassun99}. \\citet{cb00} present a critical comparison of the rotational velocity distributions of Taurus and Orion and show that they are different at the $>$3$\\sigma$ level.  However, one significant caveat in the comparison of Taurus and Orion rotational velocities is that the Taurus sample is biased towards higher masses (87\\% $>$ 0.4 M$_\\odot$ for Taurus versus 40\\% $>$ 0.4 M$_\\odot$ for Orion). In Orion it is predominantly the lowest mass stars (M $<$ 0.2 M$_\\odot$) that are the rapid rotators \\citep{herbst01}.  With this in mind, we use the mass and $v$sin$i$ estimates derived here to investigate the mass dependence of rotation in Taurus, and to establish a less biased comparison between the rotational distributions of Taurus and Orion.  Following \\citet{cb00}, the observed $v$sin$i$ values are normalized by their break-up velocity; mass and radius estimates are obtained from the adopted evolutionary models (\\S 3.2).  These values are plotted in Figure \\ref{rot_mass}, along with the sample of higher mass\\footnote{\\citet{cb00} calculate masses using use the evolutionary models of \\citet{siess00}.} Taurus stars studied by \\citet{cb00}.  None of the low mass stars and brown dwarfs studied here rotate with speeds of more than 20\\% of their break-up velocity, similar to the distribution of higher mass stars in Taurus.  A K-S test shows that the distributions of $v$sin$i$/$v_{br}$ for stars above and below 0.2 M$_\\odot$ are indistinguishable.  Even at masses which extend into the substellar regime, Taurus does not appear to produce rapidly rotating objects.  This result contrasts significantly with the large fraction (30\\%) of stars in Orion which rotate faster than 20\\% of their break-up velocity.  Thus, in agreement with \\citet{cb00}, we conclude that Taurus and Orion produce different rotational distributions.  The variations in the rotational distributions of star forming regions pose an interesting problem for theories of low mass star formation.  Two proposed explanations for the more rapidly rotating Orion population, relative to Taurus, are (1) weaker magnetic fields, and thus weaker magnetic brakes, for stars in Orion relative to Taurus \\citep{cb00}, and (2) younger ages and thus less effective braking for stars in Orion relative to Taurus \\citep{hartmann02}.  It should be realized, however, that there is at most only marginal evidence for a difference in the magnetic field strengths \\citep{gcs95, garmire00} or ages \\citep[cf.][]{luhman00a, luhman00b} of stars in Orion and Taurus. Perhaps more fundamentally, several rotational studies have challenged the role of disk braking at early evolutionary stages \\citep{stassun99, stassun01, rebull01}. \\citet{stassun01}, for example, suggest initial conditions may be more important than disk braking in determining the rotational velocities at T Tauri ages.  In support of this, we note that the mass dependence of the rotational velocities in Orion are not easily explainable through disk braking evolution alone.  Following \\citet{hartmann02}, the disk braking timescale scales as $M_\\star\\dot{M}^{-1}f$, where $M$ is the mass of the star, $\\dot{M}$ is the mass accretion rate, and $f$ is a measure of the rotational angular velocity normalized by the breakup angular velocity.  Thus, if the mass accretion rate, $\\dot{M}$, is roughly proportional to $M$ as suggested in \\S 3.3, then the disk braking timescale should be independent of mass unless $f$, the \\textit{initial} angular momentum distribution, is mass dependent.  This argues that the rapid rotation of low mass Orion members is not a consequence of evolution, but a result of the conditions set up in the formation process.  The difference between rotational distributions of Taurus and Orion may also be a result of different initial conditions, and possibly related to Taurus's high binary fraction \\citep{gws02}, low stellar density \\citep{jh79}, and less turbulent cloud cores \\citep{myers98}.  Finally, we note that the LMTTs in Taurus (M $<$ 0.2 M$_\\odot$) have radii, determined from both pre-main sequence models and luminosity estimates, that are 4-6 times their main sequence values.  These radii are roughly twice that of more massive T Tauri stars \\citep[e.g.][]{bouvier95}. Under the assumption that disk braking terminates simultaneously for high and low mass T Tauri stars, continued contraction toward the main sequence will transform the mass independent rotational distribution into a mass dependent distribution, with the lowest mass stars and brown dwarf rotating the fastest.  If the diminished frequency of accreting disks among LMTTs implies disk braking terminates more quickly at low masses, then the resulting zero-age main sequence rotational distribution may exhibit and even stronger mass dependence.  These effects, in part, may lead to the observed rapidly rotating population of very low mass stars and brown dwarfs in the field \\citep{basri00}.   \\subsection{Distinguishing CTTSs and WTTSs based on H$\\alpha$ emission}  The most common criterion used to distinguish CTTSs from WTTSs is H$\\alpha$ emission.  Although WTTSs usually exhibit some H$\\alpha$ emission, which is attributed to their active chromospheres, this emission is limited in amount by the saturation level of the chromosphere, and is limited in profile-width by the stellar rotation and mirco-turbulent, non-thermal velocities of the chromosphere \\citep[e.g.][]{jds00}.  In contrast to this, CTTSs usually show strong, broad H$\\alpha$ emission profiles generated from the high temperatures and high velocities of accreting circumstellar material \\citep[e.g.][]{hartmann94}.  These properties are illustrated in Figure \\ref{width}, which plots the EW[H$\\alpha$]s versus the 10\\%-widths of H$\\alpha$ for a large sample of T Tauri stars.  The measurements were extracted from high resolution spectra presented in \\citet{heg95}, \\citet{ab00}, \\citet{muzerolle00b}, and \\S 3.1.4.  Multiple measurements for a star are averaged and double-lined spectroscopic binaries are excluded (DQ Tau, V826 Tau, Hen 3-600A, TWA 5A, TWA 6).  From this sample, stars that exhibit an optically veiled spectrum ($r > 0.06$) are classified as CTTSs, while stars with no optical veiling are classified as WTTSs.  Using the presence of optical veiling to identify accretion, we develop empirical criteria for distinguishing CTTSs and WTTSs.  Historically, the EW[H$\\alpha$] has been used to distinguish between WTTSs and CTTSs \\citep[e.g.][]{hb88}.  As shown in Figure \\ref{width}, however, no unique value of EW[H$\\alpha$] distinguishes all CTTSs from WTTSs.  This is primarily a consequence of the \"contrast effect\" \\citep[cf.][]{bm95}. For example, H$\\alpha$ emission from equally saturated chromospheres of a late-M star and an early K-star will appear much more prominently in the M star because of its substantially diminished photospheric continuum near 6500 \\AA.  \\citet{martin98} suggests EW[H$\\alpha$] criteria that account for this spectral type dependence.  We suggest a slight modification to these values based on the large sample of T Tauri stars presented in Figure \\ref{width} that extend to cooler spectral types than previously considered.  Specifically, we propose that a T Tauri star is classical if EW[H$\\alpha$] $\\ge$ 3 \\AA\\, for K0-K5 stars, EW[H$\\alpha$] $\\ge$ 10 \\AA\\, for K7-M2.5 stars, EW[H$\\alpha$] $\\ge$ 20 \\AA\\, for M3-M5.5 stars, and EW[H$\\alpha$] $\\ge$ 40 \\AA\\, for M6-M7.5 stars. These values are determined empirically from the maximum EW[H$\\alpha$]s for non-veiled T Tauri stars within each spectral type range (see Figure \\ref{width}).  Stars with EW[H$\\alpha$] below these levels are not necessarily WTTSs, however.  Confirmation depends upon the Li abundance \\citep[cf.][]{martin98}.  Assuming that optical veiling correctly identifies accretion, the EW[H$\\alpha$] classification is correct 95\\% of the time.  We caution that these values were determined from high spectral resolution measurements (R $>$ 20,000) and the biases in EW measurements using lower resolution spectra, as noted above, may result in slightly different (larger) distinguishing values.  As Figure \\ref{width} demonstrates, optically veiled stars are also distinguishable from stars with no optical veiling based on the 10\\%-width of H$\\alpha$ emission.  In all but one case, stars with 10\\%-widths greater than $270$ km/s are optically veiled, while stars with narrower 10\\%-widths are not.  The one exception is UX Tau A, a non-veiled early-K T Tauri star with a 10\\%-width of 475 km/s.  However, extracting optical veiling measurements for early-K T Tauri stars is difficult because of both their increasing relative stellar luminosity and the difficulty in determining spectral types \\citep[the spectral type of UX Tau A ranges from K0 to K5;][]{bb90, hss94}.  Thus, UX Tau A may in fact have some low level optical veiling, as measurements by \\citet{bb90} suggest, and perhaps should be considered a CTTS.  Therefore we propose a new accretion diagnostic: T Tauri stars with 10\\%-widths $> 270$ km/s are CTTSs.  Our choice of 10\\% of the peak flux is somewhat subjective, but this level is typically low enough to avoid biases introduced by superimposed blue-shifted absorption features and high enough to distinguish above the often uncertain continuum level in late-type stars. Based on the optical veiling as an accretion diagnostic, this classification is correct 98\\% of the time (100\\% if UX Tau A is accreting).  10\\%-widths measurements are advantageous to optical veiling measurements because they (1) can be extracted over a short wavelength range, (2) do not depend on the underlying stellar luminosity, and (3) do not depend on having a properly identified comparison template.  Thus, given the strong correlation between the presence of optical veiling and broad 10\\%-widths, H$\\alpha$ 10\\%-widths may be a more accurate diagnostic of accretion, or CTTS type, than either optical veiling or EW[H$\\alpha$].   \\subsection{The Mass Dependence of Circumstellar Accretion}  In Figure \\ref{mass_acc} the mass accretion rates or mass accretion rate upper limits are plotted versus evolutionary-model masses for the 10 stars and brown dwarfs studied here, and for the low mass T Tauri star V410 Anon 13 \\citep{muzerolle00a}.  For comparison, we also plot the sample of more massive classical T Tauri stars in Taurus from \\citet{wg01}, with mass accretion rates determined from U-band excesses following the same accretion model that is used here \\citep[i.e.][]{gullbring98} and with stellar masses estimated from the same evolutionary model that is used here.  \\citet{wg01} and \\citet{muzerolle00a} have shown evidence that the mass accretion rate in Taurus decreases toward lower masses.  The very low mass accretion rates of the lowest mass stars in these studies hinted at a strong functional form ($\\sim \\dot{M} \\propto M^{3}$).  The new accretion rates presented here, however, demonstrate that stars at the stellar/substellar boundary can have accretion rates comparable to higher, more canonical mass T Tauri stars ($\\sim$ K7-M0 spectral type; M $\\sim$ 0.7 M$_\\odot$).  Thus although the average mass accretion rate appears to decrease with decreasing mass, it is a weaker dependence than initially presumed.  A mass dependence of the form $\\dot{M} \\propto M$ (i.e. $\\dot{M}/M$ is constant) is consistent with the available data (Figure \\ref{mass_acc}), but there are yet too few data points to conduct a meaningful best fit.  We note this because circumstellar disk models with this mass dependence match the observed NIR emission from young low mass stars and brown dwarfs \\citep{nt01}.  \\citet{rebull00, rebull02} also find evidence of a similar mass dependence over the mass range 0.25 - 1 M$_\\odot$, based on ultra-violet excesses of young stars in Orion Nebula cluster flanking fields and NGC 2264.  We caution, however, that the Taurus sample, as well as more complete photometric surveys \\citep[e.g.][]{rebull00, rebull02}, may still suffer selection biases caused by high accretion stars being heavily extincted and optically fainter \\citep[see][]{wg01}. A correlation between extinction and accretion could put high accretion low mass stars below detection limits, but leave higher mass high accretion stars still observable, yielding an apparent mass dependence on the average accretion rate.  Although this possibility is worth exploring further in future studies, the current emerging trend suggests at least a modest mass dependence on the mass accretion rate.  The difficulty in determining the mass dependence of the mass accretion rate stems from the selection biases in the known samples of LMTTs, as well as the paucity of \\textit{accreting} LMTTs.  To demonstrate the latter, we determine the fraction of CTTSs in Taurus with spectral types hotter than M5, and the fraction with spectral types of M5 and cooler (for Taurus, M5 $\\sim$ 0.2 M$_\\odot$).  CTTSs are distinguished from WTTSs based on H$\\alpha$ 10\\%-widths, if that information is available, or the new EW[H$\\alpha$] criterion presented in \\S 4.2.  Using the sample of single T Tauri stars listed in \\citet{wg01} with spectral types hotter than M5, 63\\% are classical T Tauri stars (30 of 48 stars).  We note that this is consistent with the \"JHKL excess fraction\" of 69\\% in Taurus \\citep[for $m_K \\le 9.5$;][]{haisch00}; roughly two-thirds of the higher mass stars in Taurus appear to be classical T Tauri stars.  In contrast to this, of the 20 T Tauri stars in Taurus of spectral type M5 and cooler (\\S 2) and with no known companions \\citep{gws02}, only 30\\% are classical (6 of 20).  This fraction may be even lower since half of those determined to be classical are determined from EW[H$\\alpha$]s measured from low resolution spectra, which biases the EWs towards larger values (see \\S 4.2).  We conclude that both the average mass accretion rate and the frequency of accreting circumstellar disks decreases toward low masses.   \\subsection{Low Mass Star and Brown Dwarf Formation}  The processes involved in the formation of low mass stellar and substellar objects are not well understood.  The isolated gravitational collapse of a dense molecular cloud core that successfully explains solar mass star formation may work successfully well into the brown dwarf (and possibly planetary mass) regime (e.g. Boss 1993).  Alternatively, if the dominant method of formation occurs within multiple systems, as appears to be the case within the Taurus molecular cloud \\citep{wg01}, dynamical interactions could eject one or more of the lowest mass hydrostatic cores immediately after formation.  The ejected core, removed from the reservoir of material, would grow from its low possibly substellar mass \\citep{rc01}.  Thus the formation of brown dwarfs may simply be a consequence of early ejection. In support of this, there is tentative evidence for an anti-correlation between the density of stars and the density of brown dwarfs in Taurus \\citep{martin01}.  If the lowest mass objects in Taurus result from low mass cores being ejected from dynamical encounters, they may have a larger radial velocity dispersion than the $\\sim$ 2 km/s observed for higher mass stars \\citep{hartmann86} and from $^{12}$CO gas measurements \\citep{ut87}.  For example, numerical simulations of triple encounters suggest typical ejection velocities of $\\sim 4$ km/s, with occasional high velocity ejections \\citep[10\\% with $\\sim 10$ km/s;][]{sd95}.  This trend is not observed for the small sample studied here.  T Tauri stars of spectral type M5 or cooler (9 objects; M $<$ 0.17 M$_\\odot$) have a velocity dispersion of $1.9 \\pm 0.5$ km/s.  Similar velocity dispersions are determined for the sub-samples that are not within binary systems (6 objects; $\\sigma_{vel} = 2.1 \\pm 0.6$), or for the single stars of spectral type M6 or cooler (4 objects; M $<$ 0.09 M$_\\odot$ $\\sigma_{vel} = 2.3 \\pm 0.8$).  The decreasing frequency of circumstellar disks for lower mass stars and brown dwarfs is consistent, however, with an ejection scenario that would likely strip a young core of its associated material. But, the slow rotation of these low mass targets suggests, at least indirectly, that they retained disks for a sufficient time to allow some disk braking to occur.  We also note that several of the lowest mass members of Taurus are in binary systems of separations $> 10$ AU. These relatively wide systems are also inconsistent with an ejection scenario, which would preferentially eject single stars.  A more complete survey is needed to address this issue properly, however \\citep[e.g.][]{gws02}.  Nevertheless, based on the radial velocity dispersion, evidence (both direct and indirect) for circumstellar disks, and the known binary LMTTs in Taurus, we conclude these objects have not been ejected at high velocities in multiple star encounters.  Their properties suggest they experienced an early evolution that is very similar to higher-mass T Tauri stars, but with less massive disks."
    },
    "0209/gr-qc0209111_arXiv.txt": {
        "abstract": "We review recent efforts to re-formulate the Einstein equations for fully relativistic numerical simulations. The so-called numerical relativity  (computational simulations in general relativity) is a promising research field matching with ongoing astrophysical observations such as gravitational wave astronomy. Many trials for longterm stable and accurate simulations of binary compact objects have revealed that mathematically equivalent sets of evolution equations show different numerical stability in free evolution schemes. In this article, we first review the efforts of the community, categorizing them into the following three directions: (1) modifications of the standard Arnowitt-Deser-Misner equations initiated by the Kyoto group, (2) rewriting of the evolution equations in hyperbolic form, and (3) construction of an ``asymptotically constrained\" system. We next introduce our idea for explaining these evolution behaviors in a unified way using eigenvalue analysis of the constraint propagation equations.  The modifications of (or adjustments to) the evolution equations change  the character  of  constraint propagation, and several particular adjustments using constraints are expected to diminish the constraint-violating modes. We propose several new adjusted evolution equations, and include some numerical demonstrations.  We conclude by discussing some directions for future research. ",
        "introduction": "\\subsection{Numerical Relativity} The theory of general relativity describes the nature of the strong gravitational field. The Einstein equation predicts quite unexpected phenomena such as gravitational collapse, gravitational waves, the expanding universe and so on, which are all attractive not only for researchers but also for the public. The Einstein equation consists of 10 partial differential equations (elliptic and hyperbolic) for 10 metric components, and it is not easy to solve them for any particular situation. Over the decades, people have tried to study the general-relativistic world by finding its exact solutions, by developing approximation methods,  or by simplifying the situations. Among these approaches, direct numerical integration  of the Einstein equations can be said to be the most robust way to study the strong gravitational field. This research field is often called ``numerical relativity\".  \\Largefbox{\\boxwidth}{ {\\bf Numerical Relativity } \\hspace*{\\fill} {\\bf Box 1.1}\\\\ $\\qquad =$ Necessary for unveiling the nature of strong gravity. For example: \\beit \\baselineskip 8pt \\item gravitational waves from colliding black holes, neutron stars, supernovae, ... \\item relativistic phenomena like cosmology, active galactic nuclei, ... \\item mathematical feedback to singularity, exact solutions, chaotic behavior, ... \\item laboratory for gravitational theories, higher-dimensional models, ... \\enit }  Numerical relativity is now an essential field in gravity research. The current mainstream in numerical relativity is to analyze the final phase of compact binary objects (black holes and/or neutron stars) related to gravitational wave observations (see e.g. the conference proceedings \\cite{PTPsupple}). Over the past decades, many groups have developed their numerical simulations by trial and error.  Simulations require large-scale computational facilities, and long-time stable and accurate calculations. So far, we have achieved certain successes in simulating the coalescence of binary neutron stars (see e.g. \\cite{binaryNS}) and binary black holes (see e.g.\\cite{binaryBH}). However, people have still been faced with unreasonable numerical blow-ups at the end of simulations.  Difficulties in accurate/stable long-term evolution were supposed to be overcome by choosing proper gauge conditions and boundary conditions. However, recent several numerical experiments show that the (standard) Arnowitt-Deser-Misner (ADM) approach \\cite{ADM,ADM-SmarrYork,ADM-York} is not the best formulation for numerics, and finding a better formulation has become one of the main research topics. A majority of workers in the field now believe in the existence of constraint-violating modes in most of the formulations. Thus, the stability problem is now shedding light on the mathematical structure of the Einstein equations.  The purpose of this article is to review the formulation problem in numerical relativity. Generally speaking, there are many open issues in numerical relativity, both theoretical (mathematical or physical) and numerical. We list major topics in Box 1.2. More general and recent introductions to numerical relativity are available, e.g. by d'Inverno (1996) \\cite{reviewdInverno}, Seidel (1996/98/99) \\cite{reviewSeidel}, Br\\\"ugmann (2000) \\cite{reviewBruegmann}, Lehner (2001) \\cite{reviewLehner}, van Putten (2001) \\cite{reviewPutten}, and Baumgarte-Shapiro (2002) \\cite{reviewBS}. \\newlength{\\listlength} \\settowidth{\\listlength}{Theoretical:} \\Largefbox{\\boxwidth}{ {\\bf Numerical Relativity -- open issues} \\hspace*{\\fill} {\\bf Box 1.2} \\baselineskip 11pt \\been \\item[0.] How to select the foliation method of space-time \\\\ $~\\qquad~\\qquad~\\quad$Cauchy ($3+1$), characteristic ($2+2$), or combined? \\hspace*{\\fill} \\enen $\\Rightarrow$ if the foliation is $(3+1)$, then $\\cdots$ \\been \\item How to prepare the initial data \\begin{list}{}{% \\setlength{\\leftmargin}{\\listlength} \\addtolength{\\leftmargin}{\\labelsep} \\setlength{\\labelwidth}{\\listlength} } \\item[Theoretical:\\hfill] Proper formulation for solving constraints? \\\\ How to prepare realistic initial data?  \\\\ Effects of background gravitational waves?  \\\\ Connection to the post-Newtonian approximation? \\item[Numerical:\\hfill] Techniques for solving coupled elliptic equations?\\\\ Appropriate boundary conditions? \\end{list} \\item How to evolve the data \\begin{list}{}{% \\setlength{\\leftmargin}{\\listlength} \\addtolength{\\leftmargin}{\\labelsep} \\setlength{\\labelwidth}{\\listlength} } \\item[Theoretical:\\hfill] Free evolution or constrained evolution? \\\\ \\underline{Proper formulation for the evolution equations?} \\hspace*{\\fill}$\\Leftarrow\\Leftarrow\\Leftarrow$ this review \\\\ Suitable slicing conditions (gauge conditions)? \\item[Numerical:\\hfill] Techniques for solving the evolution equations?\\\\ Appropriate boundary treatments?  \\\\ Singularity excision techniques?  \\\\ Matter and shock surface treatments?\\\\ Parallelization of the code? \\end{list} \\item How to extract the physical information \\begin{list}{}{% \\setlength{\\leftmargin}{\\listlength} \\addtolength{\\leftmargin}{\\labelsep} \\setlength{\\labelwidth}{\\listlength} } \\item[Theoretical:\\hfill] Gravitational wave extraction?\\\\ Connection to other approximations? \\item[Numerical:\\hfill] Identification of black hole horizons?  \\\\ Visualization of simulations? \\end{list} \\enen }  \\subsection{Formulation Problem in Numerical Relativity: Overview}  There are several different approaches to simulating the Einstein equations. Among them the most robust way is to apply 3+1  (space + time) decomposition of space-time, as was first formulated by Arnowitt, Deser and Misner (ADM) \\cite{ADM} (we call this  the ``original ADM\" system). \\footnote{ One alternative method of space-time foliation is the so-called characteristic approach ($2+2$ space-time decomposition). See reviews e.g. by d'Inverno (1996) \\cite{reviewdInverno}, Winicour \\cite{reviewWinicour}, Lehner (2001) \\cite{reviewLehner}. Even in the 3+1 ADM approach, we concentrate the standard finite differential scheme to express numerical expression of space-time. See e.g. Brewin \\cite{brewin} for a recent progress in a lattice method.}  If we divide the space-time into 3+1 dimensions, the Einstein equations form a constrained system: constraint equations and evolution equations. The system is quite similar to that of the Maxwell equations (Box 1.3), \\largefbox{\\boxwidth}{ {\\bf The Maxwell equations :} \\hspace*{\\fill} {\\bf Box 1.3} \\\\ \\baselineskip 11pt The evolution equations:  ($\\ptl_t=\\ptl / \\ptl t$) \\bear {\\ptl_t {\\bf E}} &=& \\mbox{rot~} {\\bf B} % -{4\\pi } {\\bf j}, \\qquad \\mbox{and}\\qquad {\\ptl_t {\\bf B} } = - \\mbox{rot~}  {\\bf E} \\enar Constraint equations: \\bear \\mbox{div~} {\\bf E} &=& 4 \\pi \\rho, \\qquad \\mbox{and}\\qquad \\mbox{div~}{\\bf B} = 0 \\enar } where people solve constraint equations on the initial data, and use evolution equations to follow the dynamical behaviors.  In numerical relativity, this free-evolution approach is also the standard. This is because solving the constraints (non-linear elliptic equations) is numerically expensive, and because free evolution allows us to monitor the accuracy of numerical evolution.  In black-hole treatments, recent ``excision\" techniques do not require one to impose explicit boundary conditions on the horizon, which is also a reason to apply free evolution scheme. As we will show in the next section, the standard ADM approach has two constraint equations; the Hamiltonian (or energy) and momentum constraints.   Up to a couple of years ago, the ``standard ADM\" decomposition  \\cite{ADM-SmarrYork,ADM-York} of the Einstein equation was taken as the standard formulation for numerical relativists. However, numerical simulations were often interrupted by unexplained blow-ups (Figure.\\ref{fig1}).  This was thought due to the lack of resolution, or inappropriate  gauge choice, or the particular numerical scheme which was applied.   However, after the accumulation of much experience,  people have noticed the importance of the formulation of the evolution equations, since there are  apparent differences in numerical stability although the equations are mathematically equivalent \\footnote{The word {\\it stability} is used quite different ways in the community. \\beit \\item We mean by {\\it numerical stability} a numerical simulation which continues without any blow-ups and in which data remains on the constrained surface. \\item {\\it Mathematical stability} is defined in terms of the well-posedness in the theory of partial differential equations, such that the norm of the variables is bounded by the initial data. See eq. (\\ref{energynorm}) and around. \\item For numerical treatments, there is also another notion of {\\it stability}, the stability of finite differencing schemes.  This means that numerical errors (truncation, round-off, etc) are not growing by evolution, and the evaluation is obtained by von Neumann's analysis. Lax's equivalence theorem says that if a numerical scheme is consistent (converging to  the original equations in its continuum limit) and stable (no error growing), then the simulation represents the right (converging) solution. See \\cite{Choptuik91} for the Einstein equations.  % \\enit }.  \\begin{figure}[t] \\unitlength 1mm \\begin{picture}(160,70) \\put(00,0){\\epsfxsize=75mm \\epsffile{fig1L_error-time.eps} } \\put(90,0){\\epsfxsize=75mm \\epsffile{fig1R_divergingM.eps} } \\end{picture} \\caption{Origin of the problem for numerical relativists: Numerical evolutions depart from the constraint surface. } \\label{fig1} \\end{figure}   At this moment, there are three major ways to obtain longer time evolutions. Of course, the ideas, procedures, and problems are mingled with each other. The purpose of this article is to review all three approaches and to introduce our idea to view them in a unified way. Table \\ref{table:hyprefs} is a list of references.  \\begin{enumerate} \\item[(1)] The first possibility is to use a modification of the ADM system developed by the Kyoto group \\cite{SN87,SN89} (often cited as Shibata and Nakamura \\cite{SN}) and later re-introduced by Baumgarte and Shapiro \\cite{BS}. This is a combination of new variables, conformal decomposition, rescaling of the conformal factor, and replacement of terms in the evolution equation using momentum  constraints  (see \\S \\ref{secBSSN}). \\item[(2)] The second direction is to re-formulate the Einstein equations in a first-order hyperbolic form. This is motivated from the expectation that the symmetric hyperbolic system has well-posed properties in its Cauchy treatment in many systems and also that the boundary treatment can be improved if we know the characteristic speed of the system.  In constructing hyperbolic systems, the essential  procedures are to adjust equations using constraints and to introduce new variables, normally the spatially derivatived metric  (see \\S \\ref{secHYP}). \\item[(3)] The third is to construct a system which is robust against the violation of constraints, such that the constraint surface is an attractor. The idea was first proposed as a ``$\\lambda$-system\" by Brodbeck et al \\cite{BFHR} in which they introduce artificial flow to the constraint surface using a new variable based on the symmetric hyperbolic system (see \\S \\ref{secASYMPT}). \\end{enumerate} The third idea has been generalized by us as an asymptotically constrained system.  The main procedure is to adjust the evolution equations using the constraint equations \\cite{ronbun2,adjADM,adjADMsch}. The method is also applied to explain why the above approach (1) works, and also to propose alternative systems based on the ADM \\cite{adjADM,adjADMsch} and BSSN \\cite{adjBSSN} equations. Section \\ref{secADJUSTED} is devoted to explain this idea with an analytical tool of the eigenvalue analysis of the constraint propagation.  We follow the notations of that of MTW\\cite{MTW}, i.e. the signature of the space-time is $(-+++)$, and the Riemann curvature is defined as \\bear R^{\\mu}_{~\\nu\\alpha\\beta}&\\equiv& \\partial_\\alpha \\Gamma^\\mu_{\\nu\\beta}- \\partial_\\beta \\Gamma^\\mu_{\\nu\\alpha} +\\Gamma^\\mu_{~\\sigma\\alpha}\\Gamma^\\sigma_{~\\nu\\beta} -\\Gamma^\\mu_{~\\sigma\\beta}\\Gamma^\\sigma_{~\\nu\\alpha} \\\\ R_{\\mu\\nu}&\\equiv& R^{\\alpha}_{~\\mu\\alpha\\nu} \\enar We use $\\mu,\\nu=0,\\cdots,3$ and $i,j=1,\\cdots,3$ as space-time indices. The unit $c=1$ is applied. The discussion is mostly to the vacuum space-time, but the inclusion of matter is straightforward.   \\begin{table} \\begin{center} \\begin{tabular}{cl|l|l} \\hline \\hline & & formulations  & numerical applications \\\\ \\hline \\hline \\multicolumn{4}{l}{(0) The standard ADM formulation} \\\\ \\hline &ADM & 1962 Arnowitt-Deser-Misner \\cite{ADM,ADM-SmarrYork} & $\\Rightarrow$ many \\\\ \\hline \\multicolumn{4}{l}{(1) The BSSN formulation} \\\\ \\hline &BSSN & 1987 Nakamura et al \\cite{SN87,SN89,SN} & $\\Rightarrow$ 1987 Nakamura et al \\cite{SN87,SN89} \\\\ &&& $\\Rightarrow$ 1995 Shibata-Nakamura  \\cite{SN} \\\\ &&& $\\Rightarrow$ 2002 Shibata-Uryu \\cite{binaryNS} etc \\\\ && 1999 Baumgarte-Shapiro \\cite{BS} & $\\Rightarrow$ 1999 Baumgarte-Shapiro \\cite{BS} \\\\ &&& $\\Rightarrow$ 2000 Alcubierre et al \\cite{potsdam9908,potsdam0003} \\\\ &&& $\\Rightarrow$ 2001 Alcubierre et al \\cite{binaryBH} etc \\\\ && 1999 Alcubierre et al \\cite{ABMS} & \\\\ && 1999  Frittelli-Reula \\cite{FR99}   & \\\\ && 2002 Laguna-Shoemaker \\cite{PabloDeirdre} & $\\Rightarrow$ 2002 Laguna-Shoemaker \\cite{PabloDeirdre} \\\\ \\hline \\multicolumn{4}{l}{(2) The hyperbolic formulations} \\\\ \\hline &BM & 1989 Bona-Mass\\'o \\cite{BM89,BM92,BMSS95} & $\\Rightarrow$ 1995 Bona et al \\cite{BMSS95,BMSS97,cactus1} \\\\ &&& $\\Rightarrow$ 1997 Alcubierre, Mass\\'o \\cite{Alcubierre,AM} \\\\ &&   1997 Bona et al \\cite{BMSS97} & $\\Rightarrow$ 2002 Bardeen-Buchman \\cite{BB} \\\\ && 1999 Arbona et al \\cite{ArBona}  & \\\\ &CB-Y & 1995 Choquet-Bruhat and York \\cite{CY9506071} & $\\Rightarrow$ 1997 Scheel et al \\cite{SBCSThyper} \\\\ &  & 1995 Abrahams et al \\cite{CY9506072}  & $\\Rightarrow$ 1998 Scheel et al \\cite{SBCST98} \\\\ &  & 1999 Anderson-York \\cite{AY}  & $\\Rightarrow$ 2002 Bardeen-Buchman \\cite{BB} \\\\ &FR & 1996 Frittelli-Reula \\cite{FR96} & $\\Rightarrow$ 2000 Hern \\cite{HernPHD} \\\\ && 1996 Stewart \\cite{Stewart} & \\\\ &KST & 2001 Kidder-Scheel-Teukolsky \\cite{KST} & $\\Rightarrow$ 2001 Kidder-Scheel-Teukolsky \\cite{KST} \\\\ && & $\\Rightarrow$ 2002 Calabrese et al \\cite{LSU-KST} \\\\ && & $\\Rightarrow$ 2002 Lindblom-Scheel \\cite{LindblomScheel} \\\\ && 2002 Sarbach-Tiglio \\cite{SarbachTiglio} & \\\\ & CFE & 1981 Friedrich\\cite{FriedrichCFE}  & $\\Rightarrow$ 1998 Frauendiener \\cite{Frauendiener} \\\\ &&& $\\Rightarrow$ 1999 H\\\"ubner \\cite{Hubner} \\\\ & tetrad & 1995 vanPutten-Eardley\\cite{vanPutten95}  & $\\Rightarrow$ 1997 vanPutten \\cite{vanPutten97} \\\\ &Ashtekar & 1986 Ashtekar \\cite{Ashtekar} & $\\Rightarrow$ 2000 Shinkai-Yoneda \\cite{ronbun1} \\\\ &&1997 Iriondo et al \\cite{Iriondo} & \\\\ &&1999 Yoneda-Shinkai \\cite{ysPRL,ysIJMPD} & $\\Rightarrow$ 2000 Shinkai-Yoneda \\cite{ronbun1,ronbun2} \\\\ \\hline \\multicolumn{4}{l}{(3) Asymptotically constrained formulations} \\\\ \\hline $\\lambda$-system & to FR & 1999 Brodbeck et al \\cite{BFHR} & $\\Rightarrow$ 2001 Siebel-H\\\"ubner \\cite{SiebelHuebner} \\\\ & to Ashtekar  & 1999 Shinkai-Yoneda \\cite{SY-asympAsh} & $\\Rightarrow$ 2001 Yoneda-Shinkai \\cite{ronbun2} \\\\ adjusted  & to ADM & 1987 Detweiler \\cite{detweiler}  & $\\Rightarrow$ 2001 Yoneda-Shinkai \\cite{adjADM} \\\\ & to ADM & 2001 Shinkai-Yoneda \\cite{adjADM,adjADMsch} & $\\Rightarrow$ 2002 Mexico NR Workshop \\cite{mexico} \\\\ & to BSSN & 2002 Yoneda-Shinkai \\cite{adjBSSN} & $\\Rightarrow$ 2002 Mexico NR Workshop \\cite{mexico} \\\\ & & & $\\Rightarrow$ 2002 Yo-Baumgarte-Shapiro \\cite{YBS} \\\\ \\hline \\end{tabular} \\end{center} \\caption{References to recent efforts of reformulating the Einstein equations. We list mainly those that have been applied to actual numerical comparisons.} \\label{table:hyprefs} \\end{table}    \\newpage \\setcounter{equation}{0} ",
        "conclusions": ""
    },
    "0209/astro-ph0209360_arXiv.txt": {
        "abstract": "Using the HEGRA system of imaging atmospheric Cherenkov telescopes, one quarter of the Galactic plane ($-2^\\circ < l < 85^\\circ$) was surveyed for TeV gamma-ray emission from point sources and moderately extended sources (\\O$\\;\\le0.8^\\circ$). The region covered includes 86 known pulsars (PSR), 63 known supernova remnants (SNR) and nine GeV sources, representing a significant fraction of the known populations. No evidence for emission of TeV gamma radiation was detected, and upper limits range from 0.15 Crab units up to several Crab units, depending on the observation time and zenith angles covered. The ensemble sums over selected SNR and pulsar subsamples and over the GeV-sources yield no indication for emission from these potential sources. The upper limit for the SNR population is at the level of 6.7\\% of the Crab flux and for the pulsar ensemble at the level of 3.6\\% of the Crab flux. ",
        "introduction": "\\begin{figure} \\begin{center} \\mbox{ \\epsfxsize7.8cm \\epsffile{xxx_fig1.eps}} \\caption{Observation time in hours used for the individual scan points. The large gray circles indicate the individual pointings and correspond the used FoV of the telescope system. Positions of potential TeV gamma-ray sources are marked by symbols; a filled symbol indicates a potential source for which we give an upper limit. The size of the circles for SNRs and GeV sources corresponds to the size of the source. Objects in the dashed box labeled (A) are excluded from further analysis (for explanation see text). } \\label{obs_time} \\end{center} \\end{figure} \\begin{figure} \\begin{center} \\mbox{ \\epsfxsize8.5cm \\epsffile{xxx_fig2.eps}} \\caption{Correlation between Galactic longitude and zenith angle under which the individual scan points were observed. Each dot represents a data taking period of 20 or 30 min. The solid line indicates the approximated energy threshold in TeV (right axis) as a function of Galactic longitude (Konopelko et al., 1999b).} \\label{fig_alt_l_correlation} \\end{center} \\end{figure} Systems of imaging atmospheric Cherenkov telescopes such as the HEGRA stereoscopic telescope system (Daum et al. 1997, Konopelko et al. 1999a), allow to reconstruct the directions of air showers over the full field of view, with a radius of about 2$^\\circ$ in the case of HEGRA, and can therefore be used for sky surveys (P\\\"uhlhofer et al. 1999, Aharonian et al. 2001b). Here, we report on a survey of one quarter of the Galactic disc ranging from the Galactic center (l$\\;\\approx 0^\\circ$) to the Cygnus region (l$\\;\\approx 83^\\circ$). The latitude range covered corresponds in most parts of the survey to the FoV of the HEGRA telescope system and ranges from $-1.7^\\circ$ to $1.7^\\circ$ (for more details see Fig. \\ref{obs_time}). The motivation for this survey was to search for gamma-ray point sources and moderately extended sources in the TeV energy range. Most of the potential Galactic gamma-ray sources like supernova remnants (SNR) (Green, 1998) and pulsars (PSR) (Taylor, 1993) are the remnants of young massive (Population I) stars and thus cluster along the Galactic plane and concentrate towards the Galactic center. This picture is supported by earlier $\\gamma$-ray surveys carried out with the COS B satellite (Swanenburg et al., 1981) and with the EGRET instrument (Hartmann et al. 1999, Lamb and Macomb, 1997) in the GeV range revealing an enhancement of $\\gamma$-ray sources along the Galactic plane. Both types of objects - SNRs and pulsars - are almost certainly particle accelerators and emitters of high-energy gamma radiation. Theoretical models predict typical gamma-ray fluxes from the majority of these objects are below the detection thresholds of the current generation of Cherenkov instruments (see, e.g. Drury et al. 1994, Aharonian et al. 1997 and Berezhko \\& V\\\"olk 2000a). Until now only three SNRs - SN1006 (Tanimori et al. 1998), RX J1713.7-3946 (Muraishi et al. 2000) in the southern hemisphere and Cas-A (Aharonian et al. 2001a) in the northern hemisphere show evidence for TeV gamma-ray emission. For SN1006 a flux at the level of 70\\% of the Crab flux\\footnote{To keep calculations simple, we give fluxes in units of the Crab Nebula flux (so called CU). For the Crab we take a value of:\\\\ $F(>E) = 1.75\\times10^{-11} \\left(\\frac{E}{1~TeV}\\right)^{-1.59} \\mathrm{ph~cm^{-2} s^{-1}}$\\\\ (Aharonian et al. 2000) }is reported, for RX J1713.7-3946 at the level of 80\\% and for Cas-A at the level of 3.3\\%. For the individual shell type SNRs, $\\gamma$-Cygni, IC-433, W44, W51 upper limits at the level of 20\\% to 30\\% of the Crab flux are given in Buckley (1998) and V\\\"olk (1997). For the SNR W28 an upper limit of 70\\% of the Crab flux is given in Rowell et al. (2000) and for Tycho an upper limit of 3.3\\% of the Crab is given in Aharonian et al. (2001c). Three pulsars - the Crab Nebula (Weekes 1989), PSR1706-44 (Kifune et al. 1995) at the level of 60\\% and Vela (Yoshikoshi et al. 1997) at the level of 70\\% of the Crab flux have been reported as gamma-ray emitters in the TeV regime. For a review of observations and theoretical predictions on Galactic gamma-ray sources see, e.g, Aharonian, 1999c. In addition to pulsars and SNRs, many unidentified GeV sources (Lamb \\& Macomb 1997) lie in the Galactic plane.\\\\ Both the lack of knowledge of the individual source parameters as well as the approximations used in the modeling result in large uncertainties in the predictions for individual objects by an order of magnitude or more. Hence it is desirable to observe a larger sample of source candidates beyond the few most promising representatives of each class. Given the density of source objects, a survey of the inner part of the Galactic plane provides an efficient way to search for gamma-ray emission and to average over the potential source populations. \\\\ With the HEGRA telescope system such a survey was conducted. The range of the survey, $-2^\\circ < l < 85^\\circ$, was chosen either by visibility conditions and by the density of potential gamma-ray emitters. From the location of the HEGRA telescope system at $28^\\circ 45'$ N, observation conditions are best for Galactic longitudes around $65^\\circ$. The Galactic center can only be observed at large zenith angles around $60^\\circ$, and most parts of the Galactic plane with negative longitudes are virtually inaccessible. ",
        "conclusions": "In a systematic search for point sources in the Galactic plane in the longitude range from -2$^o$ to 85$^o$ with the HEGRA IACT system no TeV gamma-ray emission was detected on a level above 4.5$\\sigma$ in a total observation time of 115 h. Upper limits for 63 SNRs, 86 pulsars and nine unidentified GeV-sources on the level between 7\\% of the Crab flux and up to 18 Crab flux units were derived, depending on observation time and zenith angle. Summation over the most promising sources for TeV gamma-ray emission within each source class did not yield an indication for emission from the SNR ensemble, the PSR ensemble or the ensemble of GeV-sources. For the ensemble of 7 GeV sources an upper limit of 5.7\\% compared to the Crab flux was derived. For the ensemble of 18 pulsars selected by characteristic age and distance a similar upper limit of 3.6\\% was produced. A theoretical estimate for these pulsars using the same conversion efficiency from rotational energy to gamma-rays as in the Crab Nebula gives a flux of approximately a factor 400 lower than the derived limit. \\\\ For an ensemble of 19 selected SNRs a limit of 6.7\\% of the Crab flux was derived. Comparing this limit with a predicted hadronic gamma-ray flux of 2.9\\% according to the DAV model and reasonable parameters rules out a strong enhancement of the emission of the SNR population compared to the model predictions.\\\\ While no new TeV sources could be established in this survey, we nevertheless note that with systems of IACTs such a survey provides an efficient method to probe extended regions of the sky with a dense population of sources, such as the Galactic plane."
    },
    "0209/astro-ph0209010_arXiv.txt": {
        "abstract": "We analyze observations of the low mass X-ray binary 2S0921-63 obtained with the gratings and CCDs on Chandra and XMM.  This object is a high inclination system showing evidence for an accretion disk corona (ADC). Such a corona has the potential to constrain the properties of the heated accretion disk in this system, and other LMXBs by extension.  We find evidence for line emission which is generally consistent with that found by previous experiments, although we are able to detect more lines.  For the first time in this source, we find that the iron K line has multiple components.  We set limits on the line widths and velocity offsets, and we fit the spectra to photoionization models and discuss the implications for accretion disk corona models.  For the first time in any ADC source we use these fits, together with density constraints based on the O VII line ratio, in order to constrain the flux in the medium-ionization region of the ADC.  Under various assumptions about the source luminosity this constrains the location of the emitting region.  These estimates, together with estimates for the emission measure, favor a scenario in which the intrinsic  luminosity of the source is comparable to what we observe. ",
        "introduction": "In spite of much study, understanding of the geometry of the gas flows and circumstellar gas in low mass X-ray binaries, (LMXBs) remains uncertain. The conventional picture consisting of a neutron star, low mass companion, and accretion disk is heavily influenced by analogy with cataclysmic variables. There are fewer direct observational constraints on the accretion disks in LMXBs than in other accretion driven systems. A major contributor to this uncertainty is the influence of X-ray heating, which makes circumstellar gas a mirror for X-rays rather than a medium which can be studied via its intrinsic emission. In addition, dilution of the circumstellar emission by the strong continuum emanating from the neutron star adds to the difficulty of finding signatures of the disk or other structures in the spectrum.  Useful clues to the geometry of LMXBs may be provided by the sources in which we are likely observing close to the plane of the accretion disk and binary orbit.   These sources may play a role in our understanding of LMXBs which is analogous to the role played by Seyfert 2 galaxies in the understanding of active galaxies; the configuration of the circumstellar gas can be studied when the direct X-rays from the continuum source are at least partially blocked. In such high inclination systems the presence of partial eclipses by the secondary star suggest the presence of an extended component of X-ray emission, likely due to scattering and emission from a cloud of highly ionized gas.  The X-rays from the neutron star may provide sufficient heating and ionization to account for the properties of this accretion disk corona, or ADC (Begelman, Mckee and Shields 1982; Begelman and McKee 1983).  In addition, the heated gas may flow outward in a wind (Woods et al., 1996; Proga and Kallman 2002) which can potentially affect the mass budget of the system. The existence of accretion disk coronae has been substantially confirmed, and the column density and size have been constrained, by fitting of models to observed X-ray light curves. Simple models treat the corona as a spherical cloud (White and Holt, 1982; McClintock, et al., 1982; Mason, 1986) in which the X-rays are reflected into our line of sight by scattering alone.  These models have also constrained the size and shape of the outer rim of the accretion disk, although the results depend on the assumed geometry for the ADC.  In addition to scattering continuum photons, the corona should also radiate in atomic emission features arising from partially ionized material. Such emission has been predicted by, e.g. Kallman and White (1989), and by Vrtilek, Soker and Raymond (1994), Ko and Kallman (1994). Strong line emission is expected under a variety of assumptions, and that the line emission (as measured by equivalent width) is strongest when the inclination of the binary system is closest to 90$^o$.  Observations of line emission have the potential to provide a sensitive test of the hypothesis that the corona is heated and ionized entirely by X-rays from the central neutron star, since the ionization balance and emission measure together serve to constrain both the gas density and the size of the emission region.  In order to carry out such a test we have used the gratings and CCD detectors on both the Chandra and the XMM satellites to observe the eclipse spectrum of 2S0921-63, a well known LMXB which shows the approximately sinusoidal lightcurve indicative of an accretion disk corona. Its period of 9.01 days makes it one of the longest period LMXB systems, and also makes it particularly suitable for eclipse studies owing to the duration of the eclipse. EXOSAT observations of the spectrum through an orbital cycle (Mason et al. 1987) showed that the spectrum softens during eclipse.  The spectrum was fit to a hard power law, with a photon index of 1.16, and a column density of 1.4$\\times 10^{21}$ (although Mason et al. find evidence for contamination at low energies which makes this value somewhat uncertain). The eclipse ($\\leq 50 \\%$ of the uneclipsed flux) lasted approximately 80 ksec. The relatively low X-ray to optical flux ratio and many recorded optical dips  (e.g. Branduardi-Raymont et al., 1981, 1983; Chevalier and Ilovaisky, 1981, 1982) are further evidence for a  hidden X-ray source. Cowley et al. (1982)  estimate the distance to be $\\simeq$7 kpc, implying a luminosity of 2.4 $\\times 10^{35}$ erg s$^{-1}$.  2S0921-63 was observed with ASCA on several occasions, including a total of 27 ksec of good data obtained during eclipse.   Fits to these data reveal a hard power law continuum and column density comparable to those observed by EXOSAT. This spectrum is significantly harder than the spectra typically observed from low inclination LMXB  systems.  The spectrum contains an iron K line at an energy of 6.75 keV, intermediate between the He-like and H-like ionization states of this element.  This line is unique among LMXBs for its equivalent width; both ASCA and EXOSAT (Asai, 2000; Gottwald et al., 1996) measure $\\simeq 120$eV for this quantity, while typical LMXBs show much weaker lines, $\\sim 40-80$ eV. The flux in the 2-10 keV band in the ASCA spectrum of 2S0921-63 is 3.2 $\\times 10^{-11}$ erg cm$^{-2}$ s$^{-1}$. In addition, the ASCA spectrum contains evidence for  line emission in the 0.5-2 keV band, not previously published.  In the following sections we describe our observations of this source using Chandra and XMM, the extraction and analysis procedures, results of model fits, and their implications for understanding of 2S0921-63 and for ADC sources in general. ",
        "conclusions": "\\subsection{Velocity Offsets and Widths}  In figure 9  we compare the velocity offsets of the lines observed by the various instruments.  The velocity offsets are determined using line wavelengths from the NIST database, where available, and from the compilation of Verner (1999) for the others. Figure 9 shows that essentially all the line wavelengths are consistent with being emitted at zero velocity, although the uncertainties are typically $\\sim$1000 km/s or greater.  Only the wavelength of the Fe XXVI L$\\alpha$ near 1.78 $\\AA$ (6.97 keV) as measured by the Chandra HETG is inconsistent with zero velocity, although the wavelength of this line as measured by the XMM PN is consistent with zero velocity. In the case of the line observed at 1.85 $\\AA$ (6.7 keV) we include the data for all 3 of the components of the 1-2 lines of Fe XXV, with the result  that the wavelengths of the forbidden and semi-forbidden lines are consistent with smaller velocity offsets than the allows line.  This suggests that, if the line-emitting material does not have a high bulk velocity relative to us, then the lines are formed predominantly by recombination.  The limits on line widths allow us to approximately constrain the location of the line emitting gas, under the assumption that the gas motions are Keplerian in the vicinity of a 1 $M_\\odot$ compact object. If so, the limits on the velocity obtained for the 1.78 $\\AA$ (6.97 keV) line using the XMM PN correspond to a radii between 8.8 $\\times 10^8$ cm and 7.2 $\\times 10^{10}$ cm.  Although there are no independent measures of the mass, it is likely that the inclination $i$ is close to 90$^o$.  This can be compared with estimates for the disk outer radius.  Using standard expressions for the Roche lobe radii (eg. Frank, King and Raine, 1992), and assuming values for the mass ratio $q$ in the range 1 -- 2.2 (Shabhaz et al., 1999) the radius of the compact object Roche lobe is in the range 0.9 -- 1.2 $\\times 10^{12}$ cm. Although estimates for the disk radius itself are less certain, it is clear that the iron emission originates at a small fraction ($\\sim$ 10$\\%$) of the probable outer disk radius. The width of the 1.94 $\\AA$ (6.4 keV) iron K$\\alpha$ line measured by the XMM PN is much greater, 0.27 keV. This  corresponds  to the Keplerian broadening at a disk radius of $7 \\times 10^7$ cm. This difference in widths is surprising if it actually reflects the range of radii where the lines are emitted.  A more likely explanation is that the width of the 1.94 $\\AA$ (6.4 keV) line is due to blending of multiple components of the fluorescence  lines from various ionization stages of iron.  If so, the estimate of radius obtained from the 1.78 $\\AA$ (6.97 keV) line remains relevant.  \\subsection{Line Luminosities, Variability, and Emission Measures}  The fluxes we derive for the various experiments are 5.4, 3.8, and 5.9  $\\times 10^{-11}$ erg cm$^{-2}$ s$^{-1}$ in the 2-10 keV band for ASCA, Chandra, and XMM respectively. The fact that the ASCA and XMM are consistent, and that the Chandra flux is slightly lower, further confirms the conclusion that the fact that the Chandra spectrum requires only a high ionization parameter component while the XMM RGS requires both high ad low ionization parameter components is not due to variability.  The lower ionization parameter inferred from the XMM fit would suggest a lower flux during the XMM observation if variability were the origin for the difference.  However, the difference is in the opposite direction, and the Chandra observation has lower flux.  From this we conclude that the difference in the best fit ionization parameter for the two observations is due to the differing wavelength bands of the two instruments together with an intrinsic distribution in ionization parameter within the source, rather than variability.  Figure 11 shows the emissivities for the various lines we observe as a function of ionization parameter, $\\xi$, as calculated for a library of single-zone xstar photoionization models.  Also included in this are lines which we do not detect, and whose absence places constraints on the conditions in the ADC.  These include principally the lines of He-like ions such as Ne IX, Mg XII, and Si XIII.  From this figure it is clear that the absence of these lines in the Chandra HETG spectra implies ionization parameter values greater than $\\xi \\sim 10^4$.  Conversely, the best fit to the XMM RGS spectrum that we find is affected by the need to fit the O VII line near 21 $\\AA$, and this requires ionization parameters $\\xi \\sim 10^1$.  The line strengths can be used to infer the emission measure of the gas in 2S0921-630 by dividing the line luminosity by the emissivity of the various lines.  As example of such a calculation is shown in figure 10, for the various strong lines detected by Chandra and XMM.  These estimates were derived by using the line emissivities shown in figure 11, and summing over an arbitrary distribution of ionization parameters binned into 9 bins equally spaced in log($\\xi$) from 1 to 5.  The result is a model luminosity for each line, $L^{mod}_i=\\Sigma_{\\xi_k} j_{i,k} EM_k$, where $ j_{i,k}$ is the emissivity for line $i$ in ionization parameter bin $k$ as shown in figure 11 and $EM_k$ is the emission measure distribution. This was fit to the observed luminosities and errors $L^{obs}_i$ and $\\sigma^{obs}_i$, derived from tables 2, 5 and 6 using a distance of 7 kpc by varying the emission measure in each bin independently until a minimum was found in the figure of merit.  We take the figure of merit to be to be $\\Sigma_i ((L^{mod}_i-L^{obs}_i)/\\sigma^{obs}_i)^2$.  This procedure is not expected to be highly accurate, owing to the simplified physical assumptions and the coarseness of the various grids, these results in the upper panel of figure 10 show that we are able to fit most line strengths to within a factor of 2. The lower panel of figure 10 shows the derived best-fit emission measure distribution, which corresponds to 3 distinct components.  This is similar to the result obtained by fitting xstar models to the Chandra and XMM data separately.  This distribution is not consistent with a power law or other smoothly varying function.  An equivalent procedure is to use the normalization calculated using the fit to the xstar model.   The general procedure for doing this is outlined in the xstar manual and in the Appendix to this paper. The result is $\\sim 10^{57}$ cm$^{-3}$, which can be compared with that expected from an ADC.  Estimates for the emission measure of an ADC depend on several  quantities which are not accurately determined, including the geometry of the corona and its effect on the transfer of radiation, the intrinsic luminosity and spectrum of the continuum X-ray source, the possible role of outflows or winds, and the presence of any other heating or ionization mechanism such as magnetohydrodynamic processes (eg. Stone and Miller, 2001). Simple estimates can be obtained by assuming the corona is hydrostatic and unaffected by MHD processes, as was done by Kallman and White (1989). If so, the corona can be crudely represented as a series of concentric cylinders, each having a Gaussian density distribution with height, and where the local scale height is:  $$z_s(R)=\\sqrt{{kT R^3}\\over{G M m_H}}$$  \\noindent where T is the local gas temperature and M is the mass of the central compact object.  The maximum density in the corona is attained when the ionization parameter is $\\Xi=\\Xi_H^*\\simeq$1--10, (eg. Krolik McKee and Tarter, 1981), where $\\Xi=L/(4 \\pi R^2 nckT)$. Then the emission measure is approximately  $$EM(R)\\simeq n_{max}^2 4 \\pi R^2 z_s = 1.2 \\times 10^{61} L_{38}^2 R_{10}^{-1/2} T_7^{-3/2} \\Xi^{-1} {\\rm cm}^{-3}$$  \\noindent where $T_7$ is the temperature in units of 10$^7$K, $R_{10}$ is the radius in units of 10$^{10}$ cm, and $L_{38}$ is the continuum luminosity in units of 10$^{38}$ erg s$^{-1}$.  This shows that the emission measures inferred from the photoionization model fits can be attained by an accretion disk corona for a plausible range of parameters, i.e. $L_{38} \\sim 10^{-2}$, and all other parameters of order unity.  A further constraint on the ADC comes from the derived density in the O VII lines.  If the intrinsic luminosity of the X-ray source is 10$^{38}$ erg s$^{-1}$, then the minimum distance at which material at the inferred O VII maximum density of 10$^{11}$ cm$^{-3}$ and ionization parameter log($\\xi)\\simeq$2 can exist under photoionization is 2.2 $\\times$ 10$^{12}$ cm. This is larger than the likely outer radius of the accretion disk, which is less than $\\sim$10$^{12}$ cm (see below).  This estimate assumes that the local radiation intensity in the ADC is determined by geometrical dilution of the X-rays from the compact object.  The most likely explanation is that the intensity of the ionizing continuum radiation within the O VII region of the ADC is reduced by scattering or absorption. If so, the amount of attenuation required in order to reduce the O VII radius to a value consistent with the disk is a factor $\\sim$100.  This differs from the simple ADC models which have been constructed by eg. Ko and Kallman (1996) and Jimenez-Garate et al. (2001), in which ionization parameters low enough to produce O VII could only occur at densities $\\sim 10^{14}$ cm$^{-3}$ or greater.  An alternative explanation is that the X-ray source is intrinsically faint, $\\sim 10^{36}$ erg s$^{-1}$ or less, or that the O VII emission region lies outside the traditional estimates for the disk outer radius.  \\subsection{Comparison with other LMXBs}  Although 2S0921-630 is not the brightest ADC source, it may the the most suitable target for studying the accretion disk corona during eclipse.  This is based on the ASCA spectrum, and is likely due to its long orbital period, which allows long observations to be carried out entirely during eclipse. An illustration of this is provided by the spectrum of the ADC source 4U1822-371 (Cottam et al., 2001), which shows many of the same features as 0921-63.  However, the line emission from 1822-371 is much weaker in spite of its greater total flux.  In addition, in 4U1822 the emission lines don't come from the corona.  They are phase-dependent and appear to come from the photo-illuminated bulge at the impact point.  This suggests that in 4U1822-371 there may be more dilution of the line emission by directly observed X-rays than is expected, or perhaps the inclination is such that we detect more direct X-rays than in 2S0921-630.  \\subsection{Neutral-Like Iron line}  The iron K line is unique among X-ray lines from abundant elements in that it can be emitted efficiently by neutral or nearly-neutral gas via inner shell fluorescence in addition to being emitted efficiently when iron is highly ionized (here and in what follows we define 'neutral-like' to be ion stages less than Fe XVII, so that the line energy is indistinguishable from neutral).  Although many LMXBs (including 2S0921-630) show emission from highly ionized iron, a uniformly illuminated accretion disk is expected to radiate a strong line from nearly neutral iron.  The strength of this line is a constraint on the strength of illumination of the accretion disk at radii $\\geq 10^8$cm, where the iron in the disk photosphere is expected to neutral-like.  A simple estimate for the flux  of a neutral-like fluorescence line emitted locally from an illuminated disk is:  $$F_{fluorescence}(R)=\\int_{\\varepsilon_{Th}}^{\\infty}{F^{continuum}_{\\varepsilon}(R)d\\varepsilon} (1-e^{-\\tau_{Fe K}}) \\omega$$  \\noindent where $\\varepsilon_{Th}$ is the threshold energy for photoionization, $\\tau_{Fe K}$ is the optical depth at this energy and is assumed to be large, $F^{continuum}_{\\varepsilon}(R)$ is the continuum flux,  $\\omega$ is the fluorescence yield and $y_{Fe}$ is the iron abundance relative to solar (Grevesse et al., 1996).  If we define the incident continuum flux in terms of the luminosity from the neutron star, $L$, the radius $R$, a factor $f\\leq$1 which accounts for non-normal incidence, attenuation between the continuum source and the disk, etc., and a factor $\\kappa$ which is the fraction of the continuum flux absorbed by iron K, we can write:  $$\\kappa f(R) {{L}\\over{4 \\pi R^2}}=\\int_{\\varepsilon_{Th}}^{\\infty}{F^{continuum}_{\\varepsilon}(R)d\\varepsilon} $$  \\noindent Then the total fluorescent line luminosity is obtained by integrating over the disk surface between two radii $R_{min}$ and $R_{max}$:  $$L_{fluorescence}=\\omega y_{Fe} \\kappa {{L}\\over{2}} \\int_{R_{min}}^{R_{max}} {{f(R)}\\over{R}} dR$$  \\noindent If we take $f(R)$=constant, then  $$L_{fluorescence}\\simeq \\omega y_{Fe} \\kappa {{L}\\over{2}} {\\rm ln}({R_{max}}/{R_{min}}) f$$  \\noindent Taking plausible values for these quantities: $\\omega$=0.34 (valid for neutral-like iron), $y_{Fe}=1$, $\\kappa$=0.2 (for an $\\varepsilon^{-1}$ power law) $L=3 \\times 10^{35}$ erg s$^{-1}$ (the measured value for a distance of 7 kpc), ${R_{max}}/{R_{min}}$=100 (an upper limit with little effect on the result), $f$=0.1 (an upper limit; cf. Ko and Kallman 1994), we get $L_{fluorescence}\\simeq 5 \\times 10^{33}$ erg s$^{-1}$. This is greater than the upper limit allowed by the observations for the 6.4 keV line, $1.8 \\times 10^{33}$ erg s$^{-1}$, suggesting that the illumination is weaker than we have assumed, $f\\leq 0.03$, or that the iron abundance is less than solar."
    },
    "0209/astro-ph0209226_arXiv.txt": {
        "abstract": "We present the X-ray analysis and the mass estimation of the lensing cluster of galaxies CL0024+17 with {\\it Chandra}. We found that the temperature profile is consistent with being isothermal and the average X-ray temperature is $4.47^{+0.83}_{-0.54}$ keV.  The X-ray surface brightness profile is represented by the sum of emissions associated with the central three bright elliptical galaxies and the emission from intracluster medium (ICM) which can be well described by a spherical $\\beta$-model. Assuming the ICM to be in the hydrostatic equilibrium, we estimated the X-ray mass and found it is significantly smaller than the strong lensing mass by a factor of 3. ",
        "introduction": "CL0024+17 is one of the most extensively studied lensing clusters of galaxies, located at $z=0.395$. Since the discovery of the multiply lensed arc system, several authors modeled the matter distributions in the cluster. Tyson, Kochanski, \\& Dell'Antonio (1998) constructed a very detailed mass map and suggested that the dark matter profile has a soft core. Broadhurst et al. (2000) measured the arc redshift to be 1.675 and also built a lens model in a simplified manner.  On the other hand, the X-ray emitting gas is an excellent tracer of the dark matter potential. Soucail et al. (2000) performed a combined analysis of the {\\it ROSAT} and {\\it ASCA} data and estimated the cluster mass within the arc radius (the X-ray mass, hereafter). They found that there is about a factor of $\\sim3$ discrepancy between the X-ray mass and the strong lensing mass (Tyson et al. 1998; Broadhurst et al. 2000). Because the {\\it ROSAT} HRI image suggested the elongated gas distribution, they considered that the discrepancy may be caused by the irregular mass distribution.  However, there were still large measurement uncertainties in both the X-ray temperature and the image morphology. It was mainly because of the heavy contamination from the bright seyfert galaxy. Thus for the cluster mass estimation the temperature determination is crucial. In this paper, we report on the accurate measurements of the temperature and the morphology with {\\it Chandra}, from which we discuss whether there is an inevitable mass discrepancy between the X-ray and the strong lensing. We use $H_0=50$ km/s/Mpc and $\\Omega_0=1$. $1\\arcmin = 383\\,h_{50}^{-1}$ kpc at $z=0.395$. ",
        "conclusions": ""
    },
    "0209/nucl-th0209074_arXiv.txt": {
        "abstract": "The influence of the in-medium mesons on the effective nucleon mass and in turn on the equation of state of hot/dense nuclear matter is discussed in the Walecka model. Due to the self-consistent treatment of couplings between nucleons and $\\sigma $ and $\\omega$ mesons, the temperature and density dependence of the effective hadron masses approaches more towards the Brown-Rho scaling law, and the compression modulus $K$ is reduced from $550$ MeV in mean field theory to an accepted value $318.2$~MeV. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/cond-mat0209482_arXiv.txt": {
        "abstract": "The most complicated phenomena of equilibrium statistics, phase separations and transitions of various order and critical phenomena, can clearly and sharply be seen even for small systems in the topology of the curvature of the microcanonical entropy $S_{B}(E,N)=\\ln[W(E,N)]$ (Boltzmann's principle $BP$) as function of the conserved energy, particle number etc.. Consequently, $BP$ allows to establish the link toward their microscopic origin and the study of the way how interacting many-body systems organize into phase-transitions. Also the equilibrium of the largest possible interacting many-body systems like self-gravitating systems is described to great extend by the topology of the entropy surface $S_{B}(E,N,\\vecbm{$L$})$ where \\vecbm{$L$} is the angular momentum. Conventional (canonical) statistical mechanics describes only a small section out of all equilibrium phenomena in nature and only in cases where the so-called ``thermodynamic limit'' applies (homogeneous phases of ``infinite'' systems interacting with short-range interactions). In this paper I present two examples of phase transitions of first order which are of fundamental importance: the liquid to gas transition in a small atomic cluster and the condensation of a rotating self-gravitating system into single stars or into multi-star systems like double stars an rings. Such systems cannot be addressed by ordinary canonical thermo-statistics.  I also give a geometric illustration how an initially non-equilibrized ensemble approaches the microcanonical equilibrium distribution. ",
        "introduction": "Many theorems we are used to in conventional macroscopic (canonical) thermo-statistics are wrong when statistical mechanics addresses small or other non-extensive systems. Here a revision of the fundamentals is demanded.  Phase separation of normal systems and also in general the equilibrium of closed non-extensive systems are not described by the canonical and grand-canonical ensembles. Only the microcanonical ensemble based on Boltzmann's entropy $S_{B}=\\ln{W}$, with $W(E)=\\epsilon_0 tr[\\delta(E-H)]$, the number of classical or quantum states, describes correctly the unbiased uniform filling of the energy-shell in phase-space.  Various ensembles like the (grand-)canonical are equivalent to the microcanonical ensemble only if the system is infinite and homogeneous, i.e. in a pure and homogeneous phase. Only then exists a one to one mapping from the conserved mechanical observables as energy $E$, particle number $N$, eventually angular-momentum \\vecbm{$L$} and others to the intensive variables temperature $T$, chemical potential $\\mu$ and eventually rotational frequency $\\omega$ and so on. Otherwise the (grand-)canonical ensembles do not reflect the equilibrized phase-space distribution of a closed ergodic Hamiltonian system~\\cite{gross124,gross140,gross158,gross174} see also Barr\\'{e} et al~\\cite{barre01}. (Grand-)canonical potentials are non-analytic at phase transitions whereas $S_B(E,N,\\cdots)$ remains multiply differentiable also there.  This program is far from only academic interest and is deeply demanded in many fields of condensed matter. A pseudo Riemannian geometry must replace Ruppeiner's Riemannian geometry of fluctuations~\\cite{ruppeiner95,andresen96}. It leads to negative heat capacities at phase-separation which cannot be explained in any canonical formalism and which are well documented experimentally c.f.~\\cite{thirring70,chbihi95,lyndenbell95a,lyndenbell99,gross171,gross172,schmidt00,dAgostino00,schmidt01} as also postulated theoretically since long cf.~\\cite{gross95,gross158,gross150,gulminelli99a,casetti99a,ispolatov01,gross181,gross187,gross190,gross191}.  Besides conventional extensive systems in the thermodynamic limit, this formulation of thermo-statistics addresses additionally the following objects: \\begin{itemize} \\item Astro-physical systems with their typical negative heat capacity are clearly outside of any canonical approach~\\cite{thirring70,lyndenbell99,casetti99a,ispolatov01,gross187,gross190,gross191}. \\item The same is true for the original goal of Thermodynamics, the description of phase separation~\\cite{gross174,gross189,gross176,gross182,gross177,gross138}, \\item and of course small systems like excited nuclei, atomic clusters etc., which are addressed more recently, where many experimental results are now cumulating~\\cite{chbihi95,lyndenbell95a,gross171,gross172,schmidt00,dAgostino00,schmidt01}, and which are exotic from the canonical point of view c.f.section(\\ref{surfS}). \\end{itemize} ",
        "conclusions": "The {\\em geometric interpretation of classical equilibrium Statistical Mechanics}~\\cite{gross186} by Boltzmann's principle (\\ref{boltzmprinciple}) offers an extension also to the equilibrium of non-extensive systems. In more fundamental, axiomatic terms, it opens the application of Thermo-Statistics to ``non-simple'' systems which are not (homogeneus) fluids or in contact with ideal gases. Surprisingly, but also understandably, this is still an open problem c.f. ref.~\\cite{uffink01} page 50 and page 72.  Because microcanonical Thermodynamics as a macroscopic theory controls the system by a few, usually conserved, macroscopic parameters like energy, particle number, etc. without fixing all $6N$ degrees of freedom, it is an intrinsically probabilistic theory. It describes all systems with the same control-parameters simultaneously. If we take this seriously and avoid the so called thermodynamic limit ($\\lim_{V\\to\\infty, N/V=\\rho}$), the theory can be applied to small systems but even to the really large, usually {\\em inhomogeneous}, self-gravitating systems, c.f.\\cite{gross187,gross190,gross191}.  Within the new, extended, formalism several principles of traditional Statistical Mechanics turn out to be violated and obsolete. E.g. at phase-separation (at negative heat capacity) heat (energy) can flow from cold to hot~\\cite{gross189}.  Or phase-transitions can be classified unambiguously in astonishingly small systems. These are by no way exotic and wrong conclusions. On the contrary, many experiments have shown their validity.  I believe this approach gives a much deeper insight into the way how many-body systems organize themselves than any canonical statistics is able to.  The thermodynamic limit clouds the most interesting region of Thermodynamics, the region of inhomogeneous phase-separation.  Because of the only {\\em one} underlying axiom, Boltzmann's principle eq.(\\ref{boltzmprinciple}), the {\\em geometric interpretation}~\\cite{gross186} keeps statistics most close to Mechanics and, therefore, is more transparent. The Second Law ($\\Delta S\\ge 0$) can even be shown to be valid in {\\em closed, small} systems under quite general dynamical conditions~\\cite{gross192}."
    },
    "0209/astro-ph0209540_arXiv.txt": {
        "abstract": "Sensitive high angular and linear resolution radio images of the 240-pc radio jet in NGC~4151, imaged at linear resolutions of 0.3 to 2.6 pc using the VLBA and phased VLA at $\\lambda$21 cm, are presented and reveal for the first time a faint, highly collimated jet (diameter $\\lesssim$1.4 pc) underlying discrete components, seen in lower resolution MERLIN and VLA images, that appear to be shock-like features associated with changes in direction as the jet interacts with small gas clouds within the central $\\sim$100 pc of the galaxy. In addition, $\\lambda$21-cm spectral line imaging of the neutral hydrogen in the nuclear region reveals the spatial location, distribution and kinematics of the neutral gas detected previously in a lower resolution MERLIN study.  Neutral hydrogen absorption is detected against component C4W (E+F) as predicted by Mundell et al, but the absorption, extending over 3 pc, is spatially and kinematically complex on sub-parsec scales, suggesting the presence of small, dense gas clouds with a wide range of velocities and column densities. The main absorption component matches that detected in the MERLIN study, close to the systemic velocity (998 \\kms) of the galaxy, and is consistent with absorption through a clumpy neutral gas layer in the putative obscuring torus, with higher velocity blue- and red-shifted systems with narrow linewidths also detected across E+F.  In this region, average column densities are high, lying in the range 2.7~$\\times$~10$^{19}$~$T_{\\rm S}$~$<$~$N_{\\rm H}$~$<$~1.7~$\\times$~10$^{20}$~$T_{\\rm S}$~\\cmsq\\ K$^{-1}$ ($T_{\\rm S}$ is the spin temperature), with average radial velocities in the range 920~$<$~V$_{\\rm r}$~$<$~1050~\\kms.  The spatial location and distribution of the absorbing gas across component E+F rules out component E as the location of the AGN (as suggested by Ulvestad et al.) and, in combination with the well-collimated continuum structures seen in component D, suggests that component D (possibly subcomponent D3) is the most likely location for the AGN. We suggest that components C and E are shocks produced in the jet as the plasma encounters, and is deviated by, dense clouds with diameters smaller than  $\\sim$1.4 pc.  Comparison of the radio jet structure and the distribution and kinematics of ionized gas in the narrow line region (NLR) suggests that shock-excitation by passage of the radio jet is not the dominant excitation mechanism for the NLR. We therefore favour nuclear photoionization to explain the structure of the NLR, although it is interesting to note that a small number of clouds with low velocity and high velocity dispersion are seen to bound the jet, particularly at positions of jet direction changes, suggesting that some NLR clouds are responsible for bending the jet. Alternatively, compression by a cocoon around the radio jet due to pressure stratification in the jet bow shock could explain the bright, compressed optical line-emitting clouds surrounding the cloud-free channel of the radio jet, as modelled by Steffen et al. ",
        "introduction": "Nuclear activity in galaxies is present over a wide range of luminosities, from the most distant and powerful quasars, to the weaker Active Galactic Nuclei (AGN) seen in nearby galaxies such as Seyferts, LINERs and even the Milky Way (e.g., Huchra \\& Burg 1992; Ho Filippenko \\& Sargent 1997). The standard model of nuclear activity involves the release of gravitational potential energy from galactic material accreted by a supermassive black hole at the galaxy center and, although radiation from the central engine is detected across the electromagnetic spectrum, it was the radio emission that first led to the discovery of powerful AGN (e.g., Baum \\& Minkowski 1960; Hazard, Mackey \\& Shimmins 1963; Bridle \\& Perley 1984).  Indeed, the presence of powerful, highly-collimated relativistic radio jets in radio-loud quasars and radio galaxies, extending well beyond their host galaxy (e.g., Fanaroff \\& Riley 1974; Bridle \\& Perley 1984), provided key evidence for the exhaust material from a black-hole driven central engine (Scheuer 1974, Blandford and Rees 1974; Bridle et al. 1994; Urry \\& Padovani 1995).  In contrast, radio-quiet quasars and Seyfert galaxies are ten times more common but 100 to 1000 times weaker at radio wavelengths than their radio-loud cousins (e.g., Goldschmidt et al. 1999); consequently nuclear starbursts have been advocated as the primary power source instead of black-hole accretion (e.g., Fernandes \\& Terlevich 1995). However, instrumental improvements and high resolution radio imaging with the VLA, MERLIN and the VLBA over the last twenty years have shown increasing evidence for collimated radio emission in the form of small-scale radio jets that are weak analogues to jets in radio-loud AGN (e.g. Ulvestad \\& Wilson 1989; Wilson 1991; Nagar et al. 1999; Kukula et al.  1999) and indicate the presence of a central black hole and accretion disk, at least in some Seyferts. Similarly, measurements of high brightness temperatures in milliarcsecond resolution images of Seyfert nuclei with flat radio spectra have also suggested support for accretion-powered central engines (Mundell et al. 2000).  Radio jets, although interesting in their own right, also provide a valuable probe of the interstellar medium close to the central engine, in particular the putative obscuring torus, which is thought to surround the central engine, providing fuel for the AGN and determining the observed optical spectral differences between type 1 (unobscured, viewed pole-on down the torus axis ) and obscured type 2 (viewed edge-on in the torus equatorial plane) (e.g., Antonucci \\& Miller 1985).  Some of the molecular gas in the torus is expected to be dissociated and ionized by the central UV/X-ray continuum source resulting in detectable quantities of neutral and ionized gas (e.g., Krolik \\& Lepp 1989; Pier \\& Voit 1995).  Absorption of the radio jet continuum emission by intervening torus gas can be measured to determine ionized gas densities via free-free absorption (e.g., Ulvestad, Wrobel \\& Carilli 1999; Ulvestad et al. 1999), neutral gas columns via $\\lambda$21-cm neutral hydrogen (\\HI) absorption (e.g., Mundell et al. 1995; Gallimore et al. 1999; Peck \\& Taylor 2002) and molecular gas content via OH absorption (e.g., Hagiwara et al. 2000). Due to the small size scales of the radio jet and the obscuring torus, high angular resolution imaging using the VLBA is required to spatially resolve gas in the absorbing region.  In this context, NGC~4151, with its large quantities of neutral gas and elongated radio continuum structure, is an ideal candidate for such a study (see Ulrich 2000 for a comprehensive review of the properties of the galaxy and its AGN).  NGC~4151 is a Seyfert type 1.5 (e.g., Osterbrock \\& Koski 1976) in an almost face-on ($ i$=21\\arcdeg), grand-design, weakly-barred spiral host galaxy, Hubble type (R')SAB(rs)ab (de Vaucouleurs et al. 1991), that contains significant amounts of \\HI\\ throughout its two optically-faint spiral arms, fat weak bar and nuclear region (Davies 1973; Bosma, Ekers \\& Lequeux 1977; Pedlar et al. 1992; Mundell et al. 1995; Mundell \\& Shone 1999; Mundell et al. 1999).  Radio continuum observations of the strong radio continuum nucleus show a linear radio structure in the form of a string of knots, elongated over $\\sim$3\\farcs5 (230 pc) in average position angle (P.A.)  $\\sim$77\\arcdeg\\ (Wilson \\& Ulvestad 1982; Johnston et al. 1982; Carral, Turner \\& Ho 1990; Pedlar et al. 1993; Mundell et al. 1995), which is embedded in diffuse emission extending over $\\sim$10\\farcs5 ($\\sim$680 pc) (Johnston et al. 1982; Pedlar et al. 1993). Carral et al. (1990) identified five main knots in the 3.5-arcsec jet at 15 GHz, annotating them C1 to C5.  $\\lambda$21-cm MERLIN observations of this central 4\\arcsec\\ region (Mundell et al. 1995) revealed localized and marginally-resolved \\HI\\ absorption, with a peak column density of $N_{\\rm H}$~$\\sim$6$\\times$10$^{19}$~$T_{\\rm S}$ cm$^{-2}$, against the component (C4) in the radio jet which is thought to contain the AGN; no absorption was detected against the other jet components to a limiting column density of $N_{\\rm H}$~$\\sim$2$\\times$10$^{19}$~$T_{\\rm S}$ cm$^{-2}$.  An east-west column density gradient was observed and, in combination with UV column densities and early VLBI images, which showed C4 to consist of two components (C4E and C4W) separated by $\\sim$7~pc, led Mundell et al. (1995) to suggest that the weaker, western component, C4W, contains the optical/UV nucleus and thus the \\HI\\ absorption is taking place against the first component of the counterjet (C4E), due to gas in the obscuring torus. Structural and spectral index information obtained from subsequent radio continuum VLBA observations of the jet at 1.6 and 5~GHz led Ulvestad et al. (1998) to suggest a similar model but with the AGN located instead at the emission peak of C4E (component E after Ulvestad et al. 1998); requiring an absence of \\HI\\ absorption against the AGN, this model predicted \\HI\\ absorption only against the start of the counterjet, i.e.  the tail of emission extending eastwards of the peak in C4E (component F after Ulvestad et al. 1998).  In this paper we re-open the discussion on the possible location of the AGN based on new spectral line observations, but whatever the precise location of AGN, NGC~4151 appears to be unusual in having neutral gas located close to the nucleus, unlike other Seyferts that show \\HI\\ absorption associated with gas located in dust disks on scales of 100$-$200~pc aligned with the host galaxy disk (e.g., Cole et al. 1998; Gallimore et al. 1999); in this respect NGC~4151 seems more similar to Compact Symmetric Objects (CSO) and Steep Spectrum Core (SSC) objects (e.g., Conway 1996; Peck 1999; Peck \\& Taylor 2001).  We present sensitive new, high resolution $\\lambda$21-cm continuum imaging, using the VLBA and phased VLA, of the entire 3\\farcs5 radio jet in NGC~4151 at an angular resolution of 40~mas (2.6 pc), along with higher resolution images of individual components with angular resolutions down to 5~mas (0.3 pc).  In addition we present analysis of the first spatially-resolved images of the associated \\HI\\ absorbing gas with an angular resolution of 10.5 mas (0.7 pc) and confirm that the neutral gas, although clumpy, is located across the whole of component C4W (E+F after Ulvestad et al. 1998).  Section \\ref{observations} details the observational parameters and data reduction performed; in Section \\ref{results} we present the results of both the sensitive, high resolution continuum imaging of the jet and the \\HI\\ distribution and kinematics. In Section \\ref{discussion} we present the arguments for the location of the AGN being associated with component D and discuss the relationship between the radio jet and the narrow line region.  We discuss the \\HI\\ absorption in comparison with Ly$\\alpha$ absorption measurements and present our conclusions in Section \\ref{conclusions}.  As discussed in Mundell et al. (1999) the heliocentric radial velocity of NGC~4151 is $\\sim$1000 \\kms, but distance estimates vary depending on the assumed value of H$_{\\rm 0}$ and whether the Virgocentric correction and the relative velocity of the Local Group with respect to the Virgo cluster are taken into account.  Distance estimates lie in the range 10$<$D$<$30.5~Mpc for 50$<$H$_{\\rm 0}$$<$100~\\kms\\ kpc$^{-1}$ and 1000$<$V$<$1523 \\kms, but since the uncertainty in the value of H$_{\\rm 0}$ is as large as the Virgocentric correction, for the purposes of this paper we assume a distance to NGC~4151 of 13.3~Mpc, using H$_{\\rm 0}$=75 \\kms\\ kpc$^{-1}$ and the heliocentric velocity of 998\\kms, giving a linear scale of 0.065~pc~mas$^{-1}$ in the galaxy. ",
        "conclusions": "\\label{conclusions}  We have used the VLBA and phased VLA at $\\lambda$21 cm to study the parsec-scale radio continuum emission and distribution and kinematics of neutral hydrogen in absorption in the Seyfert galaxy NGC~4151.  We find:  \\begin{itemize} \\item A faint, highly collimated radio jet, diameter $\\lesssim$1.4 pc, underlying brighter radio knots, seen in lower resolution MERLIN and VLA images, that appear to be shock-like features associated with changes in direction as the jet interacts with small gas clouds within the central $\\sim$100 pc of the galaxy. \\item H {\\sc i} absorption across the whole of component C4W (E+F), as predicted by Mundell et al. (1995),  with complex spatial and kinematic structure. \\item The location and distribution of the HI absorption and the structure of the radio continuum emission argue for the AGN being located in component D, rather than component E as argued by Ulvestad et al. (1998). \\item We suggest the absorbing gas lies in a thin photo-dissociated layer of clumpy neutral gas in between the molecular and ionized gas in the circumnuclear torus, with an inner radius $\\sim$3.3 (sin$^{-1}$$\\theta$) pc. \\item The location of the absorbing gas close to the AGN is unusual for a Seyfert galaxy and we argue that the \\HI\\ and continuum properties of NGC~4151 are similar to those seen in compact symmetric objects and compact steep spectrum objects. \\item Spatial association of the radio jet knots and optical line-emitting clouds with high velocity dispersion in the narrow line region suggest some interaction between the radio jet and the clumpy interstellar medium, but photoionization remains the dominant excitation mechanism for the NLR.  \\end{itemize}"
    },
    "0209/astro-ph0209589_arXiv.txt": {
        "abstract": "The EXPLORE Project is a series of searches for transiting extrasolar planets using large-format mosaic CCD cameras on 4-m class telescopes. Radial velocity follow-up is done on transiting planet candidates with 8--10m class telescopes.  We present a summary of transit candidates from the EXPLORE Project for which we have radial velocity data. ",
        "introduction": "The EXPLORE (EXtrasolar PLanet Occultation REsearch) Project is a series of searches for transiting extrasolar planets orbiting Galactic plane stars using 4-m class telescopes.  As an integral part of the search strategy, radial velocity (RV) follow-up observations for mass confirmation are done on 8--10m class telescopes.  In June 2001, we used the CTIO 4-m telescope for 11 nights (6 clear) to observe the EXPLORE~I field, located at $l=-28, b=-3$ (Mall\\'en-Ornelas et al.\\ 2002). The best 37,000 light curves were examined, and RV follow-up of three planet candidates was done on the VLT in September 2001 (Mall\\'en-Ornelas et al., in prep.).  In December 2001 we used the 3.6-m CFHT for 16 nights (14 clear) to observe the EXPLORE~II field, located at $l=203, b=0.7$ (Yee et al., in prep.).  The best 12,000 light curves were examined, and RV follow-up of 2 planet candidates and 2 additional eclipsing systems was done at Keck in February 2002 (Mall\\'en-Ornelas et al., in prep.).  Here we present a summary of the transiting planet candidates for which we have RV follow-up. ",
        "conclusions": "We have presented a status report on the transit candidates from the EXPLORE Project's first two surveys.  Out of three systems with RV follow-up in the EXPLORE~I search, one was found to be a close stellar binary plus a blended star, a second one is an unlikely planet candidate that nonetheless has not been entirely ruled out as a planet, and a third one is a promising planet candidate based on an upper mass limit.  The two candidates with RV follow-up in the EXPLORE~II search remain good planet candidates, although more RV data are required for both systems, and more photometric data are also desirable for EXP2c10s5069, which has extremely shallow eclipses detected with low S/N.  We have shown that it is possible to produce a clean set of planet candidates from a deep transit search and place meaningful mass limits (2--3 $M_J$) on the companions via follow-up RV observations.  We emphasize that in order to find one good planet candidate, it is necessary to monitor several thousands of stars with good time coverage, high photometric precision, and time sampling which is good enough to determine the eclipse shape."
    },
    "0209/astro-ph0209406_arXiv.txt": {
        "abstract": "{The distribution of gravitational torques and bar strengths in the local Universe is derived from a detailed study of 163 galaxies observed in the near-infrared. The results are compared with numerical models for spiral galaxy evolution. It is found that the observed distribution of torques can be accounted for only with external accretion of gas onto spiral disks. Accretion is responsible for bar renewal -- after the dissolution of primordial bars -- as well as the maintenance of spiral structures. Models of isolated, non-accreting galaxies are ruled out. Moderate accretion rates do not explain the observational results: it is shown that galactic disks should double their mass in less than the Hubble time. The best fit is obtained if spiral galaxies are open systems, still forming today by continuous gas accretion, doubling their mass every 10 billion years. ",
        "introduction": "Bars are a major perturbation of the gravitational potential and a highly efficient engine for the evolution of morphological and chemical properties. In purely stellar disks, bars are robust, long-lived structures (Combes \\& Sanders 1981). In reality, however, spiral galaxies contain gas which provokes the dissolution of bars (Bournaud \\& Combes 2002, hereafter BC). The bar itself initiates an important radial gas inflow, which destroys barred structure (Pfenniger \\& Norman 1990).  Since bars in gaseous disks are dissolved in a few Gyrs, the ubiquity of bars in the local Universe suggests that bars are also renewed. BC painted a scenario wherein continual gas accretion might be a crucial evolutionary factor. This gas settles in the disk, enhancing its self gravity, reducing the influence of the stabilizing central mass concentration, and a second bar develops. In this Letter, we test this hypothesis, by comparing the observed distribution of gravitational torques (in a well-defined galaxy sample) with the distribution predicted by numerical simulations incorporating gas accretion. ",
        "conclusions": "The issue of whether observations support a galaxy dissolving and reforming its bar has awaited the completion of near-infrared surveys.  We are now able to show how robust this scenario is. Isolated galaxies (non-accreting systems) cannot reproduce the observed properties at all: they would become unbarred and spiral arms would disappear; many disks would then be nearly axisymmetric after a few Gyrs. On the contrary, spiral galaxies appear to be open systems that are still forming and continuously accreting mass today. We expect a doubling in disk mass every 10 billion years. The origin of the accreted gas has not been considered, but the most likely source could be the reservoirs of gas observed outside nearly all spiral disks (Sancisi 1983). Pfenniger et al. (1994) have even postulated that the dark matter around spiral galaxies might be in the form of cold gas. Accretion rates from infalling dwarf satellites only represent a few percent of the accretion rate that we derive (Toth \\& Ostriker 1992), so that other sources of accretion must be invoked."
    },
    "0209/astro-ph0209630_arXiv.txt": {
        "abstract": "The Atacama Large Millimeter Array, or ALMA,   is an international telescope project which will be built over the coming decade in Northern Chile. With over 7000 m$^2$ of collecting area comprised of 64 12m antennas arrayed over baselines up to 14 km in extent, ALMA will provide images of unprecedented clarity and detail.  One revolutionary feature of ALMA will be its ability to combine interferometric and single telescope data, providing complete flux recovery.  ALMA will cover a spectral wavelength range from 7mm to 0.3 mm or shorter wavelengths, providing astronomy with its first detailed look at the structures which emit millimeter and submillimeter photons, the most abundant photons in the Universe. ",
        "introduction": "\\label{sect:intro}  % In this paper we review  ALMA goals, the chosen site at Chajnantor, the scope of the project, and a few of the technical hurdles and proposed solutions, as well as the ALMA schedule.  \\subsection{Goals: Science Aims and Drivers} \\label{sect:goals}  Several publications$^{1,2}$ contain contributions describing the science goals of ALMA.  In its basic formulation, the ALMA goal is to provide images of unprecedented clarity and detail in the millimeter and submillimeter spectral range.  This range contains two of the three primary peaks in the electromagnetic spectrum of the Universe; these are the two containing the preponderance of the observed energy in the Universe. The largest of these is the peak from the 3 K blackbody radiation relic of the Big Bang.  That peak occurs in the millimeter range of the spectrum, as expected for any black body radiating at such a low temperature.   The second strongest occurs at about 1.5 THz or 200 microns wavelength.  Light of these wavelengths cannot penetrate the atmosphere, as it is absorbed by water and other molecules--this maximum was identified only recently through satellite observations.  Alas, with a satellite one is limited as to the size of telescope with which one can observe and anything we can currently put into space is far too small to give good images of the energy sources comprising this second peak. From its characteristic blackbody temperature of $\\sim$20 K we are led to suspect that it is comprised of emission from the cold molecular clouds from which stars and planets in the Universe have formed, and from the young galaxies full of dust which host those molecular clouds.  Recently, some telescopes have made progress in identifying the sources of this unknown radiation.  Much of it--perhaps most of it--appears to come from tremendous episodes of star formation in galaxies at the earliest stage of their creation. Unfortunately, many sources of submillimeter radiation have not been identified optically, at least in part because their spectrum has been wholly redshifted to wavelengths blocked by the terrestrial atmosphere and in part because of tremendous amounts of dust endemic to the source, which absorb optical light and re-emit it at longer wavelengths. The only way to tell for sure is to get precise images of these sources. An instrument to provide these images must provide high resolution--matching that which will be available at other wavelengths, or from 0.01'' to 0.1'' and it must provide the sensitivity to invest these images with high dynamic range.  ALMA has been defined to achieve these goals.  It is currently under construction by an international partnership.  With construction funding begun in FY2002, ALMA will be built over the coming decade in Northern Chile. ALMA will be a revolutionary telescope, operating over the entirety of the millimeter and submillimeter wavelength band observable from the Earth's most lofty regions.  \\begin{figure} \\begin{center} \\begin{tabular}{c} \\includegraphics[height=10cm]{woottena_1.ps} \\end{tabular} \\end{center} \\caption[ Artist's rendition of ALMA in its compact configuration.  Image courtesy European Southern Observatory.] { \\label{fig:ALMA}  Artist's rendition of ALMA in its compact configuration.  Image courtesy European Southern Observatory.} \\end{figure} Table 1 summarizes the specifications for ALMA. \\begin{table}[h] \\caption{Summary of ALMA Specifications} \\label{tab:spectable} \\begin{center} \\begin{tabular}{|l|r|} % \\hline \\rule[-1ex]{0pt}{3.5ex}  Parameter & Specification \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Number of Antennas & 64  \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Antenna Diameter & 12m \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Antenna Surface Precision & $<$ 25 $\\mu$m rss \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Antenna Pointing Accuracy & $<$ 0.\"6 rss \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Total Collecting Area & $>$7000 m$^2$ \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Angular Resolution & 0\".02 $\\lambda$ (mm) \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Configuration Extent & 150 m to 14 km\\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Correlator Bandwidth & 16 GHz per baseline\\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Spectral Channels & 4096 per window\\\\ \\hline \\rule[-1ex]{0pt}{3.5ex} Number of Spectral Windows & 8\\\\ \\hline \\end{tabular} \\end{center} \\end{table} The ALMA specifications are described in more detail in the ALMA Construction Project Book, which like other project details may be found on the Worldwide Web at www.alma.nrao.edu. To fully enable the order of magnitude gain in resolution and two order of magnitude gain in sensitivity over existing instruments which are ALMA's goal, it will be located at an elevation of 5000m on the Chajnantor Alitplano near San Pedro de Atacama in northern Chile, as shown in Figure 1 and described below. Table 2 summarizes the sensitivity of ALMA in various modes and at various frequencies, given the atmosphere at the Chajnantor site.  The calculations assume 1mm of preciptable water vapor above the site, observation at 1.3 air masses (40$^o$ zenith angle) and antenna performance as given in Table 1.  \\begin{table}[h] \\caption{Summary of ALMA Sensitivities} \\label{tab:senstable} \\begin{center} \\begin{tabular}{|l|c|c|r|} % \\hline \\rule[-1ex]{0pt}{3.5ex} Band no. & Frequency & Continuum & Line (1 km s$^{-1}$) \\\\ & Range (GHz)$^a$  & (mJy; 60s) &(mJy; 60s) \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}1 & 31.3--45 & 0.02 & 5.1 \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}2 & 67--90 & 0.028&4.9  \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}3$^b$ & 84--116 & 0.027 & 4.4 \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}4 &125--163 & 0.039 & 5.1 \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}5 &163--211 & 0.52 &5.9  \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}6$^b$ &211--275 & 0.071 & 7.2 \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}7$^b$ & 275--370 & 0.120 & 10 \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}8 & 385--500 & 0.34 &26  \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}9$^b$ & 602--720 & 0.849 & 51 \\\\ \\hline \\rule[-1ex]{0pt}{3.5ex}10 & 787--950 & 1.26 & 66 \\\\ \\hline \\end{tabular} \\end{center} \\vspace{0.3cm} $^a$ Frequency ranges as given in {\\it Specifications for the ALMA Front End Assembly}, AEC Version 1, 18 October, 2000.\\\\ $^b$ These  4 bands are the highest priority bands, the others are the second priority bands, as judged by the ALMA Scientific Advisory Committee.\\\\ \\end{table} \\subsection{ALMA Partners} \\label{sect:partners} ALMA has gained widespread support as an instrument astronomy needs to develop astronomical ideas through observation in the twenty-first century. The project, previously known at the Millimeter Array (MMA), has been endorsed by both the 1991 and 2000 Astronomy and Astrophysics Survey Committees of the National Academy of Sciences (US) as among the highest priority items for astronomical facilities to be constructed.  In France, at a colloquium at Arcachon in March 1998 organized by CNRS/INSU, French astronomers put construction of ALMA, in its interim form as the Large Southern Array (LSA), as highest priority for future instruments. In Canada, ALMA was identified as top priority by the National Research Council's Long Range Planning Panel on Canada's Future in Astronomy until 2015. Dutch astronomers likewise established ALMA as the top priority for instrumentation in the coming decade. In the U.K., the Astronomy Vision Panel of PPARC identified ALMA as the highest priority medium-term project.  This large set of astronomical community endorsements has gained ALMA a number of international partners. North American ALMA partner institutions include the United States National Science Foundation, through its National Radio Astronomy Observatory facility operated by Associated Universities, Inc., and the Canadian National Research Council.  NRAO works in cooperation with a University consortium including the Owens Valley Radio Observatory of Caltech and the Berkeley Illinois Maryland Association.  European partners include the European Southern Observatory; The Centre National de la Recherche Scientifique (CNRS), France; The Max Planck Gesellschaft (MPG), Germany; The Netherlands Foundation for Research in Astronomy, (NFRA); Nederlandse Onderzoekschool Voor Astronomie, (NOVA); The United Kingdom Particle Physics and Astronomy Research Council, (PPARC); The Swedish Natural Science Research Council, (NFR); and the Oficina de Ciencia y Tecnologia and Instituto Geografico Nacional (IGN), Spain. Chile, as host nation for the ALMA project, participates through its presence on the ALMA Coordinating Committee, the ALMA Science Advisory Committee and by providing the excellent site high in the Andean Altiplano and support for it. The National Astronomical Observatory of Japan may join the ALMA consortium soon.  \\subsection{Location: the Site at Chajnantor, Chile} \\label{sect:site} In May 1998 NRAO recommended construction of the MMA on a site in Region II of northern Chile which lies on a high plain at the foot of three ancient volcanic peaks, Cerro Toco, Cerro Chajnantor and Cerro Chascon.  The site (longitude 67d 45m W, latitude -23$^o$ 01' S)lies near the Tropic of Capricorn,  about 50 km east of the historic village of San Pedro de Atacama, 130 km southeast of the mining town of Calama, and about 275 km ENE of the coastal port of Antofagasta. It lies close to the border with Argentina and Bolivia beside the paved Paso de Jama road into Argentina and gas pipelines connecting Argentina sources with Chilean mines.  The mean elevation is about 5000m (16400 ft). Several sites had been tested but as scientific interest increased for the highest frequencies, sites at the highest altitudes became favored.  Testing of the Chajnantor site began in April 1995 and continues to the present, a joint effort of NRAO and the European Southern Observatory (ESO).  The Nobeyama Radio Observatory (NRO) operates a similar testing facility nearby at Pampa la Bola. The testing operations continue with the involvement of the Chilean university community.  The land is administered by the Chilean Ministry of National Assets, having been set aside as a protected region for science by Presidential decree.  Up-to-date details of the monitoring of a number of parameters critical to ALMA's success continue and current details from the instrumentation may be found at the ALMA web site.  Some salient characteristics include: the median annual temperature is -2.5$^o$C with annual 50th percentile winds of 10.4 m s$^{-1}$.  The average barometric pressure is only 55 percent of the value at sea level.  Humidity averages 39\\% and ultraviolet radiation is about 170\\% that at sea level.  Transparency at 225 GHz has been monitored for several years; the 50th percentile zenith optical depth at this frequency is 0.061 corresponding to a column of precipitable water of a little more than 1mm$^3$.  With such a low water column, observations are possible at the atmospheric windows covered by ALMA Bands 9 and 10 (see Table 2) for roughly half of the time.  Direct observations of atmospheric transparency including the supraterahertz windows at 1.035, 1.3 and 1.5 THz have been published$^4$ showing transmission of up to 20\\% in the highest frequency window.   \\subsection{Science} \\label{sect:science}  The mission of ALMA is to produce detailed spectral line and continuum images of objects emitting radiation in the millimeter and submillimeter spectrum accessible from the surface of the Earth.  These high dynamic range images, covering a wealth of spatial scales and featuring total flux recovery, will allow ALMA scientists to study the formation of galaxies, stars and planets and the distribution of the chemical precursors necessary for life.  Specifically, ALMA will allow astronomers to address a number of topics of high interest.  As mentioned above, the photons comprising the visible/infrared and submillimeter portions of the spectrum arise from two distinct features in the overall spectrum of background radiation.  Images of the sky in these two bands are therefore complementary. The distinct nature of the submillimeter sky far from the Galactic Plane is compellingly illustrated by comparing the Hubble Deep Field (HDF) imaged at optical or infrared wavelengths  with a submillimeter view (SCUBA 850$\\mu$m $^5$). As redshift increases, the volume of the Universe surveyed increases rapidly; this fact combined with an increasingly bright spectrum being brought into the observing band by higher redshifts (the so-called K-correction--$S_\\nu \\propto \\nu^{3-4}$) ensures that the brightest sources at submillimeter wavelengths are distant (z$>$1) dusty galaxies.  Furthermore, a generally higher star formation rate in the earlier epochs$^6$ adds to the dominance of high redshift objects in the submillimeter. Comparison of the expected blackbody spectrum of the ultraluminous starburst galaxy Arp 220 as seen at high redshifts with the frequency coverage and the sensitivity of ALMA suggests that The Atacama Large Millimeter Array should be able to detect Arp 220-like dusty starburst galaxies out to a redshift of 10 or more.  Galaxies like the present day Milky Way can also be detected out to z beyond 3.  Although continuum emission from dust holds great interest, ALMA will also provide spectral line images to detail the kinematics of the gas, its excitation and its chemical and isotopic composition throughout the history of the universe.  In the submillimeter and millimeter bands, the gas and dust enveloping galactic nuclei may be imaged without the optical obscuration which affects optical or infrared observations; the kinematics may be measured on spatial scales less than 100 pc or so.  Various redshifted spectral lines--for example, all rotational transitions of CO and infrared fine structure lines such as [C II], [C I], [N II] and [O I]--can be detected and studied to derive the molecular gas content and elucidate the character of star forming activity.  Millimeter/submillimeter astronomy has discovered most of the $>$120 known interstellar molecules.  These molecules may be quite complex--up to 12 atoms have been found in those small dense regions whose imaging will be ALMA's strength.  ALMA offers complete coverage of the available windows from a site of unprecedented atmospheric clarity. ALMA improves currently available spectral sensitivity by more than an order of magnitude (the weakest lines detected today are of order 1 Jy km/s in strength; compare to the ALMA sensitivity in Table 2 above).  But as sensitivity improves, understanding of an already crowded spectrum may not be possible without an accompanying increase in spatial resolution.  This ALMA also improves, by nearly two orders of magnitude, reaching resolutions of ~10 milliarcsec.  Furthermore, the throughput, measured by the spectral bandwidth open to ALMA (16 GHz), exceeds by several times that available now.  A recent spectral survey from the 12m telescope revealed some 15,000 lines in one spectral window toward the seven sources covered (Turner, private communication); ALMA's potential is clearly tremendous.  As stars form from a mixture of these molecules and dust, disks remain to form planets.  Most T Tau stars possess disks with masses from 0.1 to 0.001 solar masses and sizes of 100-500 AU. Similar masses and sizes of disks appear to be present in protostars, embedded in dense molecular cloud cores.  As the disks evolve, the chemistry continues to evolve, adapting to local conditions.  With resolution reaching 5AU in nearby (140 pc) systems, ALMA will observe the chemical makeup of the disks as a function of vertical height (the disk `atmosphere', with ultraviolet processing important higher up and condensation onto icy dust grains important mid-plane) and of radial distance from the star.  Jets driven by the young star generate shocks into a cocoon of material surrounding the jet, chemically processing that material.  Dutrey$^7$ has summarized ALMA's contribution to molecular disk studies; many molecules have been observed in the disks about T Tau stars such as DM Tau and GG Tau, including CN, HCN, HNC, CS, HCO+, C$_2$H and H$_2$CO, suggesting a photon-dominated chemistry; even at 4'' (600 AU) resolution changes in chemical structure of disks may be noted (e.g. LkCa15; Qi$^8$).  With ALMA, these structures will be more sensitively imaged; even time-resolved changes should be distinguished. ALMA will provide the sensitivity and resolution necessary to probe the planet-forming disk midplane.  Fairly complex molecules have been detected in the interstellar medium. The simple sugar glycolaldehyde, a building block for more complex sugars ribose and glucose, was recently detected.  Detection of its chemically reduced form, the ten atom molecule ethylene glycol, has just been announced.  Searches for the simplest amino acid, the ten atom molecule glycine (NH$_2$CH$_2$COOH) have proven negative, lacking sensitivity.  Glycine has a complex spectrum, exhibiting about a dozen lines per GHz on average.  The Green Bank Telescope offers the sensitivity and wide bandwidth for a renewed search; ALMA will add to these attributes directivity, reducing spectral confusion in the complex regions where large molecules congregate.  Detection may also be possible of other prebiotic molecules, such as adenine or uracile.  Molecules formed in dense regions may become incorporated in subplanetary or planetary objects.  In cold regions, dense gas becomes highly deuterated. Elevated deuterium levels have been observed in some disks, as well as in cometary comae, suggesting some dense gas maintains its identity as planetary systems form. Evidence from the solar system suggests that some organics remain, even in the inner solar system, as they have become incorporated into discrete bodies.  Carbonate blebs in the ALH84001 Martian meteorite maintained a heterogeneous pattern of magnetization suggesting transfer of the meteorite from Mars to Earth did not result in heating above 40$^o$C, Even after incorporation into larger bodies, organics may be transferred between bodies.  Hence a direct connection may exist between complex molecules observable by ALMA in interstellar space and the complex prebiotic molecules found in small condensed objects in the solar system.  In solar system objects, ALMA will obtain unobscured subarcsecond images. As an example, the flow of sulfur oxides from the volcanoes on Io may be imaged to differentiate their origins.  ALMA will be able to detect wobbling motion in stars caused by planets orbiting them.  ALMA will be able to detect planets in the process of formation.  ALMA will only marginally detect thermal emission from nearby extrasolar large planets. Although it will not be able to image even nearby Earthlike planets, it will be able to detect the interstellar chemicals which are available for the nourishment of early life.  As stars evolve, nuclear processes in their interiors results in a striated structure.  As their energy sources change according to local physical conditions, the star must adjust hydrodynamically, often shedding shells of processed material into the interstellar medium which cool and form dust and molecules.  ALMA will image these shells to reveal the isotopic and chemical gradients that reflect the chronology of invisible stellar nuclear processing.  ALMA is the only instrument, existing or planned which provides the combination of sensitivity, angular resolution and frequency coverage to achieve these scientific goals.  By providing high fidelity images at subarcsecond resolution ALMA will complement both the present generation of gound-based telescopes, such as VLT, Gemini and Keck and their successors, GSMT or CELT, and space telescopes such as the HST and NGST. ALMA's images of cool thermal emission complements optical/infrared data provided by these telescopes to enable astronomers to explore all of the peaks of the electromagnetic spectrum at similar resolution and high sensitivity. ",
        "conclusions": ""
    },
    "0209/astro-ph0209341_arXiv.txt": {
        "abstract": "We have observed the broad-line radio galaxy Arp~102B with {\\it ASCA} in order to determine the absorbing column density towards its X-ray source and measure its X-ray spectrum. The ultimate goal was to constrain the properties of the medium responsible for the metastable \\ion{Fe}{2} absorption lines observed in the ultraviolet spectrum of this object. The 0.5--10~keV X-ray spectrum is best described by a simple power-law model of photon index $1.58\\pm0.04$ modified by photoelectric absorption with an equivalent hydrogen column density of $(2.8\\pm0.3)\\times 10^{21}~{\\rm cm}^{-2}$. An Fe~K$\\alpha$ line is not detected with an upper limit to its equivalent width of 200~eV, assuming that its full width at half maximum is 60,000\\kms.  Using the column density measured from the X-ray spectrum and the observed spectral energy distribution as constraints, we explore simple (single-zone, constant-density) photoionization models for the absorber for a wide range of densities and ionization parameters in an effort to reproduce the strengths of the ultraviolet absorption lines. We find that densities of at least $10^{11}~{\\rm cm^{-3}}$ are needed. However, a single ionization parameter cannot explain all of the observed lines. An ionization parameter between $10^{-2.5}$ and $10^{-3.5}$ is needed to explain the Mg and Fe lines and the soft X-ray absorption, but the observed lines (from Si, C, Al, and H) require different density--ionization parameter combinations. According to the models, such an absorbing medium must be located very close to the source of ionizing radiation (within 5,000 gravitational radii) and must be very compact. As such, the properties of this absorbing medium differ from those of more luminous quasars, but are reminiscent of the absorber in the Seyfert galaxy NGC~4151. We suggest that the absorber is in the form of thin sheets or filaments embedded in an outflowing wind that overlays the accretion disk of Arp~102B. This picture is consistent with all of the available constraints on the central engine of this object. In an appendix, we present the X-ray spectrum of the source MS~1718.6+4902, which happened to fall within the field of view of the {\\it ROSAT} PSPC and the {\\it ASCA} GIS during the observations of Arp~102B. ",
        "introduction": "The broad-line radio galaxy (BLRG) Arp~102B \\citep*{ssk83,chf89,ch89} is the prototype of a subset of radio-loud active galactic nuclei (AGNs) with double-peaked Balmer lines in their optical spectra \\citep[see more examples in][]{eh94}. The profiles of these double-peaked emission lines and a number of additional, extreme spectroscopic properties of these objects have led to the interpretation that they harbor line emitting accretion disks whose inner parts have the form of an optically thin, geometrically thick ion torus \\citep{ch89,eh94}. The ion torus \\citep{r82} is very similar to what is known today as an advection-dominated accretion flow (ADAF; Narayan \\& Yi 1994, 1995\\nocite{ny94,ny95}), a structure which is realizable at accretion rates considerably lower than the Eddington rate. Although alternative scenarios for the origin of the double-peaked lines have been proposed, these are disfavored by the observational evidence available today (see reviews by Eracleous 1998, 1999\\nocite{e98,e99}).  In the process of studying the UV spectroscopic properties of AGNs with double-peaked emission lines we obtained a spectrum of Arp~102B with the {\\it Hubble Space Telescope (HST)}, which revealed an unexpected complex of \\ion{Fe}{2} {\\it absorption} lines very close to its systemic redshift \\citep{h96}. This complex includes much more than the transitions from the ground state of \\ion{Fe}{2} that are often observed in quasar absorption line systems and the interstellar medium (multiplets UV1, UV2, and UV3 from the $a\\,^6D$ ground state term): it also includes multiplets from the next two lowest states (terms $a\\,^4F$ and $a\\,^4D$).  Particularly surprising is the presence of multiplets UV62, UV63, and UV64, arising from the metastable $a\\,^4D$ term, which lies 0.99--1.10~eV above the ground state. Since absorption from excited states of \\ion{Fe}{2} is incompatible with the diffuse interstellar medium of the host galaxy and with intervening absorption-line systems, the absorber is most likely intrinsic to Arp~102B and its association with the active nucleus remains to be understood. Metastable \\ion{Fe}{2} absorption lines are not observed in Seyfert galaxies very often, with NGC~4151 being one of the rare examples \\citep{kraemer01}. NGC~4151 also shows a \\ion{C}{3}]~\\lam1176 line, arising from a level 6.05~eV above the ground state \\citep{k95}. Absorption lines from the ground state of \\ion{Fe}{2} have been observed in some other Seyfert galaxies, for example, NGC~3227 \\citep{crenshaw01}.  In marked contrast with Seyfert galaxies, \\ion{Fe}{2} absorption lines from metastable states are often seen in the spectra of quasars. Examples that have been known for quite some time include Q~0059--2735 \\citep*{h87,wcp95}, Hawaii~167 \\citep{c94}, and Mrk~231 \\citep*{bmm91,s95}. Recently, a number radio-loud quasars from the FIRST radio survey \\citep*{bwh95} have been found to host metastable \\ion{Fe}{2} lines as well \\citep{b97,b00}, as well as a number of quasars from the Sloan Survey \\citep{hall02}. Finally, we ourselves have found systems of metastable \\ion{Fe}{2} absorption lines, identical to that in Arp~102B, in the BLRG 3C~332 and in the LINER NGC~1097 \\citep{h97,e02}.  The excitation state of the gas responsible for the \\ion{Fe}{2} absorption lines in Arp~102B is very similar to what is observed in Q~0059--2735. The absorber in the latter object has been studied in detail by \\citet{wcp95}, who suggested that it is made up of condensations in a hotter broad-absorption line flow. The temperature of the condensations is likely to be of order $10^4$~K, implying that, if the metastable \\ion{Fe}{2} levels are populated by collisions, the electron density should be greater than $10^6~{\\rm cm}^{-3}$.  In our effort to understand the nature of the absorber in Arp~102B, we studied the X-ray spectrum obtained with the {\\it ROSAT} PSPC and found a very large absorbing column of neutral gas, in the range $(2-8)\\times 10^{21}~{\\rm cm}^{-2}$ \\citep{h97}. Because the {\\it ROSAT} spectrum was so heavily absorbed, more stringent constraints could not be derived. Therefore, we observed Arp~102B with {\\it ASCA}. This paper is devoted to the analysis and interpretation of the {\\it ASCA} spectra, including the development and assessment of simple photoionization models for the absorbing medium. Preliminary results of this work were presented by \\citet{e02}.  In \\S2 we describe the observations and data screening and in \\S3 we present the light curve and discuss the time variability. In \\S4 we study the X-ray continuum and determine the column density of the absorbing medium, while in \\S5 we derive upper limits to the equivalent width ($EW$) of the Fe~K$\\alpha$ line. In \\S6 we construct and evaluate simple photoionization models for the medium responsible for the UV and X-ray absorption.  We discuss our findings in \\S7 where we examine the X-ray spectral properties of Arp~102B in the context of other BLRGs and consider the implications of the model results for the nature of the absorber.  In an appendix we present the X-ray spectrum of MS~1718.6+4902, which happens to be in the field of view of Arp~102B. Throughout this paper we assume a Hubble constant of 50~km~s$^{-1}$~Mpc$^{-1}$, which implies a distance to Arp~102B of 146~Mpc, given its redshift of $z=0.0244$. ",
        "conclusions": "\\subsection{The X-Ray Spectral Properties of Arp~102B in Context}  The X-ray properties of Arp~102B reinforce previously known trends of systematic differences between Seyfert galaxies and BLRGs.  As we have already remarked, simple power-law spectra, with neither a soft nor a hard excess are fairly common among BLRGs observed with {\\it ASCA}, {\\it RXTE}, {\\it BeppoSAX}, and {\\it Ginga} (Wo\\'zniak et al. 1998; Sambruna et al. 1999; Hasenkopf, Sambruna, \\& Eracleous 2002; Eracleous et al. 2000; Grandi, Urry, \\& Maraschi 2002; Zdziarski et al. 1995)\\nocite{w98,sem99,hse02,esm00,gum02,z95}.  Seyferts, on the other hand, often show a soft and/or a hard excess in their spectra. The heavy absorption by ``cold'' (i.e. neutral) matter found in Arp~102B is not uncommon in radio-loud AGNs, in contrast to Seyfert~1 galaxies whose spectra often show the signature of a ``warm'' (i.e., ionized) absorber \\citep{r97,g98,k00}.  \\citet{sem99} find such heavy absorption in 1/3 of the the BLRGs and radio-loud quasars in their collection, with columns of $10^{21}~{\\rm cm}^{-2}$, or higher (see their Figure~6).  The relatively flat spectral index of Arp~102B, although not typical of luminous BLRGs (e.g., 3C~390.3, 3C~111, 3C~120, 3C~382) is characteristic of a subclass of radio-loud AGNs, the weak-line radio galaxies (hereafter WLRGs; see Sambruna et al.  1999\\nocite{sem99} for a summary of their X-ray properties). These are radio galaxies distinguished by the low luminosity of their [\\ion{O}{3}]$\\;$\\lam5007 emission lines \\citep{t98,l94}. The combination of spectral index and X-ray luminosity of Arp~102B puts it well within the region occupied by WLRGs in Figure~2d of \\citet{sem99}. Moreover, its [\\ion{O}{3}]$\\;$\\lam5007 emission-line luminosity of $9.6\\times 10^{40}~{\\rm erg~s^{-1}}$ is comparable to the [\\ion{O}{3}] luminosities of WLRGs, although its integrated 0.1-100~GHz radio luminosity is an order of magnitude lower than the least luminous WLRGs in the \\citet{t98} sample.  The weakness of the broad, disk-like Fe~K$\\alpha$ line compared to Seyferts appears to be a hallmark of BLRGs as a class \\citep{w98,sem99,esm00}.  This difference is illustrated in Figure~6 of \\citet{hse02}, where the locations of Seyfert galaxies and BLRGs in the Fe~K$\\alpha~EW - L_{\\rm 2-10~keV}$ plane are compared. In the vast majority of cases, BLRGs fall {\\it below} the region occupied by Seyferts, with Arp~102B representing the lowest-luminosity BLRG with available data.  The weakness of the Fe~K$\\alpha$ line of Arp~102B can be understood if the inner accretion disk has the form of an ion torus or ADAF, rather than a thin disk surrounded by a hot corona, as is thought to be the case in Seyfert galaxies \\citep*{hm93,hmg94}. The geometry of an optically thin, vertically extended inner disk illuminating a geometrically thin, dense outer disk reduces the available solid angle of the fluorescing medium (the thin disk) by a factor of 2 relative to a thin disk sandwiched by a hot corona \\citep*{ch89,z99} and results in a reduction of the the Fe~K$\\alpha$ $EW$ by the same factor. This picture is also appealing because it explains some of the other properties of Arp~102B \\citep[e.g., the optical double-peaked emission lines and the shape of the SED; see][]{ch89} and because it may be applicable to other low-luminosity BLRGs \\citep{esm00}. For example, 3C~332 is a dead-ringer for Arp~102B: not only does it have a highly-absorbed X-ray spectrum and an X-ray luminosity similar to that of Arp~102B \\citep{cf95}, but it also sports optical double-peaked emission lines and metastable \\ion{Fe}{2} absorption lines in the UV.  The above discussion refers to the extremely broad, disk-like Fe~K$\\alpha$ lines found in the \\asca\\ spectra of Seyfert galaxies. However, recent observations of bright Seyfert galaxies at high spectral resolution (with the \\chandra\\ grating spectrometers) or at high $S/N$ (with the \\xmm\\ CCD cameras) have shown that the lines can be decomposed into a broad, disk-like component and a narrower, bell-shaped component. Examples include NGC~3783 \\citep{k02}, NGC~5548 \\citep{yaqoob02a}, and NGC~3516 \\citep{tur02,netzer02}. The narrower component has a width of a few thousand \\kms\\ and can be plausibly attributed to the ``broad-line region,'' which is the source of the broad, optical emission lines. The $EW$ of the narrower component is typically around 90--130 eV. Such lines would not be detectable in the \\asca\\ spectrum of Arp~102B, since their $EW$s are just below the detection limit.  \\subsection{Constraints on the Absorber, Implications, and Future Prospects}  In summary, we find that simple, single-zone models for the absorber require high densities of $n\\geq 10^{11}~{\\rm cm^{-3}}$. However, not all UV absorption lines can be produced at the same ionization parameter: the Fe and Mg lines require an ionization parameter in the range $10^{-3.5} < U < 10^{-2.5}$, while the Si, Al, C, and H lines require a higher ionization parameter of $10^{-3.0} < U < 10^{-1.5}$ (the two regions of parameter space do not overlap, however, as illustrated in Figure~\\ref{fig_grid}).  The models that reproduce the $EW$s of the Si, Al, C, and H lines predict Mg and Fe line $EW$s that are much lower than the observed values, and {\\it vice-versa}: the models that reproduce the $EW$s of the Mg and Fe lines predict Si, Al, C, and H line $EW$s that are much lower than the observed values. The models that explain the Mg-Fe absorber can also explain the X-ray absorption observed in the {\\it ASCA} and {\\it ROSAT} spectra. Including a ``far-IR bump bump'' in the SED has a negligible effect on these conclusions.  The simple, single-zone models that we have considered, although partially successful, do not provide a fully satisfactory description of the absorber. One possible way out of this difficulty would be to adjust the abundances of Mg and Fe relative to the other elements by a factor of a few.  However, more sophisticated models are probably necessary and higher-resolution spectra are sorely needed to constrain them and guide their development.  One possible effect that an improved model could incorporate is radiative transfer in an accelerating medium, since it may play a role in determining the strengths of the absorption lines. This could be an important effect, if the absorber has the form of an outflowing wind.  Another possibility relevant to the outflowing wind scenario is that the absorber spans a range of densities or that it contains pockets or layers of higher or lower density gas.  Such a structure could lead to the two phases that are apparently needed to explain the strengths of the observed absorption lines. Different phases or layers of the absorber may reveal themselves in the form of ``components'' of the line profiles in high-resolution UV spectra. In this sense, our preferred model for the Mg-Fe absorber is reminiscent of the model of \\citet{kraemer01} for NGC~4151: their high-resolution spectra allowed them decompose the absorber into distinct kinematic components, one of which was responsible for the metastable \\ion{Fe}{2} absorption lines. Their photoionization model for this particular component required a high density of $n\\geq 10^{9.5}~{\\rm cm}^{-3}$, an ionization parameter of $U=10^{-1.8}$, and a high column density of $N_{\\rm H}\\approx 3\\times 10^{21}~{\\rm cm}^{-2}$. Their interpretation was that this component of the absorber was the closest to the ionizing source.  Examining the model results from a broader perspective and comparing them with results of detailed models of absorption lines in Seyfert galaxies, we find that a range of ionization parameters in the absorbing gas is not at all uncommon; in fact, it is the rule rather than the exception. This conclusion follows from studies of the UV and X-ray absorption lines in bright Seyfert galaxies such as NGC~3783 \\citep{k01,blustin02}, NGC~5548 \\citep{crenshaw99,kaastra02}, NGC~4051 \\citep{col01}, IRAS~13349+2438 \\citep{sako01}, NGC~3516 \\citep{kraemer02a}, and Mrk~509 \\citep{yaqoob02b,kraemer02b} using high-resolution spectra. More specifically, the UV absorption lines, and sometimes even the X-ray lines, can be decomposed into several kinematic components, which require gas with a wide range of ionization parameters (this is sometimes obvious from a simple comparison of the line profiles, without the need for detailed modeling). In other cases a wide range of ionization is infered with the help of photoionization models: such models cannot reporduce the strengths of all of the observed lines under the assumption of a uniform absorber with a single ionization parameter.  What is different between Arp~102B and Seyfert galaxies, however, is the presence of metastable \\ion{Fe}{2} absorption lines and the requirement for high densities to explain them. As we have noted in previous sections, such lines are rare in Seyfert galaxies.  Taking the properties of the absorber in Arp~102B inferred from the models at face value, we may deduce its distance from the source of ionizing radiation. We find that the absorber distance is related to the model parameters via $r=7\\times 10^{16}\\,(U_{-3}\\,n_{11})^{-1/2}$~cm (where $U_{-3}=U/10^{-3}$ and $n_{11}=n/10^{11}\\;{\\rm cm}^{-3}$).  Since $U = 10^{-3.0}$ and $n = 10^{11}~{\\rm cm}^{-3}$ represent the lowest allowed photon flux incident on the face of the absorber, the absorber distance should be less than $7\\times 10^{16}$~cm. We can recast this limit in terms of the gravitational radius, $r_{\\rm g} \\equiv GM_{\\bullet}/c^2$ (where $M_{\\bullet}$ is the mass of the central black hole), as $r/r_{\\rm g} \\lsim 5,000\\; (U_{-3}\\,n_{11})^{-1/2}\\; M_8^{-1}$ (where $M_8 = M_{\\bullet}/10^8\\; M_{\\odot}$). This distance is slightly larger than the size line-emitting portion of the disk, inferred from fits to the profiles of the broad Balmer emission lines, which extends over the range $r/r_{\\rm g}=500 - 1,000$ \\citep{chf89,ch89}. We may also estimate the thickness of the absorber along the line of sight as $\\delta r \\sim N_{\\rm H}/n$, which gives $\\delta r \\lsim 3\\times 10^{10}$~cm, or $\\delta r/r_{\\rm g} \\lsim 2\\times 10^{-3}\\;M_8^{-1}$, assuming that $n\\ge10^{11}\\;{\\rm cm}^{-3}$. Comparing the \\ion{Fe}{2} absorber in Arp~102B with those in luminous quasars, we find it to be quite different: it is more dense and located closer to the ionizing source. In comparison, the \\ion{Fe}{2} absorber in Q~0059--2735 has $10^6~{\\rm cm^{-3}} < n < 10^8~{\\rm cm^{-3}}$ and $2.4\\times 10^5\\; M_8^{-1} \\lsim r/r_{\\rm g} \\lsim 1.3\\times 10^6\\; M_8^{-1}$ \\citep{wcp95}, while in the FIRST quasars studied by \\citet{dek01,dek02} the \\ion{Fe}{2} absorber is characterized by $n\\sim 10^3-10^5~{\\rm cm^{-3}}$ and $r/r_{\\rm g}\\sim 10^8\\; M_8^{-1}$.  Independently of the model results, there are two more hints afforded by the data: (1) the absorbing material is either outside of the broad-line region or mixed in with it, since the \\ion{Mg}{2} absorption lines are deeper than the broad \\ion{Mg}{2} {\\it emission} lines, and (2) the dust-to-gas ratio in the absorber is rather low, since for a Galactic dust-to-gas ratio we would expect a reddening of $E(B-V)=0.43$, which would strongly attenuate the broad UV emission lines, contrary to what is observed \\citep{h96}.  Putting all the clues together, it is plausible to think of the absorber as thin, dense sheets or filaments embedded in an outflowing wind overlaying the outer accretion disk, which is thought to be the source of the broad, double-peaked emission lines of Arp~102B. This outflowing wind may also be the source of the broad UV high-ionization and Ly$\\alpha$ emission lines \\citep[see the general discussion by][and references therein, and the specific discussion of Arp~102B by Halpern et al. 1996]{cd89}. A very similar scenario may apply to other AGNs of relatively low luminosity, which have many properties in common with Arp~102B, including \\ion{Fe}{2} absorption lines in their UV spectra, for example the BLRG 3C~332 and the LINER NGC~1097 \\citep{h97,e02}."
    },
    "0209/astro-ph0209388_arXiv.txt": {
        "abstract": "We are setting up a new search for transiting extra-solar planets using the 0.5m Automated Patrol Telescope at Siding Spring Observatory, Australia. We will begin regular observations in September 2002. We expect to find $\\sim7$ new planets per year. ",
        "introduction": "An increasing number of teams are searching for extra-solar planets using the transit method (see Horne 2002 for a review). Although the probability of observing a transit for any given star is small, using a wide-field telescope a large number of stars can be monitored, potentially yielding a higher detection rate than the radial velocity surveys (e.g. Marcy et al. 2002; Mayor et al. 2002). Furthermore, for a transiting planet orbiting a sufficiently bright star, the planet's size, as well as its actual mass and orbital characteristics (from follow-up spectroscopy) can be determined, constraining models of its structure and formation. ",
        "conclusions": ""
    },
    "0209/astro-ph0209494_arXiv.txt": {
        "abstract": "Most models of dark energy predict the beginning of the accelerated epoch at \\( z\\leq 1 \\). However, there are no observational or theoretical evidences in favor of such a recent start of the cosmic acceleration. In fact, a model of dark energy coupled to dark matter is explicitely constructed that \\emph{a}) is accelerated even at high \\( z \\); \\emph{b}) allows structure formation during acceleration; and \\emph{c}) is consistent with the type Ia supernovae Hubble diagram, including the farthest known supernova SN1997ff at \\( z\\approx 1.7 \\). It is shown that the accelerated epoch in this model could have started as early as \\( z\\approx 5 \\). ",
        "introduction": "Cosmic acceleration is one of the most exciting discoveries of the recent years. The combined data from cosmic microwave background (CMB, see Netterfield et al. 2001, Lee et al. 2001, Halverson et al. 2001), cluster masses and abundance (see e.g. Bahcall et al. 2002) and supernovae Type Ia (SNIa) Hubble diagrams (Riess et al. 1998, Perlmutter et al. 1999) indicate that the universe is currently dominated by a very weakly clustered component that is able to accelerate the expansion. This component, denoted dark energy or quintessence, is supposed to fill roughly 70\\% of the cosmic medium.  However, the nature of the dark energy is still enigmatic (Wetterich 1988; Ratra \\& Peebles 1988; Frieman et al. 1995; Caldwell et al. 1998). Its equation of state and its interaction with dark matter are in fact so far subject only to very weak constraints (Perlmutter et al. 1999, Huey et al. 1999, Baccigalupi et al. 2002, Corasaniti \\& Copeland 2002, Bean \\& Melchiorri 2002, Amendola et al. 2002) so that there exist a large variety of different dark energy models still viable (see for instance the review by Peebles \\& Ratra 2002). Perhaps the only characteristic common to almost all proposed models of dark energy is that its domination started very recently: the present epoch of acceleration was preceded by a decelerated epoch at redshift \\( z\\geq 1 \\) in which structure formed. Two exceptions to this scheme are the model of Dodelson et al. 2000, in which the dark energy is periodically dominating, and the generic quintessence model of Lee \\& Ng 2002, in which the dark energy was dominating again at very large redshifts; in both cases the expansion was however decelerated at \\(z \\) of order unity. The fact that the acceleration sets in just recently in the cosmic history is one aspect of the {}``coincidence problem{}'' that demands an explanation: see for instance the discussion based on the anthropic principle by Vilenkin (2001). Another aspect of the {}``coincidence problem{}'' is commonly phrased as why the dark energy and the dark matter densities happen to be similar just today (Zlatev et al. 1999).  The existence of a decelerated epoch rests on three arguments, one theoretical and two observational. The theoretical argument is based on the most common models of dark energy. In fact, a FRW universe with scale factor \\( a=(1+z)^{-1} \\) filled with pressureless dark matter (subscript \\( m \\)) and non-interacting dark energy (subscript \\( \\phi  \\)) with a constant equation of state \\( w_{\\phi }=1+p_{\\phi }/\\rho _{\\phi } \\) is decelerated before the epoch \\begin{equation} \\label{zacc} z_{acc}=[(2-3w_{\\phi })(\\Omega _{\\phi }/\\Omega _{m})]^{1/(3-3w_{\\phi })}-1. \\end{equation} Here and in the following the density parameters \\( \\Omega _{i} \\) refer to the present quantity of the \\( i \\)-th component. Given the current estimates \\( \\Omega _{m}\\approx 0.3\\pm 0.1,\\Omega _{\\phi }=0.7\\pm 0.1 \\) one has \\( z_{acc}\\leq 1 \\) for all values of \\( w \\). Even allowing for a large curvature, e.g. \\( |\\Omega _{k}|\\equiv |1-(\\Omega _{m}+\\Omega _{\\phi })|<0.3 \\), more than three sigma away from CMB measurements (de Bernardis et al. 2001), it is not possible to go beyond \\( z_{acc}\\approx 1.3 \\). Therefore, an hypothetical observation of a value of \\( z_{acc} \\) significantly larger than unity would signal that some of the assumptions leading to (\\ref{zacc}) are false.  The two observational arguments are as follows. First, since gravitational instability is ineffective in an accelerated regime, it seems that an extended accelerated era is in contrast with the observed large scale structure. Second, the recent supernova SN1997ff at \\( z\\approx 1.7 \\) is consistent with a decelerated expansion at the epoch of light emission (Benitez et al. 2001, Riess et al. 2002, Turner \\& Riess 2001) and seems to provide {}``a glimpse of the epoch of deceleration{}'' (Benitez et al. 2001).  The aim of this paper is to show that all three arguments are not generally true: they are in fact valid only in a restricted class of models. As a counterexample, a simple flat-space model with a constant equation of state will be explicitely constructed that can be accelerated at large \\( z \\), allows structure formation and is not in conflict with the current supernova data, including SN1997ff. Such a dark energy model, based on a coupling of dark energy to dark matter (Amendola 2000), makes very strong and unique predictions that can be easily tested in the near future.  Some of the results here reported have been already applied to a specific model of string-inspired dark energy (Amendola et al. 2002). This paper generalizes the arguments without referring to specific realizations and performs a likelihood comparison to the SNIa data. ",
        "conclusions": "This paper shows that high-\\( z \\) acceleration is a viable possibility if dark energy couples to dark matter. Although the present abundance of baryons limit the epoch of acceleration to \\( z_{acc}<5 \\) , this is still much earlier than the standard models of uncoupled dark energy, which hardly reaches \\( z_{acc}=1 \\). The future observations of SNIa at high redshift will be well suited to detect or reject an early acceleration.  A strong coupling between dark energy and dark matter can also be detected through the biasing and the rate of growth of perturbations, as discussed in Amendola \\& Tocchini-Valentini (2002). In principle, the coupling could be seen also in astrophysical objects like galaxy clusters by comparing the forces felt by dark matter halos and baryons (for instance, baryons in the intracluster gas), i.e. as an astrophysical test of the violation of the equivalence principle. This approach, discussed in detail in Gradwohl \\& Frieman (1992), requires several assumptions on the distribution of dark haloes. As emphasized by Peebles (2002), however, it has the potentiality to open unexplored paths in cosmology."
    },
    "0209/astro-ph0209177_arXiv.txt": {
        "abstract": "s{ The primordial abundance of $^4$He, $Y_P$, is one of the hottest themes in present-day astronomy, mostly due to its cosmological relevance. The disagreement between different determinations has been currently reduced to the 1-2\\% level, but these differences are still large enough to have deep implications for Big-Bang nucleosynthesis. It is therefore crucial to estimate precisely the uncertainties involved in the measurement of $Y_P$. Here, I review the methods used in the determination of $Y_P$ and the related uncertainties. I also discuss some recent results, and emphasize the assumptions underlying the differences among them. } ",
        "introduction": "Research on the abundance of primordial helium ($Y_P$) has entered its golden age. The exact value of $Y_P$ is one of the few missing pieces in the puzzle of the Big-Bang scenario, fueling a lively debate among astronomers, and motivating a huge amount of literature on the subject. The other side of the coin is that this literature is stuffed with numerically subtle argumentations, devoted to the quantitative determination of very specific aspects of the issue. Having established the basic principles, all the effort is now committed to the fine tuning of these details. $Y_P$ measurements are increasingly precise, but still not accurate enough to be used as strong cosmological constraints. The debate concerns now the third digit of $Y_P$: a scale small indeed, but where differences still matter.  My scope here is to present an overview of the subject and its implications. The most important part of this task is trying to help the reader understand, without getting lost in numbers, where the differences come from. Unfortunately, I find it impossible to be simultaneously clear and exhaustive in just a few pages. My own choice is to be as clear as I can, though it implies that many important studies on this topic will not be mentioned. I apologize for that to their authors, hoping that this review will have, at least, the effect to make readers grasp the essence of the debate, and, hopefully, want to know more. ",
        "conclusions": "From the discussion above, it is apparent that the central problem in the determination of $Y_P$ is that several physical mechanisms acting in H{\\sc~ii} regions are still not completely understood. Furthermore, although I described them separately for exposing convenience, these mechanisms interact with each other in complex ways. A huge collective effort is presently aiming to pinpoint the relevance of these effects, in part with direct observations, more often with numerical simulations. In the meanwhile, whether they actually play a role or not remains mostly a question of personal judgement, based on pieces of evidence that are more or less compelling, but rarely conclusive. Because personal judgement is so important, it is a natural question whether unconscious individual prejudices might be playing a role in obtaining one result or another. It is well known that, to some extent, this kind of bias is always present in any analysis, and that it may be particularly insidious. I therefore conclude with a personal remark: it would be extremely instructive for all of us if the relevant scientists in the field build up a kind of double-bind experiment, with both real and ``placebo'' data, to evaluate the impact of the human factor on $Y_P$ determinations."
    },
    "0209/astro-ph0209207_arXiv.txt": {
        "abstract": "{ The Master Catalogue of stars towards the Magellanic Clouds (MC2)\\thanks{\\tt http://vizier.u-strasbg.fr/MC2/} is a multi--wavelength reference catalogue. The current paper presents the first results of the MC2 project. We started with a massive cross--identification of the two recently released near--infrared surveys: the DENIS Catalogue towards the Magellanic Clouds (DCMC) with more than 1.3 million sources identified in at least two of the three DENIS filters ($I$ $J$ $K_\\textrm{s}$) and  the 2nd Incremental Release of the 2MASS point source catalogue ($J$ $H$ $K_\\textrm{s}$) covering the same region of the sky. Both point source catalogues provide an unprecedented wealth of data on the stellar populations of the Magellanic Clouds (MCs). The cross--matching procedure has been extended to optical wavelength ranges, including the UCAC1 (USNO) and GSC2.2 catalogues. New cross--matching procedures for very large catalogues have been developed and important results on the astrometric and photometric accuracy of the cross--identified catalogues were derived. The cross--matching of large surveys is an essential tool to improve our understanding of their specific contents. This study has been partly supported by the {\\sc astrovirtel}\\thanks{\\tt http://www.stecf.org/astrovirtel/} project that aims at improving access to astronomical archives as virtual telescopes. ",
        "introduction": "\\begin{figure*} \\begin{center} \\begin{tabular}{cc} \\psfig{figure=dcmcmag-enlarge.ps,clip=,height=5cm,angle=-90} & \\psfig{figure=2massmag-enlarge.ps,clip=,height=5cm,angle=-90} \\\\ \\psfig{figure=gsc2mag-enlarge.ps,clip=,height=5cm,angle=-90} & \\psfig{figure=ucacmag-enlarge.ps,clip=,height=5cm,angle=-90} \\\\ \\end{tabular} \\end{center} \\caption[example] { \\label{fig:compl} Completeness diagrams for four major surveys covering the LMC. Each plot gives the number of sources per magnitude bin. The bin size is 0.1 magnitude. Note that: 2MASS observations (upper right) are deeper than DENIS observations (upper left) in the $K_\\textrm{s}$ band; the GSC2.2 magnitudes (bottom left) show a sharp cut--off ; the UCAC1 magnitudes (bottom right) are only indicative, since the UCAC1 is not a photometric catalogue and furthermore this is only a preliminary catalogue, which means some improvements are expected in future releases. } \\end{figure*}   The Magellanic Clouds (MCs) are among the best suitable places to study the stellar evolution outside the Milky Way, because of their proximity and their various stellar populations. Near--infrared surveys provide useful data for this kind of study because of their insensitivity to interstellar reddening. The Magellanic Clouds have been recently fully observed by two major infrared surveys: the DEep Near--Infrared Survey of the Southern Sky~--~DENIS (Epchtein et al.\\ 1997)  and the Two Micron All Sky Survey -- 2MASS (Skrutskie et al.\\ 1997). A Near--Infrared Point Source Catalogue towards the Magellanic Clouds, based on DENIS data, has been published (Cioni et al.\\ 2000a; DCMC). The part of this catalogue devoted to the Large Magellanic Cloud (LMC) covers an area of $19.87 \\times 16$ square degrees centered on ($5^\\mathrm{h}27^\\mathrm{m}20^\\mathrm{s}$,~$-69^{\\circ}00\\arcmin00\\arcsec$). To compile this catalogue, the objects were required to be detected in at least two of the three DENIS bands $I (\\mathrm{Gunn-}i, 0.79\\mu \\mathrm{m})$, $J (1.22\\mu \\mathrm{m})$, $K_\\textrm{s} (2.15\\mu \\mathrm{m})$. The 2MASS project observed the whole Magellanic Clouds in three photometric bands: $J (1.23\\mu \\mathrm{m})$, $H (1.63\\mu \\mathrm{m})$ and $K_{\\mathrm{s}} (2.15\\mu \\textrm{m})$. For this work we used only the data available from the 2nd Incremental Release PSC\\footnote{\\tt http://www.ipac.caltech.edu/2mass/}, which do not cover two rectangular regions crossing the bar of the Large Magellanic Cloud and some cross--like gaps around bright stars. { ($4^{\\mathrm{h}}00^{\\mathrm{m}}00^{\\mathrm{s}}$ $<$ R.A. $<$ $7^{\\mathrm{h}}00^{\\mathrm{m}}00^{\\mathrm{s}}$; $-78^{\\circ}01\\arcmin37\\arcsec$ $<$ Dec. $<$ $-60^{\\circ}48\\arcmin00\\arcsec$)},    %  The number of sources from both surveys are recorded in Table~\\ref{tab:nbsource}. Because of different sensitivity limits, DENIS sources detected only in the $I$ and $J$ bands are often detected in $H$ and $K_\\textrm{s}$ by 2MASS. 2MASS observations are more than one magnitude deeper than DENIS in the $K_\\textrm{s}$ channel (due to a better thermalization), while they are roughly equivalent in the $J$ channel (Fig.~\\ref{fig:compl}). Thus it appeared very interesting to cross--match the two catalogues to complete the spectral range of the DCMC $IJ$--sources with the $H$ and $K_\\textrm{s}$ bands coming from 2MASS, though observations are not simultaneous.  More generally, cross--matching catalogues is highly relevant for completing the spectral or spatial coverage when there are missing or unpublished data. It is also a powerful tool to cross--validate the catalogues and search for discrepancies.  Cross--matching infrared (IR) with optical catalogues, such as DCMC/2MASS with the Guide~Star~Catalog~II (GSC2.2), helps on producing new colour--magnitude and colour--colour diagrams, thus offering multispectral views of the LMC. In the cross--matching procedure we also included the proper motions from the USNO CCD Astrograph Catalogue (UCAC1), in order to discriminate MC members from foreground stars. The resulting MC2 catalogue provides an unprecedented basis for the study of stellar populations in the Magellanic Clouds and for further cross--identifications with catalogues at other wavelengths.  { Section 2 gives an overview of each survey towards the LMC. Section 3 deals with the strategy developed to cross--match the infrared DENIS and 2MASS catalogues. Following in Sect. 4 is a comparison of the DENIS and 2MASS photometric systems. In Sect. 5 we add the optical GSC2.2 and UCAC1 catalogue to the cross--matching procedure. In Sect. 6 we present a few multispectral views of the stellar populations of the Clouds, based on the MC2 data. } \\begin{table} [h]   % \\small \\begin{center} \\caption{Number of sources as a function of detected wavebands in the DCMC and 2MASS catalogues. } \\label{tab:nbsource} \\begin{minipage}[r]{\\linewidth} \\centering \\begin{tabular}{rr|rr} \\multicolumn{4}{c}{\\bf LMC}  \\\\ \\hline \\multicolumn{2}{c|}{DCMC} & \\multicolumn{2}{c}{2MASS}\\\\ \\hline $IJK_\\textrm{s}$ & 297,031 & $JHK_\\textrm{s}$ & 1,996,382 \\\\ $IJ$ & 1,151,789 & $JK_\\textrm{s}$ & 66 \\\\ $IK_\\textrm{s}$ & 8,724 & $JH$ &  - \\\\ $JK_\\textrm{s}$ & 1,897 & $HK_\\textrm{s}$ & 4 \\\\ &  & $J$ & 11 \\\\ &  & $H$ & - \\\\ &  & $K_\\textrm{s}$ & 23 \\\\ &  & {\\it Saturated} & 259 \\\\ \\hline Total & 1,459,441 & Total & 1,996,745 \\\\ \\hline \\end{tabular} \\end{minipage} \\end{center} \\end{table} ",
        "conclusions": "The Master Catalogue of stars towards the Magellanic Clouds (MC2) is now available on the web at CDS\\footnote{\\tt http://vizier.u-strasbg.fr/MC2/}. It is a compilation of cross--identified surveys, from optical to IR. The MC2 roughly covers the following area: $4^{h}$ to $7^{h}$ in Right Ascension, and $-61^{\\circ}$ to $-78^{\\circ}$ in Declination, with slight variations according to the catalogue considered.  We are currently working on the cross--identification of analogous catalogues in the direction of the Small Magellanic Cloud and we plan to add catalogues and tables at other wavelengths: ROSAT, IRAS,and many more specific catalogues, as well the variability informations coming either from MACHO, EROS, OGLE to the second version of the catalogue.  A typical query of the MC2 returns several lines of data. {An example is given in Table~3.} Each line of data contains the name of the source for all the original catalogues, followed by the magnitudes and the proper motion when the UCAC1 is present. For each catalogue, the distance of the cross--identification is given, except for 2MASS which is taken as reference. The distance associated to a DCMC source is the distance to the 2MASS counterpart. The distance associated to a GSC2.2 source is the distance to the 2MASS counterpart, or the DCMC counterpart when there is no 2MASS counterpart. The distance associated to a UCAC1 source is the distance to the MC2 counterpart (2MASS, DCMC or GSC2.2, depending on the detection bands). At the beginning of each line, R.A. and Dec. are given. The choice of the coordinates has been as follows: when possible, we kept the 2MASS coordinates as the reference, otherwise we took the GSC2.2, then the UCAC1 and finally the DCMC ones. \\begin{table*} [h]   % \\small \\begin{center} \\caption{Subsample of the MC2. Each line of data corresponds to one point source. }  \\begin{tabular}{cccccccccccccccccccccccccccccc} \\hline R.A.     &     Dec. &        2MASS &   J  &  dJ &  H   & dH  &   Ks & dKs \\\\ \\hline 93.769831&-75.633698&0615047-753801&12.380&0.022&12.177&0.024&12.076&0.029\\\\ 93.760565&-75.632538&0615025-753757&16.072&0.089&15.975&0.178&15.273&null \\\\ 93.769231&-75.629761&0615046-753747&16.413&0.123&16.141&0.211&15.611&0.261\\\\ 93.756995&-75.621536&0615016-753717&16.192&0.098&15.352&0.106&15.443&0.215\\\\ \\hline DCMC\t  &\tI&  dJ &   J  &   dJ&\tKs &  dKs &  dist  \\\\ \\hline 061504.60-753800.7&12.999&0.006&15.788&0.188&99.000&99.000&0.853030\\\\ 061502.40-753757.5&16.501&0.062&15.866&0.230&99.000&99.000&0.621811\\\\ ..\t\t  &...   &...  &...   &...  &...   &...   &...     \\\\ ..\t\t  &...   &...  &...   &...  &...   &...   &...     \\\\ \\hline GSC2.2 &F    &dF       &J    &dJ       &V   &dV   &dist    \\\\ \\hline S1102121266 &13.24&$\\pm$0.23&13.95&$\\pm$0.17&--- &$\\pm$&0.192289\\\\ S11021214338&17.40&$\\pm$0.23&17.90&$\\pm$0.18&--- &$\\pm$&0.203876\\\\ S11021214339&17.44&$\\pm$0.23&18.29&$\\pm$0.18&--- &$\\pm$&0.554134\\\\ S11021214495&18.30&$\\pm$0.24&---  &$\\pm$    &--- &$\\pm$&0.251949\\\\ \\hline UCAC1\t&mag  &PMra &PMdec& dist   \\\\ \\hline 00697288&13.22&+27.3&-9.7 &0.045439\\\\ ...\t&...  &...  &...  &...     \\\\ ...\t&...  &...  &...  &...     \\\\ ...\t&...  &...  &...  &...     \\\\ \\end{tabular}  \\end{center} \\end{table*}  We decided to keep the distances of the cross--identifications in the MC2 to give the user the opportunity to judge the reliability of each cross--identification. We have also shown that for some DCMC point sources, this distance was not reliable. But since links for each source allow to access the complete data from the original catalogues through the VizieR search engine (Ochsenbein et al.\\ 2000), it is always possible to retrieve the strip number of the DCMC source and then go to the MC2 web site to find the shifts associated to this strip. Those links are also very valuable in order to retrieve observational data such as image or scan number, flags or whatever parameter the user would like to know from the original catalogues.  This reference catalogue is made available as a support for a number of studies concerning, e.g. the stellar populations in the Magellanic Clouds, the structure of the Clouds, or certain classes of objects (Cepheids, AGB stars, etc.). { Recent articles, such as those by Zaritsky et al. (2002), Van der Marel (2001), Nikolaev \\& Weinberg (2001) and Cioni et al.\\ (2000b) have demonstrated the power of optical and near--infrared surveys to improve our understanding on these neighbouring galaxies.}"
    },
    "0209/astro-ph0209031_arXiv.txt": {
        "abstract": "We have compiled a pseudo-snapshot data set of two-colour observations from the {\\em Hubble Space Telescope} archive for a sample of 53 rich LMC clusters with ages $10^{6}-10^{10}$ yr. We present surface brightness profiles for the entire sample, and derive structural parameters for each cluster, including core radii, and luminosity and mass estimates. Because we expect the results presented here to form the basis for several further projects, we describe in detail the data reduction and surface brightness profile construction processes, and compare our results with those of previous ground-based studies.  The surface brightness profiles show a large amount of detail, including irregularities in the profiles of young clusters (such as bumps, dips, and sharp shoulders), and evidence for both double clusters and post core-collapse (PCC) clusters. In particular we find power-law profiles in the inner regions of several candidate PCC clusters, with slopes of approximately $-0.7$, but showing considerable variation. We estimate that $20 \\pm 7$ per cent of the old cluster population of the LMC has entered PCC evolution, a similar fraction to that for the Galactic globular cluster system. In addition, we examine the profile of R136 in detail and show that it is probably not a PCC cluster.  We also observe a trend in core radius with age that has been discovered and discussed in several previous publications by different authors. Our diagram has better resolution however, and appears to show a bifurcation at several hundred Myr. We argue that this observed relationship reflects true physical evolution in LMC clusters, with some experiencing small scale core expansion due to mass loss, and others large scale expansion due to some unidentified characteristic or physical process. ",
        "introduction": "The star cluster system of the Large Magellanic Cloud (LMC) is unique in containing rich star clusters of masses comparable to Galactic globular clusters, but covering a wide age range ($10^{6}-10^{10}$ yr) and being close enough for detailed observation. It therefore offers a seemingly perfect opportunity for studies of all aspects of star cluster astronomy, from cluster formation, evolution and dynamics, to luminosity and mass function studies, as well as investigations which shed light on the evolution history of the entire LMC. It is surprising then, that while catalogues of high resolution surface brightness profiles and structural parameters exist for Galactic globular clusters (e.g., Trager, King \\& Djorgovski \\shortcite{trager}), no such uniform catalogue exists for a large sample of the rich LMC clusters.  There has however been some considerable activity in this field. In particular, significant numbers of surface brightness and/or density profiles have been published for young and intermediate age clusters (Elson, Fall \\& Freeman \\shortcite{eff}, hereafter EFF87; Elson, Freeman \\& Lauer \\shortcite{efl}; Elson \\shortcite{elsonbig}, hereafter E91; Elson \\shortcite{elsonfive}); old clusters \\cite{mateo}; and clusters with different spatial distributions -- in the LMC disk (Kontizas, Chrysovergis \\& Kontizas \\shortcite{kck}); in the LMC disk and within 5 kpc of the rotation centre (Chrysovergis, Kontizas \\& Kontizas \\shortcite{ckk}); and in the LMC halo (Kontizas, Hadjidimitriou \\& Kontizas \\shortcite{khk}; Metaxa, Kontizas \\& Kontizas \\shortcite{mkk}). These studies are all ground-based and therefore suffer from problems -- primarily crowding and seeing related -- which limit their resolution, particularly in the inner regions of clusters. This in turn renders the derivation of key parameters, such as the core radius, rather uncertain. In addition, each set of authors uses a different data set and makes different measurements (e.g., Elson and collaborators construct $V$-band surface brightness profiles; Mateo uses $B$-band profiles; and Kontizas and collaborators use number density profiles), so most of the important derived values are not strictly comparable between studies.  We have taken advantage of the presence of a large number of observations of rich LMC clusters in the {\\em Hubble Space Telescope} ({\\em HST}) archive to compile a high resolution data set, unaffected by the problems which beset ground-based studies. Rather than simply providing an atlas of surface brightness profiles from this data, the aim of the present study is to obtain a statistically homogeneous set of profiles and key structural parameters for as many clusters as possible, for the purposes of differential comparison. To this end, we require a sample as free from bias as possible, and we have therefore applied a uniform selection and reduction process to the available observations. Because these procedures are rather detailed, and because we expect our measurements to form the basis of several additional projects, in this paper we discuss at length the data set and its selection (Section \\ref{sample}) and the reduction and profile construction processes (Sections \\ref{reduction} and \\ref{sbprofile} respectively). In Section \\ref{results} we present the surface brightness profiles and structural parameter measurements, compare these results with those of the authors previously mentioned, and examine some of the interesting sub-groups present in the sample -- such as binary clusters and post core-collapse clusters. Finally, in Section \\ref{discussion} we observe and discuss the relationship between core radius and age for the LMC cluster system. This relationship has previously been studied (Elson et al. \\shortcite{efl}; E91; Elson \\shortcite{elsonfive}); however our measurements are able to provide new insight into this problem.  The data presented in Tables \\ref{data}, \\ref{ages}, \\ref{params}, and \\ref{luminmass}, and the surface brightness profiles (Fig. \\ref{plots}) are available on-line at {\\em http://www.ast.cam.ac.uk/STELLARPOPS/LMC\\_clusters/}. ",
        "conclusions": "We have compiled a pseudo-snapshot data set of two-colour observations from the {\\em HST} archive for a sample of 53 rich LMC clusters spanning the full age range $10^{6}$ to $10^{10}$ yr. The emphasis has been on trying to make this compilation and the subsequent reduction process as homogeneous as possible without sacrificing data integrity. We have also compiled literature estimates for the ages and metallicities of these clusters, again trying to maintain consistency as far as possible. From the {\\em HST} observations, we have constructed surface brightness profiles for the entire sample and obtained structural parameters for each cluster, including the core radius and power-law slope at large radii. Using these parameters we have also estimated the total luminosity and mass for each cluster. These data, along with the surface brightness profiles, are available on-line at {\\em http://www.ast.cam.ac.uk/STELLARPOPS/LMC\\_clusters/}.  The surface brightness profiles show a rich amount of detail, with young clusters in particular exhibiting bumps, shoulders and dips in their profiles. We see evidence for double clusters in our sample, as well as post core-collapse clusters. The PCC candidates are especially interesting -- the two best examples show clear power-law profiles at small radii, with slopes $\\beta \\sim 0.7$ in the $\\log\\mu - \\log r$ plane. Our sample covers twelve of the fifteen definite old LMC globular clusters, and we are able to estimate that $20 \\pm 7$ per cent of these clusters are PCC objects, matching the $20$ per cent estimated for the Galactic globular cluster system. We have also shown that R136 requires a two component fit to its profile, and that it is likely not yet in a PCC state.  If core radius is plotted against age for the entire sample, we see that while all the young clusters have compact cores, the spread in core radius increases with age, with the oldest clusters covering the full range of core radii measured. We have argued that this trend reflects real evolution in cluster structure with age. The distribution of clusters on the plot suggests a bifurcation at several hundred Myr, with most clusters maintaining small cores consistent with standard isolated globular cluster evolution, but with several moving to the upper right of the diagram and evolving large diffuse cores. We suggest that these clusters must be different to the ``standard'' clusters, either by having exceptional stellar populations, or by being subjected to an external influence. We are currently employing $N$-body simulations to explore several physical processes which fall into these categories."
    },
    "0209/astro-ph0209082_arXiv.txt": {
        "abstract": "We have carried out a survey of optically-selected dark clouds using the bolometer array SCUBA on the James Clerk Maxwell Telescope, at $\\lambda = 850$~$\\mu$m.  The survey covers a total of 0.5 square degrees and is unbiased with reference to cloud size, star formation activity, or the presence of infrared emission.  Several new protostars and starless cores have been discovered; the protostars are confirmed through the detection of their accompanying outflows in CO(2--1) emission.  The survey is believed to be complete for Class~0 and Class~I protostars, and yields two important results regarding the lifetimes of these phases.  First, the ratio of Class~0 to Class~I protostars in the sample is roughly unity, very different from the 1:10 ratio that has previously been observed for the $\\rho$ Ophiuchi star-forming region.  Assuming star formation to be a homogeneous process in the dark clouds, this implies that the Class~0 lifetime is similar to the Class~I phase, which from infrared surveys has been established to be $\\sim 2 \\times 10^5$~yr. It also suggests there is no rapid initial accretion phase in Class~0 objects.  A burst of triggered star formation some $\\sim 10^5$~yr ago can explain the earlier results for $\\rho$ Ophiuchus.  Second, the number of starless cores is approximately twice that of the total number of protostars, indicating a starless core lifetime of $\\sim 8 \\times 10^5$~yr.  These starless cores are therefore very short-lived, surviving only two or three free-fall times.  This result suggests that, on size scales of $\\sim 10^4$~AU at least, the dynamical evolution of starless cores is probably not controlled by magnetic processes. ",
        "introduction": "Systematic surveys of the earliest stages of low-mass star formation provide vital information on the physics of cloud collapse. In particular, if complete samples of protostars and pre-stellar cores can be identified, their relative lifetimes can be estimated if it is assumed that the clouds are not being observed at a special time in their evolution.  Such studies are vital for differentiating between models of star formation, which can predict very different protostellar accretion histories.  In some models, the evolution of molecular cloud cores is controlled entirely by strong magnetic fields, leading to lifetimes determined by the ambipolar diffusion timescale, which is typically $10^7$~yr (Ciolek \\& Mouschovias 1994).  However, recent 3-dimensional numerical simulations of turbulent, gravitationally unstable, clouds show that cores can form and evolve over only a few dynamical timescales, even if significant magnetic fields are present (Li et al.\\ 2000; Heitsch, Mac Low, \\& Klessen 2001).  For typical cloud conditions, this implies evolution on timescales of about $10^6$~yr.  Once a gravitationally unstable core has formed and started to collapse, the accretion rate of the protostar is governed by the initial conditions at the onset of collapse.  The idealized collapse of a singular isothermal sphere (Shu 1977) proceeds with a uniform accretion rate.  Other accretion models starting from different initial conditions predict a short phase of rapid accretion, succeeded by a more or less constant accretion rate (e.g., Foster \\& Chevalier 1993; McLaughlin \\& Pudritz 1997). If evolutionary phases of different accretion rates can be identified observationally then their relative lifetimes can then be estimated, possibly enabling the discrimination between these different models.  Complete surveys of infrared protostars have been used to obtain good estimates for the lifetime of infrared protostars, in particular the Class I, II, and III phases of protostellar evolution.  From such surveys Wilking et al.\\ (1989) and Kenyon et al.\\ (1990) conclude that the Class I phase lasts $\\sim 2 \\times 10^5$~yr in the low-mass, star-forming regions of $\\rho$ Ophiuchus and Taurus, and this lifetime is consistent with typical cloud collapse models (e.g., Adams, Lada, \\& Shu 1987).  However, the timescales associated with the earlier phases of star formation, specifically the Class~0 phase (Andr\\'e et al.\\ 1993) and the pre-collapse phase (Ward-Thompson et al.\\ 1994), are much more uncertain.  Most samples of dense cores derive originally from catalogues of optically-selected dark clouds, which were subsequently searched for associated IRAS emission (e.g., Beichman et al.\\ 1986; Clemens \\& Barvainis 1988; Benson \\& Myers 1989; Lee \\& Myers 1999; Jijina, Myers, \\& Adams 1999).  Since Class~0 protostars are typically too cold and faint to have been detected by IRAS, complete samples of both Class~0 protostars and truly ``starless'' cores (as opposed to cores lacking an associated IRAS source or 2~$\\mu$m emission) have been difficult to obtain.  In one of the few regions where complete samples of Class~0 and Class~I objects exist --- the $\\rho$ Ophiuchi cloud --- the results imply that the Class~0 phase lasts only one tenth of the Class~I phase in this region, i.e., $2 \\times 10^4$~yr (Andr\\'e \\& Montmerle 1994).  However, there are few, if any, good statistical constraints in other regions of star formation.  The advent of large bolometer arrays on millimeter and submillimeter telescopes has made the first systematic surveys for the earliest phases of star formation possible.  Typical observing wavelengths, $\\lambda \\sim 0.5$ to 1~mm, lie towards the Rayleigh-Jeans side of the Planck function for all reasonable dust temperatures $T \\ga 7$~K\\@. Millimeter and submillimeter dust emission is therefore an excellent tracer of the youngest embedded protostars and starless dense cores, which are regions of high dust column density that can be too cold for detection by far-infrared instruments.  In this paper we present the second and concluding part of a survey, using the submillimeter camera SCUBA on the James Clerk Maxwell Telescope (JCMT), of dust continuum emission from typical, nearby, molecular clouds forming low-mass stars. The observations are sufficiently sensitive to detect all the embedded protostars and starless cores above a certain mass limit, and so allow us to estimate the lifetimes of these earliest phases of star formation. In an earlier paper (Visser, Richer, \\& Chandler 2001, hereafter Paper I) we presented initial results derived from images of the smaller clouds in the sample.  Here we include the images of the larger clouds, and using the full survey present a detailed analysis of the structure of the cores, their star forming properties, and the properties of their molecular outflows.  We also estimate the relative and absolute lifetimes of the various pre-stellar and protostellar phases based on their detection rates. ",
        "conclusions": "We have surveyed a sample of optically-selected dark clouds for the submillimeter dust emission associated with embedded protostars and starless dense cores.  The optical selection criterion is equivalent to a limiting column density ($A_V \\ga 10$ mag), and avoids biases relating to the infrared properties of older protostars and other indicators of star formation potential or activity.  Furthermore, the clouds are predominantly nearby, giving a sample for which good spatial resolution can be obtained.  All clouds were imaged at $\\lambda = 850$~$\\mu$m, with a spatial resolution of $\\sim 2,000$--10,000~AU, depending on the distance.  A total of 42 dark clouds, covering an area of 0.5 square degrees, are included in the survey.  Compact submillimeter cores have been identified in the clouds, and we have established whether the cores contain embedded protostars through a combination of association with IRAS emission and/or the presence of high-velocity CO(2--1) emission from protostellar outflows.  The survey is complete for starless cores, and for Class~0 and Class~I protostars, to a mass limit of 0.015~$M_\\odot$.  Half of the clouds, 21 in total, do not contain any compact submillimeter cores, and are therefore quiescent as far as star formation activity is concerned.  The other half contain a total of 7 Class~0 protostars, 5 Class~I protostars, one Class~II source, a candidate protostar (L673$-$SMM3), and 26 starless cores.  The ratio of Class~0 to Class~I protostars found in this survey is therefore close to unity, and is not consistent with the previous result found for the $\\rho$ Ophiuchi cloud, where a ratio of 1 to 10 has been observed (Andr\\'e \\& Montmerle 1994; Motte et al.\\ 1998).  The ratio of starless cores to cores containing embedded protostars is approximately 2 to 1.  The implied lifetimes of the Class~0 and starless core phases are therefore $\\sim 2 \\times 10^5$~yr and $\\sim 8 \\times 10^5$~yr respectively.  The different ratio of Class~0 to Class~I protostars detected in our survey compared with previous work suggests star formation is highly dependent on the local environment.  Our new results provide several possibilities for the nature of low-mass star formation: {\\it (i)} Class~0 sources are at the same evolutionary stage as Class~I sources, but represent a different branch of star formation, e.g., star formation in high density environments.  However, a comparison between our study and that of Motte et al.\\ (1998) suggests the Lynds clouds are actually less dense than those in $\\rho$ Ophiuchus; {\\it (ii)} if Class~0 sources are precursors to Class~I sources, the local environment may determine the lifetime of the Class~0 phase.  For example, if the Lynds clouds are more quiescent than those in $\\rho$ Ophiuchus, they may have more time to evolve towards the singular $n \\propto r^{-2}$, which predicts a more uniform accretion rate.  However, none of the starless cores detected here has the characteristics of a singular isothermal sphere; {\\it (iii)} star formation is not steady, and relies heavily on triggering. The ratio of Class~0 to Class~I protostars in $\\rho$ Ophiuchus can then be explained by being dominated by an older, more evolved population caused by a burst of star formation some $\\sim 10^5$~yr ago.  {\\it We regard this as the most likely possibility.} Furthermore, if the lifetime of the Class~0 phase is actually similar to that of the Class~I phase, Class~0 protostars may not have such dramatically high accretion rates compared with Class~I protostars, as has previously been assumed.  The lifetime of the starless cores found in this survey is similar to the $\\sim 10^6$~yr derived from surveys using the lack of an IRAS association to define ``starless'' (e.g., Lee \\& Myers 1999).  The cores therefore last only 2--3 free-fall times before collapsing.  Temperatures of only $\\sim 10$--12~K are needed for the dominant support mechanism of the cores to be thermal pressure.  The starless cores are therefore on the verge of collapse, and it is unlikely that strong magnetic fields are important for the collapse dynamics of the cores on scales $r \\sim 10^4$~AU\\@. Further work is needed to determine the velocity structure of these cores through molecular line spectroscopy."
    },
    "0209/astro-ph0209561_arXiv.txt": {
        "abstract": "In searches for planetary transits in the field, well over half of the survey stars are typically giants or other stars that are too large to permit straightforward detection of planets.  For all-sky searches of bright $V\\la 11$ stars, the fraction is $\\sim 90\\%$.  We show that the great majority of these contaminants can be removed from the sample by analyzing their reduced proper motions (RPMs): giants have much lower RPMs than dwarfs of the same color.  We use Hipparcos data to design a RPM selection function that eliminates most evolved stars, while rejecting only 9\\% of viable transit targets.  Our method can be applied using existing or soon-to-be-released all-sky data to stars $V<12.5$ in the northern hemisphere and $V<12$ in the south.  The method degrades at fainter magnitudes, but does so gracefully.  For example, at $V=14$ it can still be used to eliminate giants redward of $V-I\\sim 0.95$, that is, the blue edge of the red giant clump. ",
        "introduction": "}  Spurred on by the detection of the transiting planetary companion to HD209458 \\citep{char00} as well as the exciting scientific results that can be extracted from intensive followup of this object \\citep{char02,cody02,hui02}, a large number of surveys are underway to detect planetary transits of stars both in clusters \\citep{str00} and the field \\citep{how00,brown99,mal01,udal02,str02}. The expected signature of a planetary transit, a $\\sim 1\\%$ drop in the stellar flux over a duration of several hours, can be mimicked by a variety of non-planetary phenomena.  These include transits by brown dwarfs or late M dwarfs, which have sizes similar to Jovian planets, grazing eclipses by ordinary stars, full eclipses by binaries that are $\\sim 100$ times fainter than the target star but lie in the same spatial resolution element, and transits of evolved stars by main-sequence stars.  While it is sometimes possible to distinguish grazing eclipses from the flatter-bottomed transits using lightcurves of sufficient photometric precision, rejection of stellar eclipses and transits usually requires several spectra: an M star companion induces radial velocity (RV) variations of $\\sim 10\\,\\kms$, whereas planetary perturbations are 10 to 100 times smaller.  Since obtaining these spectra is expensive in both telescope time and human effort, it is important to seek other robust methods of recognizing non-planetary transits.  Here we show how the great majority of evolved stars can be eliminated from the target list of transit searches using reduced proper motion (RPM) diagrams.  Field star transit surveys usually contain one to several times more evolved stars than main-sequence stars, while only the latter present useful targets. Hence, robust rejection of evolved-star contamination of transit surveys should result in substantial gains in efficiency in both the data analysis and the follow-up observations required for confirmation.  In \\S~\\ref{sec:principles}, we review the basic physical principles that allow one to recognize evolved stars from a RPM diagram. In \\S~\\ref{sec:tycho}, we assemble an an all-sky catalog of transit targets $V_T<11$.  We show that our RPM criteria reject about 60\\% of the stars in this magnitude range that remain after an initial 25\\% are already rejected because their colors are too blue.  That is, altogether 70\\% are rejected. In \\S~\\ref{sec:2mass}, we establish the criteria for rejecting evolved stars using a combination of Tycho-2 \\citep{t2} and 2MASS \\citep{2mass} data.  These criteria can be applied to stars $V\\la 12$ over the whole sky once the full 2MASS catalog is released. In \\S~\\ref{sec:ucac}, we show how the same technique can be extended to $V\\sim 12.5$ over the northern sky by combining 2MASS and UCAC \\citep{ucac} data, supplemented with data from the transit surveys themselves.  We also show that the method degrades gracefully at fainter magnitudes, and can still be used to eliminate the majority of evolved stars at $V=14$. While most of the paper focuses on the problem of selecting targets for transits of amplitude $\\sim 1\\%$, we consider the possibility in \\S~\\ref{sec:bigger} that smaller amplitudes might be reached for brighter subsamples of the main survey.  In this way, efficient transit searches might be extendable to stars of somewhat larger radius. Finally, in \\S~\\ref{sec:discuss}, we briefly discuss our results. ",
        "conclusions": "}  The RPM selection function is very efficient at removing early-type and late-type stars.  For the former, it reduces to a simple color cut that is similar to the ones already in common use.  For the latter, its robustness derives from the large gap between the giant branch and the MS. The method is relatively inefficient at excluding evolved G-type stars because here the gap is much smaller.  However, this residual G-star contamination is modest relative to the contaminants that are efficiently eliminated.  Given that the primary new gain from this method is the elimination of red giant contaminants, one might ask how difficult it would be to weed these out by other means.  Stellar transits of giants must be very finely tuned in order to mimic planetary transits.  For example, the eclipse due to a star in a 0.1 AU (12 day) orbit about a $10\\,r_\\odot$ giant would last $\\sim 1\\,$day if the inclination were anywhere near $90^\\circ$. Only if it traced a chord within $1\\%$ of the giant's limb would the eclipse be short enough to be confused with a planetary transit.  If the data were of sufficient quality, the rounded shape of the lightcurve could reveal the stellar nature of the event.  A more difficult problem is caused by triple systems containing a giant and a MS eclipsing binary.  If the binary separation is bigger than the radius of the giant, it is extremely difficult to distinguish this system from a planet transiting a dwarf star using the lightcurve alone.  Indeed, even spectroscopic observations designed to recognize the $\\sim 10\\,\\kms$ reflex motion due to an M-star companion would most likely fail to detect such a contaminant because the giant would not be markedly changing its RV.  Only a high signal-to-noise ratio spectrum would succeed in detecting the spectroscopic signature of the companion pair contributing $\\sim 1\\%$ of the light.  These difficult contaminants are easily removed using our method.  Perhaps the best aspect of our method in this respect is that it not only removes the need for followup observations of contaminant events, it removes the need to even analyze the lightcurves of the great majority of contaminating stars.  In this paper we have ignored extinction.  The effect of extinction (if it is not corrected) is to move stars along the reddening vector in the RPM diagram.  This can essentially only move stars from being rejected to accepted and not the reverse.  That is, ignoring extinction is conservative in that it adds to contamination but does not lead to missing legitimate targets.  For our primary $V_T<11$ example, shown in Figure \\ref{fig:tycrpm}, extinction is actually a very minor effect: target stars $M_{V_T}\\ga 4$ always lie $\\la 250\\,$pc, so that even in the Galactic plane the reddening is only $E(B_T-V_T)\\la 0.06$.  Hence, ignoring extinction has hardly any effect.  As the magnitude limit gets fainter, extinction becomes more important and so failure to correct for it can lead to significant increases in contamination.  We caution, however, that accurately correcting for this effect is not trivial, and that such corrections should therefore be done conservatively, i.e., by maintaining a strong bias against overestimating the extinction.  Even for fields for which the extinction at infinity is well measured (from e.g.\\  \\citealt{schlegel}), a large fraction of the dust may lie behind the target stars, which tend to be relatively close.  Thus, the 3-dimensional dust distribution must be modeled.  Moreover, for fields close to the Galactic plane, i.e., those for which the extinction correction is most important, the Schegel et al.\\ (1998) estimates for extinction at infinity are less reliable.  Hence, even more care is required in making the correction."
    },
    "0209/astro-ph0209611_arXiv.txt": {
        "abstract": " ",
        "introduction": "\\parindent=5.0mm  Our information on the dynamical parameters of the Universe describing the cosmic expansion comes from three different epochs. The earliest is the Big Bang nucleosynthesis which occurred a little over 2 minutes after the Big Bang, and which left its imprint in the abundances of the light elements affecting the baryonic density parameter $\\Omega_b$. The discovery of anisotropic temperature fluctuations in the cosmic microwave background radiation at large angular scales (CMBR) by COBE-DMR \\cite{smot}, followed by small scale anisotropies measured in the balloon flights BOOMERANG \\cite{dber} and MAXIMA \\cite{ba-ha}, by the radio telescopes Cosmic Background Imager (CBI) \\cite{pe-ma}, Very Small Array (VSA) \\cite{scot} and Degree Angular Scale Interferometer (DASI) \\cite{halv} testify about the conditions in the Universe at the time of last scattering, about 350\\thinspace000 years after Big Bang. The analyses of the CMBR power spectrum give information about every dynamical parameter, in particular $\\Omega_0$ and its components $\\Omega_b,\\ \\Omega_m$ and $\\Omega_{\\lambda}$, and the spectral index $n_s$. For an extensive review of CMBR detectors and results, see Bersanelli et al. \\cite{bersa}. Very recently, also the expected fluctuations in the CMBR polarization anisotropies has been observed by DASI \\cite{kova}.   The third epoch is the time of matter structures: galaxy clusters, galaxies and stars. Our view is limited to the redshifts we can observe which correspond to times of a few Gyr after Big Bang. This determines the Hubble constant, successfully done by the Hubble Space Telescope (HST) \\cite{free}, and the difference $\\Omega_{\\lambda}-\\Omega_m$ in the dramatic supernova Ia observations by the High-z Supernova Search Team \\cite{ries} and the Supernova Cosmology Project \\cite{perl}. The large scale structure (LSS) and its power spectrum has been studied in the SSRS2 and CfA2 galaxy surveys \\cite{daco}, in the Las Campanas Redshift Survey \\cite{shec}, in the Abell-ACO cluster survey \\cite{retz}, in the IRAS PSCz Survey \\cite{saun} and in the 2dF Galaxy Redshift Survey \\cite{peac},\\cite{coll}. Various sets of CMBR data, supernova data and LSS data have been analyzed jointly. We shall only refer to global analyses of the now most recent CMBR power spectra and large scale distributions of galaxies.  The list of other types of observations is really very long. To mention some, there have been observations on the gas fraction in X-ray clusters \\cite{evrd}, on X-ray cluster evolution \\cite{ba-ek}, on the cluster mass function and the Ly$\\alpha$ forest \\cite{wein}, on gravitational lensing \\cite{c-h-i}, on the Sunyaev-Zel'dovich effect \\cite{bi-ca}, on classical double radio sources \\cite{guer}, on galaxy peculiar velocities \\cite{zeha}, on the evolution of galaxies and star creation versus the evolution of galaxy luminosity densities \\cite{tota}.  In this review we shall cover briefly recent observations and results for the dynami\\-cal parameters $H_0,\\ \\Omega_b,\\ \\Omega_m,\\ \\Omega_{\\lambda},\\ \\Omega_0,\\ n_s,\\ w_\\lambda$ and $q_0$. In Section 2 these parameters are defined in their theoretical context, in Section 3 we turn to the Hubble parameter, and in Section 4 to the baryonic density. The other parameters are discussed in Sections 5 and 6, which are organized according to observational method: supernov\\ae\\ in Section 5, CMBR and LSS in Section 6. Section 7 summarizes our results. \\\\ ",
        "conclusions": "Information on the dynamical parameters of the Universe are coming from the Big Bang nucleosynthesis, from the fluctuations in the temperature and polarization of the cosmic microwave background radiation, from the large scale structures of galaxies, from supernova observations and from many other cosmological effects that may not yet be of interesting precision. The results of different analyses are now converging towards agreement when in the past disagreements of the order of 100\\%  have been known.  In this review we have taken the attitude that remaining disagreements reflect systematic errors coming either from the observations or from differences in the methods of analysis. We have then compiled the most precise parameter values, combined them and added our estimates of such systematic errors. This we have done for the baryonic density parameter $\\Ombh$, the density parameter of the matter component $\\Omm$, the density parameter of the cosmological constant $\\Oml$, the spectral index of scalar fluctuations $n_s$, the equation of state of the cosmological constant $w_{\\lambda}$, and the deceleration parameter $q_0$. In addition we quote the best values of the Hubble parameter $H_0$ and the total density parameter $\\Om0$ from other sources.  In Table 2 we summarize our results.  The conclusion is not new: that the Universe is spatially flat, that some 25\\%  of gravitating matter is dark and unknown, and that some 70\\%  of the total energy content is dark, possibly in the form of a cosmological constant. \\\\  \\noindent{\\bf \\it Acknowledgements:} S. M. H. is indebted to the Magnus Ehrnrooth Foundation for support."
    },
    "0209/astro-ph0209119_arXiv.txt": {
        "abstract": "{ We study the evolution of growth and decay laws for the magnetic field coherence length $\\xi$, energy $E_{\\rm M}$ and magnetic helicity $H$ in freely decaying 3D MHD turbulence. We show that with certain assumptions, self-similarity of the magnetic power spectrum alone implies that $\\xi \\sim t^{1/2}$. This in turn implies that magnetic helicity decays as $H\\sim t^{-2s}$, where $s=(\\xi_{\\rm diff}/\\xi_{H})^2$, in terms of $\\xi_{\\rm diff}$, the diffusion length scale, and $\\xi_{\\rm H}$, a length scale defined from the helicity power spectrum. The relative magnetic helicity remains constant, implying that the magnetic energy decays as $E_{\\rm M} \\sim t^{-1/2-2s}$. The parameter $s$ is inversely proportional to the magnetic Reynolds number $Re_{\\rm M}$, which is constant in the self-similar regime. ",
        "introduction": "Magnetic fields are ubiquitous in the Universe, being observed in objects from planets to galaxy clusters % (Zeldovich et al.\\ 1983, Ruzmaikin et al.\\ 1988, Kronberg 1994). In galaxies and galaxy clusters, the typical strength is of order a few $\\mu$Gauss, which is thought to be produced by dynamo action on a seed field. In galaxies the dynamo timescale is roughly a rotation period, $10^8$ yr, and a simple calculation % (Ruzmaikin et al.\\ 1988) based on the age of a typical galaxy shows that the seed field must have been about \\(10^{-20}\\) Gauss, or perhaps less in the currently favored models with a cosmological term % (Davis et al.\\ 1999).  There is no shortage of ideas for generating this seed field.  The more conventional astrophysical explanations are based on a Biermann battery operating at the era of reionization (see, e.g., % Gnedin et al.\\ 2000, and references therein).  There are more speculative ideas based on various generation mechanisms in the early Universe % (Grasso \\& Rubinstein 2001), which have the common feature of producing stochastic, homogeneous and isotropic magnetic and velocity fields, characterized by their power spectra and initial length scales. Some of these mechanisms produce stochastic fields with non-zero magnetic helicity (Joyce 1997, Cornwall 1997, Vachaspati 2001) the first of these being maximally helical. Another common feature of these generation mechanisms in the early Universe is that they last for a short time only, typically much less than the time it takes for the Universe to double in size at the time of generation, after which the magnetic fields decay.  As we are discussing mechanisms operating at the era of the electroweak phase transition ($10^{-11}\\,{\\rm s}$) or before, field is generated essentially instantaneously compared with any current astrophysical or cosmological timescale.  The subsequent decay of these primordial fields, and also of those in star-forming regions, motivates the study of  freely decaying magnetohydrodynamic (MHD) turbulence (Mac Low et al.\\ 1998, Biskamp \\& M\\\"uller 1999, M\\\"uller \\& Biskamp 2000, Christensson et al.\\ 2001). The decay will not in general just be through linear dissipation, as the fields in the early are likely to have high magnetic Reynolds number because of the high conductivity of a fully ionized relativistic plasma (we recall that $Re_{\\rm M} = \\xi v/\\eta$, where $\\xi$ and $v$ are the typical length scale and velocity of the system, and $\\eta$ the conductivity). This has been taken to mean in the cosmological context that the  magnetic field is frozen into the plasma, and the scale length of the field increases only with the expansion of the Universe. This is in general untrue, because the plasma can move, and turbulence can transfer energy to different length scales % (Brandenburg et al.\\ 1996).  Thus a study of decaying MHD turbulence is required in order to calculate quantities such as the size of the seed for the galactic dynamo or the amplitude of the perturbations in the temperature of the Cosmic Microwave Background (CMB) radiation arising from primordial magnetic field generation % (Durrer et al.\\ 1998, 2003, Caprini \\& Durrer 2001, Caprini et al.\\ 2003). Our numerical results % (Christensson et al.\\ 2001) uncovered decay laws for magnetic fields which were not those one would expect from purely linear dissipative processes. In particular we saw that helical fields decayed more slowly than non-helical, due to the fact that magnetic helicity is an invariant of ideal MHD. Helicity is known to be important in dynamo theory (Pouquet et al.\\ 1976, Meneguzzi et al.\\ 1981, Brandenburg 2001), and we shall also be able to confirm its importance in decaying turbulence. Our interest here is to try and understand the results, and to compare them with those found by Biskamp \\& M\\\"uller (1999) and M\\\"uller \\& Biskamp (2000). In doing so we have developed a new framework for understanding scaling in decaying 3D MHD turbulence, in the case where the fields are close to being maximally helical, as in the mechanisms proposed % by Joyce \\& Shaposhnikov (1997) and Vachaspati (2001). It should be noted that none of the estimates % by Durrer et al.\\ (1998, 2003), Caprini \\& Durrer (2001), and Caprini et al.\\ (2003) take into account the decay laws we find, and so our results are of direct importance for cosmology.  The decay of the magnetic field in the turbulent case is often presumed to result from an inverse cascade (Pouquet et al.\\ 1976, Meneguzzi et al.\\ 1981, Brandenburg 2001), in which power is transferred locally in $k$-space from small to large scales. However, while it is certainly true that power is transferred from small to large scales, as can been seen from the energy power spectra plotted in % Christensson et al.\\ (2001), it was not established that the power is transferred locally, despite the appearance of the term ``inverse cascade'' in the title of our paper,  and so it may be strictly incorrect to call the process a cascade. In fact, a true inverse cascade seems rather unlikely, as large scale power appears almost immediately even from initial conditions which are highly localized in $k$-space. For our analysis in this work it will not matter whether or not there is a cascade, and we will not discuss the matter further. However, we emphasize that the decay of the energy is not simple linear dissipation: there is definitely interesting non-linear physics, as shown by the appreciable Reynolds numbers (100-600) and the processing of the initial power spectra.  Various scaling arguments have been put forward to obtain the growth law for the length scale of the magnetic field and the decay law for the energy. For ideal MHD (infinite conductivity), Olesen (1997) and later Son (1999), % Field \\& Carroll (2000), % and Shiromizu (1998) % argued \\( \\xi(t) \\sim t^{2/(n+5)} \\) where $t$ is conformal time, and $n$ is the index of the initial magnetic power spectrum. Supporting evidence for this scaling law was also found in two-dimensional MHD simulations % (Galtier et al.\\ 1997). Shiromizu's results were based on a renormalization group argument, which has been revisited by Berera \\& Hochberg (2001), % who do not find evidence for an inverse cascade in the steady state. The effect of having significant helicity was supposed to modify this scaling law to \\(\\xi(t) \\sim t^{2/3}\\), \\( E_{\\rm M} \\sim t^{-2/3}\\) (Biskamp 1993, Son 1999, Field \\& Carroll 2000). Early numerical experiments with a shell model of the full MHD equations (Brandenburg et al.\\ 1996) % suggested \\(\\xi \\sim t^{0.25}\\), and gave supporting evidence to the inverse cascade. Full MHD simulations by Biskamp \\& M\\\"uller (1999) and M\\\"uller \\& Biskamp (2000) showed, in the helical case, an energy decay law \\( E_{\\rm M} \\sim t^{-1/2}, \\) supported by a phenomenological model, which we will discuss at the end of this work. Decaying non-helical turbulence was studied semi-analytically in a shell model by Basu (2000) % who found an energy decay law of $E_{\\rm M} \\sim t^{-1.2}$. The decay of helical fields was also studied semi-analytically by Sigl (2002), % giving a growth law for the length scale between $t^{1/2}$ and $t^{2/3}$. Unfortunately, direct comparison with our results is otherwise difficult because the correlation functions are expressed in real space. The importance of magnetic helicity in slowing down the decay has been recognized earlier in studies of a low-order model of three-dimensional hydromagnetic flows (Stribling \\& Matthaeus 1991), % where it was also found that a finite initial cross helicity (not studied in the present paper) can slow down the decay. However, power law exponents of the decay have not been determined for this low-order model.  In an earlier paper (Christensson et al.\\ 2001) % we performed 3D simulations both with and without magnetic helicity, starting from homogeneous and isotropic random initial conditions, with power spectra suggested by cosmological applications. We found that the coherence scale of the field grows approximately as $t^{1/2}$, with significant transfer of power to small scales in the helical case, which we ascribed to an inverse cascade. The magnetic power spectrum was self-similar with an approximately $k^{-2.5}$ behavior at high $k$. We found decay laws for the magnetic and kinetic energies of $t^{-0.7}$ and $t^{-1.1}$ in the helical case, and $t^{-1.1}$ for both in the non-helical case.  These are close to, but not identical to those found by Biskamp \\& M\\\"uller (1999), % and we suggested that their relatively large initial length scale, 25\\% of the simulation box size, might account for the difference. It should be emphasized that we are interested in the decay only while the scale length of the flows is less than the simulation volume, as we want results relevant to the early Universe where there are no boundaries.  In this paper we present a new theoretical understanding of our numerical results for the power law behavior of the length scale and the energies in the helical case. We show that the key to understanding the power laws is the self-similarity of the magnetic field, coupled with the near-invariance of the helicity, and that the crucial parameter controlling the rate of decay of the magnetic energy and the helicity is the magnetic Reynolds number.   Our theoretical model has analogies with decaying fluid turbulence in 2 dimensions, where there is also an ideal invariant, the kinetic energy, which plays a similar role to the helicity in 3D. Decaying turbulence in 2D (Ting et al.\\ 1986, Chasnov 1997) % exhibits self-similarity, and power-law decays in the kinetic energy and enstrophy (mean squared vorticity) are observed in numerical simulations at high Reynolds number (Chasnov 1997). % We reserve detailed discussion for Section~\\ref{s:TwoDTurb}. ",
        "conclusions": "\\label{s:disc}  We have studied the evolution of decaying 3D MHD turbulence involving maximally helical magnetic fields. For finite magnetic diffusivity there emerges an important quantity $s = (\\xi_{\\rm diff}/\\xi_{\\rm H})^2$, where $\\xi_{\\rm H}$ is the helicity scale defined in Eq.\\ (\\ref{length_H}), and $\\xi_{\\rm diff}$ is the diffusion scale.  We find $\\xi_{\\rm H} \\simeq vt$, where $v$ is the RMS velocity, and hence that $s \\propto Re^{-1}_{\\rm M}$, the magnetic Reynolds number evaluated using the helicity scale. The magnetic field coherence length (which can be equally well expressed as the integral, helicity or relative helicity scales) goes as $\\xi \\sim t^{1/2}$, magnetic helicity $H_{\\rm M} \\sim t^{-2s}$ and magnetic energy $E_{\\rm M} \\sim t^{-1/2 - 2s}$. A corollary is that $Re_{\\rm M}$ is constant once the system has reached self-similarity. Furthermore, we can extrapolate to the limit of very large magnetic Reynolds numbers, useful for example in the early Universe, to find $H$ constant and $E_{\\rm M} \\sim t^{-1/2}$.  Our model for the scaling laws should be compared with that of Biskamp \\& M\\\"uller (1999). % The first difference is that they assumed that the non-linear term in the evolution equation for the magnetic field was the dominant source of energy loss for the magnetic field, and that the magnetic field was asymptotically the dominant contributor to the total energy $E$, expressed as $\\Gamma \\equiv E_{\\rm V}/E_{\\rm M} \\ll 1$.  Then we can write \\ben \\dot E \\sim \\Gamma^\\half E^{3/2}/\\xi, \\een where $\\xi$ is a length scale of the magnetic field.  They then found the phenomenological relation $\\Gamma \\simeq E/H$, which, when coupled with $E \\simeq H/\\xi$ and the conservation of $H$, gives $E \\sim t^{-1/2}$.  We emphasize that this model is not inconsistent with ours. We infer $E_{\\rm V} \\sim t^{-1}$ from the relation $\\xi_{\\rm H} \\simeq vt$, and hence that $\\Gamma \\sim t^{-1/2+2s} \\sim (E/H)t^{2s}$.  The difference between Biskamp \\& M\\\"uller's assumed relation $\\Ga \\sim E/H$ and ours is small at large magnetic Reynolds numbers where $s\\to 0$. Furthermore, both approaches need to assume only that the non-linear and dissipative terms in Eq.\\ (\\ref{energy_loss}) are not sub-dominant (rather than dominant) and it turns out that both scale with time in the same way, as $E_{\\rm M}/t$.  In our simulations dissipation typically accounted for about 60\\% of the energy loss in the period $t\\simeq 40$ to $t \\simeq 80$, which means that the field is not force-free.  We believe that our model has certain advantages, in that the assumptions going into it give more physical insights than the phenomenological (and dimensionally incomplete) relation $\\Ga \\sim E/H$. Our assumptions are that the magnetic power spectrum exhibits a self-similar form (\\ref{sca_law}), with power-law behavior $k^{-z}$ at high $k$, that resistive dissipation occurs predominantly at the diffusion scale $\\xi_{\\rm diff}$, that there is a separate helicity scale $\\xi_{\\rm H}$, from which it follows that $2<z<3$.  We also assume that the eddy turn-over time $\\xi/v$ is $t$, and that the relative helicity is asymptotically constant, i.e.\\ that the energy decays as fast as possible as is consistent with helicity conservation. Given these assumptions, it follows that the dissipative and non-linear terms in the magnetic energy loss equation (\\ref{energy_loss}) scale the same way, and we can infer that $\\xi \\sim t^{1/2}$.  From this we derive a scaling law for $H$, finding as expected that $H$ is conserved in the limit of large magnetic Reynolds number, and with the assumption of constant relative helicity, we obtain that $E \\sim t^{-1/2}$ in the same limit.  In summary, for both 3D helical MHD and 2D fluids, the key to understanding the free decay of turbulence is self-similarity, coupled to a separation between the scale of the flow and the diffusion scale.  \\textit{Note added.\\ } Since the first draft of this work appeared as astro-ph/0209119,v1, there have been several developments. A new review of magnetic fields in cosmology has appeared (Giovannini 2004). % Using different scaling arguments, Campanelli (2004) % has rederived our scaling laws for decaying helical turbulence, and extended the analysis to the non-helical case, where the ideal limit has $E_{\\rm M} \\propto t^{-1}$, $E_{\\rm V} \\propto t^{-1}$, and $\\xi \\propto t^{1/2}$. Extensive 3D numerical simulations, including the interesting case of Prandtl number larger than unity, have been carried out by Banerjee \\& Jedamzik (2003, 2004). The authors did not seem to be aware of our work or of Campanelli (2004) % and did not attempt to analyse  the decay of the helicity, kinetic or magnetic energies in its light.  However, examination of their Figs.\\ 1 and 7 in the later paper show decay laws close to $t^{-0.5}$ and $t^{-1}$ for the helical and non-helical cases respectively (with Prandtl number unity). They do not quote Reynolds numbers for the simulations so an exact comparison cannot be made. There has also been further work on effects of primordial fields on the CMB (Kosowsky et al.\\ 2005, Kahniashvili \\& Kahniashvili 2005) and on the evolution of magnetic fields in the post-recombination era (Sethi \\& Subramanian 2004). %"
    },
    "0209/astro-ph0209433_arXiv.txt": {
        "abstract": "We study the merging of star clusters out of cluster aggregates similar to Knot S in the Antennae on orbits close to the one of $\\omega$ Cen by carrying out high resolution numerical N-body simulations. We want to constrain the parameter space which is able to produce merger objects with similar properties as $\\omega$ Cen. ",
        "introduction": "\\label{sec:intro}  Interactions of gas-rich disk galaxies show intense bursts of star formation.  For example HST-images of the Antennae (Whitmore et al.\\ 1999) reveal that the knots of intense star formation produce clusters of massive young star clusters.  These aggregates which we call super-clusters ($=$ cluster of star clusters; not to confuse with super stellar cluster (SSC), which are individual massive star clusters) appear to contain dozens to hundreds of massive star clusters within a region spanning only a few hundred pc to a kpc in radius.  On the other hand new observations of the central galaxy of the Fornax cluster revealed a new class of unresolved, compact objects (Hilker et al.\\ 1999, Phillipps et al.\\ 2001) with radii of a few hundred pc, which are called ultra compact dwarf galaxies (UCD).  Also in a few lenticular field galaxies (e.g.\\ NGC~1023) Larson and Brodie (2000, 2002) found star clusters with extremely large effective radii ($r_{\\rm eff} > 7$~pc) which they call faint fuzzies.  Here we show N-body results concerning the dynamical evolution of such super-cluster aggregates.  All simulations of super-clusters show a strong merging behaviour building up compact merger objects in few super-cluster crossing times (Fellhauer et al.\\ 2002).  Depending on the initial conditions of our simulations (strong or weak tidal field; massive or extended low-mass super-cluster) our resulting merger objects have similar properties like the new classes of objects above (Fellhauer \\& Kroupa 2002a/b).  But placing compact and massive super-clusters in strong tidal fields on an orbit similar to $\\omega$-Cen reveals an object which has similar properties like the most massive globular cluster (GC) in the Milky Way. $\\omega$-Cen is not only the most massive GC, it has also some strange properties like different populations of stars (different ages and metalicities).  It shows signs of rotation with a maximum rotation speed of $8$~km/s (Freeman 2001). ",
        "conclusions": ""
    },
    "0209/astro-ph0209605_arXiv.txt": {
        "abstract": "s{ The possibility of natural and abundant creation of antimatter in the Universe in a SUSY-baryogenesis model with a scalar field condensate is described. This scenario predicts  vast quantities of antimatter, corresponding to galaxy and galaxy cluster scales today,  separated from the matter ones by baryonically empty voids. Theoretical and observational constraints on such antimatter regions are discussed. } ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209575_arXiv.txt": {
        "abstract": "{We consider the transverse structure and stability properties of relativistic jets formed in the course of the collapse of a massive progenitor. Our numerical simulations show the presence of a strong shear in the bulk velocity of such jets. This shear can be responsible for a very rapid shear--driven instability that arises for any velocity profile. This conclusion has been confirmed both by numerical simulations and theoretical analysis. The instability leads to rapid fluctuations of the main hydrodynamical parameters (density, pressure, Lorentz factor, etc.). However, the perturbations of the density are effectively decoupled from those of the pressure because the beam of the jet is radiation--dominated. The characteristic growth time of instability is much shorter than the life time of the jet and, therefore, may lead to a complete turbulent beam.  In the course of the non-linear evolution, these fluctuations may yield to internal shocks which can be randomly distributed in the jet. In the case that internal shocks in a ultrarelativistic outflow are responsible for the observed phenomenology of gamma-ray bursts, the proposed instability can well account for the short-term variability of gamma-ray light curves down to milliseconds. ",
        "introduction": "Catastrophic stellar events like massive stellar core collapse (see, e.g., van Putten \\cite{Putten01}, M\\'esz\\'aros \\cite{Meszaros01} and references therein) or merging of a binary neutron star (Paszynski \\cite{Paszynski86}, Eichler et al. \\cite{Eichleretal89}, Mochkovitch et al. \\cite{Mochkovitch93}) have been proposed to explain the energetics of gamma-ray bursts (GRB).  Nowadays, an increasing amount of observational evidence stresses the association, at least, of some GRB events with massive progenitors (Bulik, Belczynski \\& Zbijewski \\cite{BBZ99}; Bloom et al. \\cite{Bloometal99}; Hanlon et al. \\cite{Hanlonetal00}; Reeves et al. \\cite{Reevesetal02}).  One of the most attractive scenarios is the collapsar model (Woosley \\cite{Woosley93}, MacFadyen \\& Woosley \\cite{MW99}) which can produce the required energy of a GRB even at cosmological distances. This model involves the collapse of the central core of a massive, evolved star to a newly-formed black hole (BH).  The progenitor star can be, for instance, a rotating Wolf-Rayet star (\\cite{Woosley93}). Hydrodynamic simulations of collapsars have been performed for a 35$M_{\\odot}$ main-sequence star whose 14$M_{\\odot}$ helium core collapses to form 2-4$M_{\\odot}$ black hole. Provided that the core has a sufficient amount of angular momentum, a massive geometrically thick accretion disc of several tenths of a solar mass can be formed around the BH. The BH accretes matter from the disc at rates of the order of solar masses per second. The burst is generated by a local energy deposition due to the annihilation of $\\nu-\\bar{\\nu}$ coming from the accretion disk or/and due to the release of the BH spin energy by means of magnetic fields. In either case, the energy is preferentially released along the rotation axis, close to the central engine and gives rise to a relativistically expanding bubble of radiation and pairs. The duration of the burst is given by the hydrodynamic time scale for the core of the star and its Helium envelope to collapse or by the viscous evolution time for the accretion disc, whichever is greater. In order to produce a long-duration GRB with a complex pulse structure, accretion is argued to proceed over a period of time comparable to the prompt phase of a GRB. The energy released due to accretion is sufficient to drive a collimated, baryon-dilute fireball that penetrates the outer layers of the star forming a relativistic jet (Aloy et al. \\cite{Aetal00}).  Recent observations indicate that the rapid temporal decay of several GRB afterglows is consistent with the evolution of a highly relativistic jet after it slows down and spreads laterally rather than with a spherical blast wave (Sari, Piran \\& Halpern \\cite{SPH99}, Halpern et al. \\cite{Halpernetal99}, Kulkarni at al. \\cite{Kulkarnietal99}, Rhoads \\cite{Rhoads99}). Therefore, formation of relativistic jets of baryon-clean material with bulk Lorentz factors $\\sim 10^{2}- 10^{3}$ represents a mayor problem in the collapsar model of GRB. Detailed simulations of formation and propagation of a relativistic jet in collapsars have been performed by Aloy et al. (\\cite{Aetal00}). Assuming an enhanced efficiency of energy deposition in polar regions (with a constant or varying deposition rate), the authors obtained an ultrarelativistic jet along the rotation axis, which is highly focused (with the half-opening angle $\\sim 6-10^{\\circ}$) and is capable of penetrating the star. The simulations were performed with the multidimensional relativistic hydrodynamic code GENESIS (Aloy et al. \\cite{Aetal99}) using a two-dimensional spherical grid. A relativistic jet in this model forms within a fraction of a second and exhibits the main morphological elements of ``standard'' jets: a terminal bow shock, a narrow cocoon, a contact discontinuity, and a hot spot. The maximum Lorentz factor is as large as $\\sim 25-30$ at break-out, and can be even greater after the break-out ($\\approx 44$ at the end of simulations). This Lorentz factor is only a bit smaller than the critical value $\\sim 10^{2}-10^{3}$ that one requires for the fireball model (Cavallo \\& Rees \\cite{CR78}, Piran \\cite{Piran99}).  According to current views, the GRB is made either as the jet encounters a sufficient amount of mass in circumstellar matter or by internal shocks in the jet (Rees \\& M\\'esz\\'aros \\cite{RM92}, \\cite{RM94}, Daigne \\& Mockkovitch \\cite{DM98}, \\cite{DM00}). Therefore, the properties of jets from collapsars are of crucial importance for understanding the mechanism of GRBs. For example, the non-uniformity across the jet can influence the structure of internal shocks and, hence, the gamma-ray emission from these shocks. A transverse gradient can also drastically change the temporal decay rate of afterglows, making the decay flatter or steeper depending on the transverse structure (M\\'esz\\'aros, Rees \\& Wijers \\cite{MRW98}). Gamma-ray light curves also show a great diversity of time dependences ranging from a smooth rise and quasi-exponential decay, through light curves with several peaks, to variable light curves with many peaks, and substructure sometimes down to milliseconds. Hydrodynamic instabilities arising in the jet can be responsible for the fine time structure of GRBs in a row with a temporal variability caused by accretion onto the BH and the circumstellar interaction. This complex time dependence can provide clues for the understanding of the geometry and physics of the emitting regions. Apart from this, instabilities and their associated fluctuating hydrodynamic motions can lead to the generation of magnetic fields in a highly conductive $e^{-} e^{+}$ plasma. In the presence of turbulent magnetic fields, the electrons can produce a synchrotron radiation spectrum (M\\'esz\\'aros \\& Rees \\cite{MR93}, Rees \\& M\\'esz\\'aros \\cite{RM94}) similar to that observed (Band et al. \\cite{Bandetal93}). The inverse Compton scattering of these synchrotron photons can extend the spectrum even into the GeV range (M\\'esz\\'aros, Rees \\& Papathanasiou \\cite{MRP94}).  In the present paper, we consider the stability properties of jets which are formed and propagate in collapsars. Large--scale hydrodynamic instabilities can be the reason for the observed morphological complexity of ``standard'' jets, and this has motivated many analytical (see, e.g., Birkinshaw \\cite{Birkinshaw84}, \\cite{Birkinshaw91}, \\cite{Birkinshaw97}, Hardee \\& Norman \\cite{HN88}, Zhao et al. \\cite{Zhaoetal92}, Hanasz \\& Sol \\cite{HS96}) and numerical (Hardee et al. \\cite{Hardeeetal92}, Hardee et al. \\cite{Hardeeetal98}, Bodo et al. \\cite{Bodoetal98}, Micono et al. \\cite{Miconoetal00}, Agudo et al. \\cite{Agudoetal01}) studies of the stability properties of jets.  Usually, the jet is considered as a beam of gas with one bulk velocity and constant density surrounded by a very narrow shear layer separating it from the external medium. One possible mechanism of destabilizing such jets is often attributed to the well-known Kelvin-Helmholtz instability which, in its classical formulation, is the instability of a tangential discontinuity between two flows, generally having different density (see, e.g., Landau \\& Lifshitz \\cite{LL78}, Chandrasekhar \\cite{Chandrasekhar81}). However, jets in collapsars show a complex structure with the presence of strong transverse shear and substantial density stratification (Aloy et al. \\cite{Aetal00}). Obviously, the stability properties of such sheared, stratified jets may well be different from those of jets with constant bulk velocity and density.  The outline of this paper is as follows. In Section 2, we represent the results of numerical calculations and discuss the transverse structure of a jet originating in a collapsar. In Section 3, we represent the linear stability analysis of a sheared, stratified jet by making use of a WKB-approximation. Finally, in Section 4, we discuss the possible role of the instability in the evolution of jets. ",
        "conclusions": "We have treated numerically and analytically the instability that can arise in jets from collapsars. The instability is caused by a combined action of shear, which is unavoidable in such jets, and an extremely high compressibility associated with a relativistic sound speed. It turns out that jets from collapsars are more unstable than, for example, standard extragalactic jets because of the relativistic compressibility. The fact of instability itself does not depend on a particular shape of the velocity profile: the instability can arise for any dependence $V(\\theta)$. However, the growth time is shorter for flows with a stronger shear. Note that only non-homogeneous perturbations in the radial direction ($k \\neq 0$) can be unstable.  {\\bf Although it is commonly believed that three dimensional studies of the stability of relativistic jets will include additional instable modes (in many cases even more unstable that the axisymmetric ones), the recent work of Hardee \\& Rosen \\cite{HR02}, has pointed the fact that shear leads both to an enhancement of the axisymmetric modes and a suppression of the asymmetric modes. Hence, a fully three dimensional study would not yield to radically different results from the ones that we obtain here.}  In the main fraction of the jet volume, the instability grows very rapidly. If we estimate the average Lorentz factor as $\\sim 10$ and the half-opening angle as $\\sim 5^{\\circ}$ then the growth time (\\ref{eq:1overtauestimate}) is of the order $0.01 r/c$. This is much shorter than the life time of the jet even at $r \\sim R$. Therefore, we can expect that shortly after jet formation the instability will generate well developed turbulent motions with substantial fluctuations of the Lorentz factor, density, pressure, etc. This conclusion is in good agreement with the results of numerical simulations (see Figs.~\\ref{fig:lorentz} and \\ref{fig:density}). During the jet's propagation, fluctuations can become noticeable at a relatively early evolutionary stage and, hence, at a small distance from the formation region because the growth time is sufficiently short even in the inner region of the collapsar. Of course, the initial amplitude of fluctuations in the jet is quite uncertain.  In our case, the initial fluctuations are triggered by numerical reasons, but it is quite likely that they mimic the irregularities of the process of accretion onto the central black hole.  Usually the time $\\tau_{0}$ required for an instability to amplify the amplitude of perturbations to a noticeable value is longer than the growth time, $\\tau$, by some factor which is typically $\\sim (5-10)$, so $\\tau_{0} \\sim (5-10) \\tau$. After this time, the fluctuations have grown by a factor $\\sim 10^{2}-10^{4}$ compared to their initial value which probably is sufficient to become significant. In the region where $\\tau_{0}$ is shorter than the propagation time scale, which can be estimated as $\\sim r/c$, fluctuations reach noticeable values. On the contrary, fluctuations seem to be insignificant in regions where $\\tau_{0} > r/c$. We can define the radius where the instability starts to manifest itself as that where the condition $\\tau_{0} \\sim r/c$ is fulfilled. Substituting $\\tau_{0}$, we obtain that \\begin{equation} (5-10) \\frac{\\theta_{0}}{\\Gamma_{a}} \\sim 1 \\xi \\end{equation} at this radius. At small $r$ ($\\sim 8 \\times 10^{8}\\,$cm well below jet break-out), the jet is less collimated ($\\theta_{0} \\approx 10-12^{\\circ}$) than on average (see Aloy et al. \\cite{Aetal00}). Therefore, the instability should manifest already in regions where $\\Gamma \\sim 1-2$.  This conclusion is in qualitative agreement with what is shown in Fig.~\\ref{fig:lorentz} where appreciable fluctuations appear in regions with a Lorentz factor between 1 and 2.  Note that the radial wave vector of unstable perturbations, $k$, should satisfy rather restrictive conditions. First, $k$ has to be sufficiently large for the applicability of the local approximation (see equation (\\ref{eq:kr})).  On the other hand, condition (\\ref{eq:1overtau}) which is necessary for the existence of the turning point, implies that \\begin{equation} \\Gamma_{a} \\frac{|V'_{a}|}{V} \\sim \\frac{\\Gamma_{a}}{x_{0}} \\gg k. \\label{eq:GV'ggk} \\end{equation} If the inequality (\\ref{eq:GV'ggk}) holds then the condition (\\ref{eq:sigmaapprox}) used in our calculations is certainly fulfilled and the parameter $\\xi$ is small.  Combining equations (\\ref{eq:kr}) and (\\ref{eq:GV'ggk}) and taking into account that $x_{0}= r \\theta_{0}$, we obtain that \\begin{equation} \\frac{\\Gamma_{a}}{\\theta_{0}} \\gg kr \\gg 1 \\end{equation} for unstable perturbations. Estimating $\\Gamma \\sim 10$ and $\\theta_{0} \\sim 5-6^{\\circ}$ in the main fraction of the jet volume, we have $100 \\gg kr \\gg 1$. Numerical simulations (see Fig.~\\ref{fig:lorentz}) indicate that the characteristic length scale of fluctuations depends on $r$ and varies within the range from $0.1r$ to $0.5r$ in good agreement with the prediction of the theory. Note that fluctuations with a shorter length scale cannot be resolved with the computational grid used in the simulations.  All this allows one to speculate that the calculated fluctuations of parameters within the jet are physical (i.e., not simply numerical artifacts) and reflect the presence of a very strong shear-driven instability. We can expect that inhomogeneities caused by this instability will produce shocks in the course of their non-linear evolution when faster fluctuations try to overtake slower ones or when fluctuations moving in the positive and negative radial directions collide.  If this is the case then internal shocks might be more or less randomly distributed and oriented within the jet forming filamentary structures. It is often supposed that shocks in a ultrarelativistic wind or jet are responsible for GRBs themselves whereas the impact against the ambient matter of this wind produces an external shock which likely produces the observed afterglows (Rees \\& M\\'esz\\'aros \\cite{RM92}, \\cite{RM94}).  Shocks can convert a portion of kinetic energy into a non-thermal gamma/X-ray transient emission which is usually ascribed to particle acceleration by shocks. Typically, the efficiency of this conversion is not high, $\\sim 1-2$ \\%, but it can be much greater ($\\sim 20-40$ \\%) for the interaction of fluctuations with very different Lorentz factors (Kobayashi, Piran \\& Sari \\cite{KPS97}, Kobayashi \\& Sari \\cite{KS01}). Since in our model the calculated fluctuations move with substantially different Lorentz factors we can expect a highly efficient transformation of their kinetic energy into radiation.  The proposed instability can also accounts for the rapid variability of the gamma-ray light curves, which lasting from tens to hundreds of seconds, exhibit variability sometimes down to milliseconds (Fishman \\& Meegan \\cite{FM95}). Likely, the most rapid temporal variability associated with the shear-driven instability has a time scale of the order of the growth time (\\ref{eq:1overtauestimate}). At the surface of a collapsar, for example, this time scale is as short as $\\sim 10^{-3}$s. Since the jet is highly inhomogeneous and the Lorentz factor varies strongly during the jet's propagation, a slower variability could also be represented in the gamma-ray light curves.  Another remarkable inference from the considered model is that the turbulent motions caused by the instability may also be important for the electron-proton energy exchange and, particularly, for the generation of the magnetic field in jets from collapsars."
    },
    "0209/astro-ph0209096_arXiv.txt": {
        "abstract": "We present {\\xmm} data of three strongly magnetic cataclysmic variables (polars) EV UMa, RX J1002-19 and RX J1007-20. These include the polar with the shortest orbital period (EV UMa) and the polar with one of the highest magnetic field strengths (RX~J1007--20). They exhibit a range of X-ray spectral characteristics which are consistent with their known magnetic field strength. We find that two of the systems show evidence for an absorption dip in soft X-rays. Their profiles are well defined, implying that the stream is highly collimated. We determine the mass transfer rate for the two systems with known distances.  We determine that the mass of the white dwarf in EV UMa and RX J1007-20 is $\\sim$1\\Msun while in RX J1002-19 it is closer to $\\sim$0.5\\Msun. ",
        "introduction": "Polars or AM Her systems are accreting binary systems in which material transfers from a dwarf secondary star onto a magnetic ($B\\sim$10--200MG) white dwarf through Roche lobe overflow. At some height above the photosphere of the white dwarf a shock forms. Hard X-rays are generated in this post-shock flow ({\\it c.f.} Wu 2000 for a review of the shock structure in these systems). Cyclotron radiation is also emitted in this post-shock flow by electrons spiralling around the magnetic field lines: this radiation is emitted in the optical band. Some fraction of the hard X-rays and cyclotron emission are intercepted by the photosphere of the white dwarf and are re-emitted at lower energies. Soft X-rays can also be produced by dense `blobs' of material which impact directly into the white dwarf.  As part of a programme to determine how the balance of soft and hard X-rays are affected by system parameters such as the magnetic field, we have observed a number of polars using {\\xmm}. These include DP Leo, WW Hor (Ramsay et al 2001), BY Cam (Ramsay \\& Cropper 2002a) and CE Gru (Ramsay \\& Cropper 2002b). Here, we present the results on three further polars, EV UMa, RX~J1002--19 and RX~J1007--20, all of which were discovered using {\\ros}. We show their main system parameters in Table \\ref{parameters}. These systems show a range of parameters, including the polar with the shortest orbital period (EV UMa), and a polar with one of the highest magnetic field strengths (RX~J1007-20). In this paper we first discuss their temporal properties and then their spectral properties.  EV UMa (RE~J1307+535) also has exhibited the highest recorded degree of polarisation in any polar (or any astrophysical object for that matter) when the circular polarisation varied from +50 to --20 percent over an orbital cycle when it was in an intermediate accretion state (Hakala et al 1994). The degree of polarisation was reduced when it was at higher accretion states (presumably because of the increased dilution of the polarisation by a bright accretion stream) and the system has a significantly different light curve (Katajainen et al 2000). Although this object has been relatively well observed in the optical and also the EUV (using the Wide Field Camera on {\\ros}, Osborne et al 1994) no X-ray observations of this object have been published.  RX J1007--20 was shown to show a prominent dip in the soft X-ray light curve, which was attributed to the accretion stream obscuring the emission from one of the accretion sites and also has a very high soft/hard X-ray ratio (Reinsch et al 1999).  Very little information has been published on RX J1002~--19.   \\begin{table} \\begin{center} \\begin{tabular}{rrr} \\hline Source & Orbital Period & Magnetic field \\\\ \\hline EV UMa & 79.69m$^{1}$ & 30--40 MG$^{1}$\\\\ RX J1002-19 & 107m$^{2}$ & \\\\ RX J1007-20 & 208m$^{2}$ & 92MG$^{3}$\\\\ \\hline \\end{tabular} \\end{center} \\caption{The main system parameters of EV UMa, RX J1002-19 and RX J1007-20. (1) Osborne et al 1994, (2) Beuermann \\& Burwitz 1995, (3) Reinsch et al 1999.} \\label{parameters} \\end{table} ",
        "conclusions": "We have presented {\\xmm} data of three polars, which display rather different characteristics. We find that in each system we cannot obtain good fits to their X-ray spectra using a single component model -- they all require a shock model plus a soft blackbody component. However, all three show very different soft/hard X-ray ratios. Further, their light curves show a range of features. We now go on to discuss our findings.  \\subsection{EV UMa - one or two accretion poles?}  The X-ray light curves and the hardness ratio of EV UMa (Figure \\ref{lightev}) suggested a different accretion region may dominate before and after the absorption dip. By examining the X-ray spectra from these phase ranges we find that there is no evidence that their spectra differ. Together with the phasing of the X-ray light curves and the optical light curves (\\S \\ref{evphase}) we suggest that there is only one dominant accretion region and the presence of negative circular polarisation for a short time in the optical data of Hakala et al (1994) is due to observing the cyclotron emitting region from beneath.  \\subsection{Soft/Hard ratio}  EV UMa, RX J1002-19 and RX J1007-20 show very different $L_{soft}/L_{hard}$ ratios, with EV UMa showing the lowest and RX J1007--20 the highest. Indeed, RX J1007--20 has a ratio which is a factor of $\\sim$20 greater than that predicted from the `standard' shock model of Lamb \\& Masters (1979) and King \\& Lasota (1979). However, the fact that it has a high magnetic field strength (92 MG, Reinsch et al 1999), a large `soft X-ray excess' is consistent with the work of Ramsay et al (1994) and Beuermann \\& Burwitz (1995) who showed that this ratio was correlated with magnetic field strength; with high field systems showing high soft/hard ratios. To account for this, the most likely solution is that dense blobs of material impact with the white dwarf directly without forming a shock (Kuijpers \\& Pringle 1982) or the shock is buried sufficiently deep for the bremsstrahlung emission to be thermalized in the local photosphere of the white dwarf (Frank, King \\& Lasota 1988). The fact that we observe strong flaring activity in the light curve of RX J1007--20 is consistent with this view.  Based on the soft/hard ratio we predict that RX J1002-19 will have a magnetic field strength which is slightly greater than that of EV UMa. If we observe the accretion region(s) of EV UMa at a high viewing angle then its soft/hard ratio is consistent with the standard shock model.  Ramsay et al (2001) showed that using a single temperature thermal bremsstrahlung model for the hard X-ray component rather than a stratified shock as used here, the resulting ratio $L_{soft}/L{hard}$ is higher: in the case of DP Leo by a factor of 2. This should be taken into account when comparing with studies using single temperature models.  \\subsection{Masses}  We can infer the mass of the white dwarf from our model fitting assuming a mass-radius relationship for the white dwarf. Using the Nauenberg (1972) relationship we find that both EV UMa and RX J1007--20 have masses of $\\sim1$\\Msun. In the case of RX J1002--19 the mass is not very well constrained, although it is likely to be less than 1\\Msun with a best fit of 0.5\\Msun.  Using the same model for the shock region as here, Ramsay (2000) fitted {\\sl RXTE} spectra from 21 mCVs and found that their masses were biased towards higher masses than that of isolated white dwarfs. There was no significant difference between the mass of the white dwarf in mCVs and non-magnetic CVs. This is consistent with the masses reported here and also the other mCVs which have been observed using {\\xmm}: CE Gru ($\\sim$1.0\\Msun, Ramsay \\& Cropper 2002b), BY Cam (0.9--1.1\\Msun, Ramsay \\& Cropper 2002a), DP Leo ($>$1.3\\Msun, Ramsay et al 2001) and WW Hor ($\\sim$1.0--1.1\\Msun, Ramsay et al 2001). It remains to be seen whether there is a systematic bias in the masses determined with this method using {\\xmm} data.  \\subsection{Mass transfer rate} \\label{transfer}  To determine the mass transfer rate we use the standard relation, $L_{acc}=GM_{wd}\\dot{M}/R_{wd}$. The accretion luminosity, $L_{acc}$, is the sum of $L_{hard}$, the unreprocessed fraction of the cyclotron luminosity, $L_{cyc}$, and the luminosity of any dense blobs of material which do not form a shock and emit in soft X-rays. We use the best fit masses determined from our model fits to determine the mass accretion rate for EV UMa and RX J1007--20 (we omit RX J1002--19 since there is no estimate for its distance).  In the case of EV UMa, the results of our spectral analysis indicate there is no evidence for blobby accretion. Using the model results of Woelk \\& Beuermann (1996) and assuming a magnetic field strength of (30-40MG) and a mass $M_{wd}$=1.0\\Msun, we find $L_{cyc}/L_{hard}\\sim$1--10 for a range of $\\dot{m}$=1--10 g cm$^{-2}$ s$^{-1}$. This implies $L_{cyc}=7.5-70\\times10^{32}$ \\ergss. The mass transfer rate is therefore $\\sim6-30\\times10^{15}$ g s$^{-1}$ and hence the fraction of the white dwarf that is accreting is $\\sim1.6-8.3\\times10^{-3}$.  In the case of RX J1007--20 the spectral results indicate that a significant proportion of the soft X-ray luminosity is in the form of blobby accretion. We therefore include $L_{soft}$ when determining the total luminosity. We again use the model results of Woelk \\& Beuermann (1996) to estimate $L_{cyc}$ and assume $B=$92MG and $M_{wd}$=1.0\\Msun. We find that $L_{cyc}\\sim10-100\\times (L_{hard}+L_{soft})$ and hence $L_{cyc}\\sim4-50\\times10^{33}$ \\ergss. This implies $\\dot{M}\\sim2-20\\times10^{16}$ g s$^{-1}$ and a fractional area of $\\sim5-60\\times10^{-3}$.  It is difficult to compare these mass transfer rates with previous results because of the different way authors account for the cyclotron flux and whether they include the soft X-ray luminosity. However, the fractional area of the white dwarf which is accreting is consistent with previous studies.  \\subsection{Absorption dips}  Absorption dips of the kind seen in EV UMa and RX J1002--19 are caused by the accretion stream obscuring the emission sites on the white dwarf as it passes through our line of sight. For an accretion region in the upper hemisphere of the white dwarf, such a dip is inevitable if the latitude of the region, $m$, is less than the binary inclination, $i$. In the case of EV UMa, Hakala et al (1994) finds the optical polarisation data is best modelled with a high inclination ($\\sim75^{\\circ}$) so it is likely that this condition holds. The inclination for J1002--19 is currently unknown. The dip ingress and egress in EV UMa is rapid: the ingress takes 20--30 sec and less than 20 sec for the egress (Figure \\ref{dip}). Such a rapid ingress and egress imply that the stream is highly collimated. For RX J1002--19 the dip structure is more complex (Figure \\ref{dip}) with a shorter dip preceding the main dip which also has a sharp profile.  The dip duration (defined as the full width half maximum of the dip profile) is 0.057 and 0.042 cycles for EV UMa and RX J1002--19 respectively (where we do not include the shorter dip). Using equation (14) of Watson et al (1989) we find that for the observed duration of the dips, $r_{s}=0.18d$ and 0.13$d$ for EV UMa and RX J1002--19 respectively where $r_{s}$ is the radius of the accretion stream and $d$ is the distance between the white dwarf and the source which is causing the absorption dip.  (We assume the stream is in the orbital plane). For a stream radius of 10$^9$ cm these radii imply $d=5.5\\times10^9$ and 7.7$\\times10^9$ cm.  To make a very approximate estimate of the radius at which the accretion stream becomes coupled by the magnetic field, $R_{\\mu}$, we use equation (1b) of Mukai (1988). If we use the best fit masses, the mass accretion rates determined in \\S \\ref{transfer}, the radius of the accretion stream, $r=1\\times10^{9}$ cm, and $B$=30MG we derive $R_{\\mu}=1.1\\times10^{10}$ and 3.5$\\times10^{10}$ cm for EV UMa and RX J1002--19. Although there is a great deal of uncertainty in some of the values of these parameters and the applicability of equation (1b) of Mukai (1988), it does suggest that the stream which is obscuring the accretion region during the dip is located in the magnetically controlled portion of the accretion flow.  We can make an estimate of the total column density of the stream by taking our best model fits then increasing the absorption until the model count rate matches the observed count rate at dip maximum for the specified energy range. We find a column density of 7--20$\\times10^{22}$ \\pcmsq and $\\sim1\\times10^{21}$ \\pcmsq for EV UMa and RX J1002--19 respectively. We can make an estimate of the number density of the stream using equation (6) of Watson et al (1995). Assuming the accretion flow is in the orbital plane and using the above values of the total column density we find $n_{13} r_{9}$=5 and 0.05 for EV UMa and RX J1002--19 respectively, where $n_{13}$ is the constant number density in units of 10$^{13}$ cm$^{-3}$ and $r_{9}$ is the radius of the stream in units of 10$^{9}$ cm. For comparison Watson et al (1995) found $n_{13}r_{9}$=5 for RX J1940-10. Taking these values at face value, they imply that either the accretion stream in RX J1002--19 has a rather small radius or a low number density.  \\begin{figure} \\begin{center} \\setlength{\\unitlength}{1cm} \\begin{picture}(8,10) \\put(-0.5,-1.4){\\special{psfile=dip.ps hscale=45 vscale=45}} \\end{picture} \\end{center} \\caption{The EPIC pn X-ray light curves (0.15--1.0keV) of EV UMa and RX J1002-19 showing the accretion stream dip in detail. The integration time is 10 and 20 sec for EV UMa and RX J1002-19 respectively. We have phased the light curves so that phase 0.0 corresponds to the centre of the dip.} \\label{dip} \\end{figure}"
    },
    "0209/astro-ph0209269_arXiv.txt": {
        "abstract": "The formation of the toroidal and jet-like structures  in the central part of the Crab Nebula is explained in the  framework of Kennel \\& Coroniti theory. The only new element introduced by us in this theory is the initial anisotropy of the energy flux in the wind. We estimate the   X-ray surface brightness of the Crab Nebula from the region of interaction of this wind with the interstellar medium and compare it with observations. ",
        "introduction": "The Crab Nebula is powered by the wind of a relativistic $e^{\\pm}$ plasma from pulsar PSR 0531+21. The wind is terminated by a shock front. The particles of the wind are redistributed in energy and their motion is randomised at the shock. Downstream of the shock (in the nebula) they emit synchrotron and inverse Compton radiation \\citep{kennel,dejager,aharonian}. Detection of these emissions is still the only way to obtain information about the wind (see however \\cite{bogah}). Observations in the X-ray \\citep{brinkman,weisskopf} and optical \\citep{hester}  have revealed a remarkable torus as well as jet-like structures in the central part of the Crab Nebula. The mechanism which produces these structures apparently gives rise to similar features observed around the Vela pulsar \\citep{pavlov00,pavlov01,helfand}, PSR 1509-58 \\citep{kaspi} and in  the supernova remnants G0.9+1 \\citep{gaensler} and G54.1+0.3 \\citep{lu}. The understanding of this mechanism will certainly give us new information about pulsar winds.  The integral characteristics of the Crab Nebula are  described by the theory of \\cite{kennel}. This theory   explains well the spectra and luminosity of the Crab Nebula in photon energy range from eV up to TeV gamma-rays \\citep{aharonian}. However, \\cite{kennel} strongly simplified the physics of the nebula. They assumed that  pulsar winds are isotropic. Therefore, this theory in it's original form is not able in principle to explain nonuniform structures observed in the Crab Nebula.  Analysis shows that magnetic collimation of the pulsar winds into jets is impossible in conventional theories of the pulsar winds \\citep{begelmanli,beskin,bogts}. Therefore, it is very difficult to interpret the observed jets as the result of collimation of the pulsar winds \\citep{lubech}.  The observation of the torus leads to the natural conclusion that the  acceleration of the wind basically occurs near  the equatorial plane \\citep{asbr}. It was shown recently  \\citep{bogkh}(hereafter Paper I) that the formation of the torus and jets directly follows from conventional theories of the pulsar winds if the longitudinal distribution of the energy flux in the wind is taken into account. This work is the direct continuation of  Paper I. Our main goal in this paper   is to estimate  the surface synchrotron brightness of the central part of the Crab Nebula and to compare  the results of these estimates  with  observations. ",
        "conclusions": "The fact that the morphology of the  central part of the Crab Nebula can be explained in frameworks of the theory developed by \\cite{rees,kennel,chevalier} is the basic result of our work. The elucidation of the nature of the torus and jets opens for us new horizons.  In particular, comparison of the observed brightness distribution of the torus with calculations based on an accurate modelling of the plasma flow in the post shock region will open the way to obtain observational information about the energy flux distribution in the pulsar wind."
    },
    "0209/astro-ph0209163_arXiv.txt": {
        "abstract": "We present the current status of { PHOENIX} model atmospheres for dwarfs of the spectral type T, typical for older field brown dwarfs and low-mass brown dwarfs. The results are based on new predictions of the CH$_4$ line opacities from theoretical calculations with the {STDS} software package, extrapolating to transitions from rotational levels up to $J\\,=\\,40$. While individual line positions and strengths are reproduced with moderate to fair accuracy, the cumulative band strength in the region of the IR methane bands is modelled much better thanks to the inclusion of large numbers of faint lines relevant at high temperatures. ",
        "introduction": "The transition between spectral classes L and T, as $T_{\\rm eff}$ drops below $\\sim$\\,1300\\,K, is characterised by a reduction of the effects of dust absorption and the appearance of CH$_4$ absorption bands in the infrared, showing up in the coolest brown dwarfs which have only recently been discovered in substantial numbers in the field. These features appearing in the K and, from spectral type T0, in the H band and growing in strength with decreasing temperature, are one of the basic characteristics defined in the classification schemes of Geballe et al.\\ (2002) and Burgasser et al.\\ (2002) for the T spectral class. As one of the dominant opacity sources in these objects besides H$_2$O and CIA H$_2$, the CH$_4$ molecular bands are also of great importance for the correct modelling of ultracool substellar atmospheres and for evolutionary calculations of old brown dwarfs. ",
        "conclusions": "STDS-created line lists produce opacity data that significantly improve on the HITRAN and GEISA databases and allow a much more detailed analysis of the IR spectra and comparison with observationally derived spectral indices for brown dwarfs. With respect to the contribution of higher polyads several uncertainties remain that could affect the quantitative determination of atmospheric parameters based on molecular absorption features. A realistic description of the atmospheric structure will also require a more sophisticated treatment of dust formation which is currently under way.  \\subsection{Acknowledgements} This work is supported by NFS grant N-Stars RR185-258, and based in part on calculations performed at the NERSC IBM SP with support from the DoE. We thank V.\\ Boudon and J.-P.\\ Champion for helpful information and S.\\ Leggett and A.\\ Burgasser for access to their observational data."
    },
    "0209/astro-ph0209480_arXiv.txt": {
        "abstract": "We report the discovery of \\psr, a radio pulsar with period $P = 98$\\,ms and dispersion measure $\\mbox{DM} = 101$\\,cm$^{-3}$\\,pc, in a deep observation with the Parkes telescope of the axially-symmetric ``Mouse'' radio nebula (\\pwn).  Timing measurements of the newly discovered pulsar reveal a characteristic age $P/2\\dot P = 25$\\,kyr and spin-down luminosity $\\dot E = 2.5 \\times 10^{36}$\\,erg\\,s$^{-1}$.  The pulsar (timing) position is consistent with that of the Mouse's ``head''.  The distance derived from the DM, $\\approx 2$\\,kpc, is consistent with the Mouse's distance limit from H{\\sc i} absorption, $< 5.5$\\,kpc.  Also, the X-ray energetics of the Mouse are compatible with being powered by the pulsar.  Therefore we argue that \\psr, moving at supersonic speed through the local interstellar medium, powers this unusual non-thermal nebula.  The pulsar is a weak radio source, with period-averaged flux density at 1374\\,MHz of 0.25\\,mJy and luminosity $\\sim 1$\\,mJy\\,kpc$^2$. ",
        "introduction": "\\label{sec:intro}  The ``Mouse'' (\\pwn\\footnote{This source previously has been referred to by several names: G359.23$-$0.82 \\cite{yb87}; G359.23$-$0.92 \\cite{pk95}; G359.2$-$00.8 ({\\sc simbad} database).  In accordance with IAU rules, we use the original name, even though the indicated position is not accurate (see Table~\\ref{tab:parms}).}; Yusef-Zadeh \\& Bally 1987)\\nocite{yb87} is among the few known non-thermal radio nebulae with axial symmetry (see Fig.~\\ref{fig:mouse}), consisting of a bright ``head'' and a long ``tail'' that is highly linearly polarized.  It is also an X-ray source \\cite{pk95,smi+99}.  The few other examples known in this class are manifestations of a pulsar bow shock:  the relativistic wind of a neutron star confined by ram pressure due to the supersonic motion of the pulsar through the local interstellar medium (ISM).  The detailed study of such objects can lead to constraints on the local ISM density and pulsar velocities, ages, spin and magnetic field evolution, and winds \\cite{cc02,vag+02}, as exemplified by the study of the ``Duck'' nebula and its pulsar \\cite{gf00}.  Also, relatively few young nearby pulsars are known.  Detecting other such nearby pulsars, as is likely to be lurking inside the Mouse, is important for accurately determining pulsar birth rates, beaming fractions and luminosity distributions (e.g., Brazier \\& Johnston 1999)\\nocite{bj99}.  The Mouse has therefore been the object of considerable interest since its discovery.  While its interpretation as a pulsar-powered nebula is appealing \\cite{pk95}, no central engine had been detected in previous radio pulsation searches.  In this Letter we report the discovery of a faint young pulsar coincident with the Mouse's head, confirming that the Mouse is a synchrotron nebula powered by a high velocity neutron star. ",
        "conclusions": "\\label{sec:disc}  The $P$ and $\\dot P$ measured for \\psr\\ imply a relatively large spin-down luminosity $\\dot E = 4\\pi^2 I \\dot P /P^3 = 2.5 \\times 10^{36}$\\,erg\\,s$^{-1}$ (where the neutron star moment of inertia $I \\equiv 10^{45}$\\,g\\,cm$^2$), small characteristic age $\\tau_c = P/2 \\dot P = 25$\\,kyr, and surface magnetic dipole field strength $B = 3.2\\times10^{19} (P \\dot P)^{1/2} = 2.5 \\times 10^{12}$\\,G.  These parameters place \\psr\\ in the group of $\\approx 20$ ``Vela-like'' pulsars now known (those with $\\dot E \\ga 10^{36}$\\,erg\\,s$^{-1}$ and $10 \\la \\tau_c \\la 100$\\,kyr).  In the remainder of this section we discuss the evidence linking \\psr\\ with the Mouse.  With the present positional accuracy (Table~\\ref{tab:parms} and Fig.~\\ref{fig:mouse}), the offset between the pulsar's timing position and that of the head of the Mouse seen in the right panel of Figure~\\ref{fig:mouse} is $7'' \\pm 37''$, and the area of the error ellipse is $5 \\times 10^{-5}\\,\\deg^2$.  With approximately 1000 pulsars known in an area $\\approx 1000\\,\\deg^2$ along the inner Galactic plane ($260\\arcdeg \\la l \\la 100\\arcdeg$; $|b|\\la 2\\fdg5$), the probability of finding one by chance this close to the Mouse's head is about $5 \\times 10^{-5}$.  We therefore regard the positional match of both sources as highly suggestive of an association.  Eventually a more precise position for the pulsar will be obtained from timing, and possibly from \\chandra\\ observations.  We now consider distance indicators.  The measured DM, together with the Cordes \\& Lazio (2002)\\nocite{cl02} model for the Galactic distribution of free electrons, implies a pulsar distance of 2\\,kpc (the older model of Taylor \\& Cordes 1993\\nocite{tc93} yields $2.1 \\la d \\la 2.8$\\,kpc).  The distance to the Mouse has been investigated using H{\\sc i} absorption measurements.  Owing to the lack of absorption against a ring located 3\\,kpc from the Galactic center, Uchida et al.~(1992)\\nocite{umy92} infer that the Mouse is located at $<5.5$\\,kpc from the Sun.  This is consistent with the pulsar distance determination, and hereafter we consider both objects to be located at $\\approx 2$\\,kpc and parametrize the distance in terms of $d_2 = d/2$\\,kpc.  The head of the Mouse has been detected in X-rays, although with limited statistics \\cite{pk95} and angular resolution \\cite{smi+99}. Sidoli et al. model the source with a power-law spectrum having unabsorbed 2--10\\,keV flux $\\approx 3 \\times 10^{-11}$\\,erg\\,cm$^{-2}$\\,s$^{-1}$.  Assuming this flux to be isotropic implies a luminosity $L_X \\sim 1.4 \\times 10^{34}\\,d_2^2$\\,erg\\,s$^{-1}$, or an efficiency for conversion of spin-down luminosity into X-ray emission of $L_X/\\dot E \\sim 0.005\\,d_2^2$.  This is apparently a factor of $\\sim 4$ larger than the comparable efficiency in the PSR~B1757$-$24/Duck pulsar wind nebula \\cite{kggl01}, a system displaying a bow-shock morphology \\cite{gf00} and seeming in many respects similar to the Mouse/\\psr, including a pulsar with comparable spin parameters.  Only X-ray observations with higher sensitivity and resolution will settle the issue, but the presently available data do indicate that the X-ray emission observed from the direction of the Mouse's head is certainly compatible with an origin in this system, located at a distance of $\\approx 2$\\,kpc.   Given the positional coincidence, consistency in distance indicators, and energetics compatible with a common source, we regard the \\psr/Mouse association as secure.  The morphology of the Mouse (bright head, coincident with \\psr, trailed to the west by a $12'$-long cometary tail; Fig.~\\ref{fig:mouse}) suggests fast motion of the pulsar through the ambient ISM.  The tail (length $L \\approx 7\\,d_2$\\,pc) presumably results from synchrotron radiation produced by the pulsar relativistic wind in the nebular magnetic field.  For a typical field of tens of $\\mu$G, the lifetime of the radiating particles is $\\sim 10^6$\\,yr.  The non-thermal spectrum steepens away from the pulsar location likely due to synchrotron losses \\cite{yb87}.  Much can be learned about the pulsar and the local ISM through a detailed study of the Mouse's head and tail, as we now outline.  As the pulsar moves supersonically through the ambient medium producing a bow shock, the morphology of the Mouse's head is expected to be shaped by ram-pressure balance between the pulsar relativistic wind and the local ISM.  In particular, the standoff radius of the shock, $R_0$, for a neutron star moving with velocity $V$ through a medium of density $\\rho$, is determined from $\\rho V^2 = \\dot E/(4 \\pi c R_0^2)$ (see, e.g., Chatterjee \\& Cordes 2002)\\nocite{cc02}.  We assume that the pulsar wind is radiated isotropically, and hereafter neglect projection effects.  We have not measured the standoff angle $\\theta_0 = R_0/d$, but infer a crude estimate from Figure~\\ref{fig:mouse} as follows.  The width of the Mouse's head in the presumed direction of motion (west to east) is about $1\\farcs6$.  It is likely that the pulsar lies within this region of intense synchrotron emission and that the apex of the bow shock lies just outside \\cite{buc02}.  Thus we estimate $\\theta_0 \\sim 1''$, parametrized as $\\theta_1 = \\theta_0/1''$.  With $\\rho = 1.37\\,m_H n$, where $n$ is the medium's number density, $m_H$ is the mass of the H atom, and the numerical factor derives from assuming a cosmic abundance of He in the ISM \\cite{cc02}, we obtain $V \\sim 570/(d_2 \\theta_1 n^{1/2})$\\,km\\,s$^{-1}$, where we have used the $\\dot E$ measured for \\psr.  For a reasonable pulsar velocity ($1000 \\ga V \\ga 100$\\,km\\,s$^{-1}$; e.g., Lyne \\& Lorimer 1994)\\nocite{ll94}, this implies an ISM of high density, $0.3 \\la n \\la 30$\\,cm$^{-3}$.  In turn this suggests that the pulsar is moving (rapidly) through a warm or cold phase of the ISM (e.g., Heiles 2001)\\nocite{hei01}, with attendant small sound speed $C$, and hence that its Mach number should be high, $V/C \\gg 10$.  We now extend our attention to the Mouse's tail.  We can obtain a crude estimate of the pulsar's age $\\tau$ by considering that it has traveled the observed length of the tail in its lifetime, $\\tau = L/V \\approx 12\\,d_2/V_{570}$\\,kyr, where $V = 570\\,V_{570}$\\,km\\,s$^{-1}$.  The age may differ significantly from this if only part of the tail has been detected or if the tail is instead caused by a relatively recent pulsar backflow (e.g., Kaspi et al.~2001)\\nocite{kggl01}.  The actual age compares to the pulsar characteristic age of $\\tau_c = 25$\\,kyr.  These are related by $\\tau = 2 \\tau_c [ 1 - (P_0/P)^{n-1} ]/(n-1)$ under the assumption of constant magnetic moment, where $P_0$ is the initial period and $n$ is the braking index of rotation \\cite{mt77}.  A number of factors may cause $\\tau$ to be larger, or smaller, than $\\tau_c$.  $P_0 \\ll P$, and $n = 3$, appropriate to magnetic dipole braking, are usually assumed. However for some pulsars $n < 3$ (Mereghetti et al.~2002\\nocite{mbbi02} and references therein).  Also, $P_0$ may be large, up to perhaps $\\approx 90$\\,ms (e.g., Camilo et al.~2002b)\\nocite{cmg+02}.  In light of these issues, we regard the pulsar's characteristic age as approximately consistent with that inferred from the length of the tail, confirming this source as a relatively young pulsar.  In summary, while it appears that the nature of the source powering the Mouse has finally been uncovered with the discovery of \\psr, much remains to be understood about this fascinating object, requiring further observational efforts.  A key measurement to be made is the proper motion of the pulsar, $\\mu = 60\\,V_{570}/d_2$\\,mas\\,yr$^{-1}$. This may be possible through \\chandra\\ observations and, depending on the amount of ``timing noise'' present in the neutron star, via radio timing.  The pulsar--bow-shock standoff distance is also of great interest.  Finally, further study of the Mouse's tail, independent characterization of the local ISM, and any improvement to the distance estimate, would be most useful."
    },
    "0209/astro-ph0209355_arXiv.txt": {
        "abstract": "{ We have conducted a NIR study of the environments of seven radio-loud quasars at redshifts 1$<$z$<$1.6. In present paper we describe deep $K$ band images obtained for the fields of $\\sim$6$\\times$6 arcmin around the quasars with 3$\\sigma$ limiting magnitudes of $K\\sim$20.5. These fields were previously studied using deep $B$ and $R$ band images (S\\'anchez \\& Gonz\\'alez-Serrano 1999). Using together optical and NIR data, it has been found a significant excess of galaxies which optical-NIR colours, luminosity, spatial scale, and number of galaxies are compatible with clusters at the redshift of the quasar.  We have selected a sample of cluster candidates analyzing the $R-K$ vs. $K$ diagram. A $\\sim$25\\% of the candidates present red optical-NIR colours and an ultraviolet excess. This population has been also found in clusters around quasars at the same redshifts (Tanaka et al. 2000; Haines et al. 2001). These galaxies seem to follow a mixed evolution: a main passive evolution plus late starformation processes. The quasars do not inhabit the core of the clusters, being found in the outer regions.  This result agrees with the hypothesis that the origin/feeding mechanism of the nuclear activity were merging processes. The quasars inhabit the region were a collision is most probably to produce a merger. ",
        "introduction": "There is much evidence for a connection between the QSO activity and clustering of galaxies (Stockton 1982; Yee \\& Green 1984, 1987; Gehren et al. 1984; Hutchings et al. 1984; Hintzen 1984; Yee 1987; Ellingson et al. 1991; Hintzen et al. 1991; Yee\\& Ellingson 1993; Fisher et al. 1996; Yamada et al. 1997; Hall et al. 1998; Hall \\& Green 1998; S\\'anchez \\& Gonz\\'alez-Serrano 1999; Cimatti et al. 2000; Wold et al. 2000). It seems that interaction/merging processes amoung cluster galaxies could rule a fundamental role in the origin/feeding process of the nuclear activity. The large fraction of host galaxies of radio quasars with distorted morphologies (e.g. Disney et al. 1995; Hutchings \\& Neff 1997; S\\'anchez 2001; S\\'anchez et al. 2002), their large luminosities at $z\\sim$1 (Carballo et al. 1998; S\\'anchez et al. 2002, and references therein), direct evidences in their spectra (Nolan et al. 2001), and the velocity distribution of galaxies around QSOs (Heckman et al. 1984; Ellingson et al. 1991) give support to this hypothesis (Yee \\& Ellingson1993).  Yee \\& Green (1987), Ellingson et al. (1991), Yee \\& Ellingson et al. (1993) and Wold et al. (2000) have shown evidences of the connection between radio emission and galaxy clustering around QSOs. They found differences between the environment of radio-loud and radio-quiet quasars. Since radio-loud QSOs are found in clusters of galaxies with an Abell richness class between 0-2 (Abell 1958), radio-quiet relatives inhabit groups of galaxies of richness similar or even lower than 0. However, these results were based in studies of overdensities of galaxies around low-$z$ quasars ($z<$0.7). At higher redshifts, the number of studies decreases (Hitzen et al. 1991; Boyle \\& Couch 1993; Hall \\& Green 1998), or were based in a reduced number of objects (Hutchings et al. 1993; Yamada et al. 1997; S\\'anchez \\& Gonz\\'alez-Serrano 1999; Teplitz et al. 1999; Cimatti et al. 2000). However, the differences between the environment of radio-loud and radio-quiet quasars appear to persist at high-$z$ (1$<z<$2).  Hall et al. (1998) and Hall \\& Green (1998) showed that the excess of galaxies around their sample of 31 radio-loud QSOs at 1$<z<$2 seems to present a two component distribution, with a peak around the quasars at $r<$35$\\arcsec$ and a smooth component that extends to r$\\sim$100$\\arcsec$ (his maximun sampled distance). They explained this two-component distribution as a cluster of galaxies embebbed in a large scale structure. In S\\'anchez \\& Gonz\\'alez-Serrano (2001), hereafter Paper I, we have analyzed the deep $B$ and $R$ band images of seven radio loud quasars at 1$<$z$<$1.6, with a larger sampled spatial region, finding a similar result. However, the explanation to this peculiar distribution was found more simple, since the quasars seem not to be in the center of the overdensities of galaxies. Once corrected for the correct overdensity centers, the two-component distribution blurs, and a typical cluster overdensity profile remains.  We present here deep $K$-band images of fields of $\\sim$6$\\arcmin$$\\times$6$\\arcmin$ around the seven radio quasars studied in Paper I (Section 2). We have looked for possible overdensities of galaxies also in this band (Section 3), and, together with the $B$ and $R$ data, we have selected a sample of candidates to cluster members (Section 4). We have tried to find if the results presented in Paper I could be confirmed with these new data. The combined $B$, $R$ and $K$ band data for the different galaxies detected have been used to study the characteristics of these cluster candidates (Section 5). Throughout this article we have assumed a standard cosmology with H$_o$=50 km s$^{-1}$ Mpc$^{-1}$ and q$_o$=0.5. It is interesting to note here that at the redshift range considered in this article the differences in the scales were small ($<$10\\%) if we use a cosmology with H$_o$=70 kms$^{-1}$ Mpc$^{-1}$ $\\Omega_m$=0.3 and $\\Omega_\\lambda$=0.7. ",
        "conclusions": "In this article we have analyzed deep $K$-band images around seven radio loud quasars, with a redshift range 1$\\la$$z$$\\la$1.6.  In Paper I, we presented the analysis over deep $B$ and $R$ band images around the same quasars. Using the present results together with the results from Paper I, we found that there is a significant excess of galaxies around these quasars, with magnitudes, colours, spatial scale and number of galaxies compatible with clusters of galaxies at the redshift of the quasars. This result agrees with previous studies of possible overdensities of galaxies around radio-loud quasars (Hintzen et al. 1991; Hall \\& Green 1998), in contrast with results obtained on similar studies over radio-quiet samples (Boyle \\& Couch 1993; Teplitz et al. 1999), although some of these studies found significant excesses around radio-quiet quasars (Haines et al. 2001). These results also agree with studies based on samples of lower redshift quasars, where the environment of radio-loud and radio-quiet quasars seem to present significant differences (Yee \\& Green 1987; Elligson et al. 1991).  There are different mechanisms that could explain the described differences: (i) it is expected that radio loud sources inhabit gas rich clusters, if the intracluster gas plays a substancial role as the fuel of the radio source. It is reasonable to assume that intracluster gas increases with cluster density, which could explain the differences in the environments of radio-loud and radio-quiet sources; (ii) extended radio emission, with steep-spectrum emission, is normally explained as the result of the interaction of the jets with the intergalactic medium (e.g. Miley 1980). This medium confines the radio emission lobes. It is expected that the intergalactic medium was denser in more populated clusters, and therefore, there could be a relation between cluster density and radio loudness. Maybe the combination of both mechanism is needed to explain the observed differences. It is interesting to note here that the second mechanism could explain the correlation between the excess of galaxies and the slope of radio emission found by Hall \\& Green (1998) and Mendes de Oliveira et al. (1998).  Two mechanisms have been proposed to explain the connection between nuclear activity and its environment: the dynamic state of the clusters and the evolution of intracluster gas (Yee \\& Ellingson 1993). In both schemes the merging/interaction processes between the host galaxy and its neighbours play a substancial role, which reinforce the hypothesis that a merging process could be the origin of nuclear activity (Barnes \\& Hernquist 1992; Shlosman 1994; Bekki 1999). There is ample evidence to support this idea, like the large fraction of host galaxies under possible interaction/merging processes, their large luminosities, and possible traces in their spectra (Hutchings 1987; Disney et al. 1995; Hutchings \\& Neff 1997; Carballo et al. 1998; S\\'anchez et al 2002; Nolan et al. 2001).  We found an envelope of red galaxies in the colour-magnitude diagram ($R-K$ vs. $K$), using the combined information from the three bands. This envelope was consistent with the CM relation found for clusters with a redshift range between 0.3$<$$z$$<$1.3. Extrapolating this relation to the redshift range of our sample (interpolating in some cases) we selected a sample of cluster candidates. Among these galaxies selected as candidates by their $R$ and $K$ magnitudes, only $\\sim$25\\% were detected in the $B$ band. This group of galaxies comprises a subsample of red galaxies with ultraviolet excess, similar to found by Tanaka et al. (2000) and Haines et al. (2001).  The colour evolution of the cluster candidates was compatible with a model in which the bulk of the galaxies follows a passive evolution typical of old galaxies ($z_{\\rm for}\\sim$4-10), whereas a fraction of $\\sim$25\\% follows a mixed evolution. This mixed evolution implies that at least 95\\% of the mass follows a passive evolution and a fraction lower than $\\sim$5\\% has undergone presents star formation, even at the observation redshift. This star formation could be induced by collision processes (Lubin et al. 1998; Tanaka et al. 2000; Haines et al. 2001). Spectroscopic studies of these galaxies are needed to clarify these results.  We have found that the quasars of our sample do not inhabit the cluster cores. The position of the quasars in the clusters is also compatible with the hypothesis of a merging process as the QSO origin. Only collisions between galaxies with relative velocity similar to their internal velocity could induce merging. This merging was a prerequisite for the infall of mass into the inner region and the formation or feeding of the nuclear source. This limits the region where collisions could lead to merging to the outer region of the clusters, since the velocity were too large in the cluster cores.  Summarizing, out results suggest that radio-loud quasars inhabit the outer regions of low populated clusters at high redshift. Clusters that show mixed evolution, which main component would be passive evolution for old galaxies with a reduced fraction under a recent starformation process. These results support the hypothesis of a merging process as the origin/feeding mechanism of nuclear activity."
    },
    "0209/astro-ph0209089_arXiv.txt": {
        "abstract": "Using the 9 years of continuous data now available from the Burst And Transient Source Experiment (BATSE) aboard CGRO, we have measured orbital periods and produced folded lightcurves for 8 High Mass X-ray Binaries (HMXB).  Given the length of the datasets, our determinations are based on many more binary orbits than previous investigations. Thus our source detections have high statistical significance and we are able to follow long-term trends in X-ray output. In particular we focus on two systems: A0538-668 and EXO2030+375 both HMXBs exhibiting Type I outbursts.  Recent work on A0538-668 \\cite{al} reported a 16.65d optical variability due to the orbital period, but only seen during minima of a longer-term variability at 421d. We searched for this signal in the BATSE dataset using an ephemeris derived from Alcock et al \\shortcite{al} \\& Skinner \\shortcite{sk}. We found no evidence for such modulation and place an upper limit of 3.0 $\\times 10^{-3}$ photon cm$^{-2}$ s$^{-1}$ in the 20-70 keV BATSE energy band , based upon statistical modelling of the signal.  Previous observations of EXO 2030+375 \\cite{rei} using RXTE ASM data indicate secondary outbursts occur at apastron passage. We present a lightcurve for an earlier epoch showing convincing evidence for such apastron outbursts. We find apastron outbursts in 3 sources, all having orbital periods greater than 41d. No such signal is conclusively detected in the more rapidly orbiting systems studied. ",
        "introduction": "The Burst And Transient Source Experiment on the Compton Gamma Ray Observatory \\cite{fish} provided near continuous monitoring of the whole sky at hard X-ray energies.  BATSE only sees the high energy tail of HMXBs, typically detected in the lowest DISCLA channel and CONT channels 1-4 covering 20-70keV.  In studying X-ray pulsars two modes of operation are possible; pulsed flux and Earth occultation. If pulsations are detected corresponding to the spin period of a known pulsar then the pulsed flux and pulse frequency can be measured with Fourier analysis or epoch folding (Bildsten, L. et al. 1997) and used to monitor the source with great sensitivity and fine time resolution. \\cite{bil} For the long term study of behavior related to orbital motion in binary systems, daily averages of the total flux are sufficient and are in addition sensitive to non-pulsed X-ray output from the source. By exploiting the Earth's limb as an occulting mask, individual X-ray sources can be isolated with an accuracy of ~1$\\degr$. Termed the Earth Occultation Technique, two measurements of the total flux from a source are made during each 90 minute satellite orbit as the source moves into and out of eclipse due to the satellite's motion, full details are given by Zhang et al \\shortcite{zha} \\& Harmon et al \\shortcite{har}.  In this paper we present folded X-ray lightcurves of Type I (normal) outbursts from 8 HMXBs in which periodic fluctuations corresponding to previously published orbital periods were significantly detected.  Such outbursts are thought to occur at each periastron passage of the neutron star. These outbursts last for 8-10 days and the X-ray luminosity increases by a factor of $\\sim$10.  In contrast, Type II are giant outbursts which do not correlate with the orbital phase but tend to last for several orbital cycles. The X-ray luminosity during the outburst may reach close to the Eddington luminosity.   In addition we studied the 30.5d modulation from LMC X-4 which was also a significant detection and investigated the X-ray behaviour of A0538-668 in the light of recent optical work \\cite{al}. The BATSE datasets for these systems cover many orbital cycles and hence present the first ever opportunity to study their long term behavior. Previously only widely spaced, scheduled or chance obsevations were available for this purpose. Usually made with different instruments having little or no overlap in terms of spectral response, extrapolating long term trends from previous data was of limited value.  The publicly available pulsed flux monitor reports produced by MSFC demonstrates that HMXBs typically exhibit active epochs lasting several months interspersed with quiescent periods. A question we addressed was that of possible changes in behaviour between these epochs. If a source is bright enough or its active epochs span enough orbital cycles then folded X-ray lightcurves from different epochs can be directly compared and variations in behavior detected.  Variability in the X-ray recurrence period can be looked for in the power spectrum of the flux history. Width of the recurrence frequency peak will be increased if that frequency varies, in other words if the exact time at which the outbursts turn on varies from orbit to orbit then this will show up in the power spectrum. If such jitter can be identified this could provide clues about the underlying physical behaviour of such systems. For example theory indicates that accretion rates are effected by a neutron star's speed relative to its surroundings. The Alfven speed in stellar wind plasma is a function of both magnetic field strength and plasma density. Thus a high outflow  velocity of a wind or circumstellar disk may inhibit accretion during periastron passage and push the X-ray outburst to a later phase. Additionally the orientation of the magnetic field frozen into a plasma strongly affects its coupling to a magnetosphere moving through it. These effects, although extremely difficult to measure directly, may leave a signature in possible chaotic fluctuations or jitter in the exact phase at which outbursts turn on. The jitter is expected to be chaotic because there is no direct connection between the plasma properties and the binary motion, yet the stability of the accretion disk depends on a dynamic balance between accretion rate onto the disk and accretion rate onto the neutron star.  Secondary outbursts occuring close to apastron of the neutron star are also a key observation in this study. Their existence and exact phase at which they occur will yield information about circumstellar disk outflow and stellar wind parameters, principally outflow velocity and density information. Higher outflow velocity is expected to delay periastron outburst turn-on and is cited as a possible cause of apastron outburst in the Bondi-Hoyle theory of accretion. Our folded lightcurves show that EXO2030+375, GX301-2 and 4U1145-619 may exhibit X-ray emision at apastron passage of the neutron star. Whether this emission occurs always or only sporadically, is impossible to determine from individual orbits, so long-term coverage is a necessity.  In these preliminary interpretations of our lightcurves we are aware that BATSE only sees the high energy tails of the emitted X-ray spectra of these sources. Lightcurve shape is thus affected by the true spectral profile of the emitted X-rays and the spectral response of the detector. The most significant effect is expected to occur at the peak of periastron outbursts, when there is the highest column density of obscuring matter and the most intense flux of hard photons, leading to X-ray reprocessing and re-radiation at lower energies.  Parameters for the 8 systems studied are shown in Table 1. Of these sources, in all cases but LMC X-4 the modulation studied is directly linked to orbital motion. In LMC X-4 the binary period of 1.4d is not seen because it is too faint in hard X-rays. The 1.4 day orbital period of LMC X-4 is above the Nyquist limit for this data (about 2.4d) so aliasing will shift this frequency in the power spectrum, but no significant signal is detected. However the well known periodicity near 30.5 days was detected and a lightcurve constructed. An improved measurement was made of X-ray modulation due to the precession of a warped accretion disk.  Note that throughout this paper dates are presented as Truncated Julian Dates (TJD), where TJD = JD - 2440000.5.  \\section[]{Data processing and data artefacts}  The BATSE dataset used in this work consists of daily averages of the flux levels derived from the occultation technique at MSFC \\cite{har}.  In the occultation technique, source fluxes are in principle measured twice every spacecraft orbit, as rising and setting steps. However, in practice, not all occultation steps are available for measurement. The most common reason steps are unavailable are (1) passage of the spacecraft through the lower Van Allen radiation belt at the South Atlantic Anomaly, when the detector high voltage was turned off; (2) periods of time when CGRO was out of line-of-sight contact with the NASA Tracking and Relay Satellites (TDRS); and (3) data which were flagged by the BATSE operations team (i.e. during gamma ray bursts, solar flares, etc.) and were not available for analysis. The oblateness of the Earth causes the CGRO orbit to precess with a period of about 53 days. During the precession cycle, high declination sources ($|\\delta|\\ge41^\\circ$) cease occulting near the orbital poles.  When occultation steps are measured, the background is estimated using 110 seconds of data before and after the occultation step, called the fitting window. Occultation steps from sources other than the measured source within this window are considered interfering sources. If occultation steps of an interfering source occur within 10 seconds of the measured source, that step is discarded. The number of interfering sources in the fit affects the systematic error in three ways (1) where sharing of flux occurs between source terms in the fitting process and (2) where sources are either included or not included in the fit because of an incorrect assumption about the sources' relative intensity to the measured source or (3) when an unknown source is present in the fitting window. Other sources of systematic error include red noise and coherent pulses from bright X-ray sources present, but not necessarily occulting, within the fitting window.  To estimate the average systematic error, a grid covering the galactic plane with points every 3 degrees along the galactic plane with two additional sets of grid points at +6 degrees and -6 degrees galactic latitude. Grid points within 2 degrees of bright occultation sources where excluded, yielding a total sample of 156 points. The average flux from each grid point was averaged over a 9 year period and the standard deviation of 1-day flux averages was computed as a function of galactic longitude. The values for the standard deviations are larger than expected for a Gaussian distribution about zero flux due to various systematic effects. We find broadening factors of about 30\\%-60\\% in excess of normal statistics, strongest near the galactic center and near bright variable sources such as Vela X-1 and Cygnus X-1. Both positive and negative trends are present in the average fluxes. For example, near EXO2030+375 ($l_{II}$ = 77.2 degrees, $b_{II}$ = -1.3 degrees), the average flux at $b_{II}$ = 0 degrees is 0.0077, 0.0009, -0.001 photons cm$^{-2} s^{-1}$ and standard deviation 1.85, 1.42, 1.26 (1 sigma = 0.01 photons cm$^{-2} s^{-1}$) for longitudes $l_{II}$ = 72.0, 78.0, and 84.0 degrees, respectively (Harmon, B.A. et al. 2002). Clearly interfering sources in various regions of the sky produce both positive and negative offsets in flux. These fluxes may be modulated at the precession period and may change sign, depending on the limb geometry.    Data gaps arise due to three specific causes: passage through the South Atlantic Anomaly when the detectors are switched off; known interfering source occulting within 10 seconds of the target; periods when the source is occulted by the earth's limb at too shallow an angle for a significant detection to be made (or is not occulted at all). The latter effect happens only for sources lying at greater than 41 degrees in declination.  Interfering sources that are positively identified should be fitted for in the occultation technique, however this is (as in the case of the background correction) a question of optimisation and so interfering sources are not always completely removed. A signature of source interference is significant negative flux indicating that a persistent source close to the target has not been fitted correctly in calculating the occultation step-height for the target source. Provided the interfering source has not varied on timescales comparable to the recurrence period of the target, it will be averaged out in the epoch folding procedure. If however the source has varied then its effects must be positively identified before a folded lightcurve can be produced.   \\begin{table*} \\caption{Summary of the observational parameters of HMXRBs studies in this work. A0538-668 is not included in the table because it was not detected in the BATSE data sample. TJD = Julian date - 2440000.5. For EXO2030+375 epoch 1 corresponds to TJD8363.5 - 9353.5 and epoch 2 to  TJD10114.5 - 10723.4. Column 4 gives the epoch of periastron passage used (Bildsten et al 1997), except in the case of 4U1145-619 and 4U1700-377 which are arbitary dates. The last two columns compare the observed peak in the Lomb-Scargle power spectrum with Monte Carlo simulated data - see text for details.} \\label{tab:summary} \\begin{tabular}{lcccccc} &&&&& \\multicolumn{1}{c}{BATSE data} & \\multicolumn{1}{c}{Simulated data} \\\\ X-ray Source & Optical  & Orbital period  & T$_{0}$  & Detection & Peak FWHM  & Peak FWHM  \\\\ & Counterpart & (days) &  (TJD)  & Significance & $\\times 10^{-4} d^{-1}$ & $\\times 10^{-4} d^{-1}$ \\\\  EXO2030+375 epoch 1         & Be & 46.01    & 8936.50 & 33.3 $\\sigma$ & 8.70 & 9.06$\\pm$0.34 \\\\ EXO2030+375 epoch 2         & Be & 46.01    & 8936.50 & 3.47  $\\sigma$ & 16.4  & 15.6$\\pm$6.0 \\\\ 4U1145-619  \t      \t     & Be & 186.5    & 8363.51 & 5.68 $\\sigma$ & 4.56  & 3.72$\\pm$1.7 \\\\ 4U1700-377                  & SG & 3.412    & 8363.54 & 241$\\sigma$ & 3.50 & 3.40$\\pm$0.06 \\\\ 4U1907+097                  & SG & 8.38     & 5578.75 & 3.41  $\\sigma$ & 4.86  & 3.61$\\pm$2.0 \\\\ Cen X-3\t\t     &    & 2.087    & 8561.50 & 29.7 $\\sigma$ & 3.27  & 3.44$\\pm$0.13 \\\\ GX301-2\t             & SG & 41.5     & 8802.79 & 139$\\sigma$ & 3.65  & 3.45$\\pm$0.09 \\\\ LMC-X4\t\t             & Be & 1.4, 30.35& 7741.99 & 86.0$\\sigma$ & 3.34  & 3.46$\\pm$0.10 \\\\ OAO1657-41\t             & SG & 10.44    & 8515.99 & 74.5 $\\sigma$ & 3.26  & 3.41$\\pm0.09$ \\\\ \\end{tabular}           \\end{table*}     Unidentified soures of interference pose an additional problem in that their position relative to the earth's limb and the target change continuously. This leads to modulation at the precession period as the interference contribution varies cyclically with the orbital plane.This type of artefact may be non sinusoidal although subtracting a sinewave with correct period and amplitude can attenuate it in some cases. In principle such signals could be used iteratively to improve the occultation analysis but this is not practical at this time. Persistent and identified transient sources of interference can be investigated on a source by source basis. In some cases datapoints or stretches of data have to be excluded, if for example a nearby source  flares unpredictably. While in some datasets the interference is clearly apparent as a completely distinct feature in the power spectrum and can be attenuated by subtracting a sinewave fit to the modulation if neccesary. Scargle \\shortcite{sca3} states that the modified power spectrum (Lomb-Scargle periodogram) is equivalent to least-squares fitting with sinewaves, hence our procedure is equivalent to filtering the data in the frequency domain.    \\subsection{Implications of uneven sampling} Unevenly sampled data cannot be correctly analysed by the direct Fourier Transform technique as the resulting power spectrum is a convolution of both the source frequencies and sampling frequencies. This leads to spurious frequencies and possible suppression of genuine signals. Specifically; uneven sampling increases leakage of power to nearby frequencies, although at the same time, aliasing may be reduced. Our data are very unevenly sampled, especially when we excluded subsets of the data e.g. data intervals selected on the basis of an ephemeris in order to search for the possible 16.65d transient modulation in A0538-668. Fortunately the Lomb-Scargle periodogram is formulated for exactly this situation and is relatively insensitive to the sampling structure (Scargle 1982, 1989).  A further complication is the non-sinusoidal nature of many of the lightcurves, generally the less sinusoidal a signal is, the more components it will have in any frequency space representation. These components are seen in power spectra as sidebands to the main frequency spike. If the shape of an otherwise regular modulation changes with time, these peaks can become broadened, shift and become blended together. Thus in our period determinations we backed up the Lomb-Scargle result with Phase Dispersion Minimisation analysis (PDM). \\cite{pdm}. Such periodograms typically contain a number of local minima corresponding to (For example) integer multiples of the true period. However the PDM technique is totally insensitive to the shape of the lightcurve and this makes for highly accurate period determinations in the case of irregular lightcurves. Periods were measured from the PDM periodogram by calculating the centroid of the feature of interest. We relied on the Lomb-Scargle periodogram for our primary detection tool because the separation of Fourier components is better suited to identification of interfering sources, sidebands, aliasing and statistical modeling of the signal.  PDM was implemented in the Starlink program 'Period'.  \\subsection[]{Evaluation of signal detection confidence levels} In detecting weak, unevenly sampled, periodic signals against a varying background, statistical significance is a key issue. The only reliable way to evaluate the confidence levels associated with a power spectrum under these conditions is by statistical simulation.  We first measured the mean and variance of each dataset, we then created a series of random numbers conforming to a Gaussian distribution with the same mean and variance as each dataset. This series was then combined with the sampling structure (Timing information) of the original dataset by simply replacing each flux measurement with a number drawn from the random gaussian distribution. Thus creating a simulation of what BATSE might have seen, had there been no periodic X-ray sources in its field of view. This dataset was then subjected to Lomb-Scargle analysis and the power of the highest peak and the highest peak in a specified frequency range recorded. This process was then repeated 1000 times after which the mean and and variance of the peak noise power (PNP)were calculated, both for the whole spectral range available and for the specified frequency range of interest. For the Lomb-Scargle periodograms presented here, the displayed confidence levels are the mean PNP + 3 standard deviations. Thus our confidence levels directly reflect how likely it is that a given signal is real or due to chance without making any assumptions about the underlying properties of the data. The confidence levels calculated by this procedure for each dataset were later used in evaluating the Lomb-Scargle periodograms generated from simulated datasets.    \\subsection[]{Jitter in outburst turn-on times}  The width (FWHM) of peaks in the power spectrum may be related to orbit-to-orbit variation in the exact time at which outbursts occur. We investigated this by comparing the FWHM of the orbital modulation peak in the Lomb-Scargle periodogram to that measured for a simulated dataset containing a sinewave of appropriate period and amplitude superimposed on a background created by randomizing the flux readings in the original dataset.  Results are shown in Table~\\ref{tab:summary} for comparison.  Further investigation of this parameter by a Monte-Carlo technique showed that the FWHM is dependent on the noise present. Specifically the FWHM becomes more variable as the S/N ratio deteriorates. Combined with the well known dependence of the FWHM on the length of the dataset, the resolution of the periodogram in respect of this parameter had to be calculated independently for each dataset. For each source a Monte-Carlo simulation was run by first finding the correct amplitude sinewave to give the same S/N ratio as that seen in the data -using the limit finding procedure described in section 3.1.1.  The flux values in the data were then repeatedly randomized and added to this signal. The Lomb-Scargle periodogram was calculated each time and the FWHM of the peak measured and recorded. After 100 cycles the mean and standard deviation of the FWHM were calculated and the results are displayed in Table~\\ref{tab:summary}.   In addition to movement in the outburst phase it is possible that other effects such as variations in the shape of the outburst profile might also give rise to a widening of the peak in the power spectrum. However, inspection of the results presented in Table~\\ref{tab:summary} shows no evidence of any variability in the power spectrum peak indicating a general lack of jitter in turn-on times and no significant variation in the outburst profiles.  \\section[]{Results} Table~\\ref{tab:summary} summarises key observational parameters of the HMXBs studied, while Figure~\\ref{fig:lightcurves} shows the X-ray lightcurves folded over 20 phase-bins. The errors shown are observational errors inherent in the occultation analysis, in producing the averaged lightcurves, the error-bars are the standard deviation of points in each phase-bin.  \\begin{figure*} \\begin{center} \\psfig{file=fig1.ps,width=15cm,height=20cm} \\caption{Phase-averaged lightcurves created from BATSE occultation flux measurements. All data are folded at the periods and phases given in Table~\\ref{tab:summary} except for 4U1907+09 which is folded at twice the period quoted in the table (see text for details).} \\label{fig:lightcurves} \\end{center} \\end{figure*}   \\begin{figure*} \\begin{center} \\psfig{file=fig2.ps,width=15cm,height=20cm} \\caption{Lomb-Scargle periodograms of BATSE occultation flux histories. The 3$\\sigma$ significance levels are plotted as dashed lines. These levels are calculated by Monte-Carlo simulations over the frequency space displayed (see Section 2.2).} \\label{fig:lsdata} \\end{center} \\end{figure*}   \\subsection{ Link between Type I X-ray outbursts and long period optical variations in A0538-668 } We tested the prediction that A0538-668 only exhibits X-ray outbursts correlated with its orbital period at times coincident with minima in its long period optical lightcurve. \\cite{al} We note that previous widely spaced X-ray observations of A0538-668 tend to support this hypothesis, however we find conclusively that between TJD 8363 - 10999 (The time span of the BATSE occultation data) no significant modulation of 20-70 keV X-ray flux at the orbital period is present at the predicted epochs. To do this we calculated an ephemeris for precise times at which this model predicts outbursts to occur. The orbital parameters were taken from Skinner \\shortcite{sk}, P=16.65d, with a zeropoint T$_{0}$=TJD 3423.96 Maxima are expected at phase 0.9 in this scheme if the X-rays follow the optical variability. The long period parameters from Alcock et al \\shortcite{al} are under this scheme: P=420.82d, with arbitrary zeropoint TJD 9001. The long period minima occur at arbitrary phase 0.55. The ephemeris was then used to isolate time intervals when X-ray outbursts were expected. The duration of the time interval examined either side of long period minima was 100 days to cover the data coincident with the optical lightcurve \"dip\" regions.  According to Skinner \\shortcite{sk}, A0538-668 was in the field of view of the Einstein satellite on four occasions. According to the ephemeris described here,  detections reported on 16 Dec 1980 and 3 Feb 1981 lie in the optical dip, and non detections on 10 Apr 1979 and 30 May 1980 lie firmly in the centre of the bright phase of the optical lightcurve.  \\subsubsection[]{BATSE detection threshold for A0538-668} By creating test datasets with the same mean and variance as the BATSE data and containing only simulated signals, we determined how strong a periodic signal must be in order to give a 3$\\sigma$ detection. This approach could also be used to model other components of a power spectrum, particularly where interfering sources and sampling structure lead to spurious frequencies.  Known interfering signals in the A0538-668 data include LMC X-4 (which is only 0.6$\\degr$ away and dominates the periodic flux in this dataset) and the spacecraft precessional modulation. These components were modelled by measuring their amplitudes in folded lightcurves at 35.5d and 52d respectively.  Using the ephemeris described above BATSE data were selected which were coincident with the four long-period optical minima observed during the MACHO observations of Alcock et al \\shortcite{al}. This resulted in four blocks of BATSE one-day averages each of duration $~$100 days spanning the time interval TJD 9124.52 - TJD 10999.5 Having failed to detect any statistically significant modulation consistent with 16.65 days, this subset of the BATSE observations of A0538-668 was then repeatedly randomized to destroy any periodic content and combined with an incrementaly increasing sinusoidal signal plus fixed sinewave components to simulate the interfering sources in the real data. After each step, the Lomb-Scargle periodogram was computed and the highest peak in a pre-determined frequency range was compared to the 3 $\\sigma$ confidence level previously calculated for that frequency range as described in section 2.3.  Once this threshold was exceeded, the amplitude of the sine wave signal was recorded.  The detectability of such a weak signal is somewhat influenced by the noise present. At the very margins of detectability the power in a given frequency can fluctuate up and down despite a gradual increase in signal amplitude.  Hence we used the criterion that the power in the selected frequency range must remain above the threshold value irrespective of chance structures in noise in order to qualify. The derived upper limit was independently tested using many different noise sets to verify this criterion was met, i.e. That a signal of the amplitude quoted would definitely have been detected by the methods described here, had it been present.  Weaker signals would not consistently rise above the threshold although detections of marginal significance are possible.  \\section[]{Discussion} \\subsection[]{EXO2030+375}  This object has been subject of several recent studies. Reig et al \\shortcite{rei} presented results of multi-wavelength observations and modeling of the accretion rate. Wilson et al (2002) report on a decade of X-ray observations combined with IR and H$\\alpha$ measurements which show a decline in the density of the circumstellar disk around the Be star. This decline was followed by a sudden drop in the X-ray flux and a turn-over from a spin-up trend to spin-down in the frequency history. This is the first Be/X-ray binary to show an extended interval, about 2.5 years, where the global trend is spin-down, but the outbursts continue. In 1995 the orbital phase of EXO 2030+375's outbursts shifted from peaking about 6 days after periastron to peaking before periastron.  The outburst phase slowly recovered to peaking at about 2.5 days after periastron. Wilson et al (2002) interpret this shift in orbital phase followed by a slow recovery as evidence for a global one-armed oscillation propagating in the Be disk. This was further supported by changes in the shape of the H$\\alpha$ profile which are commonly believed to be produced by a reconfiguration of the Be disk. The truncated viscous decretion disk model provided an explanation for the long series of normal outbursts and the evidence for an accretion disk in the brighter normal outbursts.  In the present analysis we find convincing evidence for elevated flux levels at apastron in the earlier epoch (see Figure~\\ref{fig:lightcurves}). The RXTE ASM lightcurve for epoch 2 was studied by Reig et al \\shortcite{rei}, we also analysed this data in order to compare the lightcurve and power spectrum with the BATSE results. The power spectra from both instruments appear to share the same features in epochs 1 \\& 2, peaks at $P_{orb}$ \\& $P_{orb}/2$ with powers at similar ratios. Direct comparison of the BATSE and RXTE folded lightcurves suggests that X-ray emission from EXO2030+375 may be softer during the apastron outbursts than during the normal Type I outbursts at periastron. This hypothesis is drawn from the relative amplitudes of the two peaks in the lightcurve and the differing spectral responses of the two instruments.  Comparison of the FWHM of the orbital period peak at 46d with that derived from simulated data reveals that although the phase of peak outburst intensity shifts between the two epochs, its recurrence is otherwise periodic.  There is no evidence for any gradual shift in phase in either epoch.  \\subsection[]{A0538-668}  We find that during the period coincident with the MACHO monitoring of Alcock et al \\shortcite{al}, A0538-668 did not exhibit hard X-ray outbursts at the 16.65 day orbital period during either the optical minima or maxima. The upper limit for such activity is found to be 3.0 $\\times 10^{-3}$ photon cm$^{-2}$ s$^{-1}$at 20-70 kev. We compare this limit with the X-ray spectrum obtained with Ariel V \\cite{wc} at the peak of the July 1977 outburst. By assuming a mean photon energy of 45 keV over a spectral range of 20-70 keV for the BATSE observations we estimate a maximum flux density of 2.7 $\\times 10^{-3}$ keV cm$^{-2}$ s$^{-1}$ keV$^{-1}$. The fitted spectrum of White \\& Carpenter \\shortcite{wc} is extremely steep and dominated by a thermal component. By extrapolating their model, our lower limit for detection by BATSE in the data presented here is consistent with the flux that would be expected on the basis of this model to within errors. Thus we conclude that had A0538-668 been active at its previous level BATSE would have detected it.  \\subsection[]{Centaurus X-3}  This source was detected in BATSE data at 29$\\sigma$ with a period 2.087d. Cen X-3 is the only HMXB to have a well determined orbital period derivative \\cite{nag} however this is too small to show up in our analysis.  The power spectrum in Figure~\\ref{fig:lsdata} shows two peaks, but comparison with the power spectrum for simulated data showed the higher-frequency peak to be an artefact of aliasing. The folded lightcurve Figure~\\ref{fig:lightcurves} is a simple sawtooth like form, brightening more rapidly than it decays.  \\subsection[]{4U1907+097}  Two peaks with $\\sim$3$\\sigma$ significance are found at periods of 16.8d and 7.39d.  The orbital period of the system is believed to be 8.38d (Marshall and Ricketts 1980, Cook and Page 1987) thus the 16.8d X-ray modulation is consistent with twice the binary period.  However, there is no evidence for a peak in the power spectrum at $\\sim$8.4d - this is not surprising if one considers the folded lightcurve in Figure 1 which shows just a single broad peak when folded at 16.8d. Constructing a simulated dataset containing a sinewave at 8.38d with amplitude taken as 0.002 (measured from a lightcurve folded at this period) failed to reproduce the 16.8d modulation. Thus it appears that the emission is really modulated at twice the binary period in these BATSE data.  \\subsection[]{4U1700-377}  Early BATSE occulation data from this source have already been the subject of a paper by Rubin et al., 1996.  Our PDM period agrees with 3.41180$\\pm$0.00003d from Copernicus X-ray observation \\cite{bran}. We also detect significant peaks in the power spectrum at 13.808d at a significance of 11.3$\\sigma$, and at exactly half this period. This period has been previously reported by Konig \\& Maisack \\shortcite{km97}.  We further investigated this period by subtracting the orbital modulation from the lightcurve, using Figure~\\ref{fig:lightcurves} as a template. This procedure resulted in reduction of spectral power at the orbital period by a factor of 24. No attenuation of the 13.808d period was observed, in fact a significant increase resulted. Hence we are confident that the 13.808d period is an independent modulation and not an harmonic of the orbital period.    \\subsection[]{4U1145-619}  A peak was detected in LS periodogram at 190.238d with FWHM=17.48 days. A PDM periodogram determination of 186.6d $\\pm$0.348068d measured by centroiding is consistent with the established value of 186.5 days.  \\subsection[]{OAO1657-41}  The PDM period of 10.4626$\\pm$0.0011d is in agreement with the period of 10.4436$\\pm$0.0038d \\cite{ch1} from BATSE pulse timing studies.  \\subsection[]{LMC X-4}  The 1.4d binary period is not detected. The \"average\" Nyquist period of the daily average data is 2.4d so the single-step data was also analysed, again with no detection. The 30.5d X-ray modulation believed to be due to the precession of a warped accretion disk \\cite{lang} is detected and is shown to be persistent throughout the full span of the data Figure~\\ref{fig:lightcurves}. All period finding techniques tried, (Lomb-Scargle, PDM, CLEAN) derive a period for this modulation slightly shorter than usually quoted. A PDM periodogram constructed at a resolution of 2$\\times 10^{-5}$ cycles and measured by centroiding, derived a period of 30.349d while the L-S periodogram yielded 30.355d. Monte Carlo simulations of sinusoidal signals at 30.5d and 30.35d superimposed on noise created by randomizing the original dataset show that the LS periodogram frequency resolution is 1.43765$\\times 10^{-5}$ at the S/N level seen in the data. A solution at 30.5d is thus seemingly in conflict with the BATSE dataset. A remarkable fact is the stability of this modulation over the 7 years of data, since no jitter is evident from the values in Table~\\ref{tab:summary}  \\subsection[]{GX301-2}  This system is clearly the brightest, most regular source studied here. The phase-binned folded lightcurve shows increasing flux levels consistent with phase 0.5, each of which reached type-I outburst levels. This result confirms previous work by Pravdo et al, 1995 and Koh et al, 1997 - both of these works also present evidence for apastron outbursts from this system. The power spectrum for GX301-2 shows a pattern of strong peaks which after comparison to simulated data are found to be real.  These frequency components, appearing at integer submultiples of the orbital period are a signature of the highly non-sinusoidal lightcurve. The spectrum shown is that generated after the removal of 4 particularly bright apastron flux points, as expected their inclusion boosts the $P_{orb}/2$ component. ",
        "conclusions": "Observations of accreting pulsars in high mass binary systems using the BATSE occultation technique reveal that this approach is sufficiently sensitive to derive accurate measurements of the X-ray modulation. Evidence for apastron X-ray outbursts is found in 2 sources: EXO2030+375 and 4U1145-619. All of these sources are in high mass, long period systems, EXO2030+375 and 4U1145-619 have similarities in their folded lightcurves having flux levels significantly above minimum for most of each orbit and apastron features reaching ~30-40 percent of peak flux. GX301-2 has a much flatter minimum level between emissions and the apastron emission is only ~10 percent of peak flux. These differences undoubtedly reflect the different nature of the companion star, whether Be or supergiant. The fact that none of the shorter period systems show evidence for this phenomenon may be explainable in terms of the standard Bondi-Hoyle accretion model if suitable parameters can be estimated/calculated for the outflow velocity and density of the circumstellar material/wind along the orbital trajectory of the neutron star. We hypothesise that in short period systems the radial profile of the stellar wind can never satisfy the required conditions for apastron outbursts due to the smaller apastron separation distances and higher neutron star velocities involved.  The FWHM of the orbital period peaks in the power spectra were investigated as a possible indicator of variability in the exact time at which outbursts \"switch on\". To within the uncertainties imposed by the signal/noise ratio of the data this effect was not detected in any of the sources."
    },
    "0209/astro-ph0209040_arXiv.txt": {
        "abstract": "The magnifications of the images in a strong gravitational lens system are sensitive to small mass clumps in the lens potential; this effect has been used to infer the amount of substructure in galaxy dark matter halos.  I study the theory of substructure lensing to identify important general features, and to compute analytic cross sections that will facilitate further theoretical studies.  I show that the problem of a clump anywhere along the line of sight to a lens can be mapped onto an equivalent problem of a clump in a simple convergence and shear field; clumps at arbitrary redshifts are therefore not hard to handle in calculations.  For clumps modeled as singular isothermal spheres (SIS), I derive simple analytic estimates of the cross section for magnification perturbations of a given strength. The results yield two interesting conceptual points. First, lensed images with positive parity are always made brighter by SIS clumps; images with negative parity can be brightened but are much more likely to be dimmed. Second, the clumps need not lie within the lens galaxy; they can be moved in redshift by several tenths and still have a significant lensing effect. Isolated small halos are expected to be common in hierarchical structure formation models, but it is not yet known whether they are abundant enough compared with clumps inside lens galaxies to affect the interpretation of substructure lensing. ",
        "introduction": "Strong gravitational lensing directly probes the mass distributions of cosmological objects. In quadruply-imaged quasars and radio sources, the flux ratios between the images provide a powerful test of the smoothness of galaxy mass distributions at redshifts $z \\sim 0.2$--1.  The popular Cold Dark Matter (CDM) paradigm predicts that galaxy dark matter halos are not smooth but rather full of small mass clumps (e.g., Klypin et al.\\ 1999; Moore et al.\\ 1999), which can perturb lens flux ratios in ways that cannot be explained by smooth lens models (e.g., Mao \\& Schneider 1998; Metcalf \\& Madau 2001; Brada\\v{c} et al.\\ 2002; Chiba 2002; Metcalf \\& Zhao 2002).  This ``substructure lensing'' effect has been detected both individually (Mao \\& Schneider 1998; Keeton 2001b; Brada\\v{c} et al.\\ 2002) and statistically (Dalal \\& Kochanek 2002a).  The statistical detection implies that $f_{\\rm clump} \\sim 2\\%$ (0.6--7\\% at 90\\% confidence) of galaxies' mass is in small clumps, and this result has been used to constrain the power spectrum on small scales (Dalal \\& Kochanek 2002b).  It suggests that the ``missing satellite'' problem in CDM (e.g., Klypin et al.\\ 1999; Moore et al.\\ 1999) is solved not by warm or exotic dark matter (e.g., Spergel \\& Steinhardt 2000; Colin, Avila-Reese \\& Valenzuela 2000; Bode, Ostriker \\& Turok 2001) but rather by astrophysical mechanisms that suppress star formation in small clumps (e.g., Bullock, Kravtsov \\& Weinberg 2000; Somerville 2002; Benson et al.\\ 2002). In other words, Dalal \\& Kochanek (2002a) argue, real galaxies contain numerous mass clumps that are predominantly dark.  The theory of substructure lensing is still young. To date, studies have relied mainly on Monte Carlo simulations and have explored relatively small parameter spaces.  To estimate the magnitude of the effect and demonstrate its measurability. Chiba (2002) fixed the clump population based on results from $N$-body simulations, while Metcalf \\& Madau (2001) adopted a clump mass function from simulations and varied the clump mass fraction $f_{\\rm clump}$ and the mass range. Dalal \\& Kochanek (2002a) made a measurement of $f_{\\rm clump}$ for a given clump mass, interpreting it as a weighted average over clump masses and an integral measure of small-scale power (Dalal \\& Kochanek 2002b). Further work is needed to understand how well studies of individual clumps in particular lens systems (e.g., Keeton 2001b) can provide a differential measure of the clump mass function and hence small-scale power. Even more importantly, the $f_{\\rm clump}$ that matters for lensing is the fraction of the surface mass density in clumps, which may neither be independent of position nor represent a global clump mass fraction.  The theory clearly needs to be further developed and explored before the claims from substructure lensing can be fully understood.  For that to happen, it would be extremely helpful if we could find an analytic or semi-analytic approach to the theory.  One important aspect of the theory has until now been neglected. All previous studies of substructure lensing have assumed that the clumps lie in the halos of lens galaxies, which seems reasonable in light of known mass clumps in real galaxies (globular clusters, satellite galaxies, etc.) and predictions from CDM. However, as lensing is sensitive to all mass along the line of sight we should at least consider the possibility of clumps outside the galaxy. This issue has important bearing on the interpretation of substructure lensing. Low-mass halos form in both CDM and WDM (Warm Dark Matter) scenarios, but in WDM they are disrupted when they merge with larger halos (e.g., Colin et al.\\ 2000; Knebe et al.\\ 2002). Thus, if the clumps responsible for substructure lensing are in the lens galaxy halos, they argue strongly for CDM and against WDM (Dalal \\& Kochanek 2002a, 2002b). If they can be isolated, however, their ability to rule out WDM may be weakened. Determining the relative importance of embedded and isolated clumps will require a detailed calculation of two effects: the number density of clumps in overdense regions like galaxies may be much higher than the number density of free-floating clumps (at least for CDM); but the portion of the line of sight that pierces the lens galaxy halo is a tiny fraction of the whole distance. That calculation is the focus of a separate paper (J.\\ Chen et al., in preparation) based on the formalism and general results presented here.  This paper examines analytic aspects of substructure lensing, in part to understand general features of the effect, and to obtain simple analytic estimates of the lensing cross sections to facilitate future theoretical studies. Section 2 focuses on the case of a singular isothermal sphere (SIS) as a simple lens model that is a reasonable representation of cosmological objects and yet analytically tractable. Finch et al.\\ (2002) recently gave a nice analysis of cross sections for multiple imaging for an SIS in an external shear. I extend their analysis by adding an external convergence and computing cross sections of interest for substructure lensing.  Section 3 places clumps in the context of substructure lensing and considers for the first time the problem of clumps along the line of sight.  I show that a clump at an arbitrary redshift is no more difficult to handle than a clump in the halo of the lens galaxy.  Sections 2 and 3 are independent of each other, but Section 4 combines them to give quantitative examples for two observed lenses. Some readers may wish at first to skip Sections 2 and 3, which are rather technical, and start with the examples and discussion in Section 4. Finally, Section 5 offers conclusions. ",
        "conclusions": "In the context of substructure lensing, the general problem of a mass clump anywhere along the line of sight to a strong lens can be mapped onto an equivalent problem of a clump in a simple convergence and shear field; thus, the theory can easily accommodate clumps at arbitrary redshifts. If a clump can be approximated as an SIS, the clump plus external field problem is analytically tractable, yielding simple formulas for the cross sections relevant for substructure lensing. Both of these results will make it possible to explore substructure lensing effects in large parameter spaces quickly and easily.  The analytic results already reveal two interesting features of substructure lensing. First, the sign of the perturbation depends on the parity of the original image. SIS clumps always make positive-parity images brighter, and they usually make negative-parity images fainter. There is some probability that SIS clumps can make negative-parity images brighter, but it is usually much smaller than the probability of dimming, except for clumps far in redshift from the lens galaxy. The qualitative difference between positive-parity and negative-parity images has been remarked by Metcalf \\& Madau (2001) for perturbations by galactic subclumps, and studied in detail by Schechter \\& Wambsganss (2002; also Witt, Mao \\& Schechter 1995) for perturbations by stars. The effect is qualitatively similar for the two types of perturbers, although it appears to be more dramatic for stars. It is a distinctive feature of substructure lensing (whether by stars or by larger, extended clumps) that will be useful for distinguishing the phenomenon from other systematic effects in lens flux ratios.  The second qualitative result is that a clump need not reside in the halo of the lens galaxy in order to have a significant lensing effect. Lensing is sensitive to clumps anywhere along the line of sight --- not uniformly, but with a redshift dependence that I have quantified.  The sensitivity function peaks at the lens redshift, but it has a width of several tenths in redshift and tails that extend to $z=0$ and to the source redshift.  If CDM is right, isolated small halos may be much less abundant (per unit volume) than clumps within galaxy halos, but they are viable over such a large volume that they may still be relevant for substructure lensing.  If WDM is right and galaxies are relatively smooth, isolated clumps could be the dominant cause of substructure lensing. Using predicted clump populations to evaluate the relative importance of isolated and embedded clumps is the subject of a work in progress (J.\\ Chen et al., in preparation).  This issue may ultimately affect how substructure lensing is interpreted, and how strongly it supports CDM and rules out WDM."
    },
    "0209/astro-ph0209106_arXiv.txt": {
        "abstract": "We present the relation of the spatial distribution of H$\\alpha$ kernels with the distribution of hard X-ray (HXR) sources seen during the 2001 April 10 solar flare. This flare was observed in H$\\alpha$ with the {\\it Sartorius} telescope at Kwasan Observatory, Kyoto University, and in hard X-rays (HXRs) with the Hard X-ray Telescope (HXT) onboard {\\it Yohkoh}. We compared the spatial distribution of the HXR sources with that of the H$\\alpha$ kernels. While many H$\\alpha$ kernels are found to brighten successively during the evolution of the flare ribbons, only a few radiation sources are seen in the HXR images. We measured the photospheric magnetic field strengths at each radiation source in the H$\\alpha$ images, and found that the H$\\alpha$ kernels accompanied by HXR radiation have magnetic strengths about 3 times larger than those without HXR radiation. We also estimated the energy release rates based on the magnetic reconnection model. The release rates at the H$\\alpha$ kernels with accompanying HXR sources are 16-27 times larger than those without HXR sources. These values are sufficiently larger than the dynamic range of HXT, which is about 10, so that the difference between the spatial distributions of the H$\\alpha$ kernels and the HXR sources can be explained. ",
        "introduction": "In the impulsive phase of a solar flare, precipitations of nonthermal electrons from the corona generate radiation from denser layers, such as the transition region and/or the upper chromosphere. This radiation is often observed in hard X-rays (HXRs) or microwaves. Precipitations of nonthermal electrons also cause radiation sources in H$\\alpha$ because of rapid thermalization or other mechanisms. Therefore, H$\\alpha$ kernels and HXR sources show a high correlation in their locations and light curves \\citep{Kita90}. However, the difference between the spatial distributions of H$\\alpha$ kernels and HXR sources is also well known. H$\\alpha$ images sometimes show elongated brightenings, called H$\\alpha$ flare ribbons, with many H$\\alpha$ kernels within them. The size of elemental H$\\alpha$ kernels is considered to be about $1^{\\prime\\prime}$ or even smaller \\citep{Kuro86}, which is larger than the spatial resolution achieved with H$\\alpha$ instruments ($\\sim 0^{\\prime\\prime}_{\\,\\cdot}2$) but is smaller than that with HXR instruments of about $5^{\\prime\\prime}$. On the other hand, HXR images show very few sources, sometimes only one. HXR sources are accompanied by H$\\alpha$ kernels in many cases, but many H$\\alpha$ kernels are not accompanied by HXR sources. The only exception to this ``lack of radiation sources in HXRs'' that is known, is the Bastille Day event on 2000 July 14 \\citep{Masu01}. This event shows a clear two-ribbon structure in HXRs such as in H$\\alpha$.  This difference of spatial distributions may be explained by the difference in radiation mechanisms between HXRs and H$\\alpha$. The HXR intensity is proportional to the number of accelerated electrons, and is thought to be proportional to the energy release rate \\citep{Hud91,Wu86}. Therefore, only compact regions where the largest energy release occurred are observable as HXR sources. On the other hand, the mechanisms for H$\\alpha$ radiation are much more complicated than those for HXR radiation, and to derive the effect of electrons is quite difficult \\citep{Ric83,Can84}. Some weaker H$\\alpha$ kernels may be caused by a secondary effect of precipitation or thermal conduction. However, as we mentioned above, the light curve of each H$\\alpha$ kernel has a high correlation with that of the total HXR intensity, even if their intensity is not so strong and they do not have an HXR counterpart. We suggest that the difference between the spatial distributions of H$\\alpha$ kernels and HXR sources is caused by the low dynamic range of the HXR data. In the HXR images only the strongest sources are seen, and the weaker sources are buried in the noise. We use the HXR data taken with the hard X-ray telescope (HXT; Kosugi et al. 1991) onboard {\\it Yohkoh} (Ogawara et al. 1991) in this paper. The dynamic range of the HXT images is about 10. Therefore, if the released energy at the H$\\alpha$ kernels associated {\\it with} HXR sources is at least 10 times larger than that at the H$\\alpha$ kernels {\\it without} HXR sources, then the difference of appearance can be explained.  In this paper, we measured the photospheric magnetic field strengths at each radiation source seen in H$\\alpha$ images which have much higher spatial resolution than HXR images. We also estimated the energy release rates based on the magnetic reconnection model \\citep{Iso02} at each radiation source. Then we compared the energy release rates with the spatial distribution of radiation sources in an HXR image since they suggest the site where the strong energy release occurred. To examine the difference of the amount of the released magnetic energy, we measured the photospheric magnetic field strengths at each radiation source. In \\S 2 we summarize the observational data and the results. In \\S 3 we discuss the amount of energy release at each radiation source. In \\S 4 the summary and the conclusion are given. ",
        "conclusions": "We have examined the difference between the spatial distributions of H$\\alpha$ kernels and HXR sources. The HXR sources indicate where large energy release has occurred, while the H$\\alpha$ kernels show the precipitation sites of nonthermal electrons with higher spatial resolution. We measured the photospheric magnetic field strength at each H$\\alpha$ kernel, and found that the magnetic field strengths at the H$\\alpha$ kernels accompanied by HXR sources are about 3 times higher than those at the other H$\\alpha$ kernels (without any HXR sources). We also estimated their energy release rates, by using the photospheric magnetic field strengths and by considering the dependence of $v_{i}$ on $B$ as derived by some authors (e.g. Sweet 1958; Parker 1957; Petschek 1964). The estimated energy release rates at the HXR sources are large enough to explain the difference of appearance of the H$\\alpha$ and HXR images.  The gap in our understanding of large scale structure such as H$\\alpha$ flare ribbons and compact radiation sources has been well known not only in HXRs, but also in microwaves. Even in SXRs, flare loops often appear only in a part of the whole flaring region and/or seem to connect only a few radiation sources. The lack of radiation sources does not mean that no energy release occurs there. We just cannot ``observe'' the radiation because of the dynamic ranges of the instrument for those wavelengths.  We compared the photospheric magnetic field strenghts at each H$\\alpha$ kernel with the spatial distribution of HXR sources, and found that HXR sources appear at strong magnetic regions. However, our result was based on only one event, and the statistical studies about the relation between HXR intensities and magnetic field strengths are needed. We will be able to perform more detailed analysis in the near future by using HXR images, with a higher dynamic range, obtained with the {\\it Reuven Ramaty High Energy Solar Spectroscopic Imager} ({\\it RHESSI})."
    },
    "0209/astro-ph0209330_arXiv.txt": {
        "abstract": "We construct explicit models of multi-field inflation in which the primordial metric fluctuations do not necessarily obey Gaussian statistics. These models are realizations of mechanisms in which non-Gaussianity is first generated by a light scalar field and then transferred into curvature fluctuations. The probability distribution functions of the metric perturbation at the end of inflation are computed. This provides a guideline for designing strategies to search for non-Gaussian signals in future CMB and large scale structure surveys. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/physics0209007_arXiv.txt": {
        "abstract": "We investigated the nanometer scale height fluctuations of 3T3 fibroblast cells with the atomic force microscope (AFM) under physiological conditions. Correlation between these fluctuations and lateral cellular motility can be observed. Fluctuations measured on leading edges appear to be predominantly related to actin polymerization-depolymerization processes. We found fast (5 Hz) pulsatory behavior with 1--2 nm amplitude on a cell with low motility showing emphasized structure of stress fibres. Myosin driven contractions of stress fibres are thought to induce this pulsation. ",
        "introduction": "The motility of animal cells is dominated by actin-myosin-based contraction and actin polymerization-based protrusion. The two basic types of protrusions, lamellipodia and filopodia are driven by actin polymerization-depolymerization processes \\cite{a1,a2,a3}. The physical theory of such biological motilities employing the tools of statistical physics \\cite{a4,a5} gives an idea how it works but it still needs to be developed.  A number of cellular activities can cause height fluctuations on the time and distance scales that we investigated. Actin polymerization and actin-myosin based contractions represent only one class. The rearrangement of the structure built up from intermediate filaments (IF) under the plasma membrane can be another source of vertical fluctuations. IF-s provide mechanical stability to animal cells. Any significant weakening of the IF array alters at least locally the elastic properties of the cell leading to increased susceptibility to intrinsic or extrinsic forces. Assembly or disassembly of large protein complexes in the plasma membrane or their lateral motion under the AFM tip will also result in vertical motility, not to mention endo- and exocytotic activity. Intracellular transport processes can have an effect on the vertical fluctuations, as well.  Cellular motility on the micrometer scale has been extensively investigated with video-microscopy (\\cite{a6,a7,a8} and references therein). Spatial resolution provided by optical microscopy of living cells, however does not enable researchers to observe nanometer scale motion and rearrangement of cell components. AFM is an adequate tool for such measurements \\cite{a9,a10,a11,a12}. Stress fibres (contractile bundles of actin filaments and myosin-II) play an important role in the control of cell shape and the adhesion of cells to the extracellular matrix through focal contacts. These characteristic cytoskeletal elements can be imaged with the AFM due to their high elastic modulus \\cite{a13,a14}. AFM is capable not only for recording high-resolution topographic images of living cells but also for measuring the elastic properties of them simultaneously \\cite{a15} and investigating cellular dynamics \\cite{a16,a17}.  By repetitively scanning on the surface of a cell time-lapse images can be recorded \\cite{a18,a19}. The analysis of subsequent images yielding a movie is highly informative in terms of the kinetics of the cytoskeleton. Although fast cellular motility can not be examined by the repetitive scanning procedure due to the minute-range of scanning time, nanometer scale fast motion can be probed by positioning the tip on the area of interest. In this way vertical fluctuations can be investigated almost without a limitation of the time scale.  Close to the molecular scale the motion of cells is dominated by stochastic fluctuations. With the AFM we tried to shed light on the linkage between microscopic fluctuations and organized motility. ",
        "conclusions": "The analysis of height fluctuations acquired at different locations allows a sensitive monitoring of the motility of cellular components. Both actin-myosin based contractions and actin polymerization-based filopodial and lamellipodial protrusions can be examined by this method. We found a correlation between the characteristics of vertical fluctuation and organized lateral locomotion.  We explain the observed 5 Hz pulsation of a cell with the periodic contractions of stress fibres. This type of oscillation cannot be easily identified by other techniques due to its low amplitude. Although the frequency of mechanical pulsation of cardiomyocytes is in the same frequency range ($\\sim$1.25 Hz), its amplitude is 2 orders of magnitude higher \\cite{a16}. Spontaneous oscillatory contractions of muscle fibers with a period of a few seconds are widely known for several years (e.g. \\cite{a20}). Theoretical models can explain spontaneous oscillation under certain conditions \\cite{a21}).  Slow pulsation of non-muscle cells has been observed in several cases. Microtubule depolymerization can induce rhythmic actomyosin-based contractility with a period of $\\sim$50 s in fibroblasts \\cite{a22} and oscillatory activity in the cortical microfilament system of lymphoblasts \\cite{a23}. Shape oscillations of leukocytes driven by cyclic actin polymerization has been studied by several groups \\cite{a24}. The period of this process is about  $\\sim$8 s.  The cortical tension of non-muscle cells generated by myosin-II can drive a change of shape \\cite{a25}. Myosin molecules cycle about 5 times in a second in muscle \\cite{a26}. Based on the above mentioned facts, we think that a synchronized behavior of myosin molecules in stress fibres may cause the observed pulsation. Myosin synchronization has been theoretically predicted close to the isometric condition in highly organized actin structures \\cite{a27,a28}. Further experiments are needed to elucidate the background of this phenomenon. Using drugs affecting a specific system of the cytoskeleton will help to distinguish their roles in the nanometer scale fluctuations of cells."
    },
    "0209/astro-ph0209383_arXiv.txt": {
        "abstract": "Among the billions of planetary systems that fill the Universe, we would like to know how ours fits in. Exoplanet data can already be used to address the question: How common are Jupiters?  Here we discuss a simple analysis of recent exoplanet data indicating that Jupiter is a typical massive planet rather than an outlier. A more difficult question to address is: How common are Earths? However, much indirect evidence suggests that wet rocky planets are common. ",
        "introduction": "Long before we detect Earth-like planets we will have a good general picture of the variety of massive planets in planetary systems. Since Jupiter is the most prominent feature of our planetary system, and our knowledge of other planetary systems is still rudimentary, we may, with current data reasonably hope to answer the less ambitious question:  How typical is Jupiter? The relevant analysis is now possible because a statistically significant sample is starting to emerge from which we can determine meaningful distributions in mass and period. To quantify these distributions as accurately as possible, we have identified a subsample of exoplanets that is minimally affected by the selection effects of the Doppler detection method (Fig. 1). Within this subsample, after a simple completeness correction, we quantify trends in mass and period that are less biased than trends based on the full sample of exoplanets. Straightforward extrapolations of these trends, into the area of parameter space occupied by Jupiter, indicates that Jupiter lies in a region densely occupied by exoplanets (Lineweaver \\& Grether 2002). Our analysis suggests that Jupiter is more typical than indicated by previous analyses. For example, instead of $M_{Jup}$ planets being twice as common as $2 M_{Jup}$ planets, we find they are 3 times as common. The latest exoplanet data (detected between January and August 2002) supports and strengthens this conclusion (Lineweaver, Grether \\& Hidas 2003). Our claims for Jupiter being a typical massive planet are well-defined in terms of $Msin(i)$ and period but can not yet include orbital eccentricity since the eccentricity of most exoplanets is larger than the $\\sim 0.1$ typical of our Solar System.  The frequency of Jupiter-like planets may have implications for the frequency of life in the Universe. A Jupiter-like planet shields inner planets from an otherwise much heavier bombardment by planetesimals, comets and asteroids \\clearpage \\begin{figure}[h,t!] \\plotone{LineweaverGretherHidasFig1.eps}   % \\caption{Comparison of the 101 exoplanets detected as of August 2002 to the mass and period of the planets of our Solar System. We would like to know how planetary systems in general are distributed in this plane. This figure shows that the Doppler technique has been able to sample a very specific high-mass, short-period region of the log P - log Msin (i) plane. Thus far, this sampled region does not overlap with the 10 times larger area of this plane occupied by the nine planets of our Solar System. Thus, there is room in the $\\sim 95\\%$ of target stars with no Doppler-detected planets, to harbor planetary systems like our Solar System. Null results from microlensing searches have been used to constrain the frequency of Jupiter-mass planets (Gaudi et al. 2002). Less than 33\\% of the lensing objects (presumed to be Galactic bulge M-dwarfs) have planetary companions within the dashed wedge-shaped area (the period scale, but not the AU scale, is applicable to this area). For more details see Lineweaver and Grether 2002. } \\end{figure} \\clearpage \\begin{figure} \\plotfiddle{LineweaverGretherHidasFig2.eps}{8cm}{0}{70}{60}{-180}{0} \\caption{ How common is liquid water in the Universe? In this phase diagram of water the scatter of the bodies of our Solar System is large. In the absence of any systematic pattern avoiding the liquid water region, and to the extent that other planetary systems share such random scatter, we can expect wet rocky planets to exist in a substantial fraction of planetary systems in the Universe. Diagram adapted from www.sbu.ac.uk/water/phase.html} \\end{figure}  \\noindent during the first billion years after formation of the central star. Wetherill (1994) has estimated that Jupiter significantly reduced the frequency of sterilizing impacts on the early Earth during the important epoch $\\sim 4$ billion years ago when life originated on Earth. The removal of comet Shoemaker-Levy by Jupiter in 1994, is a more recent example of Jupiter's protective role. However, we know so little about the details of the biochemistry of biogenesis that impacts may also play a constructive (not necessarily destructive) role in biogenesis. ",
        "conclusions": ""
    },
    "0209/astro-ph0209510_arXiv.txt": {
        "abstract": "We simulate convection near the solar surface, where the continuum optical depth is of order unity.  Hence, to determine the radiative heating and cooling in the energy conservation equation, we must solve the radiative transfer equation (instead of using the diffusion or optically thin cooling approximations).  A method efficient enough to calculate the radiation for thousands of time steps is needed.  We assume LTE and a non-gray opacity grouped into 4 bins according to strength.  We perform a formal solution of the Feautrier equation along a vertical and four straight, slanted, rays (at four azimuthal angles which are rotated 15 deg. every time step).   We present details of our method.   We also give some results: comparing simulated and observed line profiles for the Sun, showing the importance of 3D transfer for the structure of the mean atmosphere and the eigenfrequencies of p-modes, illustrating Stokes profiles for micropores, and analyzing the effect of radiation on p-mode asymmetries. ",
        "introduction": "Radiative energy exchange is critical in determining the structure of the upper convection zone.  Near the surface of the Sun, the energy flow changes from almost exclusively convective below the surface to radiative above the surface.  The interaction between convection and radiation near the surface determines what we observe, drives the convection, and generates the magnetic and non-magnetic activity which heats the chromosphere and corona.  Hence, the interaction between convection and radiation has significant impact on both the dynamics of convection and our diagnostics.  Escaping radiation cools the plasma that reaches the surface, which produces the low entropy plasma whose buoyancy work drives the convection.  Radiation transport determines (with convection and waves) the mean atmospheric structure.  Radiation provides the diagnostics that we use to determine the velocity, density, temperature and magnetic field on the Sun.  Radiation transport modifies the observable p-mode temperature fluctuations so as to reverse the asymmetry of the intensity vs. velocity spectral peaks.  Since the top of the convection zone occurs near the level where the continuum optical depth is one, neither the optically thin nor the diffusion approximations give reasonable results.  We need to solve the 3D, LTE, non-gray radiation transfer in our models (Nordlund 1982). ",
        "conclusions": "In order to make realistic numerical simulations it is necessary to include the relevant realistic physics.  In the case of the solar atmosphere this means including an appropriately realistic treatment of the radiative losses.  Many techniques are available to solve for radiation transport.  The problem is to devise a method that is efficient enough to be applied at each time step in a dynamic calculation.  We have described a method developed by Nordlund that relies on reducing the number of frequencies that need to be treated to a bare minimum of 4, by using a multi-group binning of the wavelengths according to their opacity.  In addition a minimal number of rays are employed, but they are rotated to cover the computational domain. Because we are only interested in the photosphere, we also can assume LTE.  Some of the approximations we use, such as assuming that all the opacities have the same depth dependence, can be easily improved.  It was instituted in a time when computer memory was small and the table size needed to be minimized.  It would be nice to replace the approximation of multi-group opacities with an opacity sampling method, so as to include the effect of Doppler shifts on the radiation absorption, but the cost will be many more frequencies that need to be solved.  For the Sun we are lucky that the radiation heating/cooling time scale is of the same order as the dynamical time scale, so that an explicit solution of the radiation transport is possible.  For giant stars, for instance, where the dynamical time scale is much longer than the Sun's, explicit solution of the radiation transport will reduce the possible time step size.  For other investigations, other physics will be needed.  To include the chromosphere it is necessary to include at least the dominant non-LTE effects of scattering and hydrogen ionization.   This is the next project we are working on in collaboration with Mats Carlsson, Viggo Hansteen and Andrew McMurray of Oslo University."
    },
    "0209/hep-th0209133_arXiv.txt": {
        "abstract": "We investigate the generation of primordial gravitational waves from inflation in braneworld cosmologies with extra dimensions. Advantage of using primordial gravitational waves to probe extra dimensions is that their theory depends only on the geometry, not on the microscopic models of inflation and stabilization. $D(D-3)/2$ degrees of freedom of the free bulk gravitons are projected onto the 3d brane as tensor, vector and scalar modes. We found the following no-go results for a generic geometry of a five (or D) dimensional warped metric with four dimensional de Sitter (inflationary) slices and two (or one) edge of the world branes: Massive KK graviton modes are not generated from inflation (with the Hubble parameter H) due to the gap in the KK spectrum; the universal lower bound on the gap is $\\sqrt{3/2}\\, H$. Massless scalar and vector projections of the bulk gravitons are absent, unlike in geometries with KK compactification.  A massless 4d tensor mode is generated from inflation with the amplitude $H/M_P$, where $M_P$ is the effective Planck mass during inflation, derived from the $D$ dimensional fundamental mass $M_S$ and the volume of the inner dimensions. However, $M_P$ for a curved dS braneworld may differ from that of the flat brane at low energies, either due to the $H$-dependence of the inner space volume or variations in the brane separation before stabilization. Thus the amplitude of gravitational waves from inflation in braneworld cosmology may be different from that predicted by inflation in 4d theory. ",
        "introduction": "\\label{sec:introduction}  Higher dimensional formulation of superstring/M theory, supergravity and phenomenological models of the mass hierarchy has most obvious relevance to cosmology. Early universe inflation can universally amplify vacuum fluctuations of all light (minimally coupled to gravity) degrees of freedom. Thus inflation serves as a powerful tool to probe the fields associated with the extra dimensions.  In this paper we consider a class of models with extra dimensions related to the braneworld scenarios. One of the most interesting recent developments of the high energy physics phenomenology is the picture of the brane worlds, where our 3+1 dimensional spacetime is the 3d brane embedded in the higher dimensional space. In application to the very early universe this leads to the brane world cosmology, where our 3+1 dimensional universe is the 3d curved brane embedded in the bulk. A lot of efforts have been devoted to the search for the new features of the background cosmological expansion, inflation in the brane world settings, cosmological fluctuations which could probe the brane world scenarios and others. Many of these problem are rather complicated. For instance, the problem of gravitational waves which might be observed on the brane requires treatment of all ${1 \\over 2}D(D-3)$ degrees of freedom of the bulk gravitons, the choice of their initial state, the issue of the KK modes, transplanckian problem and others.  Cosmological gravitational waves are a very useful tool to probe the very early universe, because their generation depends on the background geometry but not on the details of the particle physics built in the model. A historic example is the theory of cosmological gravitational waves generated from the early de Sitter stage \\cite{starob} which, ironically, was suggested even before the microscopic models of inflation were proposed.  Gravitons almost freely propagate through the expanding universe, and may leave their prints on the power spectrum and polarization of the cosmic microwave background. With the ever increasing accuracy of the CMB anisotropy experiments, it is worth to enlarge the pool of the primordial gravitational waves theories.  In the familiar model of inflation in the four dimensions, gravitational waves are generated from vacuum fluctuations during inflation. Gravitons have two polarizations, and temporal part of their wave function $h_k(t)$ obeys an equation which is similar to the one for the massless scalar field minimally coupled to gravity.  Late time asymptotics of $h_k(t)$, which is relevant to present cosmological observations, contains long wavelength modes $k$. They have (almost) scale free spectrum $k^3 \\vert h_k\\vert^2 \\approx \\text{const}$. The dimensionless almost constant amplitude \\begin{equation} \\label{ampl1} k^{3/2} h_k = {H \\over M_P}, \\end{equation} is defined by the ratio of the only two scales in the problem, which are related to the gravitational waves: the Hubble parameter during inflation $H$ and the Planck mass $M_P$ at low energies. Usually the Hubble parameter slightly varies during inflation, $H=H(t)$, which brings slight tilt to the spectrum of gravitational waves (in the {\\it r.h.s.} of (\\ref{ampl1}) $H \\to H(k)$). In this paper we will assume $H=\\text{const}$, but will devote some discussion to the case when $H$ slightly varies.  In this paper we will study the problem of the primordial gravitational waves generation from the early universe in the theories with large extra dimensions. As opposed to excitations of some other fields, gravitational waves propagate in the bulk. What information about the bulk/brane geometry can be, in principle, observable on the 3d brane through the gravitational excitations? The problem has the flavor of the well known mathematical quest of ``how one can hear the shape of a drum?'' \\cite{drum}.  To make geometrical setting definite, we assume the brane world inflation. We will consider geometry of five dimensional warped metric with four dimensional de~Sitter slices \\begin{equation} \\label{warp} ds^2 = dy^2 + A^2(y)\\, ds_4^2 \\ . \\end{equation} The warp factor $A(y)$ is determined self-consistently by the five dimensional Einstein equations, supplemented by the brane boundary conditions. This formalism is briefly outlined in Section \\ref{sec:2}. For the sake of generality, we also present results for a more general case of $D$ dimensional warped metric with $D-1$ dimensional de~Sitter slices. The class of metrics (\\ref{warp}) covers many important brane world scenarios \\cite{brane} including the Horawa-Witten theory, HW with with curved branes, the Randall-Sundrum (RS) models, RSI model with phenomenological stabilization of two branes, scalar sector of the supergravity realization of the RS model and others. The presence of a bulk scalar field $\\phi$ (a dilaton) with the potential $V(\\phi)$ or several scalars is often assumed.  The warped metric (\\ref{warp}) can be rewritten in ``conformally-factorized'' form \\begin{equation} \\label{warp2} ds^2 = \\Omega^{-2} \\left( d\\chi^2 + ds_4^2 \\right), \\end{equation} where conformal factor and conformal fifth coordinate are defined by \\begin{equation} \\Omega(\\chi) = \\frac{1}{A(y(\\chi))},\\hspace{1em} \\chi = \\int \\frac{dy}{A(y)}. \\end{equation} In our treatment of gravitational waves, beginning in Section \\ref{sec:3}, we will extensively use conformal mapping of the whole geometry (\\ref{warp2}), bulk and branes on its edges, into a simpler manifold with boundaries, with the metric \\begin{equation} \\label{conf} d{\\hat s}^2 = \\Omega^2\\, ds^2 = d\\chi^2 + ds_4^2. \\end{equation}  There are several important papers on the formalism of metric fluctuations \\cite{form} and properties of the gravitational waves in the brane world scenarios \\cite{HHR,LMW,GKL}, and will try to make connection with them in the course of the paper. We will work with the metric in the form (\\ref{warp2}), which allows separation of variables in gravitational wave equation, as it will be shown in Section \\ref{sec:4}. Conformal mapping also significantly simplifies the issue of the boundary conditions, as it will be explained in Section \\ref{sec:5}. Just as it was in the four dimensional case, in $D$ dimensional theory the graviton eigenfunctions $h_{k,m}(\\chi)$ obey an equation for the massless scalar minimally coupled to gravity.  In Section \\ref{sec:6}, we show that the spectrum of nonzero KK modes has a mass gap with the universal lower bound $ \\sqrt{\\frac{3}{2}}\\, H$, independently on the bulk bulk scalar fields potential $V(\\phi)$ and the presence or absence of the second brane. This result generalizes the finding of the ${3 \\over 2}\\, H$ gap in the Randall-Sundrum II model with a single brane in the AdS bulk \\cite{LMW}.  In Section \\ref{sec:graviton}, we show that the homogeneous zero KK mode $m=0$ is degenerate, and can be described by only two physical polarization degrees of freedom, while other three can be gauged by zero by the residual coordinate transformation. Thus, the issue of the gravi-scalar and gravi-vector modes in brane cosmology is resolved: they do not contribute.  All together this allows us to conclude that the late time asymptotic of the long-wavelength gravitational waves generated from the brane inflation is defined by the two tensor polarization of the homogeneous KK mode $m=0$, similar to what takes place in a case of inflation in four dimensions. However, as we will discuss in Section \\ref{sec:8}, it does not mean that gravitational waves from the brane inflation in all respects are identical to those in the four dimensional theory. Essential difference arises in their amplitudes. Indeed, in five dimensional theories there is additional scale $\\ell$, associated with the fifth dimension (brane separation in the HW theory, the AdS radius in the RS models, etc.) With three scales in the problem, $\\ell$, $H$ and $M_P$, one may have a dimensionless amplitude different from (\\ref{ampl1}). Indeed, for the braneworld inflation, one finds $k^{3/2} h_k=\\frac{H}{M_P} F(H\\ell)$, with some function $F$. Remarkably, the amplitude can be expressed economically in terms of the effective Planck mass for curved (de Sitter) braneworld \\begin{equation} \\label{ampl2} k^{3/2} h_k = {H \\over M_{P,\\text{eff}}} \\ . \\end{equation} Effective Planck mass $M_{P,\\text{eff}}$ during inflation in the braneworld scenario may differ from the current value $M_P \\sim 10^{19}\\,$GeV, therefore prediction for the gravitational wave amplitude from brane inflation may differ from that from inflation in 4d theory.  In Section \\ref{sec:9}, we will discuss implications of our results to the observational test of the ratio of tensor to scalar modes, transplanckian problem in the context of models with extra dimensions, and other related issues.  We evoke the ``conformally-factorized'' frame (\\ref{warp2}) mainly for treatment of the gravitational waves. However, conformal mapping of the brane worlds has merits of its own. In particular, we show that the warp geometry with branes at the edges is conformally equivalent to the portion of the spacetime with regular end of the world edges. For the case of flat 4d slices of (\\ref{warp2}), the world edges after conformal mapping are empty. ",
        "conclusions": "\\label{sec:9}  Among the many features of superstring/M theory, the existence of extra dimensions has the most direct relevance to cosmology. Gravitons propagate in all -- outer and inner -- dimensions. Bulk gravitons in $D$ dimensions have $\\frac{1}{2}D(D-3)$ components. Only two of them appear as a transverse traceless tensor mode from the point of view of a four dimensional observer, while the others are projected into the $3+1$ dimensional world as (gravi)scalars and (gravi)vectors.  Vacuum fluctuations of all light (minimally coupled) degrees of freedom are unstable during inflation and appear as long-wavelength perturbations in late time cosmology. Thus, outer space inflation acts like an amplifier and stretcher of light modes. Using inflation as a tool, we may probe extra dimensions by examining light modes associated with extra dimensions which may be produced by inflation.  It is especially appealing to investigate gravitational waves in the context of extra dimensions plus inflation. This is because the theory of small gravitational fluctuations depends on the underlying geometry but not on the microscopic model of its realization. Additionally, gravitons propagate almost freely throughout the evolution of the universe, thus carrying information about the primordial universe.  There are two distinct types of scenarios with compactification of extra dimensions from the point of view of the theory of primordial gravitational waves. One type is Kaluza-Klein compactification. The theory of gravitational waves from inflation in models with KK compactification will be considered separately. Another type of model arises from the braneworld scenarios with heavy branes (including a single brane in AdS bulk or compact warped inner space in between two branes). In this paper we investigated primordial gravitational waves in generic geometries of the warped inner space with outer space (brane) inflation.  We found the following no-go results for gravitational waves from inflation in the generic case of these braneworld scenarios:  Massless mode of bulk gravitons is projected to the usual four dimensional two component TT tensor mode, while their scalar and vector projections are absent. This is different from KK compactification where light bulk gravitons are projected to tensor, vector and scalar massless components in the four dimensional world.  Massive KK modes are not generated from inflation due to the gap in the KK spectrum. The gap is present in all models including models with two branes. The universal lower bound of the gap is \\begin{equation} \\Delta m \\geq \\sqrt{\\frac{3}{2}}\\, H. \\end{equation} This result generalizes the finding of the gap (of magnitude $\\frac{3}{2} H$) for the RS model with a single brane embedded in AdS \\cite{LMW}.  Among all massless projections of the bulk gravitons, only the four dimensional TT massless modes are generated from brane inflation. As in the standard four dimensional theory, their spectrum is scale free, but the amplitude is given by the formula \\begin{equation} k^{3/2} h_k=\\frac{H}{M_{P,\\text{inf}}}, \\end{equation} where $M_{P,\\text{inf}}$ is the four dimensional Planck mass for the curved (de Sitter) observable brane. Suppose the inner dimensions are stabilized after inflation (say due to SUSY breaking). This is not too radical an assumption bearing in mind that the usual energy scale of inflation is $H \\sim 10^{13}$ GeV and the usual energy scale of SUSY breaking is $\\sim 10^{11}$ GeV. Therefore theoretically it is not necessary that $M_{P,\\text{inf}}$ during inflation coincides with present day low energy $ M_{P}$. The difference may come from variation of the size of the inner dimension(s) (i.e. the limits of integration in formula (\\ref{size}), or from an imprint of the brane curvature $H$ on the warp factor $A(y)$. A change in $M_{P,\\text{eff}}$ by a factor of $2$ is already interesting. Thus in general, the connection between gravitational wave amplitudes and energy scale of inflation $H$ in theories with extra dimensions may be different from that in four dimensional theory.  Most important test of the primordial cosmological fluctuations from inflation is the ratio between the amplitude of tensor and scalar perturbations $T/S$. To calculate scalar perturbations, one needs to specify the model of inflation (say in terms of the scalar field potential at the brane). Already the simplest model of inflation in the RSI scenario without bulk scalars predicts that $T/S$ may be different from that predicted in the four dimensional theory. The theory of scalar perturbations in braneworld scenarios with bulk scalars is even more involved. In general, we expect that $T/S$ from braneworld inflation will be model-dependent and depend on many parameters.  Finally, we will make a comment on the transplanckian problem in the context of extra dimensions plus inflation. The idea beyond this problem is the following \\cite{trans}: In an expanding universe the physical momentum of quantum fluctuations is redshifted to smaller values. Going backwards in time, we have to deal with ever increasing momenta which at some instant will be equal to the fundamental mass scale $M_S$. At this point and beyond the quantum field theory approach to fluctuations in an expanding universe (say during inflation) should be replaced by a string theory description, which is not yet available. For quantum fluctuations from inflation in the 4d theory there is, however, a convenient phenomenological coding of the problem in terms of the Bogolyubov coefficients $\\alpha_k, \\beta_k$ for the initial vacuum state: instead of the vacuum eigenmode $f_k(\\tau)$ at infinity (given in terms of the Hankel function (\\ref{temp1})) one can use the mode function \\begin{equation}\\label{mode} f_k(\\tau) \\to \\alpha_k f_k(\\tau)+\\beta_k f_k^*(\\tau), \\hspace{1em} |\\alpha_k|^2-|\\beta_k|^2, \\end{equation} and discuss further the effects from $\\beta_k$, if any. If the sizes of the extra dimensions are larger than $M_S^{-1}$ which takes place in many phenomenological schemes (like the HW theory), then we must simultaneously deal with extra dimensions and the transplanckian problem. It is, however, straightforward to combine the formalism of cosmological fluctuations and that of the transplanckian problem. If the sizes of the extra dimensions are large enough to work there in the supergravity approximation, we can use the quantum field theory of fluctuations in the bulk, as we did in this paper. Then the transplanckian story will again be coded in terms of $\\beta_k$ for bulk quantum eigenmodes (of gravitons, for instance)."
    },
    "0209/astro-ph0209218_arXiv.txt": {
        "abstract": "{A significant fraction of galaxy clusters exhibits cluster wide radio halos. We give a simple prediction of the local and higher redshift radio halo luminosity function (RHLF) on the basis of (i) an observed and a theoretical X-ray cluster luminosity function (XCLF) (ii) the observed radio--X-ray luminosity correlation (RXLC) of galaxy clusters with radio halos (iii) an assumed fraction of $f_\\ha \\approx \\frac{1}{3}$ galaxy clusters to have radio halos as supported by observations.  We then find 300-700 radio halos with $S_{\\rm 1.4 GHz}> 1\\,$mJy, and $10^5-10^6$ radio halos with $S_{\\rm 1.4 GHz}> 1\\,\\mu$Jy should be visible on the sky.  14\\% of the $S_{\\rm 1.4 GHz}> 1\\,$mJy and 56\\% of the $S_{\\rm 1.4 GHz}> 1\\,\\mu$Jy halos are located at $z>0.3$. Subsequently, we give more realistic predictions taking into account (iv) a refined estimate of the radio halo fraction as a function of redshift and cluster mass, and (v) a decrease in intrinsic radio halo luminosity with redshift due to increased inverse Compton electron energy losses on the Cosmic Microwave Background (CMB). We find that this reduces the radio halo counts from the simple prediction by only 30 $\\%$ totally, but the high redshift ($z>0.3$) counts are more strongly reduced by 50-70\\%. These calculations show that the new generation of sensitive radio telescopes like LOFAR, ATA, EVLA, SKA and the already operating GMRT should be able to detect large numbers of radio halos and will provide unique information for studies of galaxy cluster merger rates and associated non-thermal processes. ",
        "introduction": "} \\subsection{Cluster radio halos\\label{sec:crh}}  The X-ray emitting intra-cluster medium (ICM) of a significant fraction of galaxy clusters also exhibits cluster wide radio emission, the so called {\\it cluster radio halos} \\citep[ for recent samples]{1996IAUS..175..333F, 1999NewA....4..141G, 2001ApJ...548..639K, 2000NewA....5..335G}. Cluster radio halos are central, extended over cluster-scales, unpolarised, and steep spectrum radio sources not associated with individual galaxies. It is recognised that radio halos appear in clusters which have recently undergone a major merger event \\citep{1993MNRAS.263...31T,2001ApJ...553L..15B}.  Whereas the cluster X-ray emission is due to thermal electrons with energies of several keV, the emission of the radio halo is due to synchrotron radiation of relativistic electrons with energies of $\\sim 10$ GeV in $\\sim\\mu$G magnetic fields.  The spatial distribution of the radio emission often seems to follow closely (and nearly linearly) on the large scale the distribution of the X-ray emission \\citep{2001A&A...369..441G}. In a few cases, where a cluster merger is in its early stage, detailed observations indicate that the radio halos seem to be spatially restricted to hot merger-shocked regions \\citep{2001ApJ...559..785K,2001ApJ...563...95M}.  The similarity of X-ray and radio morphologies of radio halo galaxy clusters indicates a connection between the energetics of the non-thermal component (magnetic fields and relativistic electrons) and the thermal ICM gas.  This is also supported by the strong correlation of radio halo luminosity and the host cluster X-ray luminosity \\citep[the RXLC,][ also see Fig.~\\ref{fig:lnulX}]{2000ApJ...544..686L, Feretti.Pune99}.  Since most of the thermal cluster gas was heated in cluster accretion and cluster merger shock waves \\citep[e.g.][]{1972A&A....20..189S, 1998ApJ...502..518Q, 2000ApJ...542..608M} one would suspect that also the relativistic electrons received their energy from these shocks.  The radiative lifetime of the radio emitting electrons is of the order of 0.1 Gyr \\citep[e.g.][]{1977ApJ...212....1J}. This is short compared to the shock crossing time in merger events, which is of the order of 1 Gyr. If the electrons were accelerated in the shock waves, and just are cooling behind them, the radio emission would not follow the X-ray emission, as observed in late stage merger clusters, but should be more patchy and only located close to the shock waves\\footnote{Such patches of radio emission, the so called {\\it cluster radio relics}, are indeed observed in merging clusters. They are interpreted to be either emission from shock accelerated ICM electrons \\citep{1998AA...332..395E, 1999ApJ...518..603R, 2001ApJ...562..233M} or from shock revived fossil radio cocoons \\citep{2001A&A...366...26E, 2002MNRAS.331.1011E}.}. In order to have a radio halo in the post shock region, which lasts sufficiently long to explain the X-ray emission like morphology of radio halos in later stage mergers, some fraction of the shock released energy has to be stored in some form and later given to the relativistic radio emitting electron population.   \\begin{figure}[t] \\begin{center} \\psfig{figure=lnulX.ps,width=\\figsize\\textwidth,angle=0} \\end{center} \\vspace{-0.7cm}\\caption[]{\\label{fig:lnulX} X-ray and radio luminosity of cluster of galaxies with radio halos. Data is from \\cite{Feretti.Pune99} and \\cite{2001A&A...376..803G} and the correlation power-laws are given in Sect.~\\ref{sec:corr}.  } \\end{figure}  \\subsection{Halo formation scenarios}  A suggestion for such an energy storing agent is turbulence within the cluster which may re-accelerate a low energy relativistic electron population against radiative losses \\citep[][ and many others]{1977ApJ...212....1J}. Such a primary electron model seems to be favoured observationally by spectral index steepening towards higher frequencies as observed in the case of the Coma cluster radio halo \\citep[][]{1987A&A...182...21S, 2001MNRAS.320..365B}.  Another suggestion is a shock accelerated population of relativistic protons. Over their long lifetimes they are able to inject the necessary radio emitting relativistic electrons by charged pion decay after hadronic interactions with the thermal ICM nucleons \\citep[][ and others]{1980ApJ...239L..93D}. Such a hadronic scenario for radio halo formation was shown to lead naturally to a very steep RXLC \\citep{Ringberg99Colafrancesco,2000A&A...362..151D,2001ApJ...562..233M}, as observed.  Such a scenario has -- in contrast to the primary models -- difficulties to explain a strong spectral steepening, as it seems to be apparent in the Coma cluster \\citep{2002BrunettiTaiwan}. However, measurements of the spectral indices of faint and very extended sources, in the presence of strong point sources, are an observational challenge, so that the possibility of larger uncertainties in the determined radio halo spectra can not be fully excluded yet.  The hadronic scenario will soon become further testable since the gamma radiation from the unavoidable neutral pion decay should be detectable by future gamma ray telescopes like GLAST \\citep{1982AJ.....87.1266V, 1997ApJ...477..560E, 1998APh.....9..227C, 2000A&A...362..151D, 2001ApJ...559...59M}.   There are also other suggested radio halo formation scenarios: radio halos were proposed to be superpositions of large numbers of relic radio galaxies \\citep[][ and others]{1978A&AS...34..117H}, they were proposed to be due to rapidly diffusing electrons escaping from radio galaxies \\citep[][ and others]{1979ApJ...228..576H}, and their relativistic electrons were proposed to result from annihilation of neutralinos, if neutralinos are the dominant dark matter component \\citep{2001ApJ...562...24C}. Although these are interesting possibilities, they are disfavoured by the apparent association of radio halos with merger shock waves as discussed above.  \\subsection{Scientific potential\\label{sec:scipot}}  In any scenario, cluster radio halos give us deep insight into the physics and properties of galaxy clusters. Very likely radio halos give a unique probe of non-thermal processes accompanying energetic cluster merger events.  Large numbers of galaxy clusters are expected to be found also at high redshifts by future surveys: e.g. the XMM Large Scale Structure Survey is expected to find $\\sim 10^3$ galaxy clusters up to redshift one \\citep{2002A&A...390....1R}, Sunyaev-Zeldovich effect cluster detections with the Planck satellite should find $\\sim 10^4$ galaxy clusters and the Sloan Digital Sky Survey is expected to identify $\\sim 5\\cdot 10^{5}$ clusters \\citep{2002A&A...388..732B}.  Using radio halos as tracers of cluster mergers should therefore allow detailed studies of the higher redshift cluster formation processes and properties of the accompanying cluster merger shock waves \\citep{1998ApJ...502..518Q, 2000ApJ...542..608M}. This will be possible due to the strongly increased sensitivity and resolution of next generation radio telescopes (e.g. ATA, EVLA, GMRT, LOFAR, SKA). In order to guide the design and observing strategies of these upcoming radio telescopes predictions for the number of observable radio halos are needed. It is the aim of this paper to provide such predictions, to show their dependence on parameters not yet well constrained, and to indicate their scientific potential.  \\subsection{Structure of the paper\\label{sec:struct}}  Our predictions are based on (i) estimates of the fraction of clusters containing halos, (ii) the local XCLF and various forms of evolution towards higher redshift, and (iii) the local relation between X-ray and radio halo luminosity of clusters (RXLC).  Having the halo fraction $f_\\ha$ (Sect.~\\ref{sec:frac}) and the RXLC (Sect.~\\ref{sec:corr}) the observed present XCLF (Sect.~\\ref{sec:xclf}) can be translated into the local RHLF (Sect.~\\ref{sec:RHLF}). In order to have predictions for higher redshifts, where the XCLF is not yet measured, we translate a theoretical cluster mass function into an XCLF via a mass-X-ray luminosity correlation (MXLC) of clusters of galaxies (Sect.~\\ref{sec:xclf}). This also allows predictions of the number counts of cluster radio halos as a function of apparent flux density (Sect.~\\ref{sec:RHLF}). We do this for a constant halo fraction irrespective of cluster mass and redshift, and for one which evolves as the fraction of clusters with recent mergers (Sect.~\\ref{sec:frac}). In the latter more realistic calculations we also include a possible dimming effect of halos due to higher radiative losses at higher redshifts. The cluster radio halo detection strategies and expectations are briefly discussed in Sect. \\ref{sec:diss}   Our calculations are done for a $\\Lambda$CDM-Universe with $\\Omega_0 = 0.3$, $\\Omega_\\Lambda = 0.7$, $H_0 = 50\\,h_{50}\\,{\\rm km/s}$, $\\sigma_8 = 0.9$, and $\\Gamma =0.21$. ",
        "conclusions": "}  We estimated the cluster radio halo luminosity function and the expected flux density distribution by translating an observed and a theoretical X-ray cluster luminosity function with the help of the observed cluster radio halo--X-ray luminosity correlation. A power-law form of this correlation was used to extrapolate into the observationally poorly constraint regime of (weak) radio halos of low X-ray luminosity clusters. For a simple model calculation we assumed that a fraction $f_\\ha = \\frac{1}{3}$ of all clusters contain radio halos, irrespective of redshift and cluster size. We note, that if the halo fraction for low X-ray luminosity clusters would be much lower, which cannot be excluded with the present day data, our predictions based on the above halo fraction would be overestimated. In the case that the halo fraction is the same for all cluster, but lower than assumed here, our results can simply be scaled.  \\begin{figure}[t] \\begin{center} \\psfig{figure=dNdz.ps,width=\\figsize\\textwidth,angle=0} \\end{center} \\vspace{-0.7cm}\\caption[]{\\label{fig:dNdz} Expected redshift distribution of radio halos with fluxes above flux limits as indicated in the figure. The solid lines give the most realistic model, whereas the dashed lines do not include any radio halo dimming with redshift. The histogram shows the differential  redshift distribution of the radio halo cluster sample compiled by \\cite{Feretti.Pune99} and \\cite{2001A&A...376..803G} (binned into bins of width $\\Delta z = 0.1$).} \\end{figure}  \\begin{table*}[t] \\begin{center} \\begin{tabular}{|rr|rrrrrrrl|} \\hline \\multicolumn{2}{|c|}{$S_{\\rm 1.4 GHz,min}$} & $N^{\\rm flat}_{\\rm evol.}$ & $N^{\\rm int.}_{\\rm evol.}$ & $N^{\\rm steep}_{\\rm evol.}$ & $N^{\\rm int.}_{\\rm local}$ & $N^{\\rm int.,*}_{\\rm local} $ & $N^{\\rm int.,*}_{\\rm evol.}$ & $N^{\\rm int.,*}_{\\rm evol.}$ & $\\!\\!\\!\\!\\!\\!\\!\\!(z\\!>\\!0.3)$\\\\ \\hline $1$ & $\\mu$Jy \t& 74857.9 & 36646.9 & 15388.6 & 118854.0 & 70579.6 & 23758.5 & 10784.9 & \\\\ $10$ & $\\mu$Jy \t& 19784.7 & 10269.5 &  4821.5 &  36733.0 & 19686.2 &  6812.2 &  2123.7 & \\\\ $100$ & $\\mu$Jy &  4308.1 &  2403.8 &  1298.1 &   8076.4 &  4247.7 &  1653.5 &   280.9 & \\\\ $1$ & $$mJy \t&   735.7 &   450.2 &   290.6 &   1143.8 &   664.7 &   326.4 &    20.5 & \\\\ $10$ & $$mJy \t&    93.3 &    64.2 &    52.0 &    100.0 &    71.0 &    50.1 &     0.6 & \\\\ $100$ & $$mJy \t&     8.4 &     6.7 &     7.1 &      5.6 &     5.0 &     5.7 &     0.0 & \\\\ $1$ & $$Jy \t&     0.5 &     0.5 &     0.7 &      0.2 &     0.2 &     0.5 &     0.0 & \\\\ \\hline \\end{tabular} \\end{center} \\caption[]{\\label{tab:NS} The number $N$ of expected radio halos on the full sky, which are above a given flux density $S_{\\rm 1.4 GHz,min}$ for the flat ($N^{\\rm flat}$), the intermediate ($N^{\\rm int.}$), and the steep ($N^{\\rm steep}$) radio halo--X-ray luminosity correlations displayed in Fig.~\\ref{fig:lnulX}. In addition to the model with an evolving X-ray luminosity function ($N_{\\rm evol.}$, see Fig.~\\ref{fig:dNdS}) also the radio halo number counts for a redshift independent ($=$ local) cluster distribution are given ($N_{\\rm local}$, see Fig.~\\ref{fig:dNdS}) for the intermediate RXLC. Further, the models marked by $*$ give the expected number counts assuming that the fraction of clusters with radio halos is not $f_\\ha = \\frac{1}{3}$ as assumed in the other calculations, but is given by the fraction of clusters which had a recent strong mass increase, as displayed in Fig.~\\ref{fig:frh}. In addition to this, it is assumed that the radio halo luminosity of a cluster with the same mass is lower by a factor $(1+z)^{-4}$ due to the increasing inverse Compton energy losses on the CMB. Thus, the first three columns indicate the level of uncertainty in these calculations due to the uncertainty in the RXLC, column 4 \\& 5 give an optimistic model, and the last two columns give the most likely estimate.} \\end{table*}  The above assumptions may be questioned, since both the higher merging rate of clusters of galaxies and also the increased electron inverse Compton losses at higher redshifts can modify the fraction of clusters having radio halos. For that reasons also calculations were presented in which we tried to take both effects into account.  If our assumptions hold, we are able to predict the number of detectable radio halos with upcoming sensible radio telescopes like LOFAR, ATA, EVLA, SKA, and also the existing GMRT. Detailed numbers for the different models can be found in Tab. \\ref{tab:NS}.  The LOFAR array as an example: the point source sensitivity at 120 MHz is expected to be 0.13 mJy within 1 hour integration time and a 4 MHz bandwidth. A survey covering half of the sky can be accomplished in a years timescale at this frequency and with this depth. It would find $800-1200_{-40\\%}^{+80\\%}$ radio halos\\footnote{The first (lower) number result from our most realistic model, the second (higher) from the model with constant $f_\\ha$ and constant RXLC; the error range indicates the uncertainties resulting from the possible slopes of the RXLC; a radio halo spectral index of $\\alpha_\\nu = 1$, which is a conservative assumption for this purpose, was used in the frequency interpolation.} with a significance of 10 sigma, sufficient for further follow up observations. Within this sample $140-300_{-40\\%}^{+80\\%}$ of the radio halos are expected to have redshifts larger than 0.3.  A more efficient strategy to find cluster radio halos would be to use the large future cluster catalogues from SDSS, PLANCK, XMM-Newton as a target list for deep integrations with the upcoming sensible radio telescopes. This should allow tests of many of the hypotheses (partly used in this work) on redshift and cluster size dependencies of the radio halo population, helping to establish cluster radio halos as a tool to investigate galaxy cluster formation and the non-thermal processes accompanying it."
    },
    "0209/astro-ph0209504_arXiv.txt": {
        "abstract": "In this paper we give a pedagogical review of the recent observational results in cosmology from the study of type Ia supernovae and anisotropies in the cosmic microwave background. By providing consistent constraints on the cosmological parameters, these results paint a concrete picture of our present-day universe. We present this new picture and show how it can be used to answer some of the basic questions that cosmologists have been asking for several decades. This paper is most appropriate for students of general relativity and/or relativistic cosmology. ",
        "introduction": "Since the time that Einstein pioneered relativistic cosmology, the field of cosmology has been dominated by theoretical considerations that have ranged from straightforward applications of well-understood physics to some of the most fanciful ideas in all of science. However, in the last several years observational cosmology has taken the forefront. In particular, the results of recent observations on high-redshift supernovae and anisotropies in the radiation from the cosmic microwave background (CMB) have pinned down the major cosmological parameters to sufficient accuracy that a precise picture of our universe has now emerged. In this paper, we present this picture as currently suggested by the beautiful marriage of theory and experiment that now lies at the heart of modern cosmology.  We begin our discussion, in sections II and III, with a review of the standard theory of the present-day universe that persisted, virtually unaltered, from the time of Einstein until the mid 1990s. This review will lay most of the theoretical groundwork needed for sections IV and V on the two experimental efforts that has had such a major impact over the last few years. Once the new results have been explained, we present, in section VI, the picture of our universe that has emerged from the recent results. We then conclude this paper with some brief comments on the implications of these results for our understanding of not just the present-day universe, but of its past and future. To help make this discussion more accessible, we use SI units with time measured in seconds instead of meters and with all factors of $G$ and $c$ explicitly shown unless otherwise noted. ",
        "conclusions": "In summary, the resent observational results in cosmology strongly suggest that we live in a universe that is spatially flat, expanding at an accelerated rate, homogeneous and isotropic on large scales, and is approximately 13 billion years old. The expansion of the universe is described by Eq. (63), and its metric by Eq. (64). We have seen that roughly 96\\% of the matter and energy in the universe consists of cold dark matter and the cosmological constant. We now know basic facts about the universe much more precisely than we ever have. However, since we cannot speak with confidence about the nature of dark matter or the cosmological constant, perhaps the most interesting thing about all of this is that knowing more about the universe has only shown us just how little we really understand.  As mentioned previously, the most common view of the cosmological constant is that it is a form of vacuum energy due, perhaps, to quantum fluctuations in spacetime [5]. However, within the context of general relativity alone there is no need for such an interpretation; $\\Lambda $ is just a natural part of the geometric theory [40]. If, however, we adopt the view that the cosmological constant belongs more with the energy-momentum tensor than with the curvature tensor, this opens up a host of possibilities including the possibility that $\\Lambda $ is a function of time [41].  In conclusion, it is also important to state that although this paper emphasizes what the recent results say about our present universe, these results also have strong implications for our understanding of the distant past and future of the universe. For an entertaining discussion of the future of the universe see Ref. 42. Concerning the past, the results on anisotropies in the CMB have provided strong evidence in favor of the inflationary scenario, which requires a $\\Lambda $-like field in the early universe to drive the inflationary dynamics. To quote White and Cohn, ``Of dozens of theories proposed before 1990, only inflation and cosmological defects survived after the COBE announcement, and only inflation is currently regarded as viable by the majority of cosmologists'' [17]."
    },
    "0209/astro-ph0209397_arXiv.txt": {
        "abstract": "We present the integrated properties of the stellar populations in the {\\it Universidad Complutense de Madrid} (UCM) Survey galaxies. Applying the techniques described in the first paper of this series, we derive ages, burst masses and metallicities of the newly-formed stars in our sample galaxies.  The population of young stars is responsible for the $\\rm H\\alpha$ emission used to detect the objects in the UCM Survey. We also infer total stellar masses and star formation rates in a consistent way taking into account the evolutionary history of each galaxy. We find that an average UCM galaxy has a total stellar mass of $\\sim10^{10}\\mathcal{M}_\\odot$, of which about 5\\% has been formed in an instantaneous burst occurred about $5\\,$Myr ago, and sub-solar metallicity. Less than 10\\% of the sample shows massive starbursts involving more than half of the total mass of the galaxy. Several correlations are found among the derived properties. The burst strength is correlated with the extinction and with the integrated optical colours for galaxies with low obscuration. The current star formation rate is correlated with the gas content. A stellar mass--metallicity relation is also found. Our analysis indicates that the UCM Survey galaxies span a broad range in properties between those of galaxies completely dominated by current/recent star formation and those of normal quiescent spirals. We also find evidence indicating that star-formation in the local universe is dominated by galaxies considerably less massive than $L^*$. ",
        "introduction": "The present paper is the second of a series which deals with the determination of the main properties of the stellar populations in the {\\it Universidad Complutense de Madrid} (UCM) Survey galaxies \\citep{1994ApJS...95..387Z,1996ApJS..105..343Z,1999ApJS..122..415A}.  We deal here with the integrated properties of the galaxies as a first step towards understanding their evolution.  Future developments will address the properties of the spatially-resolved stellar populations and the improvement of the modelling procedures. This will be necessary to understand the details of the star formation history of each galaxy as well as their dust extinction properties, which turn out to be one of the key points (and probably the most important one) in this field.  One of the goals of this study is to determine the nature of the galaxies which were detected by the UCM Survey. There is an extensive dataset available for the sample, including spectroscopic and photometric information covering a broad wavelength range from the optical to the near infrared (nIR), together with some radio data. The analysis of the spectroscopic observations allow us to study the emission lines formed in the ionized gas clouds surrounding young hot stars. Among these lines, the Balmer $\\rm H\\alpha$ line is one of the best tracers of the most recent star formation (\\citealt{1992ApJ...388..310K,1998ARA&A..36..189K}). It is easily observable in nearby galaxies and is less extinguished by dust than other optical emission lines ($\\rm H\\beta$, $\\rm{[OII]}\\lambda3727$\\,\\AA) and the ultraviolet continuum. The $\\rm H\\alpha$ luminosity and equivalent width are directly linked to the youngest population of stars responsible for the heating and ionisation of the gas, and thus can be used in the determination of the mass in newly-formed stars, their age, etc. Spectroscopic data can also be used to evaluate the extinction (via the Balmer decrement), the metallicity, and the excitation.  Photometric data covering a wide wavelength range can be used to carry out a population synthesis analysis of composite stellar populations. Many examples of such studies, for low and high redshift galaxies, are found in the literature.  See, e.g., \\citet{1995A&A...303...41K,1996A&A...313..377D,1999MNRAS.303..641A, 2000ApJ...536L..77B,2000MNRAS.316..357G,2000MNRAS.312..497B,2001ApJ...559..620P}. Some other authors have focused on the quantitative analysis of the optimal sets of observables and signal-to-noise ratios required to obtain robust results (see \\citealt[and references therein]{2000A&A...363..476B,2002AJ....123.1864G}).  In this respect, the combination of high-quality optical, ultraviolet and nIR data has been found provides some of the fundamental information needed to study local galaxies. To complement the broad-band photometry, emission-line fluxes can also be used in galaxies presenting star-formation activity.  The ultraviolet part of the spectrum and the emission lines are dominated by young hot stars formed recently. The nIR is essential to characterise the more evolved population, since it is less sensitive to recent bursts and dust extinction.  One principal application of this line of research is the determination of the stellar masses of galaxies, another major goal of our project.  It has been argued that nIR data, and more precisely, the K-band luminosity, can be used as a good tracer of the stellar mass \\citep{1993ApJ...418..123R,2000ApJ...536L..77B}. Based on this assumption, several nIR-based surveys have been carried out in order to use the K-band luminosity function at several redshifts to directly obtain the distribution galaxy masses (e.g., \\citealt{1996AJ....112..839C,1999ApJ...512...30C,2001ApJ...560..566K, 2001ApJ...562L.111D}). However, it is very important to test  the reliability of  the stellar masses determined using $K$-band luminosities alone. Age differences from galaxy to galaxy, or the presence of massive recent star-formation (with a mass comparable to that of the evolved population) may have an effect on the   mass-to-light ratio even in the nIR.  Indeed, some authors have recently claimed that the K-band mass-to-light ratio depends on parameters such as the galaxy colours,  clearly affecting the determination of total stellar masses \\citep{1998A&A...339..409M,2000ApJ...536L..77B,2001ApJ...550..212B,2002MNRAS.334..721G}.    \\citet[\\pone\\, hereafter]{2002MNRASnotyetI} presented the dataset and  the modelling and statistical techniques used in the current analysis. Paper~I also discusses how well our techniques are able to reproduce the observations. Using a stellar population synthesis library, and taking into account the gas emission and dust attenuation, our method is able to model successfully the observational properties of star-forming galaxies.  Several {\\it a priori} parameters of the models were tested. These include (1) the evolutionary spectral synthesis library (we used Bruzual \\& Charlot --private communication-- and \\citealt{1999ApJS..123....3L}); (2) the recent star formation scenario (instantaneous and constant star formation rates -SFR- were tested); (3) the initial mass function (\\citealt{1955ApJ...121..161S}, \\citealt{1986FCPh...11....1S} and \\citealt{1979ApJS...41..513M}); and (4) the extinction-correction recipe (\\citealt{2000ApJ...533..682C} and \\citealt{2000ApJ...539..718C}).   Among these, we found that the extinction plays a fundamental role.  We present now the results obtained from the application of our modelling procedure and statistical analysis to the UCM Survey data. Briefly, the global properties of the newly-formed stars and those of the underlying evolved population will be quantified.  These properties are derived for each individual galaxy, ensuring that the stellar content and star formation history of each object are properly taken into account.  The determination of these properties will lead to a better understanding of the observational biases of this kind of surveys.   A plan of the paper follows. First, the main properties of the UCM Survey sample will be briefly described in Section~\\ref{sample}. The population synthesis method used in this will be reminded in Section~\\ref{method} (see \\pone\\, for further details). Next, the results concerning the youngest population will be presented and discussed in Section~\\ref{results}. Following this, in Section~\\ref{masses} we will focus on the integrated stellar masses of the UCM galaxies. Finally, the conclusions will be presented. Throughout this paper we use a cosmology with $\\mathrm H_{0}=70$\\,km\\,s$^{-1}$\\,Mpc$^{-1}$, $\\Omega_{\\mathrm M}$=0.3 and $\\Lambda$=0.7. ",
        "conclusions": ""
    },
    "0209/astro-ph0209442_arXiv.txt": {
        "abstract": "We present new optical images ($B$, $V$, and H$\\alpha$) of the archetypical starburst/superwind galaxy M82 obtained with the 8.2 m Subaru Telescope to reveal new detailed structures of the superwind-driven nebula and the high-latitude dark lanes. The emission-line nebula is decomposed into (1) a ridge-dominated component comprising numerous filament/loop sub-structures whose overall morphology appears as a pair of narrow cylinders, and (2) a diffuse component extended over much wider opening angle from the nucleus. We suggest that these two components have different origins. The ridge-dominated component appears as a pair of cylinders rather than a pair of cones. Since this morphological property is similar to that of hot plasma probed by soft X-ray, this component seems to surround the hot plasma. On the other hand, the diffuse component may arise from dust grains which scatter stellar light from the galaxy. Since inner region of this component is seen over the prominent ^^ ^^ X\"-shaped dark lanes streaming out from the nuclear region and they can be reproduced as a conical distribution of dust grains, there seems to be a dusty cold outflow as well as the hot one probed by soft X-ray and shock-excited optical emission lines. If this is the case, the presence of such high-latitude dust grains implies that neutral gaseous matter is also blown out during the course of the superwind activity. ",
        "introduction": "Bursts of massive star formation (starbursts) are considered to play important roles in the evolution of galaxies from high redshift to the present day. Approximately 10\\% of nearby galaxies show evidence for such starburst activity (Weedman et al. 1981; Balzano 1983; Ho et al. 1997). In particular, some of them also show evidence for galactic-scale bipolar outflows (superwinds) driven by collective effects of a large number of supernova explosions occurred in the central region of galaxies (e.g., Chevalier \\& Clegg 1985; Tomisaka \\& Ikeuchi 1988; Heckman et al. 1990; Suchkov et al. 1994; Tenorio-Tagle \\& Munoz-Tunon 1998; Strickland \\& Stevens 2000). The superwind phenomenon is also very important as well as the starburst activity itself because it provides metal-enriched gas into the intergalactic medium.  Since the identification of optical emission-line filaments in M82 (Lynds \\& Sandage 1963; Burbidge et al. 1964), much attention has been paid to this galaxy and it is now widely accepted that M82 is the archetypal starburst galaxy with evidence for the bipolar superwinds emanating from the central region to the outer halo area (e.g., Heckman et al. 1987; Nakai et al. 1987; Bland \\& Tully 1988; Shopbell \\& Bland-Hawthorn 1998; Lehnert et al. 1999). Because of its proximity (a distance to M82, $D=$ 3.63 Mpc; Freedman et al. 1994), M82 allows us to investigate detailed characteristics of the superwind activity. Indeed, previous studies of the M82 superwind have shown that the gas in this superwind takes various phases: I) hot ($T \\sim 10^7$ K) soft X-ray emitting gas (e.g., Bregman, Schulman, \\& Tomisaka 1995; Strickland, Ponman, \\& Stevens 1997; Lehnert et al. 1999), II) warm ($T \\sim 10^4$ K) ionized gas probed by optical emission lines (e.g., Heckman et al. 1987; Bland \\&Tully 1988; McKeith et al. 1995; Shopbell \\& Bland-Hawthorn 1998), III) cool ($T \\sim 100$ K) neutral gas probed by dust grains, including i) dusty molecular-gas halo (e.g., Scarrott, Eaton, \\& Axon 1991; Sofue et al. 1992) and ii) complex dark lanes seen in the upper disk regions (Ichikawa et al. 1994; 1995), and IV) cold ($T \\sim 10 - 100$ K) molecular and atomic gas (e.g., Nakai et al. 1987; Yun, Ho, \\& Lo 1993). However, it is not yet clear how such various gaseous components are related to each other although the observational properties of each component have been discussed extensively.  In order to improve our understanding of this superwind activity, we present new sensitive high-resolution images of M82 taken with the 8.2 m Subaru telescope (Kaifu et al. 1998). These images are analyzed in various ways to reveal new detailed structures of the galactic disk (stars), dark lanes (dust grains), ridge-dominated and diffuse components (shock-heated warm gas and dust grains scattering the nuclear light, respectively) of the nebula. Combining with the previous information of X-ray image for hot ionized gas, we investigate the multi-structured superwind phenomenon in M82. ",
        "conclusions": "\\subsection{The Ridge Component}  The ridge component comprises many filaments along the superwind. Each filament shows either shell- or loop-like structures directing radially from the nuclear region. Similar filament-dominated emission-line nebulae are also found in some well-known Galactic objects with an outflow [such as M1-67: a Wolf-Rayet star with stellar wind (Grosdidier et al. 2001) and M57: a planetary nebula with a ring (Komiyama et al. 2000)]. It is widely accepted that such a filament can be a local shock front between the inner expanding matter and the outer interstellar medium. The observed filaments in M82 are also considered to be driven by shocks. This interpretation is supported by the optical emission-line diagnostics (Heckman et al. 1987) although their data probe both ridge and diffuse components.  As mentioned before, the overall structure of the ridge component looks like a pair of elongated cylinders rather than a pair of cones, and hence hereafter we simply denote it as the cylinder. The diameter of the cylinder at the galaxy disk plane (35\\arcsec~ - 40\\arcsec) is similar to, or slightly larger than, that of the nuclear starburst region traced both by near-infrared emission (30\\arcsec~ - 35\\arcsec; e.g., Lester et al. 1990) and by an ensemble of non-thermal point-like sources (e.g., Kronberg, Biermann, \\& Schwab 1985). It seems worthwhile noting that the cylinder shows the sharp boundary, in particular at NE, SE, and SSE of the nebula, causing the limb-enhanced morphology. All these features can be well reproduced by numerical simulations of the superwind in which radiative shock occurs around the expanding hot gas (e.g., Suchkov et al 1994; Strickland \\& Stevens 2000) (see section 3.3).  The hot gaseous component probed by soft X-ray emission is known to be collimated rather tightly along the minor axis of the galaxy than the optical emission-line nebula (Bregman et al. 1995; Strickland et al. 1997; Lehnert et al. 1996). We found that the ridge component surrounds the hot gaseous nebula more closely than the diffuse component. However, the tangential profiles along the galaxy disk at high latitude look different between the ridge component (showing limb-enhanced profile) and the soft X-ray nebula (showing the ridge along the centerline). Note, however, that such difference can be explained if we assume that the optical filaments are distributed on the thin surface around the inner hot gas component. All these lines of evidence suggest that the radiative shock at the thin interface between the expanding hot X-ray emitting gas and the ambient cold gaseous matter would be responsible for the ridge component.  \\subsection{The Diffuse Component}  The diffuse component was first found by Bland \\& Tully (1988) and was confirmed by Shopbell \\& Bland-Hawthorn (1998). It shows rather oval shape elongated along the minor axis of the galaxy as a whole, but is divided into NW and SE parts around the disk plane. It is extended more widely in the tangential direction (parallel to the galaxy disk) and does not show a limb-enhanced morphology, being much different from the ridge component \\footnote{ As noted in the earlier section, separating the ridge and diffuse components by the ridge-detection program could not be perfectly done because some part of the flux separated as ``diffuse'' may actually be structures which are resolved slightly around the peak of the ridges. However, since the ridge component sppares to be a collimated cylindrical structure outer parts of the nebula around the ridge component are likely to be truly ``diffuse''. Therefore the comparison of the overall shapes between the ridge and diffuse components can be made with less ambiguity. } . These observational properties suggest that this component does not have a form of thin surface but occupies a large part of the volume probed by optical emission lines.  Here a question arises as what the origin of the diffuse component is. Previous imaging polarimetries of the M82 nebula have revealed strongly-polarized emission whose $E$ vectors show circular symmetric pattern around the nucleus. Since this property is a typical signature of the reflection nebulae, the dust scattering of the galaxy light would contribute to the nebula (e.g., Bingham et al. 1976; Schmidt, Angel, \\& Cromwell 1976). It is thus suggested that the diffuse component arises from the scattering of the galaxy light at dusty halo.   In addition to the dusty halo, M82 is well known to have rich dark lanes across the galaxy disk. They are peculiar and complex in shape, and are remarkably different from those found in normal spiral galaxies. The most prominent and peculiar dark lane is seen on SE side of the disk (the front side of the disk: e.g., Shopbell \\& Bland-Hawthron 1998; McKeith et al. 1995). It runs through the nuclear region out to 2.5 arcmin (2.6 kpc) from the nucleus (Figure 5). Interestingly it curves toward both SE and SW directions at both ends of the dark lane, and can be traced up to $\\sim 1$\\arcmin~ ($\\sim 1$ kpc in projection) above/below the galaxy disk at their tips. Similar but less-prominent dark lanes are also found at another side of the disk emanating to NE and NW from the nuclear region. These two sets of the dark lanes make an ``X''-shaped morphology around the nucleus. They are considered to be distributed not within the galaxy disk but above/below the disk (e.g., Ichikawa et al. 1994; 1995). We find that the diffuse component can be seen even on the dark lanes, suggesting that dust grains responsible for the dark lanes are also responsible for the scattering of the diffuse component.   The next question is addressed to the origin of the dark lanes. The dark lanes have the following characteristics. First, the X-shaped dark lanes seem to be streaming out from the nucleus. Second, the NW boundary (and also probably the NE boundary) of the SE dark lane is smoothly curved comparing with other complicated dark lanes. Finally one-dimensional profile of the SW dark lane, cut perpendicular to the disk across the dark lane (Figure 5), shows highly asymmetric structure of the extinction, i.e., the dust extinction increases monotonically toward the NW side, and then decreases suddenly at the NW edge the SE dark lane. Such characteristics can be understood if we assume that the dust is distributed in a wide cone-like geometry and we see a higher extinction near the cone surface due to projection effect. This supports a scenario of the dusty superwind in which dust grains are provided from the nuclear region with the outflow associated with the superwind probed in optical emission-lines and soft X-ray emission. If this is the case, high-latitude dusty halo may be created in a similar way as for the dark lanes (i.e., the dusty superwind) since there are no distinct boundaries in the diffuse component dividing the dust scattering on the dark lane near galaxy plane and high-latitude halo area up to $\\gtrsim 2$ kpc from the disk. Yun et al. (1993) suggested that a part of the high-latitude HI gas of M82 may be affected dynamically by the superwind, supporting for the presence of a high-latitude cold gas associated with the superwind outflow. Although we do not rule out an alternative possibility that the past tidal interaction with the neighbor galaxy M81 has brought some dusty matter from the galaxy disk up to the halo (Yun, Ho, \\& Lo 1993; 1994), the observational properties discussed in this paper suggest that the dusty superwind scenario is more reasonable.   Such a dusty cold-gas outflow can be made as a result of the pushing out and/or dragging out process of the dusty gas around the nucleus by the expanding hot high-pressure gas within the superwind (e.g., Suchkov et al. 1994; Strickland \\& Stevens 2000). Indeed dusty outflows associated with superwinds have been suggested by optical observations (Nakai 1989; Heckman et al. 2000; Veilleux \\& Rupke 2002; Rupke, Veilleux, \\& Sanders 2002) and submillimeter observations (Alton, Davies, \\& Bianchi 1999).  Finally we give a comment on the previous claim that even the inner filamentary nebula shows strong polarization whose direction is nearly aligned to that of the outer diffuse nebula (Scarrott et al. 1991), which could be inconsistent with our conclusion since no polarized flux is expected from the shock-excited emission. We suspect that the imaging polarimetry with insufficient spatial resolution could be the cause of this apparent contradiction. Separate polarimetric measurement of the filaments in the ridge and diffuse components are required to disentangle the polarimetric properties of the narrow filamentary structures (typically $\\lesssim 2$\\arcsec~ in width) and the overlapping diffuse component.   \\subsection{Implications for Superwind Model}  Important characteristics of the M82 superwind nebula are that the shock-driven ridge component contains numerous shells and filaments and its overall shape is rather collimated out to outer region of the nebula. In recent hydrodynamical simulations of the superwind conducted by Strickland \\& Stevens (2000), two types of the interstellar medium (ISM) density distribution, thin or thick disk, are adopted to see its effect on the evolution of the wind. It turns out that the model with the thick ISM can reproduce the characteristics of M82 nebula: The thick and dense wall of ISM created around the starbursting region works to collimate the wind, and Rayleigh-Taylor instability at the wall creates small-scale structures which are eventually dragged out with the wind to form shells/filaments (Strickland \\& Stevens 2000). Although these hydrodynamical simulations with thick ISM look successful, we point out the following points which require some cautions. As Strickland \\& Stevens (2000) noted, the thick ISM cannot reproduce the observed rotation curve of M82. Therefore one may need more realistic ISM distribution models or improved treatment of gas dynamics at the circumnuclear region to reproduce both a collimated wind and the rotation curve consistently. It seems also noteworthy that the morphology of the small-scale structures of the nebula depend rather strongly on the numerical resolution in the simulation, and higher resolution tends to result in finer shell/filament structures (Strickland \\& Stevens 2000). Therefore one may need even finer-scale numerical simulation to enable us to compare sizes and spatial distributions of shells/filaments between observations and the models. Note that, although observed small-scale structures with a typical size of $90 - 350$ pc (5\\arcsec~ - 20\\arcsec) are already resolved by their simulations with a cell size of $4.9 - 14.6$ pc, the simulation resolution would not be fine enough to resolve the circumnuclear region where smaller-scale instability begins to work. Finally, we show a cartoon in which all the superwind components discussed here are summarized (Figure 6)."
    },
    "0209/astro-ph0209168_arXiv.txt": {
        "abstract": "We extract {\\it J} and {\\it K$_s$} magnitudes from the 2MASS Point Source Catalog for approximately 6$\\times 10^6$ stars with $8 \\le {\\it K_s} \\le 13$ in order to build an {\\it$A_K$} extinction map within 10$^{\\circ}$ of the Galactic centre. The extinction was determined by fitting the upper giant branch of ({\\it K$_s$, J-K$_s$}) colour-magnitude diagrams to a dereddened upper giant branch mean locus built from previously studied Bulge fields. The extinction values vary from {\\it$A_K$}=0.05 in the edges of the map up to {\\it$A_K$}=3.2 close to the Galactic centre. The 2MASS extinction map was compared to that recently derived from DENIS data. Both maps agree very well up to {\\it$A_K$}=1.0. Above this limit, the comparison is affected by increased internal errors in both extinction determination methods. The 2MASS extinction values were also compared to those obtained from dust emission in the far infrared using DIRBE/IRAS. Several systematic effects likely to bias this comparison were addressed, including the presence of dust on the background of the bulk of 2MASS stars used in the extinction determination. For the region with $3^{\\circ}<|{\\it b}|<5^{\\circ}$, where the dust contribution on the far side of the Galaxy is $\\approx$ 5 \\%, the two extinction determinations correlate well, but the dust emission {\\it$A_K$} values are systematically higher than those from 2MASS. A calibration correction factor of 76\\% for the DIRBE/IRAS dust emission extinction is needed to eliminate this systematic effect. Similar comparisons were also carried out for the $1^{\\circ}<|{\\it b}|<3^{\\circ}$ and $|{\\it b}| < 0.5^{\\circ}$ strips, revealing an increasing complexity in the relation between the two extinction values. Discrepancies are explained in terms of the calibration factor, increasing background dust contribution, temperature effects influencing the dust emission extinction and limitations in the 2MASS extinction determination in very high extinction regions ($|{\\it b}| < 0.5^{\\circ}$). An asymmetry relative to the Galactic plane is observed in the dust maps, roughly in the sense that {\\it$A_K$} values are 60\\% smaller in the south than in the north for $1^{\\circ}<|{\\it b}|<5^{\\circ}$. This asymmetry is due to the presence of foreground dust clouds mostly in the northern region of the Bulge. ",
        "introduction": "The high extinction and its patchy distribution in the Galactic centre region have been a constant problem to the study of the properties of the Bulge stellar population. Several efforts were carried out to investigate the extinction distribution close to the Galactic centre. For example, Catchpole et al. (1990) studied the distribution of stars in the central $1^{\\circ} \\times  2^{\\circ}$ of the Galaxy by means of {\\it J}, {\\it H} and {\\it K} colour-magnitude diagrams (CMDs). They derived visual absorptions in the range  $7 < {\\it A_V} < 30$. Frogel et al. (1999, hereafter FTK99) obtained extinction values varying from {\\it$A_V$}=2.41 up to {\\it$A_V$}=19.20 for 11 Bulge fields with $|{\\it b}|<4^{\\circ}$, close to the Galactic centre. Stanek (1996) derived a mean extinction of ${\\it<A_V>}=1.54$ for Baade's Window (Baade 1963). As a consequence, the investigations about the Bulge stellar content based on optical data were restricted for a long time to lower-extinction regions such as Baade's Window.  The advent of near infrared surveys such as the Two Micron All Sky Survey (2MASS, Skrutskie et al. 1997) and the Deep NIR Southern Sky Survey (DENIS, Epchtein et al. 1997) has provided fundamental tools to study the stellar population (Unavane et al. 1998) and extinction (Schultheis et al. 1999) in the inner Bulge. Recently, Dutra et al. (2002, hereafter Paper I) confirmed the existence of two new Bulge windows, W0.2-2.1 and W359.4-3.1, closer to the Galactic centre than Baade's Window; they used the {\\it JK$_s$} photometry from the 2MASS survey archive to map the extinction distribution within 1$^{\\circ}$ towards these windows. The reddening distribution in the Bulge area is also fundamental to understand the spatial distribution of globular clusters and their relation to the Bulge field stellar population itself. Barbuy et al. (1998) discussed the globular clusters projected within 5$^{\\circ}$ of the nucleus and which appear to be related to the Bulge, while Barbuy et al. (1999) discussed the properties of those found in the area covered by the present study.  \\begin{figure} \\begin{center} \\centerline{\\psfig{file=mb948f1.eps,height=6cm,width=6cm,angle=0}} \\end{center} \\caption{Extinction-corrected (filled circles) and observed (open circles) CMDs for a cell at $\\ell=1.93^{\\circ}$, ${\\it b}=0.93^{\\circ}$. The straight line represents the reference upper giant branch (Eq. 1).} \\end{figure}  In this work we use the 2MASS {\\it J} (1.25$\\mu$m) and {\\it$K_s$} (2.17$\\mu$m) photometric data to build ({\\it K$_s$, J-K$_s$}) CMDs of Bulge fields in order to map out the interstellar extinction in the central 10$^{\\circ}$ of the Galaxy. The extinction was derived by means of the upper giant branch fitting method, similar to that described in Paper I. In Sect. 2 we revise this method and describe in detail the way we build the extinction map. In Sect. 3 we compare the present results with those from  DENIS. In Sect. 4 we analyze the 2MASS and DIRBE/IRAS extinction maps, making use of simple models for the optical depth in different directions in the Galaxy, and discuss the asymmetries between observations in the northern and southern Galactic hemispheres. Finally, the concluding remarks are given in Sect. 5.  \\section[]{Building the extinction map with 2MASS data}  In Paper I we have built an extinction map in regions of low-extinction by means of upper giant branch fitting. We adopted an upper giant branch template from a composite CMD using 7 Bulge fields from FTK99. Using this template we managed to reproduce the mean extinction in Baade's and Sgr I windows. Metallicity effects were not considered in Paper I, since Ramirez et al. (2000), from a spectroscopic study of central M giant stars, pointed out that there is no evidence for metallicity gradient within the inner Bulge ({\\it R} $<$ 560 pc). Besides, FTK99 concluded that the amplitude of metallicity variations in the inner Bulge implies very small giant branch slope changes. Therefore, we assume that the metallicity variations in the inner Bulge do not affect significantly the extinction estimates and use as reference the same upper giant branch adopted in  Paper I:  \\begin{equation} (K_s)_0 = -7.81 (J-K_s)_0 + 17.83 \\end{equation}  The equation above describes appropriately the upper giant branch locus for stars with 8 $\\le {\\it K_0} \\le $12.5.  \\noindent We carried out {\\it JK$_s$} photometric extractions of stars in the 2MASS Point Source Catalog available in the Web Interface {\\it http://irsa.ipac.caltech.edu/applications/Gator/} for 61 fields with radius {\\it r}=1$^{\\circ}$ each. For most fields, we extracted stars in the range 8 $\\le {\\it K_s} \\le $ 11.5, there being typically 60,000 such stars in each. For the fields with $|{\\it b}| <$ 1.5$^{\\circ}$, we expected very high extinction (Schultheis et al. 1999); thus we extracted fainter stars, down to {\\it$K_s$}=13. This extra 1.5 mag allowed us to determine the best magnitude range for upper giant branch fitting, taking into account factors such as the increasing photometric errors and contamination by disk stars with fainter magnitude limits, and the decreasing observed extent of the upper giant branch in heavily reddened inner Bulge fields. This issue is discussed in Sect. 2.1. In these low-latitude regions the deeper 2MASS extractions led to a considerably larger number of stars per field, $\\approx$ 200,000. Considering all fields in the area, we extracted {\\it J} and {\\it$K_s$} magnitudes for approximately 6$\\times 10^6$ stars.  \\begin{figure} \\begin{center} \\centerline{\\psfig{file=mb948f2.eps,height=6cm,width=6cm,angle=0}} \\end{center} \\caption{Analysis of limiting magnitude for fields with $|{\\it b}|<1.5^{\\circ}$: CMDs for stars with (a) 8 $\\le {\\it K_s} \\le $ 11.5 and (b) 9 $\\le {\\it K_s} \\le $ 13 within a cell at $\\ell=0.87^{\\circ}$, ${\\it b}=0.27^{\\circ}$; (c) and (d) same {\\it$K_s$} intervals as in (a) and (b), respectively, for stars within a cell at $\\ell=1.93^{\\circ}$, ${\\it b}=0.93^{\\circ}$. The straight line represents the reddened reference upper giant branch. The derived extinction values in panels (a) and (b) are $A_K$= 1.88 and 1.83, respectively. For (c) and (d) $A_K$= 0.74.} \\end{figure}   \\begin{figure*} \\begin{center} \\centerline{\\psfig{file=mb948f3.ps,height=12cm,width=12cm,angle=0}} \\end{center} \\caption{{\\it$A_{K,2MASS}$} extinction map within 10$^{\\circ} \\times 10^{\\circ}$ of the Galactic centre. The contours represent the following extinction values: {\\it$A_K$} = 0.2, 0.5, 1.0, 1.5 and 2.0.} \\end{figure*}  In order to map out the extinction in the 10$^{\\circ}$ Galactic central region, we defined 961 cells with 4$^{\\prime} \\times 4^{\\prime}$ in size in each one of 61 extracted fields. The extinction was determined assuming that the upper giant branch defined by the stars in a cell has the same slope as that adopted as reference (Equation 1). We then calculated from the ({\\it K, J-K$_s$}) values of each star the shift along the reddening vector necessary to make it fall onto the reference upper giant branch. We used the relations ${\\it A_{K_s}} = 0.670 {\\it E(J-K_s)}$ and $\\frac{{\\it A_K}}{{\\it A_{K_s}}}=0.95$ (Paper I) to derive the {\\it$A_K$} values. The {\\it$A_K$} adopted for each cell was taken to be the median of the distribution of {\\it$A_K$} values for the individual stars in it; an iterative 2-$\\sigma$ clipping was applied to the distribution of {\\it$A_K$} values in order to eliminate contamination from foreground stars. Fig. 1 shows the method applied to a cell located at $\\ell=1.93^{\\circ}$, ${\\it b}=0.93^{\\circ}$; the stars in the observed CMD (open circles) have a median extinction of ${\\it A_K}=0.74$ with respect to the reference upper giant branch. The extinction-corrected CMD based on this value is also shown (filled circles) in the figure. Note that the extinction-corrected CMD is well fitted by the reference upper giant branch (straight line), indicating that the determined median extinction is representative of most stars within the cell's area.  \\subsection{The upper giant branch fitting in heavily reddened fields}  We now try to assess the existence of systematic effects on the extinction determination just outlined. Pushing the limits used in our fitting method towards fainter magnitudes is desirable only if it increases the magnitude range where the upper giant branch is clearly defined and can be fitted. Several factors may prevent fainter magnitudes from being useful in this way. The first is obviously the increasingly large photometric errors. The 2MASS {\\it JK$_s$} nominal errors are smaller than 0.04 for $J$ \\ltsima $15$ and $K_s$ \\ltsima $11.5$. Close to the centre, however, the density of stars increases very rapidly, yielding less precise photometric measurements due to source crowding. Another issue is the existence of selection effects: we fit a straight line to the upper giant branch defined in a $(K_s,J-K_s)$ CMD. The $K_s$ limit to the data will stand out clearly as a lower limit in the populated CMD region, but any $J$ band cut-off will be a diagonal line on the CMD. It will bias our best-fit line if the giant branch is crossed by this diagonal line within the magnitude range used in the fit. Finally, at fainter magnitudes the Bulge luminosity function rises rapidly, increasing the contamination by Bulge stars with less extinction than the average, which will hamper our attempt to fit the bulk of the Bulge stellar population.  All the 3 factors just mentioned are in fact present in our data to some extent. Fig. 2 shows the CMDs for two sample cells: one at $\\ell=0.87^{\\circ}$, ${\\it b}=0.27^{\\circ}$ (panels a and b), the other at $\\ell=1.93^{\\circ}$, ${\\it b}=0.93^{\\circ}$ (panels c and d). The panels on the left (right) show the data in the range 8 $\\le {\\it K_s} \\le $ 11.5 (9 $\\le {\\it K_s} \\le $ 13). The reference upper giant branch is shown in all panels, reddened by the best-fit ($A_K$,$E(J-K)$) value found in each case. Clearly, the extinction is underestimated for the lower latitude cell when stars in the 9 $\\le {\\it K_s} \\le $ 13 interval are used. The reason for that is two-fold: there is a large number of blue stars with $K_s$ \\gtsima $11$ and there is a paucity of stars in the lower right corner of the CMD in panel 2b. Both tend to shift the best fit line towards the blue.  The scarcity of faint red stars is due to the 2MASS $J$ band limit: there are essentially no stars beyond the diagonal line $J = 16.5$, which is thus the 2MASS Point Source Catalog limiting $J$ magnitude. This empirical limit is close to that proposed by the 2MASS collaboration (Skrutskie et al. 1997, see also the 2MASS web site). The faint blue stars are probably due to contamination from Bulge stars with less extinction and to the increased photometric errors in these dense fields.  Note that the fit within the range 8 $\\le {\\it K_s} \\le $ 11.5, despite the inclusion of some residual stars bluewards of the giant branch is not biased (panel 2a). For the cell at $\\ell=1.93^{\\circ}$, ${\\it b}=0.93^{\\circ}$ the situation is more reassuring, the derived $A_K$ being insensitive to the magnitude range used. We thus conclude that a more efficient use of the CMDs in the crowded and high extinction areas close to the Galactic plane is made by restricting the fit to {\\it$K_s$}=11.0. This limit is also conveniently close to where completeness effects should start to be significant in the 2MASS. For DENIS, Unavane et al. (1998) estimated the 80\\% completeness level to be at $K \\simeq 10; J \\simeq 13$ and the 2MASS data are about 1 mag fainter.  We conclude that in areas where extinction is around ${\\it A_K} \\le$ 1.5 and photometric errors are not substantially increased by crowding our fitting method does not suffer from any biases due to the $J$ band cut-off. For ${\\it A_K} \\simeq$ 2.5, the J band detection limit effects become dominant and the range available for the fit is substantially shortened, rendering the extinction determination unreliable.  \\begin{figure} \\begin{center} \\centerline{\\psfig{file=mb948f4.eps,height=6cm,width=6cm,angle=0}} \\end{center} \\caption{Histogram of {\\it$A_K$} extinction values of the 10$^{\\circ}$ extinction map.} \\end{figure}  \\begin{figure} \\begin{center} \\centerline{\\psfig{file=mb948f5.eps,height=9cm,width=6cm,angle=0}} \\end{center} \\caption{Upper panel: histogram of internal error $\\sigma_i$ values. Lower panel: Variation of $\\sigma_i$ with {\\it$A_K$} extinction values.} \\end{figure}  \\subsection{The {\\it$A_K$} extinction map}  Fig. 3 shows the {\\it$A_K$} contour map for the central 10$^{\\circ} \\times 10^{\\circ}$ of the Galaxy obtained by applying the method described in the previous section to stars down to {\\it$K_s$}=11.0. The {\\it$A_K$} contours show that the regions with ${\\it A_K} >$ 1.5 are concentrated in the area with $|{\\it b}|< 1.0^{\\circ}$. The isocontours with high values are slightly asymmetric with respect to the mid-plane and more extended in the southern hemisphere. They appear to be caused by the displaced location of the Sun from the Plane (Sect. 4). The structure with extinction values between {\\it$A_K$}=0.5-1.0 and angular dimension of $3^{\\circ} \\times 2^{\\circ}$ around $\\ell$ = 1.5$^{\\circ}$  and {\\it b} = 4.0$^{\\circ}$ is a component of the Pipe Nebula, a dark nebula recently studied in CO by Onishi et al. (1999). This nebula appears to be located at 160 pc from the Sun, as a southern extension of the Ophiuchus dark cloud complex, and is located at the edge of the ScoOB2 association (Onishi et al. 1999). The detection of the Pipe Nebula in Fig. 3 suggests that nearby clouds cannot be neglected in attempts to interpret central extinction maps. From the 32,761 cells that cover a 12$^{\\circ} \\times 12^{\\circ}$ central area of the Galaxy, we have 2MASS data for 80 \\% of them. The quadrant $0^{\\circ}<\\ell<5^{\\circ}$, $-5^{\\circ}<{\\it b}<0^{\\circ}$, which comprises Baade's Window, is so far only partially released by 2MASS.  Fig. 4 shows a histogram of {\\it$A_K$} values for the cells in the 10$^{\\circ} \\times 10^{\\circ}$ extinction map. The mean extinction in the entire map is ${\\it<A_K>} = 0.29$ with a standard deviation $\\sigma = 0.12$ from the mean. 63 \\% of the cells fall within 2-$\\sigma$ of this mean value, and 80\\% of them have ${\\it A_K} < 1.0$. The upper panel of Fig. 5 shows the histogram of internal errors in extinction determination. The mean internal error is $<\\sigma_i> = 0.08$, with a standard deviation of $0.02$ around this mean. 70\\% of the cells have internal errors within 2 standard deviations from the mean value ($0.04 \\leq \\sigma_i \\leq 0.12$). The lower panel shows the dependence of internal errors with {\\it$A_K$}; we note that for ${\\it A_K} > 1.5$ the internal errors increase significantly.  \\begin{figure} \\begin{center} \\centerline{\\psfig{file=mb948f6.eps,height=6cm,width=6cm,angle=0}} \\end{center} \\caption{Comparison between ${\\it A_{K,2MASS}}$ and ${\\it A_{K,DENIS}}$ extinction values for the region $|\\ell|<5^{\\circ}$ and $|{\\it b}|<1.5^{\\circ}$. The straight line represents the identity function.} \\end{figure}  Since the present extinction map can be useful for a wide variety of Galactic and extragalactic studies in such central directions, it will be provided in electronic form in the CDS, by columns: (1) and (2) galactic longitude and latitude of the cell centre, (3) the {\\it K} band extinction {\\it$A_K$} and (4) the uncertainty in the {\\it$A_K$} determination $\\sigma_i$. ",
        "conclusions": "We built the {\\it$A_K$} extinction map towards the central $10^{\\circ} \\times 10^{\\circ}$ of the Galaxy using the 2MASS Point Source Catalog. We extracted J and K$_s$ magnitudes for about 6 million stars in the range $8.0 \\le {\\it K_s} \\le 13.0$. The adopted map resolution is 4$^{\\prime} \\times 4^{\\prime}$. It was possible to obtain extinction values for $\\approx$80 \\% of the 32,761 cells defined in the area, where 2MASS data were currently available and a Bulge giant branch was distinct enough. The extinction affecting the bulk of the Bulge stellar population was determined by matching the upper giant branch found in the {\\it$K_s$}, ({\\it J-K$_s$}) colour magnitude diagram to the reference upper giant branch built using de-reddened Bulge fields. The extinction values vary from {\\it$A_K$}=0.05 in the edges of the map up to {\\it$A_K$}=3.2 close to the Galactic centre. The mean extinction found is ${\\it<A_K>}=0.29$ with a dispersion $\\sigma = 0.12$; 63 \\% of the cells are within 2-$\\sigma$ of the mean.  We compared our 2MASS extinction map to that of Schultheis et al. (1999) in the region $|\\ell| <5^{\\circ}$ and $|{\\it b}|<1.5^{\\circ}$, which is common to both studies. Schultheis et al. extinction map is based on DENIS photometry. We find an excellent agreement between the two extinction determinations, especially up to ${\\it A_K}=1.0$. Beyond this limit the values derived from the DENIS data are systematically larger. This small discrepancy in large extinction regions is not unexpected, considering the photometric errors, incompleteness effects, and uncertainties in extinction determination.  We also compared the present extinction map to that of Schlegel et al. (1998), which is based on dust emission in the far infrared detected by the DIRBE/IRAS instruments. As the data from the latter are affected by the entire dust column, with no depth limit, the comparison was made separately for regions of decreasing Galactic latitude {\\it b}, which supposedly correspond to increasing contribution by dust located on the background of the Galactic Centre. The background dust contribution was estimated by means of a double exponential dust distribution model. Some expected systematic biases, besides that caused by dust on the far side of the Galaxy, were also assessed and quantified.  In general, the extinction values derived from dust emission are higher than those from 2MASS, mainly close to the Galactic Plane and Centre. We detected two unexpected regions symmetric and close to the Galactic Centre where the two extinction estimates are of the same order. The lack of background dust in these low latitude regions could be explained by a process of dust grain destruction by UV emission from sources associated with continuous star formation and/or Post-AGB stars in the central parts of the Galaxy.  For the cells in the region $3^{\\circ}<|{\\it b}|<5^{\\circ}$, we observe a clear and roughly linear correlation between the {\\it$A_K$} values from 2MASS and dust emission. We also confirm, as was done in Paper I, that the {\\it$A_K$} values from 2MASS data are in general 73\\% smaller than those derived from dust emission. Since in this region the background dust contribution is less than 5 \\%, the differences between these two quantities should be smaller than observed. This discrepancy is also verified by Arce \\& Goodman (1999) in the Taurus Dark Cloud. It is probably due to systematic effects in the dust column density {\\it vs.} reddening calibration from Schlegel et al. (1998), yielding an overestimate of extinction in moderate to high extinction regions. We estimate a calibration correction factor of 76\\% for the FIR extinction values.  For the intermediate $1^{\\circ}<|{\\it b}|<3^{\\circ}$ region, the relation between DIRBE/IRAS and 2MASS extinction values departs more significantly from the identity line, as expected due to the larger contribution by background dust. In this region, the typical $A_{K,2MASS}/A_{K,FIR}$ ratio is 65\\% and could be explained by background dust contribution and the calibration factor affecting the FIR data. An enhancement in the foreground dust with respect to the dust model is observed in many cells in the northern strip. In the southern strip, several cells have $A_{K,2MASS}/A_{K,FIR}$ smaller than expected, probably due to dense dust clouds and temperature variations, currently not incorporated into our model.  For the regions very close to the Galactic Plane ($|{\\it b}| < 0.5^{\\circ}$), we have a typical value for the $A_{K,2MASS}/A_{K,FIR}$ ratio of 27\\%. Even considering the background dust contribution and the calibration factor, this ratio is still smaller than that predicted by our simple model for the dust distribution. This fact is probably due to the overestimation of the $A_{K,FIR}$ values by heated dust above that obtained from DIRBE temperature maps. Another possible contribution to this difference is the existence of systematic effects on the $A_{K,2MASS}$ values in high extinction regions ($A_{K,2MASS} > 2.5$), where the 2MASS extinction should be significantly underestimated or even unreliable.  A systematic asymmetry in the {\\it$A_K$} values relative to the plane of the Galaxy {\\it at $1^{\\circ}<|{\\it b}|<5^{\\circ}$} is observed both in the 2MASS and DIRBE/IRAS data. The behaviour and amplitude of this asymmetry with position on the sky suggest that the dominant role in creating this north-south asymmetry is a more effective presence of foreground dust clouds in the northern Galactic strips, such as the Pipe Nebula (Sect. 2.2). A possible explanation is stellar winds and supernovae from nearby OB stellar associations producing dust cloud shells (Bhatt 2000). The nearby clouds projected towards the central parts of the Galaxy at positive latitudes belong to the Ophiuchus dust complex. They are probably related to the association ScoOB2, which is at a distance of 145 pc from the Sun (Bhatt 2000; Onishi et al. 1999). ScoOB2, in turn, belongs to Upper Scorpius, which is the easternmost part of the  Sco-Cen Association, as studied by means of Hipparcos (de Zeeuw et al. 1999).  In all regions, significant substructure in the ${\\it A_{K,2MASS}} vs.$ ${\\it A_{K,FIR}}$ relation is seen, with loops and arms stretching out from the main relation. These structures are probably caused by intervening dust clouds, with different temperatures and densities for different lines of sight. One extremely interesting perspective is to model the dust distribution within the Galaxy, trying to reproduce as close as possible the details of the {\\it$A_K$} maps currently available. This effort demands models that incorporate, on top of a smooth dust distribution, the effects of individual dust clouds, spiral arms, molecular rings and other structure, possibly with variable density contrasts and temperatures. This effort is currently under way for the central region of the Galaxy."
    },
    "0209/astro-ph0209324_arXiv.txt": {
        "abstract": "We argue that the observed correlations between central black holes masses $M_{BH}$ and galactic bulge velocity dispersions $\\sigma_e$ in the form $M_{BH}\\propto\\sigma_e^4$ may witness on the pregalactic origin of massive black holes. Primordial black holes  would be the centers for growing protogalaxies which experienced multiple mergers with ordinary galaxies. This process is accompanied by the merging of black holes in the galactic nuclei. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209438_arXiv.txt": {
        "abstract": "I follow a bar from its formation, via its evolution, to its destruction and, perhaps, regeneration. I discuss the main features at each stage and particularly the role of the halo. Bars can form even in sub-maximum discs. In fact, such bars can be stronger than bars which have grown in maximum discs. This is due to the response of the halo and, in particular, to the exchange of energy and angular momentum between the disc particles constituting the bar and the halo particles at resonance with it. The bar slowdown depends on the initial central concentration of the halo and the initial value of the disc $Q$. Contrary to the halo mass distribution, the disc changes its radial density profile considerably during the evolution. Applying the Sackett criterion, I thus find that discs become maximum in many simulations in which they have started off as sub-maximum. I briefly discuss the evolution if a gaseous component is present, as well as the destruction and regeneration of bars. ",
        "introduction": "Bars can form spontaneously in galactic discs (e.g. Miller, Prendergast \\& Quirk 1970, Hohl 1971). They then evolve over a large number of rotations, changing their length, strength, shape and angular frequency. During this period they can drive spiral or ring formation, push gas to the center-most parts of the galaxy and thus trigger starbursts and activity in the nucleus, and interact with the outer disc, bulge and/or halo by exchanging energy and angular momentum with them. I will discuss a few of these processes, relying -- due to the strongly nonlinear nature of bars -- mainly on results of $N$-body simulations. ",
        "conclusions": ""
    },
    "0209/astro-ph0209112_arXiv.txt": {
        "abstract": "{It is shown that bright cluster dwarf ellipticals follow a relation between metallicity and apparent flattening. Rounder dwarfs tend to be more metal-rich. The evidence is based on colour as well as spectroscopic line-strength data from the literature. The large scatter of dEs around the mean metallicity-luminosity relation, usually ascribed to large observational errors, turns out to be an ellipticity effect. In the magnitude range $-14 \\ga M_B \\ga -18$ the metallicity of dEs depends more strongly on ellipticity than luminosity. A possible explanation is that galaxies with masses around $10^{9}$ $M_{\\odot}$ suffered a partial blowout of metal-enriched gas along their minor axis, rendering ellipticity a critical parameter for metallicity (De Young \\& Heckman 1994). ",
        "introduction": "Elliptical galaxies follow a well-known luminosity-metallicity relation, in the sense that more luminous ellipticals are observed to be globally redder, or have larger metallic absorption line strengths in their spectra than less luminous systems (Faber 1973, Visvanathan \\& Sandage 1977). Obviously this is a mass-metallicity relation, and its canonical interpretation is that less massive galaxies had more significant outflows of metal-enriched gas at an early evolutionary stage (Faber 1973, Mould 1984). The relation seems to hold for the entire luminosity range of spheroidal galaxies, down to the faintest dwarfs (Caldwell 1983, Brodie \\& Huchra 1991, Caldwell et al.~1992). However, there is also considerable scatter in the relation, and there have been attempts to find a {\\em second parameter}\\/ which governs metallicity (mass being the primary parameter). A good candidate is the internal velocity dispersion (Terlevich et al.~1984, Efstathiou \\& Fall 1984), which even more strongly correlates with metallicity than mass (Bender et al.~1993).  The largest scatter in the luminosity-metallicity relation is observed in the magnitude range of bright cluster dwarf ellipticals ($-14 \\ga M_B \\ga -18$). While this scatter could plausibly be due to measurement errors, Rakos et al.~(2001) find a weak correlation with age for a sample of Fornax cluster dwarfs observed in Stroemgren narrow-band colours.  In this letter we show that the scatter in metallicity for cluster dEs is largely explained by {\\em apparent ellipticity}\\/ (flattening). At a given luminosity, {\\em rounder dwarf ellipticals are more metal-rich}. There have already been hints that ellipticity might act as a second parameter in normal elliptical galaxies (Terlevich et al.~1984). However, for dEs the effect is so strong that ellipticity appears to be the {\\em primary}\\/ parameter. Such a metallicity-flattening relation for dEs is not implausible. The outflow of metal-enriched gas in stellar systems of intermediate mass ($M \\approx 10^{9}$ $M_{\\odot}$) is preferentially occurring along the minor axis, rendering ellipticity a critical parameter for the metallicity of present day dwarfs (De Young \\& Heckman 1994). {\\em Rounder dEs seem to have suffered less significant outflow}. ",
        "conclusions": "We have presented evidence for a metallicity-flattening relation for dwarf elliptical galaxies based on colour and metallicity data available from the literature. At a given total magnitude, rounder dEs are more metal-rich. In the narrow magnitude range of bright cluster dwarfs ($-14 \\ga M_B \\ga -18$), metallicity is more strongly correlated with ellipticity than luminosity (mass), i.e. ellipticity seems to be the primary parameter for the enrichment history of these galaxies. Possibly this holds only for {\\em cluster}\\/ dwarfs; the effect is not significant for local dwarfs. A possible explanation is provided by the scenario of De Young \\& Heckman (1994), where the outflow of gas, and hence the regulation of metallicity, depends on the intrinsic shape of a galaxy in the intermediate mass range around $10^9 M_{\\odot}$.  It would be highly desirable to strengthen the evidence with further observations, especially spectroscopically well determined metallicities for many cluster dwarfs. If confirmed, the effect is likely to be of importance for our understanding of the chemodynamical evolution of galaxies."
    },
    "0209/astro-ph0209262_arXiv.txt": {
        "abstract": "{ The X-ray background intensity around Lick count galaxies and rich clusters of galaxies is investigated in three \\ROSAT\\ energy bands. It is found that the X-ray enhancements surrounding concentrations of galaxies exhibit significantly softer spectrum than the  standard cluster emission and the average extragalactic background. The diffuse soft emission accompanying the galaxies is consistent with the thermal emission of the hot gas postulated first by the Cen \\& Ostriker hydrodynamic simulations. Our estimates of the gas temperature - although subject to large uncertainties - averaged over several Mpc scales are below $1$\\,keV, which is substantially below the temperature of the intra-cluster gas, but consistent with temperatures predicted for the local intergalactic medium. It is pointed out that the planned {\\it ROSITA} mission would be essential for our understanding of the diffuse thermal component of the background. ",
        "introduction": "The X-ray background (XRB) is mostly generated by discrete extragalactic sources (e.g. \\cite{lehmann01}, and references therein). Among those sources, various classes of AGNs constitute a dominating part. Probably 5 to 10\\,\\% of the soft XRB is produced by hot gas in clusters of galaxies. Around and below $1\\,{\\rm keV}$ hot plasma in the Galaxy also contributes to the total background flux (\\citealt{hasinger92}). Apart from the source contribution, truly diffuse emission of extragalactic origin is also expected. Using hydrodynamic simulations \\cite{cen99} investigated evolution of the primordial gas density and temperature. Baryons not condensed in stars and interstellar medium within galaxies, occupy  the intergalactic space and are spread over a wide range of temperatures and densities. This question is discussed in detail by \\cite{dave}, \\cite{bryan} and \\cite{croft}. According to all the simulations, the hottest phase is located in the high mass concentrations of clusters of galaxies and it is responsible for the cluster X-ray emission. Relatively cold phase with temperature below $10^5$\\,K reveals its presence by the Lyman alpha forest. The fraction which still escapes detection, described as {\\it Warm-Hot} Intergalactic Medium (WHIM), comprises of $30$ -- $40$\\,\\% of all baryons in the present-day universe (\\citealt{dave}). Its temperatures are between $10^5$ and $10^7$\\,K and the hottest and high density fragments surround mass concentrations as groups and clusters of galaxies. Thus, the thermal emission by WHIM should be easiest to detect in the soft X-rays in the vicinity of rich clusters and high galaxy density areas. However, as pointed by \\cite{bryan}, compact sources produce most of the background and observations put tight constraints on the level of truly diffuse XRB component.  Detailed analysis of the diffuse emission produced by the intergalactic medium has been presented by \\cite{bryan} and \\cite{croft}. Using hydrodynamic simulations they generated maps of the X-ray sky which included contribution from the WHIM component. These maps have been used to calculate the model autocorrelation function of the XRB at small angular scales (below $\\sim 10^\\prime$) as well as the cross-correlation with the galaxy distribution.  In the present paper the distribution of the XRB in the vicinity of galaxy concentrations is carefully investigated from the observational point of view. Lick counts (\\citealt{shane}, hereafter SW\\footnote{The Lick galaxy counts in $10^\\prime$ pixels have been kindly provided to us in the electronic form by Dr. M. Kurtz.}) and Abell clusters of galaxies (\\citealt{abell58}, \\citealt{abell89}) have been used to select areas of the high mass concentrations. Since the predicted WHIM temperatures are substantially lower than those of the intra-cluster gas, and the WHIM emission is softer that the average extragalactic XRB, the objective of our analysis is to look for the systematic variations of the XRB spectral slope as a function of the distance from peaks of the galaxy distribution. The expected WHIM contribution to the XRB is small in comparison to the total background flux. To increase signal-to-noise ratio we have measured the excess XRB flux around galaxy concentration using the cross-correlation technique.  The X-ray data contained in the \\ROSAT\\ All-Sky Survey (RASS) have been used.  For a comprehensive description of the RASS see \\cite{snowden90}, \\cite{voges} and \\cite{snowden95}. Due to limited angular resolution of the RASS we have concentrated on the larger scales than those investigated by \\cite{croft}. In the next section the basic concept of our investigation is presented and in Sect.~\\ref{results} results of calculations are given. We end our investigation with the short conclusions in the Sect.~\\ref{conclusions}. ",
        "conclusions": "}  A perceptible fraction of the XRB fluctuations is correlated with rich clusters of galaxies and with the overall galaxy distribution. The cross-correlations between the XRB and Abell clusters/SW galaxies extends up to several degrees. The large angular size enhancements of the XRB which surround regions of higher galaxy densities have distinctly softer spectrum than the genuine cluster emission. It is highly suggestive that these enhancements are generated by the WHIM. The halo emission is consistent with the thermal Bremsstrahlung with ${\\rm k}T$ below $1$\\,keV. Taking into account large uncertainties of the present estimates and a wide range of plasma temperatures predicted by \\cite{cen99} and \\cite{croft}, our results are in good agreement with theoretical models. Limited energy range used in this investigation combined with the moderate energy resolution of the PSPC data prevent us from more detailed conclusions.  This limitation applies also to our evaluations of the AGN spectral characteristics. Although the average X-ray colours of a population of AGNs are adequately mimicked by the thermal spectrum (Fig.~\\ref{r5_7_th}), simple power law spectral models also provide good fits (Fig.~\\ref{r5_7_pl}). Uncertainties (not shown in figures) of the AGN colours are of the order of $0.009$, $0.015$, $0.050$ for redshift bins $0.1 - 0.4$, $0.4 - 1.0$, $1.0 - 2.0$, respectively. Thus, all three samples do not exhibit significant colour differences. The majority of AGNs used in the present analysis are luminous X-ray sources and their average spectral properties are not identical with the average spectrum of weak sources which generate most of the XRB. Additionally, the integral soft XRB is contaminated by thermal emission of our Galaxy (\\citealt{hasinger92}). Importance of this effect in our calculations is illustrated in Fig.~\\ref{r5_7_pl} with dashed lines.  Most if not all shortcomings of the present investigation would be eliminated with the {\\it ROSITA} mission which is being proposed. The {\\it ROSITA} telescope mounted on board of the International Space Station would perform an all-sky survey within the energy range of $0.5-10$\\,keV and a sensitivity about 100 times better than the RASS. These characteristics are particularly suitable for the analysis of the low surface brightness features generated by the WHIM. Such investigation  requires large, unlimited field of view. Also wide energy range extending well above the WHIM domain would help to isolate the WHIM thermal emission from the non-thermal components of the X-ray emission correlated with the galaxy distribution. High sensitivity would substantially increase efficiency of the correlation method by reducing the photon noise statistics.  Finally, we would like to note that although the extended clouds of hot gas around clusters in principle are sources of the Sunyaev-Zel'dovich (S-Z) effect (\\cite{sunyaev}), simple estimates by \\cite{soltan96} and \\'Sliwa et al.\\  (in preparation) indicate that for temperatures of the gas lower than $1\\,{\\rm keV}$, the amplitude of the S-Z effect generated by halos is below the detection threshold of the {\\it COBE} DMR measurements.  \\vspace{2mm} ACKNOWLEDGEMENTS. The \\ROSAT\\ project has been supported by the Bundesministerium f\\\"ur Bildung, Wissenschaft, Forschung und Technologie (BMBF/DARA) and by the Max-Planck-Gesellschaft (MPG). This work has been partially supported by the Polish KBN grant 5~P03D~022~20."
    },
    "0209/astro-ph0209054_arXiv.txt": {
        "abstract": "The differential rotation of the sun, as deduced from helioseismology, exhibits a prominent radial shear layer near the top of the convection zone wherein negative radial gradients of angular velocity are evident in the low- and mid-latitude regions spanning the outer 5\\% of the solar radius. Supergranulation and related scales of turbulent convection are likely to play a significant role in the maintenance of such radial gradients, and may influence dynamics on a global scale in ways that are not yet understood.  To investigate such dynamics, we have constructed a series of three-dimensional numerical simulations of turbulent compressible convection within spherical shells, dealing with shallow domains to make such modeling computationally tractable.  In all but one case, the lower boundary is forced to rotate differentially in order to approximate the influence that the differential rotation established within the bulk of the convection zone might have upon a near-surface shearing layer.  These simulations are the first models of solar convection in a spherical geometry that can explicitly resolve both the largest dynamical scales of the system (of order the solar radius) as well as smaller-scale convective overturning motions comparable in size to solar supergranulation (20--40~Mm).  We find that convection within these simulations spans a large range of horizontal scales, especially near the top of each domain where convection on supergranular scales is apparent.  The smaller cells are advected laterally by the the larger scales of convection within the simulations, which take the form of a connected network of narrow downflow lanes that horizontally divide the domain into regions measuring approximately 100--200~Mm across.  We also find that the radial angular velocity gradient in these models is typically negative, especially in the low- and mid-latitude regions.  Analyses of the angular momentum transport indicates that such gradients are maintained by Reynolds stresses associated with the convection, transporting angular momentum inward to balance the outward transport achieved by viscous diffusion and large-scale flows in the meridional plane, a mechanism first proposed by \\citet{fou1975} and tested by \\citet{gil1979}.  We suggest that similar mechanisms associated with smaller-scale convection in the sun may contribute to the maintenance of the observed radial shear layer located immediately below the solar photosphere. ",
        "introduction": "\\notetoeditor{I am ok with single-column figures, except where I have indicated otherwise.  Figures 5 -- 10, 12, 13, 15, and 17 should be in color.}  \\begin{figure} \\plotone{f1.eps} \\caption{Variation of angular velocity $\\Omega/2\\pi$ with proportional radius $r/R$ at selected latitudes as inferred from helioseismic RLS inversions averaged over four years of GONG data (adapted from \\citealt{how2000a}).  Shear layers ({\\sl shaded}), as evidenced by more rapid variations of $\\Omega$ with radius, are observed near the base of the convection zone as well as near the surface, with the latter region extending from 0.95~$R$ to 1.00~$R$.  The gradients of $\\Omega$ in that near-surface shear layer at high latitudes are sensitive to the inversion method and data sets used \\citep{sch2002}}  \\label{fig:gong} \\end{figure}  Helioseismology has revealed that the differential rotation profile observed at the solar photosphere roughly extends throughout the bulk of the convection zone \\citep{tho1996,sch1998b}.  From about 0.75~$R$ to 0.95~$R$ (where $R$ is the solar radius), the angular velocity $\\Omega$ has a small radial gradient, particularly at mid latitudes, as seen in Figure~\\ref{fig:gong}.  In contrast, regions of strong radial shear are evident near both the bottom and top of the convection zone (shown shaded in Fig.~\\ref{fig:gong}), and these shear layers are believed to play important dynamical roles within the solar convection zone.  While the tachocline region at the base of the convection zone has commanded much recent attention (as it is likely the seat of the global solar dynamo and the associated 22-year magnetic activity cycle), the dynamics within the near-surface shear layer, extending from 0.95~$R$ to 1.00~$R$, are also likely to have additional dynamical consequences that affect the appearance and evolution of flows and magnetic structures visible at the surface.  Such dynamics are presently not well understood, but are now becoming accessible to study through direct numerical simulations that capture the important effects of sphericity, compressibility, and rotation.  Several questions arise about the near-surface shear layer.  First, what dynamical mechanisms within the coupling of turbulent convection with rotation leads to such a boundary layer involving negative radial gradients of $\\Omega$?  In contrast, in the bulk of the convection zone the gradients are much smaller and positive.  Second, why does this boundary layer have a depth of about 5\\% in solar radius?  Third, does the presence of such a strong radial shear zone play a significant role in the complex large-scale meandering flows and reversing meridional circulations that have been shown by helioseismology to coexist with the intense smaller-scale convection in the upper reaches of the convection zone?   These three issues motivate our studies as we seek to resolve both supergranulation and global-scale responses in our simulations of convection in rotating spherical shells. Computational constraints have encouraged us to begin by studying thin shells of such turbulent convection, encompassing at this stage only the upper portion of the solar convective envelope.  A preliminary account of such modeling is presented in \\citet{der2001}.  We shall here show that the resulting multi-scale convection is able to redistribute angular momentum so as to yield radial gradients in $\\Omega$ that are largely in accord with the helioseismic findings.  These first steps are important in defining more complex simulations to be undertaken within deep convection shells that capture much of the depth range of the solar convection zone.  The stratification within the near-surface shear layer serves to drive vigorous motions possessing a wide range of spatial and temporal scales, visible at the surface as the convective patterns of supergranulation, mesogranulation, and granulation \\citep{spr1990}. Many aspects of such small-scale but intensely turbulent convection, influenced by radiative transfer effects and complex equations of state and opacities, have been studied through three-dimensional simulations within localized planar domains positioned near the solar surface (e.g.~\\citealt{ste1998,ste2000,ste2001}). The driving in such convection is enhanced by the latent heat released within the ionization zones of helium and hydrogen that are present in the near-surface shear layer (e.g.~\\citealt{ras1993}).  These small-scale turbulent convective motions are likely to facilitate the transport of angular momentum along both radial and latitudinal velocity gradients within the shear layer, and thus may be able to affect the dynamics within the convection zone on a more global scale.  In particular, the horizontal extent and overturning time of supergranular flows suggest that such convection will be at least weakly influenced by rotational effects, which can yield Reynolds stresses of significance in transporting angular momentum within the layer.  The coupling of turbulent compressible convection with rotation has also been studied extensively in localized $f$-plane domains (e.g.~\\citealt{bru1995,bru1996,bru1998,brum2002,bra1996,cha2001}) using perfect gases, revealing that the presence of coherent structures associated with strong downflow plumes and networks play a crucial role in the redistribution of angular momentum.  Such studies are complemented by a broad range of other simulations of compressible convection that exhibit intrinsic asymmetries between upflows and downflows, of complex vorticity structures that influence the transport of heat, momentum and magnetic fields, and of rich time-dependence involving a broad range of time scales (e.g.~\\citealt{cat1991,por1994,por2000,sai2000,rob2001,tob2001}).  Velocity features larger than the spatial scale of solar supergranulation are also in evidence in the near-surface shear layer. Bands of slightly faster rotation, or torsional oscillations, that gradually propagate toward the equator as the magnetic activity cycle advances are detected in Doppler measurements of the surface \\citep{lab1982,hat1996,ulr1998}.  They are also seen in global $f$- and $p$-mode helioseismic studies (e.g.~\\citealt{kos1997,sch1999,how2000b,vor2002}), and are present over at least the outer 8\\% in radius.  Even more complex flows within the near-surface shear layer, now called solar subsurface weather \\citep{too2002}, are revealed by local-domain helioseismic techniques such as ring-diagram analyses (e.g.~\\citealt{hil1988,hab1998,hab2000,bas1999}) and time-distance methods (e.g.~\\citealt{duv1993,gil1997,duv2000,cho2001}).  Mappings of subsurface flow fields over a range of depths reveal evolving large-scale horizontal flows that are somewhat reminiscent of jet streams, meridional circulations that may possess multi-celled structures in one hemisphere and not in the other, and distinctive flow deflection in the vicinity of active complexes \\citep{hab2002}.  Although the flow speeds in the meridional circulations are only of order 20~m~s$^{-1}$, they may be quite effective in redistributing angular momentum in latitude, thereby coupling widely separated regions within the near-surface shear layer and possibly having a role in its existence.  Photospheric magnetic field observations reveal structured concentrations that also possess a wide range of spatial scales, including active regions, sunspots, pores, and emergent flux elements.  On the smallest observable scales, concentrations of filamentary magnetic flux elements are found to be laterally advected by larger-scale surface flow patterns.  Outflows associated with the convective patterns of supergranulation and granulation in particular are observed to readily advect such small-scale flux toward intercellular lanes and concentrate these fields on scales small enough for dissipation to occur \\citep{sch1997,ber1998}.  Figure~\\ref{fig:gong} indicates that the radial gradient of angular velocity $\\Omega$ is largely negative at low and mid latitudes within the near-surface shear layer, such that the rotation rate decreases by about 2--4\\% as one moves outward across the layer.  Such a radial gradient in $\\Omega$ may be interpreted as a tendency for fluid parcels in the convection zone to partially conserve their angular momentum as they move toward or away from the axis of rotation.  This idea was originally suggested by \\citet{fou1975}, and may explain why larger-scale magnetic tracers at the surface have a faster rotation rate relative to the photospheric plasma, assuming that these magnetic features are anchored at a radius below the photosphere where the rotation rate is faster.  Numerical simulations  of Boussinesq fluids confined to thin shells \\citep{gil1979} showed that angular momentum is roughly conserved along radial lines, and small-scale convective motions are able to transport angular momentum inward, thereby maintaining the negative radial gradient of rotation rate with radius for such an incompressible fluid.  Whether the same is true for a compressible fluid is one of the topics addressed in this paper.  The maintenance of the relatively small radial gradients of $\\Omega$ throughout the bulk of the convection zone must be a direct consequence of the interaction of rotation with the turbulent fluid motions that exist within the solar convection zone.  Recent three-dimensional numerical simulations of such deep convection within rotating spherical shells (e.g.~\\citealt{mie2000,ell2000,bru2001a,bru2002}) indicate that for a range of parameter values, solar-like differential rotation profiles can be established, even with viscous and thermal diffusivities in a regime far removed from their solar values.  Many of these simulations possess about a 30\\% contrast in angular velocity $\\Omega$ between the equator and high latitudes and have small radial gradients of $\\Omega$ in the mid-latitude regions, features that roughly match the helioseismic determinations of the interior rotation profile within the bulk of the convection zone.  Analyses of the angular momentum transport within these simulations indicate that the fast equatorial rotation relative to the higher latitudes is primarily maintained by a complex interplay between global meridional circulation and Reynolds stresses achieved within the domains, both of which contribute to the equatorward transport of angular momentum with latitude.  The radial velocity planforms within the more laminar convection zone simulations take the form of rotationally aligned banana cell structures, with the downflowing fluid lanes extending throughout most of the radial extent of the domain.  As the level of turbulence is increased, these organized patterns become less prominent, giving way to a network of narrower downflow lanes that form plume-like structures at the interstices in the network.  Such plumes tend to possess significant vortical motion and span the entire domain in radius.  The influence of rotation on these vertical plumes preferentially tilts these structures such that they are partially aligned with the axis of rotation, which in turn creates the Reynolds stresses that facilitate angular momentum transport within the domain.  These deep-shell simulations of the convection zone typically place the upper boundary at about 0.96~$R$, and thus do not capture the smaller scales of convection that exist closer to the surface.  Consequently, the convection in even the most turbulent of these simulations involves overall pattern scales of order $20^\\circ$--$30^\\circ$, or several hundred~Mm, although the sheets and plumes associated with the downflow network are individually narrower and more concentrated.  To understand more clearly some of the physical processes occurring within the near-surface shear layer of the solar convection zone, we have constructed numerical simulations of compressible fluids within thin spherical shells that extend up to 0.98~$R$, encompassing solely the near-surface shear layer region.  Continual advances in supercomputing technology now permit three-dimensional compressible fluid simulations that explicitly resolve spatial and temporal scales spanning several orders of magnitude.  As a result, we are able for the first time to employ direct numerical simulations to investigate the effects of supergranular-sized convection on the more global dynamics within the near-surface shear layer of the sun.  In formulating our simulations, we have adopted the viewpoint that the latitudinal variation of the angular velocity in the sun, with the equatorial regions rotating more rapidly than the poles as in Figure~\\ref{fig:gong}, is established and maintained within the bulk of the convection zone somewhere below the lower boundary of our thin shell models.  We have thus imposed a solar-like differential rotation profile as a no-slip lower boundary in three of the four simulations presented here, in order to capture some of the dynamical effects related to such an angular velocity structure.  In so doing, we are implicitly assuming that the global differential rotation profile is not substantially affected by the convection within our thin shells, even though in the sun such shearing layers could have subtle effects on these dynamics.  Our primary focus in this paper will be to investigate the angular momentum transport achieved by multi-scale convection, involving both global and supergranular scales, within shearing layers analogous to the near-surface shear layer of the sun.  We shall consider radial stratifications that resemble ones deduced from stellar structure models over the depth range being studied, though the physics of the gases is highly simplified.  After briefly discussing in \\S\\ref{sec:ashcode} the governing equations and numerical approach used in solving them, we review the parameters used to initialize our thin-shell simulations in \\S\\ref{sec:thinshellsetup}.  We next examine in \\S\\ref{sec:modelflows} the multi-scale convective velocity patterns of the mature solutions, and discuss the meridional circulation, time-evolution and angular momentum balance achieved within the thin shell domains.  Lastly, we discuss the connection between these simulations and the near-surface shear layer of the solar convection zone, and present possible directions for future research in \\S\\ref{sec:conc}. ",
        "conclusions": "\\label{sec:conc}  We have presented results of three-dimensional numerical simulations of turbulence confined to thin rotating spherical shells, seeking to understand some of the dynamical effects that supergranular scales of motion within thin shearing layers might have within the analogous layer located near the top of the solar convection zone.  We have focused our analysis on the physical processes that enable the transport of angular momentum within the thin shell domains, in order to investigate analogous angular momentum transport on the sun, and to determine the cause of the negative radial angular velocity gradients shown to exist within the near-surface shear layer of the sun.  The high-resolution simulations presented here allow horizontal structures of order 10~Mm to be explicitly resolved, thereby allowing us for the first time to incorporate dynamical scales on the order of solar supergranulation within global simulations of solar convection.  We find that the broad spectrum of scales of motion, while typically much smaller in size than the largest characteristic length scales of the convection zone, are able to influence the large-scale dynamics of the system through their ability to transport angular momentum within the shells on a global scale.  The vigorous convection realized in each of the four simulations presented here is driven by imposing the solar heat flux at the lower boundary of each domain.  Shell thicknesses of 4\\% and of 8\\% of the solar radius $R$ are considered, with the upper boundary in each case located at a radius of 0.98~$R$.  We have imposed a differentially rotating no-slip lower boundary in three of the four simulations.  Both the lower and upper boundaries are impenetrable, and the upper boundary is forced to remain at a constant entropy.  We find in all cases that near the middle of the domain the convection takes the form of a connected network of fast yet narrow downflow lanes that enclose broad regions of warmer, more slowly rising fluid.  The cells enclosed by the downflow network typically measure 100--200~Mm across, with the lanes themselves about 20~Mm wide.  In the deeper layers where the density is greater, this network loses much of its horizontal connectivity, instead forming more plume-like structures that approach the bottom of the domain, at which point the impenetrability of the lower boundary forces the fluid to be diverted horizontally.  Closer to the surface, the broad cells of upwelling fluid are found to segment into several smaller upflows having a horizontal scale comparable to that of solar supergranulation, of order 20--40~Mm.  These small-scale upflow cells appear in both the shallow- and deep-shell simulations, suggesting that the more superadiabatic stratification present near the top of each domain, rather than the depth of the shell, is the primary factor that determines the morphology of the convection near the surface.  Time series of the evolving near-surface flow field show that both the smaller upflow cells as well as the narrow downflow lanes are horizontally advected in a sustained fashion as they respond to larger-scale sweeping flows that develop nearby.  Averaging the flow fields in longitude reveal that the angular velocity decreases with radius in the low- and mid-latitude regions of each domain, with the exception of a thin viscous boundary layer that forms near the no-slip lower boundary in each case.  An analysis of the angular momentum balance shows that such negative radial gradients of angular velocity are maintained by an inward transport of angular momentum, achieved by Reynolds stresses associated with the convective motions that balance the outward transport of angular momentum from viscous diffusion and the global meridional circulation.  Such an inward transport is achieved if radially moving fluid motions, such as the broad upflows and strong downflows seen here, have the tendency to conserve their angular momentum per unit mass while moving radially throughout the shell.  The longitudinally averaged meridional velocity patterns take the form of a series of $15^\\circ$ latitudinal rolls that span the full radial thickness of each shell, with cells having a poleward surface flow tending to have a broader latitudinal extent than cells with the opposite sense.  The net effect of these rolls is to transport angular momentum poleward, as required by latitudinally varying angular momentum flux imposed by the differentially rotating lower boundary.  However, we reemphasize that the profiles of meridional circulation within these thin-shell simulations are significantly influenced by the impenetrable radial boundaries of our simulations, effectively forcing any circulation in the meridional plane to be completely enclosed within the domain.  As a result, the meridional flow profiles realized here are not expected to resemble the solar case.  The continual advance of supercomputing technology will allow simulations of convection within thin spherical shells to be extended to deeper layers in the future. Such models would preclude having to add angular momentum to the system via no-slip boundary conditions, as the differential rotation and near-surface shear layer could then be computed in a self-consistent manner. In addition, large-scale flows driven by the small-scale convective patterns that would not normally be confined to the near-surface layers (such as the meridional circulation) would then be allowed to feed back on the deeper layers below.  Preliminary attempts to construct such global models, encompassing the bulk of the convection zone as well as a more highly resolved layer where convection on supergranular scales can exist, are currently underway.  We also believe the treatment of sub-grid scale (SGS) convective motions not explicitly resolved in our simulations deserves considerable attention in the future.  The current prescription, whereby the diffusivities are enhanced over their thermal and molecular values, is adopted only for simplicity, and likely does not capture all of the relevant effects of the unresolved scales on the global dynamics.  Other treatments, such as those discussed in \\S\\ref{sec:ashcode} whose functional forms depend on the shearing properties of the resolved flows, may be more appropriate.  Nevertheless, the thin-shell simulations presented here contain highly evolving, multi-scale convective motions that are able to efficiently redistribute angular momentum in both radius and latitude.  Such motions are found to maintain the radial shear within the domains, even in this idealized environment that only approximates the near-surface shear layer of sun by decoupling it from the bulk of the convection zone.  We also speculate that convection in the near-surface layers of the sun may behave in a similar fashion, maintaining the negative radial gradients in the near-surface shear layer of the sun as deduced from helioseismology.  While we are admittedly still far removed from directly modeling a convective layer with realistic solar parameters, it is encouraging that the flow patterns realized in these simulations exhibit the multiple scales of supergranulation and the more global convection cells being deduced from local helioseismic probing of the near-surface shear layer."
    },
    "0209/gr-qc0209081_arXiv.txt": {
        "abstract": "We argue that space-time properties are not absolute with respect to the used frame of reference as is to be expected according to  ideas of relativity of space and time properties by Berkley - Leibnitz - Mach - Poincar\\'{e}. From this point of view gravitation may manifests itself both as a field in Minkowski space-time and as space-time curvature. If the motion of test particles is described by the Thirring Lagrangian, then in the inertial frames of reference, where space-time is pseudo-Euclidean, gravitation manifests itself as a field. In reference frames, whose reference body is formed by point masses moving under the effect of the field,  it appears as Riemannian curvature which in these frames is other than zero. For realization of the idea the author bimetric gravitation equations are considered. The spherically - symmetric solution of the equations in Minkowski space-time does not lead to the physical singularity in the center. The energy of the gravitational field of a point mass is finite. It follows from the properties of the gravitational force that there can exist stable compact supermassive configurations of Fermi-gas without an events horizon. ",
        "introduction": "The key reason preventing a correct inclusion of the Einstein  theory of gravitation in the interactions unification is that gravity is identified with space-time curvature. It is also a cause of such unsolved problems of the theory as an operational definition of the observable variables, the energy - momentum tensor problem and gravity quantization. In the present paper starting from \\cite{Vr_Verozub81}, \\cite{Vr_Verozub95} we consider a likely reason of gravity geometrization. We argue that the gravitation properties are not absolute with respect to the used frame of reference. In inertial frames of reference gravitation can be considered as a field in flat space-time, while in so called proper frames of reference it manifests itself as space-time curvature.  The author's gravitation equations which realize this idea are considered in details. They do not contradict available experimental data. The physical consequences resulting from the equations differ very little from the ones in general relativity if the distances from the attracting mass are much larger than the Schwarzschild radius $r_{g}$. However, they are completely different at the distances equal to $r_{g}$ or less than that. A number of new physical consequences follow from the equations. ",
        "conclusions": "It follows from the above results that the equations under consideration do not contradict available experimental data obtained in the Solar system. In paper \\cite{Vr_VerKoch2000} these equations were tested by the binary pulsar PSR1913+16 and it was found out that the results are very close to the ones in general relativity. It is a consequence of the fact that the used distances from attracting masses are much larger than the Schwarzshild radius. At the conditions the function $f(r)$ is very close to the radial distance $r$. However, the physical consequences between these equations are completely different at the distances $r\\leq r_{g}$. The events horizon is absent. There can exist supermassive configurations of the degenerated Fermi-gas. Candidates to the objects of such a kind are the galactic centers (\\cite{Vr_Eckart} $\\div$ \\cite{Vr_Bender})."
    },
    "0209/astro-ph0209525_arXiv.txt": {
        "abstract": "The statistical shapes of the highest pulse have been studied by aligned method. A wavelet package analysis technique and a developed pulse--finding algorithm have been applied to select the highest pulse from burst profiles observed by BATSE on board CGRO from 1991 April 21 to 1999 January 26. The results of this work show that the statistical shapes of the highest pulses are related to energy: the higher the energy, the narrower the pulse. However, the characteristic structures of the pulses have nothing to do with energy, which strongly supports the previous conclusion that the temporal profiles in different channels are self--similar. The characteristic structures of the pulses can be well described by a model proposed by  Norris et al. (1996). The fitting parameters are: $t_r$=0.12, $t_d$=0.16, $\\upsilon =1.09$, the ratio of $t_r$ to $t_d$ for the pulse is 0.75. The result leads to our conjecture that the mechanisms of bursts in different gamma-ray bands might be the same. The shock, either an internal or an external one, producing the pulse, might emit photons over the four energy channels in the same way. ",
        "introduction": "Gamma--ray bursts(GRBs), which are still mysterious, have very complex temporal structure. Their temporal profiles are enormously varied --- no two bursts have ever been found to have exactly the same temporal and spectral development. The temporal activity is suggestive of a stochastic process (% Nemiroff et al. 1993). The diversity of the bursts seems to be due to random realization of the same process that is self-similar over the whole range of timescale. Attempts to quantify these structures have not been successful (e.g., Fishman 1999).  Most of the observed profiles of GRBs are composed of pulses, each comprising a fast rise and an exponential decay (a FRED; e. g., Desai 1981, Fishman et al. 1994). Many methods for pulse analysis have been developed, e. g., the parametric analysis in model fitting (Nemiroff et al. 1993, Norris et al. 1996), the auto-correlation method (Fenimore et al. 1995), the nonparametric method (Li \\& Fenimore 1996% ), the peak alignment and normalized flux averaging method (Mitrofanov et al. 1996, Mitrofanov et al. 1998, Ramirez--Ruiz \\& Fenimore 1999, Ramirez--Ruiz \\& Fenimore 2000), and the pulse decomposition analysis method (Lee et al. 2000), etc. These statistical studies have revealed many observed temporal signatures of pulses. The pulses are hypothesized to have the same shape at all energies, differing only by scale factors in time and amplitude (``pulse scale conjecture''). And, the pulses are hypothesized to start at the same time, independent of energy (``pulse start conjecture''). The two conjectures were confirmed by Nemiroff (2000). In general, higher energy channels show shorter temporal scale factors (e.g., Norris et al. 1996., Nemiroff 2000). It is found that the temporal scale factors between a pulse measured at different energies are related to that energy by a power law, possibly indicating a simple relativistic mechanism is at work (Fenimore et al. 1995, Norris et al. 1996., Nemiroff 2000).   The statistical pulse shape has been well studied by the peak alignment and normalized flux averaging method. The peak aligned averaging pulse is spiky. A succinct pulse model, which well describes many pulse shapes, was proposed by Norris et al. (1996): \\begin{equation} I(t)=I_0e^{-(|t-t_{\\max }|/{t_{r,d}})^\\upsilon } \\end{equation} where $t_{max}$ is the time of the pulse's maximum intensity ($I_0$); $t_r$ and $t_d$ are the rise and decay time constants, respectively; and $\\upsilon $ is a measure of the pulse sharpness, which was referred to ``peakness'' by Norris et al. (1996).  However, both the duration and total count of pulses vary significantly. Statistical properties of the pulses revealed by the peak alignment and flux--normalized averaging method are limited. In this paper, this method is developed to study the shape of pulses in a more detailed manner.  To reach a result of high quality, we concern in this paper only the highest pulse of bursts, where one finds the highest level of signal--to--noise. We make the noise decomposition for the time profile of bursts by performing the wavelet analysis (which is described in section 2), then modify the pulse--finding algorithm proposed by Li \\& Fenimore (Li \\& Fenimore 1996) to identify the highest pulse in a burst profile (see section 3). In sections 4 and 5, we employ and develop the pulse aligned method to study the flux--normalized aligned averaging pulse shape and the count--and--duration--normalized aligned averaging pulse shape, respectively. Conclusions and discussion are presented in section 6. ",
        "conclusions": "In this paper we study the flux--normalized--peak(beginning)--aligned averaging pulses and the count-and-duration-normalized--peak(beginning)--aligned averaging pulses with the highest pulse in the profiles of GRBs. We apply the wavelet package analysis technique and a developed pulse--finding algorithm to select the highest pulse. The sample we get includes 275 bursts which fluxes range from 0.513 erg.cm$^{-1}$.s$^{-1}$ to 183.370 photons.cm$^{-1}$.s$^{-1}$.  The wavelet package analysis technique is suitable to treat those signals which cannot be analyzed by the traditional Fourier method. It is successful in de--noising the original signal and identifying the structure within a burst. The pulse--finding algorithm used in Li \\& Fenimore (1996) is developed in this paper so that the selection of pulses depends on the relative value of counts rather than the absolute value.  The number of bursts concerned in this paper is the largest one of that used for pulse analysis so far, and the sample adopted here covers the biggest flux range. Many samples of very bright bursts have been employed to study statistical properties of pulses (e.g., Norris et al. 1996). Though the numbers of bursts concerned are much smaller, the authors were able to get more pulses by selecting not only the highest pulse but also other pulses in a burst (e.g., Norris et al. 1996). Their results refer only to bright bursts, and the pulses so selected might include possible evolutionary effects of pulses.  Figures 2 and 3 support the well-known conclusion that the higher the energy, the narrower the pulse.  Different from the previous works, we make in this paper not only the normalization of the total count but also the normalization of the duration. In this way, all the pulses (strong or weak) contribute equally to the averaging pulses. And, in this way, the averaging pulses obtained stand only for the statistical shape of pulses. The effects from both the duration and the total count are removed. The results so obtained are quite different from that got by the previous method. One can find this by comparing Figs. 2 and 4.  For the pulses shown in Figs. 4 and 5, we prefer those in the latter. Since the pulses in Fig. 4 come from aligning the normalized pulses at the moment of maximum count of the pulses, those asymmetric normalized pulses would contribute differently to different sides of the averaging pulses. Thus the distribution of the peak count position in the shape of selected pulses must be at work. It would lead to a spiky shape of the averaging pulses. Fig. 4 must conceal most of the diversities of the duration and the asymmetry of pulses. Differently, the pulses shown in Fig. 5 stand only for the average shape of the original pulses. These pulses are well described by the model in Norris et al. (1996). The fitting parameters for the pulse in channel 3 in Fig. 5 are: $t% _r$=0.12, $t_d$=0.16, $\\upsilon =1.09$. The ratios of $t_r$ to $t_d$ for the pulse is 0.75.  We find that the count-and-duration-normalized--peak(beginning)--aligned averaging pulses are the same for different channels. Our results strongly support the previous conclusion that the temporal profiles in different channels are self--similar. The averaging pulse shape is independent of energy bands. Due to these results, we believe that the mechanisms of bursts in different gamma-ray bands must be the same.  The mechanism generating the bursts is still unknown. Many models for interpreting the origin and emission of the event have been proposed (e. g., Rees \\& M$\\acute{e}$sz$\\acute{a}$ros 1992, Vietri et al. 1998, Fuller et al. 1998, Dai \\& Lu 1998, Daigne et al. 1998, \\cite {Pa00}, etc.), mostly in the context of two major scenarios involving relativistic shells. An approach frequently used in these models is to identify each pulse in the light curve with a single event. Depending on the model chosen, this event could be the collision between inhomogeneities in a relativistic wind in the internal models or the ``activation'' of a region on a single external shell. Our study shows that, the shock, either an internal or an external one, producing the pulse, might produce photons over the four energy channels in the same way."
    },
    "0209/astro-ph0209239_arXiv.txt": {
        "abstract": "{In previous work, Gardiner et al. (\\cite{GKS99}) found evidence for a discrepancy between the \\teff\\ obtained from Balmer lines with that from photometry and fundamental values for A-type stars. An investigation into this anomaly is presented using Balmer line profiles of stars in binary system with fundamental values of both \\teff\\ and \\logg. A revision of the fundamental parameters for binary systems given by Smalley \\& Dworetsky (\\cite{SD95}) is also presented. The \\teff\\ obtained by fitting \\halpha\\ and \\hbeta\\ line profiles is compared to the fundamental values and those obtained from \\uvby\\ photometry. We find that the discrepancy found by Gardiner et al. (\\cite{GKS99}) for stars in the range 7000~K $\\la \\teff \\la$ 9000~K is no longer evident. ",
        "introduction": "Balmer lines are an important diagnostic of stellar atmospheric structure since they are formed at a wide range of depths within the atmosphere. In addition, the depth of formation of \\halpha\\ is higher than that of \\hbeta, thus observations of these profiles provide useful diagnostics (e.g. Gardiner \\cite{GAR00}). Balmer profiles are relatively insensitive to surface gravity for stars cooler than $\\sim$8000~K (Gray \\cite{GRA92}; see also Heiter et al. \\cite{HEI+02}). In addition, Balmer profiles are sensitive to the treatment of atmospheric convection (van't Veer-Menneret \\& Megessier \\cite{VM96}; Castelli et al. \\cite{CGK97}; Gardiner \\cite{GAR00}; Heiter et al. \\cite{HEI+02}). For stars hotter than $\\sim$8000~K, the profiles are sensitive to both effective temperature and surface gravity. However, provided we know surface gravity from some other means (e.g. from eclipsing binary systems), we can use them to determine effective temperature.  In previous work, Smalley \\& Kupka (\\cite{SK97}, hereafter SK97) found no significant systematic problems with \\uvby\\ and fundamental (and standard) stars. In fact, \\uvby\\ was found to be very good for obtaining \\teff\\ and \\logg. Using \\halpha\\ and \\hbeta\\ profiles, Gardiner et al. (\\cite{GKS99}, hereafter GKS99) found that both the Canuto \\& Mazzitelli (\\cite{CM91,CM92}) and standard Kurucz (\\cite{KUR93}) mixing-length theory without overshooting (MLTnoOV) (see Castelli et al. \\cite{CGK97}) are both in agreement to within the uncertainties of the fundamental stars. Overshooting models were always clearly discrepant. However, GKS99 found some evidence for significant disagreement between \\emph{all} treatments of convection and fundamental values around 8000 $\\sim$ 9000~K. In this region the effects of \\logg\\ cannot be ignored. In GKS99, most of the \\teff\\ stars do not have fundamental values of \\logg. Thus, possible \\logg\\ bias could have occurred. In this paper, we use binary systems with fundamental values of \\logg, determine revised fundamental values of \\teff\\ and compare the results with those from Balmer lines.  Binary systems with fundamental values of \\logg, with in some cases fundamental values of \\teff, were discussed by Smalley \\& Dworetsky (\\cite{SD95}, hereafter SD95). Their list was limited to four systems, mainly due to lack of trigonometric parallax measurements. Fortunately, the Hipparcos mission (ESA, \\cite{ESA97}) has provided trigonometric parallaxes for all of the binary systems considered by SD95. Thus we re-evaluated 15 of those systems, using the methods of SD95, with slight modifications and a discussion of the sources of uncertainty.  Theoretical Balmer line profiles are compared to observations and the required values of \\teff\\ are derived which when used in a model atmosphere will predict the correct profiles. We have considered three models of convection: the standard mixing-length theory {\\sc atlas9} models (Kurucz \\cite{KUR93}; Castelli et al. \\cite{CGK97}), with and without approximate convective overshooting, and modified {\\sc atlas9} models based on the turbulent convection theory proposed by Canuto \\& Mazzitelli (\\cite{CM91,CM92}) and implemented by Kupka (\\cite{KUP96}). ",
        "conclusions": "The availability of the Hipparcos parallax measurements has enabled the list of stars with fundamental values of both \\teff\\ and \\logg\\ to be considerably extended, from the 4 originally given by Smalley \\& Dworetsky (\\cite{SD95}), to the 15 presented here. Even when the available optical flux measurements are limited to only $UBV$ magnitudes, the quality of the final \\teff\\ values is good. In some cases, it is the uncertainty of the Hipparcos parallax measurements that limits the accuracy of the \\teff\\ obtained. The stars with IRFM values are mostly in very good agreement with the fundamental values, showing that the two methods are self-consistent and reliable. Since there are more systems with \\teff\\ values from the IRFM (e.g. Blackwell \\& Lynas-Gray \\cite{BL-G94}; Alonso et al. \\cite{ALO+95}), this method can be used to obtain `near-fundamental' values, provided we avoid binary systems with markedly dissimilar components (Smalley \\cite{SMA93}).  Balmer line profiles have been fitted to the fundamental binary systems. To within the errors of the fundamental \\teff\\ values, neither the \\halpha\\ or \\hbeta\\ profiles exhibit any significant discrepancies for the CM and MLT\\_noOV models. As in previous work, the MLT\\_OV models are found to be discrepant. Moreover, there are no systematic trends, such as offsets, between results from \\halpha\\ and \\hbeta\\ as long as $\\alpha$ in MLT models is chosen small enough (e.g. 0.5). The discrepancies exhibited by the fundamental \\teff\\ stars in GKS99 can be explained by rapid rotation in two cases and by the fact that the Balmer profiles become sensitive to \\logg\\ and less sensitive to \\teff\\ in the other two cases. However, for the time being the lack of any stars with fundamental values of both \\teff\\ and \\logg\\ in this region precludes the conclusion that there is not a problem with the models in the \\teff\\ range 8000 $\\sim$ 9000~K."
    },
    "0209/astro-ph0209149_arXiv.txt": {
        "abstract": "In a recent paper (Ruffa \\cite{ruffa}) it was proposed that the massive black hole at the Galactic center may act as a gravitational lens focusing gravitational wave energy to the Earth. Considering the gravitational wave signal emitted by galactic spinning pulsars, an enhancement in the gravitational wave intensity by a factor of a few thousand is found. For galactic and extra-galactic sources the intensity enhancement can be as high as 4,000 and 17,000, respectively. In this note we consider the probability of significant signal enhancement from galactic and extra-galactic pulsars by the proposed mechanism and find that it is actually negligible. The lensing effect due to a possible companion object (a star or the galactic center black hole) of the gravitational wave source is also investigated in the framework of the classical microlensing theory. ",
        "introduction": "Gravitational lensing of electromagnetic waves is a well known phenomenon predicted by the General Theory of Relativity  (for a review on this issue see Schneider, Ehlers and Falco \\cite{sef}). In principle, gravitational lensing of gravitational waves should occur in the same way as it does for light. The most obvious difference is that gravitational wave propagation is not disturbed by dust grains, as happens for light, so that the central part of our galaxy may be investigated by using the next generation of gravitational wave detectors.  In a very interesting paper Ruffa (\\cite{ruffa}), assuming that the mass of the Galactic center is in the form of a massive black hole with mass $% M\\simeq 2.6\\times 10^6$ M$_{\\odot}$ (for a Galactic center overview see Eckart, Genzel, Ott and Schoedel \\cite{egos}), the gravitational wave lensing problem was studied by a typical Fraunhoffer diffraction approach. It was pointed out (Ruffa \\cite{ruffa}) that extra galactic sources can be amplified by a factor of about $17,000$ and galactic neutron stars by over $% 4,000$. The author also argued that the Earth would take about 10.1 days to traverse the focused region of the extra galactic sources. For galactic bulge sources the focused region would be scaled down by a factor of nearly 3 and hence the observing time would be reduced to about 3 days.  The question that naturally arises is ``how likely is it that we see the proposed enhancement?'' To answer this question we need to estimate the number of sources that could be expected to be observed, both galactic and extra-galactic. In principle, of course, there are many extra-galactic sources as there are $\\sim 10^{11}$ galaxies. However, we need to limit the number, applying a cut-off by requiring that the expected intensity (after amplification) be greater than the sensitivity of the detector.  As will be seen, the chances of seeing the dramatic enhancement calculated by Ruffa are extremely small. This is due to the fact that Ruffa assumes a very special geometry, with the source, lens and observation point aligned. Whereas Ruffa considered Fraunhoffer diffraction, an analysis using ``geometric optics'' for gravitational waves had been undertaken, in which the special geometry of Ruffa was not assumed (De Paolis, Ingrosso and Nucita \\cite{din}). This analysis gave a much higher probability but a lower enhancement.  In this paper, we evaluate the probability of enhancement of the gravitational wave signals as a consequence of diffraction by massive compact objects in the highly aligned geometry.  We also consider the lensing effect due to a possible companion star of the gravitational wave source in the framework of the classical microlensing theory. In addition to this case we consider the gravitational wave lensing due to the massive black hole at the galactic center. As we shall see, the gravitational wave signal amplification due to the companion object (or the black hole at the galactic center) might be detectable by the VIRGO detector for some orbital parameters of the binary system. ",
        "conclusions": ""
    },
    "0209/astro-ph0209133_arXiv.txt": {
        "abstract": "{  We  present  radio  continuum  images  of  three  Galactic \\ion{H}{ii}  regions, S~201,  S~206,  and S~209  near  232, 327,  and 610~MHz using  the Giant Meterwave Radio Telescope  (GMRT).  The GMRT has a  mix of  short and long  baselines, therefore, even  though the data have  high spatial resolution,  the maps are still  sensitive to diffuse  extended  emission.   We  find that  all  three  \\ion{H}{ii} regions have  bright cores surrounded  by diffuse envelopes.   We use the high  resolution afforded  by the data  to estimate  the electron temperatures  and emission  measures of  the compact  cores  of these \\ion{H}{ii}  regions.  Our  estimates  of electron  temperatures  are consistent  with  a  linear  increase of  electron  temperature  with Galacto-centric  distance  for distances  up  to  $\\sim 18$~kpc  (the distance to the most distant \\ion{H}{ii} region in our sample). ",
        "introduction": "A number of studies have  indicated that the electron temperature \\te of  \\ion{H}{ii}  region  increases  with  increasing  Galacto-centric distance (e.g. \\cite{deh00} and  references therein). This effect is attributed  to  a  decrease  in   heavy  elements  abundances  with Galacto-centric  distance.  A  low   metal  abundance  leads  to  less effective  cooling  and  consequently  higher  electron  temperature. These  studies  are  based  either  on estimates  of  \\te  from  radio recombination lines  (RRLs) (which in  turn depend on  corrections for departures  from   local  thermodynamic  equilibrium   (LTE)  and  for collisional  broadening  effects),  or   estimates  based  on   line strengths    of   the   forbidden    line   transitions    of   Oxygen [\\ion{O}{iii}]$\\lambda\\lambda$4363, 5007 (which are strongly dependent on  temperature  variations,  if   any,  over  the  observed  volume). Further,  most   of  these  studies  are  based   on  observations  of \\ion{H}{ii} regions with Galacto-centric distances $R_G \\le 15$~kpc with  very few measurements  of \\te  beyond 15~kpc.  Consequently most determinations  of  metalicities  of  the  outer galaxy  \\ion{H}{ii} regions are based on values of  \\te taken from an extrapolation of the observed   gradient  in   temperature  up   to  about   15~kpc  (e.g., \\cite{deh00}).   Since the  O/H ratio  (a commonly  used  indicator of metal  abundance) depends sensitively  on \\te,  metalicities of the outer  galaxy \\ion{H}{ii} regions  are poorly  constrained. In  view of this, it  is important  to get independent  estimates of  the electron temperatures of \\ion{H}{ii} regions in the outer galaxy.  An independent  measurement of \\te can be  obtained from radio continuum  observations. The ionized  material in  \\ion{H}{ii} regions emits radio continuum through free-free emission.  At sufficiently low radio frequencies where the nebula is optically thick ($\\tau>>1$), the emergent  radiation  is a  black  body  spectrum,  and therefore,  the observed brightness  temperature is equal to  the electron temperature \\te. On the other hand,  at sufficiently high radio frequencies, where the optical depth $\\tau$ of  thermal electrons is low ($\\tau<<1$), the observed  brightness is proportional  to the  emission measure  of the nebula.  Most of the available  radio maps for \\ion{H}{ii} regions are at  high  radio frequencies  (i.e.  above  1.4~GHz,  e.g., Fich  1993, \\cite{balser95}). These maps show  that \\ion{H}{ii} regions often have a bright core with several  knots surrounded by an extended envelop of diffuse  emission.   These   core--envelope  structures  of  \\ion{H}{ii} regions  imply  that accurate  measurement  of  \\te  from low  radio frequency observations requires  high angular resolution, since, often only bright compact cores will  be optically thick at frequencies of a few hundred MHz. This study  presents an analysis of the low-frequency GMRT  observations  of  three  Galactic  diffuse  \\ion{H}{ii}  regions spanning Galacto-centric distances up to 18~kpc.  The  GMRT  is an  ideal  telescope  for  these observations  since  it operates at  several low radio  frequency bands, viz., 150,  232, 327, 610, and 1420  MHz and also it has a  hybrid configuration which makes it  sensitive  to  both  diffuse  emission (on  scales  up  to  $\\sim$ 45\\arcmin~ at 232, 30\\arcmin~ at 327, and 17\\arcmin~ at 610~MHz) while also having  the resolution ($\\sim$  15\\arcsec~ at 232,  10\\arcsec ~at 327, and 6\\arcsec ~at 610~MHz) to resolve the compact cores. ",
        "conclusions": "Three outer  galaxy \\ion{H}{ii} regions,  S~201, S~206 and  S~209 have been imaged at meter wavelengths  using the GMRT.  The images of these \\ion{H}{ii} regions have been obtained  at a resolution of less than a pc. This is the highest resolution achieved for any \\ion{H}{ii} region at such low radio frequencies.   All three \\ion{H}{ii} regions show structures down to our resolution  limit.  The high resolution images  near 610 MHz of these \\ion{H}{ii}  regions show a  good correspondence with  the radio continuum images  at $cm$ wavelengths.  The low  resolution radio images show that  these \\ion{H}{ii} regions  are surrounded by  large diffuse envelopes.  The  high resolution radio  images have allowed us  to get estimates of \\te of these \\ion{H}{ii} regions. From these measurements we find that : \\\\  \\noindent (1) the estimates of \\te are in general consistent with that obtained from  RRLs and [\\ion{O}{iii}]$\\lambda\\lambda$4363,  5007 line measurements, and \\\\ \\noindent (2)  the measured temperatures are consistent  with a linear increase  of  \\te  with   Galacto-centric  distance  until  $R_G  \\sim 18$~kpc."
    },
    "0209/astro-ph0209419_arXiv.txt": {
        "abstract": "The DEIMOS spectrograph has now been installed on the Keck-II telescope and commissioning is nearly complete.  The DEEP2 Redshift Survey, which will take approximately 120 nights at the Keck Observatory over a three year period and has been designed to utilize the power of DEIMOS, began in the summer of 2002.  The multiplexing power and high efficiency of DEIMOS enables us to target 1000 faint galaxies per clear night. Our goal is to gather high-quality spectra of $\\approx 60,000$ galaxies with $z>0.75$ in order to study the properties and large scale clustering of galaxies at $z \\approx 1$.  The survey will be executed at high spectral resolution, $R=\\lambda/\\Delta \\lambda \\approx 5000$, allowing us to work between the bright OH sky emission lines and to infer linewidths for many of the target galaxies (for several thousand objects, we will obtain rotation curves as well).  The linewidth data will facilitate the execution of the classical redshift-volume cosmological test, which can provide a precision measurement of the equation of state of the Universe. This talk reviews the project, summarizes our science goals and presents some early DEIMOS data. ",
        "introduction": "Our theoretical understanding of large scale structure and galaxy formation is well advanced, yet many crucial questions remain. Studies of structure depend on objects that can be seen, namely galaxies, whereas the raw medium from which galaxies formed is a mixture of both dark and baryonic matter.  It has become clear that the formation of visible galaxies in the universe is highly uneven and that galaxies are a ``biased'' tracer of the underlying mass.  Thus, to study structure in the universe at high redshift, we must be able to predict how the universe converted matter into luminous objects, requiring a thorough understanding of the physics of galaxy formation and evolution; but those processes depend on the underlying cosmological parameters and structure which we would also like to study!  Untangling galaxy evolution from cosmological evolution is extremely difficult with studies restricted to the local universe, which provide us with only a snapshot of the end result of galaxy formation, or those which include only a small number of very distant objects. Our ability to draw conclusions from galaxies in the Hubble Deep Field, for instance, is limited not only by the relatively small number of objects in the field that are bright enough to be studied from the ground, but also by the intrinsic spatial correlations between the galaxies, which causes fluctuations of measurements performed over a small volume to be much greater than simple Poisson statistics would suggest.  However, the combination of a statistically robust, large-volume, high-redshift sample with studies of the present-day universe should allow us to untangle the properties of galaxies from studies of large-scale structure, and simultaneously provide great amounts of information on galaxy formation and evolution.  The DEEP2 (DEEP Extragalactic Evolutionary Probe 2) Redshift Survey has been designed to produce such a dataset: one sufficiently rich both to refine our knowledge of fundamental cosmology and to challenge future galaxy formation models.  The DEEP2 collaboration plans to obtain spectra of $\\sim 60,000$ galaxies at high redshift using DEIMOS, a new multi-object spectrograph recently commisioned on the Keck Telescope. Details of DEIMOS are presented in this conference by Faber \\etal \\cite{faber02}. DEEP2 will provide a sample comparable in quality and volume to local surveys such as the LCRS, and thereby constrain the evolution of the properties of galaxies and of large-scale structure. The full program is expected to occupy 120 nights of Keck time, spread over a three year period.  By the end of 2002, 17 science nights will have been allocated for this program. ",
        "conclusions": "The flood of data now coming from the SDSS and 2DF projects detailing the local Universe is beginning to be complemented by data from the VLT/VIRMOS project and by the DEEP2 Redshift Survey providing detailed information on the Universe at $z \\approx 1$, thus continuing the revolution in precision cosmology and large-scale structure.  These results will rapidly expand in the coming months and will keep us all extremely busy. We intend to share our results with the public and to put our spectra online in a timely manner.  Further details on the survey can be found at the web site http://deep.berkeley.edu/."
    },
    "0209/astro-ph0209305_arXiv.txt": {
        "abstract": "We have developed a numerical code with which we study the effects of 2D perturbations on stellar structure. We present new numerical and analytical results on the heating of a main-sequence star in a binary system by its companion. ",
        "introduction": "Numerical study of stellar structure and evolution is now a well established subject with a long history. Today there is a plethora of 1D stellar evolution codes, from which astrophysicists can choose. All these models rely on the assumption that the star is spherically symmetric.  This assumption has served well for many years. However it is insufficient in cases that require stars to be non-symmetric. Stellar rotation, accretion and binary systems all add asymmetry to a spherical star. The required computational power is now available to look at these perturbations.  We have been working to develop a code that calculates 2D stellar structures in a non-computationally demanding way. So far we have studied perturbations to structure only but we plan to move into 2D stellar evolution. The two areas we have studied are rotation of solar-like stars and heating of a star by a companion star. The results are presented in this paper. ",
        "conclusions": ""
    },
    "0209/astro-ph0209596_arXiv.txt": {
        "abstract": "% As a special contribution to the proceedings of the \\sax workshop dedicated to blazar astrophysics we present a catalog of 157 X-ray spectra and the broad-band Spectral Energy Distribution (SED) of 84 blazars observed by \\sax during its first five years of operations. The SEDs have been built by combining \\sax LECS, MECS and PDS data with (mostly) non-simultaneous multi-frequency photometric data, obtained from NED and from other large databases, including the GSC2 and the 2MASS surveys. All \\sax data have been taken from the public archive and have been analysed in a uniform way. For each source we present a $\\nu f(\\nu)~vs~\\nu$ plot, and for every \\sax observation we give the best fit parameters of the spectral model that best describes the data. The energy where the maximum of the synchrotron power is emitted spans at least six orders of magnitudes ranging from $\\approx 0.1~$eV to over $100~$keV. A wide variety of X-ray spectral slopes have been seen depending on whether the synchrotron or inverse Compton component, or both, are present in the X-ray band. The wide energy bandpass of \\sax allowed us to detect, and measure with good accuracy, continuous spectral curvature in many objects whose synchrotron radiation extends to the X-ray band. This convex curvature, which is described by a logarithmic parabola law better than other models, may be the spectral signature of a particle acceleration process that becomes less and less efficient as the particles energy increases. Finally some brief considerations about other statistical properties of the sample are presented. ",
        "introduction": "Blazars emission is known to be dominated by strong and highly variable non-thermal radiation across the entire electromagnetic spectrum. Multi-frequency ground based observations, combined with data from high energy astronomy satellites, have often been used to derive the broad-band Spectral Energy Distribution (SED) of blazars, that is the source intensity as a function of energy, usually represented in the $\\nu f(\\nu)~vs~\\nu$ or $\\nu L(\\nu)~vs~\\nu$ space. These measurements are consistent with the widely accepted scenario where blazar emission is due to synchrotron radiation whose power increases with energy up to a peak value above which it drops sharply. At higher energies the spectrum is dominated by inverse Compton emission which also smoothly raises until it reaches a second luminosity peak. The often extreme observational characteristics of blazars are thought to be the result of the emission from a relativistic jet seen at a very small angle with respect to the line of sight (e.g. Urry \\& Padovani 1995), an interpretation first proposed by Blandford \\& Rees (1978). According to this scenario the position and the relative power of the synchrotron and inverse Compton peaks directly depend on important physical parameters such as the intensity of the magnetic field, the maximum energy at which electrons can be accelerated, and the relativistic motion and orientation of the emitting plasma. The synchrotron peak is located at energies ranging from less than $\\approx 0.1~$eV (or $ \\nu \\approx 10^{13}~$Hz) to well over $ 10~$keV (or $\\nu \\approx 10^{18}~$Hz) or even 100 keV in flaring states, demonstrating the existence of a wide variety of physical and geometric conditions in blazars. For these reasons the Spectral Energy Distribution of blazars has been and still is the subject of intense research activity. Figure 1 shows the expected emission from Synchrotron Self Compton models (SSC) tracing a hypothetical sequence of blazar SEDs that ranges from LBL sources where the synchrotron peak frequency (\\nupeak ) occurs at low energies to HBL objects where \\nupeak reaches the X-ray band, and up to the extremely large \\nupeak energies of the, possibly existing but still unseen, Ultra High energy peaked BL Lacs (UHBLs). As shown in Figure 1, within the broad-band energy spectrum of blazars the X-ray region is particularly important since at these energies a variety of different spectral components can be (and have been) seen. These include the flat and rising Compton component, the transition between the two regimes, and the high energy end of the synchrotron spectrum which is produced by very, sometimes extremely, energetic electrons. These crucial observations, in combination with other multi-frequency data allow the determination of the overall spectral shape and therefore the estimation of important physical parameters.  With its very wide X-ray band pass, good sensitivity and spectral capabilities {\\it Beppo\\-}SAX has provided a very important opportunity to study blazars astrophysics, especially when simultaneous multi-frequency observations could be arranged.  As a special contribution to the proceedings of the \\sax workshop dedicated to blazar Astrophysics we present the catalog of X-ray spectral fits and broad-band Spectral Energy Distribution of all the blazars observed with \\sax whose data are currently public.  \\setcounter{figure}{0} \\begin{figure}[!ht] \\vspace*{-3.0cm} \\centering \\epsfxsize=10.0cm\\epsfbox{allbl_sed.ps} \\caption[h]{The Spectral Energy Distribution of BL Lacs is shown as a sequence of Synchrotron Self Compton spectra peaking at different energies. Objects whose synchrotron component peaks at low energy are called LBLs, the maximum of their synchrotron power output occurs in the IR-Optical band, while High energy peaked BL Lacs (HBLs) peak in the UV or X-ray band. It is not known how far the sequence of \\nupeak goes on. If it continues to very high energies it could be that in some very extreme objects (UHBLs) the synchrotron component might even reach the gamma-ray band. Note that for the same peak luminosity, the radio power decreases by orders of magnitudes in going form LBLs to HBLs and possibly to UHBLs.} \\label{fig1} \\end{figure} ",
        "conclusions": "We have presented the X-ray spectrum and the Spectral Energy Distribution of a large sample of blazars observed by \\sax with the aim of providing a single homogeneous reference for this type of \\sax data. The collection of the results presented here together with all the data is also available as part of the \\sax archive at the ASI Science Data Center (ASDC) at the following address \\par \\begin{center} http://www.asdc.asi.it/blazars/ \\end{center}  Given the heterogeneous nature of the sample we have not attempted to perform any deep statistical studies. In the following we summarize our work and give some remarks about possible interpretations. More detailed statistical studies or deeper interpretations are reported elsewhere or will be the subject of future publications.   \\setcounter{figure}{3} \\begin{figure}[!ht] \\vspace*{-2.0cm} \\centering \\epsfysize=6.8cm\\epsfbox{pg1418_fit.ps} \\hspace{2.cm}\\epsfysize=6.8cm\\epsfbox{on231_fit.ps} \\caption[ht]{Spectral energy distributions of a typical LBL object (PG~1418+546, \\nupeak $ \\approx 0.4~$eV$~\\approx 8\\times 10^{13} $Hz) for which the X-ray emission is dominated by the flat inverse Compton radiation and of an Intermediate BL Lac (ON 231, \\nupeak $ \\approx 1~$eV$~\\approx 2\\times 10^{14} $Hz) where the simultaneous optical and \\sax observations (Tagliaferri et al. 2000) clearly show that the transition between the synchrotron and inverse Compton emission occurs in the soft X-ray band.} \\label{fig4} \\end{figure}  \\setcounter{figure}{4} \\begin{figure}[!ht] \\vspace*{-2.5cm} \\centering \\epsfysize=6.8cm\\epsfbox{pks2155_fit.ps} \\hspace{2.cm}\\epsfysize=6.8cm\\epsfbox{1h1430_fit.ps} \\caption[ht]{Spectral energy distributions of HBLs where the X-ray emission is completely dominated by synchrotron radiation. In the case of PKS 2155$-$304 \\nupeak is at $\\approx$ 50 eV $\\approx \\times10^{16}$ Hz while for the extreme HBL 1H 1430+423 \\nupeak is above 10 keV.} \\label{fig4} \\end{figure}  \\setcounter{figure}{5} \\begin{figure}[!ht] \\vspace*{-1.5cm} \\centering \\epsfysize=12.0cm\\epsfbox{nupeaks.ps} \\caption[h]{The distribution of the synchrotron peak frequencies in the FSRQ and in the BL Lac subsamples. While the \\nupeak values for FSRQs strongly cluster around $10^{14}$ Hz, a much wider distribution is present in the BL Lacs subsample. In the latter case the distribution is strongly biased towards high \\nupeak values by the \\sax time allocation process which favoured X-ray bright (and therefore high \\nupeak) objects that could be detected by all NFI instruments.} \\label{fig5} \\end{figure}   \\newpage  \\bigskip  \\par The SED of the 84 blazars considered in this work confirms with large statistics the widely accepted scenario where blazar emission is smooth across several decades of energy and is characterized (in a  $\\nu f(\\nu)~vs~\\nu$ representation) by two broad peaks which are usually interpreted as being due to synchrotron emission followed by inverse Compton radiation.  \\bigskip  \\par A wide range of X-ray spectral indices has been observed, ranging from very flat values in Compton dominated sources like PG~1418+546 to very steep spectral slopes in objects where the tail of the synchrotron emission just reaches the X-ray band. A sharp transition from a very steep soft X-ray component to a much flatter hard X-ray spectrum, marking the transition between the synchrotron and Compton emission, has been clearly detected in a number of objects. Examples are given in Figures 4 and 5 where, for four representative objects, we plot the observed SED together with the expected distribution from a SSC model peaking at appropriate \\nupeak values.  \\par When viewed only from the narrow soft X-ray band, as was done in the past, these SEDs clearly show that the local X-ray spectral index must be correlated to the X-ray to radio flux ratio ($f_x/f_r$) as was first found by Padovani \\& Giommi (1996) and by Lamer, Brunner \\& Staubert (1996) in large samples of BL Lacs observed with ROSAT.  \\bigskip  \\par The position of the synchrotron peak, estimated comparing the SEDs to SSC models such as those shown in Figures 4 and 5, spans at least six orders of magnitudes ranging from $\\approx 0.1~$eV  in e.g. PKS~0048$-$097 or S5 2116+81 to 10--100 keV in some extreme HBL BL Lacs like Mkn 501 and 1H 1430+423.  \\par Very strong intensity and spectral variability can occur near the synchrotron (and inverse Compton) peak. The position of this peak can move to higher energy by up to two orders of magnitude (or perhaps more) during flares.  It is not clear what is the maximum \\nupeak that can be reached and whether Ultra High energy synchrotron peaked BL Lacs (UHBL) exist. A few potential UHBL sources may be present in the Sedentary survey (Giommi, Menna \\& Padovani 1999, Perri et al. 2002), which by definition only includes extreme HBL objects, especially those few that are located within the error circle of unidentified EGRET sources. If these candidates turn out to be the real counterpart of the EGRET gamma-ray sources their \\nupeak would be so high that their synchrotron radiation would reach the gamma-ray band. One such object, 1RXS~J123511.1$-$14033 (see Figure 2i), was observed by \\sax on three occasions but always with short integration times giving inconclusive results (Giommi et al. 2002, in preparation).  \\par The observed distributions of \\nupeak values (rest frame), obtained by fitting SSC models to the multi-frequency data shown in Figure 2a-2o and 3a-3g, are plotted in Figure 6 for the FSRQ (top panel) and the BL Lac (bottom panel) subsamples. The two distributions are certainly affected by selection effects, including that induced by the \\sax time allocation process which, by necessity, favoured high \\nupeak/X-ray strong sources which could be detected by the high energy instruments. Although this bias is clearly present in the BL Lac subsample where a large fraction of the sources are X-ray bright HBL objects, in the case of FSRQs most of the objects have low \\nupeak . This is because FSRQs are in general more luminous than BL Lacs and especially because FSRQs with \\nupeak $ > 10^{16}$ Hz are very rare. To date the only FSRQ (RGB J1629+4008 = 1ES 1627+402, see Figure 3d) whose synchrotron emission reaches the X-ray band was found by Padovani et al. 2002b.  \\bigskip  \\par A logarithmic parabola model, which can describe the spectral curvature of blazars in a very wide energy band with only three parameters (see Landau et al. 1986), fits better than other models (e.g. broken power law) the spectrum of HBL objects whose X-ray emission is still due to synchrotron radiation. The average amount of spectral curvature, as measured by the $\\beta$ parameter in the log parabola model of paragraph 3.1 is $-$0.38 +/$-$ 0.1, a value somewhat steeper (possibly because of the energy dependant synchrotron cooling), but not too different, than the amount of curvature  found by Landau et al. 1986 ($-$0.22 to $-$0.09 ) in a sample of BL Lacs whose synchrotron power peaks at infra-red, optical frequencies. This similarity points to an intrinsically similar curvature in the spectrum of the emitting particles. The smoothly changing slope could be the spectral signature of a statistical acceleration mechanism where the acceleration process becomes less and less efficient as the particle's energy increases (Massaro 2002). In this scenario the widely different synchrotron \\nupeak energies in LBL and HBL objects would be the result of the inefficiency in the acceleration process that sets off at different energies.  \\bigskip"
    },
    "0209/astro-ph0209075_arXiv.txt": {
        "abstract": "We investigate the EUV and X-ray flare rate distribution in radiated energy of the late-type active star AD Leo. Occurrence rates of {\\it solar} flares have previously been  found to be distributed in energy according to a power law, $dN/dE \\propto E^{-\\alpha}$, with a power-law index $\\alpha$ in the range 1.5$-$2.6. If $\\alpha \\ge 2$, then an extrapolation of the flare distribution to low flare energies may be sufficient to heat the complete observable X-ray/EUV corona.  We have obtained long  observations of AD Leo with the {\\it EUVE} and {\\it BeppoSAX} satellites. Numerous flares have been detected, ranging over almost two orders of magnitude in their radiated energy. We compare the observed light curves with light curves synthesized from model flares that are distributed in energy according to a power law with selectable index $\\alpha$. Two methods are applied, the first comparing flux distributions of the binned data, and the second using the distributions of photon arrival time differences in the unbinned data (for {\\it EUVE}).  Subsets of the light curves are tested individually, and the quiescent flux has optionally been treated as a superposition of flares from the same flare distribution. We find acceptable $\\alpha$ values between 2.0$-$2.5 for the {\\it EUVE} DS and the {\\it BeppoSAX} LECS data. Some variation is found depending on whether or not a strong and long-lasting flare occurring in the {\\it EUVE} data is included. The {\\it BeppoSAX} MECS data indicate a somewhat shallower energy distribution (smaller $\\alpha$) than the simultaneously observed LECS data, which is attributed to the harder range of sensitivity of the MECS detector and the increasing peak temperatures of flares with increasing total (radiative) energy.  The results suggest that flares can play an important role in the energy release of this active corona. We discuss caveats related to time variability, total energy, and multiple power-law distributions. Studying the limiting case of a corona that is entirely heated by a population of flares, we derive an expression for the time-averaged coronal differential emission measure distribution (DEM) that can be used as a diagnostic for the flare energy distribution. The shape of the analytical DEM agrees with previously published DEMs from observations of active stars. ",
        "introduction": "The physics of coronal heating remains one of the most fundamental problems in stellar (and solar) astrophysics. The subject has been reviewed extensively from the point of view of theoretical concepts \\citep{ionson85, narain90, zirker93}, observational solar physics  (e.g., \\citealt{benz94}), and stellar physics (e.g., \\citealt{haisch96}), where the cited work stands exemplary for a large body of literature available. It is somewhat surprising that the nature of the ``coronal heating mechanism(s)'' still eludes agreement given high-resolution imaging of solar coronal structures or large statistical samples of stellar coronal X-ray observations. For example, there is no unequivocal agreement on whether all, or any, of the X-ray coronal energy detected from {\\it certain} classes of  stars is magnetic in origin.  Coronal heating is of particular interest to stellar astrophysics since it  relates directly to our understanding of coronal structure and dynamics, information that is usually obtained by means of indirect modeling. Apart from heating models involving acoustic heating, for example on F-type stars (see \\citealt{mullan94}; although the resulting X-ray flux would be much smaller than observed - see \\citealt{stepien89}), the currently advocated mechanisms are of two types: (i) Steady heating mechanisms, e.g., by steady electric current dissipation or MHD waves, and (ii)  heating by explosive energy release, e.g., coronal flares. The latter are attractive heating agents since  flares  do heat plasma efficiently, although only episodically  since the radiative and conductive losses rapidly cool plasma to pre-flare levels, typically within minutes to hours.  The flare heating hypothesis has gained momentum in particular from solar, but also from stellar observations during recent years. If the quasi-steady (``quiescent'') coronal emission is to be explained by flare contributions, flares must act as {\\it stochastic} heating agents.  \\citet{parker88} proposed that shuffling of magnetic field footpoints in the photosphere by the convective motions leads to tangled  magnetic field lines in the corona and thus to current sheets. With increasing winding of magnetic fields, the necessary energy may be transported into the coronal magnetic field where it is released by sudden relaxation involving reconnection.  Parker estimates that energy dissipation occurs in packets involving $10^{24} - 10^{25}$~ergs (``nanoflares'').  The flare-heating hypothesis resolves to the basic question of whether or not the {\\it statistical ensemble} of flares (in time and energy)  suffices to heat the apparently nonflaring coronae. ",
        "conclusions": "We have investigated the role of statistical flares in coronal heating of magnetically active stars. Long observations of AD Leo were obtained in order to maximize flare statistics. Flares have been suspected to play an important role in coronal energy release and subsequent impulsive heating of chromospheric material to high temperatures.  Chromospheric evaporation induced by chromospheric overpressure lifts the hot plasma into the corona where it fills closed magnetic loops. Since (solar) flares are always related not only to an increase in emission measure but to a significant increase in the average plasma temperature, they are natural candidates  to heat perhaps all of the detected coronal plasma. Recent progress in solar physics \\citep{krucker98, aschwanden00, parnell00} has added new momentum to this hypothesis.  Active (but quiescent) stellar coronae exhibit a number of features unknown to the non-flaring Sun but suspiciously reminiscent of solar (or stellar) flares: i) Very high temperatures up to 2$-$3~keV, similar to temperatures of large solar flares; ii) accompanying, strong non-thermal gyrosynchrotron radio emission attributed to relativistic electrons accelerated in the initial phase of the flare energy release (\\citealt{gudel94} and references therein) ; iii) high densities ($\\ga 10^{10}$ cm$^{-3}$) reminiscent of (solar) flare densities \\citep{gudel01a,gudel01b}; iv) \"anomalous\" elemental abundances  tentatively ascribed to the action of flares, perhaps analogs to solar Ne- and S-rich flares \\citep{brinkman01, drake01}; v) and finally, the presence of a large number of strong flares, where the rate of detected flares correlates with the quiescent emission  level \\citep{audard00}.  We have studied the distribution of EUV and X-ray flares in energy, seeking power laws of the form $dN/dE =kE^{-\\alpha}$ where $k$  is the normalization of the distribution and  $\\alpha$ determines the steepness of the distribution.  We have applied two methods, one based on the count rate distribution of binned data, and the second related to the distribution of arrival-time differences of the original   photon lists (only for {\\it EUVE}). Despite  the fundamentally different approaches,  the results of both methods are in excellent agreement for the {\\it EUVE} data  and indicate $\\alpha = (2.1-2.3) \\pm 0.1$. Simultaneous X-ray observations obtained with the {\\it BeppoSAX} LECS were treated with the first method. The results again overlap with the {\\it EUVE} results, namely $\\alpha = 2.4\\pm 0.2$. Only the MECS data show somewhat shallower distributions, with $\\alpha = (2.0-2.2)\\pm 0.2$ which can be explained by the harder sensitivity range of the MECS detector, and detection bias in terms of flare temperatures. Our results are compatible with the findings of \\citet{audard00} who applied a flare identification algorithm to explicitly record flares and to measure their energies. They are further supported by the findings of \\citet{kashyap02} who study further active stars with the {\\it EUVE} DS. At first sight, our $\\alpha$ values  support a model in which the complete coronae are heated by a statistical distribution of flares, involving flares with energies down to a few times $10^{29}$ erg (of radiated energy). Also, a model EM distribution based on the superposition of flares of different peak temperatures is compatible with the observed EM distribution. However, before we  can conclude that flares play an important role in the coronal heating process, we should keep in mind the following caveats that stellar observations of the present quality are invariably subject to:   (A) All EUV and X-ray  observations refer to the radiated energy  from the hot plasma. There is considerable  ignorance of other energy partitions that also contribute to the  flare energy budget  (\\citealt{wu86}):  Kinetic energy of the upstreaming plasma; potential energy of the lifted plasma; energy in waves; energy in accelerated particles;  and energy released at longer wavelengths. To interpret our results physically, we  adopt the following working hypothesis: i) The energy initially released in energetic particles is largely thermalized in the chromospheric evaporation process. ii) The remaining energy (kinetic and potential) is eventually thermalized  (e.g., when material drops back to the chromosphere). iii) All thermal energy is eventually radiated away during the cooling processes. We emphasize that this applies also to all energy that is conducted from the coronal loops downwards. This energy is radiated by the chromosphere. iv) The fraction of the radiative energy released in the X-ray range is similar for all flares.  While points i)--iii) are supported by observations \\citep{dennis85} and by numeric simulations (e.g., \\citealt{nagai84, antonucci87}), point iv) is relatively difficult to assess. Clearly, a considerable part of the energy  is lost at UV wavelengths not accessible to our observations. The  tendency of the flare temperature to increase with overall flare energy \\citep{feldman95,aschwanden99} would suggest that smaller flares lose a larger fraction of their energy outside the EUV/X-ray regime. This is, however, of little relevance for us: All flares considered here are quite large, with probable peak temperatures (according to the Feldman et al. relation) exceeding 20 MK; even the quiescent emission shows its peak EM between 5--10 MK (Table~\\ref{saxfit}). \\citet{hudson91} reports that approximately 2/3 of the total radiant energy of a solar flare are emitted in soft X-rays.  The total, long-term average energy loss in optical U band flares is linearly correlated with the average X-ray losses  in active stars \\citep{doyle85}. If we missed a population of very small flares with very low temperatures (e.g., comparable to microflares in the Sun), then they would simply add to the flare distribution on the low-energy  side, i.e., our distributions would become steeper still, making small flares even more crucial for the total  energy release. Further, the measured average thermal energy input (``heating rate'') during a solar flare scales linearly with the radiative loss rate in X-rays \\citep{aschwanden00}.  Finally, from the phenomenological point of view adopted in the present study, the relation between losses in the X-rays/EUV  and those at longer wavelengths can be ignored  altogether if we keep with our goal of modeling the {\\it observed coronal} emission. The latter is clearly dominated by X-ray/EUV radiation. If we successfully  explain the total X-ray/EUV emission by the radiation  of a statistical ensemble of flares, then this simply implies that there is no significant additional {\\it coronal} component. Although there may be additional energy release at lower temperatures (i.e., at chromospheric  levels), this becomes irrelevant for the question of {\\it coronal} heating.  (B) The power-law distribution of flares may change spatially on the star. Stellar observations unavoidably treat the corona as an average structure. The recent solar results with $\\alpha > 2$ were obtained in regions of the quiet Sun while most larger flares occur in active regions. It may, however,  be interesting to mention  that an observation of a {\\it single} solar X-ray bright point resulted in a power-law index similar to those obtained from the whole Sun \\citep{shimojo99}.  (C) The power-law distributions may also vary in time. \\citet{bai93} and \\citet{bromund95} find a 154~d periodicity in which $\\alpha$ changes by $\\sim 0.2-0.4$ in solar data. There may also be a dependence on the overall magnetic activity level that varies with the (cyclic or irregular) ``magnetic activity cycle''. This latter conjecture is, however, not supported by the solar studies of \\citet{feldman97} and \\citet{lu91}. We note that we found different values for $\\alpha$ depending on whether or not the early part of the DS observation (containing a large flare) was included.  (D) Although we have used a rather long observing time series (27.3 days of coverage with {\\it EUVE}, referring to segment I--IV), some chance coincidence, like the very large flare at the beginning of the observation, may introduce considerable systematic bias. We have investigated the role of this large flare on the result for $\\alpha$ and found indeed that its selective inclusion/exclusion  can shift the optimum value  by $\\Delta\\alpha \\approx 0.1$.  (E) There may be high-energy cut-offs (``roll-overs''; \\citealt{kucera97}) related to the maximum energy that can be liberated in stellar active regions. In a limited set of observations with a limited dynamic range (ratio of strongest to weakest detected flares, also depending on the noise level), the deficit of flares close to the high-energy cut-off (because of their small occurrence rate) can induce a steepening of a power law. A consequent shallower continuation of the distribution toward  flares below our detection limit would contribute less energy than estimated with our single power-law approach. The present data do not allow us to judge on the presence or absence of high-energy cut-offs. The  good representation  by single-power-law flare distributions does presently not argue for their presence. The continuation of the power law from detected flares to energies below the detection threshold has been explicitly assumed in our energy estimates, and this is no different from any previous (solar or stellar) study. For sufficiently small energies, the large number of small flares involved begin to overlap in time (the ``confusion limit'', already evident in our light curves). They can no longer be measured individually unless spatial resolution is available.  (F) Appreciable non-flare contributions to the EUV/X-ray variability are possible  (e.g., evolution of non-flaring active regions, newly emerged magnetic regions, rotational modulation of active regions). They would  normally add  to the low-level variability and may thus tend to steepen the count rate distributions.   Despite these caveats, some of which will be difficult or impossible to avoid in future observations, we presently see no compelling argument  against our basic finding, namely, that flares statistically contribute an important part to the overall coronal radiative losses, and that they are therefore good candidates for the coronal heating process per se in magnetically active stars.  Our values for $\\alpha$ are very similar to those measured for microflares in the Sun \\citep{krucker98, parnell00} despite the 6 orders of magnitude larger energies involved. This factor in energy may  partly reflect the level of magnetic activity. If so, then the role played by microflares in the Sun is played by the much larger flares relevant here in active stellar coronae.  We find independent support for flare heating of active stellar coronae in their coronal emission measure distribution. By statistically co-adding flaring emission measures by weighting them with the dwell time at a given temperature, we derived an analytical expression for the differential emission measure distribution. The DEM is characterized by a steeply rising  low-temperature part and a falling high-temperature part. The slopes and the turnover temperature are in principle determined by the flare energy power-law index $\\alpha$, the low-energy break in the distribution, and the flare heating parameter $\\zeta$ during the flare decay. Previously published DEMs of active stars (e.g., \\citealt{laming96, drake01a}) and the DEM derived here show characteristic shapes that are compatible with our expression but are not supported by quasi-static loop models. We suggest that the coronal DEMs directly reflect the operation of heating and cooling mechanisms during stochastic flares (\\citealt{gudel97, gudeleal97}).  We conclude this presentation by emphasizing two observational circumstances: i) It may be pivotal in which energy range relevant for coronal losses the observations are made. Observations that exclusively record the harder part of soft X-rays selectively favor detections of large flares and suppress the relevance of low-energy flares (due to the Feldman et al. relation; see also discussion in \\citealt{porter95}). As is to be expected from the flare-heating hypothesis, the quiescent emission is comparatively soft and is therefore also underrepresented in hard observations. We  have marginally found this effect in our MECS observations. One may wonder whether a similar effect exists for non-thermal hard X-rays often used for solar flare energy statistics. If they are generated overproportionally  in larger flares (as suggested by the ``Big Flare Syndrome'', \\citealt{kahler82}, but also by recent observations finding that microflares are radio-poor, i.e., relatively weak in the production of accelerated particles, \\citealt{krucker00}) then the statistical distributions may be biased toward too low $\\alpha$. We have selected  the energy range in which the dominant fraction of the coronal flare energy is radiated. Also, the efficiency (ratio between observed count rate and incident flux) of the DS and the LECS detectors shows only a  weak temperature dependence, i.e., the observations are equally sensitive to plasma over a wide range of relevant temperatures. ii) The power-law distribution may depend on the flare energy range considered. There are indications in solar observations to this effect, and we can safely state that the power laws found here cannot be extrapolated to arbitrary energies: There must be a low-energy break (possibly changing to a shallower distribution) in order to confine the total radiated power, and there must be a high-energy limit, corresponding to the largest physically possible flares."
    },
    "0209/astro-ph0209243_arXiv.txt": {
        "abstract": "A brief review concludes that there is now good overall agreement between theoretical estimates of the energy associated with the production of the observed metal content of the Universe and the observed extragalactic background light. In addition the overall form of the star-formation history over $0 < z < 5$ is reasonably well constrained.  The study of emission line gas in galaxies as a function of redshift provides a complementary view of chemical evolution to that obtained from studies of absorption line systems in quasar spectra. Emission line gas is more relevant for some questions, including the confrontation with models for the chemical evolution of our own and other galaxies, and Origins-related questions about the formation of stars and planets. Using relatively crude diagnostic parameters such as Pagel's $R_{23}$, the increase of the metallicity of star-forming gas with cosmic epoch can be tracked.  Observations of a large sample of CFRS galaxies at $z \\sim 0.8$ (a look-back time of 0.5$\\tau_0$) show the appearance of a significant number of low metallicity systems amongst luminous L* galaxies. However, overall there is only a modest change in mean metallicity compared with the present, $\\Delta$ log $Z = -0.08 \\pm 0.06$. These data do not support a fading-dwarf scenario for the ``faint blue galaxies''. At $z \\sim 3$ it is likely all star-forming galaxies have sub-solar metallicities.  The overall increase in the metallicity of star-forming gas with cosmic epoch matches rather well the age-metallicity relation in the solar neighbourhood and some recent models for the evolution of the global metallicity of gas in the Universe. ",
        "introduction": "As is clear from the theme of this conference, the production of the heavy elements through stellar nucleosynthesis is central to the Astronomical Search for Origins.    Clearly, the metallicity of the Universe and of objects in it provides a fundamental metric reflecting the development of structure and complexity in the Universe on galactic scales. This metric is all the more important because it is relatively easily observable and ``long-lived'' in the sense that heavy atomic nuclei, once produced, are not readily destroyed.  Thus, the metallicity of material reflects its entire evolutionary path through cosmic time - with a potential for sophistication which is impressive (see Jim Truran's contribution to these proceedings). Furthermore, in regard to the Origins theme, it is these same heavy elements that play a key role in the formation of structure on planetary scales and which are fundamental to the existence of Life itself in the Universe.  For many years, the metallicities of gas at high redshifts have been studied through the analysis of absorption line systems seen in quasar spectra.  These are discussed by Wal Sargent in these proceedings. The lines of sight to quasars probe, almost by definition, random regions of the Universe - the only possible concern being whether lines of sight passing through dusty regions are under-represented because the quasar is then eliminated from the sample (Fall and Pei 1993, see Ellison et al 2001). On the other hand, only a single line of sight through a given system is generally available, making the interpretation of the metallicity measurement in the context of larger structures, such as galaxies, non-trivial and fundamentally statistical in nature.  As an example, the relationship between the high column density ``damped Lyman $\\alpha$'' (DLA) systems and galaxies is still by no means clear.  The study of the metallicities of material in known galaxies at high redshift is at a much earlier stage of development and is inevitably of lower sophistication. However, metallicity estimates, especially of the [O/H] abundance, using diagnostics that are based solely on strong emission lines, e.g. [OII] 3727, H$\\beta$, [OIII] 4959,5007, H$\\alpha$, [NII]6584, [SII]6717,6731, are now technically feasible over a wide range of redshift and give a new perspective on the chemical evolution of galaxies.  For some purposes, the estimates of the metallicity of star-forming gas at earlier epochs that are obtained from emission lines may actually be of more relevance than the more ``global view'' that is obtained from the absorption line studies. Applications include the confrontation of models for the chemical evolution of the Milky Way or other galaxies, the comparison with the observed age-metallicity relationship seen in galaxies at the present epoch, and the use of metallicity estimates to constrain the present-day descendents of high redshift galaxies. The emission line gas in star-forming regions is also the most relevant for the planetary and astro-biological aspects of the Origins theme since it should be representative of the material out of which the stars and planets are actually being made.  In this contribution, we first review the production of metals in the Universe in the context of its stellar content and the extragalactic background light (EBL).  We then describe and discuss measurements of star-forming gas at different cosmic epochs, focussing on new results that we have obtained from a large sample at $z \\sim 1$. Throughout the discussion we adopt the so-called ``concordance'' cosmology with H$_0 = 70$ kms$^{-1}$Mpc$^{-1}$, $\\Omega_0 = 0.3$ and $\\lambda_0 = 0.7$. ",
        "conclusions": ""
    },
    "0209/astro-ph0209257_arXiv.txt": {
        "abstract": "We propose a model to explain how a Gamma Rays Burst can take place days or years after a supernova explosion. Our model is based on the conversion of a pure hadronic star (neutron star) into a star made at least in part of deconfined quark matter. The conversion process can be delayed if the surface tension at the interface between hadronic and deconfined-quark-matter phases is taken into account. The nucleation time ({\\it i.e.} the time to form a critical-size drop of quark matter) can be extremely long if the mass of the star is small. Via mass accretion the nucleation time can be dramaticaly reduced and the star is finally converted into the stable configuration. A huge amount of energy, of the order of 10$^{52}$--10$^{53}$ erg, is released during the conversion process and can produce a powerful Gamma Ray Burst. The delay between the supernova explosion generating the metastable neutron star and the new collapse can explain the delay proposed in GRB990705 \\citep{Amati00} and in GRB011211 \\citep{Reeves02}. ",
        "introduction": "The discovery of a transient (13~s) absorption feature in the prompt emission of the $\\sim 40$~s Gamma Ray Burst (GRB) of July 5, 1999 (GRB990705) \\citep{Amati00} and the evidence of emission features in the afterglow of several GRBs \\citep{Piro99,Yoshida99,Piro00,Antonelli00,Reeves02} have stimulated the interpretation of these characteristics in the context of the fireball model of GRBs. \\citet{Amati00} attribute the transient absorption feature of GRB990705 (energy released $\\sim 10^{53}$ erg assuming isotropy) to a redshifted K edge of Iron contained in an environment not far from the GRB site ($\\sim 0.1$~pc) and crossed by the GRB emission. They estimate an Iron abundance typical of a supernova (SN) environment ($A_\\mathrm{Fe} \\sim 75$) and a time delay of about 10 years between the SN explosion and the GRB event. \\citet{Lazzati01} give a different interpretation of the absorption feature, in terms of a redshifted resonance scattering feature of H--like Iron (transition 1s--2p, $E_\\mathrm{rest} =6.927$~keV) in an inhomogeneous high--velocity outflow, but invoke a Iron rich environment as well, due to a preceding SN explosion, even if a shorter time delay ($\\sim 1$~yr) between SN and GRB is inferred. A SN explosion preceding  the GRB event is also inferred for explaining the properties of the emission features in the X--ray afterglow spectrum of GRB000214 \\citep{Antonelli00} and GRB991216 \\citep{Piro00}. In the latter case it cannot be excluded that the SN explosion occured days or weeks before the GRB \\citep{Rees00}. \\citet{Reeves02}, to explain the multiple emission features observed in the afterglow spectrum of GRB011211 (time duration of $\\sim 270$~s, isotropic gamma--ray energy of $5 \\times 10^{52}$~erg), invoke a SN explosion preceding the GRB event by $\\sim 4$~days (in the isotropic limit, a minimum of 10 hrs). Even if other interpretations for the afterglow emission lines are possible which do not involve a previous SN explosion (e.g., \\citep{Rees00,Meszaros01}), this explosion seems to be the most likely way to explain the transient absorption line observed from GRB990705 \\citep{Bottcher02}. In conclusion, the previous observations suggest that, at least for a certain number of GRBs, a SN explosion happened before the GRB, with a time interval between the two events ranging from a few hours to a few years. In this context, an attractive scenario is that described by the {\\it supranova} model \\citep{Vietri98} for GRBs. In this model, the GRB is the result of the collapse to a black hole (BH) of a supramassive fast rotating neutron star (NS), as it loses angular momentum. According to this model the NS is produced in the SN explosion preceding the GRB event. The initial barionic mass $M_B$ of the NS is assumed to be above the maximum baryonic mass for non-rotating configurations. However, as also noticed by \\citet{Bottcher02}, on the basis of realistic calculations of collapsing NS \\citep{fryer1998}, in these collapses too much baryonic material is ejected and thus the energy output is expected to be too small to produce GRBs. Even if the introduction of magnetic fields or beaming could overcome this limitation, in any case, the GRB duration from a NS collapse should be very short ($\\ll1$~s), much  shorter than that observed from GRB990705.  In this Letter, we propose an alternative model to explain the existence of GRBs associated with previous SN explosions. In this model, unlike the supranova model, the NS collapse to BH is replaced by the conversion from a metastable, purely hadronic star (neutron star) into a more compact star in which deconfined quark matter (QM) is present. This possibility has already been discussed in the literature \\citep{cd96,bd00,wang,ouyed}. The new and crucial idea we introduce here, is the metastability of the purely hadronic star due to the existence of a non-vanishing surface tension at the interface separating hadronic matter from quark matter. The mean-life time of the metastable NS can then be connected to the delay between the supernova explosion and the GRB. As we shall see, in our model we can easily obtain a burst lasting tens seconds, in agreement with the observations. The order of magnitude of the energy released is also the appropriate one. ",
        "conclusions": "We propose the following origin for at least some of the GRBs having a duration of tens of seconds. They can be associated with the transition from a metastable HS to a more compact HyS or a QS. The time delay between the supernova explosion originating the metastable HS and the GRB is regulated by the process of matter accretion on the HS. While most of the stellar objects obtained by a SN explosion will possibly have a mass larger than $M_{\\mathrm{cr}}$ and will therefore directly stabilize as HyS or QS at the moment of the SN explosion, in a few cases the mass of the protoneutron star will be low enough not to allow the immediate production of QM inside the star. Only when the star will acquire enough mass, the process of QM formation could take place. Due to the surface tension between the hadronic matter and the QM the star will become metastable. The later collapse into a stable HyS or QS will generate a powerfull GRB. It can be interesting to notice that, in order to have a not too small value for $M_{\\mathrm{cr}}$, a relatively large value for the bag constant $B$ has to be choosen, $B^{1/4}\\sim$ 170 MeV, which turns out to be the prefered value in many hadronic physics calculations (see e.g. \\citep{thomas}). In this situation the final state is an HyS and not a QS.  \\bigskip \\bigskip   It is a pleasure to thank Elena Pian and Luciano Rezzolla for very useful discussions."
    },
    "0209/astro-ph0209061_arXiv.txt": {
        "abstract": "It has been shown that there is a possible mass-period correlation for extrasolar planets from the current observational data and this correlation is, in fact, related to the absence of massive close-in planets, which are strongly influenced by the tidal interaction with the central star. We confirm that the model in P\\\"atzold \\& Rauer (2002) is a good approximation for the explanation of the absence of massive close-in planets. We thus further determine the minimum possible semimajor axis for these planets to be detected during their lifetime and also study their migration time scale at different semimajor axes by the calculations of tidal interaction. We conclude that the mass-period correlation at the time when these planets were just formed was less tight than it is now observed if these orbital migrations are taken into account. ",
        "introduction": "The number of discovered extrasolar planets is increasing quickly during recent years. According to the Extrasolar Planets Catalog maintained by Jean Schneider (http://cfa-www.harvard.edu/planets/catalog.html), in May 2002, there are about 77 extrasolar planets around 69 main sequence stars. These planets with mass range from 0.16  to 17 Jupiter masses ($M_J$) have semimajor axes from 0.04 AU to 4.5 AU and also a wide range of eccentricities. Interestingly, there is a planet moving on an extremely elongated orbit ($e=0.927$) around the solar-type star HD 80606 (Naef et al. 2001). These exciting discoveries provide great opportunities to understand the formation and evolution of planetary systems.  For example, Jiang \\& Ip (2001) showed that the interaction with disc is important to explain the original orbital elements during the planetary formation. Yeh \\& Jiang (2001) analytically showed that the scattered planets should in general move on an eccentric orbit and thus the orbital circularization must be important for scattered planets if they are now moving on nearly circular orbits (See Jiang \\& Yeh 2002a, Jiang \\& Yeh 2002b for the following up).    In addition to the dynamical studies, Tabachnik \\& Tremaine (2002) used the maximum likelihood method to estimate the mass and period distributions of extrasolar planets and found there is a mass-period correlation, but they attributed their finding to the observational selection effect.  However, Zucker \\& Mazeh (2002) claimed that this mass-period correlation cannot be completely explained by the observational selection effect. They did some Monte Carlo simulations and show the real dependency between the mass and period of extrasolar planets. This mass-period correlation gives the paucity of massive close-in planets. Since they are supposed to be the easiest to detect,  Zucker \\& Mazeh (2002) said this paucity was unlikely to be the result of any selection effect.  P\\\"atzold \\& Rauer (2002) have reported the possible explanation about the absence of massive close-in planets by tidal interaction. They defined ``critical mass'' to be the maximum mass that the planet can have and survive under the tidal interaction from the central star for a given  particular semimajor axis. They determined the critical mass as function of semimajor axis for some assumed stellar dissipation factors and the ages of the planetary systems. Their results showed that most planetary systems are located at the permitted region of the ``critical mass-semimajor axis'' plot (their Figure 3) except the $\\tau$ Boo system, which needs more careful treatment for the assumed parameter values.  However, if these planets could be formed a bit farther from the central star initially, they should still survive under the tidal interaction and thus might be detected during the inward migration. One should keep in mind that the location where the planets are detected are not where they are formed. The planets from farther place could migrate inward to the region closer to the central star and probably have chances to be detected by us.  To further investigate this problem, we carefully study the planetary migration due to tidal interaction. We try to include the effect of orbital eccentricity at the beginning and we confirm that that the model used in P\\\"atzold \\& Rauer (2002) is a good approximation. We thus use the similar model in  P\\\"atzold \\& Rauer (2002) for the rest calculations. We describe our basic models for tidal interaction in Section 2 and the results will be in Section 3. We provide concluding remarks in Section 4. ",
        "conclusions": "As dynamical friction successfully explained the orbit of Sagittarius dwarf galaxy (Jiang \\& Binney 2001), the tidal interaction can indeed explain the current observed mass-period correlation reported by Zucker \\& Mazeh (2002). The results in Figure 1 give us the full picture of inward migration due to tidal interaction. We found that 0.03 AU seems to be the critical semimajor axis for the planet with mass of order of $\\tau$ Boo system to survive in 2 Gyrs. This is consistent with the current observational results that the smallest semimajor axis of discovered planet is about 0.04 AU.  On the other hand, we can also check this minimum possible semimajor axis from another point of view. In Figure 2, the time scale for a planet can survive is smaller if the planet is closer to the central star initially and  the time a planet can stay around 0.03 AU is considerablely much less than 2 Gyrs, which was regarded as the typical age of these planetary systems. Because time scale is too short, the probability to detect the planet is very small.  Moreover, we interestingly discover the observational ``critical line'' on  ${\\rm ln}(a/{\\rm AU})-{\\rm ln}(M/M_J)$  plane. All the planets on the left side of this line would migrate inward quickly to approach the central star and thus cannot be detected.  Therefore, the initial configuration on ${\\rm ln}(a/{\\rm AU})-{\\rm ln}(M/M_J)$ plane might be composed of all the points on Figure 3(b) plus those points which might have been on the left side of the ``critical line'' about 2 Gyrs ago but disappear in Figure 3(a) because these planets already fall into the central star. From this point of view, even there is correlation between mass and period for current discovered planets as claimed by Zucker \\& Mazeh (2002), this  correlation could be weaker or less obvious at the time when these planets were just formed since we can add arbitrary number of ``possible'' planets on the left side of our observational ``critical line'' if there is no difficulty to form planets there in theory. This tells us that we should be careful when we try to link the mass-period correlation to the theory of planetary formation."
    },
    "0209/astro-ph0209582_arXiv.txt": {
        "abstract": "Temperature maps are presented of the 9 largest clusters in the mock catalogues of Muanwong et al. for both the {\\it Preheating} and {\\it Radiative} models.  The maps show that clusters are not smooth, featureless systems, but contain a variety of substructure which should be observable. The surface brightness contours are generally elliptical and features that are seen include cold clumps, hot spiral features, and cold fronts. Profiles of emission-weighted temperature, surface brightness and emission-weighted pressure across the surface brightness discontinuities seen in one of the bimodal clusters are consistent with the cold front in Abell 2142 observed by Markevitch et al. ",
        "introduction": "Observations of X-ray clusters of galaxies over the last few years have shown that clusters are not the smooth, featureless systems they were expected to be.  X-ray surface brightness and temperature observations indicate the presence of substructure and support the view that cluster formation occurs through the infall and merger of subclusters.  Merger shocks would be expected to occur in such a scenario (e.g. Markevitch \\& Vikhlinin, 2001; Markevitch et al., 2002) but additionally {\\it Chandra} observations have revealed a new phenomenon of `cold fronts' (Markevitch et al, 2000; Forman et al., 2001; Mazzotta, Fusco-Femiano \\& Vikhlinin, 2002). At a cold front, the entropy jump across the sharp gas density discontinuity is in the opposite sense to that expected for a shock, with the high surface brightness side of the dense edge corresponding to low temperature.  The variety of features that have been observed in the clusters mapped so far are of great interest since they may shed light on the physical processes which are occurring as well as containing information on the stage, geometry, scale and velocity of the mergers.  A number of authors have carried out simulations of controlled, single mergers (see Ritchie \\& Thomas 2002 for a review). For example, Ritchie \\& Thomas (2002) carried out high-resolution simulations of the merger of idealized clusters containing both dark matter and gas. They studied the effect on the observable properties of clusters of single head-on and off-centre mergers between both equal and unequal mass objects.  A sequence of maps of emission-weighted temperature with superimposed X-ray surface brightness and velocity fields for mergers between equal mass systems showed the compression and shocking of the gas as the merger progressed. Recently Nagai \\& Kravtsov (2002) carried out a detailed study of cold fronts in high resolution simulations of two clusters forming in different Cold Dark Matter models (standard CDM and $\\Lambda$CDM). Their results indicate that cold fronts are probably fairly common but are non-equilibrium transient phenomena.  In this paper, we present preliminary results from temperature maps of an ensemble of clusters that form within a cosmological simulation, already shown to reproduce the observed X-ray scaling relations at low redshift (Muanwong et al.~2002). This way, we are able to directly assess the range of substructure present in the cluster population. In common with observations, we see significant temperature fluctuations in the hot gas, even when there is little information present in the surface brightness distribution. In particular, we discuss the presence of a cold front in one of our bimodal clusters, which has properties consistent with the cold front in Abell 2142 (Markevitch et al.~2000). ",
        "conclusions": "Temperature maps of the 9 largest clusters in the simulated cluster catalogues of M2002 show clearly that the clusters do contain substructure.  Features that are seen include cold clumps, some of which may be consistent with the infall of cold subclusters, hot spiral features consistent with shocks produced in mergers, and at least one map containing cold fronts.  Elliptical surface brightness contours are the norm, and rotation as expected in off-centre mergers is common. The {\\it Preheating} maps appear 'cleaner', with fewer cold clumps present than for the {\\it Radiative} maps, since the gas has been heated at z=4, wiping out many of the colder blobs. The remaining cold clumps seen in these maps did seem to be associated with motion into or through the cluster.  In this preliminary paper our main aim has been to demonstrate that simulated maps contain a variety of substructure. The scale of the structure is similar to that seen observationally and so gives some indication of the features which should be observable. Future work will focus on the evolution of structure in clusters and on a larger range of cluster masses extending down to the size of groups."
    },
    "0209/astro-ph0209311_arXiv.txt": {
        "abstract": "In the nucleated instability picture of gas giant formation, the final stage is the rapid accretion of a massive gas envelope by a solid core, bringing about a tenfold or more increase in mass.  This tends to trigger the scattering of any nearby bodies, including other would-be giant planet cores; it has been shown in past work that the typical outcome is an outer planetary system very similar to our own. Here, we show that the gravitational scattering accompanying the formation of gas giant planets can also produce, in some cases, outer planets with semimajor axes much larger than those in the Solar System, and eccentricities which remain high for tens of millions of years.  Rings, gaps and asymmetries detected in a number of circumstellar dust disks, which hint at the presence of embedded planets at stellocentric distances far beyond where planet formation is expected to occur, may be connected to such a scenario. ",
        "introduction": "During the lifetime of the nebular gas, a protoplanetary disk with the profile of the standard Hayashi (1981) model, and a surface density several times the minimum, will tend to produce bodies large enough to serve as the solid cores of giant planets, i.e. around 10 M$_{\\oplus}$ in mass, in an annulus roughly corresponding to the the Jupiter-Saturn region (Thommes, Duncan and Levison, hereafter TDL, 2002b).  The inner bound results from the increase of the isolation mass (e.g., Lissauer 1987) with stellocentric distance; inside some radius, the final protoplanet masses are too small.  The outer bound comes about from the increase of accretion timescale with stellocentric distance; beyond some radius, protoplanets cannot grow large enough during the lifetime of the nebular gas ($\\sim$ 10 Myrs, e.g., Strom, Edwards and Skrutskie 1993).  In a system which forms gas giant planets by nucleated instability (e.g., Pollack et al. 1996), one of the giant protoplanets will eventually acquire a massive gas envelope.  Using numerical simulations, TDL (1999, 2002a) demonstrated that this sudden mass increase of one of the protoplanets (by an order of magnitude or more over $\\sim 10^5$ years) tends to destabilize the orbits of its neighbours, scattering them outward onto eccentric orbits.  With most of its orbit now embedded in the still accretionally unevolved outer planetesimal disk, a scattered protoplanet then experiences dynamical friction (e.g., Stewart and Wetherill 1988), which damps its eccentricity down again.  The end result is most often a protoplanet on a nearly circular orbit, albeit with a semimajor axis significantly larger than it originally had.  In this way, a planetary system resembling the configuration of the giant planets of the Solar System is commonly produced, with the scattered and recircularized protoplanets standing in for Neptune, Uranus, and Saturn; the latter needs to acquire its own (less) massive gas envelope before the dispersal of the nebular gas, in order to match the Solar System.  In some of the simulations performed in the above work, protoplanets were so strongly scattered after ``Jupiter'''s gas accretion phase that that they ended up with apocenter distances of hundreds of AU and ultimately became unbound from the Sun.  However, in all these runs the planetesimal disk was only modeled out to a radius of 60 AU.  Here, we investigate the possibility that, given a larger (and thus likely more realistic) planetesimal disk, such bodies might be retained on stable orbits.  This would result in planetary systems with giant planets at stellocentric distances far beyond where significant accretion ought to have taken place. ",
        "conclusions": "A number of spatially resolved debris disks show features which may result from the influence of unseen (giant) planetary companions.  The problem is, these features are seen at large stellocentric distances, far beyond where giant planet formation is expected to have occurred. Examples are the dust disks around HD 141569 (Weinberger et al. 1999), HR 4796A (Schneider et al. 1999), and HD 163296 (Grady et al. 2000).  These systems possess, respectively, a 40 AU wide gap centered at a radius of 250 AU, a dust ring of radius 70 AU, and a 50 AU wide gap at a radius of 325 AU.  A seemingly less extreme case is Vega, which displays two dust emission peaks 80 AU from the star.  These have been successfully modeled as concentrations resulting from the trapping of dust particles (which spiral inward due to radiation pressure and Poynting-Robertson drag) in exterior mean-motion resonances with an eccentric gas giant (Wilner et al. 2002).  A model with a 3 M$_{\\rm Jupiter}$ planet having a semimajor axis of 40 AU and an eccentricity of 0.6 has been demonstrated to give a good match to the observations.  Large scattering events like the ones described here constitute a mechanism by which giant planets might, {\\it after} forming at orbital radii similar to those of Jupiter and Saturn, be transported out to distances comparable to those of the disk features described above. This offers one way to avoid the difficult problem of explaining the formation of giant planets at 100 AU or more, just as the original model avoids the formation timescale problem of Uranus and Neptune in our Solar System.  However, extending the model in this way introduces two principal problems.  First, the mechanism works for ``ice giants''---Uranus/Neptune mass bodies---not for bodies the mass of gas giants.  In the latter case, one would be dealing with a scenario more akin to that described by Rasio and Ford (1996), in which multiple gas giants scatter each other. Eccentricities of any strongly-scattered planets would likely remain high indefinitely, since at $\\ga$ 100 AU the mass in planetesimals is very small compared to that of a gas giant.  The other problem is that, as stated above, the probability of a very large scattering event decreases with the mass of the exterior planetesimal disk.  A minimum-mass disk allows scattering of ice giants to semimajor axes of order 100 AU to occur readily, but a minimum-mass disk is also unlikely to produce giant planets in the first place.  This will be a problem if giant planets at large stellocentric distances turn out to be ubiquitous.  The above-described features do seem to be common, given that they appear in a significant fraction of the (small number of) disks that have been directly imaged thus far.  In order to simultaneously form such large bodies and allow them to be readily scattered to large distances, one must invoke a steep density profile, or a local density enhancement where the giant planets form.  The above considerations provide ways in which the scenario proposed here can, in the near future, be observationally tested.  First, newly-formed gas giants at large stellocentric are hot enough to be detectable, in principal, by current searches (e.g., Macintosh et al. 2000).  Positive detections would immediately rule out the above mechanism as the {\\it sole} source of planets at large orbital radii. Conversely, the continued absence of such detections would suggest that whatever is producing the features we see is smaller than a gas giant.  Or, indeed, the features may have nothing to do with planets at all.  Secondly, future observations will give us a better idea of the typical masses and density profiles of dust disks and, by extension, of the characteristics of the underlying planetesimal disks.  It will then be easier to assess the likelihood of large scattering events.  In particular, it will be interesting to see what correlations, if any, exist between disk mass and the occurence rate of rings, gaps and asymmetries at large radii."
    },
    "0209/astro-ph0209127_arXiv.txt": {
        "abstract": "The Magellanic Stream is a $100\\arcdeg \\times 10\\arcdeg$ filament of gas which lies within the Galactic halo and contains $\\sim 2 \\times 10^8$ \\Msun\\ of neutral hydrogen.  In this paper we present data from the HI Parkes All Sky Survey (HIPASS) in the first complete survey of the entire Magellanic Stream and its surroundings. We also present a summary of the reprocessing techniques used to recover large-scale structure in the Stream. The substantial improvement in spatial resolution and angular coverage compared to previous surveys reveals a variety of prominent features, including: bifurcation along the main Stream filament; dense, isolated clouds which follow the entire length of the Stream; head-tail structures; and a complex filamentary web at the head of the Stream where gas is being freshly stripped away from the Small Magellanic Cloud and the Bridge.  Debris which appears to be of Magellanic origin extends out to 20\\arcdeg\\ from the main Stream filaments. The Magellanic Stream has a velocity gradient of 700 \\kms\\ from the Clouds to the tail of the Stream, $\\sim 390$ \\kms\\ greater than that due to Galactic rotation alone, therefore implying a non-circular orbit. The dual filaments comprising the Stream are likely to be relics from gas stripped separately from the Magellanic Bridge and the SMC. This implies: (a) the Bridge is somewhat older than conventionally assumed; and (b) the Clouds have been bound together for at least one or two orbits. The transverse velocity gradient of the Stream also appears to support long-term binary motion of the Clouds. A significant number of the most elongated cataloged Stream clouds (containing $\\sim 1$\\% of the Stream mass) have position angles aligned along the Stream. This suggests the presence of shearing motions within the Stream, arising from tidal forces or interaction with the tenuous Galactic halo.  As previously noted, clouds within one region of the Stream, along the sightline to the less distant half (southern half on the sky) of the Sculptor Group, show anomalous properties.  There are more clouds along this sightline than any other part of the Stream and their velocity distribution significantly deviates from the gradient along the Stream. We argue that this deviation could be due to a combination of halo material, and not to distant Sculptor clouds based on a spatial and kinematic comparison between the Sculptor Group galaxies and the anomalous clouds, and the lack of cloud detection in the northern half of the group.  This result has significant implications for the hypothesis that there might exist distant, massive HVCs within the Local Group.  Cataloged clouds within the Magellanic Stream do not have a preferred scale size.  Their mass spectrum $f(M_{\\rm HI})\\propto M_{\\rm HI}^{-2.0}$ and column density spectrum $f(N_{HI})\\propto N_{\\rm HI}^{-2.8}$ are steep compared with Ly$\\alpha$ absorbers and galaxies, and similar to the anomalous clouds along the Sculptor Group sightline. ",
        "introduction": "\\label{introduction}  Many searches have been made for streams of halo material; the remnants of Galactic satellites which are responsible for building up the Milky Way (e.g. Newberg et al. 2002; Morrison \\etal 2000; Majewski \\etal 2000; Lynden-Bell \\& Lynden-Bell 1995).  These searches have concentrated on the stellar halo, but neutral hydrogen is also a key tracer of galaxy formation and destruction in the low redshift universe (e.g. Ryder et al. 2001; Smith 2000; Hibbard \\& Yun 1999; Yun, Ho \\& Lo 1994).  Our Galaxy is a prime example of this, with neutral hydrogen streams and their remnants tracing the more recent aspects of the Milky Way's formation and evolution. The most famous Galactic halo HI stream is the Magellanic Stream. Discovered 30 yrs ago (Wannier \\& Wrixon 1972; Mathewson \\etal 1974), this complex arc of neutral hydrogen starts from the Magellanic Clouds and continues for over 100$^{\\circ}$ through the South Galactic Pole.  It has been created through the interaction of our Galaxy with the Magellanic Clouds and may represent a recent example of the accretion and merging process which created the Milky Way.   The finding of a leading  stream of material (the Leading Arm) indicates that the dominant mechanism responsible for forming the Stream is tidal (Putman \\etal 1998), however it remains unclear how much the passage through the Galaxy's diffuse corona or extended disk may have shaped this feature.  The accretion and merging of HI clouds may be occurring throughout the Local Group and there has been a great deal of controversy as to whether some of the high-velocity clouds (HVCs) represent the leftover Local Group building blocks at average distances of 700 kpc from the Milky Way (e.g. Blitz \\etal 1999; Braun \\& Burton 1999).  Many of the clouds would have HI masses of a few times $10^7$ \\Msun\\ at Local Group distances; however, if only the compact, isolated HVCs are at large distances, the HI mass of a typical intra-Local Group cloud would only be a few times $10^6$ \\Msun\\ (Putman \\etal 2002).  The detection of star-free intergalactic HI clouds in nearby groups which are kinematically similar to the Local Group would greatly support a Local Group origin for HVCs. Deep HI surveys of nearby groups have detected 0 intergalactic HI clouds to masses of a few $\\times 10^7$ \\Msun~(e.g. Zwaan 2001; Banks \\etal 1999).  Recently, de Blok et al. (2001) surveyed 2\\% of the total area of the Sculptor and Centaurus A groups to HI masses of a few times $10^6$ \\Msun\\ and also found 0 free-floating HI clouds.  Considering the mass limitations of the large scale surveys and the limited area covered in the Sculptor and Centaurus A group survey, the proposal of compact HVCs being scattered throughout the Local Group remains a possibility.  The Local Group is unvirialized and is falling together for the first time (Schmoldt \\& Saha 1998). A nearby group which may be kinematically similar to the Local Group is the Sculptor Group.  The Sculptor Group is also unvirialized and appears to form part of a large continuous filament of galaxies which includes the Local Group (Cote \\etal 1997; Jerjen \\etal 1998). There is growing evidence that intergalactic clouds (Ly$\\alpha$ absorbers) trace filaments and clusters of galaxies (Penton, Shull \\& Stocke 2000), and the unsettled environment of the Sculptor-Local Group cloud could be the ideal environment to find these intergalactic clouds in both absorption and emission.  The Sculptor Group falls along the same sightline as a section of the Magellanic Stream.  This, together with the low velocities of the Sculptor Group galaxies (down to 70 \\kms), make it very difficult to distinguish how much of the high-velocity neutral hydrogen is directly related to the Magellanic Stream and how much may be intra-Sculptor Group/Local Group material.  Mathewson \\etal~ (1975) was the first to argue that some of the HI clouds along the sightline  to the Magellanic Stream do not seem to fit into the normal velocity distribution of the Stream and may be associated with Sculptor Group galaxies. This was countered by Haynes \\& Roberts (1979) and Haynes (1979) who argued it is merely a spatial coincidence. The origin of these clouds was ultimately left unanswered, with the possibility of a detection of intergalactic HI clouds remaining alluring.   Uncovering the Magellanic Stream and its surroundings is crucial to the understanding of the formation and evolution of not only the Milky Way, but the entire Local Group, yet until the present paper, the neutral hydrogen data for the entire Stream and its surroundings remained severely under-sampled (e.g. Mathewson \\etal 1974; Bajaja \\etal 1985).  Sections of the Stream have been observed at higher spatial or velocity resolution (e.g. Cohen 1982; Haynes 1979; Morras 1983, 1985; Wayte 1989),  but these observations focused on small regions within the main filament of the Stream originally presented by Mathewson \\etal.  Here we present HI maps of the entire Magellanic Stream and its surroundings at 15.5$^{\\prime}$ resolution (equivalent to 250 pc at 55 kpc) using data from the HI Parkes All Sky Survey (HIPASS; Barnes \\etal~ 2001). Though a tidal origin may be the primary one for the Stream, there is clearly a complex history behind the formation of this feature.  The large area of the survey also provides new insight into the origin of the clouds along the Sculptor Group sightline.  We begin this paper by describing the HIPASS observations and the HVC data reduction technique (known as {\\sc minmed5}) and subsequently describe the spatial and kinematic HI structure of the Stream and the high-velocity clouds along the Sculptor Group sightline.  We then go on to discuss the origin of the Stream and the anomalous clouds along the Sculptor Group sightline. ",
        "conclusions": ""
    },
    "0209/astro-ph0209477_arXiv.txt": {
        "abstract": "The Rapid Telescope for Optical Response (RAPTOR) program consists of a network of robotic telescopes dedicated to the search for fast optical transients.  The pilot project is composed of three observatories separated by approximately 38 kilometers located near Los Alamos, New Mexico.  Each of these observatories is composed of a telescope, mount, enclosure, and weather station, all operating robotically to perform individual or coordinated transient searches. The telescopes employ rapidly slewing mounts capable of slewing a 250 pound load 180 degrees in under 2 seconds with arcsecond precision. Each telescope consists of wide-field cameras for transient detection and a narrow-field camera with greater resolution and sensitivity. The telescopes work together by employing a closed-loop system for transient detection and follow-up.  Using the combined data from simultaneous observations, transient alerts are generated and distributed via the Internet.  Each RAPTOR telescope also has the capability of rapidly responding to external transient alerts received over the Internet from a variety of ground-based and satellite sources.  Each observatory may be controlled directly, remotely, or robotically while providing state-of-health and observational results to the client and the other RAPTOR observatories.  We discuss the design and implementation of the spatially distributed RAPTOR system. ",
        "introduction": "\\label{sect:intro}  Constructing an observatory that can run in a completely autonomous manner is a significant challenge.  Not only must it be capable of scheduling its own observations, but it must robustly handle things such as changing weather conditions, software errors, and hardware failures.  Recently, several projects have met this challenge admirably.  The recent detection of a prompt optical flash from the gamma-ray burst GRB990123 by the Robotic Optical Transient Search Experiment I (ROTSE-I) at Los Alamos National Laboratory demonstrated the value of small robotic observatories\\cite{Akerlof99}.  The 80 second flash from GRB990123 was the brightest object ever observed in optical wavelengths at an absolute magnitude of -36.  The detection of GRB990123 also demonstrates how poorly the night sky is monitored for transient objects.  Gamma-ray bursts occur every day, yet the ROTSE-I detection is still the only contemporaneous optical detection of a GRB.  Undoubtedly many more events like this have been within the reach of small telescopes, yet few have been observed. The aim of the RAPTOR project is to develop a wide-field system for detecting these optical transients\\cite{Vestrand01}.  The system should have the greatest possible sensitivity for the given field of view and also be able to operate robotically.  A fast analysis pipeline is also necessary if transients are to be identified in near real-time.  The ability to reject false triggers is an absolute necessity for any program that intends to do a wide-field search for transients that occur on the timescale of minutes.  This was a significant problem in previous experiments such as the Explosive Transient Camera (ETC)\\cite{Vanderspek94}.  There can be many causes of false triggers in any astronomical imaging system; hot pixels, satellite glints, meteors, camera defects, cosmic rays, etc.  One way to reduce the number of false triggers is to require that any object must be present in more than one image to be considered a valid detection.  This technique can eliminate many false triggers, however things such as a glint from a geostationary satellite may still get through.  An additional technique for eliminating false triggers is to use coordinated observations from spatially separated observatories.  It is very unlikely that a camera defect will appear at the same location on two different telescopes.  Additionally, if the observatories are separated by a sufficient distance, parallax can be used to eliminate nearby objects such as satellites and meteors.  Developing a set of telescopes that can operate in a coordinated manner further aggravates the challenge of building fully robotic observatories.  Not only must each observatory be able to operate autonomously, they must also be able to communicate with each other and synchronize their observations.  Further, if they are going to search for transients in real-time, they must be able to process and compare their observations with each other in a fast automated way. The goal of the RAPTOR project is to build just such a system.  To meet these goals, we have decided on a system similar to human vision.  Humans use two eyes not just for depth perception, but also to filter out artifacts that appear in only one eye.  We have a central fovea which has greater resolution and color sensitivity.  Our vision process operates in a feedback loop which can quickly identify a transient object at the edge of the field of view and ``slew'' our eyes to center the object in the foveas to study the object with greater sensitivity.  Following this example, we have decided to construct two telescopes, RAPTOR A and B, which will operate simultaneously.  Each telescope will have a set of four wide field cameras and a central ``fovea'' camera.  A real-time image processing pipeline will identify transients in the wide field system.  If a transient is present in both telescopes at the same location and the same time, each telescope will re-center the object in the fovea camera.  Additionally we are constructing a third system, RAPTOR S, which will consist of a single 12 inch telescope with a transmission grating allowing low resolution spectroscopic follow-up of any transients identified by the RAPTOR A-B system. ",
        "conclusions": "\\label{sect:status}  The RAPTOR project is a group of robotic telescopes that will search for optical transients.  Each system will operate in a completely autonomous manner.  Additionally, each telescope will communicate with the others allowing them to operate in synchronization.  Using this technique, we will be able to eliminate false transient detections with high efficiency.  To accomplish the task of synchronized observations, we are developing new data acquisition software which will coordinate the activity of the RAPTOR telescopes. Three telescopes are currently under construction; RAPTOR A, B, and S.  In late February 2002, RAPTOR A had first light and began limited operation in manual mode.  Since that time, construction has finished on RAPTOR A and RAPTOR B.  We also expect construction of the RAPTOR S telescope to be completed by the end of August 2002.  Initial testing on all three telescopes indicate that they all will perform within expectations.  Several scientific observations have been made with the RAPTOR A telescope, including monitoring the eclipsing binary W Ursae Majoris and photographing the comet Ikeya-Zhang (Fig. \\ref{fig:comet}).  RAPTOR A and B are currently able to operate in a limited robotic mode using an early version of the RAPTOR DAQ system.  This early version of the DAQ system allows for simple sky patrols and alert responses.  The client/server capability of {\\sl controld} is not yet implemented, nor is the transient alert feedback loop.  However, the automated processing pipeline is running on all of the camera computers.  We expect the DAQ system to be fully functional in the coming months."
    },
    "0209/astro-ph0209194_arXiv.txt": {
        "abstract": "The active galaxy Markarian 421 underwent a substantial outburst in early 2001. Between January and May of that year, the STACEE detector was used to observe the source in \\gammaray s between the energies of 50 and 500 GeV.  These observations represent the lowest energy \\gammaray\\  detection of this outburst by a ground-based experiment.  Here we present results from these observations, which indicate an average integral \\gammaray\\  flux of $(8.0\\pm0.7\\pm1.5)\\times 10^{-10}$ ${\\rm cm}^{-2}{\\rm s}^{-1}$ above 140 GeV. We also present a light curve for Markarian 421 as observed by STACEE from March to May, and compare our temporal, as well as spectral, measurements to those of other experiments. ",
        "introduction": "The blazar Markarian 421 is one of only a handful of astrophysical objects detected in the very high energy (VHE) \\gammaray\\  regime, between 10 GeV and 100 TeV.  At a red-shift of 0.031, it is the closest BL Lac object seen by EGRET, and the first to be detected by a ground-based TeV instrument \\citep{punch92,petry96}. Along with Markarian 501 (z=0.034), it is one of the most prominent sources of extra-galactic VHE \\gammaray s, and has been regularly monitored by atmospheric Cherenkov telescopes since its TeV identification \\citep[see][for a review]{catanese99}.  Markarian 421 is an X-ray selected BL Lac object, and as such, it exhibits the double-humped spectrum that is characteristic of blazars \\citep[e.g.][]{takahashi00}. The lower energy hump, believed to arise from synchrotron radiation, peaks in the X-ray range, while the higher energy hump, generally attributed to inverse Compton (IC) scattering of soft photons, peaks in the GeV to TeV energy range.  The details of the emission processes that produce the IC hump are, as yet, unresolved. For example, the origin of the soft photons that seed the IC component of the spectrum is an outstanding question in the study of blazars. Although synchrotron self-Compton models, in which the seed photons originate from synchrotron radiation within the blazar jets, are generally favored for Markarian 421 \\citep{coppi92,coppi99}, other competing models exist.  These include external Compton models \\citep{dermer92,sikora94}, proton-induced cascades \\citep{mannheim93}, and proton synchrotron models \\citep{aharonian00,mucke01}.  Markarian 421 has been given to strong periods of flaring activity over the past few years.  Of note was a flare in 1996 during which the source was detected at more than ten times the flux of the Crab Nebula \\citep{gaidos96}.  In the early part of 2001, a rather impressive flare was again observed \\citep{iau}, with reported TeV fluxes of comparable magnitude to those recorded in 1996, but of much longer duration \\citep{krennrich01}.  It was during the 2001 flare that STACEE--48, an intermediate incarnation of STACEE (the Solar Tower Atmospheric Cherenkov Effect Experiment), was commencing operation. STACEE was able to observe the activities of Markarian 421 for much of the flaring period.  These observations are of particular note as they represent the only \\gammaray\\  detection below 200 GeV during this flare. ",
        "conclusions": "STACEE has detected the BL Lac object, Markarian 421, with high significance in the 140 GeV energy band, a hitherto unexplored region of its spectral energy distribution.  The STACEE observed flux is consistent with other VHE \\gammaray\\  observations, and the temporal evolution of the STACEE observations appears similar to observations in both the TeV and X-ray bands.  Figure \\ref{fig-flux2} indicates that the STACEE measurements occupy an important region in Markarian 421's spectral energy distribution near the peak of the IC hump.  Future measurements of the spectrum by STACEE, along with simultaneous data at X-ray and TeV energies should help to further constrain synchrotron and IC blazar models for this source.  STACEE is currently operating with 64 channels, each equipped with a flash ADC, and should be able to obtain more detailed spectral information in the near future."
    },
    "0209/astro-ph0209531_arXiv.txt": {
        "abstract": "{We present the results of $UBVJHKLM$ photometry of \\r\\ spanning the period from 1976 to 2001. Studies of the optical light curve have shown no evidence of any stable harmonics in the variations of the stellar emission. In the $L$ band we found semi-regular oscillations with the two main periods of $\\sim 3.3$\\,yr and $\\sim 11.9$\\,yr and the full amplitude of $\\sim 0\\fm8$ and $\\sim 0\\fm6$, respectively. The colors of the warm dust shell (resolved by Ohnaka \\e \\cite{ohnaka01}) are found to be remarkably stable in contrast to its brightness. This indicates that the inner radius is a constant, time-independent characteristic of the dust shell. The observed behavior of the IR light curve is mainly caused by the variation of the optical thickness of the dust shell within the interval $\\tau(V)= 0.2-0.4$. Anticorrelated changes of the optical brightness (in particular with $P \\approx 3.3$\\,yr) have not been found. Their absence suggests that the stellar wind of \\r\\ deviates from spherical symmetry. The light curves suggest that the stellar wind is variable. The variability of the stellar wind and the creation of dust clouds may be caused by some kind of activity on the stellar surface. With some time lag, periods of increased mass-loss cause an increase in the dust formation rate at the inner boundary of the extended dust shell and an increase in its IR brightness. We have derived the following parameters of the dust shell (at mean brightness) by radiative transfer modeling: inner dust shell radius $r_{\\rm in} \\approx 110\\,R_*$, temperature $T_{\\rm dust}(r_{\\rm in}) \\approx 860$\\,K, dust density $\\rho_{\\rm dust}(r_{\\rm in}) \\approx 1.1\\times 10^{-20}\\,{\\rm g\\,cm^{-3}}$, optical depth  $\\tau(V) \\approx 0.32$ at 0.55\\,$\\mu$m, mean dust formation rate $\\dot{M}_{\\rm dust} \\approx 3.1 \\times 10^{-9}\\,{\\rm M_{\\sun}\\,yr^{-1}}$, mass-loss rate $\\dot{M}_{\\rm gas} \\approx 2.1 \\times 10^{-7}\\,{\\rm M_{\\sun}\\,yr^{-1}}$, size of the amorphous carbon grains $\\la 0.01\\,\\mu$m, and $B-V \\approx -0.28$. ",
        "introduction": "R Coronae Borealis is the prototype of a small group of yellow supergiants (about 35 known members in our Galaxy) characterized by sudden declines in their optical brightness and extremely hydrogen-deficient, carbon-rich atmospheres. The visual light curve of \\r\\ contains quasi-regular low-amplitude variations without any dominating periods. These are commonly interpreted in terms of stellar pulsations and episodic deep declines, which typically last a few months and have amplitudes of up to 8\\,magnitudes. Such events are thought to be the result of obscuration by dust clouds generated spasmodically and blown away from the star by radiation pressure (Clayton \\cite{clay96} and references therein).  The near-infrared excess, discovered by Stein \\e (\\cite{stein69}), is assigned to a warm dust shell (blackbody temperature of $\\sim 900$\\,K). This extended dust shell was resolved for the first time by speckle interferometric observations with the SAO 6 m telescope (Ohnaka \\e {\\cite{ohnaka01}). The IR excess contributes about 30\\% of the total flux and is permanently present, regardless of the visual brightness of the object. The $L$ flux of \\r, which is mainly due to dust emission, varies semi-regularly with a period of 1260\\,days (Feast \\e \\cite{feast97} and references therein). The analysis of IRAS observations at 60\\,$\\mu$m and 100\\,$\\mu$m (Gillett \\e \\cite{gill86}) led to the discovery of a very extended ``fossil'' shell around \\r, whose diameter and temperature are $\\simeq 18\\arcmin$ and 30\\,K, respectively.  In this paper we focus on the investigation of the time-dependent characteristics of the extended warm dust shell and their interpretation. In the next section we present the results of our photometric observations of \\r\\ ($UBV$ in 1994--1999 and $JHKLM$ in 1983--2001) and the complete tables of photometric observations of the star in 1976--2001. The basic characteristics of the optical and IR radiation of the star during its bright state are analyzed in Sect.\\,3. Radiative transfer modeling of the warm dust shell at its average brightness is described in Sect.\\,4. The variations of the dust shell brightness and colors, and, correspondingly, its structural parameters, are investigated in Sect.\\,5. Our results are discussed in Sect.\\,6. ",
        "conclusions": "We have presented optical and IR long-term monitoring and radiative transfer modeling of \\r. These studies have allowed the derivation of various time-dependent properties of both (i)\\,the star itself at bright state and (ii)\\,an extended dust shell (similar to that observed around many supergiants) which condenses from the supergiant wind at a large distance of approximately 100 stellar radii from the star. This extended dust shell is larger than the suggested region of dust clouds (distance approx. 2--30 stellar radii, see, e.g. Clayton \\cite{clay96}, Fadeyev \\cite{fad88}), which causes the deep minima in the visual light curve. The extended dust shell (radius $\\sim 19$\\,mas) was first resolved by speckle interferometric observations with the SAO 6\\,m telescope (Ohnaka \\e \\cite{ohnaka01}).  In the $V$ and $L$ light curves probably a $\\sim$\\,4-year wind travel time from the stellar surface to the extended dust shell with a radius of approximately 110 stellar radii can be seen. The comparison of the $V$ and $L$ light curves shows that the period 1994--1998 of an increased (and decreasing) $L$ and $M$ brightness started  approximately 4 years after the end (in 1990) of a period with many deep minima in the visual light curve. This can be explained if we assume that during phases of increased stellar magnetic activity (see studies by Soker \\& Clayton \\cite{soker99}) there is a simultaneous increase of both (i)\\,the supergiant wind and (ii)\\,the rate of formation of dust clouds. If this assumption is true, the increased stellar wind period ended in 1990. The L band brightness increased until 1994. This can be explained by the increased stellar wind until 1990 and a wind travel time of 4 years from the stellar surface to the extended dust shell region. In the period 1994-1999 the L band brightness was decreasing because in this period only lower intensity stellar wind arrived, which was produced during the low-activity period 1990-1995. We believe that the variation of the stellar wind can be considered as an indirect argument in favor of active phenomena on the surface of \\r."
    },
    "0209/hep-ph0209264_arXiv.txt": {
        "abstract": "\\bigskip  Recent work has shown that dispersion relations with Planck scale Lorentz violation can produce observable effects at energies many orders of magnitude below the Planck energy $M$. This opens a window on physics that may reveal quantum gravity phenomena. It has already constrained the possibility of Planck scale Lorentz violation, which is suggested by some approaches to quantum gravity.  In this work we carry out a systematic analysis of reaction thresholds, allowing unequal deformation parameters for different particle dispersion relations. The thresholds are found to have some unusual properties compared with standard ones, such as asymmetric momenta for pair creation and upper thresholds.  The results are used together with high energy observational data to determine combined constraints. We focus on the case of photons and electrons, using vacuum \\v{C}erenkov, photon decay, and photon annihilation processes to determine order unity constraints on the parameters controlling $O(E/M)$ Lorentz violation. Interesting constraints for protons (with photons or pions) are obtained even at $O((E/M)^2)$, using the absence of vacuum \\v{C}erenkov and the observed GZK cutoff for ultra high energy cosmic rays.  A strong \\v{C}erenkov limit using atmospheric PeV neutrinos is possible for $O(E/M)$ deformations provided the rate is high enough. If detected, ultra high energy cosmological neutrinos might yield limits at or even beyond $O((E/M)^2)$. ",
        "introduction": "The principle of relativity of motion goes all the way back to Galileo~\\cite{dialog}, who noted that observers below decks in a large ship gliding across a calm sea have no way of determining whether they are in motion or at rest. Einstein's special relativity, which is founded on this principle, has been spectacularly successful in accounting for phenomena involving boost factors as high as $10^{11}$. Moreover, the Lorentz group has a beautiful mathematical structure, and this symmetry powerfully constrains theories in a way that has been very useful in discovering new laws of physics. It is natural to assume under these circumstances that Lorentz invariance is a symmetry of nature up to arbitrary boosts. Nevertheless, there are several good reasons to question exact Lorentz symmetry. ~From a logical point of view, the most compelling reason is that while $10^{11}$ is a large number, it is nowhere near infinity. There is, and will always be, an infinite volume of the Lorentz group that is experimentally untested since, unlike the rotation group, the Lorentz group is non-compact. Why should we assume that {\\it exact} Lorentz invariance holds when this hypothesis cannot even in principle be tested?  While the non-compactness reason for questioning Lorentz symmetry is perhaps logically compelling, it is by itself not very encouraging. However, there are also several reasons to suspect that there will be a failure of Lorentz symmetry at some energy or boosts. One reason is the ultraviolet divergences of quantum field theory, which are a direct consequence of the assumption that the spectrum of field degrees of freedom is boost invariant. Another reason comes from quantum gravity. Profound difficulties associated with the ``problem of time'' in quantum gravity~\\cite{Isham,Kuchar} have suggested that an underlying preferred time may be necessary to make sense of this physics.  Also tentative results in string theory~\\cite{KS89}, quantum geometry~\\cite{loopqg}, and non-commutative geometry~\\cite{Hayakawa,Carroll:2001ws, Amelino-Camelia:2001cm} approaches to quantum gravity have suggested that Lorentz symmetry may be broken in the ground state.  Finally, there have been recent hints from high energy astroparticle physics that we may already be seeing the effects of Lorentz violation (although as discussed below the most recent analyses make this seem unlikely.)  One comes from the photo-production of electron-positron pairs when cosmic gamma rays collide with photons of the infrared background. Below 10 TeV the (indirectly) observed absorption of such gamma rays by this process offers support for boost invariance up to the boost that relates the cosmic rest frame to the center of mass frame of the colliding photons. (For a 10 TeV gamma ray colliding head on with a 25 meV infrared photon this yields a boost of $10^7$.) However, according to some (but not all) models of the infrared background, there appears to be less absorption than expected for gamma rays above 10 TeV coming from the blazar Mkn 501 (located at about $157$ Mpc from us). If true this could be explained by an upward threshold shift due to a Planck scale suppressed Lorentz violating term in the dispersion relation for the gamma rays~\\cite{Protheroe:2000hp}.  The other hint comes from the cosmic ray events beyond the GZK cutoff~\\cite{G,ZK} on high energy protons. Ultra high energy protons undergo inelastic collisions with CMBR photons leading to the production of pions (the boost to the center of mass frame yields the figure of $10^{11}$ mentioned above). As a result, protons above $\\sim 5\\times10^{19}$ eV are not able to reach us from distances above a few Mpc~\\cite{Stecker68}. In spite of this prediction, cosmic rays with energy beyond $10^{20}$ eV have apparently been observed by the AGASA experiment~\\cite{GZKdata} (see also~\\cite{NW00} for a review on this issue). The nature and origin of these ultra high energy cosmic rays is unknown and several explanations have been proposed (see~\\cite{Sigl, Stecker:2002fh} for an extensive review). One proposal is that Lorentz violating terms in the dispersion relation for the proton produce an upward shift of the threshold for pion production, allowing these high energy protons to reach us~\\cite{Mestres,CG,Bertolami,ACP}. Interestingly it was argued that a universal Lorentz violating deformation of the particle dispersion relations would be capable of explaining both the TeV gamma ray absorption anomaly and the trans-GZK events~\\cite{ACP}.  The evidence for the TeV gamma ray and GZK anomalies is not convincing at this stage, however. Indeed it has been argued in~\\cite{stecker01, Stecker:2002fh} for the former and in~\\cite{Bahcall:2002wi,FLYeye02} for the latter that the data are consistent with Lorentz invariance. For us therefore the most important point is just that it is possible at all that Planck scale violations of Lorentz symmetry could be observed or constrained by current and upcoming observations. The focus of the present paper is almost entirely on the {\\it constraints} that can be imposed. Our work extends prior results~\\cite{Mestres,CG,Bertolami,Acea,ACP,Kluzniak,Kifune:1999ex,Aloisio:2000cm} in several ways: (i) combining constraints to limit parameter space of {\\it a priori} independent parameters, (ii) discovery and characterization of the asymmetric threshold effect, (iii) characterization of upper threshold effects, (iv) extending analysis for threshold effects to higher order nonlinearities.  A brief report on some of our results has already been given in~\\cite{Jacobson:2001tu}. Some of these results have been confirmed in~\\cite{Major}.  In the next section we discuss our theoretical framework and list the reactions we are going to consider. In Section~\\ref{sec:qed} we study the kinematics of some photon--electron processes in order to determine how Lorentz violating dispersion affects thresholds. The details of the photon annihilation threshold analysis are worked out in the Appendix. These results are then used to deduce observational constraints on the electron and photon deformation parameters.  Taken jointly these constraints severely restrict the parameter plane. Section~\\ref{sec:other} is devoted to the discussion of other possible interactions including hadrons or neutrinos, and in section~\\ref{sec:univ} we discuss the special case of common Lorentz violating parameters for all the particles.  Finally we present some conclusions and perspectives in section~\\ref{sec:disc}.  Throughout this paper we adopt the following notational conventions: $\\fp$ denotes a four-momentum $\\fp=(\\omega,\\p)$, and $p$ is the magnitude of the three-vector $\\p$. The metric signature is $(+,-,-,-)$. We use the energy scale $M=10^{19}$ GeV to form dimensionless Lorentz-violating parameters, since it is close to the Planck energy $M_{\\rm P}=(\\hbar c^5/G)^{1/2}\\simeq 1.22 \\cdot10^{19}$ which we are presuming sets the scale for violation of Lorentz invariance induced by quantum gravity. We often employ units in which $M=1$. ",
        "conclusions": "Upper thresholds exist for $n=3$ only below the diagonal and between the $\\b=1.5$ and $\\b=\\infty$ (which gives the same line as $\\b=1$) symmetric contours (\\ref{symmn3}). For a given $\\b$ the threshold is symmetric in the region above the line $\\widetilde{\\xi}=-4/\\b^2$ and asymmetric below, where the contour is given by the curve (\\ref{asymmn3}). The regions of symmetric and asymmetric upper thresholds for $n=3$ are shown in Figure~\\ref{fig:upperregions}. \\begin{figure}[htb] \\vbox{ \\vskip 8 pt \\centerline{\\includegraphics[width=2.7in]{upperphase}} \\caption{\\label{fig:upperregions} Regions where the upper threshold is determined by the symmetric configuration (light grey region) or the asymmetric one (dark grey region). In the white region below the light grey and below the diagonal the reaction never occurs, and in the rest of the white region there is a lower threshold but no upper threshold. \\smallskip} } \\end{figure} The boundary of the lens shaped region next to the diagonal is determined by the curve $\\widetilde{\\xi}_{\\rm join}=- (-\\widetilde{\\eta})^{2/3}$ consisting of the points where the symmetric and asymmetric segments join. The bottom of the lens meets the diagonal at $\\widetilde{\\eta}=\\widetilde{\\xi}=-1$ where the symmetric $\\b=2$ line crosses, so asymmetric upper thresholds exist only for $\\b>2$. The lower boundary of the region of upper thresholds is the $\\b=1.5$ line, which meets the diagonal at $\\widetilde{\\eta}=\\widetilde{\\xi}=-32/27$.  The possibility of upper thresholds for photon annihilation has been previously discussed by Klu\\'zniak~\\cite{Kluzniak}, who gave results for the values $\\eta=0$, $\\xi=-1$, and $\\eta=\\xi=-1$ in the $n=3$ case. It seems that only the symmetric configuration was examined in \\cite{Kluzniak}, hence his results cannot fully agree with ours in cases where the asymmetric configuration is important. For the case $\\eta=0$, and negative $\\xi$, our results show that there is a symmetric upper threshold only for $\\widetilde{\\xi}$ values above the $\\beta=1.5$ line, i.e. for $\\widetilde{\\xi}>-16/27$. Our upper threshold agrees with that of \\cite{Kluzniak} in the limit $|\\widetilde{\\xi}/4|=|\\xi m^4/4\\omega_0^3|\\ll1$. The left hand side is unity for $\\xi=-1$ and $\\omega_0\\simeq 20$ meV, hence our results agree approximately provided $\\omega_0$ is greater than about $\\simeq 40$ meV. In the diagonal case, while our results for the symmetric configuration agree in the same limit, we have seen that there is no upper threshold since asymmetric configurations exist for arbitrarily large $\\b$.  \\subsubsection{Observations and constraints from absence of deviations from standard photon annihilation}\\label{sec:photann}  The \\v{C}erenkov and photon decay constraints leave open an infinite wedge-shaped region including the diagonal in the lower left quadrant for the case $n=3$.  A constraint from agreement with standard photon annihilation would be complementary to these and hence has the potential to confine the allowed region to a small neighborhood of the origin. Such a constraint is provided by indirect observations of annihilation of high energy gamma rays from blazars on the cosmic background radiation (CBR). Since there is presently considerable uncertainty regarding both the background radiation and the nature of the sources, the constraint that can be extracted is not yet very precise however.  Another limitation of the present work arises from the fact that each observed gamma ray has the opportunity to interact with soft photons at any energy above the threshold, so to compare with observation one should compute the absorption using the Lorentz violating dispersion relation, integrating over all target frequencies.  Such an investigation lies outside the scope of the present paper, so we shall only attempt to roughly characterize how large a threshold shift might be compatible with current observations.  We now summarize the observational situation. The BL Lac objects Mkn 421 and Mkn 501 are a type of blazar emitting high energy gamma rays whose observed spectrum reaches 17 TeV in the case of Mkn 421~\\cite{Mkn421} and 24 TeV in the case of Mkn 501~\\cite{Mkn501}. The source power spectra are reconstructed accounting for absorption via photon annihilation on the intervening CBR, which ranges from the near infrared (NIR, $\\sim 1 ~\\mu$m) to the cosmic microwave background (CMBR, $\\sim 1000 ~\\mu$m). Currently we have a good knowledge of the NIR and CMBR but uncertainties remain regarding the distribution in the intermediate, mid infrared ($\\sim 10 ~\\mu$m) and far infrared ($100 ~\\mu$m), regions (see e.g. Figure 1 of~\\cite{Aharonian:2001cp} or the discussion in~\\cite{stecker01}).  Some models of the IR background imply a source spectrum for Mkn 501 with an unexpected amount of radiation (a ``pile-up'') above $10$ TeV~\\cite{Protheroe:2000hp,Aharonian:2001cp}. If such IR backgrounds are correct, the pile-up might be due to a process producing enhanced emission at energies larger than $10$ TeV~\\cite{Aharonian:2001cp}, or it might be explained by anomalously low absorption caused by an upward shift of the threshold due to Lorentz violation~\\cite{ACP, Kifune:1999ex,Kluzniak,Protheroe:2000hp,Aloisio:2000cm}. However, recent work~\\cite{SG01,stecker01} based on improved reconstructions of the FIRB and on a new analysis of the gamma ray flux from Mkn 501 supports the view that current observations are consistent with the predictions of standard Lorentz invariant theory up to 20 TeV.  Even without resolving the question of the pile-up, it seems well established that some degree of photon absorption has been observed up to 20 TeV, which already provides an interesting constraint on Lorentz violation. Moreover, it is our impression that the suggestions of an anomaly above 10 TeV will likely prove illusory as new observations are made available, confirming the results of~\\cite{SG01,stecker01}~\\footnote{After this work was completed a further observational analysis appeared~\\cite{Konopelko:2003zr}. This allows the observational basis for the constraint discussed in this paper to be solidified~\\cite{GACcom}.}. We can thus obtain observational constraints from the requirement that the Lorentz violation does not too strongly modify standard Lorentz-invariant thresholds for photon annihilation. The strength of the constraints depends of course on the order $n$ of the Lorentz deformation. The general threshold equation (\\ref{eq:gfggsapp}) shows that an order unity constraint on $\\b$ translates into an order unity constraint on $\\widetilde{\\eta}$ and $\\widetilde{\\xi}$, which corresponds to an order $\\o_0^n/m^{2(n-1)}$ constraint on $\\eta$ and $\\xi$. Since all studies seem to agree that more or less standard Lorentz-invariant absorption is occurring for gamma rays up to 10 TeV, we shall use the corresponding soft photon threshold of $\\o_0=25$ meV $\\sim$ 50 $\\m$m as a numerical benchmark. One then has $\\o_0^2/m^2\\sim 10^{-15}$ for $n=2$, $\\o_0^3/m^4\\sim 1$ for $n=3$, and $\\o_0^4/m^6\\sim 10^{15}$ for $n=4$. Hence only the $n=2$ and $n=3$ cases can provide interesting constraints. Note that in the $n=3$ case, which is of most interest to us, the dependence on $\\o_0$ is cubic, so for example a constraint at $2\\, \\o_0$ is eight times weaker than a constraint at $\\o_0$, while one at $\\o_0/2$ is eight times stronger. This means also that there could be strong deviations in absorption for, say, 20 TeV gamma rays, and yet little deviation for 10 TeV gamma rays, since the standard soft target threshold $m^2/E$ is half as large for the 20 TeV gamma rays.  To formulate the constraints we begin by identifying the contour in the $\\xi$--$\\eta$ plane, for which the threshold is not shifted away from the Lorentz-invariant value. For $n=2$ this no-shift contour is given by the diagonal $\\xi=\\eta$ (corresponding to equal speeds of light for electrons and photons), which is independent of the soft photon energy $\\omega_0$. For $n=3$ the contour is given by the joined symmetric and asymmetric $\\b=1$ contours (\\ref{symmn3}) and (\\ref{asymmn3}) converted to the unscaled parameters, \\begin{eqnarray} \\xi &=& \\displaystyle{\\frac{\\eta}{2}} \\qquad\\qquad\\qquad\\qquad\\qquad\\mbox{for $\\eta>-8\\omega^{3}_{0}/m^4$} \\label{kstsy}\\\\ \\nonumber \\\\ \\xi &=& \\displaystyle{\\eta-\\frac{4\\omega_{0}^3}{m^4}+ \\sqrt{-\\frac{8\\omega_{0}^3}{m^4}\\eta}} \\qquad\\mbox{otherwise} \\label{kstasy} \\end{eqnarray} The symmetric part is independent of $\\o_0$ but the joining point and the asymmetric part are not.  Above the no-shift contour, Lorentz violation {\\it lowers} the threshold. Since the shift would be larger for higher energy gamma rays this might, depending on the details of the IR background spectrum, enhance the ``pile-up\" in the reconstructed source spectrum if the IR backgrounds of \\cite{Protheroe:2000hp} are used, or it might produce a pile-up where one did not otherwise exist if the IR background of~\\cite{stecker01} is used.  We thus consider it unlikely that there is much downward shift of the threshold. In any case, nearly all of the region above the no-shift line is already excluded by the photon decay and \\v{C}erenkov constraints.  Below the no-shift contour, Lorentz violation {\\it raises} the threshold. We now consider the constraints this can yield in the cases $n=2$ and $n=3$.  \\paragraph{$n=2$ Photon annihilation constraints.} Constraints in the $n=2$ case have been previously examined in Ref. \\cite{SG01}, although it was not realized there that the maximum upper shift is $\\b=2$, beyond which the process does not occur at all. The $\\b=2$ contour (\\ref{n2contour}) is a line of unit slope and $\\widetilde{\\xi}$--intercept $-1$ in the scaled parameters, hence unit slope and $\\xi$--intercept $-\\o_0^2/m^2\\sim -10^{-15}$. As long as the 25 meV photons annihilate at least with 20 TeV photons (whose normal threshold is 12.5 meV), the parameters must lie above this line.  \\paragraph{$n=3$ Photon annihilation constraints.} For $n=3$ the contours of constant threshold in the scaled parameters $\\widetilde{\\eta}$ and $\\widetilde{\\xi}$ are shown in Fig.~\\ref{fig:ggstruct}. The process does not occur for parameters below a broken line consisting of the diagonal up to $\\widetilde{\\eta}=\\eta \\times m^4/\\o_0^3=-32/27$, and the line of slope $1/2$ for greater $\\widetilde{\\eta}$. If absorption at $\\o_0$ is occurring for {\\it any} hard gamma ray, the parameters must lie above this broken line, so in particular everything on and below the diagonal is excluded for $\\widetilde{\\eta}<-32/27$.  For $\\o_0=25$ meV this corresponds to $\\eta<-2.3\\cdot 32/27\\approx -2.7$. This is important, since it is a strong constraint excluding most of the diagonal, which has been preferred by some researchers~\\cite{ACP, Aloisio:2000cm}. It is likely that a much stronger constraint holds however, restricting the lower threshold at 25 meV to be not more than some number of order unity times its usual value. We have indicated in Fig.~\\ref{fig:gg-n3-ph} the form of the region below the no-shift contour and above the shift-less-than-$\\b$ contour for $\\b$ equal to 10, 5, 2 and 1.5. A stronger constraint would not exclude more of the diagonal, but it has the potential to chop off the infinite wedge of Figure~\\ref{fig:cergdec} at around the same place it excludes the diagonal. \\begin{figure}[htb] \\vbox{ \\vskip 8 pt \\centerline{\\includegraphics[width=2.4in]{gg-n3-ph}} \\caption{\\label{fig:gg-n3-ph} The unfilled region indicates parameters allowed if the lower threshold for a soft photon of 25 meV is $(a)$ not shifted down and $(b)$ not shifted up by more than 1.5, 2, 5, 10, and infinity. The upper line is the no-shift contour. No curvature due to the asymmetric solution is visible for this line because the junction point as defined in Eq.~(\\ref{kstsy}) is at $\\eta=-20$. The line for the existence of a lower threshold is the lowest line. It is coincident with the symmetric $\\b=1.5$ line below the diagonal and with the (dashed) diagonal below the crossing point. The curves stemming from the $\\b=1.5$ contour are the asymmetric contours for $\\b=10,5,2$, with lower values of $\\b$ corresponding to the curves with less slope. \\smallskip} } \\end{figure}  \\subsection{QED processes without thresholds} \\label{sec:qednoth}  We now consider two QED effects that occur in the presence of Lorentz violation without any threshold, velocity dispersion of photons in vacuo and photon splitting. The former will eventually provide competitive constraints on $\\eta$ and $\\xi$ respectively, but the latter has too slow a rate to be important.  \\subsubsection{Velocity dispersion of photons} \\label{sec:veldisp}  Gamma-ray bursts (GRB's) are explosive extragalactic events that release a large number of high energy photons with a flux that varies rapidly in time. It was therefore realized~\\cite{Acea,Ellis:1999sd} that they can provide interesting constraints or possible observations of Planck scale suppressed Lorentz violation in the dispersion relation for photons (a possibility noted long ago in~\\cite{Pav}).  The reason is that while propagating over such a long distance even tiny differences in group velocity could produce detectable time differences between the arrival at Earth of photons of different energy.  For photons with Lorentz breaking dispersion relations of order $n$, $\\xi$ is related to the fractional variation in group velocity by \\begin{equation} \\xi=\\frac {2} {n-1} \\frac {M^{n-2}} {k_1^{n-2} - k_2^{n-2}}\\frac{\\Delta c}{c}. \\end{equation} An upper limit on the difference in arrival times of photons from the same event provides an upper limit on the relative speed difference, if one assumes there is no conspiracy of different emission times cancelling different propagation times. Together with the energies of the different photons, such observations provide a constraint on $|\\xi|$.  The strongest constraint available today comes from GRB 930131~\\footnote{ Sarkar~\\cite{Sarkar:2002mg} has criticized the use of this particular gama ray burst since this object has no measured redshift, and hence an uncertain distance. Other bursts~\\cite{Ellis:1999sd} or blazar flares~\\cite{Biller} give somewhat weaker constraints. }, a gamma ray burst at a distance of 260 Mpc that emitted gamma rays from 50 keV to 80 MeV on a timescale of milliseconds~\\cite{sommer}. Schaefer~\\cite{schaefer} finds the upper limit $\\Delta c/c<9.6\\cdot 10^{-19}$ for photons of energy $k_1=78.6$~MeV, and $k_2=30$~keV. This yields the constraint $|\\xi|<122$ for $n=3$.  This is weaker than the constraint we have from photon annihilation, hence time of flight data do not at present strengthen our constraints for $n=3$. For $n=4$ dispersion the bound on $|\\xi|$ is on the order of $|\\xi|<10^{18}$, so we get no interesting constraint for $n>3$.  The situation for $n=3$ will be significantly improved in the future thanks to GLAST, the gamma ray large area space telescope, which should be able to set limits of order unity on $\\xi$~\\cite{Norris:1999nh}.  \\subsubsection{Photon Splitting}  The photon splitting processes $\\gamma \\rightarrow 2 \\gamma$ and $\\gamma \\rightarrow 3 \\gamma$, etc.\\ do not occur in standard QED.  Although there are corresponding Feynman diagrams (the triangle and box diagrams), their amplitudes vanish.  In the presence of Lorentz violation these processes are generally allowed when $\\xi>0$. However, the effectiveness of this reaction in providing constraints depends heavily on the decay rate. We now give an estimate of this rate, independent of the particular form of the Lorentz violating theory, which indicates that the rate involves at least four Lorentz violating factors, so is apparently too small to be relevant at observed photon energies.  We carry out the analysis allowing for any terms in the amplitude consistent with gauge and translation invariance.  The particular form of Lorentz violation considered in this paper also preserves rotation invariance in a preferred frame, however the following argument will not use that condition. Since gauge invariance is preserved, the amplitude for the process $\\gamma \\rightarrow N \\gamma$ should arise from a term that is a scalar formed from $N$ factors of the electromagnetic field strength $F_{ab}$ corresponding to the external photon legs. For each photon, $F^{({\\rm s})}_{ab}\\sim k_{[a}\\epsilon_{b]}$, where $k_a$ is the 4-momentum and $\\epsilon_b$ is the polarization vector.  In the Lorentz invariant case the equations of motion imply that $k_a$ is a null vector and $k_a\\epsilon^a=0$. Energy-momentum conservation then implies that these 4-momenta are all parallel, so being null they are orthogonal to each other and to all the polarization vectors. The rate thus vanishes for two different reasons. First, since the momenta are necessarily  all parallel, the phase space has vanishing volume. Second, the rate must be a scalar formed by contracting these four field strengths using only the metric. Any such contraction vanishes since it must involve contractions of the momenta with each other or with the polarizations. Hence the amplitude vanishes. In the case of an odd number of photons, another reason for  vansihing of the amplitude is Furry's theorem, which states that the sum over loops with an odd number of electron propagators vanishes.  If there is Lorentz violation then none of the above reasons for a vanishing rate apply.  First of all the $N$-odd amplitudes are no more guaranteed to vanish. Indeed for sufficiently general implementations of Lorentz violation the Furry theorem can be violated (see e.g.~the discussion of the Furry theorem and its violation in the extended QED~\\cite{Kostelecky:2001jc}). Secondly, the contractions of the field strengths might involve not just the metric but also a Lorentz violating tensor (for example $u^a u^b$ in the rotation invariant case, where $u^a$ is the unit timelike vector specifying the preferred frame.) Finally,  in the presence of Lorentz violation the photon four-momenta are in general not null vectors hence they need not be parallel and they need not vanish upon contraction. (To satisfy energy-momentum conservation $\\xi$ must be positive.)  In order for the phase space to not have vanishing volume, at least one of the 4-momenta must involve a Lorentz-violating factor $\\d=\\xi (k/M)^{n-2}$. This is not enough for the amplitude to not vanish however. For $\\g\\rightarrow N\\g$ with $N=3$ or 4 the contraction of the 3 or 4 field strength tensors $F^{({\\rm s})}_{ab}\\sim k_{[a}\\epsilon_{b]}$ using only the metric involves at least two vanishing contractions, and for larger $N$ there are more. One of those vanishing contractions can be rendered nonzero by the single Lorentz violating factor already invoked on an external photon momentum, but the other one requires either another such factor, or a Lorentz violating tensor in the operator whose matrix element is being computed. Such a tensor comes with some coefficient with dimensions determined by the dimension of the operator. We also use the symbol $\\d$ to indicate this sort of Lorentz-violating factor.  The possible contributions to the amplitude  will therefore be suppressed by at least two factors of $\\d$. The rate goes like the square of the amplitude, hence we infer that at energies well above the electron mass the decay rate must behave as $E\\d^4$ or slower, where $E$ is the initial photon energy. (There is an additional factor of $\\a^N$ if we consider standard QED diagrams for which each external photon leg comes with a factor of the electric charge in the amplitude.)  The lifetime is therefore at least of order $\\d^{-4}E^{-1}$, which for a photon of energy 50 TeV is $10^{-29}\\d^{-4}$ seconds. Such 50 TeV photons arrive from the Crab nebula, about $10^{13}$ seconds away, so the best constraint (i.e. if there is is no further small parameter such as $\\a^N$ or $1/16\\pi^2$ in the decay rate) we could possibly get on $\\d$ from photon splitting is $\\d\\lesssim 10^{-10}$. For $n=2$ this is not competitive with the other constraints already obtained. For higher $n$, each contribution arising from an operator of dimension greater than four will be suppressed by at least one inverse power of the scale $M$. For example, the contributions from $n=3$ deformations to the dispersion relation will yield $\\d\\sim\\xi E/M$. In this case the strongest conceivable constraint on $\\xi$ would be of order $\\xi\\lesssim 10^4$, and even this is not competitive with the other constraints we have found.  \\subsection{Combined Constraints}  Having completed our discussion of photon--electron processes we now turn to the determination of the global constraints that can be derived from the combination of all the above results. The photon splitting and the time of flight constraints are not as strict as those determined by the other considered interactions, at least for quadratic and cubic deformations, although in the future time of flight constraints may become competitive.  \\subsubsection{n=2}  In the case of quadratic deviations only the difference $\\xi-\\eta$ is constrained. The vacuum \\v{C}erenkov effect yields $\\xi-\\eta>-10^{-17}$, while photon decay provides the constraint $\\xi-\\eta<10^{-16}$.  Together these confine $\\xi-\\eta$ to a small neighborhood of zero. The photon annihilation ``likelihood region'' would just impose $\\xi-\\eta\\lesssim 10^{-15}$, which does not further strengthen the constraint.  \\subsubsection{n=3}  Putting together the constraints from the three photon--electron interactions previously considered we obtain a remarkably small allowed region in the $\\eta$--$\\xi$ plane (see Figure~\\ref{fig:all}). \\begin{figure}[htb] \\vbox{ \\vskip 8 pt \\centerline{\\includegraphics[width=2.7in]{total}} \\caption{Combined constraints on the photon and electron parameters, for the case $n=3$. The regions excluded by the photon decay and \\^{C}erenkov constraints are lined horizontally in blue and vertically in red respectively.  The region between the two diagonal green lines is where the threshold for the annihilation of a gamma ray with a 25 meV photon ranges from its standard value (upper diagonal green line) to not more than twice that value. The shaded patch is the part of the allowed region that falls between these gamma annihilation thresholds. The dashed line is $\\xi=\\eta$.~\\label{fig:all} \\smallskip} } \\end{figure} The photon decay and \\v{C}erenkov constraints exclude the horizontally and vertically filled regions respectively. The allowed region lies in the lower left quadrant, except for an exceedingly small sliver near the origin with $0<\\eta\\lsim 10^{-3}$ and a small triangular region ($-0.16\\lsim\\eta<0$, $0<\\xi\\lsim 0.08$) in the upper left quadrant. The discussion of the photon annihilation threshold in subsection~\\ref{sec:photann} indicates that, although no firm constraint can be given at present, the allowed region cannot lie too far from the corridor between the two roughly parallel diagonal lines. These lines indicate where the threshold for the annihilation of a gamma ray with a 25 meV photon ranges from its standard value (upper diagonal green line) to not more than twice that value.  If future observations of the blazar fluxes and the IR background yield agreement with standard Lorentz invariant kinematics, the region allowed by the photon annihilation constraint will be squeezed toward the upper line ($k_{\\rm th}\\approx k_{\\rm s}$).  Time of flight constraints for high energy photons currently constrain $\\xi$ to be less than $\\sim 100$ at best, but future observations should allow such constraints to further narrow the allowed region towards the origin.  \\subsubsection{n=4}  The case of quartic deviations is unfortunately just mildly constrained from the available observations. The order of magnitude allowed for the parameters is as small as $10^{11}$ (from \\v{C}erenkov) for the electron--photon vertex interactions.   \\label{sec:disc}  In this paper we have performed a systematic analysis of the effects of Lorentz violating dispersion on particle reactions, allowing for unequal deformation parameters for different particles.  We have analyzed the threshold kinematics and combined the observational constraints where possible.  Even when suppressed by the inverse Planck mass, such Lorentz violation can lead to radically new behavior in the kinematics of particle interactions at much lower energies. Reactions previously forbidden can be allowed, lower thresholds can be shifted and upper thresholds can be introduced.  The presence of upper thresholds is a feature of Lorentz breaking physics that is not present in Lorentz invariant physics and which can be relevant for observational constraints.\\footnote{ More complex dispersion relations can lead to multiple thresholds~\\cite{JLMth} with could have further observational effects.} Furthermore, we have found that for interactions with identical final particles, the final momenta can be distributed asymmetrically at threshold. While this is a straightforward consequence of the kinematics, it has been previously overlooked in the literature, probably because it is alien to Lorentz invariant physics.  Using these kinematical results, we have seen that a conservative interpretation of observations puts strong constraints on the coefficients $\\eta$ and $\\xi$ of order $E/M_{\\rm P}$ modifications to the electron and photon dispersion relations. The allowed region includes $\\xi=\\eta=-1$, which has been a focus of previous work~\\cite{ACP,Kifune:1999ex,Kluzniak}. The negative quadrant has most of the allowed parameter range.  Note that in this quadrant all group velocities are less than the low energy speed of light.  For modifications of order $(E/M_{\\rm P})^2$ there are no significant constraints in the electron-photon sector derivable from current observations, due to the fact that the energies of observed particles are too low.  However reactions such as proton \\v{C}erenkov (in vacuo) or pion production by cosmic rays, for which we have data at much higher energies, can provide good constraints for $(E/M_{\\rm P})^2$ modifications (although for different particle deformation parameters).  Ultra high energy cosmological neutrinos may also provide good \\v{C}erenkov constraints at this or even higher orders, since the neutrino mass is much smaller than that of any other particle. The interaction amplitudes are very suppressed however, so it is necessary to accurately calculate the rate and compare it with the travel time of the neutrino.  There are a number of ways to improve the constraints on $O(E/M_{\\rm P})$ modifications from electron-photon interactions.  Higher energy electrons would not help much since the \\v{C}erenkov constraint is already strong, while finding higher energy undecayed photons would squeeze the allowed region onto the line $\\xi=\\eta$ of Figure~\\ref{fig:all}. To further shrink the allowed segment of this line would require improved knowledge of the infrared background and a reconstruction of the source spectrum from the observed gamma rays in the presence of Lorentz violation.  Also, the constraint from time of flight measurement may become competitive using improved detectors.  Other constraints may be provided by additional interactions not considered here. For example, a possible upper threshold for $e^+\\, e^- \\rightarrow 2\\g$ cannot provide a competitive constraint in astrophysical observations since there are other processes by which observed high energy photons can be produced.  However, if future electron accelerators can reach energies above $10$ Tev then one can expect to get a good constraint from this reaction.  In addition there may be other reactions for which upper thresholds can produce useful constraints at or near currently observed energies. Reactions involving more than two types of particles, such as $\\nu \\rightarrow e^- \\, W^+$, could also give constraints. It may be possible that by considering a number of such reactions a multi-dimensional parameter space can be usefully constrained.  The idea motivating our work is that Lorentz violation may be a consequence of quantum gravity, in which case the natural scale for the Lorentz violation is the Planck scale. If, as in braneworld scenarios, the quantum gravity scale were to be around a TeV, then the natural scale for Lorentz violation induced by quantum gravity would be the TeV scale. Clearly, the only way such Lorentz violation could be compatible with observations is if it were extremely suppressed compared with this natural scale. This suggests that either TeV scale quantum gravity is wrong, or it does not violate 4d-Lorentz invariance.  In conclusion, the {\\it absence} of anomalous observations provides stringent constraints on the possibility of Lorentz violation originating at the Planck scale.  This in turn gives important information as to the viability of quantum gravity theories that predict 4-d Lorentz violation.  We can expect that, as better data at higher energies becomes available, even stronger constraints will be imposed or, alternatively, positive signatures of Lorentz violation may be found. Either way, it is clear that a useful tool for the phenomenological investigation of quantum gravity is now at hand."
    },
    "0209/astro-ph0209214.txt": {
        "abstract": "Non-axisymmetric forces are presented for a sample of 107 spiral galaxies, of which 31 are barred (SB) and 53 show nuclear activity. As a database we use JHK images from the 2 Micron All Sky Survey, and the non-axisymmetries are characterized by the ratio of the tangential force to the mean axisymmetric radial force field, following Buta $\\&$ Block (2001). Bar strengths have an important role in many extra-galactic problems and therefore it is important to verify that the different numerical methods applied for calculating the forces give mutually consistent results. We apply both direct Cartesian integration and a polar grid integration utilizing a limited number of azimuthal Fourier components of density. We find that bar strength is independent of the method used to evaluate the gravitational potential. However, because of the distance dependent smoothing by Fourier decomposition the polar method is more suitable for weak and noisy images. The largest source of uncertainty in the derived bar strength appears to be the uncertainty in the vertical scale-height, which is difficult to measure directly for most galaxies. On the other hand, the derived bar stength is rather insensitive to the possible gradient in the vertical scale-height of the disk or to the exact model of the vertical density distribution, provided that a same effective vertical dispersion is assumed in all models. In comparison to the pioneering study by Buta $\\&$ Block (2001), bar strength estimate is here improved by taking into account the dependence of the vertical scale-height on the Hubble type: we find that for thin disks bar strengths are stronger than for thick disks by an amount which may correspond to even one bar strength class.  We confirm the previous result by Buta $\\&$ Block (2001) and Block et al. (2001) showing that the dispersion in bar strength is large among all the de Vaucouleurs's optical bar classes. In the near-IR 40 $\\%$ of the galaxies in our sample have bars (showing constant phases in the m=2 Fourier amplitudes in the bar region), while in the optical 1/3 of these bars are obscured by dust. Significant non-axisymmetric forces can be induced also by the spiral arms, generally in the outer parts of the galactic disks, which may have important implications on galaxy evolution. Possible biases of the selected sample are also studied: we find that the number of identified bars rapidly drops when the inclination of the galactic disk is larger than $50^0$. A similar bias is found in The Third Reference Catalogue of Bright Galaxies, which might be of interest when comparing bar frequencies at high and low redshifts.   \\keywords {galaxies: spiral  -- galaxies: active -- galaxies: statistics -- galaxies: kinematics and dynamics} ",
        "introduction": "Bars consist mostly of old stellar population (de Vaucouleurs 1955; Elmegreen $\\&$ Elmegreen 1985), which stresses their significance as dynamically important components in galaxies. In fact, a large fraction of galaxies have bars (Block $\\&$ Wainscoat 1991; Knapen et al. 2000; Eskridge et al. 2000, Block et al. 2001), indicating that they must be long-lived phenomena in galaxies. Bars are fundamental in galaxy evolution, suggested to be driving forces for star formation, formation of rings and global spiral density waves, and even for the onset of nuclear activity. When quantified the correlations between bar strength and the other properties of the galaxies can be studied. The wavelength that best traces the dynamical mass is the near-IR, where the obscuration of dust is also less significant than in the optical region. For example, galaxies like NGC 5195, which are irregular in the optical may have regular grand-design spiral arms in the near-IR (Block et al. 1994), which emphasizes the importance of a new, more dynamical picture of the morphological structure in galaxies. A step toward that direction is the new dust penetrated classification of spiral arms in the near-IR (Block $\\&$ Puerari 1999; Buta $\\&$ Block 2001; Block et al. 2001), in which bar strength plays an important role.   As discussed by Buta $\\&$ Block (2001; BB from hereon) there are many quantitative parameters which can be used to estimate bar strengths, such as bar-interarm contrast (Elmegreen $\\&$ Elmegreen 1985) or light remaining after the disk and bulge components are subtracted (Seigar $\\&$ James 1998). The most commonly used method is the maximum ellipticity of a bar, an approach justified by the analytical models by Athanassoula (1992), who showed that the non-axisymmetric forces in the bar correlate with the bar ellipticity. This method has been recently refined by Abraham $\\&$ Merrifield (2000), who consider both the inner and outer contours of the image to better resolve the ellipticity of a bar. However, the ellipticity is not a full description of bar strength. In fact, a more physical approach has been taken by BB who estimate bar torques by calculating tangential forces in the bar region, taking into account also the underlying axisymmetric potential. Indeed, when refined, the bar torque method is probably the most promising way of estimating bar strengths.  When the bar torque method is finely tuned, future refinements will include the complex bar structures seen in many galaxies; taking more properly into account the vertical scale heights and their gradients, as well as taking into account bulge stretch scenarios upon deprojection of the images. In the bar torque method there are also different ways of evaluating the gravitational potential and it is important to verify that the different methods give mutually consistent results. For example, BB used the 2D Cartesian integration method by Quillen et al. (1994), whose new contribution in the potential evaluation was that the vertical density profile of the disk was taken into account in the convolution function. The potential was calculated in Cartesian grid by applying Fast Fourier Transform techniques (see also Elmegreen et al. 1989). On the other hand, in our study of IC 4214 (Salo et al. 1999) we evaluated the barred potential by first ``smoothing'' the image by calculating the Fourier decomposition of the surface density in a polar grid. In principle these two methods should give similar results.  In the present study bar strengths are calculated in JHK-bands for 107 spiral galaxies using the polar method (Salo et al. 1999) for the evaluation of the gravitational potential. The method has been improved  by taking into account the recent observational work showing that bars in early-type galaxies are thicker than in late-type galaxies. Also, the effects of a distance dependent scale height, detected in many boxy/peanut shaped disks, are estimated. The algorithm of calculating forces is described, the different ways of estimating the gravitational potential are compared and Fourier analysis is applied for the analysis of bars. Also, biases of the sample are studied and the distributions of bar torques among the de Vaucouleurs's optical bar classes are compared. In future the method will be further developed to better take into account the observational properties of bars and bulges in galaxies. The measurements of this paper have been used in comparisons of bar strengths between active and non-active galaxies by Laurikainen et al. (2002). ",
        "conclusions": "Non-axisymmetric forces are calculated for 107 spiral galaxies in J,H and K bands using a method where gravitational potentials are evaluated in a polar grid. Non-softened convolution function is applied and the vertical distribution of matter is approximated by an exponential function. The vertical scale-height of the disk is taken to be a certain fraction of the radial scale-length of the disk, and this ratio is assumed to be larger for the early than for the late-type galaxies. The M/L-ratio is assumed to be constant throughout the disk. The vertical mass distribution is generally assumed to obey the same formula everywhere in the galaxy, but tests were also performed to estimate the effect of radially non-constant vertical thickness. The phases of the Fourier density amplitudes are used to estimate the lengths of the bars. One of the main concerns of this study is to verify that the different methods of calculating the gravitational potential give mutually consistent results, most of the tests being carried out using a high quality H-image of NGC 1433 (Buta et al. 2001). In comparison to BB our method is more suitable for weak and noisy images. Also, it is possible to limit only to even Fourier decompositions, most likely to characterize the non-axisymmetry related to the bar. Likewise in polar method it is easy to study distant-dependent $h_z$. The isophotal ellipticities of bars are also estimated, to facilitate comparisons to bar strengths estimated from maximal forces.  The main results are the following:  (1) Cartesian and polar grid methods for the potential evaluation are compared. In the Cartesian method the image is sampled to a density array and then 2D FFT in Cartesian coordinates is applied. In the polar grid method Fourier decomposition of density is calculated in a polar grid using FFT in azimuth and direct summation over radius. We found that similar results are obtained by these two methods for good quality images, provided that enough Fourier components (up to m=6) are included, and the resolution of the Cartesian grid is sufficiently large.  (2) Bar strength is found to be rather insensitive to the vertical mass model of the disk, as long as a same vertical dispersion is assumed for all models (e.g. $h_{sech2}/h_z = \\sqrt{24}/\\pi, $ $h_{sech}/h_z = \\sqrt{8}/\\pi$). Boxy/peanut shaped structures, in terms of non-constant vertical scale-heights along the disk, were also found to be quite unimportant for the evaluation of bar strengths. These parameters affect $Q_b$ less than 5$\\%$. The largest uncertainties in $Q_b$ are associated to the large scatter in the observed vertical scale-height of the disk within one Hubble type, and to the observed uncertainties in the orientation parameters of the disks, which both may induce uncertainties of about 10-15 $\\%$ in $Q_b$.  (3) Significant non-axisymmetric forces ($Q_b>$0.05) are detected in 80 $\\%$ of the galaxies in our sample. In most cases they were interpreted as bar-like features, based on significant m=2 Fourier amplitudes in the bar regions and distinct ``butterfly patterns'' in the $F_T/<F_R>$-ratio maps. In 40$\\%$ of the galaxies ``classical bars'' were detected, determined as having $A_2/A_0 > 0.3$ and the m=2 phases maintained nearly constant in the bar region. In some of the galaxies significant non-axisymmetric forces were detected in the outer parts of the disks connected to spiral arms, corresponding even to bar strength class 3.  (4) We confirm the previous result by BB and Block et al. (2001) showing the large overlap in bar strength between the optical SA, SAB and SB classes. Actually, SB-galaxy can belong to any of the bar strength classes between 1 and 6.  (5) We found that 95 $\\%$ of SB galaxies in our sample belong to the ``classical bars'' identified in the near-IR, which means that the bars are similar. In the optical 1/3 of the ``classical bars'' are not classified as SB. Even bars which are in the optical obscured by dust and which become dust penetrated in the near-IR, cover all bar classes from 1 to 6, thus indicating that even strong bars can be obscured by dust.  (6) Bar lengths are estimated from the phases of the m=2 and m=4 Fourier components of density requiring that the phase is maintained nearly constant in the bar region. Bar length is found to correlate with the galaxy brightness $M_B$, confirming the previous result by Kormendy (1979) in the optical region.  (7) The number of SB-galaxies in our sample drops rapidly at inclinations $> 50^0$. A similar bias appears also in RC3 when limiting to galaxies brighter than 15.5 mag. This bias might have important implications while studying the frequencies of bars at low and high redshifts, especially because the bias increases toward fainter galaxies. Also, at high inclinations the number of SAB galaxies in RC3 drops much more rapidly than the number of SB-galaxies, thus being a manifestation of an ambiguous definition of the de Vaucouleurs's SAB class."
    },
    "0209/hep-ph0209322_arXiv.txt": {
        "abstract": "A fluid interpretation of Cardassian expansion is developed. Here, the Friedmann equation takes the form $H^2 = g(\\rho_M)$ where $\\rho_M$ contains only matter and radiation (no vacuum).  The function $g(\\rhom)$ returns to the usual $8\\pi\\rhom/(3 m_{pl}^2)$ during the early history of the universe, but takes a different form that drives an accelerated expansion after a redshift $z \\sim 1$.  One possible interpretation of this function (and of the right hand side of Einstein's equations) is that it describes a fluid with total energy density $\\rho_{tot} = {3 m_{pl}^2 \\over 8 \\pi} g(\\rhom) = \\rhom + \\rho_K$ containing not only matter density (mass times number density) but also interaction terms $\\rho_K$. These interaction terms give rise to an effective negative pressure which drives cosmological acceleration.  These interactions may be due to interacting dark matter, e.g. with a fifth force between particles $F \\sim r^{\\alpha -1}$. Such interactions may be intrinsically four dimensional or may result from higher dimensional physics.  A fully relativistic fluid model is developed here, with conservation of energy, momentum, and particle number.  A modified Poisson's equation is derived.  A study of fluctuations in the early universe is presented, although a fully relativistic treatment of the perturbations including gauge choice is as yet incomplete. ",
        "introduction": "Recent observations of Type IA Supernovae \\cite{SN1,SN2}, as well as concordance with other observations, including the microwave background \\cite{boom} and galaxy power spectra \\cite{2df}, indicate that the universe is flat and accelerating.  Many authors have explored possible explanations for the acceleration: a cosmological constant, time-dependent vacuum energy such as quintessence \\cite{fafm,peebrat,frieman,stein,caldwell,huey}, and gravitational leakage into extra dimensions \\cite{ddg}.  Recently, Freese and Lewis \\cite{freeselewis} (Paper I) proposed an explanation for the acceleration which involves only matter and radiation, invoking no vacuum energy or cosmological constant whatsoever.  In their model, called Cardassian, the universe has a flat geometry as required by measurements of the cosmic background radiation \\cite{boom} and yet consists only of matter and radiation.  The Friedmann equation is modified from its usual form, $H^2 = {8 \\pi \\over 3 m_{pl}^2} \\rho_M$, to \\begin{equation} \\label{eq:newfunc} H^2 = g(\\rho_M) , \\end{equation} where $H = \\dot a / a$ is the Hubble constant (as a function of time), $a$ is the scale factor of the universe, and the energy density $\\rhom$ contains only ordinary matter and radiation.  The function $g(\\rhom)$ reduces to ${8\\pi \\over 3 m_{pl}^2}\\rhom$ in the early universe, so that Eq.~(\\ref{eq:newfunc}) reduces to the ordinary Friedmann equation during early epochs including primordial nucleosynthesis.  Only at redshifts $z<{\\cal O}(1)$ does the function $g(\\rhom)$ differ from the ordinary Friedmann Robertson Walker (FRW) case; during these late epochs, $g(\\rhom)$ gives rise to accelerated expansion. In Paper I, the specific form of $g(\\rho)$ that was considered was Power Law Cardassian, \\begin{equation} \\label{eq:new} H^2 = {8\\pi\\over 3 m_{pl}^2} \\rhom + B \\rhom^n , \\end{equation} with \\begin{equation} n< 2/3. \\end{equation} The second term only becomes important once $z<{\\cal O}(1)$, at which point it dominates the equation and causes the universe to accelerate.  Other possible functions $g(\\rho)$ \\cite{freese} are discussed further below.  There remains the question of the fundamental origin of these modifications to the Friedmann equation.  There is no unique four-dimensional or even higher-dimensional theory that gives Cardassian evolution.  We consider two different motivations for these modifications: \\hfill\\break 1) These functions may arise from fundamental theories of gravity in higher dimensions, as was discussed in \\cite{freeselewis}.  Chung and Freese \\cite{chung} showed that, generically, the Friedmann equations are modified as a consequence of embedding our universe as a three-dimensional surface (3-brane) in higher dimensions. \\hfill\\break 2) Alternatively these functions may arise in a purely four-dimensional theory in which the modified right hand side of the Friedmann equation is due to an extra contribution to the total energy density.  The right hand side is treated as a single fluid, with an extra contribution to the energy-momentum tensor in (ordinary four dimensional) Einstein's equations. \\hfill\\break The two motivations may or may not be linked, in that the fluid interpretation may be intrinsically four-dimensional, or it may be an effective description of higher dimensional physics.  In this paper, we restrict our discussion to four dimensions, and treat the right hand side of Einstein's equations as a single fluid. We consider models with an extra energy density associated with matter that contributes in such a way as to drive acceleration.  This extra energy density may be intrinsically four dimensional or may serve as an effective description of higher dimensional physics.  We take the total energy density of the matter \\begin{equation} \\rho_{tot} = {3 m_{pl}^2 \\over 8\\pi} g(\\rho) = \\rho_M + \\rho_K \\end{equation} (plus possible internal thermal energy which is unimportant on cosmological scales) to contain not only the ordinary mass density $\\rho_M$ (mass times number density) but also an additional contribution $\\rho_K$. For example, in Eq.~(\\ref{eq:new}), \\begin{equation} \\rho_K = \\frac{3 m_{pl}^2}{8\\pi} B \\rhom^n. \\end{equation}  Given this total energy density, we can now compute the accompanying pressure, and find that the Cardassian contribution has a {\\it negative pressure}, $p_K<0$.  This negative pressure is responsible for the universe's acceleration. In fact, one can obtain any negative equation of state $w_K = p_K/\\rho_K <0$, including $w_K<-1$.  The fluid approach has several advantages: (i) it is fully relativistic, (ii) it allows for the conservation of energy and momentum as well as of particle number, (iii) it admits a sensible weak-field limit which leads to a modified Poisson's equation, and (iv) it permits the study of fluctuations in the early universe, the study of effects on the cosmic microwave background anisotropies, and other observables.  The primary purpose of this paper is to examine this fluid approach. However, we briefly speculate on a possible origin for this extra term $\\rho_K$ in the energy density.  It may arise from (dark) matter self-interactions that contribute a negative pressure, for example through a long-range confining force which may be of gravitational origin or may be a fifth force.  This self-interacting dark matter is different from any such component considered in the past, in that is has a negative rather than a positive pressure.  We speculate on a form of the force between particles that may be responsible for such an interaction, $F \\sim r^{\\alpha-1}$, although this Newtonian form must of course be modified on horizon scales. This description of a self-interacting dark fluid may be an effective description of a more fundamental theory.  The fluid approach does not rely on the validity of such an interpretation of self-interacting dark matter, e.g., the interactions  may be an effective description of higher dimensional physics.  We begin by reviewing the idea of Cardassian expansion in Sect.~\\ref{sec:cardass}.  We present a general fluid formulation in Sect.~\\ref{sec:basics}, and then give specific examples in Sect.~\\ref{sec:examples}. In Sect.~\\ref{sec:perturbations} we address the growth of density perturbations, and in Sect.~\\ref{sec:confine} we speculate on the possible origin of an interaction energy with negative pressure. ",
        "conclusions": "An interpretation of Cardassian expansion as an interacting dark matter fluid with negative pressure is developed. The Cardassian term on the right hand side of the Friedmann equation (and of Einstein's equations) is interpreted as an interaction term. So the total energy density contains not only the matter density (mass times number density) but also interaction terms. These interaction terms give rise to an effective negative pressure which drives cosmological acceleration.  These interactions may be due to interacting dark matter, e.g. with a long-range confining force or a fifth force between particles. Alternatively, such interactions may be an effective description of higher dimensional physics.  We have said that matter alone can be responsible for accelerated behavior.  However, if the Cardassian behavior results from integrating out extra dimensions, then one may ask what behavior of the radii of the extra dimensions is required.  Similarly, if we follow a QCD bag or other description of self-interacting dark matter, one may wonder if an equivalent vacuum description can be constructed.  Further work in search of a fundamental origin of Cardassian expansion must be studied to answer these questions in detail.  A fully relativistic fluid model of Cardassian expansion has been developed, in which energy, momentum, and particle number are conserved, the modified Poisson's equations have been derived, and a preliminary study of density fluctuations in the early universe has been presented.  One of the goals of this study is to allow predictions of various observables that will serve as tests of the model.  The Cardassian model will have unique predictions, particularly due to the modified Poisson's equations. For example, one can now calculate the effect on the Integrated Sachs Wolfe effect in the Cosmic Microwave Background \\cite{lef}.  In addition, one can now calculate the effect on cluster abundances as function of redshift. These predictions can then be tested against existing and upcoming measurements of these quantities. Comparison with existing and upcoming supernova data is being studied in another paper \\cite{wang}.  We reiterate that this fluid approach is only one of the ways that Cardassian expansion may result."
    },
    "0209/astro-ph0209617_arXiv.txt": {
        "abstract": "{ Using ISOPHOT maps at 100 and 200\\,$\\mu$m and raster scans at 100, 120, 150 and 200\\,$\\mu$m we have detected four unresolved far-infrared sources in the high latitude molecular cloud L~183. Two of the sources are identified with 1.3\\,mm continuum sources found by Ward-Thompson et~al.\\ (\\cite{wthompson94}, \\cite{wthompson00}) and are located near the temperature minimum and the coincident column density maximum of dust distribution.  For these two sources, the ISO observations have enabled us to derive temperatures ($\\sim 8.3$\\,K) and masses ($\\sim$1.4 and 2.4\\,M$_{\\sun}$). They are found to have masses greater than or comparable to their virial masses and are thus expected to undergo gravitational collapse. We classify them as pre-protostellar sources.  The two new sources are good candidates for pre-protostellar sources or protostars within L~183. ",
        "introduction": "Low-mass stars are known to form within dark clouds. To study the initial conditions for their formation, it is important to probe the physical conditions of both molecular gas and cold dust in the deep interiors of such clouds.  Dense cores have been identified inside dark clouds, some probably being in the stage of contraction to form a star. A pre-protostellar core is defined as the stage in which a gravitationally bound core has formed in a molecular cloud, but no central hydrostatic protostar exists yet within the core (see e.g.\\ Ward-Thompson et~al.\\ \\cite{wthompson94}).  Pre-protostellar cores are thought to be very cold.  Thus, in many cases they escaped detection by IRAS, limited to $\\lambda \\le 100\\,\\mu$m.  The long wavelength and multi-filter capabilities of ISOPHOT aboard the Infrared Space Observatory (ISO) (Kessler et~al.\\ \\cite{kessler96}), combined with its improved sensitivity and spatial resolution over IRAS, have been utilized in studying the far-IR emission of molecular clouds and pre-stellar and young embedded stellar objects within them.  Analysis of their physical parameters is facilitated by using the spectral energy distributions (SEDs) obtained from ISOPHOT multi-filter photometry.   \\subsection{Lynds 183}  The dark cloud/large globule L~183, frequently cited also as L~134 North, is a prototypical dense cold molecular cloud. Its visual extinction is estimated to be $\\sim 17^{\\mathrm m}$ based on its 200\\,$\\mu$m optical depth (Juvela et~al.\\ \\cite{juvela02}).  So far there is no evidence for associated newly born stars such as T~Tauri stars or IRAS point sources. However, Martin \\& Kun (\\cite{martinkun}) have found a bona fide T~Tauri star and H$\\alpha$ emission line star near L~183. These two stars are located outside our maps.  Given its short distance of $\\sim$~110 pc (Franco\\ \\cite{franco}) L~183 provides a good spatial resolution with the ISO FIR beam size (ISOPHOT's spatial resolution of 45$\\arcsec$ at 100\\,$\\mu$m corresponds to 0.02\\,pc at the distance of L~183). The location at high Galactic latitude ($b = 36$ deg) minimizes contamination with unrelated cirrus along the line of sight. The location off the Galactic plane at $z = 65$\\,pc implies that the impinging ultraviolet radiation field is strongly asymmetrical.  A Digitized Sky Survey\\footnote{The Digitized Sky Survey was produced at the Space Telescope Science Institute under U.S.\\ Government grant NAG W-2166. The images of these surveys are based on photographic data obtained using the Oschin Schmidt Telescope on Palomar Mountain and the UK Schmidt Telescope. The plates were processed into the present compressed digital form with the permission of these institutions} red plate ('Equatorial Red' survey (UK Schmidt) IIIaF + RG610) image of L~183 is shown in Fig.~\\ref{optIR}. In Fig.~\\ref{optIR}b, the original image has been smoothed with a FWHM=9\\arcsec (nine image pixels) Gaussian kernel.  The cloud has a dark core - bright rim structure, which is characteristic of clouds having high optical depth at visual wavelengths that are illuminated by the diffuse interstellar radiation field (Mattila \\cite{mattila74}; FitzGerald et~al.\\ \\cite{fitzgerald76}). The bright rim maximum occurs at a radius corresponding to extinction A$_{\\lambda} \\approx 1.5^{\\mathrm m}-2^{\\mathrm m}$.  Laureijs et~al.\\ (\\cite{laureijs91}) observed the 60\\,$\\mu$m emission to decline relative to 100\\,$\\mu$m emission in a narrow transition layer around the cloud edge. They explained this behaviour by assuming a separate grain component at 60\\,$\\mu$m that undergoes a modification of properties in a transition layer.  Laureijs et~al.\\ (\\cite{laureijs95}) have used IRAS images, CO molecular line observations and blue extinction values from star counts for a large-scale study of the L~134 cloud complex, including L~134, L~169 and L~183. Using $^{13}$CO observations, they found 18 clumps in the complex and derived their properties. The clumps follow clear size vs.\\ linewidth and luminosity vs.\\ size relationships.  An analysis and discussion of the dust and molecular gas properties in L~183 based on the ISOPHOT 100 and 200\\,$\\mu$m emission maps has recently been presented by Juvela et~al.\\ (\\cite{juvela02}).  L~183 has been a favourite target for molecular line observations which have revealed it as a rich source of molecular species (see e.g.\\ Swade \\cite{swade89a}, \\cite{swade89b}; Dickens et~al.\\ \\cite{dickens}).  Lee et~al.\\ (\\cite{lee01}) have classified L~183 as a strong infall candidate.  The optically thick CS($J$=2-1) lines in the inner region of the cloud show a double peak with the blue component brighter than the red one, which is characteristic of inward motions. Similar profiles suggesting infall motions have been observed for HCO$^+$($J$=3--2) (Gregersen \\& Evans \\cite{gregersen00}) and for CS($J$=2--1) by Snell et~al.\\ (\\cite{snell82}).  Mapping of the millimetre and submillimetre continuum dust emission by Ward-Thompson et~al.\\ (\\cite{wthompson94}, \\cite{wthompson99}) has revealed a small extended core in the center of L~183. This source has been detected at 800, 1100 and 1300\\,$\\mu$m, and has FWHM dimensions of $60\\arcsec \\times 40\\arcsec$ (0.032 $\\times$ 0.021\\,pc) at 800\\,$\\mu$m. According to Ward-Thompson et~al.\\ (\\cite{wthompson94}), the core is probably pre-protostellar in nature, i.e.\\ at an earlier stage than an accreting Class~0 protostar (for the definition of a Class~0 object see Andr\\'e et~al.\\ \\cite{andre93}).  The submillimetre emission from the core is consistent with the dust being heated externally by the general interstellar radiation field.  Recent observations by Ward-Thompson et~al.\\ (\\cite{wthompson00}) have revealed that the continuum emission extends further to the south, and that there is another emission maximum located at a position about 1.5$\\arcmin$ south of the previously detected maximum, with a FWHM size of $120\\arcsec \\times 60\\arcsec$. From sub-mm polarization observations Ward-Thompson et~al.\\ (\\cite{wthompson00}) measured a magnetic field direction which is at an angle of $34\\degr \\pm 6\\degr$ to the minor axis of the core, and found evidence for decreasing polarization at the highest continuum emission intensities.  Ward-Thompson et~al.\\ (\\cite{wthompson02a}) have mapped the core of L~183 at 90, 170 and 200\\,$\\mu$m with ISOPHOT. The core was not detected at 90\\,$\\mu$m wavelength. The dust temperature derived from 170 and 200\\,$\\mu$m data showed no temperature gradient across the core.  In this study, we search for pre- and protostellar objects in the cloud. Our maps are much bigger than previous (sub)millimetre continuum maps.  The derived fluxes of the sources between 120--200\\,$\\mu$m, located near the peak or in the Wien regime of the assumed blackbody radiation, are essential for temperature and thus mass determination of the objects.  Only if the masses of the objects are determined can we study their dynamical state.  Furthermore, we study the relation of young (proto)stellar objects with the properties of the underlying dust in the cloud such as temperature and column density.  In the context of L~183, the term 'pre-protostellar core' has been previously used to refer to the two known (sub)mm continuum sources within the cloud (Ward-Thompson et~al.\\ \\cite{wthompson94}, \\cite{wthompson00}), and also to the cloud core itself in which these sources are embedded (Ward-Thompson et~al.\\ \\cite{wthompson02a}). Throughout this article, we use the term 'source' to refer to objects in L~183 which are unresolved in our maps, such as the previously-known continuum sources. By the term 'core', we refer to the visually opaque cloud core which is larger than the sources within it and resolved by ISO observations. ",
        "conclusions": "The dark cloud L~183 has been mapped at far-infrared wavelengths with the ISOPHOT instrument aboard ISO. Four unresolved sources were detected in the cloud, called FIR~1, 2, 3 and 4, of which FIR~1 and 2 were previously known.  The SEDs of FIR~1 and 2 have been compiled by combining our ISO fluxes at 120, 150 and 200\\,$\\mu$m with longer wavelength fluxes from other studies. The SEDs are well fitted with a single modified blackbody with dust temperatures T$_{\\mathrm d} \\approx 8.3$\\,K. The total masses (gas plus dust) of the sources are $\\sim$1.4\\,M$_{\\sun}$ and $\\sim$2.4\\,M$_{\\sun}$, and the bolometric luminosities are very low, $\\sim$0.04\\,L$_{\\sun}$ and $\\sim$0.06\\,L$_{\\sun}$. Virial equilibrium consideration including magnetic, kinetic, and potential energy and external pressure shows that the sources have masses higher than or comparable to their virial masses.  FIR~3 is detected at 120, 150 and 200\\,$\\mu$m.  Its dust temperature is $\\sim$13\\,K and bolometric luminosity is $\\sim$0.03\\,L$_{\\sun}$. FIR~4 has been detected at 200\\,$\\mu$m only.  FIR~1 and 2 are probably gravitationally bound objects, rather pre-protostellar sources than Class~0 protostars. The available data do not provide definitive conclusions on the nature of the objects.  FIR~3 and 4 can be starless cold cores (either gravitationally bound or unbound), pre-protostellar cores (gravitationally bound) or protostars.  Virial equilibrium considerations cannot be made due to a lack of accurate density structure and size information. Further (sub)mm and centimetre continuum observations are required to study the nature of the sources."
    },
    "0209/astro-ph0209421_arXiv.txt": {
        "abstract": "Almost all of the $>600$ known Kuiper belt objects (KBOs) have been discovered within 50~AU of the Sun.  One possible explanation for the observed lack of KBOs beyond 50~AU is that the distant Kuiper belt is dynamically very cold, and thus thin enough on the sky to have slipped between previous deep survey fields.  We have completed a survey designed to search for a dynamically cold distant Kuiper belt near the invariable plane of the Solar system.  In $2.3~{\\rm deg}^2$ we have discovered a total of 33 KBOs and 1 Centaur, but no objects in circular orbits beyond 50~AU.  We find that we can exclude at 95\\% CL the existence of a distant disk inclined by $i\\le1\\arcdeg$ to the invariable plane and containing more than 1.2 times as many $D>185$~km KBOs between 50 and 60 AU as the observed inner Kuiper belt, if the distant disk is thinner than $\\sigma=1\\fdg75$. ",
        "introduction": "More than 500 Kuiper belt Objects (KBOs) have been discovered since the first, 1992 QB1, was detected almost a decade ago. Nearly all of these are currently within 50~AU of the sun, and only one has been detected beyond 54~AU, despite surveys sensitive enough to detect 160~km-diameter KBOs to 60~AU. Such surveys have placed strong upper limits on the density of KBOs in an outer Kuiper belt \\citep{ABM01, TLBE, JLT, Gl01}.  There are several possible explanations for the observed absence of distant KBOs, each with distinctly different predictions for the Kuiper belt beyond 50~AU, and distinctly different implications for the history of the solar system.  In the simplest view of the Kuiper Belt, the region beyond 50~AU, where the gravitational perturbations of the giant planets become negligible, has remained dynamically cold, with KBOs moving on their primordial circular and coplanar orbits. It is inferred that the region inside 48~AU has been heavily depopulated, by a factor of 10-100 in mass, over the age of the solar system as a result of planetary perturbations and mutual collisions \\citep{S96, SCb, Le93, Du95}. In the absence of such perturbations, the outer disk is expected to be a factor of 5--100 times as dense as the presently observed inner Kuiper belt.  However, as yet no discovered KBO has been determined to be in a circular orbit beyond 48~AU.  Shallow surveys, with typical magnitude limits of $R<23.5$, have searched for KBOs in more than $100\\,{\\rm deg}^2$ of sky, but objects at $\\ge50$~AU must be $\\ge280$~km in diameter to be detected in such surveys.  (In quoting sizes of KBOs in this paper, we assume a 4\\% albedo.)  It is desirable to search for smaller objects since the growth timescales for KBOs in the outer Kuiper belt, though highly uncertain, are likely longer than in the inner Kuiper belt.  Only a few square degrees, however, have been surveyed to magnitudes $R\\gtrsim24.5$.  \\citet{Ha00} has pointed out that if the outer disk were dynamically very cold, thus restricted to a thin line on the sky, it could have easily slipped between the previous sparse deep survey fields.  The most likely location for this distant cold disk would be the invariable plane, the plane normal to the total angular momentum of the solar system, which is inclined 1\\fdg6 to the ecliptic \\citep{Allen76}. Most deep KBO surveys, including our previously published results \\citep{ABM01}[ABM], have been conducted close to the ecliptic plane, possibly missing a thin disk near the invariable plane. We have therefore supplemented the ABM survey with four new fields that are strategically chosen on the sky to maximize the likelihood of detecting a thin trans-Neptunian disk near the invariable plane. We present here new limits on a putative cold disk beyond 50~AU that result from our observations in these new survey fields combined with previously published data. ",
        "conclusions": "We have provided strong limits on the existence of a distant cold disk. The simple expectation of a dense thin disk in the invariable plane appears to be ruled out. However, it is possible that a thin disk composed of objects smaller than $D=185$~km, undetectable in much of our survey, is present. This would require an extreme change in the size distribution of KBOs just beyond $\\sim$50~AU, as objects larger than $D\\approx1000$~km have been found at distances near 50~AU. It is not clear what would cause such a large and sudden change in the size distribution.  Other explanations for the observed lack of distant KBOs involve dropping the density of objects beyond 50~AU.  One possibility is that the outer Kuiper belt was dynamically excited by a stellar encounter early in the history of the solar system \\citep{Id00}.  An increase in the mean eccentricity and inclination of distant KBOs, thus lowering their apparent sky density and possibly halting accretion of large KBOs, would make a distant disk much harder to detect. Dynamical excitation by a stellar encounter results in a distinctive orbital distribution of KBOs, and could yield limits on the birth cluster environment of the solar system \\citep{AL01}.  It is also possible to explain the lack of KBOs beyond 50~AU if the primordial planetesimal disk ended at this distance. If Neptune and Uranus have migrated significantly over the age of the solar system \\citep{Ma93, Ma95, Th99}, the primordial solar nebula surface density beyond $\\sim30$~AU could have been very small initially. This could suggest that the larger KBOs actually formed interior to 30--40 AU and were displaced to their present orbits by a large scale rearrangement of orbits in the outer Solar system. Circumstellar disks around other stars have been observed to have diameters ranging between 50--1000~AU \\citep{McOd96, BrownD00}. A truncation of the solar nebula near 50~AU would fit easily within this range, with important implications for planetary formation models of our Solar System.  In order to distinguish between these theories and further constrain the distant KB population, deeper and wider survey observations of the Kuiper belt (to limiting magnitudes of $R>25$) will be necessary. Based on the limits we have found here, a survey covering approximately 20 square degrees near the invariable plane should be able to detect a distant disk dynamically excited to a thickness of 10\\arcdeg\\ with even half the number of objects as the CKB. This would constrain the density and distribution of the Kuiper belt at distances beyond 50~AU, including the Scattered Disk Objects, enabling a better comparison between our solar system and extrasolar systems."
    },
    "0209/astro-ph0209084_arXiv.txt": {
        "abstract": "We report $ISO$ SWS infrared spectroscopy of the H~II region Hubble~V in NGC 6822 and the blue compact dwarf galaxy I~Zw~36. Observations of Br$\\alpha$, [S~III] at 18.7 and 33.5$\\mu$m, and [S~IV] at 10.5$\\mu$m are used to determine ionic sulfur abundances in these H~II regions. There is relatively good agreement between our observations and predictions of S$^{+3}$ abundances based on photoionization calculations, although there is an offset in the sense that the models overpredict the S$^{+3}$ abundances. We emphasize a need for more observations of this type in order to place nebular sulfur abundance determinations on firmer ground. The S/O ratios derived using the $ISO$ observations in combination with optical data are consistent with values of S/O, derived from optical measurements of other metal-poor galaxies.  We present a new formalism for the simultaneous determination of the temperature, temperature fluctuations, and abundances in a nebula, given a mix of optical and infrared observed line ratios. The uncertainties in our ISO measurements and the lack of observations of [S~III] $\\lambda 9532$ or $\\lambda 9069$ do not allow an accurate determination of the amplitude of temperature fluctuations for Hubble~V and I~Zw~36. Finally, using synthetic data, we illustrate the diagnostic power and limitations of our new method. ",
        "introduction": "\\label{intro}  Because of their low metallicities \\citep{PE81, SKH89, IT99}, dwarf irregular and blue compact galaxies can provide valuable information for a wide variety of astrophysical problems. By comparing the low metal abundances found in dwarf galaxies to abundances found in luminous spirals, one can infer variations in star formation histories and chemical evolution. It is possible to use measurements of abundances in H~II regions in dwarf irregular galaxies to establish limits on yields from stellar and big bang nucleosynthesis \\citep{Pe92}.  It is also possible to characterize the ionizing radiation from OB stars from measurements of emission lines in H~II region spectra without resolving individual stellar spectra \\citep{VP88}.  Heavy elements such as C, N, O, Ne, S, and Ar are typically observed in H~II regions.  In order for accurate abundances to be determined, it is necessary to observe all of the ionization stages present in an H~II region, or to have a reliable method for inferring the contribution of unobserved ions.  Of the aforementioned elements, oxygen is the only one for which all important ionization stages can be easily observed at optical wavelengths. In the case of sulfur, the primary optical lines are [S~II] $\\lambda\\lambda$6717,6731 and [S~III] $\\lambda$6312. However, the [S~III] $\\lambda$6312 line is an extremely weak and temperature sensitive line, making it difficult to measure in many H~II regions. Because a significant fraction of the sulfur in an H~II region is in the ionization state S$^{+2}$ \\citep{G89}, accurate determination of S/H = N(S)/N(H) can be difficult based on optical measurements alone. The [S~III] $\\lambda\\lambda$9069,9532 lines in the near-infrared, which are intrinsically much stronger, often require a correction for atmospheric water absorption. In high-ionization nebulae S$^{+3}$, which emits only in the 10.5$\\mu$m line, can also become an important constituent.  Furthermore, the depth of particular ionization zones, temperature fluctuations ($t^2$), and other variations throughout a nebula can cause optical/UV forbidden line diagnostics to yield temperatures that are larger than ion-weighted average values in photoionization models, and they can weight emissivities toward values found in higher temperature regions of the nebulae \\citep{P67, G92, MTP98, Ee99}. These potential problems motivate the use of temperature insensitive mid- and far-infrared forbidden fine structure transitions in order to include optically unobserved ions, accurately determine nebular abundances, and to determine the amplitude and scale of temperature fluctuations inside H~II regions (e.g., Dinerstein, Lester \\& Werner 1985). The infrared portion of the spectrum contains strong emission lines such as [S~III] 18.7$\\mu$m and [S~IV] 10.5$\\mu$m. For faint extragalactic H II regions, the high background flux from earth's atmosphere precludes ground-based observations of many middle- and far-infrared emission lines at the present time. However, the low background and high sensitivity of the ISO observatory allowed observations to be made of many faint extragalactic sources.  In this paper, we present mid-infrared $ISO$ spectra of the [S~III] 18.7 $\\mu$m and 33.5 $\\mu$m lines, plus the [S~IV] 10.5 $\\mu$m line, from the H~II region Hubble~V in NGC~6822 (hereafter referred to as Hubble~V), and the blue compact galaxy I~Zw~36 (MRK~209; UGCA~281). Our goal is to compare our infrared observations with published optical observations in order to test whether the theoretically predicted ionization correction factors are correct and to determine whether temperature fluctuations are large enough to be detectable in this manner. ",
        "conclusions": "We have reported new ISO observations of the mid-infrared fine structure [S~III] and [S~IV] lines.  These lines are of great importance to the accurate determination of nebular total sulfur abundances.  This is due to the strong dependence of higher excitation lines on temperature, along with the fact that there are no strong optical lines for either of these species.  With our observations, we have shown that S$^{+3}$ can constitute a large fraction of the total sulfur abundance in extragalactic H~II regions. This means that if one is to determine accurately the total sulfur abundance without model dependence as in equation (4), then one must make infrared observations of the fine structure [S~III] and [S~IV] lines.  Once infrared observations are made, it becomes possible to test not only useful techniques such as the determination of the radiation ``hardness'' \\citep{VP88} and the determinations of the total sulfur abundance as in equation (4) \\citep{G89}, but it becomes possible to test the accuracy of photoionization models.  Our data suggest that ionization corrections for sulfur based on oxygen ion ratios and photoionization models are valid.  The presence of temperature fluctuations in nebulae can complicate the determination of nebular abundances. We have developed a new, generalized diagnostic capable of determining the amplitude of these fluctuations by assuming a Gaussian temperature distribution in these nebulae, although this method can be applied to any normalized distribution. The uncertainties in our ISO measurements and the lack of observations of [S~III] $\\lambda 9532$ or $\\lambda 9069$ did not allow an accurate determination of the amplitude of temperature fluctuations for Hubble~V and I~Zw~36 using our method. A significant challenge is presented by combining large aperture infrared spectra with relatively small aperture optical spectra. Future long slit spectrographs available with SOFIA and SIRTF will allow us to overcome this challenge. As these more powerful instruments become available, observational uncertainties should decrease, allowing a more accurate determination of the size of the temperature fluctuations and other nebular parameters.  In the future, one can consider extensions to the present analysis. For example, the potential to calculate spatially unresolved density fluctuations in a similar mathematical framework remains relatively unexplored, although \\cite{R89} has considered the biasing of IR density indicators by density fluctuations. Like temperature fluctuations, density fluctuations can also affect nebular line ratios, and they may have significant effects on nebular physical parameters which must be constrained in order to develop better nebular models. To this end, application of this diagnostic to published data on a large number of H~II regions may be useful in characterizing the variances found in H~II regions."
    },
    "0209/astro-ph0209567_arXiv.txt": {
        "abstract": "We use joint likelihood analyses of combinations of fifteen cosmic microwave background (CMB) anisotropy data sets from the DMR, UCSB South Pole 1994, Python I--III, ARGO, MAX 4 and 5, White Dish, OVRO, and SuZIE experiments to constrain cosmogonies. We consider open and spatially-flat-$\\Lambda$ cold dark matter cosmogonies, with nonrelativistic-mass density parameter $\\Omega_0$ in the range 0.1--1, baryonic-mass density parameter $\\Omega_B$ in the range (0.005--0.029)$h^{-2}$, and age of the universe $t_0$ in the range (10--20) Gyr.  Marginalizing over all parameters but $\\Omega_0$, the data favor $\\Omega_0\\simeq$ 0.9--1 (0.4--0.6) flat-$\\Lambda$ (open) models. The range in deduced $\\Omega_0$ values is partially a consequence of the different combinations of smaller-angular-scale CMB anisotropy data sets used in the analyses, but more significantly a consequence of whether the DMR quadrupole moment is accounted for or ignored in the analysis. While the open model is difficult to reconcile with the results of less exact analyses of more recent CMB anisotropy data, the lower values of $\\Omega_0$ found in this case are more easily reconciled with dynamical estimates of this parameter. For both flat-$\\Lambda$ and open models, after marginalizing over all other parameters, a lower $\\Omega_B h^2 \\simeq$ 0.005--0.009 is favored. This is also marginally at odds with estimates from more recent CMB anisotropy data and some estimates from standard nucleosynthesis theory and observed light element abundances. For both sets of models a younger universe with $t_0 \\simeq$ 12--15 Gyr is favored, consistent with other recent non-CMB indicators. We emphasize that since we consider only a small number of data sets, these results are tentative. More importantly, the analyses here do not rule out the currently favored flat-$\\Lambda$ model with $\\Omega_0 \\sim 0.3$, nor the larger $\\Omega_B h^2$ values favored by some other data. ",
        "introduction": "There has been a remarkable increase in the quality and quantity of cosmic microwave background (CMB) anisotropy measurements since the initial detection of the anisotropy on large angular scales a decade ago.\\footnote{ See, e.g., Miller et al. (2002a), Coble et al. (2001), Scott et al. (2002), and Mason et al. (2002) for recent measurements.} These measurements are becoming increasingly useful in the continuing processes of determining how well cosmological models approximate reality and for constraining cosmological parameters such as $\\Omega_0$, $h$, and $\\Omega_B$ in these models\\footnote{ Here $h$ is the Hubble constant $H_0$ in units of $100\\ {\\rm km}\\ {\\rm s}^{-1}\\ {\\rm Mpc}^{-1}$. See, e.g., Podariu et al. (2001), Wang, Tegmark, \\& Zaldarriaga (2002), Miller et al. (2002b), Sievers et al. (2002), and Lewis \\& Bridle (2002) for recent discussions of constraints on cosmological parameters.}.  In this paper we utilize the full information in the data from each experiment in an effort to place robust constraints on cosmological parameters. This is achieved through a maximum likelihood analysis of the data using realistic model anisotropy spectra. Most of the data sets we consider in this paper are small, in contrast to some of the more recent data sets (see, e.g., Netterfield et al. 2002; Stompor et al. 2001), which because of their size require a more approximate analysis technique. Ganga et al. (1997a, hereafter GRGS) extend the maximum likelihood technique to account for uncertainties, such as those in the beamwidth of the telescope and the calibration of the experiment.\\footnote{ For some of the data sets we consider in this paper foreground non-CMB contamination must also be accounted for; see, e.g., Kogut et al. (1996), de Oliveira-Costa et al. (1998), Hamilton \\& Ganga (2001), and Mukherjee et al. (2002a, 2002b) for discussions of the method used to accomplish this.} This technique has been used with model CMB anisotropy spectra in analyses of the Gundersen et al. (1995) UCSB South Pole 1994 data, the Church et al. (1997) SuZIE data, the Lim et al. (1996) MAX 4+5 data, the Tucker et al. (1993) White Dish data, the de Bernardis et al. (1994) ARGO data, the Platt et al. (1997) Python I--III data, and the Leitch et al. (2000) OVRO data (GRGS; Ganga et al. 1997b, 1998; Ratra et al. 1998, 1999a, hereafter R99a; Rocha et al. 1999; Mukherjee et al. 2002b).  Given the error bars associated with these measurements, interesting constraints on cosmological model parameters require the joint analysis of many data sets. If the measurements are acquired for regions well separated in space, or on very different angular scales, the likelihoods of the individual data sets are independent and can thus be multiplied together to construct the likelihood of the combined data set. This combined likelihood is then used to derive constraints on cosmological model parameters.  A combined analysis of the smaller-angular-scale data sets listed above, excluding the Python and OVRO data, is presented in Ratra et al. (1999b, hereafter R99b). In this paper we extend the analysis of R99b to include the Python and OVRO data, as well as the large angular scale DMR data (G\\'orski et al. 1998, hereafter G98; Stompor 1997).  In $\\S$ 2 we describe the models and cosmological parameter space we consider. See R99a for a more detailed description. In $\\S$ 3 we summarize the various combinations of data sets we consider. See R99b for further details. In $\\S$ 4 we summarize the computational techniques we use. See GRGS and R99b for more detailed discussions. In $\\S$ 5 we present and discuss results from analyses of various combinations of the smaller-angular-scale CMB anisotropy data sets, and in $\\S$ 6 we add the DMR data to the mix. We conclude in $\\S$ 7. ",
        "conclusions": "We have derived constraints on cosmological model parameters in the open and flat-$\\Lambda$ CDM models, from joint analyses of combinations of the DMR, SP94, Python I--III, ARGO, MAX 4+5, White Dish, OVRO, and SuZIE CMB anisotropy data sets. The constraints derived here are not of very high statistical significance.  The data considered here mildly favor the open case over the flat-$\\Lambda$ model, although they do not rule out the currently favored flat-$\\Lambda$ model with $\\Omega_0 \\sim 0.3$. The favored value of $\\Omega_0$ in the open ($\\Omega_0 \\sim 0.4-0.6$) and flat-$\\Lambda$ ($\\Omega_0 \\sim 0.9-1$) models are somewhat dependent on whether the DMR quadrupole moment is included in or excluded from the analysis. Resolving this issue will likely require analyses that use new higher quality large angular scale CMB anisotropy and foreground emission data.  Constraints on $\\Omega_B h^2$ and $t_0$ are only weakly dependent on the data set combination considered. The data considered here favors lower $\\Omega_B h^2 \\sim 0.005-0.009$, or younger, $t_0 \\sim 12-15$ Gyr, universes.  Some of the constraints derived here are mildly inconsistent with those derived elsewhere, but gratifyingly not so at high statistical significance. Tighter and more robust constraints on cosmological parameters will require a models-based joint analysis of a larger collection of CMB anisotropy data sets.  \\bigskip  We acknowledge valuable assistance from R. Stompor. This work was partially carried out at the California Institute of Technology IPAC and JPL, under a contract with NASA. PM, BR, and TS acknowledge support from NSF CAREER grant AST-9875031. NS acknowledges support from the Alexander von Humboldt Foundation and Japanese Grant-in-Aid for Science Research Fund No. 14540290."
    },
    "0209/hep-ph0209358_arXiv.txt": {
        "abstract": "We find that there exists a soliton-like solution ``I-ball'' in theories of a real scalar field if the scalar potential satisfies appropriate conditions.  Although the I-ball does not have any topological or global $U(1)$ charges, its stability is ensured by the adiabatic invariance for the oscillating field. ",
        "introduction": "\\label{sec:introduction}     Scalar fields play important roles in theories of the early universe. It is believed that our universe experienced quasi-exponential expansion phase ($=$ inflation ) in its very early stage, which solves the flatness and horizon problems of the standard cosmology and explains the origin of the density fluctuations of the universe such as observed by COBE~\\cite{COBE} and other experiments~\\cite{BOOMERANG,MAXIMA,DASI}. The inflationary universe scenario is realized by the vacuum energy of some scalar field (inflaton). After inflation, the inflaton starts to oscillate and decays into other particles which reheat the universe through thermalization processes.  Similar dynamics is found in the Affleck-Dine mechanism for baryogenesis~\\cite{AD} which is a promising candidate for explaining the matter-antimatter asymmetry of the universe. The mechanism makes use of a scalar field (AD field) corresponding to a flat direction in the scalar potential of the minimal supersymmetric standard model. During inflation the AD field has a large field value and oscillates when the effective mass becomes smaller than the Hubble parameter after inflation. When the AD field starts oscillation, the baryon number is generated through the baryon number violating term in the potential.  Recently, it was found that the oscillating AD field deforms into lumps of the scalar condensate called Q balls~\\cite{Kusenko,Enqvist,Kasuya1,Kasuya2}. The Q ball is a non-topological soliton and its stability comes from the global charge ($=$baryon number) conservation. The existence of the Q ball is crucial because it may significantly change the scenario of the Affleck-Dine baryogenesis~\\cite{Kasuya3}. The fragmentation into scalar lumps may also take place for the inflaton field. In fact, Enqvist {\\it et al}~\\cite{EnqvistKasuya} pointed out that the oscillating inflaton field can fragment into Q balls.  Since the Q ball is stable owing to the charge conservation, the scalar field responsible for the Q ball must be complex. Then, a question arises whether or not a real scalar field deforms into lumps similar to the Q balls. At first glance, no stable lumps are formed because any conservation quantities like a global charge do not exist for the system of a real scalar field.  However, the previous studies on the dynamics of scalar fields showed that some soliton-like objects are formed. For example, ``oscillons'' are formed for % phase the double well potentials~\\cite{Gleiser} and the axion field fragments into ``axitons''~\\cite{Tkachev}. In both cases the numerical simulations showed the existence of some scalar lumps inside which the scalar fields are rapidly oscillating. However, the reason why such quasi-stable scalar lumps are formed was not clear at all. Moreover, recently, McDonald~\\cite{McDonald} pointed out that in a hybrid inflation model the inflaton field can fragment into scalar lumps even if the scalar field has any conserved charges [see also Ref.~\\cite{rajantie}].  Thus, it has been seen that real scalar fields fragment into quasi-stable lumps in numerical simulations for various situations.  In this paper, therefore, we face the important problem concerning the real scalar dynamics, that is, what makes the scalar lump quasi-stable?  Because the conserved baryon number plays a crucial role for stability of the Q ball, we need similar conservation quantity to stabilize the real scalar lump. In classical mechanics it is well-known that the adiabatic invariant exists for oscillating phenomena~\\cite{landau}.  As will be seen later, we find that the adiabatic invariant can be extended to the field theories. Then, the existence of the stable lump can be explained by the adiabatic charge $I$ ( see the next section for definition ) for the oscillating scalar field.  We call this scalar lump ``I-ball'', since the adiabatic charge $I$ plays the same role as the global $U(1)$ charges in the case of Q balls.  We obtain the condition on the form of the scalar potential for the I-ball formation and derive the equation which determines the field configuration of the I-ball.  In particular, it is found that the adibaticity requires the scalar potential to be dominated by a quadratic term.  We also perform numerical simulations to confirm the existence of the I-balls for two types of simple potentials. ",
        "conclusions": "\\label{sec:con}   We have studied the system of a real scalar field and found the solution of the quasi-stable scalar lump, I-ball. The stability of the I-ball can be explained by the adiabatic invariant charge $I$, which does stem from the dynamics of the system, not any symmetries.  For the I-ball solution to exist, the scalar potential should be dominated by the quadratic term ($m^2\\phi^2$) and satisfy the condition (\\ref{eq:ibc}) which is almost same as that for the Q-ball.  Furthermore, we have performed numerical simulations and have found that the quasi-stable I-balls are really produced and their properties are in agreement with the theoretical predictions.  Since scalar fields prevail in theories of the early universe, the I-balls may be formed and play important roles in various cosmological processes~\\cite{KKT}.       \\subsection*"
    },
    "0209/astro-ph0209201_arXiv.txt": {
        "abstract": "Abstract: We apply a model of dynamo-driven mass loss, magnetic braking and tidal friction to the evolution of stars with cool convective envelopes; in particular we apply it to binary stars where the combination of magnetic braking and tidal friction can cause angular-momentum loss from the {\\it orbit}. For the present we consider the simplification that only one component of a binary is subject to these non-conservative effects, but we emphasise the need in some circumstances to permit such effects in {\\it both} components. \\par The model is applied to examples of (i) the Sun, (ii) BY Dra binaries, (iii) Am binaries, (iv) RS CVn binaries, (v) Algols, (vi) post-Algols. A number of problems regarding some of these systems appear to find a natural explanation in our model. There are indications from other systems that some coefficients in our model may vary by a factor of 2 or so from system to system; this may be a result of the chaotic nature of dynamo activity. ",
        "introduction": "In a previous paper (Nelson \\& Eggleton 2001; Paper I) we constructed a large grid of models of Case A binary evolution, according to the assumption of conservative evolution. We found that these fitted reasonably the parameters of certain observed `hot Algols', i.e. semidetached binaries in which {\\it both} components were earlier than about spectral type G0. We also found that agreement was quite poor for some `cool Algols', by which we mean those in which at least one component is later than $\\ts $G0. Since several of the latter appeared to disagree on account of having less angular momentum, and/or less total mass, than the theoretical models, we suggested that the discrepancy is due to dynamo activity in stars with cool convective envelopes. Such activity can reasonably be expected to be substantially greater than in single stars of the same spectral type, because components in Algols are typically rotating much faster than single stars, or stars in binaries where the orbital period is quite long. This activity may reasonably be expected to carry off both mass and angular momentum, but whether it can carry off the right amounts is not clear, and is the main subject of this paper. \\par In another paper (Eggleton 2001; Paper II), PPE suggested a simplistic model of dynamo activity, suitable for inclusion in a binary-evolution code. In the present paper we present some results. The model is simplistic in the sense that it gives the mass-loss rate, the overall poloidal magnetic field, and the consequential magnetic braking rate, as functions of just four parameters: the mass, luminosity, radius and rotational period of the star. In order that the magnetic braking should be able to drain angular momentum from the {\\it orbit}, it is necessary to include also a model of tidal friction. Of course this is also necessary to get the star to spin faster in a binary than it would if it were single. We follow the prescription of Eggleton \\& Kiseleva (2001: Paper III), but specialising to the simple case of only two bodies, and stellar spin parallel to orbital spin. \\par In Section 2 we discuss briefly the implementation of these models in a stellar evolution code, noting that there are some considerable approximations which influence particularly low-mass systems, where quite probably {\\it both} components contribute to mass loss and angular-momentum loss. In this paper we allow only the initially more massive star to be subject to these processes. This is forced on us by numerical considerations, but we hope to circumvent them in the future. In Section 3 we discuss our results, and consider the further evolution of such systems. In Section 4 we consider what may be required in future modeling. ",
        "conclusions": "We have presented a simplistic formulation of mass loss driven by dynamo activity, angular momentum loss driven by magnetic braking, and tidal friction, that in the first instance has no free parameters. It is calibrated to agree with the present-day Sun, and is scaled according to the depth of convective envelope and the Rossby number in a realistic way. It appears to account for a fairly wide range of phenomena: the surprisingly slow rotation in BY Dra, the surprisingly low mass in the more evolved component of Z Her, and the surprisingly low angular momentum of Algols such as DN Ori and S Cnc. Variants of this model, in which some coefficients are altered by factors of up to $\\ts 2$, may account for a wider range of observed objects, such as AS Eri, R CMa and AA Dor. Given that the dynamo activity which is the basis of both the mass loss and the magnetic braking is an inherently chaotic process, it is certainly not surprising that factors of two or more should be necessary as between one system and another. \\par The same formulation can of course be applied to other classes of object: contact binaries, cataclysmic variables, pre-cataclysmic systems, and low-mass X-ray binaries. We hope to pursue these in a future paper. Estimates have already been given by in Paper II. \\par An important improvement that will be necessary to understand some further systems, such as cool double-subgiant binaries, is to include star 2 within the formalism; for the present only star 1 is allowed to be subject to these non-conservative processes. This will be quite a major undertaking, since it will be necessary to solve for both components {\\it simultaneously}. However this is also necessary if one is to follow the evolution of contact binaries, since an additional model, for heat transport between the two components, is necessary there. We hope to produce this in due course. \\par A different but very important way of pursuing the same topics is to model these interactions in a fully 3-D stellar model, or pair of models. For example, it would be desirable to model tidal friction in such a way, to make a better estimate of the viscous timescale that is incorporated in the constant factor of Equation (9). If MHD is included, presumably just in the frozen-in approximation, then 3-D calculations could also serve to calibrate the other non-conservative processes. Most importantly, they could also give us insight into the poorly-understood process of heat transport in contact binaries. \\par The DJEHUTY project is currently being developed at the Lawrence Livermore National Laboratory, with a view to tackling the 3-D structure of stars. At present grids of $\\ts 10^8$ meshpoints are available, and as computer power increases we hope to improve this to $\\ts 10^{10}- 10^{11}$. Of course, one would not evolve such a 3-D model for several Gigayears, but only for modest times like $\\ts 1\\thin$yr. This should allow us to make a 1-D average of such processes as tidal friction, and incorporate them in a 1-D code such as the one used here."
    },
    "0209/astro-ph0209347_arXiv.txt": {
        "abstract": "The results of a search for cool subdwarfs are presented. Kinematic (U, V, and W) and stellar parameters (\\teff, log g, [Fe/H], and $\\xi_t$) are derived for 134 candidate subdwarfs based on high resolution spectra.  The observed stars span 4200K $<$ \\teff~$<$ 6400K and $-2.70<$ [Fe/H] $<$ 0.25 including only 8 giants (log g $<$ 4.0).  Of the sample, 100 stars have MgH bands present in their spectra.  The targets were selected by their large reduced proper-motion, the offset from the solar metallicity main sequence, or culled from the literature.  We confirm the claims made by \\citet{ryan89} regarding the NLTT catalog being a rich source of subdwarfs and verify the success of the reduced proper-motion constraint in identifying metal-poor stars. ",
        "introduction": "\\label{sec:intro}  The driving force behind our understanding of the chemical evolution of the Galaxy is the interpretation of observed abundance ratios.  Self-consistent analyses of large high quality data sets have revealed detailed abundance patterns (e.g., \\citealt{bdp93}, \\citealt{mcwilliam95}).  Recent attempts to understand the observed abundance trends include endeavors by \\citet{timmes95}, \\citet{goswami00}, and \\citet{alc01} who predict the evolution of the abundances of all elements from carbon through zinc. Cool stars provide a unique opportunity to test directly these models of Galactic chemical evolution through the abundances of low ionization potential trace elements (e.g., Rb, Cs) and isotopic ratios measured from molecular bands (e.g., Mg from MgH, Ti from TiO). Presently these tests cannot be carried out due to the dearth of known cool metal-poor stars.  In order to rectify this situation we have commenced a search for cool subdwarfs, unevolved metal-poor stars that fall below the solar metallicity main sequence in color-magnitude diagrams.  Searches undertaken by \\citet{carney87} and \\citet{ryan89} selected candidates drawn from proper-motion catalogs.  These searches were successful in identifying metal-poor dwarfs for subsequent detailed abundance analysis. However, both searches neglected cool subdwarfs. An alternative to searches for metal-poor stars based on proper-motion catalogues are the objective prism surveys which identify candidate metal-poor stars by the weakening of Ca {\\scshape ii} H and K. Such studies have been conducted by \\citet{bond70}, \\citet{bond80}, \\citet{beers85}, \\citet{beers92}, and others. Selection criteria which rely exclusively upon Ca features as metallicity indicators strongly bias the sample towards warmer stars. Since metal lines weaken with increasing \\teff, a temperature estimate is required before deciding if lines are abnormally weak. In the absence of temperature indicators, a cool metal-poor star will have Ca {\\scshape ii} features comparable to a warmer solar metallicity star. More current work by \\citet{christlieb00} on the stellar content of the Hamburg/ESO Survey has dramatically increased the yields of metal-poor stars.  Candidate metal-poor stars are identified from their location in {\\em $(B-V)$ $-$ Ca line strength} space. By comparing the Ca line strength between stars of similar \\teff's, this increases the likelihood that a metal deficiency is the cause of the relative weakening of the Ca line in a candidate. With the remarkable success rate of 60\\% for stars below [Fe/H] $=-2.0$, twice as good as the HK survey by Beers and collaborators, we look forward to the results of the Hamburg/ESO Survey when applied to cool subdwarfs.  In this paper we present the results of our search for cool subdwarfs, stellar (\\teff, log g, [Fe/H], and $\\xi_t$) and kinematic parameters (U, V, and W) for 134 stars. Our sample includes 80 stars with no prior metallicity estimates.  In \\S\\ref{sec:criteria} we outline and justify the selection criteria.  The observations will be described in \\S\\ref{sec:data} and the analysis in \\S\\ref{sec:analysis}.  A discussion will be presented in \\S\\ref{sec:discussion} and concluding remarks given in \\S\\ref{sec:remarks}. ",
        "conclusions": "\\label{sec:remarks}  We present stellar parameters for 134 candidate subdwarfs selected by their reduced proper-motion, offset from the solar metallicity, or from the literature.  Our goal was to provide a large database of cool subdwarfs.  Our selection criteria were successful in identifying cool stars (69 with \\teff$<4800$K) and metal-poor stars (27 with [Fe/H]$\\le-1.50$).  Of our sample, 11 stars were sufficiently cool to provide measurable MgH lines with [Fe/H]$<-1.50$. Armed with a sample of cool subdwarfs, we can begin to exploit their unique qualities.  For a subset of our sample, we have measured the Mg isotopic abundance ratios and compared the observed trends with predictions from models of Galactic chemical evolution.  This work will be presented in a future paper.  We also intend to make further observations of candidate subdwarfs.  Targets will be selected from the revised NLTT catalog \\citep{salim02}.  Cool \\hipparcos subdwarf candidates will be observed and we anticipate that Christlieb and collaborators will identify cool subdwarfs in the Hamburg/ESO survey data."
    },
    "0209/astro-ph0209171_arXiv.txt": {
        "abstract": "We have updated our PHOENIX model atmospheres and theoretical spectra for ultracool dwarfs with new opacity data for methane based on line strength predictions with the STDS software. By extending the line list to rotational levels of $J\\,=\\,40$ we can significantly improve the shape of the near-IR absorption features of CH$_4$, and in addition find an enhanced blanketing effect, resulting in up to 50\\% more flux emerging in the $J$ band than seen in previous models, which may thus contribute to the brightening in $J$ and blue IR colors observed in T dwarfs. ",
        "introduction": "Absorption bands of H$_2$O and CH$_4$ dominate the IR spectra of dwarfs cooler than $\\sim$\\,1400\\,K and are a backbone of the T dwarf classification schemes defined by Geballe et al.\\ (2002) and Burgasser et al.\\ (2002). Accurate and complete opacity data for these molecules are thus essential for modelling the atmospheres of T dwarfs. We have updated the PHOENIX model atmosphere code with a new methane list based on theoretical calculations with the {Spherical Top Data System (STDS)} ({Wenger \\& Champion 1998}). Compared to our AMES-Cond models ({Allard et al.\\ 2001}), with $\\sim$\\,37\\,000 CH$_4$ lines from the HITRAN and GEISA databases, this has added 1.2$\\cdot\\,10^7$ lines (Homeier et al.\\ 2002). ",
        "conclusions": "We could significantly improve our synthetic IR spectra for T dwarfs with the new STDS line list for methane. In addition, the new models produce a shift of emergent flux towards the $J$ band due to stronger blanketing, reproducing well the blue IR colors of T dwarfs. To clarify the exact role of the onset of CH$_4$ in the rapid brightening in $J$ observed at the L-T transition, a new set of models will be necessary that also provides a consistent treatment of the settling of the cloud layer below the photosphere expected to occur in about the same temperature range (Allard et al., these proceedings).  \\vspace*{2.4ex} \\noindent \\textbf{Acknowledgments.} We thank D.\\ Alexander, V.\\ Boudon, J.-P.\\ Champion and U.\\ J{\\o}rgensen for helpful discussions and S.\\ Leggett for access to observational data. This work is supported by NFS grant N-Stars RR185-258."
    },
    "0209/astro-ph0209165_arXiv.txt": {
        "abstract": "We investigate the effect of rotating, triaxial halos on disk galaxies through an extensive set of numerical $N$-body simulations. Our simulations use a rigid potential field for the halos and bulges and collisionless particles for the disks. The triaxiality and the rotation rate of the halo are varied, as well as the masses of all three galaxy components. We analyze both the bar stability and the spiral response of the disks under these conditions.  We characterize most of our models by the mass ratio of the disk to the halo at 2.3 disk scale lengths, \\mratio. For models with a mass ratio greater than 0.8, a halo pattern speed \\omh\\ = 6.7 \\kmskpc, and a intermediate-to-major axis ratio $q_b=0.85$, a strong bar will develop within 3 Gyr, even for models with a bulge mass $M_b=0.3M_d$. Models in which the bulge mass is reduced by half develop bars earlier and with lower \\omh.  We create an artificial Hubble sequence of disk galaxies by varying the bulge-to-disk ratio of our models from 0 to 2.5. The torque induced by a rotating, non-axisymmetric halo creates bisymmetric spiral structure in the disk.  We find that the pitch angle of the spiral arms in these models follows the same general trend found in observations of spiral galaxies, namely that later type galaxies have higher pitch angles. Our simulations follow closely the observational relation of spiral pitch angle with maximum rotational velocity of the disk, where galaxies with faster rotation have more tightly wound spiral arms. This relation is followed in our simulations regardless of whether the dominating mass component of the galaxy is the disk, the halo, or the bulge. ",
        "introduction": "Galaxies are not spherical objects. Although it is difficult to measure the intrinsic shapes of astronomical bodies from their projected light distributions, inferences can be make from their observed ellipticities. Many studies of the distribution of observed axis ratios of galaxies and galaxy clusters (\\eg\\ Ryden, 1992, 1996; Tremblay \\& Merritt 1996) have shown that they are well represented by a population of intrinsically triaxial objects. Similar statistical analysis performed by Alam \\& Ryden (2002) with a much larger sample of galaxies from the Sloan Digital Sky Survey Early Data Release has strengthened the claim that the observed distributions of ellipticities cannot be fit by projections of oblate or prolate spheroids.  This type of analysis is only available in the cases where the mass distribution creates light. The shapes of dark matter halos, therefore, must be inferred from even more indirect means. Proxy measures of the shape of our own halo offer widely varying results. The tidal stream of the Sagittarius dwarf galaxy implies a nearly spherical halo (Ibata \\etal\\ 2001), in contrast to the results of star counts, which give a minor to major axis ratio of $\\sim$ 0.6 (Siegal \\etal\\ 2002). The dark matter halos created in numerical studies of structure formation are non-axisymmetric systems. $N$-body simulations of dissipationless collapse in the cold dark matter (CDM) scenario by Dubinski \\& Carlberg (1991) showed that the resulting virialized halos deviated substantially from spherical symmetry and from axisymmetry, with average axis ratios $\\langle c/a\\rangle=0.5$ and $\\langle b/a\\rangle=0.7$, where $a$, $b$ and $c$ are the major, intermediate, and minor axes respectively. Simulations of larger cosmological volumes, which more adequately reproduce the hierarchical nature of structure formation in CDM, have also shown that dark matter halos are triaxial (Barnes \\& Efstathiou 1987; Warren \\etal\\ 1992). Analysis of the currently favored $\\Lambda$CDM cosmology has shown similar results (Jing \\& Suto 2002).  The simulations cited above consist only of a collisionless dark matter component and include no gas dynamics. The introduction of a gas and stellar disk in the equatorial plane of the halo would certainly have effects on the internal structure of the halo. Fully self-consistent cosmological simulations of dissipational structure formation have problems reproducing more than general galactic attributes (\\eg\\ Murali \\etal\\ 2002), and the resulting halo shape has not been the focus of these efforts.  However, Dubinski (1994) showed that the adiabatic growth of a disk-like potential in a triaxial halo did not influence $\\langle c/a\\rangle$, while the intermediate to major axis ratio was only slightly increased to $\\langle b/a\\rangle \\gtrsim 0.7$.  Models and simulations of disk galaxies, however, usually employ spherical halos, which do not reflect these results from CDM simulations. The purpose of most current investigations of galaxy-scale simulations has been modeling transformations of galaxies through major mergers (\\eg\\ Barnes 1992; Dubinski, Mihos, \\& Hernquist 1999), minor encounters (\\eg\\ Struck 1997; Thakar \\& Ryden 1998) and gas-dynamical effects (\\eg\\ Mihos \\& Hernquist 1996; Springel 2000). The triaxiality of the halo does not immediately effect the results of these simulations and the implications of using triaxial halos have not been explored.  Early studies of the stability of disk galaxies (e.g. Ostriker \\& Peebles 1973; Hohl 1976; Efstathiou \\etal\\ 1982) demonstrated that a cold, rotationally supported disk is dramatically unstable to bar formation. This fact was used as justification for the existence of heavy, spherical, dark matter halos enveloping disk galaxies. This dynamically hot component of the galaxy inhibits bar formation. It has also been shown that a dense core gives a galaxy an inner Lindblad resonance, which does not allow swing amplification of waves through the center of the galaxy and thus acts as a bar suppressant (Sellwood 1989)\\footnote{Efstathiou \\etal\\ (1982) did perform two simulations with dense centers but found that it did not make their disks stable. Sellwood proposed that the low resolution of their simulations was responsible for the results of their ``bulge'' simulations.}. In all cases the stabilizing mass component has been assumed to have spherical symmetry. If these galaxies were enveloped by triaxial halos, they would be more susceptible to bar instabilities. A systematic exploration of the stability of disks with non-axisymmetric halos has not been conducted. Curir \\& Mazzei (1999) presented fully self-consistent gas-dynamical simulations of disk galaxies with triaxial halos, but only used two halo models and did not employ bulges. The resolution in their simulations was severely limited ($N_{disk}=3,000$), far less than the number of particles used by Efstathiou \\etal\\ (1982) which Sellwood (1989) found to be insufficient.  The notion that the potential field of a galaxy may contain non-axisymmetric features is not a new one; it is, in fact, readily apparent from the bars seen in many galactic disks. The effect of an oval distortion, like a bar, on a gas disk has been proposed as a driving mechanism for spiral density waves (Lin 1970). This idea has been explored numerically as well (Sanders \\& Huntley 1976; Sanders 1977; Huntley \\etal\\ 1978). These early simulations generally were limited to two-dimensional, massless disks driven by an analytic distortion to the symmetric potential, but were successful in creating significant spiral response in the disk. However, bar forcing does not reproduce tightly bound spiral patterns. And in comparison to Sanders \\& Huntley (1976), the response of the outer disk is weak when a more realistic bar potential, one that drops off faster with radius, is used in the simulation (Sanders \\& Tubbs 1980). Observationally spiral patterns are as common in barred galaxies as galaxies without bars. Also, there is evidence that bars and spiral arms have different pattern speeds, showing that one feature may not drive the other (Sellwood \\& Sparke 1988). A triaxial halo would not be subject to these concerns since the potential of the dark matter would dominate the outer regions of the disk, especially low mass disks which are stable to bar formation.  The bulge-to-disk ratio of a spiral galaxy, as well as the tightness of the spiral pattern, are two integral criteria of the Hubble classification scheme. On the average, there exists a smooth increase in pitch angle with later Hubble types (Kennicutt 1981). Such a correlation, however, is little more than a consistency check between classification parameters, and there exists significant scatter in the measured pitch angle for a given Hubble type.  Even with this scatter, the correlation between arm pattern and galaxy type has prompted several theoretical studies on the origin of the this relation (see Kennicutt \\& Hodge 1982, and references therein). Kennicutt (1981) found that the correlation between pitch angle and maximum rotational velocity of the disk was as good as the theoretical model predictions, suggesting that the mass distribution and its resulting rotation curve is an important determinant in producing the shape of spiral arms. A more massive or more concentrated bulge will induce higher rotation velocities and loosely follow the correlation with Hubble type as well.  Our simulations include a collisionless stellar disk and a stiff, rotating halo. Most simulations include a bulge, which is also stiff. Our use of dissipationless simulations with halos which are not fully self-consistent allows us to cover a significant range of parameter space for all three galaxy components; the disk, bulge, and halo.  The structure of this paper is as follows: \\S 2 presents our initial conditions as well as the $N$-body techniques used in this study. \\S 3 shows the analysis and results of our fiducial simulation. \\S 4 describes our suite of different galaxy models. In \\S 5 we investigate the stability of our models with different halo and bulge masses, as well as different halo rotation rates. \\S 6 presents our results for the spiral morphology of our simulations. We present results for models which follow the bulge-to-disk ratios of the Hubble sequence and models which vary widely in rotation velocity. We compare our results to observations of spiral galaxies. ",
        "conclusions": "We have presented simulations of galaxy disks embedded in rigid, rotating, triaxial halos. We have constructed a suite of models in which the masses of the disk, bulge and halo have all been varied. With these models, the effects of changing the halo rotation rate and flattening have been investigated. Our results have focused on the formation of bars and spiral structure. Models with a mass ratio \\mratio\\ $> 0.8$ and \\omh\\ $=0.09$ develop a strong bar, even if a significant bulge is present. For all models, the torque created by the halo induces spiral structure in the disk. The average pitch angle of this structure, \\pave, is strongly anti-correlated with the maximum rotation speed of the model.  It is clear that estimations of the stability of disk galaxies cannot ignore the shape of the halo. Models with significant amounts of dark matter and a significant bulge (like D1 and D2, which have rotational velocities of over 200 \\kms\\ in our system of units) can become bar dominated with halo rotation rates of \\omh\\ = 0.09 and greater. Maximal disk models, like D5, are very susceptible to bar instabilities if torqued by the halo. In all the simulations presented in \\S 5, the only set of halo and bulge parameters for which models D3--D5 were stable against bar formation was $M_b=0.3M_d$ and \\omh\\ = 0.06. Even with those parameters, model D3 still developed a low amplitude but persistent bar. A simple criterion for determining whether a disk galaxy will develop a bar, such as that determined by Efstathiou \\etal\\ (1982), is not readily apparent from the data. With more free parameters, $M_b$, $q_b$, and \\omh, in our simulations, it may be that a simple criterion cannot quantify the stability of all our models used in our simulations.  The correlation of rotational velocity and pitch angle in our models provides a physical explanation of the same correlation found in observations of spiral galaxies. The pitch angle of the arms induced by an external torque is related to the total mass of the system, loosely through $v_{max}$, but more tightly to the dimensionless rotational energy parameter \\ep. The differences seen in our results for different halo rotation rates and flattenings fit well with the amount of scatter seen in the correlation of $v_{max}$ with \\pave\\ for real galaxies. This result does not have to be unique to the properties of a galaxy's halo; it has been shown that encounters with satellite galaxies, which would also create a time-varying potential field in which to disrupt the disk, induce two-armed spiral structure in disk galaxies (\\eg\\ Toomre \\& Toomre 1972). But encounters with smaller galaxies are by their nature transient and it remains that spiral structure is a ubiquitous feature of disk galaxies both with and without satellite perturbers.  In Kennicutt (1981), and references therein, the properties of the spiral arms are measured through their blue light, usually from H{\\small II} regions, dust lanes, or the blue continuum, all of which are associated with star formation. Our collisionless simulations better represent the red stellar population which accounts for most of the mass in a galactic disk. Images of spiral galaxies in the near infrared have shown that spiral structure is evident in old stellar populations (Eskridge \\etal\\ 2002). The arms observed are smoother and wider than those seen in blue light, but generally follow the same morphology.  The Fourier decomposition method used in this paper has been applied to observational data of H{\\small II} regions in spiral galaxies (Puerari \\& Dottori 1992; Garc\\'ia-Gom\\'ez \\& Athanssoula 1993). The Fourier spectra obtained resemble in many ways the spectra of our simulated galaxies, often with several distinct peaks inside one spectrum. The pitch angle was calculated by these authors from the value of $\\alpha$ for which the amplitude is highest. This definition of $p$ reproduces the general trend with Hubble type as described by Kennicutt (1981) but does not reflect the natural variations in pitch angle within a single galaxy seen both in the Fourier spectra of the observations and in Kennicutt's calculations. Kennicutt noted that the variations of $p$ for a typical galaxy were significantly larger than the measurement error. An average value of the pitch angle better represents the multiple spiral components.  Hydrodynamic simulations by Bekki \\& Freeman (2002) have shown that the extended spiral structure seen in the blue compact dwarf galaxy NGC 2915 could be due to a rotating triaxial halo. Their simulations contained a disk fully composed of gas particles and a halo 100 times more massive than the disk, effectively making the disk massless. Our collisionless simulations show spiral structure in the inner regions of galactic disks. Since gas fractions in disk galaxies are usually no higher than 30\\%, the gas should be gravitationally coupled to the stars and follow the spiral patterns seen in our simulations. The dissipation and subsequent star formation in the gas would lead to self-consistent spiral structure which would outlast a collisionless disk and work to preserve its structure as well.  An open question is the correct value or range of values to use for \\omh. The figure rotation of NGC 2915 has been calculated to be 8.0 \\kmskpc\\ (Bureau \\etal\\ 1999), which is 0.107 in our $N$-body units. These authors also propose that halo figure rotation, a distinct phenomenon from halo angular momentum, is present in a significant fraction of dark matter halos in CDM simulations.  Regardless of the the details in determining \\omh\\ or \\pave, it has been clearly demonstrated that an external torqueing mechanism can reproduce the type of correlation observed between spiral pitch angle and rotational velocity. Even in simulations in which the magnitude of the applied torque is the same, \\ie\\ the density profile, flattening, and rotation rate of that halo are all held constant, the pitch angle of the spiral structure varies according to the total mass of the system."
    },
    "0209/astro-ph0209486_arXiv.txt": {
        "abstract": "We explore the implications of type Ia supernovae (SNIa) observations on flat cosmological models whose matter content is an exotic fluid with equation of state, $p=-M^{4(\\alpha +1)}/\\rho^{\\alpha}$. In this scenario, a single fluid component may drive the Universe from a nonrelativistic matter dominated phase to an accelerated expansion phase behaving, first, like dark matter and in a more recent epoch as dark energy. We show that these models are consistent with current SNIa data for a rather broad range of parameters. However, future SNIa experiments will place stringent constraints on these models, and could safely rule out the special case of a Chaplygin gas ($\\alpha=1$) if the Universe is dominated by a true cosmological constant. ",
        "introduction": "According to the standard cosmological scenario ($\\Lambda $CDM, QCDM) that has emerged at the end of the last century, the Universe is dominated by two unknown components with quite different properties: pressureless cold dark matter (CDM), which is responsible for the formation of structures, and negative-pressure dark energy, that powers the accelerated expansion. There are several candidates for these two components. For the CDM, the leading particle candidates are the axion and the neutralino, two weakly interacting massive particles. The preferred candidates for dark energy are vacuum energy - or a cosmological constant $\\Lambda $ - and a dynamical scalar field (quintessence) \\cite{denergy}. At the cosmological level, the direct detection of each of these two components involves observations at different scales. Since it is not supposed to cluster at small scales, the effect of dark energy can only be detected over large distances, where the accelerated expansion is observed. On the other hand, the CDM can be detected by its local manifestation on the motion of visible matter or through the bending of light in gravitational lensing.  An interesting question that arises is: could this two phenomena - accelerated expansion and clustering - be different manifestations of a single component? In principle the answer is yes, if, for instance, the Universe is dominated by a component with an appropriate exotic equation of state (EOS). We will generically refer to any kind of such unifying dark matter-energy component as {UDM}. \\footnote{% Following the current jargon, another possible denomination for UDM would be ``quartessence'' since in this scenario we have only one additional component, besides ordinary matter, photons and neutrinos, and not two like in $\\Lambda $CDM and QCDM.}  The above question has been addressed in some works recently \\cite {wetterich,bilic,davidson,padmanabhan}. For instance, in Ref. \\cite {padmanabhan} it was investigated the possibility that a tachyonic field, with motivation in string theory, could unify dark energy and dark matter and explain cosmological observations in small and large scales. Here we investigate observational limits on a simple realization of UDM: a fluid with the following equation of state \\cite{bilic,makler,bento,kamenshchik}, \\begin{equation} p=-\\frac{M^{4(\\alpha +1)}}{\\rho ^{\\alpha }}.  \\label{eostoy} \\end{equation} The particular case $\\alpha =1$ is known as Chaplygin gas and its cosmological relevance, as an alternative to quintessence, has been pointed out in \\cite{kamenshchik}. In \\cite{bilic}, it has been shown that the inhomogeneous Chaplygin gas represents a promising model for dark matter-energy unification. Some possible motivations for this scenario from the field theory point of view are discussed in \\cite {kamenshchik,bilic,bento}. The Chaplygin gas appears as an effective fluid associated with $d$-branes \\cite{kamenshchik,bazeia}. The same EOS is also derived from a complex scalar field with appropriate potential and from a Born-Infeld Lagrangian \\cite{bilic}. More recently, by extending the work of Bili\\'{c} {\\it et al.} \\cite{bilic}, Bento {\\it et al.} \\cite{bento} also discussed the particle physics motivation for the EOS (\\ref{eostoy}). The fluid given by this EOS is sometimes called generalized Chaplygin gas (GCG).  It is interesting to notice that this model can also be obtained from purely phenomenological arguments, by requiring that an exotic fluid unifies the dark-matter/dark-energy behavior as a function of its density and that it is stable and causal \\cite{makler}. The simplest EOS satisfying this criteria is given by eq. (\\ref{eostoy}).  Let us consider the homogeneous case of the GCG Universe. The energy conservation can be written as \\begin{equation} d\\rho=-3\\left( \\rho + p\\right)\\frac{da}{a}, \\end{equation} where $a$ is the scale factor. By solving this equation we may express the energy density in terms of the scale factor: \\begin{equation} \\rho = M^4 \\left[ B \\left(\\frac{a_0}{a}\\right)^{3(\\alpha +1)}+1 \\right] ^{1/( \\alpha +1) },  \\label{rhotoy} \\end{equation} where $a_0$ is present value of scale factor and $B$ is an integration constant. When $a/a_0\\ll 1$, we have $\\rho \\propto a^{-3}$ and the fluid behaves as CDM. For late times, $a/a_0\\gg 1$, and we get $p=-\\rho =-M^4 =const.$ as in the cosmological constant case. There is also an intermediate phase where the effective EOS is $p=\\alpha \\rho$ \\cite{kamenshchik}. Once we have $\\rho$ as a function of the scale factor it is simple to find the Hubble parameter. Since observations of anisotropies in the cosmic microwave background (CMB) indicate that the Universe is nearly flat \\cite{balbi}, here we restrict our attention to the zero curvature case. We also neglect radiation, that it is not relevant for the cosmological tests we discuss in this work.  >From the Friedmann equation with $k=0$ we have \\begin{equation} H^{2}\\left( z\\right) =H_{0}^{2}\\left[ \\Omega _{M}^*\\left( 1+z\\right) ^{3\\left( \\alpha +1\\right) }+\\left( 1-\\Omega _{M}^*\\right) \\right] ^{1/\\left( \\alpha +1\\right) }, \\end{equation} where $z=a_{0}/a-1$ is the redshift, and we have conveniently defined $ \\Omega _{M}^*=B/(B+1)$, or equivalently \\begin{equation} B=\\frac{\\Omega _{M}^*}{1-\\Omega _{M}^*}.  \\label{B} \\end{equation} Further, we also have \\begin{equation} M^{4}=\\rho _{c0}\\left( 1-\\Omega _{M}^*\\right) ^{\\frac{1}{\\left( \\alpha +1\\right) }},  \\label{m4} \\end{equation} where $\\rho _{c0}$ is the present value of the critical density. For these models the deceleration parameter can be written as \\begin{equation} q=-\\frac{\\stackrel{.}{H}}{H^{2}}-1=\\frac{\\frac{\\Omega _{M}^*}{2}-\\left( 1-\\Omega _{M}^*\\right) \\left( 1+z\\right) ^{-3(1+\\alpha )}}{\\Omega _{M}^*+\\left( 1-\\Omega _{M}^*\\right) \\left( 1+z\\right) ^{-3(1+\\alpha )}}, \\end{equation} and the redshift $z_{\\ast }$, at which the Universe started its accelerating phase is given by, \\begin{equation} 1+z_{\\ast }=\\left( \\frac{2\\;(1-\\Omega _{M}^*)}{\\Omega _{M}^*}\\right) ^{\\frac{1 }{3(\\alpha +1)}}. \\end{equation} An accelerating Universe at present time ($q_{0}<0$) implies that $\\Omega _{M}^*<2/3$, and from (\\ref{B}) we have $0<B<2$; the lower limit follows from the fact that we assume $\\Omega _{M}^*>0$. Moreover, if $\\alpha $ is not very close to $-1$, from (\\ref{m4}), we obtain $M\\sim 10^{-3}$ eV. It would be desirable that a fundamental theory, aimed to describe the UDM, sheds some light on the origin of this mass scale. Thus, at this point this model is not free of some tuning. However, once the origin of the above mass scale is explained, the so called dark matter-energy ``coincidence problem'' is not present in this scenario.  In a GCG Universe, if the parameter $\\alpha $ is positive, the adiabatic sound velocity, ${c_{s}}^{2}=dp/d\\rho =-\\alpha p/\\rho $, is real and therefore, the fluid component is stable. If $\\alpha $ is negative and there is only adiabatic pressure fluctuations, they accelerate the collapse producing instabilities that turn the model for structure formation unacceptable \\cite{fabris1,hu}. Moreover, to obey causality, the sound velocity in this medium has to be less or equal than the light velocity. Since the maximum allowed sound velocity of this fluid (which occurs in the regions where $p\\rightarrow -\\rho $) is given by $\\sqrt{\\alpha }$, this condition imposes $\\alpha \\leq 1$. The Chaplygin gas, $\\alpha =1$, is the extreme case, where the sound velocity can be nearly the speed of light. The case $\\alpha =0$ is equivalent to $\\Lambda $CDM and is, of course, well motivated. In this paper, we discuss the GCG model from a phenomenological point of view. Hence, although we are aware that most likely $0\\leq \\alpha \\leq 1$, we also include in our analysis the region where $\\alpha $ is negative, but larger than $-1$. If $\\alpha =-1$ we obtain a de Sitter Universe. The situation $\\alpha <-1$ seems unphysical, since the energy density of UDM would be increasing with the expansion of the Universe. In fact, as we shall see, age constraints can safely exclude regions in the parameter space with very negative values of $\\alpha $.  In the forthcoming section we will see what constraints to the model described above are set by present and future SNIa observations. Recently, some constraints from SNIa on related models where obtained in Ref. \\cite {fabris2}. The work presented here differs from \\cite{fabris2} in the following aspects: a) Following the idea of unification, we have not included an additional dark matter component and we have considered the more general case in which $\\alpha $ is not necessarily equal to unity. b) When analyzing current SNIa data we perform a Bayesian approach in which the intercept is marginalized c) We also investigate the predicted constraints on the models from future SNIa observations. ",
        "conclusions": "We derived constraints, from current and future SNIa observations, in a scenario where both the accelerated expansion and CDM are manifestations of a single component. We considered the special case of a generalized Chaplygin gas. For the homogenous model, an important difference between UDM and models with $\\Lambda $ or scalar fields is that in the former there is a transformation of effective CDM into effective dark energy that produces the accelerated expansion.  Our results show that the GCG is consistent with current SNIa data, for any value of $\\alpha $ in the considered range, although values of $\\alpha \\sim 0.4$ are favored. If the accelerated expansion is caused by a cosmological constant, than SNAP data should be able to rule out the Chaplygin ($\\alpha =1 $) gas and alternatively, if the Universe is dominated by the Chaplygin gas a cosmological constant would be ruled out with high confidence.  For simplicity, we have discussed in this letter the case of a Universe composed of UDM only. Of course, one should also include the baryonic component, whose energy density scales differently from the UDM. When baryons are included in the Hubble parameter the picture does not change, although some details do. For instance, if we introduce $\\Omega_b$ and perform the analysis with the current supernovae data, the results for $\\Omega _{M}^*$ stay almost unchanged, but the best fit value for $\\alpha$ decreases ($\\alpha \\sim 0.15$ for $\\Omega_b \\sim 0.04$, instead of $\\alpha \\sim 0.4$ for $\\Omega_b = 0$). Also, the age constraints on $\\alpha$ are weaker. For instance, for $\\Omega_b = 0.04$ we can exclude negative values of $\\alpha$ close to $-1$ only for $\\Omega _{M}^*\\lesssim 0.3$. In the case of the data expected from SNAP, we redid the analysis of the preceding section for $\\Omega_b=0.04 \\pm 0.004$, assuming a Gaussian distribution. We marginalized over $\\Omega_b$ and noticed that the contours increase only slightly.  The GCG seems to be a promising model for unifying dark matter and dark energy. More generically, the idea of UDM (``quartessence'') has to be explored further, both from the particle physics point of view - to provide a fundamental theory to it -, as well as from the observational side, to constrain UDM models guiding us to unveil its nature.  \\bigskip {\\bf Note added:} After this manuscript was submitted for publication, another paper using the GCG and SNIa obervations appeared on the web \\cite{avelino}. Their results are similar to ours, although they do not set constraints on the parameter $\\alpha$ of the GCG equation of state.  \\begin{figure*} \\hspace*{0.1in} \\psfig{file=fig4.eps,height=10.0cm,width=10.0cm} \\vspace*{0.1in} \\caption{Predicted 68 and 95 confidence level contours for the SNAP mission are shown. We considered a fiducial model with $\\Omega _{M}^*=0.3$ and $\\alpha =1$. For the figure the Hubble intercept is not supposed to be known.} \\end{figure*}  \\bigskip {\\bf Acknowledgments}  IW is partially supported by the Brazilian research agencies CNPq and FAPERJ. SQ is partially supported by CNPq. MM wishes to acknowledge the hospitality of Fermilab."
    },
    "0209/astro-ph0209353_arXiv.txt": {
        "abstract": "The system of neutrino-antineutrino $(\\nu\\bar{\\nu})$ - plasma is considered taking into account their weak Fermi interaction. New fluid instabilities driven by strong neutrino flux in a plasma are found. The nonlinear stationary as well as nonstationary waves in the neutrino gas are discussed. It is shown that a bunch of neutrinos, drifting with a constant velocity across a homogeneous plasma, can also induce emission of lower energy neutrinos due to scattering, i.e. the decay of a heavy neutrino $\\nu_{H}$ into a heavy and a light neutrino $\\nu_L$ ($\\nu_H\\rightarrow\\nu_H\\nu_L$) in a plasma. Furthermore we find that the neutrino production in stars does not lead in general to energy losses from the neutron stars. ",
        "introduction": "Our understanding of the properties of neutrinos in a plasma has recently undergone some appreciable theoretical progress \\cite{Zel83,Ora,Tsin98,TsinP98,Vol}. The interaction of neutrinos with a plasma particles, the creation of $\\nu\\bar{\\nu}$ pairs, the emission of neutrinos due to the collapse of a star are of primary interests \\cite{Zel83,Wil,Bet,Raf,Taj} in the description of some astrophysical events such as supernova explosion. One of the key processes upon the explosions are the large-scale hydrodynamic instabilities as well as $\\nu$ driven plasma instabilities. These processes are also believed to have occurred during the lepton stage of the early universe. During the formation of a neutron star the collapsed core of the supernova is so dense and hot that $\\nu$ and $\\bar{\\nu}$ are trapped and are thus unable to leave the core region of the neutron star. The rates of escape of $\\nu$ and $\\bar{\\nu}$ are very small, inside the star an equilibrium state is reached, which includes the $\\nu\\bar{\\nu}$ concentration. Recently, the neutrino transport phenomenon during the Kelvin-Helmholtz phase of birth of a neutron star in the diffusion approximation was investigated by Pons et al. \\cite{Pons}. A detailed knowledge of transport properties of the neutrinos in these extreme environments must include the neutrino-electron Fermi weak interaction coupling. There are different mechanisms involved in the creation of the $\\nu$, as well as pairs. The collective effects of the stellar plasmas can significantly alter the production rate of neutrinos. It is widely thought that the most dramatic plasma process is the decay of photons as plasmons into neutrino pairs. It was pointed out by Adams et al. \\cite{Ada} that the neutrino pair radiation can be the dominant energy loss mechanism for the plasma of the very dense stars, as well as white dwarfs, red giants and supernovas. It has been also shown \\cite{Gvo} that a $\\nu_H$ can undergo radiative decay into a photon and a $\\nu_L$ ($\\nu_H\\rightarrow\\gamma\\nu_L$) in the presence of a strong magnetic field, with the strength greater than the critical value given by $H_{cr}=m_e^2c^3/e\\hbar= 4.41\\cdot 10^{13}$ Gauss. The increase (decrease) of the $\\nu$ ($\\bar{\\nu}$) energy within the plasma was also demonstrated \\cite{TsinP98}. This may have a significant and potentially detectable effect. Observationally, the recent results from the Kamiokande group provide strong evidence for the existence of neutrino oscillations. These results, together with the increased confidence in ability to produce and manipulate intense muon beams makes feasible the future neutrino factories based on muon storage rings. A muon storage ring, as presently envisioned, would have energy in the 10-50 GeV range, and produce directed beams of intense neutrinos, and naturally such an intense stream of neutrinos can exist in astrophysical and cosmological plasmas. Thus, it is of interest to examine if new physical processes that are caused by the intense flux of neutrinos, can appear in such a plasma, i.e., of particular conceptual interest are effects which have no counterpart in vacuum.   In this paper we consider a medium of neutrino gas and plasma in semitransparent regions, where collective process plays an important role. In our consideration we assume that the interaction between neutrinos is weak in comparison with the neutrino-electron interaction, i.e., the neutrino gas is ideal. We can also ignore the spin of neutrinos in an isotropic plasma, since in this case the energy of neutrinos is independent of the spin operator. With this modeling, here we show that there exists the opposite effect that has been proposed by others \\cite{Ora,Ada}: namely that the neutrinos production does not generally lead to energy losses from a hot and dense system. ",
        "conclusions": "We have studied the problem of weak Fermi interaction of a neutrino beam with a plasma. Novel fluid instabilities driven by the neutrino flux in a plasma are observed. The range of wavelengths corresponding to several instabilities, discussed in this paper, for relevant parameters of the neutrino gas and plasma is $\\lambda\\sim 1-10^{-9}cm$. We have found that for the neutrino beam there exists a Cherenkov type of emission of low energy neutrinos; this is different from the usual Cherenkov effect, though. It should be emphasized that in our case the physical mechanism is that the high energy neutrino does not just excite the plasma wave but also can scatter from density perturbations of electrons and can emit the low energy neutrino. We also found that in the usual hot plasma environments such as supernova explosions, the effect considered above prevents neutrino escape. Since the low energy neutrinos cannot stream away from the central regions of the plasma one can have an accumulation of such neutrinos in a dense stellar medium. Thus, the above discussed processes can considerably modify certain phases of evolution in a stellar model. It should be emphasized that unlike the previous models \\cite{Ada},\\cite{chi} on neutrino pair emission by a stellar plasma, the emission of low energy neutrinos by high energy neutrinos, as discussed in this paper, does not depend on the density and temperature of a plasma. The growth rate found for some typical neutrino parameters is $5.4\\cdot 10^8 sec^{-1}$. Which is much larger than the collision frequency of neutrinos, $10^5 sec^{-1}$. Thus in this case the medium for the neutrino is \"collisionless\".  The nonlinear stationary as well as nonstationary waves in the neutrino gas are also discussed. The dispersive effect, which is responsible for the existence of the nonlinear stationary waves, e.g. solitons, in the neutrino gas is due to the electron density modulation by the neutrino ponderomotive force \\cite{TsinP98}. Our results indicate that a neutrino beam is subject to wave breaking and shock wave formation. Thus, high energy neutrinos will lose energy also by wave breaking in addition to Cherenkov emission, leading to plasma heating (and neutrino cooling). Whereas, the shock waves can produce a relativistic energy flow. In addition, the solitons can also be a potential candidates for the generation of relativistic particles. These particles then release the energy and produce the observed radiation in gamma-ray bursts (GRBs). Hence, one has a origin of radiation in the modeling of burst sources, which requires a discussion of particle acceleration processes. Since the simplest, most conventional, and practically inevitable interpretation of the observations of GRBs is that GRBs result from the conversion of the kinetic energy of ultra-relativistic particles to radiation. It is well accepted that many of them originate in the very distant, early universe. In the early universe, the processes discussed in the paper can also lead to formation of nonlinear structures, contributing to the formation of the large scale structure of the Universe \\cite{Mis}. Which, is believed, grew gravitationally out of small density fluctuations \\cite{wei}. The effect of this density variation in the early universe was left on the cosmic microwave background radiation in the form of spatial temperature fluctuations. It should be emphasized that the gravity, however, cannot produce these fluctuations, but increase alone. Therefore a discussion of the physical models of generation of the initial matter density fluctuations is very crucial. There is no comprehensive model at present that can explain their origin. The weak Fermi interaction in a plasma, as discussed in this paper, can be an alternative and a new source for the required density perturbations. As demonstrated, a long lived nonlinear structures, which carry large amount of mass and energy, are generated in such a system. Since an initial localization of mass and energy is exactly that the gravity needs for eventual structure formation, weak Fermi interaction may have provided a decisive element in the formation of a large scale map of the observable Universe.  \\appendix"
    },
    "0209/astro-ph0209509_arXiv.txt": {
        "abstract": "The Taiwanese-American  Occultation Survey (TAOS) will detect objects in the Kuiper Belt, by measuring the rate of occultations of stars by these objects, using an array of three to four 50cm wide-field robotic telescopes.  Thousands of stars will be monitored, resulting in hundreds of millions of photometric measurements per night. To optimize the success of TAOS, we have investigated various methods of gathering and processing the data and developed statistical methods for detecting occultations.  In this paper we discuss these methods. The resulting estimated detection efficiencies will be used to guide the choice of various operational parameters determining the mode of actual observation when the telescopes come on line and begin routine observations. In particular we show how real-time detection algorithms may be constructed, taking advantage of having multiple telescopes.  We also discuss a retrospective method for estimating the rate at which occultations occur. ",
        "introduction": "Since the middle of the last century, there has been increasing speculation that a residual protoplanetary disk existed beyond Neptune consisting of a vast number of remnants of the accretional phase of the early evolution of the solar system.  This belt is the source of most short--period  comets, those with periods of 200 years or less \\cite{edge}, \\cite{kuiper}, \\cite{fern}. Observational success was first achieved with the discovery of 1992QB1 \\cite{jewitt5}. Major observational efforts since then have identified about 500 objects, the largest having a diameter of about 900km. Studies of the Kuiper belt have been reviewed in  \\cite{weiss}, \\cite{stern} and \\cite{jewitt3}.  At 50AU (one AU is the average distance from the Sun to Earth, $1.49 \\times 10^8$ km,), we can currently only directly observe objects larger than around 100km, since smaller objects do not reflect sufficient light. Thus, other methods are needed to detect smaller objects, which are far greater in abundance. The idea of the occultation technique \\cite{bailey}, \\cite{axel2} is simply the following: One monitors the light from a sample of stars that have angular sizes smaller than the expected angular sizes of Kuiper Belt Objects (KBO's) we hope  to detect. An occultation is manifested by detecting the partial or total reduction in the flux from one of the stars for a brief interval, when a KBO passes between it and the observer. This technique will allow the detection of objects with a radius of only a few kilometers and/or larger objects lying beyond 100AU, objects which have thus far been undetectable by direct observation.   The Taiwanese-American Occultation Survey (TAOS), a collaboration involving the Lawrence Livermore National Laboratory (USA), Academia Sinica and National Central University (both of Taiwan), will use this stellar occultation technique with an array of three or four wide-field robotic telescopes to estimate the number of KBOs of size greater than a few kilometers. Each of these 50cm telescopes will be pointed at the same 3 square degrees of the sky and will record light from the same approximately 2,000 stars. The telescope array will be located in the Yu Shan (Jade Mountain) area of central Taiwan (longitude $ 120 ^{\\circ}$  50' 28'' E; latitude 23$^{\\circ}$ 30' N).  The detection scheme will need to operate in real-time, as the data is being gathered, in order to alert more powerful telescopes for followup. It is anticipated that a large amount of data will be generated on a nightly basis, yielding about 10,000 gigabytes of data and $10^{10}-10^{12}$ occultation tests per year. The challenge is to detect among these only small number of occultations, perhaps tens or hundreds (the uncertainty in this number reflects our ignorance). ",
        "conclusions": "\\label{Conclusion}  We have shown how multiple telescopes may be used to control the false alarm rate for real-time detection, by using thresholds determined in a data-based manner. Realistic simulations illustrated the effectiveness of this procedure and may be used to guide the choice of various operational parameters.  We have argued that once a substantial body of data has been gathered, an alternative method based on the FDR procedure can be used to estimate the occultation rate.  Simulations indicate that this method can produce useful results. It is important to note that different statistical procedures are appropriate for real time detection as compared to retrospective analysis.  The data archive resulting from TAOS will be a valuable resource for other astronomical research as well.  For example, little is known about stellar variability on a sub-second level and TAOS will provide a vast collection of such time series. There may well be unexpected phenomena in the archive and in order not to miss them we face a problem of searching for needles in a haystack when we don't know what needles look like. How can we automate serendipity?   \\newpage"
    },
    "0209/astro-ph0209059_arXiv.txt": {
        "abstract": "We consider the growth of a protoplanetary embryo embedded in a planetesimal disk. We take into account the dynamical evolution of the disk caused by (1) planetesimal-planetesimal interactions, which increase random motions and smooth gradients in the disk, and (2) gravitational scattering of planetesimals by the embryo, which tends to heat up the disk locally and repels planetesimals away. The embryo's growth is self-consistently coupled to the planetesimal disk dynamics. We demonstrate that details of the evolution depend on only two dimensionless parameters incorporating all the physical characteristics of the problem: the ratio of the physical radius to the Hill radius of any solid body in the disk (which is usually a small number) and the number of planetesimals inside the annulus of the disk with width equal to the planetesimal Hill radius (which is usually large). We derive simple scaling laws describing embryo-disk evolution for different sets of these parameters. The results of exploration in the framework of our model of several situations typical for protosolar nebula can be summarized as follows: initially, the planetesimal disk dynamics is not affected by the presence of the embryo and the growth of the embryo's mass proceeds very rapidly in the runaway regime. Later on, when the embryo starts being dynamically important, its accretion slows down similar to the ``oligarchic'' growth picture. The scenario of orderly growth suggested by Safronov (1972) is never realized in our calculations; scenario of runaway growth suggested by Wetherill \\& Stewart (1989) is only realized for a limited range in mass. Slow character of the planetesimal accretion on the oligarchic stage of the embryo's accumulation leads to a considerable increase of the protoplanetary formation timescale compared to that following from a simple runaway accretion picture valid in the homogeneous planetesimal disks. ",
        "introduction": "\\label{sect:intro4}    It is now widely believed (e.g. Safronov 1991, Ruden 1999) that terrestrial planets (planets like Earth and Venus) have formed by agglomeration of numerous rocky or icy bodies called planetesimals (similar to the present-day asteroids, comets, and Kuiper Belt objects). Understanding this accretion process is central to understanding planet formation.  Since the pioneering work of Safronov (1972) it has been known that the dynamics of the planetesimal disk, namely the evolution of planetesimal eccentricities and inclinations, is a critical factor in this process because relative planetesimal velocities determine to a large extent the accretion rate of protoplanetary bodies. The dynamical state of the disk is affected by several processes. Gravitational scattering between planetesimals leads to the growth of their epicyclic energy, thereby increasing relative velocities in the disk. This interaction can proceed in two different regimes depending on the relation between the random velocity of planetesimals $v$ relative to local circular motion and the differential shear across the Hill radius of two interacting bodies $r_H$: \\begin{eqnarray} r_H=a\\left(\\frac{m_1+m_2}{M_c}\\right)^{1/3}, \\label{eq:Hill_def} \\end{eqnarray} where $m_1$ and $m_2$ are the masses of the interacting planetesimals, $M_c$ is the mass of the central star and $a$ is the distance from the central star at which the interaction occurs. Scattering is said to be in the shear-dominated regime when $v\\lesssim \\Omega r_H$, and in the dispersion-dominated regime when $v\\gtrsim \\Omega r_H$ (Stewart \\& Ida 2000), where $\\Omega=\\sqrt{G M_c/a^3}$ is the rotation frequency of a Keplerian disk. Scattering in the first regime is strong for planetesimals separated by $\\sim r_H$ and they can increase their random velocities very rapidly as a result of it. In the second regime scattering is mostly weak even at close encounters because relative velocities are large; in this regime it requires much more  time to change the kinematic properties of planetesimals. Since scattering in the shear-dominated case is so efficient it is likely that this scattering would quickly heat up the disk and bring it into the dispersion-dominated regime. For example, a $50$-km planetesimal (corresponding to a mass $\\approx 10^{21}$ g) at $1$ AU would enter the dispersion-dominated regime for interaction with other planetesimals when its epicyclic velocity is as small as $3$ m~s${-1}$ (corresponding eccentricity is only $10^{-4}$). Thus, it is plausible that at the evolutionary stage when the first massive bodies start appearing in the system, planetesimals interact with each other in the dispersion-dominated regime --- an assumption which we will be using throughout this paper.  Later in the course of the nebular evolution, another dynamical excitation mechanism emerges: gravitational interaction with the newly born massive bodies. We call these bodies embryos (following Safronov 1972) since they are the precursors of the protoplanets; we will require that their masses be greater than the planetesimal masses. When these embryos become massive enough they excite planetesimal velocities in their vicinity and also change the spatial distribution of planetesimals around them. The importance of these effects was first emphasized  by Ida \\& Makino (1993), who used  N-body simulations to study planetesimal disks with embedded embryos.   Depending on the dynamical state of the planetesimal disk, protoplanetary growth can proceed in several different regimes. Safronov (1972) argued that emerging embryos heat the planetesimal disk up to the point where the random velocities of constituent planetesimals are of the order of the escape velocity from the most massive embryos. It is not clear, however, whether the embryo has enough time to increase planetesimal velocities to such large values given inefficiency of scattering in the dispersion-dominated regime. Stewart \\& Wetherill (1988) have noted the importance of the dynamical friction in the redistribution of random energy among planetesimals of different masses and between planetesimals and embryos. They argued in favor of much smaller planetesimal epicyclic velocities which facilitates planetary accretion. However the spatially inhomogeneous character of the dynamical effects of the embryos on the disk were overlooked in their study.  To illustrate the difference between these two scenarios let us estimate the embryo's accretion rate $dM_e/dt$ in the two-body approximation\\footnote{See also Kokubo \\& Ida (1996, 1998) for a similar discussion.} (neglecting the gravity of the central star): \\begin{eqnarray} \\frac{d M_e}{dt}\\simeq\\pi R_e^2 \\Omega m N\\frac{v}{v_z} \\left(1+\\frac{2GM_e}{R_e v^2}\\right), \\label{eq:accr_rate} \\end{eqnarray} where $m$ is the average planetesimal mass, $M_e$ is the mass of the planet, $R_e$ is its radius, $N$ is the surface number density of planetesimals, and $v_z$ is the average planetesimal velocity in the vertical direction (determining the disk thickness and, thus, the local volume density of planetesimals; it is usually of the same magnitude as $v$). The second term in brackets describes gravitational focussing, which can strongly increase the collision cross-section over its geometric value $\\pi R_e^2$ when the planetesimal velocity is smaller than the escape speed from the embryo's surface $\\sqrt{2GM_e/R_e}$. Safronov's (1972) assumption means that focussing is weak. Then one can easily show that \\begin{eqnarray} \\frac{1}{M_e}\\frac{d M_e}{dt}\\propto M_e^{-1/3}, \\label{eq:orderly} \\end{eqnarray} i.e. the embryo's growth rate slows down as its mass increases. This type of planetary growth is known as {\\it orderly} growth, because it implies that many embryos grow at roughly the same rate. On the contrary, if one neglects the embryo's effect on $v$ and $v_z$ and assumes focussing to be strong (following Wetherill \\& Stewart 1988) one finds that \\begin{eqnarray} \\frac{1}{M_e}\\frac{d M_e}{dt}\\propto M_e^{1/3}, \\label{eq:runaway} \\end{eqnarray} i.e. embryo's growth accelerates as its mass increases. This corresponds to the so-called {\\it runaway} accretion regime (Wetherill \\& Stewart 1989), which allows massive bodies grow very quickly\\footnote{Formally the condition (\\ref{eq:runaway}) means that embryo reaches infinite mass in finite time given unlimited supply of material for accretion.} in contrast with the orderly growth picture. Note that the presence of the dynamical friction is not crucial for the onset of runaway growth (it only speeds it up) --- what is important is the regulation of planetesimal velocities by planetesimal-planetesimal scattering, rather than embryo-planetesimal scattering which makes $v$ and $v_z$ independent of $M_e$.  However, it is also possible that effects of the embryo on the velocity dispersion are important but not so strong as orderly scenario assumes; in this case planetesimal velocities can be increased up to several $\\Omega R_H$, where $R_H=a(M_e/M_c)^{1/3}$ is the Hill radius of the embryo (note that it is considerably larger than the Hill radius $r_H$ characterizing planetesimal-planetesimal scattering). Then formula (\\ref{eq:accr_rate}) yields equation (\\ref{eq:orderly}) but with a different proportionality constant. This regime is called {\\it oligarchic} (Kokubo \\& Ida 1998) and is different from orderly growth because of the role of the gravitational focussing (it is very important in this regime) and the values of planetesimal velocities (they are much smaller than the embryo's escape speed). As a result, more massive embryos grow slower than the less massive ones (similar to orderly growth) but embryos still grow faster than planetesimals in their vicinity [similar to runaway growth, see Ida \\& Makino (1993)]. The accretion in the oligarchic regime is considerably slower than in the runaway regime but faster than in the orderly regime. For further convenience we briefly summarize major properties of all 3 accretion regimes: \\begin{itemize} \\item orderly --- growth of velocity dispersion in the disk is dominated by the embryo, gravitational focussing is weak; \\item runaway --- velocity dispersion growth is determined solely by the planetesimal-planetesimal scattering, focussing is strong; \\item oligarchic --- velocity dispersion growth is dominated by the embryo, focussing is strong. \\end{itemize}  The growth timescale of the embryo is a very important quantity. The need for the giant planets to accrete their huge gaseous mass onto their rocky cores (core-instability model, see Mizuno 1980; Stevenson 1982; Bodenheimer \\& Pollack 1986) prior to the dissipation of the gaseous nebula (Hollenbach \\etal 2000) constrains the growth timescale in the giant planet zone  --- between $5$ and $30$ AU in the protosolar nebula --- to be $\\lesssim 10^6-10^7$ yr. This interval must accommodate several distinct evolutionary stages --- planetesimal formation from dust, planetesimal coagulation into protoplanetary embryos, and finally, catastrophic collisions between embryos to form rocky cores of giant planets. Thus, embryos have only part of this time available for their growth. Orderly growth is characterized by very long timescales --- of order $10^8-10^9$ yr (Safronov 1972). On the contrary, runaway growth allows embryos to reach $\\sim 10^{26}$ g (approximately the mass of the Moon) in about $10^5$ yr (Wetherill \\& Stewart 1993). Runaway growth stops at approximately this mass because the embryo accretes all the solid material within its ``feeding zone'' --- an annulus with the width of several $R_H$ (Lissauer 1987). This mass is sometimes called the {\\it isolation} mass $M_{is}$ (see \\S \\ref{subsect:num_setup}).  The dynamical effects of the embryo on the disk, as described numerically by Ida \\& Makino (1993) and Tanaka \\& Ida (1997) and analytically by Rafikov (2001, 2002a, 2002b, hereafter Papers I, II, \\& III correspondingly) and Ohtsuki \\& Tanaka (2002) are likely to change this conclusion. The results of these studies indicate that even if the mass of the embryo is much smaller than the isolation mass it will dominate the dynamical evolution of the nearby region of the disk, which brings the accretion into the much slower oligarchic mode. In this case the timescale required to achieve  $M_{is}$ becomes longer than that predicted by a simple picture of the runaway growth.  In this paper we study the coupled evolution of the planetesimal disk and protoplanetary embryo. Our goal is to determine (1) which conditions need to be fulfilled for the aforementioned regimes to be realized in protoplanetary nebula, (2) whether there are transitions between them and when do they occur, (3) which regime sets the timescale for the planetary growth in the disk. To answer these questions we consider a very simple model of the embryo-planetesimal disk system. We describe the behavior of the disk by representing it as a planetesimal population with a single nonevolving mass $m$ [similar to ``monotrophic'' model of coagulation of Malyshkin \\& Goodman (2001)]. This assumption might seem to be a gross oversimplification but even this toy model can still give us a lot of insight into the details of embryo-disk dynamics. The most important omission for the dynamics of the system which we make by employing this assumption is the absence of dynamical friction within the planetesimal disk. But, as we have mentioned above, this cannot preclude the runaway growth from happening. Also, when runaway or oligarchic regimes take place, the most massive body in the system grows faster than lower mass planetesimals because the growth rate increases with mass in these accretion regimes, see equations (\\ref{eq:accr_rate}) and (\\ref{eq:runaway}). This justifies our assumption of nonevolving planetesimal mass. We devote more time to the discussion of the accuracy of our approximations in \\S \\ref{sect:discussion}.  The spatially resolved evolution of the kinematic properties of the disk and the growth of the embryo's mass are considered simultaneously and self-consistently within the framework of our model. This distinguishes our approach from both conventional coagulation simulations which neglect  spatial nonuniformities of the disk properties caused by the embryo's presence (Wetherill \\& Stewart 1993; Kenyon \\& Luu 1998; Inaba \\etal 2001) and purely dynamical estimates which do not allow the embryo's mass to vary\\footnote{Probably the closest analog of our method is the multi-zone coagulation simulations of Spaute \\etal (1991) and Weidenschilling \\etal (1997).} (Ida \\& Makino 1993; Paper I; Ohtsuki \\& Tanaka 2002). To describe the coupling between the disk dynamics and embryo's mass growth we use a set of equations describing both the planetesimal-planetesimal and the embryo-planetesimal gravitational interactions which were derived in Papers II \\& III correspondingly. Their validity has been checked elsewhere using direct integrations of planetesimal orbits in the vicinity of the embryo (Paper III). These analytical equations are solved numerically in \\S \\ref{sect:numerical}. In addition, to highlight the importance and better understand the behavior of different physical processes operating in the system we provide simple scaling estimates for the evolution of the disk kinematic properties and the growth of the embryo's mass in \\S \\ref{sect:scaling}. We discuss the applications and limitations of our analysis in \\S \\ref{sect:discussion}.  We will often use the model of the Minimum Mass Solar Nebula (MMSN, Hayashi 1981) to obtain typical numerical values of physical quantities. In particular, we will use the following distribution of the surface mass density of the solids in MMSN (Hayashi 1981): \\begin{eqnarray} \\Sigma=20~ \\mbox{g~cm}^{-2}~\\left(\\frac{a}{1~\\rm{AU}}\\right)^{-3/2}, \\label{eq:MMSN} \\end{eqnarray} which is obtained assuming that solids constitute $\\approx 1\\%$ of the nebular mass. ",
        "conclusions": "\\label{sect:conclusion}    In this paper we have self-consistently studied the dynamical evolution of a planetesimal disk coupled to the growth of a single massive protoplanetary embryo. We are able to demonstrate that the evolution of the embryo-disk system proceeds in two consecutive stages: \\begin{itemize} \\item It starts with a rapid runaway growth of the embryo, during which its mass is not large enough to affect the disk dynamics and the spatial distribution of planetesimals. On a rather short timescale (comparable to that arising in conventional coagulation simulations) the embryo reaches the mass at which it takes over the control of the disk heating and runaway growth stops. \\item After a short transient stage, the embryo-disk system settles into the asymptotic regime in which the timescale of the embryo's mass growth is comparable to the timescale on which the disk is heated by the embryo's gravity. \\end{itemize} During the last stage, the embryo dynamically excites planetesimal epicyclic motions in its vicinity and repels the planetesimals forming a depression in the surface density of the guiding centers (see Figure \\ref{fig:time_ev}). This effect produces a negative feedback for the embryo's accretion rate for two reasons: rapid growth of the random velocities of planetesimals leads to weaker gravitational focussing while the decrease of the instantaneous surface density (see Figure \\ref{fig:ninst_3_3}) reduces the amount of material available for accretion. As a result, growth of the protoplanetary embryo proceeds slower than in homogeneous planetesimal disks. This implies that conventional ``particle-in-a-box'' coagulation simulations (Wetherill \\& Stewart 1989; Kenyon \\& Luu 1998; Inaba \\etal 2001) might be underestimating the timescale of the protoplanetary embryo's growth, sometimes by rather large factors (up to $\\sim 10$). Multi-zone coagulation simulations (Spaute \\etal 1991; Weidenschilling \\etal 1997) should give a more reliable description of the protoplanetary disk evolution.   We have presented our equations in a dimensionless form which has allowed us to uncover a very important property of the problem we consider: its outcome depends on only two dimensionless parameters --- the number of planetesimals inside the annulus of the disk with width equal to the planetesimal Hill radius $r_H$, $N a r_H$, and the ratio of the physical to Hill radii of a solid body (planetesimal or embryo) $p$. All astrophysically relevant characteristics of the system ($m$, $a$, $M_c$, $\\Sigma$, $\\rho$) are combined into these two parameters which greatly simplifies the exploration of the parameter space. We have numerically studied in \\S \\ref{sect:numerical} the evolution of the system for a number of pairs of $N a r_H$ and $p$; combined with our analytical developments in \\S \\ref{sect:scaling} these results allow us to formulate a set of simple scaling relations (\\ref{eq:star_eredict})-(\\ref{eq:m_tau_eredict}) which can be used to predict the embryo-disk evolution for different sets of $N a r_H$ and $p$.  To summarize we can say that the embryo's growth starts in the runaway regime, but then switches into the  ``oligarchic'' growth of Kokubo \\& Ida (1998). We have demonstrated that the embryo's accretion never proceeds in the orderly regime with weak gravitational focussing suggested by Safronov (1972) --- planetesimal velocities are always smaller than the escape velocity from the embryo's surface. The embryo formation timescale and some details of the embryo-disk evolution are subject to a number of uncertainties related to the simplicity of our model, as we discussed in \\S \\ref{sect:discussion}. Nevertheless, we believe that our major conclusions are robust: the results of our qualitative analysis of \\S \\ref{sect:scaling} and numerical calculations presented in \\S \\ref{sect:numerical} suggest that the dynamical interaction between the protoplanetary embryo and planetesimal disk is a very important issue which should certainly be addressed when realistic coagulation simulations are performed. The general agreement between the results of our simple  analysis of \\S \\ref{sect:scaling} with more accurate numerical calculations of \\S \\ref{sect:numerical} encourages the studies of more complex and more realistic problems in the same spirit. In the future we are going to improve our treatment of the coupled planetesimal disk and embryo evolution by relaxing simplifying assumptions employed in this paper. We are also going to include additional mechanisms which are important in modelling of planetary formation --- effects of multiple embryos, migration, gas drag, etc."
    },
    "0209/astro-ph0209573_arXiv.txt": {
        "abstract": "We present new theoretical evaluations of optical and near-IR Surface Brightness Fluctuations (SBF) magnitudes for single-burst stellar populations in the age range $t=5-15$ Gyr and metallicity from $Z_{\\sun}/200$ to $2Z_{\\sun}$. Our theoretical predictions can be successfully used to derive reliable distance evaluations. They also appear to be a new and valuable tool to trace the properties of unresolved stellar populations. ",
        "introduction": "The SBF method is commonly used to derive accurate extragalactic distances (Tonry et al. 2001). The basic idea arises from the evidence that the spatial distribution of the light of nearby galaxies is ``bumpy'', while in more distant ones it appears quite smooth. On CCD images of galaxies the level of bumpiness was quantified by Tonry \\& Schneider (1988) by defining the apparent SBF magnitudes $\\bar{m}=-2.5~log \\bar{f}$, where $\\bar{f}$ is the ratio of the pixel to pixel flux variance to the average pixel flux. This value depends on the \\emph{number} and \\emph{kind} of unresolved stars actually ``located'' inside the pixels. Thus, if the absolute SBF magnitudes are evaluated by using a stellar population synthesis code (as we do), two major information on distant galaxies can be inferred: \\emph{i) the distance }and \\emph{ii) the stellar population properties}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209329_arXiv.txt": {
        "abstract": "We derived expressions for energy, momentum and angular momentum losses due to gravitational and electromagnetic radiation from the closed superconducting chiral cosmic strings of arbitrary form. The expressions for corresponding radiation rates into the unit solid angle have the form of four-dimensional integrals. In the special case of piece-wise linear strings these formulas are reduced to sums over the kinks. We calculate numerically the total radiation rates for three examples of string loops in dependence of current along the string. ",
        "introduction": "We study the gravitational and electromagnetic radiation of energy, momentum and angular momentum of superconducting closed cosmic strings with chiral current. Cosmic strings are linear topological defects, that may have been created during phase transitions in the early Universe (see e.~g. reviews in \\cite{Vilenkin2,Kibble2}). Oscillating ordinary cosmic strings (without current) radiate only gravitational waves. The corresponding energy radiation  was studied in \\cite{Vachaspati1,Burden1,Garfinkle1,Vilenkin5,Allen1,Allen2}. Besides energy, gravitational waves from strings take also momentum \\cite{Vachaspati1,Allen3,Durrer} and angular momentum \\cite{Durrer}. It was found that the rates (averaged per oscillation period) of energy $\\dot{E}$, momentum $\\dot{P}$ and angular momentum $\\dot{L}$ losses can be expressed in the following form: $\\dot{E}^{\\rm{gr}}=\\Gamma_{E}^{\\rm{gr}} G\\mu^{2}$, $\\dot{P}^{\\rm{gr}}=\\Gamma_{P}^{\\rm{gr}} G\\mu^{2}$, $\\dot{L}^{\\rm{gr}}= \\Gamma_{L}^{\\rm{gr}} \\mathcal L G\\mu^{2}$, where $\\Gamma_{E}^{\\rm{gr}}\\sim 100$, $\\Gamma_{P}^{\\rm{gr}}\\sim 10$ and $\\Gamma_{L}^{\\rm{gr}}\\sim 10$ are numerical coefficients, depending on the particular string configuration, $\\mathcal{L}$ is length of the string, $\\mu$ is string mass per unit length and used units $\\hbar=c=1$.  Witten \\cite{Witten1} showed that in some field theory models the cosmic strings can carry superconducting current which is coupled to some gauge field. In the case of electromagnetic gauge field the superconducting cosmic string loops would radiate not only gravitational waves, but also electromagnetic ones. Equations of motion of superconducting cosmic strings can be solved analytically \\cite{Carter1,Davis2,Vilenkin3} if (i) the influence of gauge field on the string motion is negligible and (ii) the current on the string $j^{\\mu}$ is chiral, i.~e. $j^{\\mu} j_{\\mu} = 0$. The gravitational and electromagnetic energy radiated by the single cusp on chiral cosmic string was studied by Blanco-Pillado and Olum \\cite{Blanco} in the case of small current. The opposite case for current which is close to maximum value was considered in \\cite{Babichev}. If the string carries the current, then the coefficients $\\Gamma_{E}^{\\rm{gr}}$, $\\Gamma_{P}^{\\rm{gr}}$ and $\\Gamma_{L}^{\\rm{gr}}$ which determine the gravitational radiation depend on the current along the string. Corresponding expressions for electromagnetic radiation have the similar form: $\\dot{E}^{\\rm{em}}=\\Gamma_{E}^{\\rm{em}} \\mu q^{2}$, $\\dot{P}^{\\rm{em}}=\\Gamma_{P}^{\\rm{em}} \\mu q^{2}$, $\\dot{L}^{\\rm{em}}=\\Gamma_{L}^{\\rm{em}} \\mathcal{L} \\mu q^{2}$, where numerical coefficients $\\Gamma_{E}^{\\rm{em}}$, $\\Gamma_{P}^{\\rm{em}}$ and $\\Gamma_{L}^{\\rm{em}}$ also depend on the current on the string.  In this paper we present the results for gravitational and electromagnetic radiation of energy, momentum and angular momentum from chiral cosmic string loops for any value of superconducting current. Due to the periodic motion of the cosmic string loops the rates of radiation losses can be expand in series $\\dot{E}=\\sum\\dot{E}_{l}$, $\\dot{P}=\\sum\\dot{P}_{l}$, $\\dot{L}=\\sum\\dot{L}_{l}$. Here $\\dot{E}_{l}$, $\\dot{P}_{l}$ and $\\dot{L}_{l}$ are correspondingly the energy, momentum and angular momentum rates in the $l$-th radiation mode. Usually the total rates per unit time (averaged over the period) are calculated by summing of losses in different modes. In practical numerical calculations the values of $\\dot{E}$, $\\dot{P}$ and $\\dot{L}$ are determined with the accuracy up to the $l$ of a few hundred. Such calculations may be not correct because of the slow convergence of the corresponding sums over $l$ as was pointed out by Allen et al. \\cite{Allen1}. See however \\cite{Allen1,Allen2,Allen3} where for some special cases of ordinary string loops the summation over $l$ was done and analytic expressions for the total energy and momentum rates into gravitational waves were obtained. We perform in the following the summation over radiation modes analytically and derive the formulas for energy, momentum and angular momentum rates into the both the gravitational and electromagnetic radiation from chiral string loops of general configuration. The corresponding radiation rates into the unit solid angle are reduced to the four-dimensional integrals which in general case can be calculated only numerically. For chiral piece-wise linear loops these formulas lead to analytic expressions for the energy, momentum and angular momentum radiation into the unit solid angle.  We considered three examples of chiral string loops and calculated the total radiated energy, moment and angular moment per unit time in dependence of current on the string. The first and the second examples are piece-wise linear loops (or \"kinky\" loops), and the third example is a hybrid of piece-wise loop and smooth loop (namely, the $a$-loop is smooth and $b$-loop is piece-wise loop). Unfortunately we are unable to present the results for kinkless cosmic loops (i.~e. for smooth $a$ and $b$-loops), because it would take an enormous amount of computer time.  The paper is organized as follows. In Section~\\ref{sec:11} we review some general properties of chiral cosmic strings. In Section~\\ref{sec:gr} we derive new expressions for gravitational radiation rates of energy, momentum and angular momentum by chiral loops of general configuration into the unit solid angle. These expressions are reduced to the four-dimensional integrals where the summation over all radiation modes were performed analytically. In Section~\\ref{sec:el} we derive the similar formulas for electromagnetic radiation rates. In Section~\\ref{sec:ex} we present numerical calculations of electromagnetic and gravitational radiation rates for some illustrative examples of chiral loops and study the properties of chiral strings radiation in dependence of current. In conclusion Section~\\ref{sec:co} we describe the obtained results and discuss some qualitative features of gravitational and electromagnetic radiation from chiral loops. ",
        "conclusions": "\\label{sec:co}  We found the general formulas for gravitational and electromagnetic energy, momentum and angular momentum radiation rates into the unit solid angle from chiral cosmic string loops. The main new result of the paper is the presentation of the radiation rates from oscillating string loops in the integral form. In the corresponding integrals the summations of infinite mode series have been performed analytically. The derived expressions for $d \\dot{E}/d\\Omega$, $d\\dot{\\mathbf{P}}/d\\Omega$ and $d\\dot{\\mathbf{L}}/d\\Omega$ contain four-dimensional integrals, which depend on the particular loop configuration. The derived integral presentation is especially convenient for numerical calculations in comparison with a weakly convergent summation over modes. To find the total rates of radiated energy, momentum and angular momentum one should integrate the obtained expressions over unit sphere. The final expressions for gravitational and electromagnetic radiation can be written in the following form: \\begin{eqnarray} \\label{gamma1} \\dot{E}^{\\rm{gr}}=\\Gamma_{E}^{\\rm{gr}} G\\mu^{2},\\, \\dot{P}^{\\rm{gr}}=\\Gamma_{P}^{\\rm{gr}} G\\mu^{2},\\, \\dot{L}^{\\rm{gr}}=\\Gamma_{L}^{\\rm{gr}} \\mathcal L G\\mu^{2},\\nonumber\\\\ \\dot{E}^{\\rm{em}}=\\Gamma_{E}^{\\rm{em}} \\mu q^{2}, \\dot{P}^{\\rm{em}}=\\Gamma_{P}^{\\rm{em}} \\mu q^{2}, \\dot{L}^{\\rm{em}}=\\Gamma_{L}^{\\rm{em}} \\mathcal L \\mu q^{2}, \\end{eqnarray} where numerical coefficients $\\Gamma$ depend on the loop configuration (and also on the current along the loop). Applying our formulas to some examples of chiral string loop configurations we calculated numerically coefficients $\\Gamma$ as functions of $k$.  In this paper the following three family of examples have been considered: (i) the piece-wise linear kinky loop with $a$ and $b$-loop consisting of two straight parts (2-2 piece-wise loop); (ii) the piece-wise linear loop such that $a$-loop consists of two segments and $b$-loop consists of three segments (2-3 piece-wise loop); (iii) the hybrid loop in which $a$-loop is circle and $b$-loop consists of two straight parts (hybrid kinky loop). For first and second examples the four-dimensional integrals in our expressions for radiated energy, momentum and angular momentum become the multiple sums over the kinks. These sums can be analytically calculated using the symbolic computation on computer (e.~g. ``Mathematica'' packet). To obtain the radiation for third example (hybrid loop) we calculated two-dimensional integrals (originated from the smooth $a$-loop) and summed over the kinks of $b$-loop. Unfortunately, we could not carry out the calculations for strings with $a$ and $b$ loops being arbitrary smooth closed curves because the corresponding calculations of four-dimensional integrals take an enormous amount of time.  For considered examples we observe weak oscillations of the electromagnetic radiation as a function of mode number. These oscillations (accompanied with a general decreasing of the radiation rate with mode number) have the different periods depending on the current along the string: the larger current the smaller the period of oscillations. This effect does not take place for gravitational radiation.  The total gravitational radiation of energy, momentum and angular momentum behave in a similar way. They increase slowly with $k$, when $k$ is small (and the current is large) and rapidly increase at large $k$ (or at large current). In total the gravitational radiation rates are increasing monotonous functions of $k$. For the electromagnetic radiation the situation is different: the losses of energy, momentum and angular momentum into electromagnetic waves for all considered examples have maximum near $k\\sim 0.9$, i.~e. when the current is rather small. For considered examples the maximal coefficients in (\\ref{gamma1}) have the following values: \\begin{eqnarray} \\label{gamma2} \\Gamma_{E}^{\\rm{gr}}&\\simeq& 50, \\quad \\Gamma_{P}^{\\rm{gr}}\\simeq 1, \\quad \\Gamma_{L}^{\\rm{gr}}\\simeq 3, \\nonumber\\\\ \\Gamma_{E}^{\\rm{em}}&\\simeq& 2, \\quad \\Gamma_{P}^{\\rm{em}}\\simeq 0.1, \\quad \\Gamma_{L}^{\\rm{em}}\\simeq 0.1. \\end{eqnarray}  We also have found, that for some non-symmetric examples of chiral loops, the radiated angular momentum $\\dot{\\mathbf{L}}$ into electromagnetic and gravitational waves is not exactly opposite to the angular momentum of the loop $\\mathbf{L}_{st}$, but slightly differ from it (even when there is no current on the string), unlike the other types of loops considered by Durrer \\cite{Durrer}."
    },
    "0209/astro-ph0209603_arXiv.txt": {
        "abstract": "We compare the energy and count fluxes obtained by integrating over the finite bandwidth of BATSE with a measure proportional to the bolometric energy flux, the $\\varphi$-measure, introduced by Borgonovo \\& Ryde.  We do this on a sample of 74 bright, long, and smooth pulses from 55 GRBs.  The correction factors show a fairly constant behavior over the whole sample, when the signal-to-noise-ratio is high enough. We present the averaged spectral bolometric correction for the sample, which can be used to correct flux data. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209372_arXiv.txt": {
        "abstract": "We present an up-to-date critique of the physical basis for the spectral ageing method.  We find that the number of cases where this method may be meaningfully applied to deduce the ages of classical double radio sources is small indeed.  This critique is much more than merely a re-expression of anxieties about the calibration of spectral ageing (which have been articulated by others in the past). ",
        "introduction": "Many people (e.g.\\ Winter et al.\\ 1980, Myers \\& Spangler 1985, Alexander \\& Leahy 1987) have observed that spectral indices change along the lobes of classical double radio galaxies.  The general trend observed is that the lobe spectra are flatter in the outermost regions near the hotspot and steeper in the regions nearer the core.  Often the observed change in spectral index, or the spectral gradient, is steady and systematic.  \\subsection{Two {\\em a priori} interpretations}  It is not widely acknowledged that there are in principle two physical interpretations of this behaviour.  The traditional interpretation of spectral gradients goes as follows: the radiating electrons nearer the core were dumped by the hotspot much earlier in the past than the radiating electrons near the hotspot now, and so the former will have undergone greater synchrotron cooling compared with the latter.  A radiating population whose energy distribution is initially a power-law, which suffered only synchrotron losses, would show a `break' in this power-law at later times.  This break frequency moves to lower frequencies as more time elapses (Kardashev 1962, Pacholczyk 1970, Jaffe \\& Perola 1973) predicting steeper measured spectral indices for the older emission. Thus far, there is consistency with observations.  But an alternative physical picture explains the observations just as well: a gradient in magnetic field along the lobe together with a curved energy electron spectrum will result in a spectral gradient being observed along the lobe (see Fig.\\,1).  Indeed, Rudnick, Katz-Stone and Anderson's (1994) analysis of multi-frequency images of Cygnus\\,A show no evidence for any variation in the shape of $N(\\gamma)$ across different regions of the lobe.  \\begin{figure} \\hbox{ \\psfig{figure=Blundell.fig2.eps,width=0.9\\textwidth} } \\caption{Illustration of how, when one observes a particular frequency $\\nu$ across the lobe, the combination of a gradient in magnetic field together with a curved electron energy distribution $N(\\gamma)$ will inevitably lead to a gradient in spectral index $\\alpha$ ($ \\equiv \\partial \\log S_{\\nu} / \\partial \\log \\nu $).} \\end{figure}  Without considering the underlying physics more deeply, one {\\em cannot} distinguish between these two possibilities.  So we now examine in turn the individual and collective assumptions which go into these two pictures. ",
        "conclusions": "The traditional way of interpreting observed gradients in spectral index along the radio lobes is the simple spectral ageing picture; we find that the assumptions which underlie this model, both individually and collectively, are flawed.  A combination of a gradient in magnetic field (which is physically plausible) together with a curved distribution in electron energy (which is measured) produces the same observed behaviour.  In addition, this second model explains not just the observed spectral gradients {\\em above} the break frequency but also those {\\em below} the break frequencies."
    },
    "0209/astro-ph0209144_arXiv.txt": {
        "abstract": "{We present JHK images of the young ($\\sim 20$ Myrs) nearby ($\\sim 48$ pc) stars HD 199143 and HD 358623 (van den Ancker et al. 2000) with high sensitivity and high dynamic range in order to search for (sub-)stellar companions around them. The images were obtained in JHK with the speckle camera SHARP-I in July 2001 and in H with the infrared imaging camera SofI in December 2000, July 2001, and December 2001, all at the ESO 3.5m NTT. We present a companion candidate with a 2 arcsecond offset being almost 2 mag fainter than HD 358623 with proper motion (over one year baseline) consistent with the known proper motion of the primary star HD 358623 A and $6 \\sigma$ deviant from the assumption that it is a non-moving background object. Then, we obtained a spectrum in the H-band (with SofI at the NTT) of this companion showing that it has spectral type M2 ($\\pm 1$), consistent with its JHK colors (for negligible extinction) and with being a companion (i.e. at same age and distance) of HD 358623 A (K7-M0), given the magnitude difference. Also, a companion candidate with a 1 arcsecond offset being 2 mag fainter than HD 199143 is detected, but clearly resolved from the primary only with SHARP-I, so that we have no proper motion information. Also, we could not obtain a spectrum with SofI due to the high dynamic range. The JHK colors of this candidates and the magnitude difference between primary (F8) and companion candidate are consistent with a spectral type M0-2. This companion candidate was predicted by van den Ancker et al. (2000) to explain some unusual properties of the primary star HD 199143 A. We determine the ages and masses of all four objects from theoretical tracks and isochrones, all four stars appear to be co-eval with $\\sim 20$ Myrs. Both these companions were presented previously by Jayawardhana \\& Brandeker (2001), but only single-epoch images in J and K obtained in May 2001. Our limits for additional detectable but undetected companions are such that we would have detected all stellar companions with separations $\\ge 0.5^{\\prime \\prime}$ (24 AU at 48 pc). ",
        "introduction": "HD 199143 and HD 358623 were presented as young, nearby, co-moving stars by van den Ancker et al. (2000; henceforth vdA00). HD 199143 is at a distance of $47.7 \\pm 2.4$~pc as measured by Hipparcos. Because HD 199143 and HD 358623 are only five arc min apart from each other and both show activity (a youth indicator) and the same proper motion, vdA00 suggested that they form a (small, but possibly larger) young nearby moving group (Capricornius), similar to the HorA, Tuc, TW Hya, and $\\beta$ Pic moving groups. HD 199143 has similar $UVW$ space motion as the Tuc and TW Hya stars (vdA00; Zuckerman \\& Webb 2000). Four more member candidates were presented by van den Ancker et al. (2001; henceforth vdA01) selected within $5^{\\circ}$ around HD 199143 by strong ROSAT X-ray emission and, partly, by proper motion. It would be very important to find more members to this new association, not only to study the formation and mass function of these new young nearby groups, but also because such young nearby stars are very well suited for direct imaging searches of sub-stellar companions.  HD 199143 (also called BD$-17^{\\circ}6127$ and SAO 163989) has spectral type F8 (Houk \\& Smith-Moore 1988; vdA00) and displays anomalous ultra-violett emission, Ca H \\& K emission, and fast rotation (vdA00). Also, HD 199143 shows N- and Q-band excess (vdA01) as well as IRAS $12 \\mu m$ excess emission, but very low upper limits in the other IRAS bands (vdA00), so that this star shows moderate, but not strong infrared (IR) excess emission. However, vdA00 argue that all its features could be explained by a faint T-Tauri-like companion, which produces itself the UV and IR excess emission and whose circumstellar material accretes onto HD 199143 and thereby spins up the primary. Mora et al. (2001) measured a rotational velocity of $v \\cdot \\sin i = 155 \\pm 8$ km/s.  HD 358623 (also called BD$-17^{\\circ}6128$ and AZ Cap) has spectral type K7-M0 with H$\\alpha$ emission, strong flaring, and strong Li 6708\\AA~absorption (Mathioudakis et al. 1995; vdA00), as well as N- and Q-band excess (vdA01), all typical for a T~Tauri star. The Tycho proper motion of HD 358623 ($\\mu _{\\alpha} = 59 \\pm 3$ mas/yr and $\\mu _{\\delta} = -63 \\pm 3$ mas/yr) is identical to that of HD 199143 ($\\mu _{\\alpha} = 59.2 \\pm 1.1$ mas/yr and $\\mu _{\\delta} = -61.55 \\pm 0.85$ mas/yr). Recently, Zuckerman et al. (2001) argued that both HD 199143 and HD 358623 are actually part of the $\\beta$ Pic moving group.  In May 2001, Jayawardhana \\& Brandeker (2001; henceforth JB01) observed the two stars in J \\& K with the AO system ADONIS at the ESO 3.6m telescope on La Silla. Near each of the two stars, they detected one companion candidate. The close and faint object near HD 199143 is red ($J-K=1.4$ mag) and, hence, either sub-stellar or the (reddened) companion expected by vdA00, whose circumstellar material could explain the anomalous features of the primary HD 199143. The companion candidate near HD 358623 is less than 2 mag fainter than the primary in J \\& K (JB01) and could be an M-type stellar companion.  We observed the two stars with high sensitivity and high dynamic range in order to detect faint companions, namely in the H-band in Dec. 2000, July 2001, and Dec. 2001 and in the J- and K-bands in July 2001 (speckle and normal IR imaging). Young nearby stars like HD 199143 and HD 358623 are well-suited for direct imaging of sub-stellar companions, both brown dwarfs and giant planets, because young sub-stellar objects are still relatively bright (e.g. Burrows et al. 1997), so that they are less difficult to detect in the PSF wing of a much brighter star. See, e.g., Lowrance et al. (1999, 2000), Neuh\\\"auser et al. (2000b), and Guenther et al. (2001) for imaging and spectroscopy of brown dwarf companions of the young stars TWA-5 and HR 7329 in the TW Hya and Tuc associations. The brown dwarf near TWA-5 was the first sub-stellar companion around a pre-main sequence star confirmed by both proper motion and spectroscopy, and also the first around a spectroscopic binary (Torres et al. 2001), and the first around a star with evidence for a disk (Jayawardhana et al. 1999). While the first four brown dwarfs confirmed as companions all orbit M-type stars, an A-type star also can have a brown dwarf companion (e.g. HR 7329 with spectral type A0). Therefore, both HD 199143 (F8) and HD 358623 (K7-M0) are promising targets for the direct imaging search for sub-stellar companions.  The probability for the two companion candidates detected by JB01 to be unrelated background objects happen to lie in the line-of-sight next to the primary stars is very small (JB01). However, one should not rely on such probabilities, even when observing only a small sample. Some previous very faint sub-stellar companion candidates (e.g. Terebey et al. 1998, Neuh\\\"auser et al. 2000a) with very low background probability were found to be background stars by follow-up spectroscopy (Terebey et al. 2000, Neuh\\\"auser et al. 2001). This shows how important it is to take multi-epoch images and spectra.  We present our imaging observations in Sect. 2 and the resulting photometry for the two stars and their companion candidates in Sect. 3. Astrometry is presented in Sect. 4 to check whether the companion candidates are co-moving with their putative primary stars. Then, in Sect. 5, we present an H-band spectrum of one of the two companion candidates. We conclude in Sect. 6. ",
        "conclusions": "We have shown that the HD 358623 primary A and its companion candidate B indeed show the same proper motion and that the spectral type of the companion (M2) is consistent with the observed colors and magnitude differences, so that it is a truely bound companion.  We would like to point out again the high precision achieved in the relative astrometry: After just one year, we could measure the proper motion ($\\sim 100$ mas) of both HD 358623 A and B with sufficient precision to show that they form a common proper motion pair, using the 150 mas pixel scale of the SofI small field and several non-moving background stars.  \\begin{figure} \\vspace{-3cm} \\vbox{\\psfig{figure=dyn_range_vda.ps,width=15cm,height=11cm,angle=270}} \\caption{Dynamic ranges achieved. We plot the log of the flux ratio between the $3 \\sigma$ background noise level and the peak intensity of HD 358623 (and, on the right hand side, the magnitude difference) versus the separation to the primary's photocenter (and, on the top axis, the projected physical separation at 48 pc), for the two detectors used: Sofi at the La Silla NTT (H-band) and SHARP-I at the NTT (K-band). The dotted lines show the expected flux ratios for 42 and 13~M$_{\\rm Jup}$ masses (according to Burrows et al. 1997) for 48 pc and 20 Myrs, next to HD 358623. Every stellar companion above the H burning mass limit (75~M$_{\\rm Jup}$) would have been detected outside of $0.5^{\\prime \\prime}$ (24 AU at 48 pc). The dynamic range curve for SHARP-I is limited by the $6^{\\prime \\prime} \\times 6^{\\prime \\prime}$ size of the quadrants (we placed the primaries only onto the two best (lower, western) quadrants).} \\end{figure}  The mean apparent angular separation between HD 358623 A and B ($2.205 \\pm 0.028^{\\prime \\prime}$) corresponds to a projected physical separation of $105.2 \\pm 6.6$~AU (at the Hipparcos distance of HD 199143, which is presumably the same as for HD 358623, see vdA00). The projected physical separation between HD 199143 A and B is $48.8 \\pm 3.9$~AU, namely $1.023 \\pm 0.031^{\\prime \\prime}$ (our separation measured with SHARP-I) at $47.7 \\pm 2.4$~pc.  Let us investigate the sensitivity limits determined for the dynamic range achieved in the images: The flux ratio is determined from our actual images of the two stars in all SofI and SHARP-I images as the $3 \\sigma$ background noise level on $7 \\times 7$ pixel boxes as approximate PSF areas and devided by the peak intensity. We compare the observed dynamic ranges with expected flux ratios for possible companions of different masses (calculated following Burrows et al. 1997) next to HD 358623 (Fig. 5). The MPE speckle camera SHARP-I clearly gives the best dynamic range. In the SHARP images, we should have detected all stellar companions above $\\sim 0.1$~M$_{\\odot}$ outside of $\\sim 0.5^{\\prime \\prime}$. Brown dwarf companions with $\\sim 25$~M$_{\\rm Jup}$ would have been detectable at $\\sim 3^{\\prime \\prime}$ separations, more massive ones at smaller separations (between $\\sim 0.5$ and $3^{\\prime \\prime}$).  \\begin{table} \\begin{tabular}{lccccc} \\multicolumn{6}{c}{\\bf Table 4. Physical properties.} \\\\ \\hline Object      & T$_{\\rm eff}$ & B.C. & L$_{\\rm bol}$ & mass & age \\\\ & [K]  & [mag] & [L$_{\\odot}$]     & [M$_{\\odot}$] & [Myr] \\\\ \\hline HD 199143 A & 6200 & 0.16  & $2.40 \\pm 0.25$   & $\\sim 1.25$ & $\\sim 20$ \\\\ HD 199143 B & 3720 & 1.43  & $\\sim 0.1$        & $\\sim 0.60$ & $\\sim 20$ \\\\ \\hline HD 358623 A & 3955 & 1.00  & $0.24 \\pm 0.5$    & $\\sim 0.90$ & $\\sim 20$ \\\\ HD 358623 B & 3580 & 1.64  & $0.05 \\pm 0.01$   & $\\sim 0.55$ & $\\sim 20$ \\\\ \\hline \\end{tabular} \\end{table}  Next, we can compute the luminosities of the four objects studied. We assume the Hipparcos distance towards HD 199143 A ($47.7 \\pm 2.4$ pc) for all four objects. From the known $V-H$ color indices and/or known spectral types, we can estimate the effective temperatures and bolometric corrections B.C. (taken from Kenyon \\& Hartmann 1995). Temperatures, B.C., and luminosities are listed in Table 4.  We placed the stars into the H-R diagram and compared their locations with theoretical tracks and isochrones by Palla \\& Stahler (1999) and Baraffe et al. (1998) to estimate masses and ages. Rough values are given in Table 4. All four stars appear to be co-eval with an age of $\\sim 20$ Myrs.  In conclusion, we find that all our data are consistent with HD 199143 and HD 358623 each having an early-M type stellar companion. In the case of HD 358623 B, the spectral type is confirmed by JHK colors and a spectrum and companionship is also confirmed by common proper motion. For HD 199143 B, spectrum as well as proper motion and, hence, companionship, still have to be confirmed."
    },
    "0209/astro-ph0209234_arXiv.txt": {
        "abstract": "\\vspace*{0.25cm} We report on a VLBA imaging study of the nearby bright southern blazar PKS~1921--293 (OV--236). High resolution VLBA observations, made at four frequencies (5, 12, 15, and 43~GHz) over the period 1994-2000, have revealed a strongly curved jet extending out to about 50 parsecs from the presumed central engine. Two epoch VLBA observations, each simultaneously carried out at both 5 and 43\\,GHz, show a large position angle difference of 51$^\\circ$ -- 67$^\\circ$ between the jet emission at 5 and 43\\,GHz. Although the core of PKS~1921--293 has one of the highest brightness temperatures measured in any compact radio source, unlike other bright blazars it is not a source of $\\gamma$--ray emission. However, there is evidence in these images for superluminal motion within the central region (a few parsecs from the core) and within the north-east diffuse emission region. In all six-epoch 43\\,GHz images, two equally compact bright components within the central parsec are seen. \\\\ ",
        "introduction": "\\vspace*{0.25cm} PKS~1921--293 (OV--236) is identified with a 17.5 V-magnitude quasar. At a redshift of 0.352 (Wills \\& Wills 1981), PKS\\,1921$-$293 is one of the closest members of its class. An angular resolution of 1~mas corresponds to a linear resolution of 4.6~pc (assuming H$_0$~=~65~km~s$^{-1}$~Mpc$^{-1}$ and q$_0$~=~0.5).  PKS~1921--293 is one of the strongest and most compact extragalactic radio sources known, which makes it a prime candidate for high-resolution VLBI observations. PKS~1921--293, together with 3C\\,273B and 3C\\,279, is currently among the brightest extragalactic sources in the sky at millimeter wavelengths, having a flux density at 3\\,mm greater than 5\\,Jy since 1990 with a peak of 15.3\\,Jy in April 1994 (Tornikoski et al. 1996). Its compactness, implied by its flat radio spectrum, has been confirmed by space VLBI (VSOP) observations (Shen et al. 1999). On projected space-ground baselines of 25,000\\,km (about three times longer than the longest ground baselines), PKS~1921--293 had a correlated flux density of 1.0\\,Jy (1.6\\,GHz), and a derived core brightness temperature limit of 3.0$\\times10^{12}$\\,K (in the rest frame of the quasar). \\\\  \\begin{figure}[ht] \\begin{center} \\vspace*{-1cm} \\leavevmode\\psfig{file=Shen_Zhiqiang_fig1.ps,angle=-90} \\end{center} \\caption{Two VLBA images of PKS\\,1921--293 simultaneously observed at 5\\,GHz (upper panel) and 43\\,GHz (lower panel) in September 2000. Note the difference in the scale since the 43\\,GHz image occupies only 1\\% area (within the {\\bf bold} square surrounding the origin) of the 5\\,GHz image. The part of the mm-jet flow from which the proper motion was obtained is indicated by a vector with the starting position (filled diamond in 43\\,GHz image) detected at epoch 1994.32 (see section 4). This mm-jet component is not seen in the 2000 image. For a discussion of components C1 and C2 within the central region of radius 1\\,pc ({\\bf bold} circle in 43\\,GHz image), see section 5.} \\label{fig:1} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209528_arXiv.txt": {
        "abstract": "s{Weakly Interacting Massive Particle (WIMP) direct detection experiments are closing in on the region of parameter space where neutralinos may constitute the Galactic halo dark matter. Numerical simulations and observations of galaxy halos indicate that the standard Maxwellian halo model is likely to be a poor approximation to the dark matter distribution. We examine how halo models with triaxiality and/or velocity anisotropy affect exclusion limits, before discussing the consequences of the possible survival of small scale clumps.} ",
        "introduction": " ",
        "conclusions": "We have seen that even if the local WIMP distribution is smooth its velocity distribution may deviate significantly from the standard Maxwellian, and this has a non-negligible effect on exclusion limits from WIMP direct detection experiments, affecting the limits from different experiments differently. The effect on experiments which can detect the angular and/or time variation of the event rate will be more significant. Constraints (and in the future possibly best fits) calculated assuming a standard Maxwellian halo could be erroneous, even worse if only the signals expected from the standard halo model (e.g. a sinusoidal annual modulation with peak in June) are searched for, a real WIMP signal could be overlooked. On the other hand, more optimistically, if WIMPs were detected it might then be possible to derive useful information about the local velocity distribution, and hence the formation of the Galactic halo."
    },
    "0209/astro-ph0209002_arXiv.txt": {
        "abstract": "We investigate the possibility of mapping large-scale anti-cyclonic vortices, resulting from a global baroclinic instability, as pre-cursors of planet formation in proto-planetary disks with the planned Atacama Large Millimeter Array (ALMA). On the basis of three-dimensional radiative transfer simulations, images of a hydrodynamically calculated disk are derived which provide the basis for the simulation of ALMA. We find that ALMA will be able to trace the theoretically predicted large-scale  anti-cyclonic vortex and will therefore allow testing of existing models of this very early stage of planet formation in circumstellar disks. ",
        "introduction": "Klahr \\& Bodenheimer~(2002) show that a global baroclinic instability represents a source for turbulence leading to angular momentum transport in Keplerian accretion disks with a radial gradient in entropy. Their hydrodynamical simulations show that this baroclinic flow is unstable and produces pressure waves, Rossby waves, and vortices in the $r-\\phi$ plane of the disk. Most interestingly, these hyper-dens anti-cyclonic vortices form out of little background noise and become long-lasting features, which have been suggested to lead to the formation of planets (e.g.\\ Adams \\& Watkins 1995). These pre-planetary matter concentrations have a surface density up to four times higher then the ambient medium. They furthermore concentrate dust in their centres, stressing their importance for the planetary formation process (e.g.\\ Barge \\& Sommeria 1995). ",
        "conclusions": "\\label{concl}  We found that ALMA will be able to map the anti-cyclonic hyper-dens vortices from the global baroclinic instability representing an early stage of planet formation. In nearby star-forming regions, the main parameters of the predicted global baroclinic instability in proto-planetary disks can be derived within several minutes of integration time. For even closer objects (within a distance of about 50\\,pc) a detailed structure of the spiral density pattern will be possible. We want to stress that the largest planned baselines and high frequencies are required for these observations."
    },
    "0209/hep-ph0209192_arXiv.txt": {
        "abstract": "If the observed deficit of solar neutrinos is due to neutrino oscillations, neutrino conversions caused by the interaction of their transition magnetic moments with the solar magnetic field (spin-flavour precession) can still be present at a subdominant level. In that case, the combined action of neutrino oscillations and spin-flavour precession can lead to a small but observable flux of electron antineutrinos coming from the sun. Non-observation of these $\\bar{\\nu}_e$'s could set limits on neutrino transition moment $\\mu$ and the strength and coordinate dependence of the solar magnetic field $B_\\perp$. The sensitivity of the $\\bar{\\nu}_e$ flux to the product $\\mu B_\\perp$ is strongest in the case of the vacuum oscillation (VO) solution of the solar neutrino problem; in the case of the LOW solution, it is weaker,  and it is the weakest for the LMA solution.  For different solutions, different characteristics of the solar magnetic field $B_\\perp(r)$ are probed: for the VO solution, the $\\bar{\\nu}_e$ flux is determined by the integral of $B_\\perp(r)$ over the solar convective zone, for LMA it is determined by the magnitude of $B_\\perp$ in the neutrino production region, and for LOW it depends on the competition between this magnitude and the derivative of $B_\\perp(r)$ at the surface of the sun. ",
        "introduction": "The observed deficit of solar neutrinos \\cite{SNE} compared to the expectations based on the standard solar model \\cite{SSM} and the standard electroweak model \\cite{SM} is now firmly established to be due to non-standard neutrino properties. In particular, the SNO Collaboration has demonstrated \\cite{SNO} that a significant fraction of solar $\\nu_e$ is converted into some other active neutrino species, which can be $\\nu_\\mu$, $\\nu_\\tau$, $\\bar{\\nu}_\\mu$ or $\\bar{\\nu}_\\tau$. The most plausible and widely accepted explanation of the observed solar neutrino deficit are neutrino oscillations; however, some alternative possibilities are not ruled out yet. One of them is neutrino spin-flavour precession \\cite{SV} due to the interaction of neutrino transition (flavour off-diagonal) magnetic moments with the solar magnetic field. Unlike the ordinary neutrino spin precession \\cite{FS}, the spin-flavour precession (SFP) can take place even if neutrinos are Majorana particles; in this case it converts left-handed $\\nu_e$ into right-handed $\\bar{\\nu}_\\mu$ or $\\bar{\\nu}_\\tau$, which would be in accord with the SNO findings. Neutrino SFP can be resonantly enhanced in matter \\cite{LM,Akh1}, very much similarly to the resonance amplification of neutrino oscillations, the MSW effect \\cite{MSW}.  SFP of solar neutrinos, both resonance and non-resonance, can very well account for the observed solar neutrino deficit. It yields an excellent fit of all currently available solar neutrino data (see, e.g., [10 -- 18] for recent analyses), even somewhat better than that of the large mixing angle (LMA) oscillation solution, which is the best one among the oscillation solutions. However, to account for the solar neutrino data, SFP requires relatively large values of the neutrino transition magnetic moment, $\\mu \\sim 10^{-11} \\mu_B$ for peak values of the solar magnetic field $B_0\\sim 100$ kG. Although such values of $\\mu$ are not experimentally excluded, they are hard to achieve in the simplest extensions of the standard electroweak model.  In the present paper we shall be assuming that the neutrino transition magnetic moment and/or solar magnetic field strength are significantly below the values necessary for the SFP mechanism to account for the solar neutrino deficiency (though not completely negligible). Our assumption is that it is neutrino oscillations that solve the solar neutrino problem, while the SFP is present at a subdominant level. What can then the observable effects of SFP be? Its influence on the survival probability of solar $\\nu_e$ will be small and essentially indistinguishable from a small change of the neutrino oscillation parameters. However, the combined action of neutrino oscillations and SFP can lead to a qualitatively new effect which is absent when only oscillations or only SFP are operative  -- the production of a flux of electron antineutrinos \\cite{LM,Akh2,next,Raju,APS,Baha}. Since all the currently favoured oscillation solutions of the solar neutrino problem -- LMA and LOW MSW solutions and vacuum oscillations (VO) -- require the solar neutrino oscillations to be driven by a large mixing angle \\cite{osc},  an observable flux of solar $\\bar{\\nu}_e$ can in principle be produced.  In  the present paper we address the question of what can be learned about the neutrino transition magnetic moments $\\mu$ and the solar magnetic field by studying the solar $\\bar{\\nu}_e$'s. In particular, we discuss the bounds on $\\mu$ and the strength and coordinate dependence of the solar magnetic field that can be derived from the current upper limits on the solar $\\bar{\\nu}_e$ flux as well as from future experiments in case the flux of $\\bar{\\nu}_e$ from the sun is not observed.  Experimentally, $\\bar{\\nu}_{e}$'s have a very clear signature and can be easily distinguished from the other neutrino species. The main problem with detecting $\\bar{\\nu}_{e}$'s from the sun is the background of electron antineutrinos from nuclear reactors. This background is a steeply decreasing function of neutrino energy; it becomes negligible for $E>$ (5 -- 8) MeV. Therefore only the solar $^8$B neutrinos can contribute to an observable flux of solar $\\bar{\\nu}_{e}$'s and the energy interval to be studied is $E\\simeq$ (5 -- 15) MeV.  Neutrino magnetic moments can also manifest themselves through the additional contribution to the $\\nu e$ scattering cross section in the solar neutrino detectors (see, e.g., \\cite{nuscat} for recent discussions). However, these contributions can only be noticeable if $\\mu\\aprge 10^{-10}\\mu_B$. While such large values of $\\mu$ are consistent with the current laboratory upper bounds \\cite{lab}, they exceed the astrophysical bounds $\\mu <$ (1 -- 3) $\\times 10^{-12}\\mu_B$ \\cite{astro} by more than an order of magnitude. In our study we shall be assuming the astrophysical upper bounds to be satisfied and so shall neglect the effects of neutrino magnetic moment on neutrino detection. ",
        "conclusions": "We calculated the probability of production of solar $\\bar{\\nu}_{eR}$'s assuming that the solar neutrino deficit is due to neutrino oscillations while the spin-flavour precession caused by the interaction of neutrino transition magnetic moments with the solar magnetic field is present as a subdominant process. We considered the SFP in perturbation theory and obtained analytic expressions for the transition probability $P(\\nu_{eL} \\to\\bar{\\nu}_{eR})$ valid for the LMA, LOW and VO solutions of the solar neutrino problem. We compared these analytical expressions with the results of numerical integration of the system of differential equations (1)-(4) and found very good agreement in all the cases.  For each of the solutions of the solar neutrino problem we then obtained simplified approximate expressions for $P(\\nu_{eL}\\to\\bar{\\nu}_{eR})$, which allowed us to relate this probability with simple characteristics of the solar magnetic field $B_\\perp(r)$. For different solutions, different characteristics of the solar magnetic field $B_\\perp(r)$ are probed: for the VO solution, the $\\bar{\\nu}_e$ flux is determined by the integral of $B_\\perp(r)$ over the upper 2/3 of the solar convective zone, for LMA it is determined by the magnitude of $B_\\perp$ in the neutrino production region, and for the LOW solution it depends on the competition between this magnitude and the derivative of $B_\\perp(r)$ at the surface of the sun.  The accuracy of the simplified expressions for $P(\\nu_{eL}\\to\\bar{\\nu}_{eR})$ is also different for different solutions: the error is less than 3\\% for the LMA solution and about 20\\% for the VO solution, while for the LOW solution the simplified expression is only correct within a factor of three or four \\footnote{Assuming that condition (\\ref{newcond}) is satisfied. Otherwise, the description of the LOW case is similar to that of LMA, and the accuracy of the approximate expression is as good as it is in the LMA case.}. Since the efficiency of SFP depends on the product of neutrino magnetic moment and magnetic field strength, only this product and not $\\mu$ and $B_\\perp$ separately can be probed by studying the solar $\\bar{\\nu}_e$ flux $\\Phi_{\\bar{\\nu}_e}$. Comparing eqs. (\\ref{prob2}), (\\ref{prob3}) and (\\ref{prob5}) we find that the sensitivity of the $\\bar{\\nu}_e$ flux to the product $\\mu B_\\perp$ is strongest in the case of the VO solution of the  solar neutrino problem; it is weaker in the case of the LOW solution and weakest for the LMA solution. This, however, does not necessarily mean that the LMA solution has the lowest sensitivity to the neutrino magnetic moment: the $\\bar{\\nu}_{eR}$ flux in that case depends on the magnetic field in the core of the sun which may well be much stronger than the field in the convective zone, relevant for the VO case.  We shall now discuss the present experimental upper bounds on $\\Phi_{\\bar{\\nu}_e}$ as well as the sensitivity of the future experiments, and their implications. Currently, the most stringent upper bounds on $\\Phi_{\\bar{\\nu}_e}$ come from the LSD experiment, $\\Phi_{\\bar{\\nu}_e}< (1.7\\times 10^{-2})\\,\\Phi_{^8\\!B}$ at 90\\% CL \\cite{LSD}, and from the Super-Kamiokande experiment, $\\Phi_{\\bar{\\nu}_e}< (1.2 - 1.6) \\times 10^{-2}\\,\\Phi_{^8\\!B}$ at 90\\% CL \\cite{Smy}. The Super-Kamiokande bounds were presented for several energy bins with $E\\ge 8$ MeV. Future experiments are expected to improve these bounds (or discover the flux of solar $\\bar{\\nu}_{eR}$): KamLAND will be able to put a limit of $10^{-3}\\, \\Phi_{^8\\!B}$ at 95\\% CL on the solar $\\bar{\\nu}_{eR}$ flux after one year of operation \\cite{KamLAND}, and Borexino should be able to reach a similar sensitivity after a few years of data taking \\cite{Lothar}. The current limit $\\Phi_{\\bar{\\nu}_e} \\aprle 1.5\\%\\,\\Phi_{^8\\!B}$ implies, in the case of the LMA solution, a bound \\be \\left[\\frac{\\mu}{10^{-12}\\mu_B}\\frac{B_\\perp(r_i)}{10\\,\\mbox{kG}}\\right] \\aprle 10^4\\,, \\label{bound1} \\ee where $B_\\perp(r_i)$ is the average solar magnetic field in the neutrino production region, $r\\aprle 0.1R_\\odot$. An experiment with the tritium radioactive antineutrino source has been recently proposed with the goal of putting an upper limit of $3\\times 10^{-12}\\mu_B$ on the neutrino magnetic moment $\\mu$ or measuring it if it is above this value \\cite{source}; if $\\mu\\simeq 3\\times 10^{-12}\\mu_B$ is found, the bound (\\ref{bound1}) would imply $B_\\perp(r_i) \\aprle 3\\times 10^7$ G. Note that this limit is more stringent than the astrophysical one obtained from the requirement that the pressure of the solar magnetic field should not  exceed the matter pressure ($B_\\perp \\aprle 10^9$) \\cite{astrolim}.  Conversely, if a reliable quantitative model of the solar magnetic field is developed, eq.~(\\ref{bound1}) will limit the neutrino magnetic moment. For $B_\\perp(r_i)$ close to the above-mentioned astrophysical bound, the limit on $\\mu$ would be $\\mu \\aprle 10^{-13}\\mu_B$, which is more than an order of magnitude more stringent than the expected limit from the planned laboratory experiment \\cite{source}. Unfortunately, no compelling model of the solar magnetic field exists at present.  Similar considerations apply to the LOW and VO cases. For LOW, assuming that condition (\\ref{newcond}) is satisfied, the current limits on the solar $\\bar{\\nu}_{eR}$ flux lead to \\be \\kappa \\left[\\frac{\\mu}{10^{-12}\\mu_B}\\frac{B_0}{10\\,\\mbox{kG}} \\right] \\aprle 10\\,, \\label{bound2} \\ee where $\\kappa$ and $B_0$ parameterize the derivative of the solar magnetic field at $r=R_\\odot$, see eq.~(\\ref{param}). For $\\mu\\simeq 3\\times 10^{-12} \\mu_B$, eq. (\\ref{bound2}) limits this derivative to be $|B_\\perp'(r)|_{R_\\odot}\\aprle 2.2\\times 10^2$ kG/$R_\\odot$. Conversely, if the actual value of $|B_\\perp'(r)|_{R_\\odot}$ is close to this value, the limit on $\\mu$ from (\\ref{bound2}) would be competitive with the expected upper bound from the planned laboratory experiment \\cite{source}. If condition (\\ref{newcond}) is not satisfied, the preceding discussion of the LMA case applies to the LOW case as well.  In the VO case, the current upper bounds on $\\Phi_{\\bar{\\nu}_e}$ imply \\be \\left[\\frac{\\mu}{10^{-12}\\mu_B}\\frac{\\overline{B}}{10\\,\\mbox{kG}} \\right] \\aprle 5\\,, \\label{bound3} \\ee where $\\overline{B}$ is the average magnetic field in the interval $0.817 R_\\odot\\le r \\le R_\\odot$ defined in (\\ref{aver}). For $\\mu=3\\times 10^{-12}$ this gives $\\overline{B}<17$ kG. If, alternatively, some model considerations establish that the average field $\\overline{B}$ is, for example, 100 kG, eq. (\\ref{bound3}) would lead to the limit $\\mu < 5\\times 10^{-13}$.  With the expected upper bound on the flux of solar $\\bar{\\nu}_{eR}$ from KamLAND, all the limits that we discussed above (eqs. (\\ref{bound1}) -- (\\ref{bound3})) will be strengthened by about a factor of four.   In the VO case, the flux of solar $\\bar{\\nu}_{eR}$'s will have seasonal variations, similar to those of the $\\nu_{eL}$ flux. In the LOW and VO cases, for which the $\\bar{\\nu}_{eR}$ production is mainly driven by the magnetic field in the convective zone, the solar $\\bar{\\nu}_{eR}$ flux may also vary with time due to the 11-year variations of this magnetic field. For all the solutions, the solar $\\bar{\\nu}_{eR}$ flux should, of course, also exhibit $\\sim 7\\%$ variations due to the variations of the distance between the sun and the earth. Low statistics may, however, make these variations difficult to detect.     \\vspace{0.1cm} \\noindent  {\\em Acknowledgements.} We are grateful to A. Mour\\~ao for useful discussions. E.A. was supported by the Calouste Gulbenkian Foundation as a Gulbenkian Visiting Professor at Instituto Superior T\\'ecnico."
    },
    "0209/astro-ph0209552_arXiv.txt": {
        "abstract": " ",
        "introduction": "Subdwarf B (sdB) stars are thought to be helium burning stars with low mass hydrogen envelopes. Several evolutionary paths have been proposed to explain the formation of these systems. One of these scenarios is the evolution of the sdB progenitor within a binary system. In fact \\cite{m01} found that out of a sample of 36 sdBs, 21 of them reside in close binary systems. This result combined with conclusions reached by other authors \\cite{gls00} suggest that two-thirds of sdBs are in binary systems.  With this in mind we have looked systematically at bright sdB stars from the PG survey. By taking spectra at several different epochs we have measured the radial velocity shifts caused by the motion of the sdB star within the binary. Our data have been taken over a long time base line (2 years) which allowed us to find longer period binaries than known before. Here we present results for 29 sdB systems. The methods we used to measure the radial velocities, to fit the radial velocity data and to select the best alias are described in detail in \\cite{l02}. ",
        "conclusions": "We find that most of the systems with orbits known up to now show a range of parameters (periods and masses of the companions to the sdB stars) that can be explained with an evolutionary model that consists of the formation of a common envelope after the onset of mass transfer when the sdB progenitor was at the tip of the red giant branch. Large orbital period systems can be explained by assuming either high common envelope ejection efficiencies or large metallicities for the sdB progenitor. But we find that these models cannot explain simultaneously large period systems and short period systems with large mass companions like KPD1930+2752 \\cite{mmn00}."
    },
    "0209/astro-ph0209078_arXiv.txt": {
        "abstract": "We propose that coronagraphic imaging in combination with moderate to high spectral resolution from the outset may prove more effective in both {\\it detecting} extrasolar planets and characterizing them than a standard coronagraphic imaging approach. We envisage an integral-field spectrograph coupled to a coronagraph to produce a datacube of two space dimensions and one wavelength. For the idealised case where the spectrum of the star is well-known and unchanging across the field, we discuss the utility of cross-correlation to seek the extrasolar planet signal, and describe a mathematical approach to completely eliminate stray light from the host star (although not its Poisson noise). For the case where the PSF is dominated by diffraction and scattering effects, and comprises a multitude of speckles within an Airy pattern typical of a space-based observation, we turn the wavelength dependence of the PSF to advantage and present a general way to eliminate the contribution from the star while preserving both the flux and spectrum of the extrasolar planet. We call this method ``spectral deconvolution''. We illustrate the dramatic gains by showing an idealized simulation that results in a $20$-$\\sigma$ detection of a Jovian planet at 2~pc with a 2-m coronagraphic space telescope, even though the planet's peak flux is only 1\\%\\ that of the PSF wings of the host star. This scales to detection of a terrestrial extrasolar planet at 2~pc with an 8-m coronagraphic Terrestrial Planet Finder (TPF) in $\\sim 7$~hr (or less with appropriate spatial filtering). Data on the spectral characteristics of the extrasolar planet and hence on its atmospheric constituents and possible biomarkers are obtained naturally as part of this process. ",
        "introduction": "The detection of a large number of extrasolar planets and planetary systems through precision radial velocity surveys has revolutionized thinking on the frequency and characteristics of planets (Marcy and Butler 1998). It is now clear that extrasolar planets are common. Along with the discovery that life can exist and thrive in an extreme range of environments on Earth, this has given a strong stimulus and boost to the search for life elsewhere in the Universe. Major design studies have been undertaken to consider competing concepts for the Terrestrial Planet Finder (TPF) whose primary goal is to locate terrestrial extrasolar planets in the habitable zone (where liquid water may be present) and to characterize their atmospheres. It is thought that a combination of disequilibrium chemical processes will betray the presence of biological activity (Woolf and Angel 1998, see also {\\it http://tpf.jpl.nasa.gov}).  Because of the faintness of planets, it has generally been assumed that characterization of Earth-like planets will be carried out at low spectral resolution. This maximizes the number of photons from the planet, per spectral resolution element, and allows the use of broad, deep molecular absorption bands to probe the atmosphere and to seek evidence of life (Kasting 1996, Woolf and Angel 1998, Schindler, Trent \\&\\ Kasting 2000, Des Marais et al 2001). Here, we consider use of higher spectral resolution observations from the outset, although retaining the imaging context. The methods described will still work, however, to varying degrees at low resolution. Allowing ourselves to consider high resolution enables us to explore more novel approaches and understand some of the basic dependencies.  High spectral resolution observations have of course been obtained for the ground-breaking precision radial velocity measurements, and in addition high spectral resolution searches have been performed in order to attempt detection of extrasolar planets through the reflection of stellar photospheric emission, coupled to sophisticated numerical methods (Collier Cameron et al 2002). Detection of planetary atmospheres with high signal-to-noise transit spectroscopy has been both proposed and observed (Brown 2001, Charbonneau et al. 2002).  Classically, direct imaging of faint companions to bright stars has been carried out with coronagraphic observation. There has been a great deal of work devoted to improving the performance of coronagraphs with a multitude of analogue methods and data analysis techniques (e.g. Roddier \\&\\ Roddier 1997, Rouan et al 2000, Spergel and Kasdin 2001; Kuchner \\&\\ Traub 2002 and many others). Other approaches include manipulation of speckle patterns to reduce PSF haloes or recognize planetary signatures within the speckle noise, including the dark-speckle coronagraph (Boccaletti et al 1998) and the use of dichroics and differencing of speckle images (Racine et al. 1999, Marois et al 2000).  Our goal here is to detect and image extrasolar planets directly, that is separated from their host star, through photons reflected or emitted from their surface or atmosphere. The concept is to use a coronagraph together with integral-field spectroscopy (e.g. Bacon et al 1995). Recognition of the planetary spectrum within the multiple spectra into which the image is divided may be achieved using pattern recognition techniques or by manipulation of the stray light of the host star with the goal of eliminating it. The process can be aided by the orbital velocity of the planet which displaces its spectrum in wavelength from that of the star. The concepts outlined may be applicable to either space or ground-based observation.  Firstly we consider the case where the spectrum of the star is well-known and consistent across the field of view (i.e. spatially invariant; the same at all points of interest in its PSF). In \\S~3.1 we explore the statistical properties of straightforward cross-correlation of the spectra using a (matched) template. We show the trade-space between signal-to-noise ($S/N$) and rejection of stray light from the host star. In \\S~3.2 we present a mathematical approach which ensures {\\it complete} elimination of stray light from the host star. By using Gram-Schmidt orthonormalization we show that the $S/N$ of the correlation functions may be preserved while offering formally complete rejection of the host star light. In the absence of a matched template, analysis of variance methods, \\S~3.3, may be used to provide information on whether there are multiple spectroscopic components (i.e. planets or other companions) present. We develop spectral templates and idealized simulations for these concepts, which are more applicable to ground-based observation, in \\S~4.  In \\S~5.0 we discuss the alternate case that the PSF of the coronagraphic observation is dominated by diffraction, or diffraction plus scattering, and introduce the concept of ``spectral deconvolution''. This situation is likely to apply to space-based observation, primarily, although there may be some ground-based opportunities using adaptive optics. We use the wavelength dependence of the PSF to advantage. By changing the image scale inversely to the wavelength the PSF is transformed into a simple, slowly varying function that can be removed e.g. by fitting low order polynomials as a function of wavelength. The extrasolar planet on the other hand is moved radially in the datacube by this process, and hence becomes a high frequency component which is ignored by the starlight rejection procedure. Reconstruction of the subtracted datacube back onto the original spatial scale and summation along the spectral dimension reveals the extrasolar planet signal. This method may prove exceptionally powerful and allows essentially {\\it an ideal Poisson-noise limited imaging detection, which simultaneously provides a spectrum of the extrasolar planet.} In \\S~6, we speculate on future directions a spectroscopic approach might take, and \\S~7 summarizes our conclusions. ",
        "conclusions": "We have considered  the use of integral-field spectroscopy together with coronagraphy as an interesting approach to planet {\\it detection} as well as characterization. The concept is to acquire a datacube of images, each at a different wavelength, of a star with planets shining by reflected light of the star modified by the integrated albedo (or emission) of the planet. The star is presumed to be occulted by the coronagraph, however there remains significant leakage or stray light from the star that dominates the problem of finding extrasolar planets. A potentially effective compromise modification to an integral-field spectrograph would be construction of a spectral datacube using narrow-band filters sequentially or a long-slit moved perpendicular to its length.  In the limiting case where the spectrum of the host star is the same across the field of view (or a region of interest within the field of view), we may use pattern recognition techniques to find extrasolar planets within the noisy spectra of the field. This is effectively done with cross-correlation using matched templates that in turn offer scientific insight into the nature of the spectrum of the planet. The $S/N$ of the cross-correlation approach is the $S/N$ of the ideal imaging case mutiplied by the standard deviation of the planet's spectrum (normalized to unit mean), which leads to pressure to use higher spectral resolution. Substantial gains can be made in rejecting the systematic contamination from the host star. In this limiting case, we may also use a general analysis of variance approach to estimate whether there is any source within the field other than just the host star, irrespective of its spectrum.  A method is described which mathematically completely eliminates the parasitic emission from the host star, although not the Poisson noise of that emission. This technique uses Gram-Schmidt orthonormalization, with the stellar spectrum and planet's template providing basis vectors from which orthonormal vectors can be constructed. The $S/N$ of the Gram-Schmidt method is essentially the same as that of the cross-correlation technique, but with the advantage that stray light is {\\it completely} eliminated.  In the case where the optics are diffraction limited, although not necessarily perfect, we describe a quite different technique to remove the stray light from the star. We call this ``spectral deconvolution'', applicable to space-based observation and perhaps some ground-based adaptive optics approaches. In the diffraction limit, the spatial scales of the parasitic emission are proportional to wavelength. By numerically adjusting the image scales in the datacube to that of a reference wavelength, the PSF changes with wavelength are much less dramatic. We can fit a low order spectral function at each pixel of the resampled (``shrunk'') datacube and subtract it to approach the Poisson noise limit. However since the planet now occupies many spatial pixels and only a limited spectral window in the resampled datacube, its contribution to the flux is overlooked by the fitting procedure. When the star-subtracted datacube is reconstructed back onto the original scale, the planet spectrum is realigned and the planet may be seen by collapsing the datacube in the spectral dimension to make an image. This method offers, with a single observation, essentially {\\it Poisson limited detection capability.} At the same time, and as part of the discovery observation, it provides a spectrum of the extrasolar planet from the outset and suitable for scientific investigation into the composition of the extrasolar planet's atmosphere and possible biomarkers.  The parameter space encompassed by integral-field spectroscopy with coronagraphic imaging is vast, covering spatial extent, spatial resolution, spectral coverage, spectral resolution for telescopes of various sizes both ground and space based. The technical and practical difficulties are non-trivial, and there is a great deal of work required to understand which regimes may be most effectively pursued with techniques similar to those described here, or with much more sophisticated pattern recognition techniques, yet the promise seems sufficiently high that we feel the effort is likely to be worthwhile. If these methods do prove practical, the design of future instruments, and perhaps even missions optimized for planet detection and characterization, may be influenced, since it not only improves detection capability, but at the same time and as part of the discovery observation itself, can offer vital information on the character of the planet's spectrum, including the presence of biomarkers and evidence for life."
    },
    "0209/astro-ph0209622_arXiv.txt": {
        "abstract": "We review the morphological and spectral energy distribution characteristics of the dust continuum emission (emitted in the 40-200$\\,{\\mu}$m spectral range) from normal galaxies, as revealed by detailed ISOPHOT mapping observations of nearby spirals and by % ISOPHOT observations of the integrated emissions from representative statistical samples in the local universe. ",
        "introduction": "The sensitivity of ISO and its spectral grasp extending to 200$\\,{\\mu}$m made it the first observatory capable of routinely measuring the infrared emission corresponding to the bulk of starlight absorbed by interstellar dust in ``normal''\\footnote{We use the term ``normal'' to denote star-forming systems not undergoing a starburst, and not dominated by AGN activity.} galaxies. Here we review ISO's view of the morphological and spectral energy distribution (SED) characteristics of the dust continuum emission (emitted in the 40-200$\\,{\\mu}$m spectral range) from normal galaxies, and its interpretation. In this review we only discuss the results from the ISOPHOT instrument (Lemke et al. 1996) on board ISO. For the spectral observations of these systems we refer to the review at this meeting by Helou.  \\begin{figure}[htb] \\includegraphics[scale=0.5]{TuffsRJ1_1.eps} \\caption{Predicted colour ratios for standard filter combinations 170/100$\\,{\\mu}$m (solid line); 100/60$\\,{\\mu}$m (dotted line); 60/25$\\,{\\mu}$m (dashed line) and 25/12$\\,{\\mu}$m (dot-dashed line) as a function of the strength of the local ISRF $\\chi$, where $\\chi=1$ near the sun. The calculations were made for spherical grains of astrophysical silicate, with the optical properties given by Laor \\& Draine (1993). A grain size distribution $n(a)\\,da\\,\\propto\\,a^{-3.5}$ was assumed, where $a$ is the grain radius ($0.001\\,\\le\\,a\\,\\le\\,0.25\\,{\\mu}$m). The colour of the radiation field illuminating the grains is fixed to that determined for the solar neighbourhood by Mezger, Mathis \\& Panagia (1982). } \\end{figure}  Although technically more demanding than observations in the Mid-Infrared (MIR) regime, only observations in the Far-Infrared (FIR) directly probe the role played by dust in the energy budget of star-forming galaxies. All star-forming galaxies are at least in part optically thick in the ultraviolet (UV)-optical regime, and the absorbed energy is predominantly re-radiated in the FIR. But the real investigative power of FIR astronomy lies in the fact that even for optically thin components of the interstellar medium, the large grains which dominate the FIR emission are in (or near to) equilibrium with the ambient interstellar radiation field (ISRF). Therefore, the grains act as test particles with FIR colours characteristic of the intensity and colour of the ISRF. This is illustrated in Fig.~1, which shows the predicted variation of infrared colours with radiation field intensities, for standard filter combinations of the ISOPHOT and ISOCAM instruments (on board ISO), and of the IRAS survey.  In particular, a filter set covering the range 60 to 170\\,$\\,{\\mu}$m probes intensities in the ISRF ranging from those expected for HII regions to those expected in the outskirts of disks of normal galaxies. By contrast, the MIR colour ratios are almost independent of the intensity of the ISRF, since they are determined by the relative abundance of small, impulsively heated grains. ",
        "conclusions": ""
    },
    "0209/astro-ph0209308_arXiv.txt": {
        "abstract": "The heavy elements formed by neutron capture processes have an interesting history from which we can extract useful clues to and constraints upon both the characteristics of the processes themselves and the star formation and nucleosynthesis history of Galactic matter. Of particular interest in this regard are the heavy element compositions of extremely metal-deficient stars. At metallicities [Fe/H] $\\leq$ --2.5, the elements in the mass region past barium (A $\\gtaprx$ 130-140) have been found (in non carbon-rich stars) to be pure $r$-process products. The identification of an environment provided by massive stars and associated Type II supernovae as an $r$-process site seems compelling. Increasing levels of heavy $s$-process (e.g., barium) enrichment with increasing metallicity, evident in the abundances of more metal-rich halo stars and disk stars,  reflect the delayed contributions from the low- and intermediate-mass (M~$\\sim$ 1-3 M$_\\odot$) stars that provide the site for the main $s$-process nucleosynthesis component during the AGB phase of their evolution. New abundance data in the mass region 60 $\\ltaprx$ A $\\ltaprx$ 130 is providing insight into the identity of possible alternative $r$-process sites. We review recent observational studies of heavy element abundances both in low metallicity halo stars and in disk stars, discuss the observed trends in light of nucleosynthesis theory, and explore some implications of these results for Galactic chemical evolution, nucleosynthesis, and nucleocosmochronology. ",
        "introduction": "Element abundance patterns in metal-poor halo field stars and globular cluster stars play a crucial role in guiding and constraining theoretical models of Galactic nucleosynthesis. These patterns can also provide significant clues to the natures of the nucleosynthesis mechanisms themselves. Nowhere is this more true than for the case of the neutron-capture ($n$-capture) processes that are understood to be responsible for the synthesis of the bulk of the heavy elements in the mass region A $\\gtaprx$ 60: the $s$-process and the $r$-process. Nucleosynthesis theory identifies quite different physical conditions and astrophysical sites for these two distinct processes. $r$-Process nuclei are effectively {\\it primary} nucleosynthesis products, formed under dynamic conditions in an environment associated with the evolution of massive stars (M $\\gtaprx$ 10 M$_\\odot$) to supernova explosions of Type II and the formation of neutron star remnants. $s$-Process nuclei are understood to be products of neutron captures on preexisting silicon-iron ``seed'' nuclei, occurring under hydrostatic burning conditions both in the helium burning cores of massive stars and particularly in the thermally pulsing helium shells of asymptotic giant branch (AGB) stars. In this picture, the first heavy (A $\\gtaprx$ 60) elements introduced into the interstellar gas component of our Galaxy are expected to have been $r$-process nuclei formed in association with massive stars, on time scales $\\tau_{star}$ $\\ltaprx$ 10$^8$ years. Most of the $s$-process nuclei, on the other hand, are first introduced into the ISM on time scales ($\\sim$ 10$^9$ years) characteristic of the lifetimes of their stellar progenitors (M $\\sim$ 1-3 M$_\\odot$).   That the general features of this simple model are correct is confirmed by the finding that $r$-process contributions dominate the heavy element abundances in extremely metal-poor halo and globular cluster stars, while significant $s$-process contributions are first identifiable at metallicities of order [Fe/H] \\footnote{ We adopt the usual spectroscopic notations that [A/B]~$\\equiv$ log$_{\\rm 10}$(N$_{\\rm A}$/N$_{\\rm B}$)$_{\\rm star}$~--~log$_{\\rm 10}$(N$_{\\rm A}$/N$_{\\rm B}$)$_{\\odot}$, and that log~$\\epsilon$(A)~$\\equiv$ log$_{\\rm 10}$(N$_{\\rm A}$/N$_{\\rm H}$)~+~12.0, for elements A and B. Also, metallicity will be assumed here to be equivalent to the stellar [Fe/H] value.} $\\approx$ --2. Observational constraints on $r$-process nucleosynthesis sites are  examined in \\S 3. The implications of the observed scatter in the level of $r$-process nuclei relative to iron and trends in the heavy element abundances as a function of [Fe/H] over the early star formation history of the Galaxy are considered in \\S 4. We also consider the use of the abundance data - particularly that involving the nuclear chronometers $^{232}$Th and $^{238}$U - for cosmochronology (\\S 5). Finally, a brief examination of neutron-capture nucleosynthesis at low metallicities is presented in \\S 6. We note there the importance of the carbon-rich stars. Although  this review does not include a detailed examination of this class of metal-poor stars, the abundance patterns in these stars may provide insight into the earliest phases of Galactic nucleosynthesis. First, in order to provide a basis for our subsequent discussions of the heavy element abundance patterns in the stellar populations of our Galaxy, we briefly review in the next section the current status of theoretical models for $s$-process and $r$-process nucleosynthesis. Throughout this paper, our emphasis will be on (first) providing the best observational data currently available concerning heavy element abundances in stars in our Galaxy and (second) the use of this data to educate us concerning the physical characteristics, sites, and time scales of $s$-process and $r$-process nucleosynthesis. ",
        "conclusions": "  \\begin{itemize}  \\item[]{1.} Observations leave absolutely no doubt but that the first contributions to the abundances of the heavy $r$-process elements in our Galaxy occurred at an early epoch, and well in advance of the first substantial $s$-process contributions. At metallicities [Fe/H] $\\ltaprx$ --2.5, the elements in the range  Z $\\ge$ 56 are virtually pure $r$-process products. $r$-Process synthesis of A $\\gtaprx$ 130-140 isotopes happens early in Galactic history, prior to contribution of heavy $s$-process isotopes  from AGB stars.  \\item[]{2.} The $r$-process mechanism for the synthesis of the A $\\gtaprx$ 130-140 isotopes (the ``main'' or ``strong'' component) is extremely robust. This is reflected in the fact that the abundance patterns in the most metal-deficient (oldest) stars, which may have received contributions from only one or at best a few $r$-process events, are nevertheless indistinguishable from the $r$-process abundance pattern that characterizes solar system matter. (The abundance patterns for the three stars CS 22892-052, HD 115444, and BD +173248 shown in Figure 2 reveal this remarkable agreement for stars of low metallicity but high [$r$-process/Fe].)  \\item[]{3.} The Ba/Eu and particularly La/Eu ratios reveal that the first significant (or major) introduction of heavy $s$-process isotopes (the Ba peak nuclei and beyond: the main $s$-process component) occurred at a metallicity [Fe/H] $\\sim$ -2. The time at which this occurred, presumably set by the lifetime of the low mass ($\\sim$ 1-2 M$_\\odot$) AGB star $s$-process nucleosynthesis site, is of order 10$^9$ years.  \\item[]{4.} The increased scatter in [$r$-process/Fe] at low metallicities [Fe/H] presumably reflects a significant and increasing level of inhomogeneity present in the gas at that epoch. It also provides evidence for the fact that only a fraction of the massive star (M $\\gtaprx$ 10 M$_\\odot$) and associated Type II supernova environments can have contributed to the synthesis of the heavy $r$-process isotopes. The data shown in Figure 4 reveal levels of $r$-process enrichment relative to iron of factors of approximately 50.  \\item[]{5.} The observations, particularly in $r$-process rich-stars, indicate that the heavier (Ba and above, Z $\\ge$ 56, or A $\\gtaprx$ 130-140) neutron-capture elements are consistent with a scaled solar system $r$-process curve. The data, although still incomplete, seem to indicate that the lighter neutron-capture elements (for example Ag) are not consistent with (i.e., fall below) that same scaled $r$-process curve. This behavior is shown in Figure 2 for CS~22892-052,  and indicates that two  distinct $r$-process environments may be required to synthesize both ends of the abundance distribution. These observations support earlier suggestions of two $r$-processes based upon solar system meteoritic (isotopic) data (Wasserburg et al. 1996). Analogously to the $s$-process, we can attribute the heavier neutron capture elements to a ``strong'' robust $r$-process, with a ``weak'' $r$-process responsible for the synthesis of the lighter elements below barium. Analyzing the lighter abundance data is complicated by the possibility of the production of $s$-process nuclei (in the weak $s$-process) from massive stars that might contribute to the production of Sr, Y and Zr only. Additional  spectroscopic studies of the abundances of the lighter neutron-capture elements, in metal deficient stars, will be  required to sort all of this out.  \\item[]{6.} The identification of the $r$-process site with massive star environments implies that the critical nuclear chronometers for dating the Galaxy were formed early in Galactic history. This strongly supports the use of the $r$-process isotopes $^{232}$Th, $^{235}$U, and $^{238}$U as reliable chronometers of the Galactic nucleosynthesis era. A more detailed discussion of this and related issues will be provided in a forthcoming paper (Burles {\\it et al.} 2002).  \\item[]{7.} For the $r$-process poor stars at low metallicity the data seem to indicate that lighter elements such as  Sr have high abundances with respect to heavier neutron-capture elements such as Ba. Also, for these stars (at least based upon the data available for the star HD 122563), the abundances of the heavier neutron-capture elements seems to depend more on atomic number (and to decline faster with atomic number) than the standard $r$-process production normally does. These results suggest that the early nucleosynthesis history of the Galaxy was quite complex with yields coming from various $s$- and $r$-process components and from synthesis sites with a variety of progenitor mass ranges.   \\end{itemize}"
    },
    "0209/astro-ph0209414_arXiv.txt": {
        "abstract": "We present {\\it Rossi X-ray Timing Explorer} observations of the X-ray pulsar SMC X-1. The source is highly variable on short time scales ($<$ 1 h), exhibiting apparent X-ray flares occupying a significant fraction ($\\sim$3 \\%) of the total observing time, with a recurrence time of $\\sim$100 s. The flares seem to occur over all binary orbital phases, and correlate with the overall variability in the light curve. We find a total of 323 discrete flares which have a mean full width half maximum of $\\sim$18 s. The detailed properties of SMC X-1 do not vary significantly between the flares and the normal state, suggesting that the flare may be an extension of the normal state persistent emission with increased accretion rates. The flares resemble Type II X-ray bursts from GRO J1744--28. We discuss the origin of the SMC X-1 flares in terms of a viscous instability near the inner edge of the accretion disk around a weakly magnetized X-ray pulsar, and find this is consistent with the interpretation that SMC X-1 is in fact an intermediate-stage source like GRO J1744--28. ",
        "introduction": "Neutron star X-ray binaries are generally categorized into two groups: low mass X-ray binaries (LMXBs) and X-ray pulsars. The surface magnetic field of the central neutron star in an LMXB is thought to be $\\sim$10$^8$ G. The mass accretion with this magnetic field is most likely spherical, so that no significant inhomogeneity in the X-ray emission over the neutron star surface is expected -- i.e., no persistent coherent pulsations are observed. The strong magnetic field, $\\sim$10$^{12}$ G, of an X-ray pulsar, on the other hand, can funnel the accretion matter onto the magnetic pole, which makes the central neutron star appear as a pulsar.  Of particular interest are the so-called ``intermediate-stage sources'' speculated to lie between the LMXBs and X-ray pulsars, including  ``the Rapid Burster\" (MXB 1730--355), GRO J1744--28 (``the Bursting Pulsar\"), and SAX J1808.4--3658  (``the accreting millisecond pulsar\") (Lewin et al. 1976; Fishman et al. 1995; in 't Zand et al. 1998). Both the Rapid Burster and GRO J1744--28 exhibit Type II X-ray bursts; however, only the former shows Type I X-ray bursts while only the latter has apparent coherent pulsations (Lewin et al. 1996). SAX J1808.4--3658, on the other hand, shows both Type I bursts and coherent pulsations (Wijnands \\& van der Klis 1998; Chakarbarty \\& Morgan 1998), but not Type II bursts. The magnetic field strengths of these sources have been inferred to be $\\sim$10$^{8-11}$ G, between those of LMXBs and X-ray pulsars.  Another possible intermediate-stage source is the X-ray pulsar SMC X-1, which has similar properties to GRO J1744--28, including its fast spin period ($\\sim$0.72 s for SMC X-1; $\\sim$0.47 s for GRO J1744--28), steady spin-up, and inferred magnetic field ($\\sim$10$^{11}$ G ) (Bildsten \\& Brown 1997; Li \\& van den Heuvel 1997). In addition, once SMC X-1 was observed with an X-ray burst that resembles Type II bursts (Angelini, Stella, \\& White 1991). It may be possible, therefore, that SMC X-1 and GRO J1744--28 belong to a distinctive group of X-ray binaries, ``bursting pulsars\", which show both coherent pulsations and Type II X-ray bursts (Li \\& van den Heuvel 1997). To investigate this important possibility, we analyze all publicly available RXTE data for SMC X-1, searching for phenomena that may be related to X-ray bursts. We report that SMC X-1 in fact exhibits active flares resembling Type II bursts from GRO J1744--28. ",
        "conclusions": "Through the analysis of the all publicly available RXTE data toward the X-ray pulsar SMC X-1, we find that the source is highly variable on short time scales ($<$ 1 h), and that Gaussian flares occur over all orbital phases with a $\\sim$100 s recurrence time scale. The flares occupy $\\sim$3 \\% of the total observing time, and the flaring activity is proportional to the overall variability of the source. While the pulse peak ratio shows a small systematic change along with the flaring activity, the PDSs, pulse profiles, softness ratios, and X-ray spectra during the flares are very similar to those outside the flares, indicating that the flares are probably extensions of a normal state just with increased accretion rates. This supports the interpretation that the SMC X-1 flares have their origin in an accretion disk instability and the suggestion that it may belong to a distinctive group of ``bursting pulsars\" with the ``bursting pulsar\" GRO J1744--28, owing to its $\\sim$10$^{11}$ G surface magnetic field. A viscous instability near the inner-edge of the accretion disk might be responsible for the SMC X-1 flares, although detailed studies on this scenario need to be done in the future."
    },
    "0209/astro-ph0209400_arXiv.txt": {
        "abstract": " ",
        "introduction": "This year is a decade since the first detection of the anisotropy of the Cosmic Microwave Background at large angular scales ($\\ge 10^0$) \\cite{str92}, \\cite{smo92}. Today the CMB anisotropy (CMBA) has been detected also at intermediate ($\\sim (1^0 - 10^0)$) and small angular scales ($\\le 1^0$), so the CMBA angular spectrum is now reasonably known down to the region of the first and second Doppler peaks \\cite{boo00}, \\cite{boo01a}, \\cite{boo01b}. Its shape gives information e.g. on the spectrum of the primordial cosmological perturbations or can be used to test the inflation theory but rises new questions to which CMBA cannot answers. Responses can on the contrary be obtained looking at the CMB polarization (CMBP) produced by Thomson scattering of CMB photons on the matter anisotropies at the recombination epoch. In particular one can hope to use CMBP to disentangle the effects of fundamental cosmological parameters like density of matter, density of dark energy etc., effects anisotropy do not separate.This is among the goals of space and ground based experiments like \\cite{Map}, \\cite{Planck1}, \\cite{Planck2}, \\cite{ami}, \\cite{que}, \\cite{gerva}, \\cite{newboo} and is the main goal of SPOrt a polarization dedicated ASI/ESA space mission on the International Space Station \\cite{cor99}. The relevance of the CMB polarization was remarked for the first time by M. Rees \\cite{ree65}. Since him many models of the expected features of the CMBP have been published (see for instance \\cite{saz95}, \\cite{ng96}, \\cite{mel97}). They stimulated the search for CMBP, but in spite of many attempts so far no CMB polarization has been detected. This observation is in fact extremely difficult because the expected signal is at least an order of magnitude smaller than the amplitude of the CMBA. Moreover foregrounds and their inhomogeneities cover the polarized fraction of the CMB or mimic CMBP spots, making the signal to noise (CMBP $/$ polarized foreground) ratio unfavorable. In this paper we will concentrate on methods for improving this ratio and for disentangling CMBP and polarized foregrounds.  In the microwave range the galactic foregrounds include: \\begin{itemize} \\item synchrotron radiation (strongly polarized), \\item free-free emission (polarization negligible), \\item dust radiation. \\end{itemize}  Because here we are interested in polarization, in the following  we will neglect the free-free emission, whose expected level of polarization is negligible. Moreover the polarized signals produced by dust, if present, (e.g. \\cite{set98}, \\cite{fos01}), may be treated as an addition to the synchrotron effects. In fact, as it will appears in the following, the important quantities in our analysis are the statistical properties of the foreground spatial distribution and, by good fortune, the spatial distribution of the dust polarized emission is similar to that of the synchrotron emission, since behind both radiation types there is the same driving force, the galactic magnetic field which alignes dust grains and guides radiating electrons.  Separation of foregrounds and CMBA was successfully solved when the CMB anisotropy was discovered \\cite{smo92}. When we go from CMBA to CMBP the separation of foreground and background however is more demanding and the problem of discriminating foreground inhomogneities from true CMB spots severe. Approaches used in the past e.g. \\cite{dod97}, \\cite{teg99}, \\cite{sto01}, \\cite{teg96}, \\cite{kog00}) were essentially based on the differences between the frequency spectra of foregrounds and CMB, therefore require multifrequency observations.  In this paper we suggest a different method which  takes advantage of the fact that the measured values of the parameters we use to describes the polarization of the diffuse radiation when measured at a given frequency in different directions behave as stochastic variables. Because the mathematical and statistical properties of these variables for synchrotron and CMB are different, we suggest to use statistical methods for analyzing single frequency maps of the diffuse radiation and disentangling their main components, synchrotron and CMBP. This method was proposed and briefly discussed in \\cite{saz01}. Here we present a more complete analysis. ",
        "conclusions": "Observations of the CMB polarization are hampered by the presence of the galactic polarized foreground. Only above $\\sim 50 GHz$ the cosmic signal is definitely above the galactic synchrotron and direct observations of CMBP are, in principle, possible. Between $\\sim 30$ GHz and $\\sim 50$ GHz the level of the polarized component of the synchrotron foreground is at least comparable to the CMBP level, but the observational situation is still insufficient to evaluate precisely its contribution to maps of the diffuse polarized radiation measured by a telescope. Below $\\sim 30$ GHz the galactic signal is definitely dominant. So far a common approach for studies of CMBP was to make accurate maps of the diffuse polarized radiation from a given region of sky at many frequencies and disentangle the various contributions modelling their frequency dependence and spatial distribution.  We have shown that an alternative way is to take advantage of the different statistical properties of the spatial distribution of the main components of the polarized diffuse radiation: CMBP and synchrotron (plus dust) galactic foreground. By measuring the $E$ and $B$ modes of the polarized radiation we can build an estimator which improves the background/foreground ratio by a factor sufficient to allow firm recognition and extraction of the CMBP contribution from single frequency maps at least down to 25 GHz (17 GHz in the most favorable conditions) at angular scales $\\leq 0.7^o$ ($l\\geq 250$)."
    },
    "0209/astro-ph0209546_arXiv.txt": {
        "abstract": "The starburst galaxies M82 and NGC253 have been proposed as the primary sources of cosmic rays with energies above $10^{18.7}$ eV. For energies $\\agt 10^{20.3}$~eV the model predicts strong  anisotropies. We calculate the probabilities that the latter  can be due to chance occurrence. For the highest energy cosmic ray events in this energy region, we find that the observed directionality has less than  1\\% probability of occurring due to random fluctuations. Moreover, during the first 5 years of operation at Auger, the observation of even half the predicted anisotropy has a probability of less than $10^{-5}$ to occur by chance fluctuation. Thus, this model can be subject to test at very small cost to the Auger priors budget and, whatever the outcome of that test, valuable information on the Galactic magnetic field will be obtained. ",
        "introduction": "Soon after the microwave echo of the big bang was discovered, Greisen, Zatsepin, and Kuzmin (GZK) noted that the relic photons make the universe opaque to cosmic rays (CRs) of sufficiently high energy~\\cite{Greisen:1966jv}. This occurs, for instance, for protons with energies beyond the photopion production threshold ($\\Delta (1232)$ resonance). After pion production, the proton (or perhaps, instead, a neutron) emerges with at least 50\\% of the incoming energy.  A similar phenomenon (of energy degradation) occurs for nuclei due to processes of photodisintegration. Therefore, the characteristic attenuation length for extremely high energy ($10^{20}~{\\rm eV} \\alt E \\alt 10^{20.5}~{\\rm eV}$) hadrons is less than 100~Mpc, decreasing down to 10~Mpc with rising energy~\\cite{Stanev:2000fb}. The survival probability for extremely high energy (EHE) $\\gamma$-rays (propagating on magnetic fields $\\gg10^{-11}$~G) to a distance $d$, \\mbox{$P(>d) \\approx \\exp[-d/6.6~{\\rm Mpc}]$}, becomes less than $10^{-4}$ after traversing a distance of 50~Mpc~\\cite{Elbert:1994zv}. This implies that the GZK sphere~\\cite{gzk-sphere} represents a small fraction of the size of the universe. Consequently, if the CR sources are universal in origin, the energy spectrum should not extend (except at greatly reduced intensity) beyond $\\sim 10^{20}$~eV, a phenomenom known as the GZK cutoff. Even though the Haverah Park~\\cite{Ave:2001hq}, Yakutsk~\\cite{Efimov:rk}, Fly's Eye~\\cite{Bird:wp}, and HiRes~\\cite{:2002ta} data show statistically significant evidence for such a cutoff~\\cite{Bahcall:2002wi} (more than 5$\\sigma$ independent of the sample used as a basis for extrapolation), the AGASA ground array detected a handful of events with energies $\\agt 10^{20}$~eV~\\cite{Takeda:1998ps}, as opposed to about 2 expected from the GZK cutoff. Moreover, within statistical uncertainty (which is large above $10^{20}$~eV) the flux of CRs above $10^{18.7}$~eV reported by the AGASA Collaboration~\\cite{Takeda:1998ps} is consistent with a $E^{-2.7}$ spectrum up to the highest observed energies, suggesting that a single acceleration mechanism is responsible for all the events beyond that energy, unless of course a very unlikely matching of spectra can account for the smoothness of the CR energy distribution.  In order to analyze the effect of energy losses on the observed spectrum, it is convenient to introduce the accumulation factor $f_{\\rm acc}$, defined as the ratio of energy-weighted fluxes for ``low'' ($10^{18.7}$ eV -- $10^{19.5}$ eV) and EHECRs. With this in mind, if the Earth is located in a typical environment and all CR-sources have smooth emission spectra, the observed spectrum above $10^{18.7}$~eV should have an offset in normalization between low and EHE given by $f_{\\rm acc}$. For CR protons and nuclei with uniform distribution of sources active over cosmological times, the cutoff due to photopion and photodisintegration processes relates the accumulation factor to a ratio of attenuation lengths~\\cite{Farrar:2000nw} and leads to $f_{\\rm acc} \\sim 100$. The smoothness of the observed CR spectrum~\\cite{Takeda:1998ps}, {\\it viz.} $f_{\\rm acc} \\sim 1$, seems to indicate that the power of nearby sources must be comparable to that of all other sources (redshift $z>0.5$) added together.  The simplest explanation, i.e., nearby sources are significantly more concentrated, does not seem to be the case. Specifically, if one simply assumes that the distribution of CR sources follows the distribution of normal galaxies, the local overdensity is only a factor of two above the mean, and thus insufficient to explain the measured flux above $10^{20}$~eV~\\cite{Blanton:2000dr}. Furthermore, the arrival direction of the super-GZK events is consistent with an isotropic distribution of sources (even when some level of clustering was already detected~\\cite{Uchihori:1999gu}), in sharp contrast to the anisotropic distribution of light within 100~Mpc~\\cite{Waxman:1996hp}. A way to avoid the problems with finding plausible astrophysical explanations is to look for solutions involving physics beyond the standard model~\\cite{Bhattacharjee:1998qc}. While the invocation of such new physics is an intringuing idea, there are now constraints that call into question the plausibility of some of these ideas~\\cite{Protheroe:1996pd}.    Recently, it was suggested that the observed near-isotropy of arrival directions could be due to a diffuse propagation of EHECRs~\\cite{Lemoine:1999ys}. In this work, we examine specific candidate sources for this hypothesis. These are the starburst galaxies M82 and NGC253 which have been shown to reproduce the main features of the observed flux~\\cite{Anchordoqui:2001ss}. In particular, we study here the critical aspect of a residual anisotropy that emerge  beyond the GZK energy limit after deflection in Galactic and extragalactic magnetic fields. Specifically, we estimate the probability that an apparent correlation between the arrival directions of the highest energy events and the  two starbursts can originate as a purely random fluctuation. After that, we study the sensitivity of Auger Observatory to the model. ",
        "conclusions": "We have made a definite prediction for future observations at the Auger Observatory: if the origin of CRs above $10^{18.7}$~eV are nearby starburst galaxies, {\\em the incoming CR flux will show a strong dipole anisotropy in the harmonic decomposition at energies beyond $10^{20.3}$~eV.} Because of its well-defined prediction, the model can be tested at the 5$\\sigma$ level in five years of running at Auger. Therefore, we strongly recommend that the Auger Collaboration take into account the next-door galaxy NGC253 in their first anisotropy prescription for super-GZK CRs. The confirmation of the starburst hypothesis would provide, as spinoff, direct evidence for the global structure of the Galactic magnetic field.  \\hfill"
    },
    "0209/astro-ph0209293_arXiv.txt": {
        "abstract": "I will start by discussing the evolutionary status of the white dwarf {\\em progenitors}, the hot UV bright stars. Observations of UIT-selected UV bright stars in globular clusters suggest that a high percentage of them manage to evolve from the horizontal branch to the white dwarf region without passing through the thermally pulsing AGB phase, thereby avoiding the planetary nebula stage.  The white dwarf {\\em successors} are stars experiencing a very late helium core flash while already on the helium white dwarf cooling curve. While they have been around theoretically for quite some time strong candidates could be verified only quite recently.  And, last not least, the {\\em white dwarfs} themselves offer new opportunities to derive distances and ages of globular clusters, which I will discuss. For a discussion of white dwarfs in binaries see the reviews by Adrienne Cool and Frank Verbunt in this volume. ",
        "introduction": "UV bright stars have been classically defined as stars brighter than the horizontal branch and bluer than the red giant branch (Zinn et al.\\ 1972). Such stars are also brighter in $U$ than any other cluster star. UV bright stars are produced by evolution from the \\begin{itemize} \\vspace*{-1ex} \\item {\\em horizontal branch} (HB) towards the asymptotic giant branch (post-HB stars) \\vspace*{-1ex} \\item {\\em extreme HB} (EHB, T$_{\\rm eff} >$23,000~K) directly towards the white dwarf domain (post-EHB stars) \\vspace*{-1ex} \\item {\\em asymptotic giant branch} towards the white dwarf domain (post-AGB stars), possibly showing up as central stars of planetary nebulae. \\vspace*{-1ex} \\end{itemize}  Ground based searches like the one of Zinn et al. (1972) found primarily cool UV bright stars (T$_{\\rm eff} \\le$9000~K). Spectroscopic analyses of the few hot UV bright stars found this way turned up only post-AGB stars. This finding is puzzling as a minimum mass of 0.565~M$_\\odot$ is required to reach the thermally pulsing AGB (Sch\\\"onberner 1983), which makes this evolutionary stage difficult to reach for the low mass stars in today's globular clusters.  The post-AGB phase also has a shorter lifetime than the other channels producing UV bright stars, making the observation of such stars even further unlikely.  \\begin{figure} \\plotfiddle{moehler_1.ps}{8.3cm}{270}{45}{45}{-180}{255} \\caption{Results of spectroscopic analyses of UV bright stars compared to evolutionary tracks. Open circles mark stars detected by optical searches (Conlon et al. 1994; Dixon et al. 1994, 1995; Glaspey et al. 1985, Heber \\& Kudritzki 1986, Moehler et al. 1998b, Rauch et al. 2002), filled squares mark UV bright stars detected as such by UIT (Moehler et al. 1998a, Landsman priv. comm.). The post-AGB and post-early AGB tracks are from Sch\\\"onberner (1983), the zero-age HB/EHB (ZAHB/ZAEHB) and post-EHB tracks are from Dorman et al. (1993). The numbers give the stellar mass of the track in units of 0.001~M$_\\odot$.} \\end{figure}  However, optical searches are obviously biased against finding hot stars, whose flux maximum moves ever farther to the ultraviolet with increasing temperature. Therefore searches in the satellite ultraviolet are much more promising and indeed the Ultraviolet Imaging Telescope (Stecher et al. 1997) found quite a few new hot UV bright stars in the globular cluster it surveyed. Spectroscopic analyses of these stars by ground based (ESO, Moehler et al. 1998a) and HST (Landsman, priv. comm.) observations showed stars evolving away from the extreme HB and also post-early AGB stars, that left the AGB {\\em before} the thermal pulses started, but no new ``classical'' post-AGB stars (see Moehler 2001 for more details). These new results are in much better agreement with the expectations from stellar evolution theory with respect to evolutionary life times (and thus observability) and minimum masses. At the same time they also suggest an explanation for the {\\em lack of planetary nebulae} in globular clusters observed by Jacoby et al. (1997): None of the UV bright stars detected solely by UIT will produce a planetary nebula -- post-EHB stars never reach the AGB and the evolution of post-early AGB stars proceeds so slowly that by the time they are hot enough to excite the remnants of their AGB envelopes these remnants have evaporated. ",
        "conclusions": ""
    },
    "0209/gr-qc0209051_arXiv.txt": {
        "abstract": "We present a new \\emph{covariant} and \\emph{gauge-invariant} perturbation formalism for dealing with spacetimes having spherical symmetry (or some preferred spatial direction) in the background, and apply it to the case of gravitational wave propagation in a Schwarzschild black hole spacetime. The 1+3 covariant approach is extended to a `1+1+2 covariant sheet' formalism by introducing a radial unit vector in addition to the timelike congruence, and decomposing all covariant quantities with respect to this. The background Schwarzschild solution is discussed and a covariant characterisation is given. We give the full first-order system of linearised 1+1+2 covariant equations, and we show how, by introducing (time and spherical) harmonic functions, these may be reduced to a system of \\emph{first-order} ordinary differential equations and algebraic constraints for the 1+1+2 variables which may be solved straightforwardly. We show how both the odd and even parity perturbations may be unified by the discovery of a covariant, frame- and gauge-invariant, transverse-traceless tensor describing gravitational waves, which satisfies a covariant wave equation equivalent to the Regge-Wheeler equation for both even and odd parity perturbations. We show how the Zerilli equation may be derived from this tensor, and derive a similar transverse traceless tensor equivalent to this equation. The so-called `special' quasinormal modes with purely imaginary frequency emerge naturally. The significance of the degrees of freedom in the choice of the two frame vectors is discussed, and we demonstrate that, for a certain frame choice, the underlying dynamics is governed purely by the Regge-Wheeler tensor. The two transverse-traceless Weyl tensors which carry the curvature of gravitational waves are discussed, and we give the closed system of four first-order ordinary differential equations describing their propagation. Finally, we consider the extension of this work to the study of gravitational waves in other astrophysical situations.   \\pacs{} ",
        "introduction": "The 1+3 covariant approach has proven to be an extremely useful technique for developing a detailed understanding of many aspects of relativistic cosmology, both in terms of fully nonlinear GR effects and through the application of the gauge-invariant, covariant perturbation formalism (see~\\cite{HvE}, for example) to the formation and evolution of density perturbations~\\cite{BE} in the universe and to the physics of the cosmic microwave background~\\cite{CL,MGE}, amongst other things (see~\\cite{HvE} for a comprehensive review). Its strength in cosmological applications lies in the fact that it is well adapted to the system it is describing: all essential information can be captured in a set of (1+3) covariant variables (defined with respect to a preferred timelike observer congruence), that have an immediate physical and geometrical significance. They satisfy a set of \\emph{evolution} and \\emph{constraint} equations, derived from Einstein's field equations, and the Bianchi and Ricci identities, which form a closed system of equations when an equation of state for the matter is chosen. The covariant and gauge-invariant linearisation procedure is easy and transparent: it consists of deciding which variables are `first order' (or `of order $\\epsilon$') and those which are `zeroth order'~-- i.e., those which do not vanish in the background, which is usually a Friedmann-Lema\\^\\i tre-Robertson-Walker (FLRW) model. Products of first-order quantities can then be ignored in the equations. Whenever the background is a homogeneous and isotropic FLRW model, all projected vectors and tensors are first-order, so there is no vector-tensor and tensor-tensor coupling in the equations. Harmonic functions can then be introduced which re-write the equations in scalar form; the resulting system is then in the form of algebraic constraints and some \\emph{first-order} ordinary differential equations; the solution is then straightforward. The key point of the approach is that it deals with physically or geometrically relevant quantities, such as the fractional density gradient, ${\\cal D}_a$, or the electric and magnetic parts of the Weyl tensor, $E_{ab}$ and $H_{ab}$, respectively, which represent the non-local parts of the gravitational field, and which describe, amongst other things, the propagation of gravitational waves (GW). The variables are also (coordinate) gauge invariant (although there is a frame-gauge freedom in the choice of~$u^a$~-- see below).   The aim of this paper is to extend the gauge-invariant, covariant perturbation formalism to an astrophysical setting. The 1+3 approach is not appropriate for many situations where such techniques would seem highly desirable: when the spacetime in inhomogeneous, for example, the 1+3 equations usually become intractable. However, by introducing an additional frame vector, assuming that the background spacetime has some preferred spatial direction (such as in spherically symmetric, or more general \\emph{locally rotationally symmetric} spacetimes, or $G_2$ spacetimes) we can in many cases recover all of the advantages of the 1+3 equations, but in a \\emph{1+1+2 covariant} framework~\\cite{henk}. In this paper we introduce the 1+1+2 formalism, and apply it to linear perturbations of a Schwarzschild spacetime. Not only is this a first step in applying the procedure to more general astrophysical situations (such as perturbations of the interior of compact objects, collapsing and exploding stars, etc.), it also represents an important field of study in itself: with the development of large gravitational wave detectors (e.g.,~\\cite{LIGO}) an improved understanding of the problem of GW propagation around compact objects is certainly timely. The power of the 1+1+2 technique is clearly shown by the significant results we are able to obtain, relatively simply. For example, we show here that the full description of gravitational waves around a Schwarzschild black hole is governed by closed covariant wave equation, unifying both parities in a single covariantly defined gauge- and frame-invariant transverse-traceless 2-tensor, $W_{ab}$.   Linear perturbations of Schwarzschild black holes have been conventionally studied through perturbations of the metric tensor (or via the Newman-Penrose formalism~\\cite{chandra}). In the metric approach, fluctuations of the spacetime geometry are characterised by perturbations in the metric tensor; these fluctuations are determined by closed wave equations~-- the Regge-Wheeler equation for odd parity perturbations~\\cite{RW} and the Zerilli equation in the even parity regime~\\cite{zerilli}. These wave equations act on linear combinations of the functions (and their derivatives) appearing in the perturbed metric, but these functions do not determine directly the gravitational waves which they represent; a general coordinate transformation would preserve neither the perturbation functions themselves, nor the wave equations which they satisfy. The approach we develop here is completely covariant, so such issues do not arise. Instead, corresponding wave equations we derive here are formed from covariant and gauge-invariant variables which have a physical significance; furthermore, they do not require a harmonic splitting for their derivation.  The formalism we develop here relies on a further splitting of the spacetime using a radial vector $\\n^a$, in addition to the usual splitting with the timelike vector $u^a$ used in the 1+3 approach. We split the Ricci and Bianchi identities using $u^a$ and $\\n^a$ into a coupled set of \\emph{first order} differential equations, plus some constraints. The differential operators we use are along the two vector fields which give us a set of \\emph{evolution} and \\emph{propagation} (along the `radial' direction) equations, while a derivative formed from a projection orthogonal to $u^a$ and $\\n^a$ gives a small number of constraints. The differential equations involve the covariant variables derived from splitting the Weyl tensor (and more generally the Ricci tensor, but this is zero here as we only consider vacuum perturbations), and the kinematics of $u^a$ and $\\n^a$. As our background is static and spherically symmetric, we may use harmonic functions for our evolution and projected derivatives, putting our equations into the form of a first order system of ordinary DE's and constraints, which can then be tackled relatively easily.   Previously,~\\cite{GS,MGG,ST,J} (and references therein) developed approaches to stellar and black hole perturbations  similar to the method presented here in the sense that they use two orthogonal vectors to form their time and space derivatives. These approaches are fundamentally different from our approach, however, in that they formulate their differential equations as \\emph{second order} PDE's derived from Einstein's field equations (EFE), the solutions of which give the metric functions (as in all metric perturbation approaches). On the other hand, our system of DE's is manifestly first order, as it is derived from the first order Ricci and Bianchi identities~\\footnote{We use the Ricci identities applied to our vectors $u^a$ and $\\n^a$, so they are second order DE's of $u^a$ and $\\n^a$; however, our covariant objects are different projections of $\\del_au_b$ and $\\del_a\\n_b$, so that the Ricci identities are first order DE's in these covariant objects.}, as it involves physical or geometric quantities, and not the metric functions. Second order wave equations may be derived if desired. This is one of the key properties of covariant or tetrad methods. This change in derivative level is conceptually analogous to the change in going from the Lagrangian to Hamiltonian formulations of classical mechanics, from configuration space to phase space.           The layout of the paper is as follows: in the following section we discuss the merits of a 1+1+2 decomposition of the field equations and set out the 1+1+2 covariant formalism. Then, in section~\\ref{the-equations} we present the full set of 1+1+2 covariant, gauge-invariant, first-order equations, linearised about a Schwarzschild background and introduce the (spherical and temporal) harmonics on the `sheet', which enable the equations to be reduced to a set of coupled ODEs for the 1+1+2 covariant variables. In section~\\ref{RWsec} we prove the existence of a transverse-traceless (TT) tensor that satisfies a closed wave equation equivalent to the Regge-Wheeler equation, valid for harmonics of either parity. Following this we discuss the even parity variable which satisfies a wave equation equivalent to the Zerilli equation; we demonstrate here the existence of an odd parity counterpart, but defer the derivation of the `Zerilli tensor' until later, Sec.~\\ref{zertens}. Then, in section~\\ref{solution-section} we describe in detail the (matrix-based) method of solution of the linear, first-order system of ODEs for the harmonic components, emphasising the significance of the freedom to choose the frame vectors and showing that with an appropriate choice we can reduce the whole solution for both parities to a single 2-dimensional ODE. The wave equations for the TT electric and magnetic Weyl tensors are also presented, and the closed four-dimensional ODE which they also satisfy is discussed. Finally, we discuss the results we have obtained.    We follow the notation and conventions of~\\cite{HvE}. ",
        "conclusions": "We have presented a new perturbation formalism for dealing with systems with spherical symmetry in the background, and we have applied this to the simplest of such systems, the Schwarzschild black hole. Our 1+1+2 splitting allowed the Schwarzschild solution to be given in covariant form. We then demonstrated that we can derive the usual equations governing perturbations of the spacetime, namely the Regge-Wheeler equation~(\\ref{RW}), and the Zerilli equation~(\\ref{zerilli}), while discussing in detail our new method. We have also shown that there exist Regge-Wheeler and Zerilli \\emph{tensors} which unify the odd and even parity perturbations; indeed, the Regge-Wheeler tensor was shown to satisfy a closed \\emph{covariant} wave equation which governs the dynamics of the whole problem. This sets the basis for future studies of more general astrophysical systems, which only require an appropriate change of the background, for which the possible applications are myriad.  The method has several important aspects. The first is the covariant spacetime splitting itself. In general, the two threading vectors $u^a$ and $n^a$ may be chosen arbitrarily, defining the sheet  on which the vectors and tensors exist (strictly speaking, in general the sheet is not a true surface, but a collection of tangent planes). This makes the approach a halfway house between the 1+3 approach and the orthonormal tetrad approach, and provides a completion of the covariant formalism. (Recall that a unique tetrad cannot be defined in isotropic or locally rotationally symmetric spacetimes~\\cite{HvE,henk}, so this is the ideal compromise between the two in such cases.) For systems with spherical symmetry in the background, we have seen that when  $\\n^a$ is radial in the background,  the perturbed spacetime becomes a tractable problem, because all vectors and tensors become first order, allowing decomposition with suitable harmonic functions~-- spherical harmonics, in this case. Time harmonics are also introduced to simplify the solution, allowed when the background is static, but these are not strictly necessary, as the dot-derivative is a scalar operator, and can be dealt with by standard techniques. \\forget{(We may regain the dot-derivative solution to this problem by replacing $i\\omega\\rightarrow d/d\\tau$ in all the equations, as already discussed.)} So far this merely writes the equations in an alternative form. At this point finding the solution is relatively trivial: our approach simply requires solving a linear system of algebraic equations, \\emph{and this is all there is to it}. The underlying dynamics are then \\emph{automatically} given by the small system of differential equations that remain~-- wave equations, if desired, then may be derived (where they exist) by differentiating this equation.  An important physical aspect of our approach is that it uses a set of covariantly defined quantities with genuine physical significance, which makes it clear which objects are crucial for the detection and measurement of gravitational waves. Put simply, GW detectors essentially measure gravitational tidal forces; that is, they are sensitive to the dynamical Weyl curvature, encoded in the electric Weyl tensor, $E_{ab}$, and this dynamical Weyl curvature forces the working parts of any GW detector through the right hand side of the geodesic deviation equation. We have shown that there is a gauge-invariant TT tensor that describes GWs of either parity, and is closely related to the variation of the radial tidal force. Thus it is clear that our formalism is dealing with real, physically measurable, objects from the start. Indeed, we have discussed how a subset of four of all thirty-three variables is sufficient to determine the full dynamics of the spacetime.    There are many possible extensions of the work we have presented. The obvious extension is to perturbations of static stars, but we envisage that our method will be widely applicable to many other astrophysical situations, such as systems with a \\emph{dynamical} background~-- e.g., collapsing stars, type Ia supernovae, etc.~-- where the scalar background equations have two (non-tensorial) derivatives in them. In the static background case, where we can introduce time-harmonic functions in the perturbed equations, solving the equations is a simple problem of solving a linear system of equations, and one is left with a first order non-autonomous system of ordinary differential equations, whose dimension is small compared to the original system, plus linear relations among the remaining variables in terms of the basis vector of the dynamical system. All the physics of the spacetime is contained in this small dynamical system. With a dynamical background, it may not be as simple as this, but we do not envisage it being much more difficult to find the full solution; one may have to be careful choosing one's observers (perhaps comoving with the matter in the case of a collapsing star). In any event, there is much that can be achieved. In any situation where there is a preferred spacelike vector field present, the covariant 1+1+2 sheet formalism should provide new insight."
    },
    "0209/astro-ph0209016_arXiv.txt": {
        "abstract": "As the chiral symmetry is widely recognized as an important driver of the strong interaction dynamics, current strange stars models based on MIT bag models do not obey such symmetry. We investigate properties of bare strange stars using the Cloudy Bag Model, in which a pion cloud coupled to the quark-confining bag is introduced such that chiral symmetry is conserved. The parameters in the model, namely the bag constant and strange quark mass are determined self-consistently by fitting the mass spectrum of baryons. Then the equation of state is obtained by evaluating the energy-momentum tensor of the system. We find that the stellar properties of the Cloudy Bag Strange Stars are similar to those of MIT Bag Models. However, the decay of pions is a very efficient cooling way. In fact it can carry out most the thermal energy in a few milliseconds and directly convert them into 100MeV photons via pion decay. This may be a very efficient $\\gamma$-ray burst mechanism. Numerical results indicate that temperature of a Cloudy Bag strange star is sufficiently lower than a MIT one for the small gap energy of color superconductivity($\\Delta$ = 1MeV). On the other hand, large gap energy ($\\Delta$ = 100MeV) can suppress the pion emissivity and hence the cooling curves of Cloudy model and MIT model are almost identical. The long term cooling behaviors of both MIT model and Cloudy model are determined by the color-flavor locked phase. The surface luminosity of a bare strange star is higher than that of a neutron star until $10^6s$ and $10^8s$ for ($\\Delta$ = 100MeV) and ($\\Delta$ = 1MeV) respectively. After this period, the surface luminosity of a bare strange star becomes lower than that of a neutron star even rapidly cooling mechanisms, e.g. direct URCA process or pion condensation, exist in the neutron stars. Hence, the cooling behavior may provide a possible way to distinguish a compact object between a neutron star, MIT strange star and Cloudy Bag strange star in observations. ",
        "introduction": "It has been argued that strange quark matter, consisting of $u$-, $d$- and $s$-quarks, is energetically the most favorable state of quark matter~\\citep{bo71, ta79, wi84}. \\citet{wi84} suggested that there are two ways to form strange quark matter: the quark-hadron phase transition in the early universe and conversion of neutron stars into strange ones at ultrahigh densities. If this strange matter hypothesis is correct, then it has profound impact on physics and astrophysics. Pulsars could be strange stars and not neutron stars as previously thought, and there could even be many strange dwarfs and strange planets in the universe~\\citep{al86, ha86, ma91, gl97, xu2001}.   Several mechanisms have been proposed for the formation of strange quark stars. For example, strange stars are expected to form during the collapse of the core of a massive star after a supernova explosion~\\citep{da95}. Another possibility is that some rapidly spinning neutron stars in low-mass X-ray binaries (LMXBs) can accrete sufficient mass to initiate a phase transition to become strange stars~\\citep{ch96}. Some of the millisecond pulsars may be strange stars, because LMXBs are believed to be the progenitors of millisecond pulsars. Strange stars have also been proposed as sources of unusual astrophysical phenomena, such as soft $\\gamma $-ray repeaters~\\citep{ch98c, cd2002}, pulsating X-ray bursters~\\citep{ch98b}, cosmological $\\gamma $-ray bursts~\\citep{ch96, da95, cheng2001}, SAX J1808.4-3658~\\citep{li99} etc. The discovery of kHz quasi-periodic oscillations in LMXBs~\\citep{zh98} implies that the compact stellar object must have a very soft equation of state (EOS), which is consistent with that of strange stars~\\citep{kl98}. Recently, Hong et al. ~\\citep{h2001}have argued that the first order phase transition of the color superconductivity occurring at the end point of some massive stars can release sufficient energy to produce hypernovae, which are considered as progenitors of cosmological $\\gamma$-ray bursts.    However, important as this strange-matter hypothesis maybe, it remains notoriously difficult to be proven or refuted on either observational or theoretical grounds. The binding energy per nucleon for strange matter is estimated to be close to that of Fe$^{56}$, but it has not been possible to calculate it from QCD to within the few percents level required to address the question. Model calculations of the binding energy are clearly inadequate and unreliable~\\citep{fa84, ma91, gl97}.  Lacking an accurate and reliable method to calculate the strange matter EOS, the structure and stability of strange stars have been addressed~\\citep{fa84, al86, ha86, ma91, gl97} typically in simplified models such as the MIT Bag Model~\\citep{ch74}. The quarks are treated as relativistic free particles confined in an impenetrable bag. Approximately then, the pressure comprises of the fermi gas pressure of the quarks minus a bag pressure, which mimics the strong interaction that holds the quarks together. Perturbative corrections to such a simple EOS can be made based on one-gluon exchange calculations~\\citep{fa84}. With this EOS, it has been shown that self-bound stars exist with virtually no lower limit on the stellar mass~\\citep{al86, ha86}; these stars are basically giant hadrons, held together by strong interaction rather than gravity.  It is arguable whether meaningful refinements of theoretical studies of strange stars can be made at this stage. Uncertainties in the strong interaction calculations easily make or kill strange stars, and arguments over details in model-dependent EOS's are probably futile. Nevertheless, one could ask what kinds of effects commonly accepted ingredients of strong interaction have on the stability and structure of strange stars. Whereas such studies are not meant to be quantitatively accurate, one may possibly gain insight into the physics of compact stars.  With this in mind, we study the implications of chiral symmetry on the structure of strange stars. Chiral symmetry is widely recognized as an important driver of the strong interaction dynamics, witness the celebrated experimental success of partially conserved axial current (PCAC), which signifies the importance of chiral SU(2)xSU(2) symmetry~\\citep{bh88}. Phenomenologically, the partial breaking of chiral symmetry is associated with the emergence of pion as a pseudo-Goldstone boson and hence related to the small pion mass~\\citep{bh88}; the coupling of pions to quarks/hadrons is thus a natural result of chiral dynamics and an essential ingredient of strong interaction physics.  There are many hadron models that incorporate various features of QCD ~\\citep{bh88}. Whereas the MIT Bag Model has the asymptotic freedom and the confinement of quarks built in, it is well known to violate chiral symmetry badly. By coupling the pions to the quarks, thus savaging the axial current at the bag, one can restore chiral symmetry. There are many variants of such chiral bag models~\\citep{my84}, and one can imagine these as either modifying the original bag model to include a meson cloud at the bag, or extending the sigma model to a nucleon with quark structure. For our purpose, a simple, representative chiral bag model suffices, and we have thus chosen the Cloudy Bag model~\\citep{th83, mi84}. ",
        "conclusions": "We have used the Cloudy Bag model to describe the properties of strange stars. The model parameters, i.e., the bag constant $(B=(120\\, {\\rm MeV})^4)$ and the strange quark mass $(m_s = 220 {\\rm\\, MeV})$, are fixed by comparing with hadron masses. We found that the star is surrounded by a thin pion cloud, which is not significant in the stellar mass and radius. Hence, stellar properties are similar in the Cloudy Bag and MIT case, except that the former has slightly larger radius and smaller mass due to a larger $B$. The maximum mass of Cloudy Bag strange stars is about $1.9 \\,M_\\odot$.  For the rotation properties of Cloudy Bag strange stars, the approximate formulae were employed. We found that the Kepler limit for Cloudy Bag strange stars with the best-fit parameters is about 6500\\,s$^{-1}$, which is lower than the MIT one by $\\sim 30\\%$. The maximum mass of a stable rotating star is about $2.3\\,M_\\odot$, similar to the value $2.4\\,M_\\odot$ of a MIT strange star.   The cooling properties of bare strange stars sensitively depend on the gap energy. For small gap energy ($\\Delta$ = 1MeV), the cooling properties of MIT strange stars and Cloudy strange stars are very much different because the pion emission can cool the strange star rapidly. For the large gap energy ($\\Delta$ = 100MeV), the quarks become superconducting immediately after the star is formed, and the pion emission is totally suppressed. MIT strange stars and Cloudy strange stars become no difference. Furthermore, large gap energy also creates the effect of color-flavor locked, which significantly reduces the electron concentration and its capacity. This makes the cooling curves lower than those with small gap energy. From fig13 and fig14, we can see that the surface luminosity of a bare strange star is higher than that of a neutron star until $10^6s$ and $10^8s$ for ($\\Delta$ = 100MeV) and ($\\Delta$ = 1MeV) respectively. After this period, the surface luminosity of a bare strange star becomes lower than that of a neutron star even rapidly cooling mechanisms, e.g. direct URCA process or pion condensation, exist in the core of neutron stars. Hence, the cooling behavior may provide a possible way to distinguish a compact object between a neutron star, MIT strange star and Cloudy Bag strange star in observations.  It is interesting to note that although the MIT and Cloudy strange stars are hotter than neutron stars before  $\\sim 10^{8}$\\,s for $\\Delta$ = 100MeV and  $\\sim 10^{11}$\\,s for $\\Delta$ = 1MeV respectively(cf. fig10), the thermal radiation of neutron stars are still stronger than those of strange stars Until until $10^6s$ and $10^8s$ for ($\\Delta$ = 100MeV) and ($\\Delta$ = 1MeV) respectively. It is because the cooling mechanism at the later stage is dominated by the black body radiation, which is strongly suppressed for strange stars as pointed in section IVB.  >From fig13 and fig14, we can see that the cooling curves of a MIT strange star and a cloudy strange star are the same when the cooling mechanism is dominated by blackbody radiation. However, the radiation properties of a MIT strange star and a cloudy strange star are very much different in the early stage ( t $< 10^6s$ ) for small gap energy case. The cloudy strange star is first cooling very rapidly by emitting pions, which decay to 100MeV photons and its temperature drops quickly below that of MIT strange star. Therefore the MIT strange star is cooling mainly by emitting electrons/positrons. The total energy of a cloudy strange star carried away by 100MeV photons is $\\sim 10^{52}ergs$, which makes the evaporation of pions a very efficient $\\gamma$-ray burst mechanism.  Finally, we want to remark that the cooling curves of strange stars are lower than those cooling curves of neutron stars, which are consistent with the current observed data ~\\citep{ts98}. Therefore we actually suggest that the cooling curves may be one possible way to differentiate strange stars from neutron stars. For example, strange stars can behave as pulsars if they have strong magnetic field~\\citep{xu2001}~\\citep{xb2001}. Therefore their ages can be estimated by their period and period derivative. If these pulsars are found to have thermal emission lower than that of the standard cooling neutron stars, then we can further to see if there are any e$^{\\pm}$ annihilation lines, which are red shifted larger than those of neutron stars (cf. fig. 5). The features of thermal X-ray emitted from these pulsars associated with bare strange stars are also very different from those of neutron stars~\\citep{xu2002}.  This work is partially supported by a RGC grant of the Hong Kong Government."
    },
    "0209/astro-ph0209220_arXiv.txt": {
        "abstract": "{An empirical calibration is presented for the synthetic Lick indices (e.g.~\\mg,~\\fe,~\\hb, etc.) of Simple Stellar Population (SSP) models that for the first time extends up to solar metallicity.  This is accomplished by means of a sample of Milky Way globular clusters (GCs) whose metallicities range from $\\sim \\zsun/30$ to $Z\\sim\\zsun$, thanks to the inclusion of several metal rich clusters belonging to the Galactic bulge (e.g., NGC~6553 and NGC~6528).  This metallicity range approaches the regime that is relevant for the interpretation of the integrated spectra of elliptical galaxies. It is shown that the spectra of both the globular clusters and the Galactic bulge follow the same correlation between magnesium and iron indices that extends to elliptical galaxies, showing weaker iron indices at given magnesium indices with respect to the predictions of models that assume solar-scaled abundances.  This similarity provides robust empirical evidence for enhanced \\afe~ratios in the stellar populations of elliptical galaxies, since the globular clusters are independently known to have enhanced \\afe~ratios from spectroscopy of individual stars.  We check the uniqueness of this $\\alpha$-overabundance solution by exploring the whole range of model ingredients and parameters, i.e. fitting functions, stellar tracks, and the initial mass function (IMF). We argue that the {\\it standard} models (meant for solar abundance ratios) succeed in reproducing the Mg-Fe correlation at low metallicities ($\\zh\\lapprox-0.7$) because the stellar templates used in the synthesis are Galactic halo stars that actually are $\\alpha$-enhanced. The same models, however, fail to predict the observed Mg-Fe pattern at higher metallicities ($\\zh \\gapprox-0.7$) (i.e., for bulge clusters and ellipticals alike) because the high-metallicity templates are disk stars that are not $\\alpha$-enhanced.  We show that the new set of SSP models which incorporates the dependence on the \\afe~ ratio (Thomas, Maraston \\& Bender~2002) is able to reproduce the Mg and Fe indices of GCs at all metallicities, with an $\\alpha$-enhancement~$\\afe=+0.3$, in agreement with the available spectroscopic determinations. The \\hb~index and the higher-order Balmer indices are well calibrated, provided the appropriate morphology of the Horizontal Branch is taken into account. In particular, the Balmer line indices of the two metal rich clusters NGC~6388 and NGC~6441, which are known to exhibit a tail of warm Horizontal Branch stars, are well reproduced.  Finally, we note that the Mg indices of very metal-poor ($\\zh \\lapprox-1.8$) populations are dominated by the contribution of the lower Main Sequence, hence are strongly affected by the present-day mass function of individual globular clusters, which is known to vary from cluster to cluster due to dynamical effects. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209199_arXiv.txt": {
        "abstract": "We present an analysis of precision radial velocity measurements for 580 stars from the Keck survey. We first discuss the detection threshold of the survey, and then describe a Bayesian approach to constrain the distribution of extrasolar planet orbital parameters using both detections and upper limits. ",
        "introduction": "Attempts to characterize the distribution of mass, orbital radius, and eccentricity of extrasolar planets are hampered by the lack of knowledge of detection sensitivities. Following our previous study of the Lick survey (Cumming, Marcy, \\& Butler 1999, hereafter CMB), the aim of this work is to carefully assess the detection threshold and selection biases for a sample of stars from the Keck survey, and use this information to constrain the distribution of orbital parameters. ",
        "conclusions": ""
    },
    "0209/astro-ph0209150_arXiv.txt": {
        "abstract": "We use high resolution numerical simulations to study whether gravitational instabilities within circumstellar discs can produce astrometrically detectable motion of the central star. For discs with masses of $M_{\\rm disc} = 0.1 \\ M_*$, which are permanantly stable against fragmentation, we find that the magnitude of the astrometric signal depends upon the efficiency of disc cooling. Short cooling times produce prominent filamentary spiral structures in the disc, and lead to stellar motions that are potentially observable with future high precision astrometric experiments. For a disc that is marginally unstable within radii of $\\sim 10 \\ {\\rm au}$, we estimate astrometric displacements of $10-10^2 \\ \\mu {\\rm arcsec}$ on decade timescales for a star at a distance of 100~pc. The predicted displacement is suppressed by a factor of several in more stable discs in which the cooling time exceeds the local dynamical time by an order of magnitude. We find that the largest contribution comes from material in the outer regions of the disc and hence, in the most pessimistic scenario, the stellar motions caused by the disc could confuse astrometric searches for low mass planets orbiting at large radii. They are, however, unlikely to present any complications in searches for embedded planets orbiting at small radii, relative to the disc size, or Jupiter mass planets or greater orbiting at large radii. ",
        "introduction": "High precision astrometry is a powerful tool to search for companions to nearby stars.  It also has the potential to discover significant numbers of extrasolar planetary systems.  In this paper we discuss the potential of astrometry as a probe of self-gravitating discs around pre-main sequence stars.  A gaseous disc with sound speed $c_s$, surface density $\\Sigma$, and epicyclic frequency $\\kappa$ is described as self-gravitating if the Toomre (1964) $Q$ parameter, \\begin{equation} Q = { {c_s \\kappa} \\over {\\pi G \\Sigma} }, \\end{equation} is of order unity. In discs where self-gravity is important, the outcome can either be fragmentation into one or more bound objects, or a quasi-steady state in which gravitational instabilities lead to the outward transport of angular momentum. Local simulations suggest that the boundary between these possibilities is set by the ratio of the local dynamical time-scale $\\Omega^{-1}$ to the time-scale on which the disc radiates thermal energy. Fragmentation occurs whenever the cooling time $t_{\\rm cool} \\ale 3 \\Omega^{-1}$, while longer cooling times lead to stable angular momentum transport (Gammie 2001).  Circumstantial evidence suggests that self-gravity could play a role in protoplanetary discs as late as the optically visible Classical T Tauri phase, which lasts for several Myr (Haisch, Lada \\& Lada 2001). Evidence that relatively old disc may be self-gravitating comes, first, from models of FU Orionis outbursts, which require a low efficiency of disc angular momentum transport to reproduce the observed $\\sim 10^2 \\ {\\rm yr}$ time-scales. If the viscosity is parameterized using the Shakura-Sunyaev (1973) prescription, $\\nu = \\alpha c_s h$, where $h$ is the vertical scale height, FU Orionis models suggest a quiescent $\\alpha \\sim 10^{-4}$ (Bell \\& Lin 1994; Bell et al. 1995). For a given accretion rate, small values of $\\alpha$ imply high surface densities, so that the disc would be self-gravitating at $r \\sim 1 \\ {\\rm au}$. Second, theory suggests that angular momentum transport ought to be suppressed in cool regions of the disc where the gas is poorly coupled to magnetic fields (Matsumoto \\& Tajima 1995; Gammie 1996; Fleming, Stone \\& Hawley 2000; Sano et al. 2000; Reyes-Ruiz 2001; Sano \\& Stone 2002). Again, this suggests that self-gravity may set in at radii of a few au as the first significant non-magnetic source of angular momentum transport (Armitage, Livio \\& Pringle 2001). Ascertaining when self-gravity is at work within the disc requires either the observation of spiral patterns using extremely high resolution imaging, or detection of the astrometric motion of the stellar photocentre induced by the self-gravitating disc. It has been shown (Adams et al. 1989) that self-gravitating perturbations with $m=1$, can force the central star to move from the centre of mass.  In this paper, we use numerical simulations to quantify the magnitude of the astrometric displacement. This has previously been studied by Boss (1998), who simulated a disc with a mass of $\\approx 0.2 M_*$ and found a large motion of the star, of the order of 0.1~au. This corresponds to milliarcsecond displacements at the distance of the nearest star-forming regions, which would be easily detectable by any of the forthcoming high precision astrometry experiments. The disc simulated by Boss (1998), however, was highly unstable, and subsequently fragmented with the formation of substellar objects. Although promising for giant planet formation (Armitage \\& Hansen 1999; Boss 2000), prompt fragmentation implies that extremely fortuitous timing would be needed for the astrometric detection of self-gravitating discs. We concentrate instead on marginally unstable discs, which are not vulnerable to fragmentation and could potentially exist around many Classical T Tauri stars. ",
        "conclusions": "We have used numerical simulations of discs around Young Stellar Objects to quantify the astrometric displacement of the star caused by a self-gravitating disc. By modelling the energy balance of the disc using a cooling time formalism (Gammie 2001), we have shown that the magnitude of the displacement depends upon how stable the disc is against fragmentation into bound substellar objects. For a cooling time of $5 \\Omega^{-1}$ (within a factor of two of the fragmentation boundary at $\\approx 3 \\Omega^{-1}$), a disc mass of $0.1 \\ M_*$, and a self-gravitating disc radius of the order of 10~au, we obtain relatively large displacements in the $10-10^2 \\ \\mu {\\rm arcsec}$ range (for a star at a distance of 100~pc). This magnitude of astrometric signal is potentially observable with any of several upcoming high precision astrometric instruments, although the presence of a circumstellar disc is likely to complicate such observations, especially if time dependent perturbations are present within the disc. Although the gravitationally driven instabilities modelled in this work are intrinsically time dependent, their growth saturates (Laughlin et al. 1997) and their filamentary nature (Nelson et al. 1998) makes their structure approximately azimuthally symmetric. Their presence is therefore unlikely to further complicate observations. Discs with cooling times longer than $5 \\Omega^{-1}$, which are correspondingly more stable, have significantly smaller stellar motions. A detailed model for the angular momentum transport processes in the disc, together with an assessment of the heating and cooling processes at work, would be required to determine whether protoplanetary discs fall into the parameter regime that produces the largest displacements.  Our estimate for the angular displacement of the star caused by a self-gravitating disc is an order of magnitude smaller than that found by Boss (1998). Although there are numerous differences in the initial conditions and numerical techniques used, we believe that the smaller displacement that we obtain primarily reflects our choice of a more stable disc model. By design, our disc is permanently stable against fragmentation into substellar objects, whereas the disc simulated by Boss (1998) broke up towards the end of the simulation. Taken together, our results, plus those of Boss (1998), suggest that a relatively narrow window of self-gravitating disc conditions could lead to small but long-lived astrometric wobbles ($10-10^2 \\ \\mu {\\rm arcsec}$ at 100~pc). More dramatic displacements -- perhaps of the order of a mas at the same distance -- are also possible, but only as a precursor to fragmentation of the disc into substellar objects. Our simulations suggest that the disc induced wobble could mimic that of a planet a few times smaller than Jupiter and orbiting at a large radius, relative to the disc size. Most T Tauri stars have disc masses smaller than that used in this simulation and would consequently induce a correspondingly smaller motion of the central star, reducing  the likelihood of confusion between stellar motion due to disc instabilities and that due to an orbiting planet. The disc induced stellar motions are therefore not likely to be a serious obstacle to the astrometric detection of planets orbiting at small radii (relative to the disc size). For planets orbiting at large radii, the planet mass which could be mimiced by a disc instability depends on the disc mass (relative to the central star) but is likely to be less than a Jupiter mass for most T Tauri disc systems.  \\begin{figure} \\centerline{\\psfig{figure=fig7.ps,width=2.5truein,height=2.6truein}} \\caption{\\label{tc10rt} Time evolution of the distance of the central star from the center of mass for $t_{cool}=10 \\Omega^{-1}$. At the end of the simulation the system seems to have settled into a steady state in which the central star is orbiting the center of mass with a orbital radius of $\\sim 10^{-3}$ radial units.} \\end{figure}  \\begin{figure} \\centerline{\\psfig{figure=fig8.ps,width=2.5truein,height=2.6truein}} \\caption{\\label{tc5rt} Time evolution of the distance of the central star from the center of mass for $t_{cool}=5 \\Omega^{-1}$. At the end of the simulation the system seems to have settled into a steady state in which the central star is orbiting the center of mass with a orbital radius of $\\sim 5 \\times 10^{-3}$ radial units.} \\end{figure}  \\begin{figure} \\centerline{\\psfig{figure=fig9.ps,width=2.5truein,height=2.6truein}} \\caption{\\label{tc5xy} Projection of the position of the central star onto the orbital plane.  The center of mass is at $x=0, y=0$. The orbit is approximately circular and has substructure due to perturbations from higher order modes.} \\end{figure}"
    },
    "0209/astro-ph0209215.txt": {
        "abstract": "We review the present status of Cosmic Microwave Background (CMB) anisotropy observations and discuss the main related astrophysical issues, instrumental effects and data analysis techniques. We summarise the balloon-borne and ground-based experiments that, after $COBE$-DMR, yielded detection or significant upper limits to CMB fluctuations. A comparison of subsets of combined data indicates that the acoustic features observed today in the angular power spectrum are not dominated by undetected systematics. Pushing the accuracy of CMB anisotropy measurements to their ultimate limits represents one of the best opportunities for cosmology to develop into a precision science in the next decade. We discuss the forthcoming sub-orbital and space programs, as well as future prospects of CMB observations. ",
        "introduction": "\\label{intro}  The Cosmic Microwave Background (CMB) radiation has played a central role in modern cosmology since the time of its discovery by Penzias and Wilson in 1965 \\cite{penwil65}. The existence of a background of cold photons was predicted several years before by Gamow, Alpher and Herman \\cite{gamow48,alpher48} following their assumption that primordial abundances were produced during an early phase dominated by thermal radiation. Traditionally, the CMB is considered one of the three observational pillars supporting the cosmological scenario of the Hot Big Bang, together with light elements primordial abundances (see, e.g., \\cite{light}) and the cosmic expansion \\cite{hubble26}. In recent years, CMB measurements are widely considered as the single most fruitful field in observational cosmology, thanks to the richness and precision of information it can provide directly from the early universe.  In the first decade or so after the discovery, experiments and theoretical work focussed on establishing the nature of the CMB itself, leading to a compelling evidence of a cosmic origin of the radiation. By the early 80's the Hot Big Bang prediction of a highly isotropic background \\cite{alpher48} with a nearly Planckian spectrum was remarkably supported by observation. The growing CMB community then shifted the interest to observations of first-order deviations from the ``idealised'', unperturbed scenario. Spectrum experiments searched for spectral distortions \\cite{danedezo77} capable, if detected, to set constraints on cosmological parameters, such as the baryon density $\\Omega_b$, and on the thermal history of the universe \\cite{burigana91}. Measurements of the CMB angular distribution were carried out with increasing accuracy to detect CMB anisotropy.  A new phase was opened up by the successful outcome of the $COBE$ mission in the early 90's. The FIRAS experiment \\cite{fixsen96} established that the CMB spectrum is planckian within limits as tight as 0.03\\% in the frequency range 60-600~GHz with a temperature of $T_0 = 2.725\\pm 0.002$~K. Coupled with sub-orbital measurements at low frequencies \\cite{bersa94,deami91,bensa93,smoot87,siro91,ruben00} very stringent constraints to distortion parameters were placed ($|\\mu|<9\\times 10^{-5}$ for ``chemical'' distortions, $y<1.5\\times 10^{-5}$ for Compton distortions and $|Y_{ff}| <3\\times 10^{-6}$ for free-free distortions) leading to tight upper limits on energy injections in the early universe.  The FIRAS results illustrate the level of precision that cosmology can seek based on observations of the CMB. The possibility of ``precision cosmology\" with the CMB comes from the combination of three factors. First, measurements of the CMB can be made with exquisite accuracy thanks to the extraordinary progress achieved by microwave and sub-mm technology in recent years. Second, we have now good evidence that astrophysical emission in the microwaves, that adds to the cosmological signal, does not prevent in principle the observation of the subtle intrinsic characteristics of the CMB. And, third, the theoretical interpretation of CMB data is relatively simple, since the features we observe in the microwave background carry information directly from epochs when all the processes were still in the linear regime.  In addition to the FIRAS results, the other major result of $COBE$ was the first unambiguous detection, by the DMR instrument, of anisotropies at a level $\\Delta T/T \\approx 10^{-5}$ on large angular scales \\cite{smoot92}. This breakthrough immediately stimulated the realisation of many new experiments aiming at measuring the CMB angular distribution with increasing resolution and sensitivity. As we shall see in detail, today more than 20 independent projects carried out with different technologies and from a variety of observing sites have reported anisotropy detection at similar $\\Delta T/T$ levels over a wide range of angular scales. At present, these are far from the precision measurements achieved by FIRAS on the CMB frequency spectrum, but the new generations of space-based anisotropy experiments is designed to eventually reach FIRAS-like accuracy in the angular power spectrum. It is important however to highlight a fundamental difference: while FIRAS made a very accurate measurement of an essentially null result (no spectral distortions detected), the present anisotropy data already demonstrate the presence of angular structure at ``non-zero'' levels. The statistical properties of these fluctuations and their dependence upon cosmological parameters can be easily computed and compared to the observed maps, without the complications that arise in non-linear processes. The details of the angular power spectrum are at reach and can reveal a wealth of genuine new information on the early universe.  In this work we present an overview of the present status of the observations of CMB anisotropy covering the ``post-$COBE$ era''. This is an extremely rapidly evolving field (e.g. \\cite{hudodel,scowhisil}) and new important results are published almost every month. This makes a comprehensive review a very hard, if not impossible, task. In particular, the NASA $MAP$ satellite, launched in June 2001, is carrying out its first survey as we write. By the time this work is published, the first release of the $MAP$ data will be shortly expected, and will have a major impact in the field. Furthermore, the planned ESA mission {\\sc Planck} will have reached a more mature stage and some information given here may turn out to be incomplete.  While a detailed discussion of CMB polarisation is out of the scope of this review, it will be briefly mentioned and discussed, when relevent, in connection with temperature anisotropy. The paper is organised as follows. In Section~\\ref{cmbanirev} we briefly discuss the standard scenario for the origin of CMB fluctuations and outline the scientific information they encode. In Section~\\ref{astrolimit} we address the astrophysical limitations faced by observations, mainly represented by confusion emission of galactic and extragalactic origin. We then discuss observational issues typical of CMB anisotropy experiments (Section~\\ref{obsissue}), the most important systematic effects that they have to fight to reach the desired precision (Section~\\ref{siste}), and some of the aspects related to the analysis of CMB data (Section~\\ref{analysis}), in particular for what concerns the challenges posed by the large data sets expected in the near future. Then in Section~\\ref{experiments} we attempt an overview of the anisotropy experiments carried out in the past decade, and we give a synthesis of the observational status. Finally, we outline the main features of the $MAP$ and {\\sc Planck} missions and future sub-orbital programs, and discuss some prospects for the future of CMB studies. ",
        "conclusions": "\\subsection{The Future} \\label{future}  The {\\sc Planck} mission is expected to bring the long-lasting effort of measurements of the temperaure power spectrum to its completion. {\\sc Planck}'s precision in the determination of the $C_\\ell$ may be analogous to that achieved by FIRAS in the measurement of the frequency spectrum \\cite{fixsen97, mather94}. After {\\sc Planck}, it is reasonable to expect a decrease of activity in traditional anisotropy experments, similar to the one we have witnessed in spectrum projects after FIRAS\\footnote{Actually, a few high precision experiments have been performed in recent years \\cite{staggs96a, staggs96b} and some new plans are being made \\cite{kogut02}}. However, we argue that a lot of exciting CMB observations should happen after {\\sc Planck}, and indeed in some areas {\\sc Planck} will act as a pioneering mission rather than as a conclusive one. Two main research directions can be anticipated: precision measurements of CMB polarisation; and deep imaging at sub-arcminute-scales.  {\\it Polarised anisotropy --} Though extraordinary, the cosmological information obtainable with temperature anisotropy alone is far less than what could be achieved with a future high-precision full-sky observation of the CMB polarisation. Linear polarisation in the microwave background arises from Thomson scattering of anisotropic radiation at last scattering \\cite{rees68}, with an expected amplitude 1\\% to 10\\% of the temperature anisotropy. The polarisation depends sensitively on the fluctuations on the LSS, and thus encodes a wealth of cosmological information, some of which is complementary to the temperature anisotropies. Current upper limits (see \\cite{hedman01} and references therein) are at a level 10-15 $\\mu$K, but the expected signal ($\\sim 5$~$\\mu$K r.m.s., peaking at multipoles $\\ell \\sim 1000$) is now in reach of experiments underway: it is likely that statistical detection at sub-degree scales will be achieved before $MAP$ and {\\sc Planck} data become available (see Section~\\ref{subsec:ongoing}) \\footnote{A claim of detection of polarisation anisotropy using the DASI interferometer has appeared \\cite{kovac02} just before sending this manuscript for printing}. Even the remarkable sensitivity of {\\sc Planck} to $Q$ and $U$ (10--20~$\\mu$K per $10'$ pixel) will be far from fully extracting the information encoded in the cosmic polarised signal. {\\sc Planck} polarisation data can provide fundamental tests on structure formation from initial adiabatic perturbations and break parameter degeneracy, e.g. between tensor mode amplitude and reionisation optical depth \\cite{efstathiou02}. However, much will be left for further experiments to measure with increasing precision the signature of a background of gravitational waves (tensor modes) as anticipated by inflationary models \\cite{selzal97, kamionkowsky97}. A high resolution, high sensitivity polarisation map can be used to discriminate between different inflation models \\cite{caldwell96} and, in particular, to determine the inflationary energy scale. A broad range of inflation models fit the present data, with a flat geometry ($\\Omega_0 \\approx 1$) and nearly scale invariant primordial spectrum ($n_S \\approx 1$). %In principle, inflation %could have occurred at energies anywhere from the electro-weak %scale (100 GeV) to the Planck energy scale ($10^{19}$ GeV). %Current CMB observations limit the energy scale to $\\sim 10^{16}$ %GeV, but these are based on temperature $\\Delta T/T$ data which %cannot unambiguously distinguish between the contributions of %gravitational waves and of density perturbations. Polarisation maps will be needed to determine the inflation energy scale, thus probing ultra-high energy physics to levels beyond what can be obtained with any conceivable terrestrial particle accelerator. Even more subtle, but not less interesting, will be to observe the effects of weak gravitational lensing through the distortion of the CMB polarisation on small scales \\cite{zaldarriaga97}.  Significant progress can be anticipated in CMB polarisation experiments from ground and balloon, but a space mission will be needed in the end to avoid atmospheric effects and to map the entire sky -- which is needed to extend the measurements to low $\\ell$'s and to optimise polarised foreground separation \\cite{bacci01, kogut00}. In fact fluctuations in the synchrotron diffuse component and from extragalactic sources have poorly known and possibly significant impact on the polarised signal. The sensitivity requirements for a full-sky, high-precision polarisation survey are extremely demanding. The expected polarisation amplitude induced by gravitational waves is $\\sim 0.1\\%$ to 1\\% of temperature $\\Delta T/T$: a high signal-to-noise imaging requires a noise per pixel $\\sim$0.05~$\\mu$K, i.e. about 300 times better than {\\sc Planck}. This is beyond the foreseeable future, but an intermediate step with sensitivity 0.5 $\\mu$K per $10'$ resolution element can be approached extrapolating existing technology. Multi-frequency observations, possibly with matching FWHM, would be needed for foreground separation. It is difficult to anticipate which systematic effects could represent limiting factors at sub-$\\mu$K levels; it is likely that thermal stability at the cryogenic temperatures needed by the ultra-high sensitivity detectors (either radiometers or bolometers) will be a major challenge; also, a multi-channel off-axis instrument, pushed at extreme sensitivities, would place critical requirements for ultra-low cross-polarisation.  {\\it Fine scale CMB -- } On angular scales $\\sim 1'$ and below the CMB is influenced by interaction with intervening ionised material (see Sect.~\\ref{subsec:secondary_anisotropies}). The CMB passed through the cosmic ``dark ages'', before star and quasar formation, from which direct observation is extremely difficult to obtain. The study of arcmin scale features in the CMB may well turn out to be the only, or one of the most powerful, technique to gather observational evidence from the very early processes of structure formation. Using the CMB as a high redshift backlight, deep arcminute CMB imaging can probe the early history of galaxy clusters and their gradual acceleration as they fell in their potential wells. In addition, CMB fine scale observations can provide images of the largest structures in the universe as they started to dissipate the heat of their gravitational collapse. {\\sc Planck} will open up this field, but its $>5'$ angular resolution is not sufficient to fully cover this promising area of CMB studies. Furthermore, accurate S-Z measurements can be made to very high redshifts \\cite{stebbins97}, all the way back to the cluster formation era \\cite{aghanim97}. A future sky survey of clusters velocities based on their S-Z features throughout the entire Hubble volume could, in principle, map the evolution of velocity fields over much of the history of the universe.  The S-Z effect from clusters is by far the dominant secondary source (see Sect.~\\ref{subsec:secondary_anisotropies}). Once this will be well identified and mapped, sky regions free from S-Z and other local foregrounds could be searched at $\\sim 1'$ scales for fainter secondary signatures, such as those from gravitational collapse of large scale ($\\sim 100$\\,Mpc) structures, bulk motion of plasma (the ``Ostriker-Vishniac effect''), effects of the evolution of gravitational potentials on CMB photons (the ``integrated Sachs-Wolfe effect'' and the ``Rees-Sciama effect''), lensing-induced signatures from clusters \\cite{seljak00}, signature of local ionisation events \\cite{platania02}, and details of the ionisation history of the universe.  It is easy to expect that the $MAP$ and (especially) {\\sc Planck} surveys will trigger new observations in selected sky areas searching for detailed, physically interesting features. New instruments and observing strategies will be developed for the purpose. At wavelengths $\\lambda \\approx 1~$mm, a $\\lesssim 1'$ resolution translates in a telescope aperture typically $D \\sim 10$~m. The brightest clusters give a S-Z thermal component $\\sim 1$~mK, while the kinetic effects is expected to be $\\sim 10$ times lower. Signatures such as filaments from in-falling clusters are expected at $\\sim$10~$\\mu$K level. Current bolometric or HEMT-based instruments (either filled aperture arrays or interferometers) are able to approach sub-$\\mu$K sensitivity in very localised areas. High precision fine-scale CMB imaging would call for very wide frequency coverage, with ancillary monitoring at low frequency ($<5$~GHz) and high frequency ($>5$~THz) to safely remove unrelated foregrounds as well as  backgrounds (such as primary CMB anisotropy, a source of confusion in this context!). These observations can be expected to progress for several years with ground-based instrumentation. Eventually, an arcminute-scale, full-sky survey to map the evolution of the large scale velocity field will only be possible with a space programme, currently out of reach.  \\subsection{Conclusions} \\label{sec:conclusions}  Observations of the CMB anisotropy can be considered at present as the most powerful method to obtain high quality data on the early universe. We have reviewed the great effort that is being carried out by sub-orbital projects after the $COBE$-DMR discovery. In spite of the major experimental challanges, a coherent picture is emerging. Both balloon-borne and ground-based experiments independently demonstrate the acoustic pattern of the angular power spectrum showing the first three peaks, and both provide a determination $\\sim 10\\%$ accuracy of the angular scale of first peak consistent with a flat geometry, $\\Omega_0 \\approx 1$. This is a truly remarkable achievement. However, it is only a foretaste of the precise (percent-level) determination of several cosmological parameters attainable by the forthcoming generations of experiments, culminating with the {\\sc Planck} space mission. Beyond {\\sc Planck}, it is likely that CMB observations will concentrate on precise polarisation measurements and deep sub-arcmin imaging searching for secondary anisotropies.  Precision measurements of the CMB temperature and polarisation fluctuations are able to shed light on very high energy phenomena occurring in the primordial cosmic environment. Cosmologists need to look more and more towards particle physics to interpret the physical processes probed by CMB data. On the other hand, exploiting the natural laboratory of the early universe to study ultra-high energy scales is of great interest for particle physics. The enormous data sets expected from the planned and future CMB missions are likely to attract the interest of a wider scientific community."
    },
    "0209/astro-ph0209366_arXiv.txt": {
        "abstract": "{We report ISO-LWS far infrared observations of CO, water and oxygen lines towards the protobinary system IRAS4 in the NGC1333 cloud. We detected several water, OH, CO rotational lines, and two [OI] and [CII] fine structure lines. Given the relatively poor spectral and spatial resolution of these observations, assessing the origin of the observed emission is not straightforward.  In this paper, we focus on the water line emission and explore the hypothesis that it originates in the envelopes that surround the two protostars, IRAS4 A and B, thanks to an accurate model. The model reproduces quite well the observed water line fluxes, predicting a density profile, mass accretion rate, central mass, and water abundance profile in agreement with previous works.  We hence conclude that the emission from the envelopes is a viable explanation for the observed water emission, although we cannot totally rule out the alternative that the observed water emission originates in the outflow.  The envelopes are formed by a static envelope where the density follows the $r^{-2}$ law, at $r \\geq 1500$ AU, and a collapsing envelope where the density follows the $r^{-3/2}$ law. The density of the envelopes at 1500 AU from the center is $\\sim 4 \\times 10^6$ cm$^{-3}$ and the dust temperature is $\\sim 30$ K, i.e. about the evaporation temperature of CO-rich ices.  This may explain previous observations that claimed a factor of 10 depletion of CO in IRAS4, as those observations probe the outer $\\leq 30$ K region of the envelope. The water is $\\sim 5 \\times 10^{-7}$ less abundant than H$_2$ in the outer and cold envelope, whereas its abundance jumps to $\\sim 5 \\times 10^{-6}$ in the innermost warm region, at $r\\leq 80$ AU where the dust temperature exceeds 100 K, the evaporation temperature of H$_2$O-rich ices.  We derive a mass of 0.5 \\solarmass for each protostar, and an accretion rate of $5 \\times 10^{-5}$ \\solarmassyr, implying an age of about 10000 years, if the accretion rate remains constant.  We finally discuss the difference between IRAS4 and IRAS16293-2422, where a similar analysis has been carried out. We found that IRAS4 is probably a younger system than IRAS16293-2422.  This fact, coupled with the larger distance of IRAS4 from the Sun, fully explains the apparent difference in the molecular emission of these two sources, which is much richer in IRAS16293-2422. } ",
        "introduction": "The south part of the NGC1333 reflection nebulae, in the Perseus cloud, is an active star forming region, containing many infrared sources associated with molecular flows and numerous Herbig-Haro objects.  IRAS4 was first identified by \\citet{Jennings86}, and further observations \\citep{Sandell91} revealed IRAS4 it was a binary system resolved into two components, named IRAS4A and IRAS4B, and separated by 31$\\arcsec$. Interferometric observations \\citep{Lay95,Looney00} have shown further multiplicity of the two sources.  IRAS4A is itself a binary system with a separation of 10\\arcsec, and there is some evidences that IRAS4B could also be a multiple system, with a separation of 0.5\\arcsec.  The distance of the NGC1333 cloud is much debated. \\citet{Herbig83} found a distance of 350 pc for the Perseus OB2 association \\citep[a more recent estimate based on the Hipparcos data gives 318$\\pm$27;][]{deZeeuw99}, but extinction observations towards NGC1333 itself \\citep{Cernis90} suggest that it may be as close as 220 pc. Assuming a distance of 350 pc, \\citet{Sandell91} measured a system total luminosity of 28 \\solarlum (11 \\solarlum at 220 pc) equally shared between IRAS4A and B.  They derived an envelope mass of 9 and 4 \\solarmass respectively (3.5 and 1.5 \\solarmass at 220 pc). This relatively large mass, together with the low bolometric luminosity suggest that both sources are deeply embedded and probably very young. They have been classified as \\emph{Class 0} sources \\citep{Andre93}. IRAS4A and B are both associated with molecular outflows, detected in CO, CS \\citep{Blake95} and SiO \\citep{Lefloch98} millimeter transitions.  The outflow originating from IRAS4A is very highly collimated, whereas that originating from IRAS4B is rather compact and unresolved in single dish observations \\citep{Knee00}.  The dynamical ages of both outflows are a few thousands years.  In the past years, many observational studies have been focused on the continuum emission of IRAS4.  Recent works include maps of the region obtained with IRAM at 1.3 mm \\citep{Lefloch98} and with SCUBA at 450 and 850 $\\mu$m \\citep{Sandell01}.  An accurate modeling of the continuum emission has been very recently carried out by \\citet[][ hereafter JSD02]{Jorgensen02}, who reconstructed the dust temperature and density profiles across the two envelopes.  The molecular line emission is probably a better and certainly a complementary tool to probe the dynamical, chemical and physical structure of the envelopes of IRAS4.  The last decade has seen flourishing several studies of molecular line profiles \\citep[e.g.][]{Gregersen97, Evans99} and line spectra \\citep{Blake95}, all having in common the goal of reconstructing the physical structure of the protostellar envelopes.  Specifically, \\citet{Blake95} carried out a multifrequency study of several molecules in IRAS4, including H$_2$CO and CH$_3$OH. Their two major results regarding the structure of the IRAS4 envelopes are: 1) a large depletion, around a factor 10-20, of CO and all molecules in the envelope, and 2) the presence of a region with an increased abundance of CS, SiO and CH$_{3}$OH, that the authors attribute to mantles desorption caused by grain-grain collisions induced by the outflows originating from the two protostars.  More recently interferometric observations by \\citet{DiFrancesco01} \\citep[see also][]{Choi99} detected an inverse P-cygni profile of the H$_{2}$CO $3_{2,1}-2_{1,1}$ line on a $2''$ scale towards both IRAS4A and B, providing the least ambiguous evidence of infall motion towards a protostar ever.  From a simple two-layer modeling, they derived an accretion rate of $1.1 \\times 10^{-4}$ and $3.7 \\times 10^{-5}$ \\solarmassyr, an inner mass of 0.7 and 0.2 \\solarmass, and an age of 6500 and 6200 yr (assuming constant accretion rate) for IRAS4A and IRAS4B respectively.  In this paper we concentrate on the far infrared (FIR) line spectrum, and in particular the water line spectrum observed with the {\\it Long Wavelength Spectrometer} \\citep[][ herein after LWS]{Clegg96} on board ISO \\citep{Kessler96} in the direction of IRAS4.  The goal of this study is to check whether the observed water line emission can be attributed to the thermal emission of the envelopes surrounding the IRAS4 protobinary system.  Water lines have in fact been predicted to be a major coolant of the gas in the collapsing envelopes of low-mass protostars (Ceccarelli et al. 1996, hereafter CHT96; Doty \\& Neufeld 1997).  Given the relatively large range of level energies (from $\\sim 100$ to $\\sim 500$ K) and spontaneous emission coefficients (from $10^{-2}$ to $\\sim 1$ s$^{-1}$) of the water transitions observed by ISO-LWS, the observed lines can in particular probe the innermost regions of the envelope.  This makes the analysis of the ISO-LWS water lines a precious and almost unique tool (when considering the water abundance across the envelope).  The reverse of the coin is that assessing the actual origin of the water emission is somewhat difficult and still debated, as the spectral and spatial resolutions of ISO-LWS are relatively poor to disentangle the various components falling into the beam.  For example, strong molecular line emission is often associated with the outflows emanating from young protostars \\citep[e.g.][]{Bachiller97}.  As already mentioned, the line emission from CO, CS and other molecules are certainly contaminated by the outflowing gas in IRAS4.  Nonetheless, low lying lines seem to be more affected than high lying lines in first instance, and different molecules suffer differently from this ``contamination'', as proved by the \\citet{DiFrancesco01} observations.  Although water has been predicted to be very abundant in shocked gas, the published ISO observations show that the water emission is usually stronger towards the central sources and weaker, if detected at all, in the direction of the peaks of the outflows powered by low mass protostars \\citep[see][ for a review]{Ceccarelli00a}. When water lines are detected in clear-cut shocked regions, the water abundance seems to be lower than that predicted by the models, like in the case of HH54 \\citep{Liseau96} or HH7-11 \\citep{Molinari99, Molinari00}, or in the outflows of IRAS4 (see next section). Finally, SWAS observations seem to support the evidence that the water abundance in the shocked regions is a few times $10^{-6}$ \\citep{Neufeld00}. These facts, together with the apparent correlation between the observed water emission and the 1.3 mm continuum, and the lack of correlation with SiO emission\\footnote{SiO emission is usually associated with the outflow strong shocks \\citep[e.g.][]{Bachiller97}, and it is believed to be a product of grain mantle desorption \\citep{Caselli97,Schilke97}} in low mass protostars \\citep{Ceccarelli99} play in favor of a relatively low contamination of the ISO-LWS observed water emission by the outflow and encourage us to explore in detail this hypothesis for the IRAS4.  In the specific case of IRAS4, the {\\it Submillimiter Wavelengths Astronomical Satellite} \\citep[SWAS;][]{Melnick00} observed the ground o-H$_{2}$O line at 557 GHz (Neufeld et al. 2000, Bergin et al. 2002). Given its relatively large linewidth ($\\sim$18 km s$^{-1}$) the 557 GHz line is certainly dominated by the outflow emission.  Nonetheless, this does not imply that the ISO FIR water lines also originate in the outflow, and this for two reasons.  First, the beamwidth ($\\sim 4'$) of the SWAS observations, being about 3 times that of ISO-LWS, encompasses the entire outflow, whereas the ISO observations do not encompass the two emission peaks of the outflow (see also \\ref{origin}), but only the envelope.  Second, the 557 GHz transition, being the water ground transition, is more easily excited than the FIR water lines, and therefore the latter probably probe different regions. In fact, Bergin et al. (2002) find that most of the 557 GHz line must originate in a component colder, hence different, by that probed by the FIR water lines, even under the assumption that they probe the outflow.  To summarize, decicing whether the observed FIR water emission in IRAS4 originates in the outflow or in the envelope remains an open question, based on the available present observations. In this article we explore in detail the latter hypothesis and submit it to the scrutiny of an accurate modeling, trying hence to answer to the question on a theoretical basis.  At this scope we used the CHT96 model, already successfully applied to the solar type protostar IRAS16293-2422 which allowed to explain more than two dozen observed ISO-LWS water lines and ground-based millimeter SiO and H$_2$CO lines \\citep{Ceccarelli00a,Ceccarelli00b}. One of the major results of that work is the prediction of the existence of a hot core like region in the innermost part of the envelope of IRAS16293-2422, in which the dust temperature exceeds the evaporation temperature of interstellar ice ($\\simeq$ 100 K). These studies have been confirmed by the recent analysis by \\citet{Schoier02} of several other molecular transitions. Such hot cores are well studied around massive protostars where -- driven by reactions among the evaporated ice molecules in the warm gas -- their chemical composition differs substantially from that of quiescent clouds \\citep{Walmsley89, Charnley92}.  Hot cores around low mass protostars may actually have a different chemical composition \\citep{Ceccarelli00b}. This molecular complexity may be of prime interest on account of a possible link to the chemical history of the solar nebula and hence the molecular inventory available to the forming Earth and other solar system planets and satellites.  In order to understand the physical and chemical processes that take place during the first stages of star formation, it would be necessary to undertake a work similar to that the one done on IRAS16293-2422 on a larger sample of protostars.  In this paper we present a study of the structure of the envelope of NGC1333-IRAS4, obtained using ISO-LWS observations of the H$_2$O far-infrared lines.  A preliminary analysis of the same set of data has already been presented in \\citet{Ceccarelli99} and \\citet{Caux99b}.  Here we revisit the data using a new calibration and compare the observations with the CHT96 model predictions, testing a large range of model parameters.  This study is part of a large project aimed to model the water emission in several low mass protostars. The water observations are complemented with formaldehyde and methanol ground based observations, to have a complete budget of the most abundant molecules in the innermost regions of the protostellar envelopes (Maret et al. in preparation). Finally, the structure obtained by the analysis of these observations will be compared with that independently obtained by continuum observations by JSD02.  The outline of the article is the following.  In \\S \\ref{observations} we present the data, in \\S \\ref{modeling} we describe the modeling of the observed lines and in \\S \\ref{discussion} we discuss the physical and chemical structure of the envelope, namely the density and temperature profiles, as well as the abundances of the major species across the envelope.  Besides, the central mass of the protostar and its accretion rate can also be constrained by these observations and modeling, yielding an alternative method to measure these two key parameters.  In \\S \\ref{discussion} we compare the results of the present study with previous studies of IRAS4.  Finally, we discuss the similarities and differences between IRAS4 and IRAS16293-2422, and highlight the importance of complementary ground-based, higher spatial and spectral resolution observations to understand the physical and chemical processes taking place in the innermost regions of low-mass envelopes. ",
        "conclusions": ""
    },
    "0209/astro-ph0209321_arXiv.txt": {
        "abstract": "We report the discovery of a new \\HI\\ ring around the S0 galaxy NGC 1533. The ring orbits at a radius of 35 kpc, well outside the optical extent of the galaxy. We have conducted N-body/SPH numerical simulations to show this \\HI\\ ring could be the merger remnant of a tidally destroyed galaxy. We find no optical component associated with the \\HI\\ ring. However, observations hint at \\ha\\ emission associated with the SE part of the ring only. The \\ha\\ is in the form of a few very small isolated emission line regions. The large \\HI\\ velocity dispersions (up to 30 \\kms) and velocity gradients (up to 50 \\kms kpc$^{-1}$) in this region indicate the \\ha\\ emission could be due to star formation triggered by clouds colliding within the ring. ",
        "introduction": "In the local Universe galaxies continue to interact and merge. These mergers provide feedback into the intergalactic medium; either directly or via star formation. S0 galaxies are synonymous with non-regular \\HI\\ distributions - leading to the common perception that this gas was acquired via accretion or mergers. Simple passive evolution of elliptical and S0 galaxies is inconsistent with observations. To explain the observations, merging and star formation in these galaxy types must have occurred from $z\\sim1$ to the present (Kauffmann, Charlot \\& White 1996). In contrast to early-type spirals, S0 galaxies exhibit a wide range of \\mhilb\\ ratios. Wardle \\& Knapp (1986) argue this is evidence for an external origin of \\HI\\ in S0s. This acquisition of \\HI\\ is perhaps more notable in S0s, since their intrinsic \\HI\\ content is low. Similar events in spiral galaxies may not be detected as easily.  The formation of \\HI\\ rings are rare. Rings are known around individual galaxies such the spiral galaxy NGC 628 (Briggs, 1982) and the elliptical galaxy IC 2006 (Schweizer, van Gorkom \\& Seitzer 1989). \\HI\\ rings are also found in galaxy groups enclosing more than one galaxy, for example, the M96 group (Schneider, 1985) and the galaxy group LGG 138 (Barnes, 1999). The Cartwheel galaxy, in the \\HI-rich Cartwheel group, also exhibits an \\HI\\ ring (Higdon 1996).  Explanations for each of these rings invariably involves some form of merging or accretion. The origin of NGC 628's \\HI\\ ring is uncertain, but the absence of a massive companion points toward the acquisition of a gas-rich dwarf galaxy. The \\HI\\ ring around IC 2006 is thought to be the remnant of the merger that created the elliptical or perhaps a later accretion event. Barnes (1999) proposed the \\HI\\ ring in LGG 138 was created by a gas-sweeping collision between one of two bright galaxies and an intruder. Analysis of the ring in the M96 group is complicated by the number of galaxies in the vicinity. The distribution of \\HI\\ in M96 itself suggests it is interacting with the ring and perhaps accreting \\HI\\ onto its own faint optical outer ring. The Cartwheel galaxy is believed to have formed from a small late type spiral with a large low surface density gas disk. Higdon (1996) suggests that another member of the group, G3, passed through this disk and `splashed-out' the \\HI\\ to form the ring.  Optical counterparts to these \\HI\\ rings are rarer still. A very faint dwarf galaxy, Leo dw A resides in the M96 ring (Schneider, 1989). The \\HI\\ in LGG 138 aligns with a colour break in stellar populations in the South-western region of the ring. Barnes (1999) suggests this is due to star formation triggered by an expanding density wave, together with the stellar remnant of the intruder. On the other hand, the Cartwheel galaxy was first noted for its remarkable optical ring and the \\HI\\ observations followed. In this case, the `splashed-out' \\HI\\ is thought to have caused a propagating burst of massive star formation (Higdon, 1996).  When galaxies collide during a merger, they can produce strong shocks, and stars may then form in the cool, compressed gas behind the shock front. For example, large \\HI\\ and CO velocity dispersion and gradients, in the youngest star forming regions in the Antennae galaxies, indicate stars were produced by colliding gas clouds (Zhang, 2001). Alternatively, gravitational instabilities can cause collapse and formation of stars. The balance between self gravity, velocity dispersion and the centrifugal force in a disk, leads to a critical surface density, $\\Sigma_{crit}= \\kappa\\sigma_v/\\pi G$, where $\\kappa$ is the epicycle frequency and $\\sigma_v$ is the velocity dispersion of the gas. According to the Toomre criterion (Toomre, 1964), large scale star formation occurs when $Q$ $(=\\Sigma_{crit}/\\Sigma_{gas}$) is less than one. The \\HI\\ ring in the Cartwheel galaxy satisfies this criterion (Higdon, 1996).  Here we present the \\HI\\ ring surrounding the S0 galaxy NGC 1533. The ring was discovered serendipitously as part of a subset of galaxies from HIPASS (see e.g. Meyer et al., 2002) chosen for mapping with the Australia Telescope Compact Array (ATCA)\\footnote{The Australia Telescope Compact Array is part of the Australia Telescope which is funded by the Commonwealth of Australia for operation as a National Facility managed by CSIRO.}. NGC 1533 is located 1$^{\\circ}$ from the centre of the Dorado group. Throughout this paper we assume a distance to NGC 1533 of 21$\\pm$4 Mpc (Tonry et al., 2001). ",
        "conclusions": ""
    },
    "0209/astro-ph0209117_arXiv.txt": {
        "abstract": "We explore the constant intensity cut method that is widely used for the derivation of the cosmic ray energy spectrum, for comparisons of data obtained at different atmospheric depths, for measuring average shower profiles, and for estimates of the proton-air cross section from extensive air shower data. The constant intensity cut method is based on the selection of air showers by charged particle or muon size and therefore is subject to intrinsic shower fluctuations. We demonstrate that, depending on the selection method, shower fluctuations can strongly influence the characteristics of the selected showers. Furthermore, a mixture of different primaries in the cosmic ray flux complicates the interpretation of measurements based on the method of constant intensity cuts. As an example we consider data published by the Akeno Collaboration. The interpretation of the Akeno measurements suggests that more than $60-70\\%$ of cosmic ray primaries in the energy range $10^{16}-10^{17}$ eV are heavy nuclei. Our conclusions depend only weakly on the hadronic interaction model chosen to perform the simulations, namely SIBYLL and QGSjet. ",
        "introduction": "Introduction}  Measuring extensive air showers (EAS) is currently the only way to study the cosmic ray spectrum and chemical composition at energies above 10$^{14}$ eV, as well as the basic properties of hadronic interactions at $\\sqrt{s}$ above 1.8 TeV.  EAS can be detected with air shower arrays which measure densities of shower particles such as electrons, muons, photons, and sometimes hadrons arriving at the detector. These densities are typically fit to lateral distribution functions to derive the total number of charged particles, electrons $N_e$ and muons $N_\\mu$ at detector level. The particle numbers are functions of the primary cosmic ray energy $E$ and the mass number $A$ of the primary particle, and depend on the atmospheric depth of the observation level. At energies $E$ \\gapproxeq $10^{17}$ eV  the shower evolution can also be directly observed by measuring the fluorescence light from the atmospheric nitrogen that is excited by the ionization of the charged shower particles. In the following we will concentrate on air shower arrays. Imaging methods such as fluorescence or Cherenkov light techniques will be discussed elsewhere \\cite{Alvarez-Muniz02c}.  One of the classical methods in the analysis of air shower data is the constant intensity cut method. The idea is based on the fact that, due to the isotropy of the primary cosmic ray flux, showers generated by primary particles of the same energy and composition will arrive at the detector with the same frequency, assuming 100\\% detection efficiency. Selecting showers arriving at the detector with the same frequency under different zenith angles allows the measurement of the mean longitudinal shower profile. At large atmospheric depths after shower maximum, the shower size decreases approximately exponentially with depth with a length scale commonly referred to as the {\\it attenuation} length.  On the other hand, selecting showers with the same features (i.e. shower size, muon size etc.) at observation  level and different incident angles allows the measurement of the {\\it absorption} length which determines how the flux of the selected showers decreases with atmospheric depth.   Measurements of the attenuation length are commonly used to correct observed particle densities to those of equivalent vertical showers. By unfolding the geometry-related attenuation of showers an experiment can use the measured intensities of showers with fixed size to derive the primary all-particle flux.  The absorption length is inherently related to the mean free path of the EAS initiating primary particle. For example, the rate of proton air showers having the first interaction point ($X_{\\rm int})$ at a slant depth greater than $X$ decreases as $\\exp(-X/\\lambda_{\\rm int})$, where $\\lambda_{\\rm int}$ is the mean free path for $p-{\\rm air}$ collisions. On this basis, several methods of extracting the $p-{\\rm air}$ cross section\\footnote{The measured mean free path corresponds to the particle production cross section. For a discussion of the difference between the inelastic and production cross sections in relation to air showers see \\protect\\cite{Nikolaev93b,Engel98c}.} from measurements of EAS have been applied in air shower experiments \\cite{Hara83,Baltrusaitis84,Honda93,Aglietta97,Aglietta99,Aglietta99b}.  Air shower arrays cannot measure the depth of the first interaction of the primary particle generating the observed shower, which directly relates to the mean free path. The decrease with zenith angle of the frequency of showers having the same electron $N_e$ and muon $N_\\mu$ sizes at observation level is studied instead. In the absence of intrinsic shower fluctuations these measurements would reflect the depth distribution of primary interactions. However, the longitudinal development of showers is itself subject to large fluctuations. To disentangle these fluctuations from those of the first interaction point is not an easy task.  This problem is usually addressed by introducing a coefficient ($k$) which relates the observed shower absorption length ($\\Lambda_{\\rm obs}$) and the inelastic cross section through the equation $\\Lambda_{\\rm int}= k\\times \\lambda_{\\rm obs}$~\\cite{Elsworthetal82}. The numerical value of $k$ has to be obtained from simulations of EAS. The coefficient $k$ reflects the influence of the features of the hadronic interactions model on the fluctuations in the shower development. The value of $k$ depends on the cross sections, secondary particle multiplicity and elasticity in the hadronic interaction model. Due to the necessary extrapolation of hadronic multiparticle production to unmeasured regions of the phase space and to high energy, the extracted cross section becomes model dependent.  Further difficulties in determining the inelastic p-air cross section from EAS measurements are related to experimental uncertainties and limitations in the determination of the development of  air showers, and also to the fact that the cosmic ray flux might be ``contaminated'' with primaries heavier than protons which in principle tend to decrease the observed mean free path.   In this article we shall study the importance of intrinsic shower fluctuations for the experimentally observed attenuation and absorption lengths by considering two examples:\\\\ (i) the reconstruction of the primary cosmic ray spectrum using charged particle shower sizes, and\\\\ (ii) the measurement of the proton-air cross section, following closely the method applied first by the Akeno group~\\cite{Hara83,Honda93} which we call the constant $N_\\mu, N_e$ method.  In the process of (ii) we found that the Akeno observations can best be understood if there is a large fraction of heavy nuclei present in the cosmic ray beam from 10-100 PeV.   The paper is structured as follows. In Section~\\ref{cuts} we study the possible errors introduced in the derivation of the primary cosmic ray spectrum by shower fluctuations when the constant intensity method is applied. Section~\\ref{protons} consists of three parts. Part A summarizes the basics of the constant $N_e-N_\\mu$ method. In part B we describe the predictions of this method for proton induced showers and in part C we discuss the more realistic situation of a mixed primary cosmic ray composition. Section~\\ref{conclusions} concludes the paper. ",
        "conclusions": "Discussion and conclusions}  We have investigated the influence of air shower fluctuations on the widely used constant intensity cut method. We consider two types of applications: the classic integral approach to the derivation of the cosmic ray energy spectrum, and the differential $N_\\mu, N_e$ cut used for the derivation of the proton-air cross section.  We find that the constant intensity cuts method can work for comparisons of data taken at different atmospheric depths and different angles. This is however possible only when the chemical composition of the primary cosmic rays is well known. The use of incorrect chemical composition can lead to a shift in the normalization of the energy spectrum. In the case of energy dependent composition the normalization errors for different cosmic ray flux components could also affect significantly the derived spectral index $\\gamma$. Such shifts are also possible close to the detector $N_e$ threshold, where measurements at different zenith angles would detect showers of different composition. The larger the zenith angle $\\theta$, the lighter would be the composition of the detected showers.    The influence of shower fluctuations is much bigger when the constant intensity cut is used in a differential way to compare showers with same electron and muon sizes detected at different zenith angles.  The selection by the muon size $N_\\mu$ with constant intensity cuts is indeed a very good method and leads to a good angle independent energy selection. This is the result of the much slower absorbtion of the shower muons as well as to the smaller $N_\\mu$ fluctuations in showers with fixed primary energy.   The constant $N_e-N_\\mu$ method, which is used for derivation of the proton-air production cross section, is dominated by fluctuations even in the case of a pure proton composition. The accuracy of this method improves with the selection of showers with large $N_e$ for a fixed $N_\\mu$ bin, where an experiment would run out of statistics.  A possible improvement of the method would be to use Monte Carlo shower simulations to determine zenith angle dependent $N_e$ bins. This, however, would represent a new method which is very different from the original idea of constant intensity cuts.  The Akeno data that were used for the derivation of the proton-air production cross section can be interpreted in terms of cosmic ray composition. The angle independent exponential slope of the shower absorption length indicates a substantial fraction of heavy primaries in the energy range of 10$^{16}$ - 10$^{17}$ eV. The comparison to showers simulated with the QGSjet98 and SIBYLL 2.1 hadronic interactions models require 60-70\\% of iron in the primary cosmic ray flux to explain the absorption length derived by the Akeno group.  These conclusions depend only mildly on the hadronic interaction model used in the simulation.   \\noindent {\\bf Acknowledgments} We thank A.A. Watson for helpful discussions. J.A. Ortiz is supported by CAPES ``Bolsista da CAPES - Bras\\'{\\i}lia/Brasil'' and acknowledges Bartol Research Institute for its hospitality. This research is supported in part by NASA Grant NAG5-10919. RE, TKG \\& TS are also supported by the US Department of Energy contract DE-FG02 91ER 40626. The simulations presented here were performed on Beowulf clusters funded by NSF grant ATM-9977692."
    },
    "0209/astro-ph0209267_arXiv.txt": {
        "abstract": "We present CO(1-0) and CO(2-1)  maps of the Seyfert galaxy NGC 5728. Although a stellar nuclear bar structure is clearly identified in the near-infrared images in the central 10\", inside the ring identified as the ILR of the primary bar, there is no nuclear bar structure in the molecular gas. Instead, the CO emission reveals an elongated structure, 15\" in length, beginning at the nucleus (defined by the radio center) aligned with the jet/ionisation cone, at a PA of 127 degrees. This morphology, not frequently observed in Seyfert galaxies, may be interpreted in terms of enhanced CO excitation along the walls of the cone. Kinematical perturbations along the cone support this scenario. At larger-scale, CO emission is tracing the primary bar, and outer ring structure. The total molecular mass, estimated from the CO emission, is M(H$_2$) = 3.1 10$^9$ M$_\\odot$. ",
        "introduction": "NGC 5728 is a well known prototype of embedded bars. It is classified as (R1)SAB(r)a, and a Seyfert 2 (distance 33 Mpc -- 1''=180 pc). Its nuclear bar is oriented at 45 degrees of the main bar (Shaw et al 1993), and could rotate faster.  Petitpas \\& Wilson (2002) have obtained a CO map of NGC 5728 with the OVRO interferometer. They found a total CO mass of 3.1 10$^8$ M$_\\odot$. However, they underestimate the extended component, filtered out by the interferometer, and miss a significant part of the CO emission, due to restricted velocity coverage. We report here CO(1-0) and CO(2-1) observations with the IRAM-30m telescope, with 23 and 12\" beams.  \\begin{figure}[h] \\centering \\includegraphics[width=11.5cm]{combes2_fig1.ps} \\caption{Contours of the CO(1-0) emission, superposed on the 2-mass NIR image ({\\it left}), and at the same scale an adaptive optics NIR image (obtained with CFHT, Combes et al in prep) showing the stellar nuclear bar ({\\it right}).} \\label{fig1} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209098_arXiv.txt": {
        "abstract": "s{The \\emph{XMM--Newton} data presented here are the first X--ray observation of the Seyfert 2 galaxy NGC4138. Although the galaxy has been pointed by ROSAT--HRI, it was not detected, with a flux upper limit of about $1.1\\times10^{-13}$~erg~cm$^{-2}$~s$^{-1}$ in the $0.2-2.4$~keV energy band~\\cite{HAL}. \\emph{XMM--Newton} performed the observation on 26 November 2001. The source is detected for the first time in X--rays with $F_{0.2-2.4}=1.0\\times 10^{-13}$~erg~cm$^{-2}$~s$^{-1}$, in agreement with the upper limit of ROSAT~\\cite{HAL}. The source spectrum is typical of Seyfert 2 galaxies. We find heavy obscuration ($N_H\\approx 8\\times 10^{22}$~cm$^{-2}$) and a flat photon index ($\\approx 1.6$). The source intrinsic luminosity is about $5\\times 10^{41}$~erg/s in the $0.5-10$~keV energy band.} ",
        "introduction": "NGC4138 is a spiral galaxy of Hubble type SA(r)0+ distant 17~Mpc, with a $D_{25}=2.57'$. Its morphology appears undisturbed, although the galaxy has a dust lane in the southeastern side. The nucleus is classified as Seyfert 1.9~\\cite{HO}. Optical observations~\\cite{JO} have shown evidence for a counterrotating disk, with a velocity dispersion systematically lower than that of the primary disk. The counterrotating disk may be either the result of a recent merger (it appears to be still forming) or the continuum infall of material with opposite spin vector into NGC4138. ROSAT--HRI observed the galaxy for $5.8$~ks without detecting it. Halderson et al.~\\cite{HAL} gave an upper limit to the $0.1-2.4$~keV flux of $1.15\\times 10^{-13}$~erg~cm$^{-2}$~s$^{-1}$. \\emph{XMM--Newton} EPIC instrument detected the source at a flux level of about $1.0-1.1\\times 10^{-13}$~erg~cm$^{-2}$~s$^{-1}$ in the same band.  \\begin{figure}[h] \\begin{center} \\epsfig{figure=foschini_p_fig1.ps, width=10cm} \\end{center} \\caption{Image from the Digitized Sky Survey with EPIC--MOS2 contours superimposed (energy band $0.5-10$~keV). North is up and East is left. \\label{images}} \\end{figure}  \\begin{figure}[h] \\begin{center} \\epsfig{figure=foschini_p_fig2.eps, width=10cm} \\end{center} \\caption{Radial profile of NGC4138. Solid line: PSF at 5 keV; Points: source counts in the energy range $0.5-10$~keV from EPIC--MOS2.\\label{radial}} \\end{figure}  \\begin{figure} \\begin{center} \\epsfig{figure=foschini_p_fig3.eps, angle=270, width=10cm} \\end{center} \\caption{Best fit model for NGC4138: it consists of a power law plus scattered component, and two gaussian emission lines at about $0.8$~keV and $6.4$~keV. See text for more details. \\label{spec}} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209051_arXiv.txt": {
        "abstract": "The correlation between a galaxy's morphology and its observed optical spectrum is investigated. As an example, 4000 galaxies from the 2dF Galaxy Redshift Survey, which possess both good quality spectra and have visually determined morphologies are analysed. Of particular use is the separation of Early and Late type galaxies present in a redshift survey since these can then be used in their respective redshift-independent distance estimators ($D_n-\\sigma$ and Tully-Fisher). It is determined that galaxies in this sample can be relatively successfully separated into these two types by the use of various statistical methods.  These methods are briefly outlined in this paper and are also compared to the default 2dFGRS spectral classification $\\eta$. In addition it is found that the 4000\\AA\\ break in the spectrum is the best discriminant in determining its morphological type. ",
        "introduction": "\\label{section:intro}  The classification of galaxies according to their observed morphologies has proved to be a very useful way of characterising different galaxy populations (see e.g. Hubble 1936).  However, the high-resolution imaging data required to make such accurate morphological classifications, over a wide range of redshifts, is often unavailable in large galaxy redshift surveys. In this paper, a selection of different statistical methods are investigated, with the aim of establishing a quantifiable link between a galaxy's morphology and its observed optical spectrum; thousands of which are now available through the advent of large galaxy redshift surveys such as the Sloan Digital Sky Survey (York \\etal\\ 2000) and the 2dF Galaxy Redshift Survey (2dFGRS, Colless et al. 2001). In particular, a set of 4000 galaxies which have already been observed in the 2dFGRS, for which the morphology and spectrum have been determined, is used as a training set. It has long been established that a substantial link exists between the overall structure of a galaxy and the chemical properties reflected in its spectrum - which quantifies its stellar and gas composition (e.g. Morgan \\& Mayall 1957). However, efforts to quantify this relationship have been somewhat hampered due to the lack of large, representative, data-sets.  For example, Folkes, Lahav \\& Maddox (1996) used a sample of only 26 unique galaxy spectra and morphologies in a similar analysis to that presented here. This is fortunately no longer an issue with the advent of fibre-based galaxy redshift surveys, which are able to acquire several hundred galaxy spectra per hour.  These surveys are producing extremely large data-sets for which the spectrum of a galaxy is known but its detailed structural parameters generally are not, or are very difficult to determine accurately from the photometry of the input catalogue over a representative redshift range. Fortunately, the spectrum of a galaxy is generally a more robust quantity to measure over a variety of redshifts and as such if a substantive link between optical spectra and these parameters can be determined this will greatly enhance our ability to probe the properties of our local galaxy population.  Another more specific advantage of being able to determine a galaxy's morphology from its spectrum is that it allows the identification of targets for peculiar velocity follow-ups (using either $D_n-\\sigma$ for elliptical galaxies or the Tully-Fisher relation for spirals).  In this paper two statistical techniques are investigated, to determine how accurately morphology can be estimated from a galaxy's optical spectrum.  Both methods are `supervised' - meaning that they require a training set of galaxies with both visually determined morphologies and observed spectra. The first method to be implemented is Fisher's linear discriminant (Section~4), which attempts to determine an optimal linear combination of inputs (the spectrum) to distinguish between several outputs (the morphologies).  The second method is an Artificial Neural Network (ANN, Section~5) which creates non-linear combinations of the input, and outputs a selection of class probabilities.  The possible biases introduced into this work by systematic effects are described in Section~6 and then the success rates of each method are compared to those achieved using the default 2dFGRS spectral classification parameter, $\\eta$, in Section~7. ",
        "conclusions": "\\label{section:conclusion}  Establishing a firm link between a galaxy's morphology and its spectrum is advantageous for several reasons.  For instance, galaxy spectra can be accurately determined to much greater redshifts and for fainter objects than morphologies.  Also, most large redshift surveys currently taking place will contain many thousands of galaxy spectra but little information relating to the optical morphologies of those galaxies.  In particular the separation of different morphological types of galaxies in these redshift surveys will be very useful as a means of separating objects for follow-up observations to determine independent distance measurements using either $D_n-\\sigma$ or the Tully-Fisher relation.  In this paper I have tried to quantify the link between galaxy spectra and morphology using several advanced statistical methods; namely, Fisher's linear discriminant and Artificial Neural Networks. The best results produced suggest that it is possible to use optical galaxy spectra to create galaxy samples containing 70\\% of the Early type galaxies present and 80\\% of the Late types respectively. The contamination between these samples depends on the morphological mix of the survey under consideration.  In the case of the $\\bj$-selected 2dFGRS the most significant contamination will be of mis-classified Late types in the Early type sample ($\\sim40\\%$ contamination), in the case of a near-infrared selected sample this situation will be reversed.  Essentially the results obtained using more advanced statistical techniques (Sections 4 and 5) are comparable to those that could be obtained simply using the default 2dFGRS spectral classification $\\eta$ (Madgwick \\etal\\ 2002) which can be accessed from the 2dFGRS database\\footnote{ Spectral extension parameter {\\tt ETA\\_TYPE}}.  This is an interesting result and certainly adds significantly to the physical interpretation of this parameter.  Another interesting aspect of this analysis is that the Fisher discriminant (Section 4) identified the 4000\\AA\\ break to be the most essential element of a galaxy's spectrum for the purposes of estimating its morphology.  This result is somewhat expected since the general correlation between galaxy morphology and colour is already well established. However, it is intriguing to see this result derived in a quantitative manner from the observed spectra themselves.  The results presented in this paper are essentially limited by the coarseness of the morphological classification adopted, which for practical reasons can only be divided into two separate types (rather than a more realistic sequence of types).  As larger samples of more accurately morphologically classified galaxies become available it will be interesting to repeat the analysis presented here, in order to determine whether even more information can be recovered to link a galaxy's morphology and spectrum."
    },
    "0209/astro-ph0209501_arXiv.txt": {
        "abstract": "{  Many binary systems are in a state of strong interaction, where mass exchanges, accretion disks, common envelopes provide circumstellar matter which can have significant effects in microlensing light curves of background sources. Such chromatic absorption effects provide very interesting information on the nature of this diffuse matter and the physics of the interaction between the two stars. ",
        "introduction": "It is well known that about one half of the stars populating our Galaxy are part of gravitationally bound multiple systems. On the other hand, binary systems acting as microlenses produce drastic modifications into microlensing light curves (Mao \\& Paczynski 1991). At least $10\\%$ of all microlensing events clearly show the typical signatures of binary lenses, including the presence of sharp caustic crossing peaks.  A conspicuous fraction of binary systems is in a state of tight interaction, where several physical phenomena show up and accompany the basic gravitational rotation. Tidal interactions, mass exchanges, the appearance of accretion disks and circumstellar matter around compact objects in their various shapes widely enrich the phenomenology of these systems. For a detailed review on strongly interacting binary systems (hereafter SIBS), we refer the reader to (Carroll \\& Ostlie 1996), where a full chapter is devoted to the subject.  In the present paper, we want to point out at all the possible signatures of SIBS in microlensing observations, with particular reference to chromatic absorption by circumstellar matter in microlensing light curves. These effects can be appreciated in careful observations and could provide an alternative tool to detect SIBS and determine their characteristics in a qualitative and quantitative way.  In the binary lens equation, two scales are present: the total Einstein radius $R_\\mathrm{E}$, proportional to the square root of the total mass, and the separation between the binaries. In practice, the system will best show up its binary nature through microlensing when these two scales are of the same order. Of course, not all close binary systems will reveal their nature in a microlensing event. The separation between the two stars cannot be lower than $0.1 R_\\mathrm{E}$ (Gaudi \\& Gould 1997). If the two stars are too close, they will effectively behave as a single lens and their binary nature would be hidden to microlensing. Typical Einstein radii in galactic observations are of the order of AU, which are compatible with most classes of interacting binary systems, characterized by separations of the order of tenths of AU. However, if we also take into account astrometric observations of the center of light deformations during microlensing, the efficiency in the detection of close binaries may be decisively increased, so that even binaries with separations of the order $0.01 R_\\mathrm{E}$ can be detected (Chang \\& Han 1999).  The rotation periods of interacting binaries may range from few hours to several months. As regards those binaries which can yield distinctive features of binary microlensing (namely those which have separations larger than 0.1, 0.2 AU), for solar mass stars, the typical periods are of the order of some tens of days. This means that the rotation of the lens should be often included in a detailed study of the microlensing event. A comprehensive study of this problem has been already performed by Dominik (1998) and will not be discussed here. ",
        "conclusions": "The presence of diffuse matter within SIBS shows up through absorption of one or more microlensing images. If the diffuse matter is localized around one star, we expect a partial depression of caustic-crossing or secondary peaks, depending on the geometry of the event. If the matter englobes the whole system, then all microlensing images are simultaneously affected and a general suppression of the curve occurs, which may significantly alter the original microlensing curve. Since the modifications are more evident in blue than in red, a careful analysis in different color bands is mandatory to unambiguously detect and study these events in a serious way. Moreover, if it is possible to take a spectrum during the microlensing event, the appearance of characteristic absorption lines will be the clear signature of the presence of the diffuse medium. In this case, we would also have a way to analyze the chemical nature of this diffuse matter.  Microlensing may provide a very powerful tool for a deep physical investigation of SIBS. It may help to estimate the distribution of diffuse matter in these systems, also retrieving information on their density profiles. Eventual turbulent density fluctuations would be apparent as random noise on the light curve if their length scales are comparable to the extension of the absorbed images. In this case, a statistical study of this noise would provide information on these fluctuations. Moreover, microlensing comes with its standard information on the mass ratio and the separation of the stars along with other parameters which can be useful to give a complete characterization of the system.  An important observation is that interacting binaries are detected by traditional optical observations only when their orbital plane is edge-on with respect to the observer (unless they are in the solar neighbourhood, so that the rotation can be detected by astrometric measurements, or the two components are separated interferometrically). The binary nature of the object is then revealed by the mutual eclipses produced by rotation, or by spectroscopical measurements of radial velocities. If the orbital plane is disposed face-on with respect to the observer, it is necessary to resort to indirect proofs (e.g. X-ray emission from accretion disk, cataclysmic variables). In microlensing surveys, instead, both situations are practically detectable with the same chances. It is thus possible to enlarge the sample of known binary systems and study them from a different perspective."
    },
    "0209/astro-ph0209392_arXiv.txt": {
        "abstract": "{  We present results from a {\\it Chandra X-ray Observatory} study of the field X-ray source population in the vicinity of the radio galaxy MRC 1138-262.  Many serendipitous X-ray sources are detected in an area of 8$'\\times$8$'$ around the radio source and 90\\% are identified in our deep VLT images.  The space density of such sources is higher than expected on the basis of the statistics of {\\it ROSAT} and {\\it Chandra} deep surveys.  The most likely explanation is in terms of a concentration of AGN associated with the protocluster at $z = 2.16$ which was found around the radio galaxy in previous studies. Two sources have a confirmed spectroscopic redshift close to that of the radio galaxy, and for three more sources other observations suggest that they are associated with the protocluster. Four of these five X-ray sources form, together with the radio galaxy, a filament in the plane of the sky. The direction of the filament is similar to that of the radio source axis, the large scale distribution of the other protocluster members, the 150 kpc-sized emission-line halo and the extended X-ray emission associated with the radio galaxy.  The majority of optically identified X-ray sources in this field have properties consistent with type I AGN, a few could be soft, low luminosity galaxies, one is probably an obscured (type II) AGN and one is a star.  These statistics are consistent with the results of deep X-ray surveys. ",
        "introduction": "Since its launch, the {\\it Chandra X-ray Observatory} has been providing the deepest and sharpest images of the X-ray sky ever: one of its most remarkable characteristics is its high spatial resolution ($<$ 1$''$) and astrometric precision ($\\sim 1''$), well matched to typical optical and near infrared (NIR) imaging resolutions. This allows unambiguous identifications of faint X-ray sources and the possibility to study the morphologies of the X-ray emitting host galaxies.  We have recently observed the radio galaxy MRC 1138-262, and the surrounding field, with {\\it Chandra} to study the extended X-ray emission associated with this radio source (Carilli et al.\\ 2002), which was previously detected with {\\it ROSAT}.  This radio galaxy at redshift 2.16 has a quite complex optical and NIR morphology, resembling a massive galaxy in the early stages of formation and it is embedded in a giant Ly$\\alpha$ halo (Pentericci \\etal\\ 1997).  Deep VLT observations have shown that MRC 1138-262 resides at the center of a protocluster consisting of at least 20 confirmed cluster members (Pentericci \\etal\\ 2000, Kurk \\etal\\ 2002b).  {\\it Chandra}'s great sensitivity allows the detection of many X-ray emitters in the field besides the main target, even in observations with modest exposure times (a few ten thousand seconds in our case).  In this paper we report on these serendipitous X-ray sources: we find an excess of soft X-ray sources in the field, as compared to the predictions for a non-cluster field based on the $\\log N - \\log S$ relation derived from deep {\\it ROSAT} and {\\it Chandra} measurements.  We present optical identifications of the X-ray sources, discuss their nature and their possible relation to the protocluster structure at redshift 2.16.  Throughout the paper we assume a flat, $\\Lambda$-dominated universe with $H_{0} = 65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_M = 0.3$ and $\\Omega_{\\Lambda} = 0.7$. ",
        "conclusions": "We have reported on the serendipitous X-ray emitters detected by {\\it Chandra} in the field around the radio galaxy MRC 1138-262. We have presented optical identifications for 15 of the 18 sources from deep VLT observations.  Based on the X-ray and optical properties of the sources, we conclude that most are type I AGN, one is a type II AGN and one is probably a star.  Compared to the $\\log N - \\log S$ relation measured in deep {\\it Chandra} observations, we find an excess of soft X-ray sources in the field of MRC 1138-262.  The most probable explanation is that several of the X-ray emitters are associated with the $z \\sim 2.16$ protocluster around the radio galaxy. Indeed, two of these have been confirmed spectroscopically to be broad emission line AGN at a redshift similar to that of the central galaxy.  Additional spectroscopy is needed to determine how many of the X-ray emitters are really associated with the $z \\sim 2.16$ protocluster.  Furthermore, future observations of AGN in other distant cluster of galaxies will help to clarify whether the concentration of AGN around MRC 1138-262 is unusual or just typical for such a structure at high redshift."
    },
    "0209/physics0209016_arXiv.txt": {
        "abstract": "We discuss the experimental consequences of hypothetical time variations of the fundamental constants. We emphasize that from a purely phenomenological point of view, only dimensionless fundamental constants have significance. Two classes of experiments are identified that give results that are essentially independent of the values of all constants. Finally, we show that experiments that are generally interpreted in terms of time variations of the dimensioned gravitional  constant $G$ are better interpreted as giving limits on the variation of the dimensionless constant $\\alpha_G=Gm_p^2/\\hbar c$. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209447_arXiv.txt": {
        "abstract": "Routine observations of the Crab Nebula for a total of about 250~hrs, performed with the HEGRA stereoscopic system of 5 imaging atmospheric \\v{C}erenkov telescopes in the standard o\\-pe\\-ra\\-tio\\-nal mode, have proven the energy threshold of the system to be 500~GeV for small zenith angles ($\\theta \\leq 20^\\circ$). A {\\it topological trigger} applied along with the {\\it convergent} observational mode allows to reduce noticeably the energy threshold of the system down to 350~GeV. Here we present the relevant Monte Carlo si\\-mu\\-la\\-tions as well as the analysis results of 15~hrs Crab Nebula data taken in such an observational mode. From the Crab Ne\\-bu\\-la data, the final energy threshold was found to be 350~GeV. The estimated $\\gamma$-ray flux from the Crab Nebula above 350~GeV is $(8.1\\pm 0.1_{stat}\\pm0.2_{syst})\\times 10^{-11}\\,\\, \\rm ph\\,cm^{-2}\\,s^{-1}$, which is consistent with recent measurements reported by the STACEE, CELESTE, CAT, and Whipple groups.  \\vspace{3mm}  \\noindent {\\it PACS:} 95.55.Ka; 95.55.Vj; 96.40.Pq  \\noindent {\\it Keywords:} Imaging Atmospheric \\v{C}erenkov Technique; Very High Energy Gamma Ray Astronomy; ",
        "introduction": "Ground-based TeV $\\gamma$-ray astronomy has effectively ex\\-ploi\\-ted i\\-ma\\-ging at\\-mo\\-sphe\\-ric \\v{C}e\\-ren\\-kov te\\-le\\-sco\\-pes (IACT) in the study of $\\gamma$-ray emission arising from a few well established sources, as well as for a general search amongst large numbers of $\\gamma$-ray source candidates \\cite{cw99}. Given the rather steep energy spectra measured for dis\\-co\\-ver\\-ed $\\gamma$-ray sources as well as for those pre\\-dic\\-ted as potential can\\-di\\-da\\-tes, a substantial advancement in the sen\\-si\\-ti\\-vi\\-ty of IACT techniques can be reached if con\\-si\\-de\\-ra\\-ble lowering of the energy threshold\\footnote{The position of a peak in the differential $\\gamma$-ray detection rate, $E$ [TeV], is usually assumed as the effective energy threshold of the instrument.} of the instrument is achieved. For such spectra the low energy threshold provides a high $\\gamma$-ray rate, which gives a high sensitivity of the instrument. Nowadays a number of instruments perform observations at the effective energy threshold as low as 250~GeV \\cite{cw99}.  Further reduction of the e\\-ner\\-gy thre\\-shold could be a\\-chie\\-ved by using of {\\it (i)} larger optical reflectors (10-20~m in dia\\-me\\-ter) to increase the photon collection efficiency for the low energy $\\gamma$-ray showers; {\\it (ii)} imaging cameras equipped with pixels of a relatively small angular size ($0.1^\\circ - 0.15^\\circ$), to reduce the background night sky light per pixel; and finally, {\\it (iii)} by operating a number of telescopes simultaneously in the {\\it stereoscopic} observational mode, which allows to reduce the trigger threshold by requiring coincidences in a number of telescopes.  The latter approach was proven by the system of 5 imaging atmospheric \\v{C}erenkov telescopes built by the HEGRA ({\\it High Energy Gamma-Ray Astronomy}) collaboration at La Palma, Canary Islands. Even for rather small optical reflectors of 8.5$\\rm \\,\\,m^2$ for each of the telescopes and relatively modest angular size of the ca\\-me\\-ra pixels of 0.25$^\\circ$ the energy threshold of the system is about 500~GeV using the nominal observational mode. The corresponding effective dynamic energy range of the $\\gamma$-ray observations with the HEGRA system of IACTs extends from 500~GeV up to 20~TeV, as shown by the observations of the well established TeV $\\gamma$-ray source~- the Crab Nebula \\cite{me00}. The extended HEGRA data sample for the Crab Nebula permitted, thanks to the good energy resolution of the HEGRA system of about 10-20\\%, measurements of the Crab Nebula spectrum with high accuracy over the entire dynamic energy range.  Lowering the e\\-ner\\-gy thre\\-shold of the \\v{C}e\\-ren\\-kov te\\-le\\-sco\\-pes is not only motivated by the enhancement of the instrument sensitivity but it is also physically important to extend the spectral studies of the $\\gamma$-ray emission to lower energies. For example, inverse Compton (IC) modeling of the $\\gamma$-ray spectrum of the Crab Nebula, taken along with the EGRET detection at GeV energies \\cite{aa96,dj96}, predicts a substantial spectral flattening down to energies of 200-300~GeV, whereas above 1~TeV the spectrum has rather a power-law shape of a spectral index of -2.6. The measurement of such change in the spectrum slope would favor the IC model of $\\gamma$-ray emission. First evidence for deviation from a straightforward power law was suggested by the WHIPPLE group (see~\\cite{hillas2,krenn}) and, subsequently, it was confirmed by other measurements made within that energy range by STACEE \\cite{o01}, CELESTE\\cite{n01}, and CAT \\cite{pir01}.  In this paper we discuss a new approach to detect low energy $\\gamma$-rays with the system of HEGRA IACTs by use of a {\\it topological} system trigger along with the {\\it convergent} observational mode. It allows to reduce considerably the trigger threshold and cor\\-re\\-spon\\-din\\-gly the e\\-ner\\-gy thre\\-shold down to 300-350~GeV. Relevant Monte Carlo simulations (see Section~III) and trigger tests (see Section~IV) have been done in order to prove the efficiency of such observational technique. Finally, we performed 15~hrs observations of the Crab Nebula using the topological trigger technique with the convergent observational mode. Analysis of these data has shown an excess attributed to $\\gamma$-ray showers at energies well below 500~GeV (Section~V-VI). Assuming the shape of the energy spectrum as measured by HEGRA \\cite{me00} we estimated the integral flux of $\\gamma$-rays from the Crab~Nebula above 350~GeV (Section~VII). ",
        "conclusions": "We performed observations of the Crab Nebula with the HEGRA system of IACTs using the {\\em topological trigger mode}, and demonstrated for the first time that it is possible to lower the actual energy threshold of the system by a factor of 1.4 without major hardware changes and, at the same time, keeping the event rate at a sustainable level for the currently used DAQ. The Crab Nebula data were taken in fall of 2000 for a total observational time of 15~hrs to check the performance of the system in such observational mode. Here we present the result of the data analysis and give an estimate of the integral $\\gamma$-ray flux from the Crab Ne\\-bu\\-la above 350~GeV. Our estimate of the Crab Nebula flux is consistent with previous measurements made by STACEE, CELESTE, and CAT groups, and may be further interpreted as an Inverse Compton TeV emission coming from the plerion in the Crab Nebula.  This technique will be applied in the near future in the observations with the forthcoming H.E.S.S. ({\\it High Energy Stereoscopic System}) system of telescopes, in par\\-ti\\-cu\\-lar to search for sub TeV $\\gamma$-ray emission from pulsars~\\cite{dejager2}. By means of a topological trigger one can achieve a significant reduction of the energy threshold and that gives a considerable advantage in such observations, due to the very steep energy spectrum of the GeV-TeV pulsed emission as measured by the EGRET detector in the GeV energy range from a number of such objects. It might be also valuable to apply such technique in search for {\\it BL Lac objects at rather large red shifts}, which are expected to have a very steep energy spectrum due to the IR absorption of the $\\gamma$-rays onto the extragalactic background light.  {\\bf Acknowledgments.} The support of the HEGRA ex\\-pe\\-ri\\-ment by the German Mi\\-nis\\-try for Re\\-search and Tech\\-no\\-lo\\-gy BMBF and by the Spanish Research Council CICYT is acknowledged. We are grateful to the Instituto de Astrof\\'{\\i}sica de Canarias for the use of the site and for providing excellent working conditions.  \\newpage"
    },
    "0209/astro-ph0209179_arXiv.txt": {
        "abstract": "{ The detection of planetary transits in stellar photometric light-curves is poised to become the main method for finding substantial numbers of terrestrial planets. The French-European mission COROT (foreseen for launch in 2005) will perform the first search on a limited number of stars, and larger missions \\emph{Eddington} (from ESA) and \\emph{Kepler} (from NASA) are planned for launch in 2007. Transit signals from terrestrial planets are small ($\\Delta F/F \\simeq 10^{-4}$), short ($\\Delta t\\simeq 10$ hours) dips, which repeat with periodicity of a few months, in time series lasting up to a few years. The reliable and automated detection of such signals in large numbers of light curves affected by different sources of noise is a statistical and computational challenge. We present a novel algorithm based on a Bayesian approach. The algorithm is based on the Gregory-Loredo method originally developed for the detection of pulsars in X-ray data. In the present paper the algorithm is presented, and its performance on simulated data sets dominated by photon noise is explored. In an upcoming paper the influence of additional noise sources (such as stellar activity) will be discussed. ",
        "introduction": "\\label{intro}  The search for rocky, terrestrial planets around other stars is a key research topic in astrophysics for the next decade. Following the first exo-planet detection around a sun-like star \\citep{mq95}, gaseous giants around other solar-type stars have been shown to be relatively common \\citep{bmf01}. The mass function of the current crop of extra-solar planets grows rapidly toward the lower masses \\citep{bmf01}, showing that low-mass planets must be common. However, the radial velocity technique, which has resulted in the detection of the exo-planets detected so far, is limited to planetary masses somewhat smaller than Saturn, and cannot reach the domain of terrestrial planets. This is due to astrophysical effects, such as microturbulence in the star's atmosphere, rather than instrumental limitations.  The most promising approach for the detection of (significant numbers) of terrestrial planets around stars other than the Sun appears to be the search for planetary transits, i.e.\\ dips in the light curve of the parent stars caused by the planet transiting in front of the stellar disk. The flux dip caused by the transit is also small, $\\Delta F/F = (R_{\\mathrm{p}}/R_*)^2$, which for the transit of an Earth-Sun system gives $\\Delta F/F = 10^{-4}$. This is well below the scintillation noise caused by the Earth's atmosphere \\citep[see]{f+00}, so that high-accuracy space-based photometry will be needed for the detection of such events. The probability of occurrence of a transit depends on the inclination of the planetary orbit relative to the line of sight (which must be close to $i = 90$ degrees), and is relatively small (for a set of randomly oriented Sun-Earth systems $p \\simeq 0.5\\%$), so that searches for planetary transits must be based on observation of large samples of target stars. A typical transit duration will be of order $\\Delta t \\simeq 10$ hours, and the transit periodicity will be the same as the orbital period of the planet, typically several months.  A number of space missions wholly or partially dedicated to the search for planetary transits are either in development or in the planning stage. The CNES/European satellite COROT is planned for launch in 2005, while the ESA mission \\emph{Eddington} and the NASA mission \\emph{Kepler} are planned for launch in 2007. Given the intrinsically statistical nature of planetary transit searches, these missions will acquire large number of stellar light curves, ranging from thousands for COROT to hundreds of thousands for \\emph{Eddington} and \\emph{Kepler}. Also, some smaller searches are being conducted for limited time periods (and concentrating on larger planets) using e.g.\\ HST (\\citealp{gbg+2000}) or ground-based telescopes (e.g.\\ \\citealp{ddk+2000}).  The analysis of data from such searches, and in particular the detection of transits with a high degree of certainty and a low false alarm rate, is a challenging task. The transit signal is weak ($\\Delta F/F = 10^{-4}$), and concentrated in a small fraction of the total signal: for a habitable planet orbiting a K5V star the orbital period will be roughly 4 months, so that for a 1 year light curve three events will be present. As each transit lasts $\\approx$ 10 hours, the transit signal is present in only $\\approx 0.3\\%$ of the total light curve.  In the Euclidean regime, the number of stars in a given field increases toward fainter magnitudes by a factor of $\\approx 4$ per magnitude. This is the case for the range of magnitudes and the low Galactic target latitudes of interest for currently planned missions . Therefore, most of the detected planets will be in the light curves of the fainter (and thus statistically noisier) stars, impying the need for effective robust data analysis algorithms able to reliably detect transits ``hidden in the noise''. At the same time, the large number of light curves which will need to be analyzed, each with a large number of points (of order $10\\,000$ points for a year of data) makes the use of efficient algorithms necessary, and rules out brute force approaches.  Some ground- (\\citealp{ddk+2000}) and HST-based (\\citealp{gbg+2000}) transit searches, which deal with relatively small numbers of light curves, use a detection approach based on comparing large numbers of model transits to the light curves and minimising a $\\chi^2$ statistic (or a linear statistic in the case of Doyle). These approaches are computationally very intensive, and thus may be unsuitable for the routine processing of the large number of light curves which will be produced by upcoming space missions.  As an alternative, transit detection algorithms based on Bayesian methods have recently been the subject of some attention. They have the advantage of combining computational efficiency with flexibility. While a global statistic can be used for the detection, information is directly available to reconstruct the detected signal if wanted, therefore providing a tool to discriminate between planetary transits and other types of periodic signals (\\citealp{dbd2001}), as well as directly measuring additional planetary characteristics such as the planet's radius.  In the present paper we present a novel algorithm for the detection of planetary transits based on the method developed by \\citet{gl92b} (hereafter referred to as GL method) for the search of pulsed emission from pulsars in X-ray data. While the algorithm was developed to be ``general purpose'', we have tuned it with the parameters of the upcoming \\emph{Eddington} planet finding mission in mind. The present paper discusses the characteristics of the algorithm on the basis of extensive simulations for the case in which the light curve is dominated by photon noise. Its performance in the case in which stellar activity is the dominating noise source will be the subject of a future paper.  Bayesian algorithms for the detection of planetary transits are also being developed in the context of the COROT mission. In particular, an approach based on expansion of the light curve into a truncated Fourier series is being investigated (\\citealp{ddb2001}). Perfoming the detection in the Fourier domain can make the algorithm computationally sensitive to data gaps and sampling rates. Here we explore a more robust \\emph{direct space} approach.  The GL (\\citealp{gl92b}) method, was initially developed for the detection of X-ray pulsars (where Poisson statistics dominate) and later extended to the Gaussian noise case (\\citealp{gre99}). At the flux levels of interest for the transit searches for \\emph{Eddington}, the photon shot noise per detection element (which is Poissonian) can be very well represented by Gaussian noise (see Sect.~\\ref{trans}). The original formulation of the GL algorithm is well-suited to the detection of periodic signals of unknown shape. However, in the planetary transit problem we have strong prior information about the transit shape. In this paper we modify the GL algorithm to perform more optimally for planetary transit detection. We do this by allowing one of the bins to have a variable width, to represent the out of transit constant signal level. This formulation also permits the phase of the transits to be identified, a task the original GL method is not suited for (see Sect.~\\ref{bay}). The fitted parameters are the period, duration and phase of the transit. The shape of the transit can then be reconstructed from the phase-folded light curve.  The simulated light curves are described in Sect.~\\ref{lc}. The algorithm is outlined in Sect.~\\ref{algo} and compared with the original GL algorithm in Sect.~\\ref{comp}. Sect.~\\ref{perf} describes the evaluation of the algorithm's performance by determining the number of false alarms and missed detections in a large sample of simulated light curves with and without transits. Conclusions and options for future work are presented in Sect.~\\ref{concl}. ",
        "conclusions": "\\label{concl}  A novel algorithm to detect transits due to extra-solar planets in stellar light curves has been developed and tested. The algorithm, based on a Bayesian approach, has proved successful in the tests performed so far, which include the effects of photon noise and data gaps. Using the photometric accuracy and throughput expected for the \\emph{Eddington} mission, we are able to detect an Earth-sized planet orbiting a K5V-type star with a period of 4 months down to an apparent stellar magnitude of $V \\simeq 14.5$.  Randomly distributed data gaps lasting up to two hours each and covering up to 20\\% of the light curve do not significantly affect the performance of the algorithm. The minimum number of transits in one light curve required for high confidence detections is three, however the algorithm's performance degrades gracefully for small number of transits, so that detections are possible for individual transits, albeit at a lower confidence level. This will allow for the detection of larger planets in long-period orbits (analogous to the gaseous giants of our solar system), likely to transit only once in the three year planet detection phase planned for the \\emph{Eddington} mission.  The most serious additional noise source to perturb planetary transit detections from space, is likely to be intrinsic stellar micro-variability (mostly activity-induced). At the moment it is also the least well investigated. The consequences of activity on the detection efficiency (using simulated light curves based on the solar light curves recorded by the VIRGO instrument on board SOHO, which spans all solar activity levels, from solar minimum to solar maximum) will be the subject of a future paper, in which the feasibility and effectiveness of using color information, as well as a number of pre-processing techniques such as whitening, will also be investigated.  The algorithm we have developed and discussed here has the potential to form part of a powerfull, multi-stage approach to analysing transit lightcurves. A more optimised processing method will be discussed in a separate paper. It will include a variability filtering stage, followed by distinct detection and parameter estimation stages, using a combination of a matched filter approach and of the present algorithm.  The performance of the algorithm presented here shows that the search of planetary transits with amplitudes comparable to the intrinsic noise level of the data set is fully feasible, and thus represents an important element in the development of the future generation of transit-based planet finding missions."
    },
    "0209/physics0209002_arXiv.txt": {
        "abstract": "The AMS spectrometer will be installed on the International Space Station in 2005. Among other improvements over the first version of the instrument, a ring imaging Cherenkov detector (RICH) will be added and should open a new window for cosmic-ray physics, allowing isotope separation up to A$\\approx$25 between 1 and 10 GeV/c and element identification up to Z$\\approx$25 between threshold and 1 TeV/c/nucleon. It should also contribute to the high level of redundancy required for AMS and reject efficiency albedo particles. A second generation prototype has been operated for a few months : the architecture and the first results are presented. ",
        "introduction": "The AMS spectrometer \\cite{barrau} will be implemented on the International Space Station in 2005. The instrument will be made of a superconducting magnet which inner volume will be mapped with a tracker consisting of 8 planes of silicon microstrips with a set of detectors for particle identification placed above and below the magnet: scintillator hodoscopes, electromagnetic calorimeter (ECAL), transition radiation detector (TRD) and ring imaging Cherenkov (RICH). This contribution is devoted to a study a second generation prototype aiming at the RICH testing. \\\\  The physics capability of the RICH counter has been investigated by simulations \\cite{buenerd}. It should provide unique informations among the AMS detectors by several respects : \\begin{itemize} \\item Isotopes separation up to A$\\approx$25 at best, over a momentum range extending from about 1-2 GeV/c up to around 13 GeV/c. \\item Identification of chemical elements up to Z$\\approx$25 at best, up to approximately 1 TeV/nucleon. \\item High efficiency rejection of albedo particles for momenta above the threshold, between 1 GeV/c and 3.5 GeV/c depending on the type of radiator. \\end{itemize} The RICH counter will allow to collect a unique sample of nuclear astrophysics data with unprecedented statistical significance over a momentum range totally unexplored for the most interesting isotopes. Fig. \\ref{fig:alexi} shows, as an example, the $^{10}$Be to $^{9}$Be ratio with 6 weeks of counting time \\cite{alexi}. Both the number of events and the covered energy range will dramatically improve the available data (lower left points on the plot).  \\begin{figure}[h] \\begin{center}             % \\includegraphics[scale=.3]{be10.eps} \\caption{\\label{fig:alexi} Expected statistics for the $^{10}$Be in 6 weeks of counting with AMS \\cite{alexi}.} \\end{center} \\vspace{-0.2cm} \\end{figure}  Recent works \\cite{donato} have emphasized the importance of measuring cosmic nuclei spectra for: 1) Setting strong constraints on the astrophysical and cosmic ray propagation parameters of the galaxy : the diffusion coefficient normalisation and its spectral index, the halo thickness, the Alfv\\'en velocity and the convection velocity; 2) Increasing the sensitivity to new physics search for supersymmetric particles or primordial back holes; 3) Testing for the nature of the cosmic-ray sources : supernovae, stellar flares, Wolf-Rayet stars, etc ... ",
        "conclusions": ""
    },
    "0209/astro-ph0209386_arXiv.txt": {
        "abstract": "It has been suggested by a number of authors that the 2.7$\\K$ cosmic microwave background (CMB) radiation might have arisen from the radiation from Population III objects thermalized by conducting cosmic graphite/iron needle-shaped dust. Due to lack of an accurate solution to the absorption properties of exceedingly elongated grains, in existing literature which studies the CMB thermalizing process they are generally modelled as (1) needle-like spheroids in terms of the Rayleigh approximation; (2) infinite cylinders; and (3) the antenna theory. We show here that the Rayleigh approximation is not valid since the Rayleigh criterion is not satisfied for highly conducting needles. We also show that the available intergalactic iron dust, if modelled as infinite cylinders, is not sufficient to supply the required opacity at long wavelengths to obtain the observed isotropy and Planckian nature of the CMB. If appealing to the antenna theory, conducting iron needles with exceedingly large elongations ($>10^4$) appear able to provide sufficient opacity to thermalize the CMB within the iron density limit. But the applicability of the antenna theory to exceedingly thin needles of nanometer/micrometer in thickness needs to be justified. ",
        "introduction": "} The 2.7$\\K$ cosmic microwave background (CMB) is generally interpreted as being relic radiation from the early hot universe of a big bang origin. Alternative attempts at explaining the observed CMB as a post-big bang phenomenon have been continuously made in terms of emission from ``Population III'' objects at high redshift (presumably either a pre-galactic generation of very massive stars or black hole accretion flows) thermalized by hollow spheres (Layzer \\& Hively 1973) or long slender conducting cosmic whiskers or ``cosmic needles'' (Hoyle, Wickramasinghe, \\& Reddish 1968; Wickramasinghe et al.\\ 1975; Rana 1980; Wright 1982; Hoyle, Narlikar, \\& Wickramasinghe 1984; Hawkins \\& Wright 1988; Hoyle \\& Wickramasinghe 1988; Bond, Carr, \\& Hogan 1991; Wickramasinghe 1992; Wickramasinghe et al.\\ 1992; Wickramasinghe \\& Hoyle 1994; Aguirre 2000).  The reason for invoking ``conducting needles'' is because neither spherical grains (both dielectric and metallic) nor dielectric needles have high enough opacity in the far infrared (IR) to be an efficient thermalizing agent unless an unreasonably large amount of dust is invoked. This can be seen from the absorption cross section expressions of spherical or spheroidal grains. Let $\\epsilon(\\lambda) = \\epsre + i\\epsim$ be the dust complex dielectric function at wavelength $\\lambda$. The absorption cross section $\\cabs$ per unit volume ($V$) for spheres in the Rayleigh regime (Bohren \\& Huffman 1983) is \\begin{equation} \\cabs/V \\approx \\frac{18\\pi}{\\lambda}\\frac{\\epsim}{(\\epsre+2)^2+\\epsim^2} ~~~. \\end{equation} For dielectric spheres, $\\cabs \\propto \\lambda^{-1}\\epsim \\propto \\lambda^{-2}$ approaches zero as $\\lambda\\rightarrow \\infty$ since at far-IR $\\epsre$ approaches a constant $\\gg\\epsim$ while $\\epsim \\propto \\lambda^{-1}$; for metallic spheres with a conductivity of $\\sigma$, $\\cabs\\propto \\lambda^{-1}\\epsim^{-1}\\propto \\lambda^{-2}$ also approaches zero as $\\lambda\\rightarrow \\infty$ since $\\epsim = 2\\lambda\\sigma/c \\propto \\lambda$ ($c$ is the speed of light) and $\\epsre \\ll \\epsim$. Let needle-shaped grains be represented by thin prolate spheroids of semiaxes $l$ along the symmetry axis and $a$ perpendicular to the symmetry axis. In the Rayleigh limit, the absorption cross section per unit volume for needle-like prolate grains ($l\\gg a$) is approximately \\begin{equation}\\label{eq:needle} \\cabs/V \\approx \\frac{2\\pi}{3\\lambda}\\frac{\\epsim} {\\left[L_{\\|}(\\epsre-1) + 1\\right]^2 + \\left(L_{\\|}\\epsim\\right)^2} \\end{equation} where $L_{\\|}\\approx \\left(a/l\\right)^2\\ln(l/a)$ is the depolarization factor parallels to the symmetry axis. For dielectric needles, $\\cabs \\propto \\lambda^{-1}\\epsim \\propto \\lambda^{-2}$ at far-IR since we usually have $L_{\\|}(\\epsre-1) + 1 \\gg L_{\\|}\\epsim$ while $\\epsre$ is insensitive to $\\lambda$ at long wavelengths; for metallic needles, it appears at first glance that, for a given value of $\\epsim$ (at a given $\\lambda$) -- no matter how large -- one can always find a sufficiently long needle with $L_{\\|}\\epsim \\simlt 1$ and $L_{\\|}(\\epsre-1) \\ll 1$ (Greenberg 1972) so that $\\cabs \\propto \\lambda^{-1}\\epsim \\propto \\sigma$ which can be very large.\\footnote{% This should not be considered inconsistent with the Kramers-Kronig relation since for a given elongation $\\ltoa$ there exists a long-wavelength cutoff ($\\lambda_0$) for $\\cabs$: $\\cabs \\propto \\lambda^{-2}$ as $\\lambda > \\lambda_0$ (see Eqs.[\\ref{eq:wright}-\\ref{eq:cabs0}]). } Therefore, it is possible for metallic needles with high electrical conductivities to provide a large quantity of opacity at long wavelengths to thermalize the cosmic background.  For conducting needles Eq.(\\ref{eq:needle}) can be expressed as (Wright 1982) \\begin{equation}\\label{eq:wright} \\cabs = \\frac{\\cabs^{0}}{1 + \\left(\\lambda/\\lambda_0\\right)^2} \\end{equation} where the long-wavelength cutoff $\\lambda_0$ is \\begin{equation}\\label{eq:cutoff} \\lambda_0 \\equiv \\frac{\\rho c}{2} \\frac{1 + L_{\\|}\\left(\\epsre-1\\right)}{L_{\\|}} \\approx \\frac{\\rho c}{2} \\frac{\\left(l/a\\right)^2}{\\ln\\left(l/a\\right)} \\end{equation} and \\begin{equation}\\label{eq:cabs0} \\cabs^0 \\equiv \\frac{4\\pi V}{3\\rho c} \\frac{1}{\\left[1 + L_{\\|}\\left(\\epsre-1\\right)\\right]^2} \\approx \\frac{4\\pi V}{3\\rho c} \\end{equation} where $\\rho = 1/\\sigma$ is the dust material resistivity. It is seen that Eq.(\\ref{eq:cutoff}) establishes a lower bound on the elongation $l/a$ of the needles which absorb strongly at wavelengths out to $\\lambda_0$.  Eqs.(\\ref{eq:needle}-\\ref{eq:cabs0}) have been widely used in obtaining the dust opacity in the far-IR and microwave range (Rana 1980; Wright 1982; Hawkins \\& Wright 1988; Hoyle \\& Wickramasinghe 1988; Bond, Carr, \\& Hogan 1991; Wickramasinghe et al.\\ 1992; Wickramasinghe \\& Hoyle 1994). However, none of these has explicitly taken into account the criterion to which the Rayleigh approximation is applicable (Bohren \\& Huffman 1983): \\begin{equation}\\label{eq:rayleigh} \\frac{2\\pi l}{\\lambda} \\ll 1 ~; ~~~ |m|\\frac{2\\pi l}{\\lambda} \\ll 1 \\end{equation} where $m(\\lambda) = \\mre + i\\mim$ is the complex refractive index ($\\epsilon = m^2$). For metals at long wavelengths we have $\\mre \\approx \\mim \\approx \\left(\\sigma\\lambda/c\\right)^{1/2}$. Therefore, the Rayleigh approximation (Eq.[\\ref{eq:rayleigh}]) establishes an upper bound on the needle length: \\begin{equation}\\label{eq:lmax} l \\ll \\frac{1}{2\\pi}\\left(\\frac{\\lambda c}{\\sigma}\\right)^{1/2} = \\frac{1}{2\\pi}\\left(\\lambda \\rho c\\right)^{1/2} ~~~. \\end{equation} The reason for applying this criterion for limiting the needle size is that, when it is not satisfied, the cross sections given by Eq.(\\ref{eq:needle}) are overestimates of the true cross sections. It is only when all elements within the particle radiate in phase with each other (i.e. negligible phase shift of light within the particle) that we can achieve the high absorptivities (Greenberg 1980). The implication of Eq.(\\ref{eq:lmax}) for cosmic needles is significant. For example, for iron needles of $\\rho = 10^{-16}\\s$ to absorb efficiently out to $\\lambda_0 = 5\\mm$, Eq.(\\ref{eq:cutoff}) requires an elongation of $l/a \\approx 1600$ (also see Wright 1982). To satisfy the Rayleigh criterion, Eq.(\\ref{eq:lmax}) leads to $l \\ll 1.9\\mum$. A combination of Eq.(\\ref{eq:cutoff}) and Eq.(\\ref{eq:lmax}) requires the needle radius $a \\ll 12\\Angstrom$. It is unlikely that such tiny iron needles exist in astrophysical environments. After all, for stacks of layers of $2\\times 2$ and $3\\times 3$ iron atoms the needle radius would already be $\\approx 2.8, 4.2\\Angstrom$, respectively. To be conservative, we therefore take the minimum radius of iron needles to be $\\amin = 3.5\\Angstrom$. In the following text, we will take the Rayleigh criterion to be \\begin{equation}\\label{eq:lmax0} |m|\\frac{2\\pi \\lmax}{\\lambda} \\approx 0.1 ~; ~~~ \\lmax \\approx \\frac{1}{20\\pi}\\left(\\lambda \\rho c\\right)^{1/2} \\end{equation} where $\\lmax$ is the maximum value of the needle length $l$ for which the Rayleigh approximation is still valid.  For a given wavelength $\\lambda$ and a given dust conductivity $\\sigma$ which is dependent on dust material and temperature we can obtain from Eq.(\\ref{eq:cutoff}) $\\ltoamin$ -- the lower limit on the needle elongation which displays appreciable opacity at wavelengths up to $\\lambda$ (following Wright 1982, we take $\\lambda=5\\mm$ for discussion); and from Eq.(\\ref{eq:lmax0}) $\\ltoamax = \\lmax/\\amin$ -- the upper limit on the needle elongation to which the Rayleigh approximation (Eq.[\\ref{eq:needle}]) is applicable.  In Figure \\ref{fig:l2aminmax} we present $\\ltoamin$ and $\\ltoamax$ for cosmic iron needles (see \\S\\ref{sec:ironnk}) thermalizing the background radiation emitted at redshift $z$ and observed at wavelength $\\lambda=5\\mm$. It is seen in Figure \\ref{fig:l2aminmax} that even with $\\amin=3.5\\Angstrom$, $\\ltoamax \\ll \\ltoamin$ for $z$ up to 200. This clearly indicates that it is not appropriate to adopt the Rayleigh approximation (Eq.[\\ref{eq:needle}]) when studying the CMB thermalization by cosmic iron needles.    \\begin{figure}[h] \\begin{center} \\epsfig{ file=elongation.cps, width=\\figwidth,angle=0} \\end{center}\\vspace*{-1em} \\caption{ \\label{fig:l2aminmax} \\footnotesize Lower [$\\ltoamin$] and upper [$\\ltoamax$] limits on the elongation of iron needles (assumed to thermalize background radiation emitted at $z$ and observed at $\\lambda=5\\mm$) as a function of redshift $z$ obtained respectively from the long-wavelength opacity consideration (Eq.[\\ref{eq:cutoff}]) and the Rayleigh criterion (Eq.[\\ref{eq:lmax0}]). It is apparent that the Rayleigh approximation (Eq.[\\ref{eq:needle}]) is not valid for studies of the CMB thermalization by iron needles since $\\ltoamax \\ll \\ltoamin$. } \\end{figure}  The absorption cross sections of needles may be approximated by those of infinite cylinders provided $l$ exceeds $a$ by a factor of $\\sim 4$ (Wickramasinghe 1973) or $\\sim 9$ (Lind \\& Greenberg 1966). For needles with a radius less than $\\sim 10\\Angstrom$ the classical scattering theory does not apply (Platt 1956; Greenberg 1960). We will adopt the infinite cylinder results but take a cutoff at $\\lambdac \\approx 400\\,l$ (Platt 1956). Similar to Li \\& Draine (2001), we assume a continuous distribution for the absorption properties of classic infinite cylinders and Platt particles ($a\\simlt 10\\Angstrom$): \\begin{equation}\\label{eq:cabsnew} \\cabs(\\lambda) = \\cabs^{\\rm inf}\\left[\\xiqm\\etacut + \\left(1-\\xiqm\\right)\\right] ~, \\end{equation} \\begin{equation}\\label{eq:xiqm} \\xiqm(a) = {\\rm min}\\left[1, (a_{\\xi}/a)^3\\right], ~~~~ a_{\\xi} = 10{\\rm \\AA} ~, \\end{equation} \\begin{equation}\\label{eq:etacut} \\etacut(\\lambda, \\lambdac) = \\frac{1}{\\pi} \\arctan\\left[\\frac{10^3 (y-1)^3} {y}\\right] + \\frac{1}{2}, ~~~ y=\\lambdac/\\lambda, \\lambdac = 400\\,l \\end{equation} where $\\cabs^{\\rm inf} (=2a \\Qabs)$ is the absorption cross sections for infinite cylinders ($\\Qabs$ is the absorption cross section per unit length divided by $2a$), and $\\etacut$ is a cutoff function.  In the far-IR and microwave regions, for metallic needles both inductance and charge separation effects are ignorable (Hoyle \\& Wickramasinghe 1988). Therefore, conducting needles can be treated as an antenna (Wright 1982): $\\cabs/V = 4\\pi/\\left(3\\rho c\\right)$ for $\\lambda < \\lambda_0$ where $\\lambda_0$ is the same as in Eq.(\\ref{eq:cutoff}). Taking into account the quantum mechanical effect, the absorption cross section of an antenna can be expressed as \\begin{equation}\\label{eq:ant} \\cabs(\\lambda)/V = \\frac{4\\pi}{3\\rho c}\\left[\\xiqm\\etacut(\\lambda,\\lambdac) + \\left(1-\\xiqm\\right)\\right] /\\left[1 + \\left(\\lambda/\\lambda_0\\right)^2\\right] \\end{equation} where the $1/\\left[1 + \\left(\\lambda/\\lambda_0\\right)^2\\right]$ term accounts for the cutoff at $\\lambda_0$ which has been justified by Wright (1987).   In Figure \\ref{fig:cabs2v} we compare the absorption cross sections (per unit volume) at $\\lambda=5\\mm$ calculated from infinite iron cylinders and iron antennae with a radius of $a=0.1\\mum$ and a range of elongations at $T_{\\rm d}=2.7\\K$. Although the Rayleigh approximation is not valid for iron needles capable of efficiently supplying far-IR and microwave opacity, we also present in Figure \\ref{fig:cabs2v} results obtained from the Rayleigh approximation (Eq.[\\ref{eq:needle}]) since it is widely used in literature. It is seen in Figure \\ref{fig:cabs2v} that the infinite cylinder model predicts a much larger 5$\\mm$ opacity for $\\ltoa <3\\times 10^4$ and a much smaller one for $\\ltoa > 3\\times 10^4$. The antenna model shows a rapid drop at $\\ltoa < 1.2\\times 10^5$ which corresponds to a cutoff wavelength $\\lambda_0\\approx 5\\mm$ (see Eq.[\\ref{eq:cutoff}]). This is because the long-wavelength cutoff becomes smaller as $\\ltoa$ decreases so that the iron opacity at 5$\\mm$ decreases rapidly, too. This trend is also seen in the Rayleigh curve.\\footnote{% The reason why the onset $\\l/a$ value of the drop in the $\\cabs/V$ curve for the Rayleigh approximation model differs from that for the antenna model is because the second term in the r.h.s of Eqs.(\\ref{eq:cutoff}-\\ref{eq:cabs0}) does not always hold. } The dramatic differences among the absorption  cross sections calculated from the three methods will have dramatic effects on the CMB thermalization model (see \\S\\ref{sec:cmb}).  \\begin{figure}[h] \\begin{center} \\epsfig{ file=cabs2v.cps, width=\\figwidth,angle=0} \\end{center}\\vspace*{-1em} \\caption{ \\label{fig:cabs2v} \\footnotesize Absorption cross sections per unit volume at $\\lambda=5\\mm$ as a function of elongation for iron needles with a radius of $a=0.1\\mum$ at $T_{\\rm d}=2.7\\K$ calculated from the infinite cylinder representation (solid line), Rayleigh approximation (dashed line), and antenna theory (dotted line). Note that the Rayleigh approximation results should not be considered too seriously since the Rayleigh criterion which requires $\\ltoa < 0.03$ is never satisfied. } \\end{figure}  It is the purpose of this {\\it Letter} first to show (see above) that the widely adopted Rayleigh approximation is not applicable to studies of the CMB as a result of thermalization by cosmic needles, and then to estimate the quantity of dust required to thermalize the background radiation using the absorption cross sections of either infinite cylinders or antennae. We will only consider iron grains since they absorb more efficiently in the far-IR than graphite and also because the upper bound on the total amount of microwave radiation generated by graphite needles was shown considerably smaller than the observed CMB and it was also shown that the optical depth of the graphite needle-containing cloud is not sufficiently large for the cloud to radiate like a black body (Sivaram \\& Shah 1985). Condensed in supernova ejecta, iron grains may likely form as slender whiskers by the ``screw dislocation'' mechanism which has been attested experimentally and (at least some fraction of them) are then expelled into extragalactic space without significant destruction due to sputtering (see Hoyle \\& Wickramasinghe 1988 and references therein). It is interesting to note that small iron particles were among the materials initially proposed to be responsible for the interstellar reddening, based on an analogy with small meteors or micrometeorites supposedly fragmented into finer dust (Schal\\'{e}n 1936; Greenstein 1938). In \\S\\ref{sec:ironnk} we calculate the dielectric functions of iron grains based on the Drude theory. In \\S\\ref{sec:cmb} we calculate the extinction optical depth caused by cosmologically distributed iron needles and estimate the amount of iron needles required to thermalize the CMB and compare it with the available intergalactic iron density. Concluding remarks are given in \\S\\ref{sec:conclusion}. ",
        "conclusions": "\\label{sec:conclusion} A wide variety of work have proposed the cosmic needle model as the CMB thermalizing agent: if cosmic metallic needle-shaped grains absorb strongly at all wavelengths from IR to microwave wavelengths it is possible to ascribe the observed background radiation at frequencies greater than $1\\cm^{-1}$ as originating from thermalization, by these slender needles, of the radiation of Population III objects. It is pointed out here that in many of these results insufficient attention was given to the limits of applicability of the small particle (Rayleigh) approximation. It is shown that the widely adopted Rayleigh approximation is not applicable to conducting needles capable of supplying high far-IR and microwave opacities. Due to lack of an accurate solution to the absorption properties of slender needles, we model them either in terms of infinite cylinders or the antenna theory. It is found that the available intergalactic iron dust, if modelled as infinite cylinders, is not sufficient to produce a sufficiently large optical depth at long wavelengths required by the observed isotropy and Planckian nature of the CMB. In the context of the antenna theory, conducting needles with exceedingly large elongations ($>10^4$) appear to be capable of satisfying the optical depth requirement without violating the iron density limit. But the applicability of the antenna theory to exceedingly thin needles of nanometer/micrometer in thickness needs to be justified."
    },
    "0209/gr-qc0209078_arXiv.txt": {
        "abstract": "Testing of the gravitation equations, proposed by one of the authors earlier, by a binary pulsar is considered. It has been shown that the formulas for the gravitation radiation of the system resulting from the equations do not contradict the available observations data ",
        "introduction": "In paper \\cite{Verozub1} gravitation equations which do not lead to a physical singularity in the center of the spherically symmetric field were proposed. These equations also predict that there can exist stable supermassive compact configurations of the degeneration Fermi - gas without an events horizon \\cite{Verozub2}. The equations do not contradict classical tests at the distances from the center which are much larger than the Schwarzschild radius. In the present paper we find the power of the gravitational-wave radiation from a close binary system and use the result to find the deceleration of the orbital period of the pulsar PSR1913+16 conditioned by the gravitational radiation. In this case we deal with a moderately strong gravitation field and use the definition of gravitational energy that follows from the gravitation equations under consideration. ",
        "conclusions": ""
    },
    "0209/hep-ph0209093_arXiv.txt": {
        "abstract": "K-essence is a possible candidate for dark energy of the Universe. In this paper we consider couplings of k-essence to the matter fields of the standard electroweak theory and study the effects of the cosmological CPT violation induced by the CPT violating {\\bf Ether} during the evolution of the k-essence scalar field on the laboratory experiments and baryogenesis. Our results show that the matter and antimatter asymmetry can be naturally explained {\\it via} leptogenesis without conflicting with the experimental limits on CPT violation test. The mechanism for baryogenesis proposed in this paper provides a unified picture for dark energy and baryon matter of our Universe and allows an almost degenerate neutrino mass pattern with a predicted rate on the neutrinoless double beta decays accessible to the experimental sensitivity in the near future. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209045_arXiv.txt": {
        "abstract": "White Dwarf luminosities are powerful age indicators, whose calibration should be based on reliable models. We discuss the uncertainty of some chemical and physical parameters and their influence on the age estimated by means of white dwarf cooling sequences. Models at the beginning of the white dwarf sequence have been obtained on the base of progenitor evolutionary tracks computed starting from the zero age horizontal branch and for a typical halo chemical composition (Z=0.0001, Y=0.23). The uncertainties due to nuclear reaction rates, convection, mass loss and initial chemical composition are discussed. Then, various cooling sequences for a typical white dwarf mass (M=0.6 M$_\\odot$) have been calculated under different assumptions on some input physics, namely: conductive opacity, contribution of the ion-electron interaction to the free energy and microscopic diffusion. Finally we present the evolution of white dwarfs having mass ranging between 0.5 and 0.9 M$_\\odot$. Much effort has been spent to extend the equation of state down to the low temperature and high density regime. An analysis of the latest improvement in the physics of white dwarf interiors is presented. We conclude that at the faint end of the cooling sequence ($log L/L_{\\odot} \\sim -5.5$) the present overall uncertainty on the age is of the order of 20\\%, which correspond to about 3 Gyr. We suggest that this uncertainty could be substantially reduced by improving our knowledge of the conductive opacity (especially in the partially degenerate regime) and by fixing the internal stratification of C and O. ",
        "introduction": "Our WD models have been obtained by means of a full evolutionary code. In particular we have used the FRANEC in the version described by Chieffi \\& Straniero (1989). The input physics have been completely revised in order to account for the peculiar conditions developed in WD interiors.  \\subsection{Opacity} We have adopted the radiative opacities of OPAL (Iglesias \\& Rogers 1996)  for high temperature (log$T[K]> 4.0$) and the Alexander \\& Ferguson (1994) molecular opacities for the low temperatures (log$T[K]\\leq 4.0$). The conductive opacities have been derived from the work made by Itoh and coworkers (Itoh et al. 1983, Mitake, Ichimaru, \\& Itoh 1984, Itoh, Hayashi, \\& Kohyama 1993). Additional models have been obtained by using the Hubbard \\& Lampe (1969) prescriptions and the table provided by A. Potekhin (see Potekhin et al. 1999). As it is well known (see e.g. Mazzitelli 1994, D'Antona \\& Mazzitelli 1990), there is a region in the $T-\\rho$ plane not covered by the OPAL radiative opacities and where the electron conductivity is not yet dominant. Thus one has to extrapolate the radiative opacities in order to fill this gap. The problem is alleviated by the use, for the lower temperature, of the Alexander \\& Ferguson (1994) opacities, which extend at larger densities with respect to the OPAL. Present models have been computed adopting a linear extrapolation of the radiative opacity beyond the upper density provided by OPAL. We have tested the reliability of this choice by changing the extrapolation method. Negligible effects on the cooling sequences have been found.   \\subsection{Nuclear reaction rates and neutrinos} Nuclear reaction are almost extinct during the WD evolution, but they play a relevant role in determine the chemical structure of the model at the beginning of the cooling sequence. We have used the rates tabulated by the NACRE collaboration (Angulo et al. 1999), except for the $^{12}$C$(\\alpha,\\gamma)^{16}$O. During the He-burning this reaction competes with the $3\\alpha$ and regulates the final C/O ratio in the core (Iben 1972). The great influence on the resulting WD evolution has been deeply investigated (D'Antona \\& Mazzitelli 1990; Salaris et al. 1997). We have used two different rates, namely the one reported by Caughlan et al. (1985) and that of Caughlan \\& Fowler (1988). Note that the difference between these two compilation may be roughly considered as representative of the present experimental uncertainty level (see e.g. Buchmann 1996). Nuclear reaction rates are corrected for the weak and intermediate electron screening by using the prescriptions of Graboske et al. 1973 and  De Witt, Graboske, \\& Cooper 1973, and for the strong screening by Itoh, Totsuji, \\& Ichimaru 1977 and Itoh et al. 1979.  The efficiency of neutrino emission processes are taken from: Haft, Raffelt, \\& Weiss 1994 (plasma neutrinos), Itoh et al. 1989 (photo and pair neutrinos).   \\subsection{Equation of state} As recognized long ago (Salpeter 1961) Coulomb interactions play a relevant role in WD interior, finally leading to crystallization of the stellar core (Lamb \\& Van Horn 1975). Thus, a detailed treatment of the thermodynamic behavior of both liquid and solid matter is a necessary physical ingredient to study the evolution of cold WDs. The high pressure experienced in the whole core ensures the complete ionization of carbon and oxygen. Thus, we have updated and extended the EOS for fully ionized matter described by Straniero (1988). In particular, we have revised the treatment of the electrostatic corrections up to the liquid-solid transition and beyond. The free energy in the fluid phase can be written as $$ F_L=F_i^{id}+F_e^{id}+F_i^{ex}+F_e^{ex}+F_{ie} $$ where $F_i^{id}$ and $F_e^{id}$ are respectively the contribution of the ideal gas of ions and electrons. To compute these terms, as in Straniero (1988), we have assumed that ions follow the Boltzmann distribution, while electrons are described by Fermi-Dirac integrals, for an arbitrary degree of degeneracy and relativistic state. $F_i^{ex}$ is the excess  of ionic free energy due to ion - ion Coulomb interactions. For this contribution we adopted the analytical expression by Potekhin \\& Chabrier (2000) that in the region of high ionic coupling parameter ($\\Gamma= (Ze)^2/(a k_B T) \\geq 1$) fits the accurate results of recent Monte Carlo simulations  of one component plasma (De Witt \\& Slattery 1999), while for $\\Gamma < 1$ it approaches to the Cohen \\& Murphy (1969) expansion and reproduces the Debye-Huckel limit for vanishing $\\Gamma$  (Landau \\& Lifshitz 1969). The quantum diffraction correction to ionic free energy has been neglected in the liquid phase. $F_e^{ex}$ represents the excess electron free energy due to electron - electron interactions. For this term we adopted the non relativistic expression by Tanaka, Mitake, \\& Ichimaru 1985. We neglected the relativistic $F_e^{ex}$. This is a reasonable approximation since at high density, when degenerate electrons become relativistic, the electron coupling parameter $\\Gamma_e \\ll 1$. Finally, for $F_{ie}$, that describes the excess due to ion - electron interactions, we adopted the analytical expression given by Potekhin \\& Chabrier (2000). As it is well known, the contribution of the ion-ion interactions increases with density, eventually leading to crystallization. According to Potekhin \\& Chabrier (2000), the liquid-solid phase transition has been assumed at $\\Gamma_c=175$. In the solid phase ($\\Gamma \\geq \\Gamma_c$) the free energy can be written as $$ F_S=F_e^{id}+F_i^{s}+F_e^{ex}+F_{ie}+F_i^{qm} $$ where the electron terms $F_e^{id}$ and  $F_e^{ex}$ are the same as in the fluid phase. For the free energy of the ionic crystal $F_i^{s}$ we used the analytical expression by Farouki \\& Hamaguchi (1993) obtained by fitting numerical models of molecular dynamic. The contribution of the ion - electron $F_{ie}$ in the solid phase is from Potekhin \\& Chabrier (2000). Finally, the term $F_i^{qm}$ represents the quantum correction to the thermodynamic behavior of the ionic Coulomb crystal. This contribution is very important in late WD evolutionary phases, in fact, as the WD cools down, the crystallized CO core reaches a quantum state (diffraction parameter $\\hbar \\Omega_P/(k_B T) > 1$, where $\\Omega_P$ denotes the ion plasma frequency). When this occurs, the ionic contribution to the specific heat decreases as $T^3$, thus depleting the main thermal reservoir of the star: it is called Debye cooling phase. For this term we adopted the analytical expression described by Stolzmann \\& Blocker (2000), which is based on the free energy of a Coulomb crystal studied by Chabrier, Ashcroft, \\& De Witt (1992). The various thermodynamic quantities for pure carbon and pure oxygen have been obtained by analytically deriving the corresponding free energy. Finally, an additive volume interpolation is used to calculate the thermodynamic quantities of a CO mixture\\footnote{We neglect the contribution of $^{22}$Ne.}.  To describe the outer layers of partially ionized helium and hydrogen, we adopted the EOS of Saumon, Chabrier, \\& Van Horn (1995, SCVH). This EOS requires both high pressure and low pressure extensions. At high pressure (P$>10^{19}$ dyn/cm$^2$), which is in any case large enough to guarantee a full ionization of H and He, we extended the SCVH EOS following the same procedure described above for the CO core. The match between the two EOS is generally good. At low pressure, we have used a perfect gas (including H, H$^+$, H$_2$, H$^-$, He, He$^+$ and He$^{++}$) plus radiation. In such a case the classical Saha equation has been used to derive  the population of the various species. For each temperature, the precise value of the maximum pressure of the perfect gas (or the minimum of the SCVH) has been varied in order to guarantee a smooth transition between the two EOS.  Tables of this EOS are available on the web.   \\subsection{Model Atmospheres} In order to fix the external boundary condition of a stellar model, appropriate model atmosphere are needed. In the last decade WD atmosphere theory has significantly improved (Bergeron, Waesemael, \\& Fontaine 1991; Bergeron, Saumon, \\& Waesemel 1995; Bergeron, Waesemel, \\& Beauchamp 1995; Saumon \\& Jacobson 1999; Bergeron 2000). As it is well known, in the high density/low temperature atmospheres of an old WD the emerging electromagnetic flux significantly departs from the black body spectrum. This is mainly due to the molecular hydrogen recombination and the consequent collision - induced absorption (CIA) by H$_2$ - H$_2$ collisions. CIA is the main source of opacity in the infrared and the main cause of increasingly blue color indices for decreasing effective temperature (T$_e$) in cold WDs (Hansen 1998, 1999; Saumon \\& Jacobson 1999; Bergeron 2000). Detailed model atmospheres are also crucial ingredient to transform the theoretical quantities (L, T$_e$) into the observational ones (magnitude, color). In the present computations we have used a solar scaled T-$\\tau$ relation (see Chieffi \\& Straniero for details) up to the onset of the cooling sequence. Then, namely when log L/L$_\\odot \\sim 0$, we adopt the model atmospheres of Bergeron, Saumon, \\& Wesemael (1995), plus a low temperature extension including the effects of the CIA, which has been kindly provided us by P. Bergeron. Since no metals are included in these models, they are consistent with the usual hypothesis of complete sorting of the external layers (see next subsection).    \\subsection{Treatment of convection and other mixing phenomena} Convective boundaries are fixed by using the Schwarzschild criterion. Concerning the progenitor evolution, the extension (in mass) of the convective core during the central He burning have a great influence on the amount of C (and O) in the most internal layers of a WD (Imbriani et al. 2001 and Dominguez et al. 2001). During this phase, we use the algorithm described by Castellani et al. (1985) to take into account the growth of the convective core induced by the conversion of He (low opacity) into C and O (large opacity) and the resulting semiconvection. Breathing pulses, which occur when the central He mass fraction decreases below about 0.1, are usually neglected (but see the discussion in section 3.3). Note that any mechanism that increases the size of the well mixed region during the final part of the He-burning (mechanical overshoot, semiconvection, breathing pulses or rotational induced mixing) leads to a reduction of the resulting amount of C in the central region of the WD (see e.g. Imbriani et al. 2001).  Theoretical studies (Fontaine \\& Michaud 1979, Iben \\& MacDonald 1985, 1986, Althaus et al. 2002), supported by observational evidences (see e.g. Bergeron, Ruiz, \\& Legget 1997), indicates that, as a consequence of the gravitational settling of heavy elements, DA WDs should have a practically pure H envelope and a pure He mantel. In addition, as suggested by Salaris et al. (1997), the carbon and oxygen profiles left by the He-burning is smoothed by Rayleigh-Taylor instabilities. In order to account for the element diffusion, when log L/L$_\\odot \\sim 0$, we adjust the composition of the envelope and the mantel: all the residual H is putted on the top of a pure He layer. The total mass of H and He is conserved. We also modify the internal C and O profile according the the prescription of Salaris et al. (1997).  Owing to the larger $\\Gamma$, oxygen crystallization occurs when carbon is still liquid. For a certain time, this occurrence produces an unstable stratification of the liquid phase and, in turn, an efficient mixing (Isern et al. 1997, Salaris et al. 1997). This phenomenon is presently not included in our calculations. ",
        "conclusions": "In the last decade cosmochronology based on WDs has become a promising tool of scientific inquiry thanks to  both the availability of a large amount of high quality data for cool and faint WDs and the improvement of the understanding of the high density plasma physics. However, as demonstrated by the rather large discrepancies among the recently published theoretical cooling sequences, a firm calibration of the age-luminosity relation is still not available.  We have analyzed some of the main sources of uncertainty affecting the theoretical cooling time. Physical and chemical parameters characterizing the white dwarfs and the progenitors evolutions have been revised.  Concerning progenitors, we found that the larger uncertainty is due to the combined action of convective mixing a nuclear reactions operating during the central He-burning phase. Both these processes are largely affected by theoretical and experimental uncertainties. They determine the amount of C (and O) left in the core of the WD and, in turn, have a great influence on the predicted cooling rate. A conservative analysis allow us to conclude that the overall impact of the uncertainties due to the progenitors evolution on the estimated WD ages at logL/L$_\\odot = -5.5$ is of the order of 2 Gyr. We hope that this uncertainty will be significantly reduced in the next future thanks to the renewed effort of nuclear physicists in measuring the $^{12}$C$(\\alpha,\\gamma)^{16}$O reaction rate at low energy (see Gialanella et al. 2001 and Kunz et al. 2001).  Concerning WD physics we emphasize the relevance of a reliable description of the electron conductivity which is the main mechanism of energy transport in WD interiors. Actually, a consistent part of the large discrepancies found in the comparisons of published cooling sequences may be due to the adopted conductive opacity. The old Hubbard and Lampe (1969) conductive opacity are probably overestimated, even in the weakly degenerate regime. The latest computations by Itoh and coworkers and Potekhin et al. (1999) imply a substantial reduction of the age. At log$L/L_\\odot = -5.5$, we obtain models $\\sim 2.5$ Gyr younger than that obtained with HL69.  Let us conclude by noting that most of the uncertainties discussed in this paper mainly affect old WDs, but have negligible influence on the age estimated for young WDs, whose luminosity is larger than $10^{-4}$ L$_\\odot$. This implies that the presently available theoretical scenario may be safely used to estimate the age of young stellar systems as, for example, the intermediate age Open Clusters (t$<$3 Gyr).     \\noindent {\\bf Acknowledgments:}  We wish to thank P. Bergeron for kindly providing us the models atmosphere and S. Degl'Innocenti, Luciano Piersanti and Jordy Isern for the many helpful comments. We thank also S. Shore for the careful reading of the manuscript and the several useful discussions. We are deeply indebted to V. Castellani for continuous encouragements and for a critical review of the manuscript. This work has been supported by the MIUR Italian grant Cofin2000."
    },
    "0209/astro-ph0209273_arXiv.txt": {
        "abstract": "{Recent results from cosmic microwave background (CMB) experiments verify several of the predictions of inflation, while ruling out a number of alternative structure-formation scenarios.  Given the successes of the theory, the obvious next step is to press ahead and test inflation to the edge of all our current and forthcoming observational abilities.  According to the inflationary paradigm, galaxies and their large-scale distribution in the Universe are remnants of inflation and can thus be studied to learn more about inflation in the same way that experimental particle physicists study the remnants of high-energy collisions.  Here I discuss how studies of galactic substructure, galaxies, clusters, large-scale structure, and the CMB, may be used to learn more about inflation.} \\resumen{Resultados recientes de experimentos del fondo de microondas c\\'osmico (FMC) prueban varias de las predicciones de la inflaci\\'on y descartan muchos de los escenarios de formaci\\'on c\\'osmica alternativos.  Dado el \\'exito de la teor\\'{\\i}a, el siguiente paso obvio  es probar la inflaci\\'on al filo de nuestras capacidades observacionales actuales y por venir. De acuerdo al paradigma inflacionario, las galaxies y su distribuci\\'on a gran escala son remanentes de la inflaci\\'on por lo que pueden ser estudiadas para explorarla m\\'as a fondo, de la misma manera que un f\\'{\\i}sico de part\\'{\\i}culas experimental estudia los remanentes de colisiones de alta energ\\'{\\i}a. A continuaci\\'on har\\'e un an\\'alisis de c\\'omo estudios de subestructura gal\\'actica, galaxias, c\\'umulos, estructura a gran escala y el FMC pueden ser usados para aprender m\\'as sobre la inflaci\\'on.}  \\addkeyword{Cosmology}   \\begin{document} ",
        "introduction": "\\label{sec:intro}  Inflationary cosmology \\cite{Gut81,Lin82a,AlbSte82} has in recent years had a number of dramatic successes.  The inflationary predictions of a flat Universe and nearly scale-invariant primordial density perturbations with Gaussian initial conditions \\cite{GutPi82,Haw82,Lin82b,Sta82,BarSteTur83} have been found to be consistent with a series of increasingly precise cosmic microwave background (CMB) experiments \\cite{Miletal99,deBetal00,Hanetal00,Haletal02,Masetal02}.  Theorists discuss open-Universe and alternative structure-formation models, such as topological defects, with {\\it far} less frequency than they did just three years ago.  Historically, when experimental breakthroughs confirm a particular theoretical paradigm and eliminate others, progress can be made at the edges---i.e., precision tests of the new standard model.  In the case of inflation, a number of important questions should be addressed.  For example, what is the physics responsible for inflation?  What is the energy scale of inflation? In particular, we really do not understand why the simple slow-roll model of inflation---really no more than a toy model---works so well.  Might deviations from the simplest model expected in realistic theories lead to small deviations from the canonical predictions of inflation?  For example, is the density of the Universe precisely equal to the critical density?  Are there deviations from scale invariance on small distance scales that arise as a consequence of the end of inflation?  Might there be some small admixture of entropy perturbations in addition to the predominant adiabatic perturbations?  Are there small deviations from Gaussian initial conditions?  A variety of forthcoming CMB experiments will test the flatness of the Universe with additional precision and determine the primordial spectrum of perturbations with increasing accuracy. CMB experiments and galaxy surveys and weak-lensing maps that determine the mass distribution in the Universe today will test Gaussian initial conditions.  Our understanding of galactic substructure may shed light on the end of inflation. Experimentalists are beginning to contemplate programs to detect the unique polarization signature due to inflationary gravitational waves.  Here, I briefly review several new probes of possible relics of inflation; namely inflationary gravitational waves, non-Gaussianity, and galactic substructure. ",
        "conclusions": ""
    },
    "0209/astro-ph0209429_arXiv.txt": {
        "abstract": "We are investigating the hypothesis that Compact High--Velocity Clouds (CHVCs) are the left-over building blocks of Local Group galaxies. To this end, we are searching for their embedded stellar populations using FORS at the VLT. The search is done with single-star photometry in V and I bands, which is sensitive to both, young and old, stellar populations. Five CHVCs of our sample have been observed so far down to I=24. We pointed the VLT towards the highest HI column density regions, as determined in Effelsberg radio data. In an alternate approach, we searched 2MASS public data towards those 5 CHVCs down to K=16. While the VLT data probe the central regions out to distance moduli of about 27, the 2MASS data are sensitive to a population of red giant stars to distance moduli of about 20. The 2MASS data, on the other hand, cover a much wider field of view than the VLT data (radius of 1 degree versus FORS field of 6.8 arcmin). We did not find a stellar population intrinsic to the CHVCs in either data. In this paper,  we illustrate our search methods. ",
        "introduction": "Recent cold dark matter simulations of the formation and evolution of galaxies predict the existence of a significantly higher amount of substructure around big galaxies like the Milky Way \\cite{KLY}, \\cite{MOO} than observed in the form of dwarf galaxies \\cite{MAT}. One solution for this so-called dwarf galaxy crisis could be that the predicted subhalos have been overlooked observationally and are hidden among the population of Compact High-Velocity Clouds (CHVC). Blitz et al. (1999) suggested that isolated CHVCs might be the leftover building blocks predicted in the CMD scenario with mean distance of about 1~Mpc. Braun \\& Burton (1999, 2000) identified an intial catalog of 65 CHVCs. We here present deep optical VLT imaging and 2MASS archival studies of five CHVCs to test them for the presence of a stellar population.  \\begin{figure}[] % \\centerline{ \\epsfig{file=hopp-u.fig1.eps,width=0.8\\textwidth} } \\caption[]{I, V-I CMDs of 4 CHVC fields. The stellar populations in all fields resemble a galactic, high latitude field population. There is no trace of either a young, or an old stellar population intrinsic to any of the CHVCs. In the lower right panel, Giradi et al. (2000) isochrones (1/5 solar) are overplotted for 10~Myr, 100~Myr, 1~Gyr, 10~Gyr and a distance of 1~Mpc to indicate the expected range of the intrinsic stellar populations. } \\label{hoppu.FIG1} \\end{figure}  \\begin{figure}[] % \\centerline{ \\epsfig{file=hopp-u.fig2.eps,width=0.8\\textwidth} } \\caption[]{I, V-I CMD of the field of HIPASS~J1712-64. Globular cluster ridge-lines of Da Costa \\& Armandroff (1990) are overplotted. } \\label{hoppu.FIG1} \\end{figure} ",
        "conclusions": "We conclude that the observed compact clouds do not host an intrinsic stellar population. Our conclusions agree with and extend those reported by Simon \\& Blitz (2002), who did not detect stars in CHVCs on processed POSS scans."
    },
    "0209/astro-ph0209103_arXiv.txt": {
        "abstract": "Current layered accretion models neglect the properties of the ``dead zone''. However, as argued here from simple considerations, the thickness of this zone is a critical quantity when the disc is in hydrostatic equilibrium. It controls not only the structure of the superficial, active layers, but also the mid-plane density and the total disc mass, and should therefore be introduced in models  of that kind, steady or not. But in the absence of intrinsic heating, the dead zone must have a tiny size which, given the non-stationary and turbulent character of the global flow, makes very likely its mixing together with the two active layers. ",
        "introduction": "Gammie (1996) proposed a two-phase accretion scenario for T Tauri discs to cure the expected inefficiency of the magneto-rotational instability (hereafter, MRI) in the cold, outer regions of the disc (beyond about $0.1$ AU from the central proto-star) due to the low abundance of electrons. In this model, only the superficial layers of the outer disc are made ``active'' thanks to incoming interstellar cosmic rays (that can reactivate the MRI), leaving a non-accreting ``dead zone'' around the equatorial plane. Interestingly, the so-called layered accretion disc model makes two major predictions: i) an infrared excess (a common feature in the spectrum of T Tauri stars; e.g. Bertout 1989) caused by a positive, radial gradient of the disc accretion rate, and ii) some ability to develop accretion bursts (the admitted interpretation for FU Orionis events; e.g. Hartman, Kenyon 1996; Kley, Lin 1999) due to mass accumulation in the outer regions. Some aspects of layered accretion have recently been investigated : evolution of the solar nebula (Stepinski 1999), occurrence of eruptive events through time-dependent simulations (Armitage et al. 2001), linear stability properties (Reyes-Ruiz 2001), possible applications to accretion in active galactic nuclei (Menou, Quataert 2001) and in binaries (Menou 2002), and the ionization of accreted material (Fromang et al. 2002). As outlined by these authors, there are still many uncertainties in such a toy-model. Apart from the hypothesis (we shall not discuss it here) that the source of angular momentum transport is only the MRI (see Stone et al. 2000, and references therein), the constraint imposed on the surface density $\\Sigma_{\\rm a}$ of the active layers (to be constant in space and time) appears as the strongest assumption. Misguidedly, the subsequent derivation of steady state profiles for the temperature, density, and disc thickness violates mass conservation (see also Menou 2002), or \\begin{equation} 4 \\pi R \\partial_t \\Sigma_{\\rm a} - \\partial_R \\dot{M}_{\\rm a} \\ne 0 \\end{equation} due to the non vanishing gradient of the total accretion rate, $\\dot{M}_{\\rm a}$. Another difficulty which is outlined here concerns the properties of the dead zone. None of the existing models have yet accounted for the structure of this zone. We argue here that it plays a crucial role. In particular, we demonstrate from simple arguments that it is not correct to find any reliable solution to the problem of layered accretion (steady or not) without specifying the thickness of the dead zone, either in a fully {\\it ad-hoc} manner (i.e. by hand), or by considering explicitly its structure through coupled equations, as soon as the disc is in hydrostatic equilibrium. Finally, it turns out that this zone should be mixed due to its small vertical extent. ",
        "conclusions": "Despite its ``passive'' role with respect to the assumed mechanism of angular-momentum transport and heat generation, the dead zone is an essential component of layered accretion-disc models. Whatever the importance of self-gravity, this zone mechanically supports the active layers, and thus determines the global properties of the disc as a whole. Any layered model of that kind, steady or not, where the disc has reached hydrostatic equilibrium (see also Glassgold et al. 2000), must take the dead zone into account, and in particular its thickness, at least in the form of a free parameter. Probably, a model accounting for more physical mechanisms (and containing more degrees of freedom) could modify some issues raised here, and even might produce steady state solutions. However, it appears clearly that, in the absence of internal heating, the dead zone must have a small thickness with respect to the active layers; otherwise, the disc would be extremely massive. Also, without any noticeable extent, the dead zone should not survive between the two turbulent layers, but mix with them. Besides, observations of discs around young stars indicate that they have a mass not in excess of the central mass (e.g. Calvet et al. 2000) as well as an outward decrease of the surface density (Beckwith et al. 1990; Dutrey et al. 1998), both properties of which are not compatible with the layered accretion model.\\\\  It is a pleasure to thank my colleagues S. Collin, D. Gautier, F. Hersant, D. Richard, and J.-P. Zahn for stimulating discussions. I am grateful to the referee for important suggestions and comments to improve the paper."
    },
    "0209/astro-ph0209472_arXiv.txt": {
        "abstract": "Subdwarf B (sdB) stars are thought to be core helium burning stars with low mass hydrogen envelopes. In recent years it has become clear that many sdB stars lose their hydrogen through interaction with a binary companion and continue to reside in binary systems today. In this paper we present the results of a programme to measure orbital parameters of binary sdB stars. We determine the orbits of 22 binary sdB stars from 424 radial velocity measurements, raising the sample of sdBs with known orbital parameters to 38. We calculate lower limits for the masses of the companions of the sdB stars which, when combined with the orbital periods of the systems, allow us to discuss approximate evolutionary constraints. We find that a formation path for sdB stars consisting of mass transfer at the tip of the red giant branch followed by a common envelope phase explains most, but not all of the observed systems. It is particularly difficult to explain both long period systems and short period, massive systems. We present new measurements of the effective temperature, surface density and surface helium abundance for some of the sdB stars by fitting their blue spectra. We find that two of them (PG0839+399 and KPD1946+4340) do not lie in the Extreme Horizontal Branch (EHB) band indicating that they are post-EHB stars. ",
        "introduction": "Subdwarf B (sdB) stars can be identified with models for Extreme Horizontal Branch (EHB) stars. The surface gravities and temperatures of sdB stars suggest that they have helium cores of mass $\\sim 0.5\\,\\msun$ and thin hydrogen envelopes of mass $\\leq 0.02\\,\\msun$ \\cite{h84,s94}. A recent asteroseismological study of an sdB star results in a value for its mass of 0.49$\\pm$0.02\\,$\\msun$ \\cite{b01}. Several evolutionary scenarios have been proposed to explain the formation of sdB stars, in particular the loss of the hydrogen envelope. Evolution within a binary star is an effective method for envelope removal, and yet it is hard to see why this should have happened to a horizontal branch star since it would have been much larger during its preceding red giant stage. A solution to this problem was presented by D'Cruz et al.\\ \\shortcite{dc96} who found that if a red giant star with a degenerate helium core loses its hydrogen envelope when it is within $\\sim$0.4\\,magnitudes of the tip of the red giant branch, the core can go on to ignite helium, despite the dramatic mass loss, and may then appear as an sdB star. The advantage of this model is that it very nicely explains the masses of sdB stars as a consequence of the core mass at the helium flash. D'Cruz et al.\\ \\shortcite{dc96} supposed that mass loss occurred because of an enhancement of the stellar wind, but it could as well have been driven by binary interaction.  If sdB stars do form within binary systems and if they still have their companions, then the companions must be low-mass main-sequence stars or compact stellar remnants to avoid outshining the sdB star. If so, it is probable that in many cases the companions were unable to cope with the mass transferred from the sdB progenitor and a single ``common'' envelope formed around the two stars. Driving off such envelopes drains energy and angular momentum from the binary orbit, which as a result becomes much smaller than it was at the start of mass transfer \\cite{w84}. It is therefore possible that many sdB stars are now members of close binary systems. Maxted et al.\\ \\shortcite{m01} found exactly this, discovering 21 binary sdB stars in a sample of 36, suggesting, after allowance for detection efficiency, that some two-thirds of all sdB stars are in short period binary systems ($P \\la 10\\,\\mathrm{d}$). The other third seems to be made up a combination of long period binary stars that avoided a common envelope phase \\cite{gls00} and apparently single sdB stars.  If D'Cruz et al.'s \\shortcite{dc96} model of the formation of sdB stars is correct, then the stage immediately prior to mass transfer in binary sdBs is well defined. This, together with the fact that the binary does not have enough time to change its orbital period significantly following its emergence from the common envelope, makes the sdB stars a superb population for testing models of the common envelope phase. Moreover, the detection of sdB binary stars is not compromised by the strong and poorly understood selection effects that plague other populations of close binary stars, such as the cataclysmic variable stars. The properties of sdB binaries (e.g. their orbital period distribution) can be compared fairly directly with the results of binary population synthesis codes and are therefore a strong test of population synthesis models for binary stars.  Following on from the detection of many binary stars by Maxted et al. \\shortcite{m01}, we started a project to measure their orbits. The orbit of one of the new binary stars has been presented in Maxted et al.\\ \\shortcite{m02a}. In this paper we present the orbits of a further 22 systems.  We then consider the implications of the known sdB binary stars for their evolution. It should be noted that our sample is biased against sdB stars with G/K-type companions as the majority of our stars were selected from the PG survey which excludes most stars that show a Ca{\\sc ii} H-line. ",
        "conclusions": "We have confirmed the binary nature of 22 subdwarf B stars and have measured their orbital parameters. This work increases the sample of sdB binaries with known orbital parameters to 38. The observations extend over several months allowing us to detect orbital periods of the order of tens of days, longer than any previously measured.  We have measured T$_{\\rm eff}$, $\\log g$, $\\log (\\rm He/H)$ for the sdBs where previous measurements of these quantities did not exist. When we place the results in the T$_{\\rm eff}$-$\\log g$ plot we find that two out of the 22 sdBs are post-EHB stars.  The large range of orbital periods and companion masses is a challenge to simple theories for the formation of sdB stars. Although binary-induced mass-loss at the tip of the red giant branch is able to explain most systems, it appears unlikely to be the only formation route. Full population synthesis will be needed to establish the viability of alternative paths.  Amongst sdB stars with known orbits, those with low-mass main-sequence or brown dwarf companions have particularly short periods. It seems likely that a fraction of such systems, particularly those of very low companion mass, may not have survived the common envelope phase. We suggest that these could now be the single sdB stars."
    },
    "0209/astro-ph0209191_arXiv.txt": {
        "abstract": "A degeneracy in strong lens model is shown analytically. The observed time delays and quasar image positions might {\\it not} uniquely determine the concentration and the extent of the lens galaxy halo mass distribution. Simply hardwiring the Hubble constant ($H_0$) and the cosmology ($\\Omega, \\Lambda$) to the standard $\\Lambda$CDM cosmology values might {\\it not} fully lift this degeneracy, which exists rigourously even with very accurate data.  Equally good fits to the images could be found in lens mass models with either a mostly Keplerian or a flat rotation curve. This degeneracy in mass models makes the task of getting reliable $H_0$ and $\\Lambda$ from strong lenses even more daunting. ",
        "introduction": "One of the promises of gravitational lenses is to measure the Hubble constant $H_0$ (Refsdal 1964) from the observed time delays among the images of a variable background quasar source lensed by a foreground galaxy.  Given a model for the spatial distributions of the stars and dark matter in the lens galaxy, the time delay $\\Delta t_{\\rm obs}$ multiplied by the speed of light $c$ is simply proportional to the absolute distances to the lens and the source, hence $c\\Delta t_{\\rm obs}$ scales with the size of the universe $c/H_0$.  While the time delays are now routinely measured for many systems (see Schechter 2000), a reliable determination of $H_0$ has been hampered to some extent by the intrinsic degeneracy in models of the dark matter potential of the lens (Williams \\& Saha 2000; Saha 2000; Zhao \\& Pronk 2001).  The general trend is that a model with a dense dark matter halo gives a small $H_0$ with \\begin{equation}\\label{trend} H_0 \\Delta t_{\\rm obs} \\propto \\left[2\\Sigma_{\\rm crit}(z_l,z_s)-\\Sigma_{\\rm *}(R_E)-\\Sigma_{\\rm h}(R_E)\\right], \\end{equation} where $\\Sigma_{\\rm crit}(z_l,z_s)$ is the critical density for the lens at redshift $z_l$ and the source at $z_s$, and $\\Sigma_{\\rm *}(R_E)$ or $\\Sigma_{\\rm h}(R_E)$ is the typical surface density of luminous or dark matter at within the Einstein ring $R_E$ (Falco, Gorenstein, \\& Shapiro 1985; Kochanek 2002).  Given that the value of $H_0$ is now well constrained by other independent methods, such as the HST Key Project (Freedman et al. 2001), it is interesting to reverse the angle of the question, and use equation$~$(\\ref{trend}) to put more stringent constraint on the dark matter potential of the lens. More specifically in this {\\sl Letter}, we would like to ask the question: how narrow is the allowed parameter space for the lens dark halo which is consistent with a given set of images, time delays, cosmology and $H_0$? In the interest of clarity, we will consider only analytical lens models with a simplified geometry for a hypothetical image and lens system. We believe our arguments should apply qualitatively to real galaxy lenses as well, and a more detailed application to the quadruple system PG1115+080 is given in a follow-up paper (Zhao \\& Qin 2002). ",
        "conclusions": "It is possible to construct many very different models with positive, smooth and monotonic surface densities to fit the image positions. There are also no extra images.  These models fit the same images, time delay, $H_0$ and cosmology.  Some fit the same lens light profile and image flux ratio as well.  Hence the models are virtually {\\sl indistinguishable} from lensing data.  There are severe degeneracies in inverting the data of a perfect Einstein cross to the lens models, even if given the Hubble constant and cosmology.  These rigourous findings with analytical models are also consistent with earlier numerical models of Saha \\& Williams (2001) and semi-analytical models of Zhao \\& Pronk (2001).  Among the acceptable models the rotation curve can be Keplerian or flat (Fig.~\\ref{vk}a), so lensing data plus $H_0$ cannot uniquely specify the lens mass profile. Among models in the literature, isothermal models and other simple smooth models of dark matter halos of gravitational lenses often predict a dimensionless time delay $H_0 \\Delta t_{\\rm obs}$ much too small (e.g., Schechter et al. 1997) to be comfortable with the observed time delays $\\Delta t_{\\rm obs}$ and the widely accepted value of $H_0 \\sim 70\\kmsMpc$.  Naively speaking the high $H_0$ suggests a strangely small halo as compact as the stellar light distribution (Kochanek 2002). But our analytical models suggest that there are still many other options.  The high $H_0$ implies that $\\kappa$ is small $\\sim 0.1-0.2$ at the images, but this does not necessarily imply a rapidly falling density. A high $H_0$ does not necessarily mean no dark halo, and models with a flat rotation curve does not always yield a small $H_0$ (e.g., compare lensIII and lensIV in Figure~\\ref{vk}b, both satisfy $h_0=0.7$). We also comment that it will be difficult to determine the cosmology from strong lensing data alone because the non-uniqueness in the lens models implies that the combined parameter $\\omega(H_0,\\Omega,z_l,z_s)$ is poorly constrained by the lensing data, even if $H_0$, $z_l$ and $z_s$ are given.  We thank the referee P. Saha for insightful comments on the cause of the degeneracy. This work was supported by the National Science Foundation of China under Grant No. 10003002 and a PPARC rolling grant to Cambridge.  HSZ and BQ thank the Chinese Academy of Sciences and the Royal Society respectively for a visiting fellowship, and the host institutes for local hospitalities during their visits."
    },
    "0209/astro-ph0209228_arXiv.txt": {
        "abstract": "Fireballs of gamma-ray bursts are partially made of free neutrons. Their presence crucially changes the mechanism of the fireball deceleration by an external medium. At late stages of the explosion, neutrons fully decouple from ions and coast with a constant Lorentz factor $\\Gamma_n$. As the ion fireball decelerates, the neutrons form a leading front. This front gradually decays, leaving behind a trail of decay products mixed with the ambient medium. The kinetic energy of the decay products far exceeds the medium rest-mass energy, and the trail has a Lorentz factor $\\gamma\\gg 1$ at radii $R<\\Rtr\\approx 10\\Rb\\approx 10^{17}$~cm, where $\\Rb\\approx 10^{16}(\\Gamma_n/300)$~cm is the mean radius of neutron decay. The ion fireball sweeps up the trail material and drives a shock wave in it. Thus, observed afterglow emission is produced in the neutron trail. It can naturally re-brighten or display a spectral transition at $R\\approx\\Rtr$ where the impact of neutrons turns off. Absence of any neutron signatures would point to an extremely low baryon loading of the fireballs and a strong dominance of a Poynting flux. ",
        "introduction": "The presence of neutrons in gamma-ray bursts (GRBs) and their possible role for the observed emission was realized in the past few years (Derishev, Kocharovsky, \\& Kocharovsky 1999a,b; Bahcall \\& M\\'esz\\'aros 2000; M\\'esz\\'aros \\& Rees 2000; Fuller, Pruet, \\& Abazajian 2000; Pruet \\& Dalal 2002). In an accompanying paper, we study in detail the nuclear composition of GRB fireballs and show that the presence of neutrons among the ejected baryons is practically inevitable (Beloborodov 2003, Paper~1). One implication is an observable multi-GeV neutrino emission from inelastic neutron-ion collisions in the fireball. Here, we focus on a different aspect. We show that the neutrons have a dramatic impact on the explosion dynamics at radii as large as $10^{17}$~cm and propose a novel mechanism for GRB afterglow emission.  Let us remind what happens in a relativistic explosion without neutrons (see M\\'esz\\'aros 2002 for a review). The ejected fireball with mass $\\Mej$ and Lorentz factor $\\Gej$ sweeps up an ambient medium and gradually dissipates its kinetic energy. The dissipation rate peaks at a characteristic ``deceleration'' radius $\\Rdec$ where half of the initial energy is dissipated. This radius corresponds to swept-up mass $\\mdec=\\Mej/\\Gej$. Further dynamics is described by the self-similar blast wave model of Blandford \\& McKee (1976). How does this picture change in the presence of neutrons? ",
        "conclusions": "Dynamics of neutron-fed explosions is a clean physical problem: existence of neutrons in GRBs and their mean lifetime of 15~min is all we needed to construct the model. We focused here on the simple case where all neutrons have equal Lorentz factor $\\Gn$, and data may require a multi-shell picture with variable $\\Gn$. This extension could affect our model in case of high variations $\\Delta\\Gn>\\Gn$ (though at large $R$, where neutrons die out exponentially, it is sufficient to consider a highest $\\Gn^{\\rm max}$ as slower neutrons have shorter lifetimes and their population is smaller by $\\exp[-\\Gn^{\\rm max}/\\Gn]$). Another simple extension is beamed ejecta. Then at late stages the ion fireball spreads laterally while the neutron beaming remains constant.  In contrast to dynamics, emission is really complicated. In general, radiation from collisionless blast waves is not derived from first principles, and neutrons do not make the problem simpler. The afterglow emission is believed to be synchrotron, and it depends on poorly understood generation of magnetic field and electron acceleration. The standard model without neutrons relies on the field generation in the shock by the two-stream instability (Sagdeev 1966, Medvedev \\& Loeb 1999, Gruzinov 2001). The shock front is, however, extremely thin ($\\delta\\sim c/\\omega_p\\ll R/\\Gamma^2$), and the postshock field decays quickly. The model needs a significant remnant field in an extended layer behind the shock, which is uncertain. This problem is alleviated in neutron-fed explosions. Here, the leading neutron front is an additional dissipation region maintained in a turbulent state with generated magnetic fields. The N-shell has thickness $\\Delta$ comparable to $R/\\Gn^2$. It can produce a significant synchrotron radiation by itself\\footnote{The N-shell emission decays exponentially on observed timescale $\\Rb/c\\Gn^2=\\taub/\\Gn\\sim 1-10$~s and it can be related to smooth (fast-rise-exponential-decay) pulses in prompt GRBs.} and leave behind remnant fields and hot plasma to be used by the ensuing shock wave for afterglow emission. We emphasize differences from a customary external shock: the neutron-trail shock propagates in a relativistically moving, dense, hot, and possibly magnetized medium. The fate of magnetic fields in such shocks and the resulting emission is an interesting issue for a future study.  The neutron impact ceases at $\\Rtr\\approx 10^{17}$~cm, which can leave an imprint on the observed afterglow. For example, the shock dissipation can have a bump (Fig.~1), and a spectral transition is also possible. The arrival time of radiation emitted at $\\Rtr$ is $\\approx\\Rtr/2\\Gamma^2c$ (counted from the arrival of first $\\gamma$-rays). It may be as long as 30 days or as short as a few seconds, depending on the fireball Lorentz factor $\\Gamma(\\Rtr)$. Remarkably, $\\Rtr$ is almost independent of the ambient medium, and its observational signature would give information on $\\Gamma(\\Rtr)$. Recent early observation of a GRB afterglow (GRB~021004) discovered an interesting re-brightening at $10^3$~s which would correspond to $\\Gamma(\\Rtr)\\approx 30(\\Gn/300)^{1/2}$. Also, we do not exclude a possible relevance of neutrons to the 20~day bumps observed in a few GRBs, as the time coincides with $\\Rtr/c$.  Neutron signatures should be absent if the fireball is strongly dominated by a Poynting flux and has extremely low baryon loading. Then the neutron component decouples early, with a modest Lorentz factor $\\Gn$, and decays at small radii. The upper bound on $\\Gn$ due to decoupling is $\\Gn\\approx 300(\\dM_\\Omega/10^{26})^{1/3}$ where $\\dM_\\Omega$ [g/s] is the mass outflow rate per unit solid angle of the fireball (see \\S~4.1 in Paper~1).  We focused here on the neutron front and did not account for the $\\gamma$-ray precursor that impacts the blast wave dynamics at $R<\\Racc=0.7\\times 10^{16}(E_\\gamma/10^{53})^{1/2}$~cm, where $E_\\gamma$ [erg] is the isotropic energy of the GRB (see Beloborodov 2002 and refs. therein). The analysis in this Letter is strictly valid for afterglows emitted at $R>\\Racc$. Then the radiation-front effects, including the gap opening at $R<0.3\\Racc$, occur at smaller radii, and apply to an earlier afterglow. For a dense medium, where $\\Rdec<\\Racc$, effects of the neutron and $\\gamma$-ray fronts should be studied together."
    },
    "0209/gr-qc0209090_arXiv.txt": {
        "abstract": "We compute the spectrum of normalizable fermion bound states in a Schwarzschild black hole background.  The eigenstates have complex energies.  The real part of the energies, for small couplings, closely follow a hydrogen-like spectrum.  The imaginary parts give decay times for the various states, due to the absorption properties of the hole, with states closer to the hole having shorter half-lives.  As the coupling increases, the spectrum departs from that of the hydrogen atom, as states close to the horizon become unfavourable.  Beyond a certain coupling the $1S_{1/2}$ state is no longer the ground state, which shifts to the $2P_{3/2}$ state, and then to states of successively greater angular momentum.  For each positive energy state a negative energy counterpart exists, with opposite sign of its real energy, and the same decay factor.  It follows that the Dirac sea of negative energy states is decaying, which may provide a physical contribution to Hawking radiation. ",
        "introduction": "Quantum theory in a black hole background has been extensively studied by many authors.  Detailed discussions of this problem are contained in the books by Birrell \\& Davies~\\cite{bir-quant} and Chandrasekhar~\\cite{cha83}, and the review paper by Brout \\textit{et al.}~\\cite{bro95}.  Much of the attention in this work is focussed on the wave equation and its scattering properties.  Detailed studies of the Dirac equation in a black hole background are less common. Indeed, the lowest order scattering cross section for a fermion in a black hole background has only recently been computed~\\cite{DL2002,LD-erice01}.  In this paper we investigate another previously neglected aspect of quantum mechanics in a black hole background.  This is the existence of the bound state spectrum for particles orbiting a spherically--symmetric point source.  That is, we study the gravitational analogue of the hydrogen atom orbitals.   There has been strangely little effort devoted to the study of the bound state spectrum, despite the fundamental importance of the electromagnetic analogue.  But it is clear that these states must exist --- how else can one provide a quantum description of a particle in orbit around a black hole?  These states must also be essential in the quantum description of the capture process.  The problem was discussed in 1974 by Deruelle and Ruffini~\\cite{der74}, who described the existence of resonance states in the Klein--Gordon equation. Further significant contributions were made in a series of papers by Gaina and coauthors~\\cite{gai87,gai88,gai92}.  These papers give various analytic expressions for the real and imaginary parts of the energy in a series of limiting cases.   Much of the study of quantum mechanics in a black hole background has focussed on the related, though distinct, problem of finding the quasi-normal mode spectrum.  Quasi-normal modes are purely ingoing at the horizon, and outgoing at infinity~\\cite{cha83}.  These boundary conditions produce a spectrum of eigenstates with complex-valued energies.  The significance of these quasi-normal modes comes from their use in describing black hole oscillations.  But the boundary condition at infinity implies that these modes are not normalizable, so cannot represent bound states.  The problem of interest here is to find these bound states, so we seek solutions which are purely ingoing at the horizon, and which fall off exponentially at infinity.    For a particle of mass $m$ in the field of a black hole of mass $M$ the dimensionless coupling strength is defined by \\begin{equation} \\alp = \\frac{mM}{m_p^2} \\end{equation} where $m_p$ is the Planck mass.  In this paper we compute the fermion bound state spectra for $\\alpha$ in the range $0\\cdots 6$.  If the bound particle is assumed to be an electron, this range corresponds to black holes of masses up to $1\\times10^{15}$kg, which is the scale appropriate for primordial black holes formed in the early universe. Computing the energy spectrum is more complicated than the hydrogen atom case for two main reasons.  The first is that the radially-separated Dirac equation contains three singular points, only two of which are regular.  There is no special function theory appropriate for the study of such equations, so we have to resort to a range of numerical techniques to find the spectrum.  The second problem is that the singularity at the centre of a Schwarzschild black hole acts as a current sink.  All normalizable states must therefore decay in time, and we must search for eigenstates over the two-dimensional space of complex energies.  The states we construct therefore all have a finite half-life, so can be viewed as resonance states.  The interpretation of these states is discussed in section~\\ref{disc}.   Despite these difficulties, the problem can be tackled numerically, and we present a range of results for the real and imaginary parts of the energy.  These are sufficient to predict how the spectrum will behave for larger values of the coupling constant.  The first result, which is entirely to be expected, is that the orbitals become increasingly tightly bound as the coupling increases.  It follows that, for a given state, the energy will initially decrease with $\\alp$, but will eventually turn round and start increasing as the particle spends too much time inside the classical radius of minimum energy.  States with higher angular momentum then become energetically favourable as $\\alp$ increases.  For example, we show that beyond $\\alp\\approx 0.6$ the $1S_{1/2}$ state is no longer the ground state. While the real part of the energy behaves in quite a complicated fashion, the imaginary part, which controls the decay rate, simply increases in magnitude.  This is also as one would expect.  The closer the orbital density is to the singularity, the greater the probability of capture.    We start by discussing the Dirac equation in a Schwarzschild background in an arbitrary gauge.  This is helpful in establishing a range of gauge-invariant results.  In particular, the energy conjugate to time translation symmetry is confirmed to be a gauge invariant quantity.  This is important in order to guarantee that the quantity is a physical observable.  We next establish the behaviour of the wavefunction around the horizon, which is sufficient to establish that the states must decay exponentially with time.  We then turn to a specific choice of gauge that is well-suited to numerical solution. We solve the equations by simultaneously integrating out from the horizon and in from infinity.  We then vary the energy to ensure that the solutions match at some finite radius.  This process guarantees that we find a global, normalizable bound state.  A set of spectra are obtained, and the density is plotted for a range of states.  Decay rates and expectation values for the distance are also presented.  We end with a discussion of the significance of these bound states, and the possible physical processes that they may generate.  Except where stated otherwise, natural units with $G=\\hbar=c=1$ are assumed throughout.  We employ a spacetime metric with signature $-2$. ",
        "conclusions": "\\label{disc}   We have demonstrated the existence of a complicated spectrum of bound states for a quantum fermion in a black hole background.  Each state represents a spatially-normalizable solution to the Dirac equation in a Schwarzschild background.  The fact that time-separable solutions exist is simply established in one particular gauge, which casts the equation in a Hamiltonian-like form.  A study of the behaviour of the wavefuntion under gauge transformations show that time-separability is a gauge-invariant feature.  The spectrum itself is determined by boundary conditions applied at the horizon and at infinity.  These alone are sufficient to imply the existence of an imaginary (decay) contribution to the energy.  The physical explanation for this is provided by the singularity, which acts as a current sink.    The qualitative features of the spectrum can be understood in terms of simple semi-classical models, but a full quantitative understanding only seems possible through a mixture of computational methods.  The work in this paper can clearly be extended in a number of ways.  We have only plotted the spectrum at low coupling strengths of $\\alpha \\sim 1$, but astrophysical values can be far larger than this, with $\\alpha \\sim 10^{15}$ for solar mass black holes.  For larger $\\alpha$, the ground state will be one of high angular momentum.  In this regime the spectrum will be quite different to that of the hydrogen atom.  One important question is precisely how great a binding energy can be achieved.  In figure~\\ref{higher_kappa} we see that at around $\\alp=5$ we are achieving total energies of $0.88mc^2$, which is significantly lower the classical value of $0.94mc^2$.  This suggests that more energy may be available in accretion processes than is traditionally thought.   As well as increasing $\\alp$, it would be of considerable interest to repeat this work for the case of a Kerr black hole.  In this respect a useful start has been made in~\\cite{D00-kerr}, where the Kerr solution is written in a form which generalises the `Newtonian' gauge employed in this paper.  The calculations for the Kerr case are more complicated, however, because the angular separation constants are energy-dependent~\\cite{cha83,gai88}.  There are also signs that the horizon structure of the Kerr solution will complicate the fairly straightforward picture presented here.  The problem can be seen by analysing behaviour in a Reissner--Nordstrom background using the setup of this paper.  For this case we find that the regular solutions at the outer horizon do not match onto regular solutions at the inner horizon.  So quantum mechanics predicts that the probability density will pile up around the inner horizon in a similar manner to the classical picture.  Behaviour of this type is inevitable, as the Reissner--Nordstrom singularity does not act as a sink, and the Hamiltonian is Hermitian.  Since the current must still be inward pointing at the outer horizon, the probability density has to pile up somewhere.  It seems likely that a similar picture holds for the Kerr solution, but detailed calculations are required to confirm this.   The energy spectra presented in this paper raise a number of fundamental issues, which demonstrate the limitations in our current understanding of the interaction between gravity and quantum theory. It is unusual to obtain a decay law from quantum mechanics without some form of approximation.  That we do so in the present case is a consequence of the fact that the system is open.  States are allowed to decay onto the singularity, but no accompanying emission is considered.  A complete treatment of the problem as a closed system would require a quantum theory of the singularity,  and such a theory does not yet exist.   The decay rates represent one feature of the quantum-mechanical description of the capture process.  But, as well as decay, the quantum description of a particle falling onto the singularity of a black hole can involve a series of quantum jumps to lower energy orbits.  This quantum description alters the physics of the process quite dramatically from the classical picture.  As the particle undergoes a series of transitions we expect that it should radiate, which does not happen classically.  Quite what form this radiation should take (electromagnetic, gravity waves?) is unclear.  Also, as a transition takes place we should keep careful track of the evolution of the matter stress-energy tensor to tell us where the radiated energy is concentrated.  A related problem this exposes is that we have not considered back reaction on the gravitational field, which could alter this picture.   The quantum treatment of a particle in a gravitational field exhibits a curious anti-parallelism with the electromagnetic case.  In classical electrodynamics a charged particle in orbit around a point source should radiate, making atoms unstable.  This problem is resolved by quantum mechanics, which predicts the existence of stable, non-radiating bound states.  The reverse is true of gravitation. Classically, a particle can orbit a black hole in a geodesic outside the horizon, and such an orbit is stable.  But quantum theory changes this, and states that no totally stable orbits exist, due to the finite probability of the particle finding itself inside the horizon and ending on the singularity.  While the time-scales involved in these decays may be of limited interest astrophysically, such processes are clearly of fundamental importance in understanding the interplay between quantum theory and gravitation.    A final point to raise here is that the spectrum of real energies derived here has a mirror image of negative energy bound states.  Each of these negative energy states also has a finite lifetime.  If we model the vacuum in terms of a Dirac sea of filled negative energy states, we must include the bound states as well as the free continuum.  It then follows that the vacuum itself is decaying --- the black hole is sucking in the vacuum.  Such a loss of negative energy states is seen as a creation of positive energy modes, which could contribute to Hawking radiation.  This contribution appears to have been neglected in previous calculations, which concentrate only on the scattered states~\\cite{bir-quant}.  It is well known in calculations of the Lamb shift, for example, that ignoring the bound states in the calculation gives the wrong answer~\\cite{itz-quant}.  It would be of great interest to assess the contribution played by bound states to the gravitational analogue of this process."
    },
    "0209/astro-ph0209534_arXiv.txt": {
        "abstract": "\\noindent We have obtained deep optical, long-slit spectrophotometry of the Galactic {\\hii} regions M\\,17, NGC\\,3576 and of the Magellanic Cloud {\\hii} regions 30~Doradus, LMC~N11B and SMC~N66, recording the optical recombination lines (ORLs) of {\\cii}, {\\nii} and {\\oii}.  A spatial analysis of 30 Doradus is performed, revealing that the forbidden-line {\\foiii} electron temperature is remarkably constant across the nebula. The forbidden-line {\\opp}/{\\hp} abundance mapped by the {\\foiii} $\\lambda$4959 collisionally excited line (CEL) is shown to be consistently lower than the recombination-line abundance mapped by the {\\oii} V\\,1 multiplet at 4650\\,\\AA. In addition, the spatial profile of the {\\cpp}/{\\opp} ratio derived purely from recombination lines is presented for the first time for an extragalactic nebula. Temperature-insensitive ORL {\\cpp}/{\\opp} and {\\npp}/{\\opp} ratios are obtained for all nebulae except SMC~N66. The ORL {\\cpp}/{\\opp} ratios show remarkable agreement within each galactic system, while also being in agreement with the corresponding CEL ratios. The disagreement found between the ORL and CEL {\\npp}/{\\opp} ratios for M~17 and NGC~3576 can be attributed to the {\\nii} V~3 and V~5 ORLs that were used being affected by fluorescent excitation effects.  For all five nebulae, the {\\opp}/{\\hp} abundance derived from multiple {\\oii} ORLs is found to be higher than the corresponding value derived from the strong {\\foiii} $\\lambda\\lambda$4959, 5007 CELs, by factors of 1.8--2.7 for four of the nebulae.  The LMC~N11B nebula exhibits a more extreme discrepancy factor for the {\\opp} ion, $\\sim$5.  Thus these {\\hii} regions exhibit ORL/CEL abundance discrepancy factors that are similar to those previously encountered amongst planetary nebulae.  Our optical CEL {\\opp}/{\\hp} abundances agree to within 20-30 per cent with published {\\opp}/{\\hp} abundances that were obtained from observations of infrared fine-structure lines. Since the low excitation energies of the latter make them insensitive to variations about typical nebular temperatures, fluctuations in temperature are ruled out as the cause of the observed ORL/CEL {\\opp} abundance discrepancies. We present evidence that the observed {\\oii} ORLs from these {\\hii} regions originate from gas of very similar density ($<$~3500~{\\cmt}) to that emitting the observed heavy-element optical and infrared CELs, ruling out models that employ high-density ionized inclusions in order to explain the abundance discrepancy. We consider a scenario whereby much of the heavy-element ORL emission originates from cold ($\\leq$~500~K) metal-rich ionized regions. These might constitute halos that are being evaporated from much denser neutral cores. The origin of these metal-rich inclusions is not clear -- they may have been ejected into the nebula by evolved, massive Of and Wolf-Rayet stars, although the agreement found between heavy element ion ratios derived from ORLs with the ratios derived from CELs provides no evidence for nuclear-processed material in the ORL-emitting regions. \\\\  \\noindent {\\bf Key Words:} ISM: abundances -- {\\hii} regions: general -- stars:individual R~139, R~140, P~3157 ",
        "introduction": "In recent years deep spectroscopic studies, coupled with new theoretical results in atomic physics, have cast new light on a well established field of modern astrophysics: the study of elemental abundances in H\\,{\\sc ii} regions (ionized gas clouds marking the birthplaces of stars within the galaxy and in external galaxies) and planetary nebulae (PNe; ionized ejected envelopes of evolved low- to intermediate-mass stars). On the one hand, an accurate knowledge of abundances in galactic and extragalactic H\\,{\\sc ii} regions is of paramount importance for constraining galactic chemical evolution models (e.g. Shields 2002). On the other hand, abundance studies of PNe provide useful constraints on nucleosynthetic theories and our understanding of the late stages of stellar evolution (Kingsburgh \\& Barlow 1994; Henry, Kwitter \\& Bates 2000). Nebular abundances of heavy elements such as C, N, O and Ne, relative to H, have traditionally been derived from observations of strong, and thus easy to measure, collisionally-excited ionic lines (CELs); e.g. C~{\\sc iii}] $\\lambda\\lambda$1906, 1909 {\\fnii} $\\lambda\\lambda$6548, 6584, {\\foiii} $\\lambda\\lambda$4959, 5007 and {\\fneiii} $\\lambda\\lambda$3869, 3967. However, the analysis of observations of weak heavy-element optical recombination lines (ORLs) from PNe, including publications by Peimbert, Storey \\& Torres-Peimbert (1993), Liu et al. (1995a; 2000; 2001b), Luo, Liu \\& Barlow (2001), Garnett \\& Dinerstein (2001), and the recent work of Tsamis et al. (2002a for observations and CEL analysis; 2002b for ORL analysis), have yielded CNONe abundances for PNe that are systematically higher than those derived using the standard CEL method. The most promising explanation for these results, at least for PNe, posits the existence within nebulae of a hitherto unknown, low-temperature component enhanced in heavy elements and emitting mostly in ORLs, intermingled with a hot component of more normal composition from which the CEL emission originates (Liu et al. 2000; Liu 2002a,b; P\\'{e}quignot et al. 2002a, b). The exact nature and origin of the putative ORL-emitting component is currently a matter of intense debate.  The ORL results for PNe become a matter of concern for nebular abundance studies, all the more so because nebular CEL abundances have been long plagued by lingering doubts about their reliability, arising from their exponential sensitivity to the adopted nebular electron temperature and their dependence on the adopted nebular electron density (for lines of low critical density, {\\crd}; Rubin 1989). On the other hand, elemental abundances relative to hydrogen derived from ratios of ORLs [e.g. from the $I$(C~{\\sc ii} $\\lambda$4267)/$I$(\\Hb) intensity ratio for the derivation of {\\cpp}/{\\hp}, as opposed to using the CEL/ORL $I$($\\lambda$1908)/$I$(\\Hb) ratio] are nearly independent of both the adopted temperature and density; this means that abundance determinations employing ORLs should be more accurate than those using CELs, since in real nebulae much of the emission of CELs can be biased towards regions of high electron temperature, as well as towards regions having electron densities less than the critical density of the CEL. As a result, ORL abundance studies of ionized nebulae are now coming to the fore, thanks also to rapid progress in detector technology.  To date, deep abundance studies of PNe (summarised by Liu 2002b) have yielded ORL abundances for C, N, O and Ne which, for the majority (90--95\\%) of nebulae, are typically a factor of 2--3 larger than those obtained from UV, optical or infrared CELs. For the remaining 5-10\\% of PNe, even larger discrepancy factors (5--80) are found between the heavy element abundances derived from ORLs and CELs. The situation for {\\hii} regions is less clear thus far than for PNe, mainly on account of the small number of {\\hii} regions so far observed specifically for the purpose of detecting heavy element ORLs and deriving abundances from them. Peimbert et al. (1993) found ORL O$^{2+}$/H$^{+}$ abundances for M\\,42 and M\\,17 that were a factor of two larger than those found from the [O~{\\sc iii}] optical CELs, while Esteban et al. (1998, 1999) found ORL O$^{2+}$ abundances that were larger than the CEL values by a factor of 1.5 for M\\,42 and a factor of two for M\\,8. Clearly, if {\\hii} regions are generally found to yield heavy element abundances from ORLs that exceed those from CELs by similar factors to those summarised above for PNe and for M\\,8, M\\,17 and M\\,42, then this could have serious implications for our understanding of the chemical evolution of galaxies, which to date has relied to a large extent on CEL abundance analyses of {\\hii} regions located in our own and other galaxies. The current contribution aims to increase the number of {\\hii} regions with ORL abundance analyses, by presenting deep, medium-resolution, long-slit optical spectrophotometry of the bright Galactic {\\hii} regions M\\,17 and NGC\\,3576, and the Magellanic Cloud {\\hii} regions 30~Doradus, LMC~N11B and SMC~N66. These nebulae were selected to be of relatively high excitation for H~{\\sc ii} regions, with O$^{2+}$ being the dominant ion stage of oxygen. In Section~2 we describe our optical spectroscopic observations and present a thorough list of emission line fluxes and dereddened intensities. In Section~3 we describe the extinction corrections and the plasma temperature and density analysis. In Section~4 we present an abundance analysis using optical collisionally excited lines. In Section~5 we present an abundance study using optical recombination lines, discussing the relative intensities of {\\oii} ORLs and complications arising from the presence of bright dust-scattered starlight within the nebulae. In Section~6 high-resolution long-slit spectra of 30~Doradus are used to map the electron temperature and density, and the ionic abundances across the nebular surface. Finally, we discuss the implications of the results from this extensive study in Section~7 and state our conclusions in Section~8.  \\setcounter{table}{0} \\begin{table*} \\centering \\begin{minipage}{135mm} \\caption[Journal of H~{\\sc  ii} regions observations.]{Journal of observations.} \\begin{tabular}{lccccrrc} \\noalign{\\vskip3pt} \\noalign{\\hrule} \\noalign{\\vskip3pt} H {\\sc ii} region  &Date    &$\\lambda$-range  &FWHM   &PA &RA   &DEC      &Exp. \\\\ &(UT)       &(\\AA)       &(\\AA)   &(deg)   &(2000)    &(2000)  &(sec)\\\\ \\noalign{\\vskip3pt} \\noalign{\\hrule} \\noalign{\\vskip3pt} \\multicolumn{8}{c}{ESO 1.52-m}\\\\ \\noalign{\\vskip3pt} M\\,17             &07/7/96  &3995--4978    &1.5    &$-$21 &18 20 40.0    &$-$16 09 29 &3$\\times$1200\\\\ M\\,17                      &13/7/96     &3535--7400         &4.5    &$-$21 &\"     &\" &60, 300, 600\\\\ NGC\\,3576                  &11/2/97     &3995--4978         &1.5    &$-$50  &11 11 56.9          &$-$61 17 25               &6$\\times$1800\\\\ \\noalign{\\vskip3pt} \\multicolumn{8}{c}{AAT 3.9-m}\\\\ \\noalign{\\vskip3pt} NGC\\,3576   &08/02/95   &3509--3908 &1  &$-$50    &11 12 00.5 &$-$61 18 24  &300, 1800\\\\ NGC\\,3576       &\"       &3908--4305    &1      &$-$50       &\"       &\"       &2$\\times$1800\\\\ NGC\\,3576       &\"       &3635--7360    &8.5    &$-$50       &\"       &\"       &120, 300, 600\\\\ \\noalign{\\vskip3pt} \\multicolumn{8}{c}{NTT 3.5-m}\\\\ \\noalign{\\vskip3pt} 30 Doradus             &15/12/95    &3635--4145         &2 &76 &05 38 45.6          &$-$69 05 24   &3$\\times$1200\\\\ 30 Doradus             &\"       &4060--4520     &2  &76 &\"      &\"    &3$\\times$1200\\\\ 30 Doradus         &\"       &4515--4975     &2  &76 &\"      &\"    &300, 4$\\times$1200\\\\ 30 Doradus         &\"       &6507--7828             &3.8    &76 &\"          &\"    &3$\\times$1200\\\\ 30 Doradus         &\"           &3800--8400             &11     &76 &\"              &\"         &60, 300, 600\\\\ \\noalign{\\vskip3pt} LMC N11B                   &16/12/95    &3635--4145 &2      &$-$57 &04 56 47.0          &$-$66 25 11         &600\\\\ LMC N11B                   &\"           &4060--4520             &2      &$-$57 &\"          &\"         &2$\\times$1800\\\\ LMC N11B                   &\"           &4515--4975             &2      &$-$57 &\"          &\"         &2$\\times$1800\\\\ LMC N11B                   &\"           &6507--7828             &3.8    &$-$57 &\"          &\"         &600\\\\ LMC N11B                   &\"           &3800--8400             &11     &$-$57 &\"          &\"         &600\\\\ \\noalign{\\vskip3pt} SMC N66                    &16/12/95    &3635--4145 &2      &$-$57 &00 58 55.2          &$-$72 12 32         &600\\\\ SMC N66                    &\"           &4060--4520             &2      &$-$57 &\"          &\"         &2$\\times$1800\\\\ SMC N66                    &\"           &4515--4975             &2      &$-$57 &\"          &\"         &2$\\times$1800\\\\ SMC N66                    &\"           &6507--7828             &3.8    &$-$57 &\"          &\"         &600\\\\ SMC N66                    &\"           &3800--8400             &11     &$-$57 &\"          &\"         &300, 600\\\\  \\noalign{\\vskip3pt} \\noalign{\\hrule} \\end{tabular} \\end{minipage} \\end{table*} ",
        "conclusions": "The current dataset supported by published far-IR observations all but rules out the notion that Peimbert-type temperature fluctuations are the cause of the high metal abundances derived from optical recombination lines with respect to the lower abundances yielded by forbidden lines in {\\hii} regions. Neither do our results advance the case for the existence of Viegas \\& Clegg-type \\emph{high density, ionized} condensations in this sample of nebulae that could compromise the observationally derived {\\foiii} temperature, since the relative intensities of high-order {\\hi} Balmer lines {\\em and} the relative intensities of {\\oii} V\\,1 multiplet components both indicate that recombination is occurring in regions having electron densities similar to those indicated by the standard CEL nebular density diagnostics. Instead, our analysis points towards an origin for much of the enhanced heavy-element optical recombination-line emission (enhanced relative to the optical and infrared CEL lines from the same ions) in ionized regions that are of \\emph{low} density and which are also cold, so that they do not emit CELs. Thus a resolution of the ORL/CEL problem appears to require the existence of a hitherto unseen component in these {\\hii} regions, consisting of cold, rarefied and metal-rich, ionized gas.   \\vspace{7mm} \\noindent {\\bf Acknowledgments}  YGT acknowledges the award of a Perren Studentship. We thank Daniel P\\'{e}quignot for insightful comments and helpful discussions and the referee, Dr. R. Rubin, for his very careful reading of the manuscript."
    },
    "0209/astro-ph0209252_arXiv.txt": {
        "abstract": "The short term variability of the Galactic cosmic ray flux (CRF) reaching Earth has been previously associated with variations in the global low altitude cloud cover. This CRF variability arises from changes in the solar wind strength. However, cosmic ray variability also arises intrinsically from variable activity of and motion through the Milky Way. Thus, if indeed the CRF climate connection is real, the increased CRF witnessed while crossing the spiral arms could be responsible for a larger global cloud cover and a reduced temperature, thereby facilitating the occurrences of ice ages. This picture has been recently shown to be supported by various data \\citepp{Shaviv2001}. In particular, the variable CRF recorded in Iron meteorites appears to vary synchronously with the appearance ice ages.  Here we expand upon the original treatment with a more thorough analysis and more supporting evidence. In particular, we discuss the cosmic ray diffusion model which considers the motion of the Galactic spiral arms. We also elaborate on the structure and dynamics of the Milky Way's spiral arms. In particular, we bring forth  new argumentation using HI observations which imply that the galactic spiral arm pattern speed appears to be that which fits the glaciation period and the cosmic-ray flux record extracted from Iron meteorites.  In addition, we show that apparent peaks in the star formation rate history, as deduced by several authors, coincides with particularly icy epochs, while the long period of 1 to 2 Gyr before present, during which no glaciations are known to have occurred, coincides with a significant paucity in the past star formation rate. ",
        "introduction": "\\label{sec:intro}  It has long been known that solar variability is affecting climate on Earth.  The first indication for a solar--climate connection can be attributed to William \\citet{Herschel}, who found that the price of grain in England inversely correlated with the sunspot number.  He later suggested that it was due to changes in the solar irradiance \\citepp{Herschel2}. The irradiance variability is probably not large enough to explain the climatic variability observed by Herschel, nevertheless, synchronous temperature and solar variations do exist. For example, typical surface  temperatures during  northern summers were found to differ by  $0.5^{\\circ}$K to $1.5^{\\circ}$K  between  solar minima and solar maxima  \\citepp{Labitzke1992}.  Over the past century, Earth has experienced a gradual, though non-monotonic warming.  It is generally believed to be a result of a greenhouse effect by anthropogenic fossil fuel emissions.  However, a much better fit is obtained if part of the warming is attributed to a process, or processes, correlated with the solar activity, thus explaining for example, the non-monotonic global temperature change \\citepp{Christ1991,Soon1996,Beer2000}.  Moreover, the part of the climatic variability which is synchronized to the solar activity is larger than could be expected from just the 0.1\\% typical change in the solar irradiance \\citepp{Beer2000,Soon2000}.  Namely, the variability in the thermal flux itself appears to be insufficient to explain, for example, the global temperature variations observed.  If one goes further back in time, then climatic variability on the time scale centuries, is too correlated with solar activity.  Cold episodes in Europe such as the Maunder, Sp\\\"orer and Wolf Minima clearly correlate with peaks in the $^{14}$C flux, while warm episodes, such as the ``medieval warm'' period during which Vikings ventured across the Atlantic, correlate with minima in the $^{14}$C flux (e.g., \\citealtm{CloudExperiment}).  This flux itself is anti-correlated with the solar activity through the solar wind which more effectively reduces the Galactic cosmic ray flux that reaches Earth (and produces $^{14}$C) while the sun is more active.  On a somewhat longer time scale, it was even found that climatic changes in the Yucat\\'an correlate with the solar activity \\citepp{Hodell2001} (and possibly with the demise of the Maya civilization).  While on even longer time scales, it was shown that the monsoonal rainfall in Oman has an impressive correlation with the solar activity, as portrayed by the $^{14}$C production history \\citepp{Neff2001}.  Two possible path ways through which the solar activity could be amplified and affect the climate were suggested. First, solar variations in UV (and beyond) are non-thermal in origin and have a much larger relative variability than that of the total energy output. Thus, any effect in the atmosphere which is sensitive to those wavelengths, will be sensitive to the solar activity. UV is absorbed at the top part of the atmosphere (at typical altitudes of 50 km), and is therefore responsible for the temperature inversion in the Stratosphere. Any change in the UV heating could have effects that propagate downward. In fact, there is evidence that it can be affecting global circulations and therefore also climate at lower altitudes. For example, it could be affecting the latitudinal extent of the Hadley circulation \\citepp{Haigh1996}.  Another suggested proxy for a solar-climate connection, is through the solar wind modulation of the galactic cosmic ray flux, as first suggested by \\citet{Ney1959}.  Ney pointed out that cosmic-rays (CRs) are the primary source for ionization in the Troposphere, which in return could be affecting the climate.  First evidence in support was introduced by \\citet{Tinsley1991} in the form of a correlation between Forbush events and a reduction in the Vorticity Area Index in winter months.  Forbush events are marked with a sudden reduction in the CRF and a gradual increase over a typically 10 day period.  Similarly, \\citet{Pudovkin} reported a cloud cover decrease (in latitudes of 60N-64N, where it was measured) synchronized with the Forbush decreases.  Later, an effect of the Forbush decreases on rainfall has also been claimed \\citepp{Lebvedev} -- an average 30\\% drop in rainfall in the initial day of a Forbush event (statistically significant to 3$\\sigma$) was observed in 47 Forbush events recorded during 36 years in 50 meteorological stations in Brazil.  While in Antarctica, \\citet{Egorova} found that on the first day after a Forbush event, the temperature in Vostok station dramatically increased by an average of 10$^\\circ$K, but there was no measurable signal in sync with solar proton events. On the longer time scale of the 11-year solar cycle,  an impressive correlation was found between the CRF reaching Earth and the average global low altitude cloud cover \\citepp{Sven1997,Marsh2000}.  Although the above results are empirical in nature, there are several reasons to believe why the cosmic-ray route could indeed be responsible for a connection between the solar variability and cloud cover.  First, CRs are modulated by the solar activity.  On average, the heliosphere filters out 90\\% of the Galactic CRs (e.g., \\citealtm{Perko1987}).  At solar maximum, this efficiency increases as the solar wind is stronger.  Since it takes time for the structure of the heliosphere to propagate outward to the heliopause at 50 to 100 AU and for the CRs to diffuse inward, the CR signal reaching Earth lags behind all the different indices that describe the solar activity (e.g., the sunspot number or the 10.7 cm microwave flux which is known to correlate with the EUV flux).  The cloud cover signal is found to lag as well behind the solar activity and it nicely follows the lagging CRF. Second, both a more detailed analysis \\citepp{Marsh2000} and an independent study \\citepp{SecondAnalysis} show that the correlation is only with the low altitude cloud cover (LACC).  Among the different possible causes which can mediate between the solar variability and climate on Earth, only Galactic CRs can affect directly the lower parts of the atmosphere.  It is the Troposphere where the high energy CRs and their showers are stopped, and are responsible for the ionization.  EUV variability will affect (and ionize) the atmosphere at higher altitudes ($\\gtrsim$ 100 km). As mentioned, thermal heating by Ozone absorption could possibly affect also low altitudes, however, the CRF-cloud cover connection is seen only in low altitude clouds.  Solar CRs (which are less energetic than Galactic CRs) are not only stopped at similarly high altitudes, the terrestrial magnetic field also funnels them towards the poles.  On the other hand, the LACC-CRF correlation is seen globally.  Last, \\citet{Tinsley1991} who first found a correlation between Forbush events and a reduction in the Vorticity Area Index in winter months, showed that these events correlate significantly better with the cosmic ray flux than with the UV variations (which generally start a week before the Forbush events). They also suggested that the UV cannot be responsible for this Tropospheric phenomenon since the time scale for the Stratosphere to affect the Troposphere is longer than the Forbush--VAI correlation time scale.  Although the process of how CRs could affect the climate is not yet fully understood, it is very likely that the net ionization of the lower atmosphere (which is known to be governed by the CRF, e.g., \\citepp{Ney1959}) plays a major role, as the ionization of the aerosols could be required for the condensation of cloud droplets \\citepp{Dickinson1975,Kirkby2000,Harrison2000}.  An experiment is currently being planned to study the possible cosmic-ray flux -- cloud-cover connection. It could shed more light and perhaps solidify this connection \\citepp{CloudExperiment}. Moreover, some physical understanding appears to be emerging \\citepp{Yu}. Interestingly, the latter work may explain why the apparent effect is primarily on the lower troposphere and why the global warming of the past century has been more pronounced at the surface than at higher altitudes.  Hence, the evidence shows it to be reasonable that solar activity modulates the cosmic ray flux and that this can subsequently affect the global cloud cover and with it the climate.  Assuming this connection to be true, {\\em we should expect climatic effects also from intrinsic variations in the CRF reaching the solar system}.  With the possible exception of extremely high energies, CRs are believed to originate from supernova (SN) remnants (e.g., \\citealtm{Longair1994}, \\citealtm{Berez1990}). This is also supported with direct observational evidence \\citepp{SNRs}. Furthermore, since the predominant types of supernovae in spiral galaxies like our own, are those which originate from the death of massive stars (namely, SNe of types other than Ia), they should predominantly reside in spiral arms, where most massive stars are born and shortly thereafter die \\citepp{Drag1999}. In fact, high contrasts in the non-thermal radio emission are observed between the spiral arms and the disks of external spiral galaxies. Assuming equipartition between the CR energy density and the magnetic field, a CR energy density contrast can be inferred. In some cases, a lower limit of 5 can be placed for this ratio \\citepp{SNRs}.  Thus, when the sun passes through the Galactic spiral arms, an increased CRF is expected.  If the CRF-LACC connection is real, this will increase the average LACC and reduce the average global temperature.  The lower temperatures will then manifest themselves as episodes during which ice ages can occur.  Moreover, if the Milky Way as a whole is more active in forming stars, more massive stars will die and produce CRs.  We show in this work that both these effects appear to be supported by various data.  \\citet{Shaviv2001} studied this conjecture and found evidence which supports it, thereby strengthening the possibility of a CRF--climate connection.  In this work, we elaborate the original treatment by performing a more thorough analysis.  We significantly extend the discussion on the dynamics of the spiral pattern of the Milky Way as it is important for determining the reoccurrences of ice age epochs, and introduce a new argument that helps determine the pattern speed.  We also introduce more evidence in the form of an apparent correlation between the recorded Milky Way activity (as described by the star formation rate, SFR) and the occurrence of ice age epochs.  In particular, it is shown that the lack of glaciation activity on Earth between 1 and 2 Gyr BP (before present) appears to correlate with a dip in the Star formation rate in the same period (at least, as obtained by several but not all authors!). On the more speculative side, since the SFR activity may correlate with the activity in the LMC and with its estimated passages through perigalacticon, ice-ages could be attributed, to some extent, to fly-by's of the LMC.  It is also interesting to note that other mechanisms have been previously proposed to link the Galactic environment with climate variability on Earth.  The first such mechanism was proposed long ago by \\citet{Hoyle1939} who argued that an encounter of the Solar System with an interstellar cloud might trigger an ice age epoch by {\\em increasing} the solar luminosity, which produces an over compensating increase in cloudiness.  The increased luminosity is a result of the accretion energy released.  However, it is currently believed that radiation driving has a positive feedback (e.g., \\citealtm{Rind1993}), not a strong negative one.  Namely, an increase in the solar luminosity will result with an increase of the temperature, not a decrease.  Nonetheless, encounters with interstellar clouds could still have a temperature reducing effect if sufficient quantities of dust grains are injected to the upper atmosphere to partially shield the solar radiation \\citepp{Yabushita1985}.  These events are more likely to occur during spiral arm crossing, since it is there where dense molecular clouds concentrate.  However, since they require high density clouds, it seems unlikely that they can explain several $10^7$ yr glaciation epochs each spiral crossing.  A second mechanism has to do with the shrinking of the heliosphere. \\citet{Begelman1976} have shown that while crossing moderately dense ISM clouds with densities of $10^2$ to $10^3$ cm$^{-3}$, the bow shock of the heliosphere will be pushed further in than 1 AU. As a consequence, the slowing down effect that the heliosphere has on Galactic cosmic rays, will cease to work and the flux of Galactic low energy CRs will be significantly increased.  On the other hand, the charged particles comprising the solar wind will not reach Earth. Either way, the flux of low energy charged particles reaching Earth could be significantly altered.  Although these particles are not known to have a climatic effect at the moment, such an effect cannot be ruled out.  Unlike the previous mechanisms, if this route can work, it may require significantly less dense ISM clouds which are more frequent.  However, it is still unclear whether a several $10^7$ yr long glaciation event can be obtained via this route.  A third mechanism operating mainly during spiral arm crossing is the perturbation of the Oort cloud and injection of comets into the inner solar system. \\citet{Napier1979} and \\citet{Alvarez1980} discussed the effects that grains injected into the atmosphere by cometary bombardment will have on the climate by blocking the solar radiation. \\citet{Hoyle1978} proposed that a cometary disintegration in the vicinity of Earth's orbit would similarly inject grains into the atmosphere.  One should note that these mechanisms all predict ice-age epochs in synchronization with the spiral arm crossing. This is counter to the model described here in which a phase lag exists.  There were also proposals that related the Galactic year (i.e., the revolution period around the galaxy) to climate on Earth (\\citealtm{SG1973},\\citealtm{Williams1975}, \\citealtm{Frakes1992} and references therein).  For example, \\citet{Williams1975} suggested that IAEs on Earth are periodic, and that this rough $\\sim 150$ Myr period is half the Galactic year.  Williams raised the possibility that this Galactic-climate connection could arise if the disk is tidally warped (e.g., by the LMC), but did not mention a specific mechanism that can translate the warp into a climatic effect. On the other hand, \\citet{SG1973} suggested that climatic variability may arise if the solar orbit around the galaxy is eccentric and if, for some unknown physical reason, the solar luminosity is sensitive to the galactic gravitational pull.  Another interesting suggestion for an  extraterrestrial trigger for the ice-age epochs, has been made by \\citet{Dilke1972}; (see also \\cite{Dilke1974}), who showed that the solar core may be unstable to convective instability under the presence of chemical inhomogeneities induced by the nuclear burning. These authors have argued that both the time scales and luminosity variations involved could explain the occurrence of IAEs.  In addition to the extraterrestrial factors, there are also terrestrial factors which are in fact most often claimed by the paleoclimatological community to affect climatic variability on geological time scales. These are the continental geography, sea level, atmospheric composition, and volcanic, tectonic and even biological activity. It is likely that, at least to some extent, many of the aforementioned terrestrial and extraterrestrial factors affect the global climate. Therefore, one of the main questions still open in paleoclimatology is the relevant importance of each climatic factor.  We begin by reviewing the observations and measurements.  These include a summary of the glaciation epochs on Earth, the dynamics and star formation history of the Milky Way, and the CRF history as derived from Iron/Nickel meteorites.  Some of these results are described here for the first time.  We then proceed to describe the model which relates the Galactic environment to climate on Earth though the variability in the CRF, assuming CRs do affect the climate, and follow with the predictions of the model.  The model's backbone is the solution of the problem of CR diffusion while incorporating that the CR sources reside primarily in the spiral arms, and adding the climatic effect that the CRs may have.  Then, we continue with a comparison between the proposed theory and observations.  We show that an extensive set of tests employing currently available data points to the consistency of the theory.  \\begin{table} \\caption{Acronyms} { \\vskip 0.5cm \\begin{center} \\begin{tabular}{c l} \\hline Notation & Definition \\\\ \\hline BP & Before Present \\\\ CR & Cosmic Ray \\\\ cR & Co-Rotation  \\\\ CRF & Cosmic Ray Flux \\\\ HI, HII & Atomic, Ionized Hydrogen \\\\ IAE & Ice-Age Epoch \\\\ LACC & Low Altitude Cloud Cover \\\\ LMC & Large Magellanic Cloud \\\\ MW & Milky Way \\\\ SFR & Star Formation Rate \\\\ SN & Supernova \\\\ \\hline \\end{tabular} \\end{center} } \\end{table} ",
        "conclusions": "\\newcounter{myenum}  The thesis presented here relates the following topics: The Milky Way dynamics, Cosmic Ray diffusion in the Galaxy, the CR record in Iron meteorites, the effect of Cosmic Rays on climatology, and glaciology. This is achieved by considering the intimation that CRs can affect the global cloud cover and that the CR flux from the Galaxy should be variable. Before summarizing  the main conclusions relating the above topics, ``preliminary'' conclusions on each of the various topics can be drawn from the background analysis.  \\smallskip \\noindent {\\bf Milky Way Dynamics}: By studying the 4-arm structure seen in HI, extending to roughly $2 R_\\odot$, stringent constraints can be placed on the spiral pattern and dynamics of the Milky Way: \\begin{enumerate} \\item Since observations of HI {\\em outside} the solar circle suffer no velocity-distance ambiguity, the observations of 4-spiral arms should be considered robust.  {\\em If one further assumes} that these arms are density waves (which is by far the most consistent explanation for spiral arms) then these arms cannot extend further out than the 4 to 1 Lindblad resonance.  This implies $\\Omega_p \\lesssim 16 \\pm 2.5$ \\omunit.  This number agrees with roughly half the determinations of $\\Omega_p$, which cluster around this value.  \\item Several additional considerations point to the above value being not an upper limit for $\\Omega_p$, but the actual value of it.  \\item The 4-arm spiral cannot extend much further in from the solar galactocentric radius. Several possibilities were raised. Since the MW has a bar, at least two different pattern speeds exist in the MW.  \\item On a more speculative note, long term activity in the star formation rate appear to correlates with activity in the LMC and possibly with its perigalacticon passages, suggesting that long term star formation in the Milky Way and LMC are related.  \\setcounter{myenum}{\\value{enumi}}  \\end{enumerate}  \\noindent {\\bf CR Diffusion:}  \\begin{enumerate}  \\setcounter{enumi}{\\value{myenum}}  \\item Since CR sources are clearly more common to the Galactic spiral arms, models for CR propagation in the MW should take this fact into account.  Otherwise, significant discrepancies can arise.  In particular, quantities such as the ``age'' of the cosmic rays reaching Earth can be distorted.  \\item For typical diffusion model parameters, the CRF is expected to vary by ${\\cal O}(1)$, which is consistent with radio observations of external spiral galaxies.  \\item The distribution of CRs is both lagging behind the spiral crossings (when defined for example by the HII regions) and skewed towards later times.  The skewness arises from the asymmetry introduced by the spiral arm motion.  The lag in the peaks arises because the SN explosions are lagging the ionizing photons that produce the HII regions, and which were used here (and in \\cite{Taylor1993}) to define the location of the arms.  \\setcounter{myenum}{\\value{enumi}}  \\end{enumerate}  \\noindent {\\bf CR record in Iron meteorites}:  \\begin{enumerate}  \\setcounter{enumi}{\\value{myenum}}  \\item The ``standard'' method for extracting historic variability of the CRF, by comparing $^{41}{\\rm K}/^{40}$K dating to dating with a short lived unstable nucleotide (such as $^{10}\\mr{Be}/^{21}\\mr{Ne}$) are effective only at extracting ``recent'' changes (over several Myrs) or secular changes over longer durations, but they are not effective at extracting the signal expected from the periodic variability of the spiral arm passages.  \\item An effective method for finding the periodic CRF variations is a statistical analysis of CR exposure ages.  It assumes that the rate at which ``new exposed surfaces'' appear in Iron meteorites does not vary rapidly.  \\item This method shows that the exposure age data is consistent with it being periodic, with a period of $143 \\pm 10$ Myr, and a CRF contrast of $\\max (\\Phi) / \\min (\\Phi) \\gtrsim 3$.  It uses 50 meteorites which were dated only with $^{41}{\\rm K}/^{40}{\\rm K}$. The method reveals that statistically, it is unlikely that the meteor exposure ages were generated by a random process.  \\setcounter{myenum}{\\value{enumi}}  \\end{enumerate}  \\noindent {\\bf The Milky Way -- Climate connection}: By comparing the above findings to the appearance of glaciations on Earth, the following conclusions can be deduced:  \\begin{enumerate}  \\setcounter{enumi}{\\value{myenum}}  \\item Periodic glaciation epochs on earth in the past 1 Gyr can be consistently explained using the proposed scenario: Periodic passages through the Galactic spiral arms are responsible for an increased CRF, an increased LACC, a reduced global temperature and consequent ice-ages.  \\item Long term glaciation activity is apparently related to the global SFR activity in the Milky Way.  This may be related to LMC perigalacticon passages which would imply that the nearby extragalactic environment could be added to the factors affecting the global climate.  \\setcounter{myenum}{\\value{enumi}}  \\end{enumerate}  The evidence upon which the above conclusions were based, is the following:  \\begin{enumerate}  \\renewcommand{\\labelenumi}{\\Roman{enumi}}  \\item The period with which spiral arm passages have occurred using the HI data alone (and which should also correspond to the CRF period), is $163 \\pm 50$ Myr.  By combining the result of the 8 total measurements that scatter around this value, the predicted spiral arm crossing period is $134 \\pm 22$ Myr.  The period with which the CRF appears to vary using Fe/Ni meteorites, is $143 \\pm 10$ Myr. The period with which glaciations have been observed to reoccur on Earth is $145.5 \\pm 7$ Myr on average.  Clearly, all three signals are consistent with each other.  \\item By comparing the actual prediction for the location of the spiral arms to the glaciations, a best fit of $143 \\pm 5$ Myr is obtained. The phase of all the three signals using this periodicity is then found to fit the predictions. In particular: \\begin{itemize}  \\item The average mid-point of the glaciations is lagging by $33\\pm 20$~Myr after the spiral arm crossing, as is predicted ($31\\pm 8$~Myr, from a possible range of $\\pm 75$~Myr).  \\item The CR exposure ages of Iron meteorites cluster around troughs in the glaciations.  \\end{itemize}   \\item A random phenomenon for the appearance of glaciations is excluded with very high statistical significance.  \\item Long term variability in the appearance glaciations correlates with the observed SFR variability of the Milky Way.  In particular, the lack of apparent glaciation between 1 and 2 Gyr BP, correlates with a particularly low SFR in the milky way (less than half of today), while a high SFR rate between 2 and 3 Gyr BP (with a peak towards 2 Gyrs), correlate with the glaciations that Earth experienced between 2 and 3 Gyrs BP.  \\end{enumerate}  \\noindent {\\bf Additional conclusions}: More conclusions can be reached by considering the above results.  They do not pertain to climatic variability, but instead, they can be drawn by assuming that long term climate variability is indeed a measure of CRF variability, as the above results seem to endorse.  \\begin{enumerate}  \\setcounter{enumi}{\\value{myenum}}  \\item Once more accurate ``Eulerian'' measurements of the Galactic pattern speed will be available, interesting constraints could be placed on solar migration in the galaxy.  This is achieved by comparing the ``Eulerian'' measurements of the pattern speed to the ``Lagrangian'' pattern speed (as measured by the moving solar system), which is now known to an accuracy of $\\pm 3\\%$.  \\item The variability observed in the CRF record in Iron meteorites (see point 10) can be used to place constraints on models for the CR diffusion in the Milky Way.  This is in addition to the currently used constraint on the survival ratio of spallation products (primarily Be).  Thus, from CRF variability, we find $D \\lesssim 2 \\times 10^{28} {\\rm ~cm}^2/{\\rm sec}$ and $l_H \\lesssim 2$ kpc.  \\item The 4-arm spiral pattern has been stable for at least a billion years. A better understanding of past Glaciations (e.g., 2-3 Gyrs ago) could place interesting constraints on the lifetime of the spiral arms which cannot be done by any other means.  \\end{enumerate}  As apparent from the above list, there are several interesting ramifications to the picture presented here. However, one which bears particular interest on global warming and which was not discussed here, is that it now appears even more plausible that cosmic rays indeed affect the global climate, as suggested for example by \\citet{Ney1959}, \\citet{Tinsley1991}, and \\citet{Sven1997}. This implies that it is now more plausible that solar variations are affecting climate through modulation of the cosmic ray flux. This would explain an important part of the global warming observed over the past century. It is because of this important aspect, in particular, that we are obligated to iron out the still unclear points  raised in the caveats section, whether it be the paleoclimatological record, the possible mechanism by which cosmic rays affect climate,  or our astronomical understanding of the structure and dynamics of the Milky Way."
    },
    "0209/astro-ph0209064_arXiv.txt": {
        "abstract": "We review the characteristics of nucleosynthesis in 'Hypernovae', i.e., core-collapse supernovae with very large explosion energies ($ \\gsim 10^{52} $ ergs).  The hypernova yields show the following characteristics: 1) The mass ratio between the complete and incomplete Si burning regions is larger in hypernovae than normal supernovae.  As a result, higher energy explosions tend to produce larger [(Zn, Co, V)/Fe] and smaller [(Mn, Cr)/Fe], which could explain the trend observed in very metal-poor stars.  2) Because of enhanced $\\alpha$-rich freezeout, $^{44}$Ca, $^{48}$Ti, and $^{64}$Zn are produced more abundantly than in normal supernovae.  The large [(Ti, Zn)/Fe] ratios observed in very metal poor stars strongly suggest a significant contribution of hypernovae.  3) Oxygen burning takes place in more extended regions in hypernovae to synthesize a larger amount of Si, S, Ar, and Ca (\"Si\"), which makes the \"Si\"/O ratio larger.  The abundance pattern of the starburst galaxy M82 may be attributed to hypernova explosions.  We thus suggest that hypernovae make important contribution to the early Galactic (and cosmic) chemical evolution. ",
        "introduction": "One of the most interesting recent developments in the study of supernovae (SNe) is the discovery of some very energetic Supernovae (SNe), whose kinetic energy (KE) exceeds $10^{52}$\\,erg, about 10 times the KE of normal core-collapse SNe (hereafter $E_{51} = E/10^{51}$\\,erg).  Type Ic supernova (SN~Ic) 1998bw was probably linked to GRB 980425 (Galama et al. 1998), thus establishing for the first time a connection between gamma-ray bursts (GRBs) and the well-studied phenomenon of core-collapse SNe.  However, SN~1998bw was exceptional for a SN~Ic: it was as luminous at peak as a SN~Ia, indicating that it synthesized $\\sim 0.5$ \\Msun\\ of \\Nifs, and its KE was estimated at $E \\sim 3 \\times 10^{52}$ erg (Iwamoto et al. 1998; Woosley et al. 1999).  Because of its large KE, SN~1998bw was called a ``Hypernova (HN)\".  Subsequently, other ``hypernovae\" of Type Ic have been discovered or recognized, such as SN~1997ef (Iwamoto et al. 2000; Mazzali, Iwamoto, \\& Nomoto 2000), SN~1997dq (Matheson et al. 2001), SN~1999as (Knop et al. 1999; Hatano et al. 2001), and SN~2002ap (Mazzali et al. 2002). Another possible hypernovae, although of Type IIn, were SNe~1997cy (Germany et al. 2000; Turatto et al. 2000) and 1999E (Rigon et al. 2002).  Figure 1 shows the near-maximum spectra and the absolute V-light curves of Type Ic hypernovae.  These hypernovae span a wide range of properties, although they all appear to be highly energetic compared to normal core-collapse SNe.  SN 1999as is the most luminous supernova ever discovered, reaching a peak magnitude $M_{\\rm V} < - 21.5$, while the brightness of SN 2002ap appears to be similar to that of normal core collapse SNe.  In the following sections, we summarize the properties of these hypernovae as derived from optical light curves and spectra.  We then show that nucleosynthesis in hypernovae is quite distinct from the case of ordinary supernovae, thus making a unique contribution to galactic chemical evolution.  \\begin{figure} \\begin{center} \\begin{minipage}[t]{0.45\\textwidth} \\plotone{f1s02ap.epsi} \\end{minipage} \\begin{minipage}[t]{0.45\\textwidth} \\plotone{f2s02ap.epsi} \\end{minipage} \\end{center} \\caption{Left: The near-maximum spectra of Type Ic SNe and hypernovae: SNe 1998bw, 1997ef, 2002ap, and 1994I. Right: The observed $V$-band light curves of SNe 1998bw ({\\em open circles}), 1997ef ({\\em open triangles}), 2002ap ({\\em stars}), and 1994I ({\\em filled circles}) (Mazzali et al. 2002). } \\label{fig:02ap} \\end{figure} ",
        "conclusions": "We have shown that signatures of hypernova nucleosynthesis are seen in the abundance patterns in very metal poor stars and the starburst galaxy M82.  (See also the abundance pattern in X-ray Nova Sco; Israelian et al. 1999; Podsiadlowski et al. 2002).  We suggest that hypernovae of massive stars may make important contributions to the Galactic (and cosmic) chemical evolution, especially in the early low metallicity phase.  The IMF of Pop III stars might be different from that of Pop I and II stars, and that more massive stars are abundant for Pop III."
    },
    "0209/astro-ph0209587_arXiv.txt": {
        "abstract": "{We report on XMM-Newton observations of GRO J1655-40 and GRS 1009-45, which are two black hole X-ray transients currently in their quiescent phase. GRO J1655-40 was detected with a 0.5 - 10 keV luminosity of $5.9 \\times 10^{31}$ erg s$^{-1}$. This luminosity is comparable to a previous Chandra measurement, but ten times lower than the 1996 ASCA value, most likely obtained when the source was not yet in a true quiescent state. Unfortunately, XMM-Newton failed to detect GRS 1009-45. A stringent upper limit of $8.9 \\times 10^{30}$ erg s$^{-1}$ was derived by combining data from the EPIC-MOS and PN cameras.  The X-ray spectrum of GRO J1655-40 is very hard as it can be fitted with a power law model of photon index $\\sim 1.3\\pm 0.4$. Similarly hard spectra have been observed from other systems; these rule out coronal emission from the secondary or disk flares as the origin of the observed X-rays. On the other hand, our observations are consistent with the predictions of the disc instability model in the case that the accretion flow forms an advection dominated accretion flow (ADAF) at distances less than a fraction ($\\sim$ 0.1 - 0.3) of the circularization radius. This distance corresponds to the greatest extent of the ADAF that is thought to be possible. ",
        "introduction": "Soft X-ray transients -- SXTs (sometimes called X-ray novae) are semi--detached binaries in which the accreting (primary) star is a black hole (BH) or a neutron star, and the mass--losing secondary is usually a late type star. These systems typically brighten in X-rays by as much as 10$^7$ in a week and then decay back into quiescence over the course of a year. The maximum outburst luminosities $L_{\\rm max}$ seen in SXTs are typically $\\sim(0.2-1)$ of the Eddington luminosity $L_{\\rm Edd}$. Successive outbursts are usually separated by years to decades of quiescence (see e.g. Tanaka \\& Shibazaki \\cite{ts96}; Chen et al. \\cite{csl97}, for reviews).  Quiescent states of SXTs are very interesting for at least two reasons. First, they provide the strongest evidence now available for the existence of stellar--mass black holes. The detection of the secondary in quiescence allows a determination of the mass function of the binary system, which is an absolute lower limit on the primary mass $M_1$. The mass function exceeds 3 M$_\\odot$ in eight systems; in five others, constraints on the inclination angle of the system and the secondary mass result in values for $M_1$ that are in the range 5 -- 10 M$_\\odot$ (Narayan et al. \\cite{ngm02}). Thus there are now thirteen SXTs with primary masses that exceed the maximum stable mass of a neutron star ($\\simless 3$ M$_\\odot$).  The second, more mundane reason why quiescent states are interesting is that quiescence provides a strong test of outburst--cycle models. SXT outburst cycles are well described by the disc instability model (Dubus et al. \\cite{dhl01}; see Lasota \\cite{l01} for a complete review of the disc instability model). In quiescence, the disc is non--steady (a property too often forgotten by too many authors). Its temperature and viscosity are low, and the disc is unable to transport all of the mass supplied by the secondary to the primary; the mass of the disc slowly builds up and the temperature rises finally to the hydrogen ionization temperature. At this moment, the disc becomes thermally and viscously unstable due to strong opacity variations. Propagating heat fronts bring the entire disc into a hot state; in this outburst state, the mass transfer rate is large, and the disc empties until it cannot sustain this regime any longer; it then returns to quiescence.  Even though the disc instability paradigm is widely accepted and its particular realizations are often successful (requiring sometimes additional assumptions, see e.g. Esin et al. \\cite{elh00}), the physics of the quiescent state is still rather controversial. As a prime example, the origin of ``viscosity\" in this state is not really known (see e.g. Gammie \\& Menou, \\cite{gm98}; Menou \\cite{m02}; Lasota \\cite{l02}), and according to the simplest version of the disc-instability model (the version originally used to explain dwarf-nova outbursts; Lasota \\cite{l01}) the very long SXT recurrence times require unusually low values of the viscosity parameter $\\alpha$. However, Dubus et al. (\\cite{dhl01}) showed that the combined effects of disc irradiation during outbursts (King \\& Ritter \\cite{kr98}) and disc truncation during quiescence (Menou et al. \\cite{mhln00}) result quite effortlessly in long recurrence times for standard values of the viscosity parameter, making the disc instability model for SXTs a working and testable hypothesis.  Narayan et al. (\\cite{nmy96}; see also Lasota et al. \\cite{lny96} and Narayan et al. \\cite{nbm97}) noticed that spectra of quiescent SXTs cannot be produced by geometrically thin, optically thick accretion discs and suggested that the inner flow in such systems forms an advection dominated accretion flow ADAF. In any case, according to the disc instability model a disc extending down to the neutron star surface or the last stable orbit around a black hole, can be in a cold, neutral state everywhere only for vanishingly low accretion rates close to the central object (Lasota \\cite{l96}): the maximum accretion rate onto the compact object would be $\\sim 4000 \\left(M_1/\\msol\\right)^{1.77} (r_{\\rm in}/r_{\\rm s})^{2.65}$ g s$^{-1}$, where $r_{\\rm s}$ is the Schwarzschild radius (Hameury et al. \\cite{hmd98}), much too low to produce the X-ray luminosity observed in quiescent SXTs. On the other hand the ADAF model reproduces well the observed luminosities, spectra (Narayan et al. \\cite{nbm97}; Quataert \\& Narayan \\cite{qn99}) and observed delays between the rises to outburst in optical and X-rays (Hameury et al. \\cite{hlmn97}).  As first pointed out by Narayan et al. (\\cite{ngm97}), the presence of the radiatively inefficient ADAFs in quiescent SXTs allows one to compare black holes with compact bodies endowed with material surfaces. Because bodies such as neutron stars must re-emit the heat left over in the accretion flow accumulating at their surface they should be brighter than black holes accreting at the same rate since in this case the residual energy is lost forever past the event horizon. X-ray observations of quiescent SXTs showed that this is indeed the case (Narayan et al. \\cite{ngm97}; Menou et al. \\cite{men99}; Garcia et al. \\cite{gmn01}). However, the ADAF origin of X-rays in quiescent SXTs has been questioned by several authors (see e.g. Bildsten \\& Rutledge \\cite{br01}; Nayakshin \\& Svensson, \\cite{ns01}). High quality X-ray observations can solve this controversy.  In this paper, we report on XMM-Newton observations of \\object{GRO J1655-50} (AO-1 GTO) and \\object{GRS 1009-45} (AO-1 GO).We briefly describe previous observations of these transients in section 2; we present our observations in section 3, and we discuss their implications in section 4. ",
        "conclusions": "We have observed and detected GRO J1655-40 during a 40 ksec XMM-Newton observation, with a luminosity of 5.9 10$^{31}$ erg s$^{-1}$; GRS 1009-45 was undetected, leading to a relatively small upper limit on luminosity of 8.9 10$^{30}$ erg s$^{-1}$ for a distance of 5 kpc. These observations are consistent with the disc instability model if the accretion disc is truncated at a radius of 0.1 - 0.3 times the circularization radius, where the accretion flow forms an ADAF. The spectrum appears to be quite hard, much harder than one would expect from coronal emission from the secondary star or from the accretion disc itself. The quality of the data is however not sufficient to constrain and distinguish between various types of optically thin flows in the vicinity of the black hole.  There is no sign of variability on a time scale of hours, and the flux measured by XMM-Newton from GRO J1655-40 is consistent with the Chandra value (Kong et al. \\cite{kmg02}), and about 10 times lower than that which was measured with ASCA by Asai et al. (\\cite{adh98}) and Ueda et al. (\\cite{uit98}) between two outbursts, indicating that the system was not fully in quiescence at that time, and that the mass transfer rate from the secondary was high."
    },
    "0209/astro-ph0209314_arXiv.txt": {
        "abstract": "In this series of two papers we describe a project with the Space Telescope Imaging Spectrograph (STIS) on Hubble Space Telescope (HST) to measure the line-of-sight velocities of stars in the central few arcsec of the dense globular cluster M15. The main goal of this project is to search for the possible presence of an intermediate mass central black hole. This first paper focuses on the observations and reduction of the data. We `scanned' the central region of M15 spectroscopically by consecutively placing the $0.1''$ HST/STIS slit at 18 adjacent positions. The spectral pixel size exceeds the velocity dispersion of M15. This puts the project at the limit of what is feasible with STIS, and exceedingly careful and complicated data reduction and analysis were required. We applied corrections for the following effects: (a) drifts in the STIS wavelength scale during an HST orbit; (b) the orbital velocity component of HST along the line-of-sight to the cluster, and its variations during the HST orbit; and (c) the apparent wavelength shift that is perceived for a star that is not centered in the slit. The latter correction is particularly complicated and requires many pieces of information: (1) the positions and magnitudes of all the stars near the center of M15; (2) the accurate positionings of the STIS slits during the observations; (3) and the HST/STIS point-spread function (PSF) and line-spread function (LSF). To address the first issue we created a stellar catalog of M15 from the existing HST/WFPC2 data discussed previously by Guhathakurta \\etal (1996), but with an improved astrometric and photometric calibration. The catalog is distributed electronically as part of this paper. It contains 31,983 stars with their positions and U, B and V magnitudes. To address the second issue we model the observed intensity profiles along the STIS slits to determine the slit positionings to $0.007''$ accuracy in each coordinate. To address the third issue we obtained observations of a bright field star to which we fitted multi-Gaussian PSF and LSF models. Upon reduction of the M15 spectroscopy we ultimately obtain 19,200 one-dimensional STIS spectra, each for a different aperture position in M15, with a velocity scale accurate to better than $2.5 \\kms$. We develop an algorithm that co-adds the spectra for individual apertures and use it to extract spectra of individual stars with minimum blending and maximum $S/N$. In Paper~II (Gerssen et al.) we use these spectra to extract reliable line-of-sight velocities for 64 stars, half of which reside within $R = 2.4''$ from the cluster center. These velocities constrain the central structure, dynamics and mass distribution of the cluster. ",
        "introduction": "\\label{s:intro}  This is the first in a series of two papers in which we present the results of a study with the Hubble Space Telescope (HST) of the line-of-sight velocities of stars in the central few arcsec of the globular cluster M15 (NGC 7078). The present paper discusses the spectroscopic observations with the HST Space Telescope Imaging Spectrograph (STIS) and the extraction and calibration of the stellar spectra. It also discusses the construction of a photometric catalog from HST imaging with the Second Wide Field and Planetary Camera (WFPC2), which proved essential in the reduction and interpretation of the data. The second paper (Gerssen \\etal 2002, hereafter Paper~II) discusses the determination of line-of-sight velocities from the spectra, and the implications for the dynamics and structure of M15. The main motivation for our study is to better constrain the possible presence of an intermediate-mass black hole in the center of M15. Such a black hole was hinted at by previous work (e.g., Peterson, Seitzer \\& Cudworth 1989; Gebhardt \\etal 2000a), but the limited quality of the data precluded very strong conclusions (see the review in van der Marel 2001).  The globular cluster M15 at a distance of 10 kpc has one of the highest central densities of any globular cluster in our Galaxy. The high density has likely affected the stellar population, as evidenced by the presence of two bright X-ray sources (White \\& Angelini 2001) and several millisecond pulsars (Phinney 1993). A variety of interesting dynamical and structural evolutionary phenomena have been predicted to occur at high stellar densities, including mass segregation and the possible formation of a central black hole (Hut \\etal 1992; Meylan \\& Heggie 1997; see the additional discussion in the introductory section of Paper II). As a result, M15 has been one of the globular clusters for which the structure and dynamics have been most intensively studied in the past decade.  Studies of the dynamics of M15 have focused primarily on the determination of line-of-sight velocities of individual stars, using either spectroscopy with single apertures, long-slits or fibers (Peterson \\etal 1989; Dubath \\& Meylan 1994; Dull \\etal 1997; Drukier \\etal 1998), or using imaging Fabry-Perot spectrophotometry (Gebhardt \\etal 1994, 1997, 2000a). Integrated light measurements using single apertures have also been attempted (Peterson \\etal 1989; Dubath, Meylan \\& Mayor 1994), but are only of limited use; integrated light spectra are dominated by the light from only a few bright giants, and as a result, inferred velocity dispersions are dominated by shot noise (Zaggia \\etal 1992, 1993; Dubath \\etal 1994).  Line-of-sight velocities are now known for $\\sim\\! 1800$ M15 stars, as compiled and analyzed by Gebhardt \\etal (2000a). The projected velocity dispersion profile increases inwards from $\\sigma = 3 \\pm 1 \\kms$ at $R=7$ arcmin, to $\\sigma = 11 \\pm 1 \\kms$ $R=24''$. The velocity dispersion is approximately constant at smaller radii, and is $\\sigma = 11.7 \\pm 2.8 \\kms$ at the innermost available radius $R \\approx 1''$. The rotational properties of M15 are quite peculiar. The position angle of the projected rotation axis in the central region is $\\sim\\!  100^{\\circ}$ different from that at larger radii, and the rotation amplitude increases to $V_{\\rm rot} = 10.4 \\pm 2.7 \\kms$ for $R \\lta 3.4''$, so that $V_{\\rm rot} / \\sigma \\approx 1$ in this region. A plausible explanation for these properties is still lacking.  To improve our understanding of the central mass distribution in M15, it is of crucial importance to obtain more stellar velocities close to the center. However, velocity determinations at $R \\lta 2''$ are very difficult due to severe crowding and the presence of a few bright giants in the central arcsec. Gebhardt \\etal (2000a) used an Imaging Fabry-Perot spectrophotometer with adaptive optics on the CFHT, and obtained FWHM values as small as $0.09''$. However, the Strehl ratio was only $\\lta 6$\\%, so that even in these observations the light from the fainter turnoff and main-sequence stars in the central arcsec was overwhelmed by the PSF wings of the nearby giants. As a result, there are only 5 stars with known velocities within $R \\leq 1.3''$ from the cluster center. To improve upon this situation we initiated the project on which we report here, in which we used HST/STIS to map the center of M15 spectroscopically at high spatial resolution. In the future, additional kinematic data on M15 may come available from stellar proper motion measurements with HST, using techniques such as those described by Anderson \\& King (2000).  However, at present no such measurements are available.  This paper is organized as follows. Section~\\ref{s:imaging} describes the construction of a stellar catalog from WFPC2 data. This catalog is essential for a proper interpretation of the M15 spectra. It is distributed electronically as part of the present paper. Section~\\ref{s:calibspec} describes STIS spectra of a bright field star which were obtained to calibrate the STIS point spread function (PSF) and line-spread function (LSF). Section~\\ref{s:specM15} describes the M15 STIS spectroscopy, the observational setup and data reduction, and an analysis of the achieved slit positioning accuracies. The velocity calibration accuracy of the spectra is the most crucial aspect of the observations, and this issue is discussed in detail. Section~\\ref{s:indstars} discusses the algorithm adopted for the extraction of individual stellar spectra. These spectra form the end-product of the present paper. They are used in Paper~II to determine line-of-sight velocities and to study the structure and dynamics of M15. Section~\\ref{s:conc} presents concluding remarks. ",
        "conclusions": "\\label{s:conc}  The globular cluster M15 is one of the densest globular clusters in our Galaxy. A variety of interesting dynamical and structural evolutionary phenomena have been predicted to occur at high stellar densities, including mass segregation and the possible formation of a central black hole. Stellar kinematical data are required required to address these issues observationally. While M15 has been one of the globular clusters that has been most intensively studied in the past decades, ground-based data have only been able to provide limited information in the central few arcsec. We therefore executed a project to map the center of M15 spectroscopically at high spatial resolution with HST/STIS.  Our observational project is at the limit of what is feasible with HST/STIS. The pixel size of our spectra corresponds to $15.86 \\kms$, which exceeds the central velocity dispersion of M15 ($\\sim 12 \\kms$; Gebhardt \\etal 2000a). Extreme care therefore had to be taken in the reduction and velocity calibration of the data. The main result of the present paper is that we have in fact succeeded in extracting accurately calibrated spectra from our data. The implications of this for our understanding of M15 are discussed in Paper II.  To be able to properly interpret the long-slit spectra, we started by creating a stellar catalog for M15. We used the HST/WFPC2 data discussed previously by G96, but we performed an improved astrometric and photometric calibration. The final stellar catalog contains 31,983 stars with their positions and U, B and V magnitudes. It is spatially complete within $14.9''$ from the cluster center, and extends as far as $2.0$ arcmin from the cluster center in some directions. The catalog contains stars down to magnitude $V \\approx 22.5$, and it is photometrically complete for stars brighter than $V \\approx 19$.  For proper calibration of the M15 spectra we performed a set of observations of a bright field star, with the star placed at various positions with respect to the slit. After basic data reduction and correction for the velocity of HST around the Earth we used these spectra for modeling the PSF and LSF of STIS. This is particularly important for understanding the apparent velocity shift for a star that is offset from the center of the slit. We derive a simple multi-Gaussian model for the PSF that reproduces the observed shifts with an RMS residual of only $2.1 \\kms$ (for a known stellar position with respect to the slit). Spatial undersampling causes artificial undulations in the spectra, but this only affects low frequencies and does not affect the determination of radial velocities through cross-correlation.  The M15 observations consisted of 6 HST visits spread over 2 years. The central region of the cluster was `scanned' by consecutively placing the $0.1''$ slit at 18 adjacent positions. The intensity profiles along the slits were used to accurately determine the actual slit positions on the sky. With use of our WFPC2 catalog and PSF model we find that these positions can be determined to an accuracy of $0.007''$ in each coordinate. Positional errors in our knowledge of the slit positions translate directly into velocity errors, but with the achieved positional accuracy these errors are no larger than $1.2 \\kms$.  The reduced and calibrated data set contains 19,200 STIS spectra, each for a different aperture position in M15. We developed an algorithm that co-adds the spectra for individual apertures, so as to extract spectra of individual stars with minimum blending and maximum $S/N$. Our data allowed us to extract spectra with an average $S/N > 5.5$ per pixel for a total of 131 stars, out of which 15 stars are within $1''$ from the cluster center, 39 stars are within $2''$, and 87 stars are within $5''$. In Paper~II we show that reliable line-of-sight velocities can be derived for 64 of these stars, and that this provides enough new information to significantly improve our understanding of the dynamics and structure of M15."
    },
    "0209/astro-ph0209122_arXiv.txt": {
        "abstract": "Recent {\\em Chandra} observations have raised expectations that the objects RX J185635-3754 and 3C58 are either bare strange quark stars or stars with extended quark cores.  However, these observations can also be interpreted in terms of normal neutron stars.  Essential requirements for either explanation are that they simultaneously account for the observed (i) spectral features (i.e., thermal or nonthermal, possible lines), (ii) bounds on the inferred mass ($M$) and radius ($R$), and (iii) the cooling curves (effective temperatures $T$ and luminosities $L$ vs. age). Compact stars offer the promise of delineating QCD at finite baryon density~\\cite{LP01} in a fashion complementary to relativistic heavy ion collider experiments, which largely address the finite temperature, but baryon poor regime. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209408_arXiv.txt": {
        "abstract": "s{ The {\\bf AERQS} is a project aimed at the construction of an all-sky statistically well-defined sample of local ($z\\le 0.3$) and bright ($R\\le 15.5$) QSOs. Present uncertainties on the theoretical modeling of the QSO evolution are to be ascribed to the relatively small number of objects observed in this magnitude and redshift domain. We are filling this gap with a sample of 392 AGN covering all the sky at high Galactic latitude: we present here the Clustering analysis and the Luminosity Function of AGN at low redshift. The {\\bf AERQS} sets an important zero point about the properties of the local QSO population and the evolutionary pattern of QSOs between the present epoch and the highest redshift.} ",
        "introduction": "The analysis of the Luminosity Function (LF) and Clustering of QSOs is a fundamental cosmological tool to understand their formation and evolution. QSOs, with normal galaxies, supply detailed informations on the distribution of Dark Matter Halos that are generally thought to constitute the ``tissue'' above which structures form and grow. The lighting up of galaxies and QSOs depends on how the baryons cool within the dark matter halos and form stars or begin accreting onto the central BH, ending up as the only directly visible peaks of a much larger, invisible structure. The detailed analysis of the distribution of those peaks in the universe, i.e. their number and clustering, can give important constraints on their evolution and distinguish between the various favored models that explain the formation of cosmic structures. The theoretical interpretation of the QSO statistical properties is not easy due to the number of ingredients used in theoretical models and the degeneracies among them. In particular there is not a unique combination of these parameters to reproduce the LF and clustering of QSOs in the redshift range $0.3\\le z\\le 2.2$. We have started a project, the {\\bf AERQS} to find bright AGN in the local Universe and fill the existing gap in present day large surveys: it is paradoxical that in the era of 2QZ and SDSS, with thousands of faint QSOs discovered even at high redshift (z=6.28), still there are very few bright QSOs at low-z, compared to the numbers at $z\\sim 2$. One of the main reasons is that the surface density of low-z and bright QSOs is very low, of the order of few times $10^{-2}$ per $deg^{2}$. Another reason is that with optical information only is very difficult to isolate efficiently bright QSOs from billions of stars in large areas: a survey based on different selection criteria is required. We have used the X-ray emission, a key feature of the AGN population: QSO candidates have been selected by cross-correlating the X-ray sources in the RASS-BSC with {\\em optically bright} objects in photographic plates. The X-ray emission is used only to compute an ``X-Optical color'' for the selection of AGN. Therefore {\\em the result of our selection cannot be considered an identification of X-ray sources}. The {\\bf AERQS} is divided in three subsamples, two in the northern hemisphere (GSC and USNO, described in Grazian et al. 2000[\\cite{paper1}]) and the DSS sample in the South (Grazian et al. 2002[\\cite{paper2}]). The spectroscopic identification of the candidates has been completed, ending with 392 AGN (60$\\%$ previously unknown) over $\\sim 14000$ $deg^2$ in the redshift range $0.02\\le z\\le 2.04$. The redshift distribution shows a peak around $z\\sim 0.1$ with an extended tail up to $z=0.4$. Five AGN with $0.6\\le z\\le 2.04$ are possibly magnified/lensed objects. ",
        "conclusions": "We have built a well defined sample of {\\em 392 optically bright ($R\\le 15.5$) AGN (60$\\%$ are new identifications) in the redshift range $0.02\\le z\\le 0.30$}, filling the gap between the local Universe and the highest redshifts. The {\\em QSO Optical LF} at $0.04 \\le z \\le 0.15$ is computed and shown to be consistent with a LDLE Evolution of the type derived by La Franca \\& Cristiani[\\cite{lfc97}] in the redshift range $0.3 \\le z \\le 2.2$. The {\\em TPCF of QSOs} is constant or mildly increasing at high-z and increasing towards low-z. The QSO LF, together with the correlation function, at $z\\le 0.3$ constitutes a {\\em fundamental zero point} for a detailed modeling of the formation and evolution of cosmic structures in the Universe."
    },
    "0209/astro-ph0209136_arXiv.txt": {
        "abstract": "A long-standing problem in supernova physics is how to measure the total energy and temperature of $\\nu_\\mu$, $\\nu_\\tau$, $\\bar{\\nu}_\\mu$, and $\\bar{\\nu}_\\tau$.  While of the highest importance, this is very difficult because these flavors only have neutral-current detector interactions.  We propose that neutrino-proton elastic scattering, $\\nu + p \\rightarrow \\nu + p$, can be used for the detection of supernova neutrinos in scintillator detectors.  It should be emphasized immediately that the dominant signal is on {\\it free} protons.  Though the proton recoil kinetic energy spectrum is soft, with $T_p \\simeq 2 E_\\nu^2/M_p$, and the scintillation light output from slow, heavily ionizing protons is quenched, the yield above a realistic threshold is nearly as large as that from $\\bar{\\nu}_e + p \\rightarrow e^+ + n$.  In addition, the measured proton spectrum is related to the incident neutrino spectrum. The ability to detect this signal would give detectors like KamLAND and Borexino a crucial and unique role in the quest to detect supernova neutrinos.  These results are now published: J.~F.~Beacom, W.~M.~Farr and P.~Vogel, Phys.\\ Rev.\\ D {\\bf 66}, 033001 (2002) [arXiv:hep-ph/0205220]; the details are given there~\\cite{elastic}. ",
        "introduction": "When the next Galactic supernova occurs, approximately $10^4$ detected neutrino events are expected among the several detectors around the world.  It is widely believed that these $10^4$ events will provide important clues to the astrophysics of the supernova as well as the properties of the neutrinos themselves.  Interestingly, recent breakthroughs in understanding solar and atmospheric neutrinos each occurred when the accumulated samples of detected events first exceeded $10^4$.  But will we have enough information to study the supernova neutrino signal in detail?  Almost all of the detected events will be charged-current $\\bar{\\nu}_e + p \\rightarrow e^+ + n$, which will be well-measured, both because of the large yield and because the measured positron spectrum is closely related to the neutrino spectrum. Because of the charged-lepton thresholds, the flavors $\\nu_\\mu$, $\\nu_\\tau$, $\\bar{\\nu}_\\mu$, and $\\bar{\\nu}_\\tau$ can only be detected in neutral-current reactions, of which the total yield is expected to be approximately $10^3$ events.  However, in general one {\\it cannot} measure the neutrino energy in neutral-current reactions. This talk presents an exception.  These four flavors are expected to carry away about 2/3 of the supernova binding energy, and are expected to have a higher temperature than $\\nu_e$ or $\\bar{\\nu}_e$.  However, there is no experimental basis for these statements, and present numerical models of supernovae cannot definitively address these issues either.  If there is no spectral signature for the neutral-current detection reactions, then neither the total energy carried by these flavors nor their temperature can be separately determined from the detected number of events.  But it is crucial that these quantities be {\\it measured}.  Both are needed for comparison to numerical supernova models.  The total energy is needed to determine the mass of the neutron star, and the temperature is needed for studies of neutrino oscillations.  At present, such studies would suffer from the need to make model-dependent assumptions.  This problem has long been known, but perhaps not widely enough appreciated.  In this talk, I clarify this problem, and provide a realistic solution that can be implemented in two detectors, KamLAND (already operating) and Borexino (to be operating soon).  The solution is based on neutrino-proton elastic scattering, which has never before been shown to be a realistic detection channel for low-energy neutrinos.  In this talk, I will focus on just the problem of measuring the temperature and total energy of $\\nu_\\mu$, $\\nu_\\tau$, $\\bar{\\nu}_\\mu$, and $\\bar{\\nu}_\\tau$, since everything else in understanding supernova neutrinos depends on it. ",
        "conclusions": "We have shown that neutrino-proton elastic scattering, previously unrecognized as a useful detection reaction for low-energy neutrinos, in fact has a yield for a supernova comparable to $\\bar{\\nu}_e + p \\rightarrow e^+ + n$, even after taking into account the quenching of the proton scintillation light and assuming a realistic detector threshold.  In addition, the measured proton spectrum is related to the incident neutrino spectrum.  We have shown explicitly that one can separately measure the total energy and temperature of $\\nu_\\mu$, $\\nu_\\tau$, $\\bar{\\nu}_\\mu$, and $\\bar{\\nu}_\\tau$, each with uncertainty of order 10\\% in KamLAND.  This greatly enhances the importance of detectors like KamLAND and Borexino for detecting supernova neutrinos.  \\begin{figure} \\centerline{\\includegraphics[width=18pc]{fits.eps}}\\vspace{-1cm} \\caption{\\label{fig:fits} Scatterplot of $10^3$ fitted values, in the $E^{tot}$ and $T$ plane, for the labeled ``true'' values, where $E^{tot}$ is the total portion of the binding energy carried away by the sum of $\\nu_\\mu$, $\\nu_{\\tau}$, $\\bar{\\nu}_{\\mu}$, and $\\bar{\\nu}_{\\tau}$, and $T$ is their temperature.  The values of $E^{tot}$ and $T$ were chosen such that the numbers of events above threshold were the same.  The measured shape of the proton spectrum breaks the degeneracy between these two parameters. Without that spectral information, one could not distinguish between combinations of $E^{tot}$ and $T$ along the band in this plane that our three example regions lie along.} \\end{figure}  For Borexino, the useful volume for supernova neutrinos is 0.3 kton, and the hydrogen to carbon ratio in the pure pseudocumene (C$_9$H$_{12}$) is $1.3:1$~\\cite{SNborexino}, so there are about 4.7 times fewer free proton targets than assumed for KamLAND.  However, the quenching is less in pure scintillator (KamLAND is about 20\\% pseudocumene and 80\\% paraffin oil~\\cite{kamland}), and the errors on $E^{tot}$ and $T$ scale as $1/\\sqrt{N}$, so that the precision in Borexino should be about 20\\% or better.  Other techniques for bolometric measurements of supernova neutrino fluxes have been studied.  Detectors for elastic neutral-current neutrino scattering on electrons~\\cite{cabrera} and coherently on whole nuclei~\\cite{coherent} have been discussed, but never built. If neutrino oscillations are effective in swapping spectra, then the temperature of the ``hot'' flavors may be revealed in the measured positron spectrum from $\\bar{\\nu}_e + p \\rightarrow e^+ + n$; two recent studies have shown very good precision ($< 5\\%$) for measuring the temperatures and the total binding energy~\\cite{barger,minakata}.  However, they assumed exact energy equipartition among the six neutrino flavors, whereas the uncertainty on equipartition is at least 50\\%~\\cite{raffeltproc}.  Nevertheless, under less restrictive assumptions, this technique may play a complementary role.  Finally, since for different cross sections, the neutral-current yields depend differently on temperature, comparison of the yields may provide some information~\\cite{relyields}.  However, there are caveats.  In neutrino-electron scattering, the neutrino energy is not measured because the neutrino-electron angle is much less than the angular resolution due to multiple scattering.  The scattered electrons, even those in a forward cone, sit on a much larger background of $\\bar{\\nu}_e + p \\rightarrow e^+ + n$ events, so it is difficult to measure their spectrum~\\cite{SNpoint}; also, their total yield is only weakly dependent on temperature.  At the other extreme (see Fig.~3 of Ref.~\\cite{relyields}), the yield of neutral-current events~\\cite{SNsk} on $^{16}$O depends strongly on a possible chemical potential term in the thermal distribution.  It is important to note that the detection of recoil protons from {\\it neutron}-proton elastic scattering at several MeV has been routinely accomplished in scintillator detectors (see, e.g., Ref.~\\cite{routine}).  Since both particles are massive, the proton will typically take half of the neutron energy.  This reaction provides protons in the same energy range as those struck in neutrino-proton elastic scattering with $E_\\nu \\sim 30$ MeV.  This is a very important proof of concept for all aspects of the detection of low-energy protons.  Though low-energy backgrounds will be challenging, it is also important to note that the background requirements for detecting the supernova signal are approximately 3 orders of magnitude {\\it less} stringent than those required for detecting solar neutrinos in the same energy range (taking quenching into account for our signal). Borexino has been designed to detect very low-energy solar neutrinos, and KamLAND hopes to do so in a later phase of the experiment.  These measurements would be considered in combination with similar measurements for $\\nu_e$ and $\\bar{\\nu}_e$ from charged-current reactions in other detectors.  Separate measurements of the total energy and temperature for each flavor will be invaluable for comparing to numerical supernova models~\\cite{SNmodels,SNnew}.  They will also be required to make model-independent studies of the effects of neutrino oscillations~\\cite{SNosc}.  If the total energy release $E_B$ in all flavors has been measured, then \\begin{equation} E_B \\simeq \\frac{3}{5} \\frac{G M_{NS}^2}{R_{NS}}\\,, \\end{equation} thus allowing a direct and unique measurement of the newly-formed neutron star properties, principally the mass $M_{NS}$~\\cite{NS}."
    },
    "0209/astro-ph0209300_arXiv.txt": {
        "abstract": "The Rapid Telescopes for Optical Response (RAPTOR) experiment is a spatially distributed system of autonomous robotic telescopes that is designed to monitor the sky for optical transients. The core of the system is composed of two telescope arrays, separated by 38 kilometers, that stereoscopically view the same 1500 square-degree field with a wide-field imaging array and a central 4 square-degree field with a more sensitive narrow-field ``fovea\" imager.  Coupled to each telescope array is a real-time data analysis pipeline that is designed to identify interesting transients on timescales of seconds and, when a celestial transient is identified, to command the rapidly slewing robotic mounts to point the narrow-field ``fovea'' imagers at the transient. The two narrow-field telescopes then image the transient with higher spatial resolution and at a faster cadence to gather light curve information. Each ``fovea\" camera also images the transient through a different filter to provide color information.  This stereoscopic monitoring array is supplemented by a rapidly slewing telescope with a low resolution spectrograph for follow-up observations of transients and a sky patrol telescope that nightly monitors about 10,000 square-degrees for variations, with timescales of a day or longer, to a depth about 100 times fainter. In addition to searching for fast transients, we will use the data stream from RAPTOR as a real-time sentinel for recognizing important variations in known sources. All of the data will be publically released through a virtual observatory called SkyDOT (Sky Database for Objects in the Time Domain) that we are developing for studying variability of the optical sky. Altogether, the RAPTOR project aims to  construct a new type of system for discovery in optical astronomy---one that explores the time domain by ``mining the sky in real time\". ",
        "introduction": "\\label{sect:intro}  %  While it has been known for centuries that the optical sky is variable, monitoring the sky for optical transients with durations of less than a day is a rich area of research that remains largely unexplored.\\cite{Paczynski00,Nemiroff99} The fact that spectacular celestial transients exist was clearly demonstrated by the detection \\cite{Akerlof99}, with the Robotic Optical Transient Search Experiment I (ROTSE-I) telescope located at Los Alamos National Laboratory (LANL), of an optical transient associated with a Gamma Ray Burst  (GRB) at redshift z=1.6. The optical flash generated by that cosmological explosion lasted about 80 seconds and reached an astounding peak apparent magnitude of 9, making it the most luminous optical source ever measured by man ($M_V=-36.4$).  However, without the real-time position provided by a high-energy satellite that cued robotic optical telescopes to slew to the correct position, the remarkable transient, that was observable potentially even with binoculars, would have been missed.  There are also reasons to suspect the existence of celestial optical transients that cannot be found through sky monitoring by high-energy satellites. Theoretical models of GRBs have suggested that so-called optical ``orphan\" transients might be a thousand times more common than the GRBs\\cite{Granot02} and/or that optical transients could be precursors to GRBs\\cite{Paczynski02}. Further, our knowledge of the variability of optical sky is so incomplete that it is likely that there are new, as yet undiscovered, classes of rapid optical transients that are completely unrelated to high-energy transients. However, to search for optical transients, astronomers need a wide-field optical monitoring system that can autonomously locate, in real time, celestial transients with timescales as short as a minute. Such a system has never been constructed.  Beyond the study of optical transients with timescales of minutes, there is a long and diverse list of other interesting scientific subjects that need wide-field optical monitoring for efficient exploration.\\cite{Paczynski97,Paczynski01b}. The subjects range from searches for killer asteroids and extrasolar planetary systems, to eclipsing and pulsating stars, to novae and supernovae, to large amplitude outbursts from active galactic nuclei. Once found, those rare objects or fleeting events are best studied with large, narrow-field, telescopes---but those powerful instruments have very limited capability to find the interesting targets. Wide-field optical monitors can therefore enable otherwise impossible observations.  The idea of monitoring the dynamic optical sky is an old one, but it is only with 21st century information technology that one can seriously contemplate a systematic monitoring of all the sky for optical transients with durations as short as a minute. For ground-based telescopes, the optical sky contains half a trillion optical resolution elements. Monitoring the entire visible sky, with that resolution, for transients that last only a minute will generate a continuous data rate approaching 10 Gigabytes per second. To be effective, that huge data stream needs to be mined in real time to locate the transient and cue follow-up observations. Everyone agrees that such a system is likely to make important discoveries, but clearly the construction of such a high-resolution system that monitors the full sky, all of the time, is challenging. Here we discuss a program underway at Los Alamos National Laboratory that is taking the first steps toward that goal. ",
        "conclusions": "RAPTOR is designed to explore the dynamic optical sky for transients with durations as short as one minute. It is the first robotic optical telescope that autonomously finds and follows up on optical transients in real time. The core of the system is two wide-field telescope arrays, separated by 38 kilometers, that spectroscopically view the same 1500 square-degree piece of sky. The stereoscopic viewing, when coupled with a fast analysis pipeline, allows the system to reject false positives and robustly identify real celestial transients in real time. It also allows the system to track near earth objects that are closer than one quarter of the way to the moon in 3D. Having identified a real celestial transient, the system generates an alert and begins follow-up observations with higher spatial resolution fovea cameras and a low spectral resolution spectrograph. This integrated system will allow RAPTOR to explore a new region of discovery space in optical astronomy by mining the sky in real time."
    },
    "0209/astro-ph0209593_arXiv.txt": {
        "abstract": "\\baselineskip 10pt  Relativistic jets in Galactic superluminals and extragalactic AGN may be surrounded by a wind near to the central engine.  Theoretical analysis and numerical simulation reveal considerable stabilization  of relativistic jet flow by a wind to helical and higher order asymmetric modes of jet distortion.  When velocities are measured in the source (inlet) frame, reduction in the absolute velocity difference between jet and wind, $\\Delta v = v_{jet} - v_{wnd}$, provides stabilization in addition to stabilization provided by a high jet Lorentz factor, but a high Lorentz factor wind is not needed to stabilize a high Lorentz factor jet.   However, the fundamental pinch mode is not similarly affected and knots with spacing a few times the jet radius are anticipated to develop in such flows.  Thus, we identify a mechanism that can suppress large scale asymmetric structures while allowing axisymmetric structures to develop. Relativistic jets surrounded by outflowing winds will be more stable than if surrounded by a stationary or backflowing external medium.  Knotty structures along a straight jet like that in 3C\\,175 could be triggered by pinching of an initially low Mach number jet surrounded by a suitable wind.  As the jet enters the radio lobe, suppression of any surrounding outflow or backflow associated with the high pressure lobe triggers exponential growth of helical twisting. ",
        "introduction": "Numerical studies (see \\citet{mku01} and references therein) and theoretical work \\citep{bb99} indicate that relativistic jets may be surrounded by a more slowly moving wind.  Flow in the environment outside the jet can have important consequences for its stability, and the twisted normal mode structures that appear on resolved jets.  In this paper we present results from a numerical simulation of a precessed elliptically distorted relativistic jet.  The simulation was designed to explore the interaction between helical twisting arising from precession and the twin helically twisted filaments predicted to accompany elliptical distortion.  In the simulation the jet develops a velocity shear (wind) layer which has significant consequences for the evolution of the initial perturbation.  The results are relevant to the development of structures in relativistic Galactic and extragalactic jets.  \\vspace{-0.5cm} ",
        "conclusions": "In the simulation small perturbations, jet precession and elliptical cross section distortion, were applied at the inlet to study the interaction between helical and elliptical jet distortion.  Somewhat surprisingly, as the jet is predicted to be Kelvin-Helmholtz unstable, the numerical simulation showed a decline in the amplitude of the helical and elliptical surface mode perturbations across the computational grid.   The simulation also revealed a relatively short wavelength fundamental pinch mode, moving at nearly the jet speed and with a constant amplitude across the computational grid.  Although there was internal jet structure associated with pinch and elliptical body modes just outside the inlet, there was a lack of internal jet structure in the outer 3/4 of the computational grid.  Previous simulations at lower Lorentz factor contained internal structure across the computational grid \\citep{hhrg01}.  Development of a shear layer of thickness $\\Delta r \\lesssim$ 2R  at $r \\geq R$  outside the jet spine is responsible for the lack of growth of the asymmetric modes and for a larger growth rate for the fundamental pinch mode.   Similar results have been found theoretically and confirmed by simulation for non-relativistic MHD jets \\citep{hr02}. The effect of a shear layer, modeled as a simple wind, is shown to reduce the growth of perturbations significantly. Suppression of growth by a wind is related to reduction in the velocity shear $\\Delta v = v_j - v_e$ and a high Lorentz factor wind is not needed to stabilize a high Lorentz factor flow. The theory indicates that growth of helical and elliptical surface mode distortions is reduced by a factor of 2 for a c/2 wind and would be reduced by a larger factor for the $> c/2$ wind observed in the simulation at large distances from the inlet. Other change in the normal mode solutions within the computed frequency range, such as the disappearance of a maximum growth rate, greatly enhanced growth rate of the fundamental pinch mode, and suppression of the body mode growth rate occurs as the solutions are very sensitive to small changes in conditions when the velocity shear is weakly supersonic. Suppression of the body modes explains the lack of internal jet structure in the simulation.  Our present results suggest that a suitable shear layer or wind could partially stabilize a low Mach number relativistic astrophysical jet to helical and higher order modes of asymmetric jet distortion while leaving the fundamental pinch mode to grow.  This provides a trigger for knots moving with the jet speed and with spacing a few times the jet radius near to the central engine. From this result we conclude that jets and, by implication, the accretion process onto the central black hole can be steadier than previously thought, while still producing rapidly moving knots with quasi-periodic spacing.  We note that the mechanism found here is quite different from that investigated by \\citet{aetal01}, in which a shock moving with the jet fluid excited trailing components that could be identified with the first pinch body mode and were moving much slower than the jet fluid.  Spatial change in the wind speed such as a reduced wind speed at larger distance from the origin and/or decrease in sound speeds and increase in the jet Mach number would lead to a change in dynamical behavior. Rapidly moving knots near to the central engine might trigger and be replaced by more slowly moving knots associated with rapidly growing pinch body modes on the supersonic jet.  More rapidly growing asymmetric structures might also be expected to accompany this change. Additional rapid growth of asymmetries would accompany jet propagation through backflow from a high pressure radio lobe as a result of increase in the velocity shear, $\\Delta v = [v_j - (-v_e)]$, and significant increase in the growth rate (see eqs.[6a,b]) and decrease in the growth length (eq.[7]).  Knotty jet structure is observed in over half of the extended 3CR quasars observed by \\citet{betal94}.  The jets may be relatively straight over much of their length, e.g., 3C\\,175, or exhibit significant curvature on the parsec and larger scales, e.g., 3C\\,204 [see \\citet{betal94}; \\citet{hetal99}].  The jet in 3C\\.175 (Figure 7) provides a possible illustration of the effects of external flow on jet dynamics. \\vspace {3.0 in} \\begin{figure}[h] \\figurenum{7} \\special{psfile=3c175ac.ps hscale=120 vscale=120 voffset=-365 hoffset=-130} \\caption{\\footnotesize \\baselineskip 8pt Radio image of 3C\\,175 at 6~cm (image courtesy of NRAO). \\label{fig7t}} \\end{figure}  \\noindent A string of knots is observed along a straight jet that relatively abruptly bends and terminates in a hot spot and ``U'' shaped region at the outer edge of the radio lobe. VLBI observations suggest at least one knot like component beyond the core aligned with the large scale jet that could be moving superluminally \\citep{hetal02}, as would be expected if knots were associated with triggering of the fundamental pinch mode in a wind outflow near to the central engine.  At larger distance the jet is not likely embedded in a rapid outflow and knots could be slowly moving pinch body modes or simple line-of-sight crossing effects associated with twisted filaments related to the elliptical surface and/or elliptical body mode.  There is little appearance of asymmetric structure before the bend which is coincident with the trailing edge of the radio lobe in radio intensity images. The relative abruptness of the bend, although much less abrupt if deprojected, is suggestive of exponential growth of the helical mode, possibly triggered by backflow from the high pressure outer edge of the radio lobe."
    },
    "0209/astro-ph0209246_arXiv.txt": {
        "abstract": " ",
        "introduction": "In a series of papers exploring seven lines-of-sight (LOS) in the Local Interstellar Medium (LISM) the Far Ultraviolet Spectroscopic Explorer (FUSE) team (\\cite{f02}; \\cite{h02}; \\cite{k02}; \\cite{leh02}; \\cite{lem02}; \\cite{s02}; \\cite{w02}) has presented results on the column densities of \\d1 and \\o1 (along with \\n1, which will not be considered in this paper) but, not of \\h1 (due to the absence of Ly$\\alpha$ within the FUSE spectral range).  These data, which have been summarized in Moos \\etal (2002), are employed in the analysis presented here.  While five of the seven absorbing clouds lie within $\\sim 80$~pc of the Sun (within the Local Bubble), the other two clouds are further away, $\\sim 100 - 200$~pc.  Moos \\etal (2002) conclude that it is likely the deuterium abundance is represented by a single value for the five sightlines in the ``near\" LISM: D/H $ = 1.52 \\times 10^{-5}$.  While the uncertainty in this mean is $\\pm ~0.08 \\times 10^{-5}$, a better measure of the uncertainty might be the weighted standard deviation which is $\\pm ~0.18 \\times 10^{-5}$ (Moos, Private Communication).  It is also claimed that within the Local Bubble the \\d1/\\o1 ratio is constant and they suggest that, as a result, the \\o1 column densities can serve as a proxy for \\h1 in the Local Bubble. The FUSE team, while cautioning that their results are subject to small number statistics, note an increasing dispersion in \\d1/\\h1 with increasing distance from the Sun and suggest that this could be due to real variations among the LISM deuterium abundances.  However, there is no claim of evidence for an anti-correlation between \\d1 and \\o1 over the very limited range in metallicity they have explored thus far.  These issues are reconsidered here.  Using the FUSE data (specifically, Tables 3 \\& 4 of Moos \\etal 2002), and the same caveats concerning the limited size of their sample, it is shown that their data is not inconsistent with small, anti-correlated variations in D/H and O/H. If so, it becomes problematic to use \\o1 as a proxy for the \\h1 column densities undetermined from the FUSE data.  The question of variation or not can only be resolved by more data, especially, to echo the conclusion of Moos \\etal (2002), HST measurements of the \\h1 column densities and gas velocity structure.  However, if the variations suggested here are supported by further data, they offer the promise of sufficiently large \\d1/\\o1 variations within a few kpc of the solar system that further FUSE data should have no trouble digging the signal out of the noise.  In \\S2 the FUSE data is used to address the question of variablility in D/H, O/H, and \\d1/\\o1 in the LISM.  Having raised the possibility of variability, the correlations of D/H and \\d1/\\o1 with O/H are further explored in \\S3 where it is suggested that D/H may be anti-correlated with O/H.  In \\S4 two, likely extreme, forms for such variation are considered and compared and the corresponding predicted and FUSE-derived abundances of deuterium and oxygen are compared.  In \\S5 our conclusions and the prospects for future resolution of the issues raised here are discussed. ",
        "conclusions": "The FUSE data have been used to revisit the question of possible abundance variations in the LISM.  The sample is painfully limited (seven LOS; only five with \\h1 column density determinations with quoted uncertainties) but, within the statistical errors, the ananlysis presented here provides a hint of some variations in the local oxygen abundance (by the excess dispersion around the mean abundance) which may be anticorrelated with some variations in the LISM deuterium abundance. This is in contrast to the conclusions of Moos \\etal (2002).  Among the seven FUSE LOS and the two additional LOS considered by Moos \\etal (2002), two outliers are identified: BD~+28$^{\\circ}$4211 and $\\delta$~Ori~A.  The former, from the FUSE data set, has the smallest statistical errors for the D- and O-abundances, and thus dominates the FUSE mean abundance determinations (largely due to the very small error adopted for the \\h1 column density determination).  When this LOS is excluded from the sample, the mean D- and O-abundances increase slightly: $<y_{\\rm D}> = 1.7 \\pm 0.1$, $<y_{\\rm O}> = 3.9 \\pm 0.3$.  The remaining FUSE data, while not inconsistent with a constant D-abundance in the LISM, still have an unexpectedly large dispersion around the mean O-abundance, suggesting that there may be real oxygen abundance variations along nearby LOS.  If, indeed, there are variations in O/H in the LISM, they might be {\\it anti}-correlated with variations in D/H since as gas is cycled through stars deuterium is destroyed. The FUSE data set is, indeed, not inconsistent with a constant product of deuterium and oxygen abundances.  If this anticorrelation is confirmed by further data, there is both good news and bad news.  The bad news is that as FUSE expands its horizon beyond the LISM, it is unlikely that the ratio of \\d1 to \\o1 column densities ($z \\equiv 10^{2}$D/O) can serve as a surrogate for independent \\h1 column density measurements in the determination of D- and O-abundances.  The good news is that even within a few kpc of the Sun, based on estimates of the oxygen and deuterium abundance gradients in the Galaxy (\\cite{mv00}, \\cite{cm00}) $y_{\\rm O}$ and $y_{\\rm D}$ will vary sufficiently so that the amplification of their ratio, $z$, will result in $z$-variations (\\eg by roughly a factor two over $\\sim 2$~kpc) which will be more easily seen above the background of the statistical uncertainties.  It should be noted that even if the rather strong anticorrelation, consistent with the current FUSE data set, is confirmed locally, such a strong anticorrelation is unlikely to extend to much lower oxygen abundances.  Indeed, as pristine gas from the early universe begins to be processed through stars, the heavy element abundances, oxygen in this case, will quickly increase from their zero primordial values before very much gas has been cycled through stars, destroying deuterium.  As a result, for a long time (as measured by metallicity) the deuterium abundance will not deviate noticeably from its relic value, while the oxygen abundance will increase by orders of magnitude (the deuterium ``plateau'').  For example, if within the Galaxy a factor two lower oxygen abundance (than in the LISM) were accompanied by a factor two higher deuterium abundance, the result would be a D-abundance indistinguishable from the current estimates of the relic primordial D-abundance inferred from observations of gas in high redshift, low metallicity QSO Absorption Line Systems (\\cite{bt98a};~\\cite{bt98b}; ~\\cite{om01};~\\cite{pb01};~\\cite{ddm01};~\\cite{lev02}).  Indeed, the mean LISM D-abundance proposed here, $y_{\\rm D} = 1.7 \\pm 0.1$, is already indistinguishable from that suggested by Pettini \\& Bowen (2001; PB) for a high redshift (z~$ \\sim 2$), low metallicity ([Si/H]~$ \\la -2$) QSOALS: $y_{\\rm D}({\\rm PB}) = 1.65 \\pm 0.35$. The deuterium abundances derived from observations of the other QSOALS range from $y_{\\rm D}({\\rm QSOALS}) \\approx 2.5$ to 4.0. Therefore, it might be anticipated that future FUSE data along LOS within a few kpc of the Sun might be capable of mapping the evolution of deuterium back to the primordial deuterium plateau, providing a valuable complement to the very difficult searches for primordial-D in the QSOALS.  \\noindent {\\bf Acknowledgments}  I gratefully acknowledge helpful and informative correspondence with several members of the FUSE team, in particular Ed Jenkins, Warren Moos, Ken Sembach and George Sonneborn.  I also thank the referee for several helpful suggestions which have been incorporated in the manuscript. This research is supported at The Ohio State University by DOE grant DE-AC02-76ER-01545.  Some of this work was done while visiting IAGUSP, Brasil and I thank them for their hospitality."
    },
    "0209/astro-ph0209520_arXiv.txt": {
        "abstract": "It is now commonly believed that Soft gamma-ray repeaters (SGRs) and Anomalous X-ray pulsars (AXPs) are magnetars --- neutron stars powered by their magnetic fields.  However, what differentiates these two seemingly dissimilar objects is, at present, unknown.  We present \\chandra\\ observations of \\rxj, the quiescent X-ray counterpart of \\sgr, famous for the intense burst of 5 March 1979.  The source is unresolved at the resolution of \\chandra.  Restricting to a period range around 8~s, the period noted in the afterglow of the burst of 5 March 1979, we find evidence for a similar periodicity in two epochs of data obtained 20 months apart. The secular period derivative based on these two observations is $6.6(5)\\times 10^{-11}\\,$s$^{-1}$, similar to the period derivatives of the magnetars.  As is the case with other magnetars, the spectrum is best fitted by a combination of a black body and a power law.  However, quite surprisingly, the photon index of the power law component is $\\Gamma\\sim 3$ --- intermediate to those of AXPs and SGRs.  This continuum of $\\Gamma$ leads us to suggest that the underlying physical parameter which differentiates SGRs from AXPs is manifested in the power law component. Two decades ago, \\sgr\\ was a classical SGR whereas now it behaves like an AXP. Thus it is possible that the same object cycles between SGR and AXP state.  We speculate that the main difference between AXPs and SGRs is the geometry of the $B$-fields and this geometry is time dependent.  Finally, given the steep spectrum of \\rxj, the total radiated energy of \\rxj\\ can be much higher than traditionally estimated. If this energy is supplied by the decay of the magnetic field then the inferred $B$-field of \\rxj\\ is in excess of $10^{15}\\,$G, the traditional value for magnetars. Independent of this discussion, there could well be a class of neutron stars, $10^{14}\\ale B \\ale 10^{15}\\,$G, which are neither radio pulsars nor magnetars. ",
        "introduction": "\\label{sec:introduction}  The soft gamma-ray repeater \\sgr\\ played a key role in our understanding of high energy transients. It was from this source that an intense burst was observed on 5 March, 1979 \\citep{mgia+79,cdpt+80}. The burst was followed  by an ``afterglow'' emission with an apparent 8-s periodicity.  The source of the burst was quickly localized to the supernova remnant N49 (also known as SNR 0525$-$66.1) in the Large Magellanic Cloud \\citep{eklc+80}.  Observations with \\rosat\\ identified a quiescent and bright ($L_X\\sim 10^{36}\\,$ erg s$^{-1}$) X-ray counterpart, \\rxj\\ \\citep*{rkl94}.   The intense burst of 5 March 1979 and the luminous afterglow with 8-s periodicity provided the first and strongest evidence for super-strong magnetic field strengths, $B\\sim 10^{15}\\,$G. Such strong fields are needed to both confine the radiating plasma as well as allow the radiation to escape \\citep{dt92,paczynski92}. However, such highly magnetized neutron stars or ``magnetars'' were originally motivated by theoretical considerations --- namely strong convection would naturally lead to growth of magnetic fields during the process of the collapse of the proto-neutron star core \\citep{dt92,td93}.  Separately, another group of neutron stars, the so-called Anomalous X-ray pulsars (AXPs), were recognized as a new class of neutron stars (\\citealt*{pth95}; \\citealt{ms95}). The AXPs were noted for a narrow period distribution, between 6 and 20 s; luminous X-ray emission, $L_X\\sim 10^{35}\\,$erg and apparent lack of a donor star. The sources were ``anomalous'' in that the source of the quiescent emission was neither rotational (from the known $\\dot P$) nor accretion (apparent lack of companion).  Various authors speculated and suggested that AXPs are also magnetars --- specifically, their X-ray emission to arise from the decay of a magnetar-like field strength \\citep{td93}.   \\begin{deluxetable}{r  c c c c l l} \\tablecaption{Position of \\rxj} \\tablewidth{0pt} \\tablehead{ \\colhead{ObsId} & & \\colhead{$x$} & \\colhead{$y$} & & \\colhead{$\\alpha-05^{\\rm h}26^{\\rm m}$} & \\colhead{$\\delta+66\\degr04\\arcmin$} \\\\ \\cline{3-4} \\cline{6-7} & & \\mc{2}{c}{(pixels)} & & \\colhead{(sec)} & \\colhead{(arcsec)} \\\\ } \\startdata 747  && 4160.596(8) & 4135.884(8) && 00.8791(6) & $-$36.180(4)\\\\ 1957 && 4090.665(8) & 4025.371(8) && 00.9094(6) & $-$36.424(4)\\\\ 2515 && 4091.27(3)  & 4025.06(3)  && 00.911(4) &  $-$36.45(1)\\\\ \\tableline Average && & && 00.8948(4) & $-$36.307(3) \\\\ \\enddata \\tablecomments{Positions are J2000.  The values in parentheses above are 1-$\\sigma$ statistical uncertainties.  There is an additional 1-$\\sigma$ position uncertainty of $\\approx 0\\farcs6$ in each coordinate due to aspect uncertainties.} \\label{tab:position} \\end{deluxetable}  The discovery of periodicity in SGRs \\citep{kds+98} and the overlap of $P$ and $\\dot P$ between AXPs and SGRs continued to motivate a unified magnetar framework for both these objects. In particular, the magnetic field strength inferred from $P$ and $\\dot P$ (vacuum dipole framework) led to estimates of about $10^{14}\\,$G for both these objects, within a factor of few of that estimated for AXPs and SGRs.  Toward the end of nineties, thanks to large area radio pulsar searchers, astronomers became aware of a growing group of radio pulsars \\citep{ckl+00} with similarly long periods and with inferred magnetic field strengths approaching $10^{14}$\\,G (hereafter HBPSRs).  These pulsars possess no special attributes linking them to either the AXPs (no steady bright quiescent X-ray emission; \\citealt*{pkc00}) or the SGRs (no bursting history).  Thus periodicity alone does not appear to be a sufficient attribute for classification.   Nonetheless, the recent discovery of bursts of radiation --- similar to the minor bursts seen from SGRs --- from two AXPs are strong empirical confirmation of a link between AXPs and SGRs (\\citealt*{gkw02}; \\citealt{kg02}).  However, we are still at a loss what specific physical parameter[s] differentiates AXPs from SGRs.   One plausible notion is that AXPs and SGRs are linked temporally. Specifically, three out of the six AXPs are associated with supernova remnants (SNRs) whereas only \\sgr\\ has a plausible SNR association \\citep{gsgv01}.  Taken at face value, these data suggest that AXPs evolve into SGRs.   However, this hypothesis has two problems.  First, the rotational periods of SGRs are similar to those of AXPs, about 10-s.  Second, inferred magnetic field strengths of SGRs are similar to (and perhaps even larger than) those of AXPs \\citep{hurley99,mereghetti99}.  Thus, there is no strong period or B-field evolution between the two groups.  In our opinion, the above two objections are sufficiently severe that we must continue searching for underlying physical parameter[s] that differentiates between AXPs and SGRs.  To this end, investigating the properties of the quiescent emission, which in practice means spectroscopic and rotational properties, appear promising. Here we report investigation of the quiescent X-ray emission of \\sgr, comparing and contrasting the quiescent emission with those of AXPs and other SGRs. ",
        "conclusions": "Here, we report \\chandra\\ observations of the X-ray counterpart of \\sgr. We have three primary results from these observations: (1) We have determined an accurate position for \\rxj\\ (Table~\\ref{tab:position}).  (2) We can rule out pure blackbody (BB) model for the X-ray spectrum.  Instead we find that the best fit model requires both a BB component and a power-law (PL) component; the photon index, $\\Gamma\\sim 3.1$, is steep (Table~\\ref{tab:specfit}).  (3) Restricting the period search to a range of 8-s (and its harmonic) we detect periodicity with $P\\sim 8\\,$s in both datasets. If we assume the period evolves secularly then $\\dot P\\sim 6.5\\times 10^{-11}$ s s$^{-1}$.  We now discuss these points in more detail.   The accurate position\\footnotemark\\footnotetext{The position reported here has been corrected using the latest aspect solutions and has higher precision than that given in \\cite{kkvk+01}.} in conjunction with \\textit{Hubble Space Telescope} (\\hst) images enabled us to place the most stringent limits to the optical emission from \\rxj\\ \\citep{kkvk+01}. These are the best limits to quiescent optical/IR emission from an SGR. In particular, in \\citet{kkvk+01} we investigated $F_{XR}$, the ratio of the integrated flux in the X-ray band (i.e.\\ $\\nu f_\\nu$) to that in the optical R band. As noted by \\citet*{hvkk00}, AXPs are distinguished by an unusually large $F_{XR}\\sim 10^4$.  \\rxj\\ possesses a similarly large $F_{XR}$ \\citep{kkvk+01} --- further evidence of commonality between \\sgr\\ and the AXPs.   Next, we draw attention to the fact that $\\Gamma$ of \\rxj\\ is decidedly steeper than the value of $\\sim$2 found for the quiescent emission from other SGRs \\citep{hurley99,ktw+01,fkkf01}, but is similar to the values of 3 to 4 for AXPs \\citep{mereghetti99}.  We view this similarity with considerable interest since AXPs are unique among neutron stars for their steep spectra.  Furthermore, we note that a significant fraction of luminosity for both SGRs and AXPs comes out in the X-ray band. Thus any commonality in the X-ray spectrum takes on additional importance.  Indeed, spectral dissimilarity is the reason why the 7.7-s X-ray pulsar 4U 1626$-$67 is not considered to be an AXP even though this source shares many attributes with AXPs but has a flat X-ray spectrum \\citep{awnk+95}.   The possible detection of periodicity in the quiescent emission with $P\\sim 8$, similar to the value of the period in the afterglow of 5 March 1979 \\citep{mgia+79,cdpt+80} is in accord with what has been seen in other SGRs. In particular, a period of about 5-s was detected in the afterglow of the giant flare of 27 August 1998 from SGR 1900+14 \\citep{fhd+01} and a similar period was also noted in the quiescent emission \\citep{hlkm+99}. Returning to \\rxj\\, if we accept the $\\dot P$ (based on only two epochs) represents the secular period derivative, then the characteristic age, $P/2\\dot P\\sim 2,000\\,$yr and inferred vacuum dipole field strength, $B^2=10^{39} P\\dot P$, is $B\\sim 7\\times 10^{14}\\,$G. The age is comparable to the estimated age of the SNR N49, $\\sim 5000\\,$yr \\citep{vblr92} and the inferred $B$ values are similar to those inferred for other magnetars and AXPs \\citep{kds+98,mereghetti99}."
    },
    "0209/astro-ph0209185_arXiv.txt": {
        "abstract": "One of the observational evidences in support of the {\\it thermonuclear runaway model} for the classical nova outburst relies on the accompanying nucleosynthesis. In this paper, we stress the relevant role played by nucleosynthesis in our understanding of the nova phenomenon by constraining models through a comparison with both the atomic abundance determinations from the ejecta and the isotopic ratios measured in presolar grains of a likely nova origin. Furthermore, the endpoint of nova nucleosynthesis provides hints for the understanding of the mixing process responsible for the enhanced metallicities found in the ejecta, and reveals also information on the properties of the underlying white dwarf (mass, luminosity...).  We discuss first the interplay between nova outbursts and the Galactic chemical abundances: Classical nova outbursts are expected to be the major source of $^{13}$C, $^{15}$N and $^{17}$O in the Galaxy, and to contribute to the abundances of other species with $\\rm A < 40$, such as $^7$Li or $^{26}$Al. We describe the main nuclear path during the course of the explosion, with special emphasis on the synthesis of radioactive species, of particular interest for the gamma-ray output predicted  from novae ($^7$Li, $^{18}$F, $^{22}$Na, $^{26}$Al). An overview of the recent discovery of presolar nova candidate grains, as well as a discussion of the role played by nuclear uncertainties associated with key reactions of the NeNa-MgAl and Si-Ca regions, are also given. ",
        "introduction": "The high peak temperatures achieved during nova explosions, T$_{peak} \\sim (2-3) \\times 10^8$ K, suggest that abundance levels of the intermediate-mass elements in the ejecta must be significantly enhanced, as confirmed by spectroscopic determinations in well-observed nova shells. This raises the issue of the potential contribution of novae to the Galactic abundances, which can be roughly estimated as the product of the Galactic nova rate, the average ejected mass per nova outburst, and the Galaxy's lifetime. This order of magnitude estimate points out that novae scarcely contribute to the Galaxy's overall metallicity (as compared with other major sources, such as supernova explosions), nevertheless they can substantially contribute to the synthesis of some largely overproduced species (see Table 1, for a sample of publications addressing nucleosynthesis in classical novae). Hence, classical novae are likely sites for the synthesis of most of the Galactic $^{13}$C, $^{15}$N and $^{17}$O, whereas they can partially contribute to the Galactic abundances of other species with $\\rm A < 40$, such as $^{7}$Li, $^{19}$F, or $^{26}$Al \\cite{Sta98,Jos98}.  \\begin{figure} \\includegraphics[height=.4\\textheight,clip=]{T1.eps} \\includegraphics[height=.4\\textheight,clip=]{T3.eps} \\caption{(Left) Mean overproduction factors, relative to solar,  in the ejecta of a 1.15 M$_\\odot$ CO novae. (Right) Same for a 1.35 M$_\\odot$ ONe novae. Figures are based on hydrodynamic calculations reported in \\cite{Jos98}.} \\end{figure}  Overproduction factors, relative to solar, corresponding to hydrodynamic calculations of nova outbursts on top of a 1.15 M$_\\odot$ CO and a 1.35 M$_\\odot$ ONe white dwarf, are shown in Figure 1. Because of the lower peak temperatures achieved in CO models, and also because of the lack of significant amounts of seed nuclei in the NeNa-MgAl region, the main nuclear activity in CO novae does not extend much beyond oxygen, as seen from the overproduction plot. In contrast, ONe models show a much larger nuclear activity, extending up to silicon (1.15 M$_\\odot$ ONe) or argon (1.35 M$_\\odot$ ONe).  Hence, the presence of significantly large amounts of intermediate-mass nuclei in the spectra, such as phosphorus, sulfur, chlorine or argon, may reveal the presence of an underlying massive ONe white dwarf. Another trend derived from the analysis of the nucleosynthesis accompanying nova outbursts is the fact that the O/N and C/N ratios decrease as the mass of the white dwarf (and hence, the peak temperature attained during the explosion) increases.  \\subsection{Abundance Determinations in the Ejecta from Novae}  In order to constraint the models, several works have focused on a direct comparison of the atomic abundances inferred from observations of the ejecta with the theoretical nucleosynthetic output (see \\cite{Jos98,Sta98}, and references therein). Despite of the problems associated with the modeling of the explosion \\cite{Sta02}, such as the unknown mechanism responsible for the mixing between the accreted envelope and the outermost shells of the underlying white dwarf \\cite{Cal02,Dur02}, or the difficulties to eject as much material as inferred from observations \\cite{Sho02}, there is an excellent agreement between theory and observations as regards nucleosynthesis (i.e.,  including atomic abundances -H, He, C, O, Ne, Na-Fe-, and a plausible endpoint for nova nucleosynthesis). In some cases, such as for PW Vul 1984, the agreement between observations and theoretical predictions (see \\cite{Jos98}, Table 5, for details) is really overwhelming. The reader is referred to \\cite{Geh98} for an extended list of abundance determinations in the ejecta from novae, and to \\cite{Sch02, Van02} for recent efforts to improve the abundance pattern for QU Vul 1984 and V1974 Cyg 1992, respectively.  Since the nuclear path is very sensitive to details of the evolution (chemical composition, extend of convective mixing, thermal history of the envelope...), the agreement between inferred abundances and theoretical yields not only validates the thermonuclear runaway model, but also poses limits on the (yet unknown) mixing mechanism itself: for instance, if mixing occurs very late in the course of the explosion, the accumulation of larger amounts of matter in the envelope will be favored (since the injection of significant amounts of the triggering nucleus $^{12}$C will be delayed). Hence, one would expect to end up with a more violent outburst, characterized by a higher T$_{peak}$, exceeding in some cases $4 \\times 10^8$ K, and, as a result, a significant enrichment in heavier species, beyond calcium, in the ejecta from novae involving very massive white dwarfs, a pattern never observed so far.  \\subsection{Presolar Grains: Gifts from Heaven}  Infrared \\cite{Eva90,Geh98} and ultraviolet observations \\cite{Sho94} of the temporal evolution of nova light curves suggest that  novae form grains in the expanding nova shells. Both CO and ONe novae behave similarly in the infrared right after the outburst. However, as the ejected envelope expands and becomes optically thin, such behavior dramatically changes: CO novae are typically followed by a phase of dust formation corresponding to a decline in visual light, together with a simultaneous rise in the infrared emission \\cite{Eva02,Geh02}. In contrast, it has been argued that ONe novae (that involve more massive white dwarfs than CO novae) are not so prolific producers of dust as a result of the lower mass, high-velocity ejecta, where the typical densities can be low enough to enable the condensation of appreciable amounts of dust. Hints on the condensation of dust containing silicates, silicon carbide, carbon and hydrocarbons have been reported from a number of novae (see \\cite{Geh98} for a recent review).  Up to now, the identification of presolar nova grains, presumably condensed in the shells ejected during the explosion, relied only on low $^{20}$Ne/$^{22}$Ne ratios (attributed to $^{22}$Na decay), but quite recently five silicon carbide and two graphite grains that exhibit isotopic signatures characteristic of nova nucleosynthesis have been identified \\cite{Ama01,Ama02}. They are characterized by very low $^{12}$C/$^{13}$C and $^{14}$N/$^{15}$N ratios,  $^{30}$Si excesses and close-to- or slightly lower-than-solar $^{29}$Si/$^{28}$Si ratios, high $^{26}$Al/$^{27}$Al ratios (determined only for two grains) and low $^{20}$Ne/$^{22}$Ne ratios (only measured in the graphite grain KFB1a-161). Such a promising discovery provides a much valuable source of constraint for nova nucleosynthesis (since contrary to the atomic abundance determinations derived from nova ejecta, measurements provide more accurate isotopic ratios) and opens interesting possibilities for the future.  Theoretical isotopic ratios for a variety of nuclear species, ranging from C to Si, based on hydrodynamic computations of the nova outburst, have been reported by different authors \\cite{Sta97,Jos01b,Jos01d,Jos02a}. A more detailed analysis, which focuses on the different chemical pattern expected for CO and ONe novae, will be presented elsewhere \\cite{Jos02b}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209466_arXiv.txt": {
        "abstract": "We present a new source separation method which maximizes the likelihood of a model of \\emph{noisy} mixtures of stationary, possibly Gaussian, independent components.  The method has been devised to address an astronomical imaging problem.  It works in the spectral domain where, thanks to two simple approximations, the likelihood assumes a simple form which is easy to handle (low dimensional sufficient statistics) and to maximize (via the EM algorithm). ",
        "introduction": "\\subsection{Astronomical components}\\label{eq:astro}  Source separation consists in recovering components from a set of observed mixtures. Component separation is a topic of major interest to the Planck space mission, to be launched in 2007 by ESA to map the cosmic microwave background (CMB).  The blackbody temperature of this radiation as a function of direction on the sky will be measured in $m=10$ different frequency channels, corresponding to wavelengths ranging from $\\lambda=350\\,$microns to $\\lambda=1\\,$cm.  In each channel, the temperature map will show not only the CMB contribution but also contributions from other sources called \\emph{foregrounds}, among which Galactic dust emission, emission from very remote (and hence quasi point-like) galaxy clusters, and several others. It is expected that (after some heavy pre-processing), the map built from the $i$-th channel can be accurately modeled as $y_i(\\vec r)=\\sum_{j=1}^{n}a_{ij} s_j(\\vec r) + n_i(\\vec r)$ where $s_j(\\vec r)$ is the spatial pattern for the $j$-th component and $n_i(\\vec r)$ is an additive detector noise. In other words, cosmologists expect to observe a noisy instantaneous (\\emph{i.e.} non convolutive) mixture of essentially independent components (independence being the consequence of the physically distinct origins of the various components). Even though recovering as cleanly as possible the CMB component is the primary goal of the mission, astrophysicists are also interested in the \\emph{other} components, in particular for collecting data regarding the morphology and physical properties of Galactic foregrounds (dust\\ldots) and the distribution of galaxy clusters.    This paper deals with \\emph{blind} component separation.  Blindness means recovering the components without knowing in advance the coefficients of the mixture.  In practice, this may be achieved by resorting to the strong but often plausible assumption of mutual statistical independence between the components. The motivation for a blind approach is obvious: even though some coefficients of the mixture may be known in advance with good accuracy (in particular those related to the CMB), some other components are less well known or predictable.  It is thus very tempting to run blind algorithms which do not require \\emph{a priori} information about the mixture coefficients.     \\subsection{Blind separation methods}\\label{sec:bsm}    Several attempts at blind component separation for CMB imaging have already been reported.  The first proposal, due to Baccigalupi \\textit{et al.} use a non Gaussian noise-free i.i.d. model for the components\\cite{2000MNRAS.318..769B}, hence following the `standard' path to source separation.  One problem with this approach is that the most important component, namely the CMB itself, seems to closely follow a Gaussian distribution.  It is well known that, in i.i.d. models, it is possible to accommodate at most one Gaussian component. It does not seem to be a good idea, however, to use a non Gaussian model when the main component itself has a Gaussian distribution.  Another reason why the i.i.d. modeling (which is implicit in `standard' ICA) probably is not appropriate to our application: most of the components are very much dominated by the low-frequency part of their spectra.  Thus sample averages taken through the data set tend not to converge very quickly to their expected values.  This may explain why Fourier methods, presented below, seem to perform better.   Thus, rather than exploiting the non Gaussianity of (all but one of) the components, one may think of exploiting their spectral diversity. A very sensible approach is proposed by Pham: using the Whittle approximation of the likelihood, he shows that blind separation can be achieved by jointly diagonalizing spectral covariance matrices computed over several frequency bands~\\cite{pham:eusipco2000}.  This conclusion however is reached only in the case of noise-free models. Therefore, it is not appropriate for CMB imaging where a very significant amount of noise is expected.  In this paper, we follow Pham's approach but we take additive noise into account, leading to a likelihood function which is no longer a joint diagonality criterion, thus requiring some new algorithmics.  We present below the form taken by the EM algorithm when applied to a set of spectral covariance matrices.  This approach leads to an efficient algorithm, much faster than the algorithms obtained via the EM technique in the case of non Gaussian i.i.d. modeling as in~\\cite{EMMG:icassp97} or~\\cite{snoussi:MaxEnt2001}.   \\subsection{A stationary Gaussian model}  Our method is obtained by starting from a stationary Gaussian model. For ease of exposition, we assume that the observations are $m$ times series rather than $m$ images (extension to images is straightforward).  The $m\\times 1$-dimensional observed process $y(t)=[y_1(t); \\ldots; y_m(t)]$ is modeled as \\begin{equation} \\label{eq:model} y(t)=As(t) + n(t) \\end{equation} where $A$ is an $m\\times n$ matrix with linearly independent columns. The $n$-dimensional source process $s(t)$ (the components) and the $m$-dimensional noise process $n(t)$ are modeled as real valued, mutually independent and stationary with spectra $S_s(\\nu )$ and $S_n(\\nu )$ respectively.  The spectrum of the observed process then is \\begin{equation} \\label{eq:specmody} S_y(\\nu ) = A S_s(\\nu ) A\\adj + S_n(\\nu ) . \\end{equation} The $\\dagger$ superscript denotes transconjugation even though transposition would be enough almost everywhere in this paper (our method is easily adapted to deal with complex signals/mixtures). The assumption of independence between components implies that $S_s(\\nu )$ is a diagonal matrix: \\begin{displaymath} [ S_s(\\nu ) ]_{ij} = \\delta_{ij} P_i(\\nu ) \\end{displaymath} where $P_i(\\nu )$ is the power spectrum of the $i$th source at frequency $\\nu $. For simplicity, we also assume that the observation noise is uncorrelated both in time and across sensors: \\begin{equation} \\label{eq:whitenoise} [S_n(\\nu )]_{ij} = \\delta_{ij} \\sigma_i^2 \\end{equation} meaning that the noise spectral density $\\sigma_i^2$ on the $i$th detector does not depend on frequency $\\nu $. In summary the probability distribution of the process is fully specified by $m\\times n$ mixture coefficients, $m$ noise levels and $n$ power spectra. ",
        "conclusions": ""
    },
    "0209/astro-ph0209095_arXiv.txt": {
        "abstract": "In this contribution, first results of deep VLT photometry ($V,I$) in the central region of the Hydra~I and Centaurus galaxy clusters are presented. In both galaxy clusters, many star clusters have been identified down to the turnover magnitude of the globular cluster luminosity function at $V\\simeq26.0$ mag. They are distributed not only around the several early-type galaxies, but also in the intra-cluster field, as far as 250 kpc from the cluster centers. Outside the bulges of the central galaxies in Hydra~I and Centaurus, the intra-cluster globular cluster system is dominated by blue clusters whose spatial distribution is similar to that of the (newly discovered) dwarf galaxies. ",
        "introduction": "The centers of galaxy clusters are the densest regions of galaxy populations in the Universe. They are the places where the most frequent interactions between galaxies are expected to have taken place during the cluster formation epoch (and maybe also in the present). Some striking properties of galaxy cluster centers are: 1) a very rich globular cluster system (GCS) around the central galaxy (e.g. Harris 1991), 2) an extended stellar halo (cD halo) around the central galaxy (e.g. Schombert 1988), and 3) an abundant population of early-type dwarf galaxies clustered towards the center (e.g. Ferguson \\& Binggeli 1994). How do these findings come together? Can they be the result of a common scenario in which galaxy disruption played a major role (see Hilker et al. 1999)? Nearby galaxy clusters provide an ideal laboratory to study the different stellar components in detail.  The Hydra~I Galaxy cluster is dynamically evolved, has a regular core shape and an isothermal X-ray gas halo that can be followed out to about 160 kpc. The Centaurus cluster is dynamically young with two merging sub-groups, a main cluster component (Cen30) around the cD galaxy NGC~4696 and a smaller group component (Cen45) around NGC~4709. Both galaxy clusters are located at a distance of about 45 Mpc.  Both galaxy clusters were observed at dark time and under photometric conditions in the filters $V$ and $I$ with FORS1 at the VLT (ESO, Paranal). The seeing in all fields was in the range 0.5 to 0.7 arcsec, thus providing a very homogeneous data set.  \\begin{figure} \\plotone{mhilker.f1.eps} \\caption{The distribution of blue (light dots) and red (dark dots) globular clusters in the center of the Hydra~I (left) and Centaurus (right) cluster is shown. The circles indicate the location of the major galaxies, and bright triangles are dwarf galaxy candidates in Hydra~I. } \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209430_arXiv.txt": {
        "abstract": "This article presents and discusses a method for measuring the proper motions of the Galactic dwarf spheroidal galaxies using images taken with the Hubble Space Telescope.  The method involves fitting an effective point spread function to the image of a star or quasi-stellar object to determine its centroid with an accuracy of about 0.005~pixel (0.25~milliarcseconds) --- an accuracy sufficient to measure the proper motion of a dwarf spheroidal galaxy using images separated by just a few years.  The data consist of images, dithered to reduce the effects of undersampling, taken at multiple epochs with the Space Telescope Imaging Spectrograph or the Wide Field Planetary Camera.  The science fields are in the directions of the Carina, Fornax, Sculptor, and Ursa Minor dwarf spheroidal galaxies and each has at least one quasi-stellar object whose identity has been established by other studies.  The rate of change with time of the centroids of the stars of the dwarf spheroidal with respect to the centroid of the quasi-stellar object is the proper motion.  Four independent preliminary measurements of the proper motion of Fornax for three fields agree within their uncertainties.  The weighted average of these measurements is $\\mu_{\\alpha} = 49 \\pm 13$~milliarcseconds~century$^{-1}$ and  $\\mu_{\\delta} = -59 \\pm 13$~milliarcseconds~century$^{-1}$.  The Galactocentric velocity derived from the proper motion implies that Fornax is near perigalacticon, may not be bound to the Milky Way, and is not a member of any of the proposed streams of galaxies and globular clusters in the Galactic halo.  If Fornax is bound, the Milky Way must have a mass of at least $(1.6 \\pm 0.8) \\times 10^{12}~\\mathcal{M}_{\\odot}$. ",
        "introduction": "\\label{intro}  There are nine known dwarf spheroidal (dSph, hereafter) galaxies in relative proximity to the Milky Way, (see Mateo 1998 and van den Bergh 2000 for reviews).  The nearest, Sagittarius, is only about 16~kpc from the Galactic center and it is moving on a polar orbit.  The most distant dSph, Leo~I, is 250~kpc from the Galactic center.  The average distance from the Sun of the whole population of dSphs is roughly 100~kpc.  Although these distances are small compared to the size of the Local Group, they are sufficiently large to have made proper motion measurements very difficult using images obtained with ground-based telescopes.  The effect of large distance on the size of the proper motion can be lessened by a long interval between image epochs --- the time baseline.  However, many of the nine dSphs were discovered only in the past several decades (the most recent, Sagittarius, in 1994) and, thus, the available images of dSphs with a quality high enough for proper motion measurements have insufficient time baselines.  Despite these difficulties, several groups have reported proper motions for a few dSphs.  Scholtz~\\&~Irwin (1993) used Schmidt plates with a time baseline of about 35~years to measure a proper motion, $(\\mu_{\\alpha},\\mu_{\\delta})$\\footnote{Throughout this article, $\\mu_{\\alpha}$ is the proper motion in arcseconds in the direction of increasing right ascension, \\textit{i.e.}, the actual shift on the sky in that direction.  The same is true for $\\mu_{\\ell}$, the proper motion in the direction of increasing galactic longitude.}, for Draco of $(90\\pm50,100\\pm50)$ milliarcseconds per century (mas~cent$^{-1}$, hereafter) and a proper motion for Ursa Minor of $(100\\pm80,100\\pm80)$~mas~cent$^{-1}$.  Schweitzer~\\etal (1995) used a variety of larger-scale plates with a time baseline of about 50~years to measure a proper motion for Sculptor of $(36\\pm22,43\\pm25)$~mas~cent$^{-1}$.  Similarly, Schweitzer \\etal (1997) used Palomar 5~m and KPNO 4~m plates with a time baseline of 42~years to measure a proper motion for Ursa Minor of $(58\\pm8,26\\pm10)$~mas~cent$^{-1}$.  Finally, Irwin \\etal\\ (1996) used Schmidt plates to measure a preliminary proper motion for Sagittarius of 210$\\pm$70~mas~cent$^{-1}$.  All of the above proper motions are corrected for the motion of the Local Standard of Rest (LSR, hereafter) and the peculiar motion of the Sun with respect to the LSR.  The small number and, in most cases, the large uncertainties of the existing measurements emphasizes the difficulties inherent in measuring the proper motion of a dSph.  Accurate proper motions of dSphs would explain or help to answer a number of outstanding questions about the dSphs themselves, the Milky Way, interactions between dSphs and the Milky Way, and the formation of galaxies.  Probably the most fundamental question is whether a dSph is gravitationally bound to the Milky Way. Satellite status is almost universally assumed for the known dSphs --- confirming this assumption requires, among other things, a knowledge of the proper motion of each dSph.  Using the Hubble Space Telescope (HST, hereafter) for astrometry has advantages over using a ground-based telescope.  HST has superior angular resolution which, in principle, allows a more precise measurement of the location of an object.  This in turn reduces the required time baseline between images of a dSph to just a few years. As an example, consider an ``average'' dSph at a heliocentric distance of 100~kpc with an assumed tangential velocity with respect to the Sun of 220~km~s$^{-1}$.  Then the expected heliocentric proper motion of the dSph is $46$~mas~cent$^{-1}$.  Assuming a plate scale of 51~mas~pixel$^{-1}$ for the CCD imager in HST cameras, the expected shift in the position of the dSph over a one-year period is about $0.009$~pixel.  Several ground-based astrometric programs (Monet \\etal\\ 1992; Tinney \\etal\\ 1995; Tinney 1996) have shown that shifts of this magnitude can be measured with CCD detectors.  CCD detectors used in differential, small-angle astrometric measurements have significantly smaller systematic errors compared to measurements with a long time baseline using photographic plates, which are almost always acquired on different telescopes and which have increased uncertainties because of digitization on scanning machines.  A single HST image of a stellar field is undersampled because the full width at half of maximum (FWHM, hereafter) of a stellar point-spread function (PSF, hereafter) is only about 1.0 pixel for the HST imaging detectors.  Undersampling causes an image to have a ``digitized'' appearance and it makes measuring the centroid of a star to an accuracy of few thousandths of a pixel impossible with just a single image of the star.  Lauer (1999) and Anderson \\& King (2000) discuss these and other problems associated with undersampling and offer remedies for these shortcomings.  They show that accurate centroids of undersampled stellar images can be measured from multiple images of a science field which are dithered in a pattern which, for example, is an $N \\times N$ grid of $1/N$ subpixel steps (Lauer 1999), where $N^2$ is the number of images.  Such dithered images allow the construction of a well-sampled effective point-spread function (ePSF, hereafter), which is the convolution of the PSF of the telescope and the function representing the spatial response of a pixel in the CCD detector.  The value of the ePSF at a specific distance and direction from its center is the response (formally the fluence: photon counts per readout, though this article uses the more colloquial term flux) measured by a pixel at that distance from the centroid of the star. The centroid of an object and its brightness can be determined by fitting the ePSF to the measured flux in an array of pixels corresponding to the object.  Even though the individual functions in the convolution defining the ePSF can be difficult, or even impossible, to ascertain, the ePSF can be constructed without knowing an explicit form for these functions.  Anderson \\& King (2000) describe in detail a method for constructing the ePSF and a procedure for fitting the ePSF to the data in order to determine the centroid and total flux of an object.  Bernstein (2002) provides additional discussion of these topics.  Our method of constructing the ePSF, discussed in this article, is very similar to that described in Anderson \\& King (2000).  The expected small size of the proper motion of a dSph requires that the reference frame be defined by extragalactic objects.  The images of quasi-stellar objects (QSOs) are more compact at a given brightness than are the cores of ordinary galaxies, making a QSO a better reference frame than such a core.  For this reason (and because bright QSOs are useful probes for gas in dSphs), Cudworth, Olszewski, \\& Schommer (1986), Tinney, Da Costa, \\& Zinnecker (1997), and Tinney (1999) (and references therein) undertook searches for QSOs behind the dSphs.  Of the resulting QSOs, those that provide the required inertial reference frame with the minimum exposure time have magnitudes comparable to those of the brightest stars of the dSph that are in a typical HST field.  We are carrying out a program to measure proper motions for four dSphs --- Carina, Fornax, Sculptor, and Ursa Minor --- from HST images taken at three epochs, each separated by 1 -- 2 years.  This article serves the following purposes.  1) It describes our method for deriving an ePSF and for fitting it to the data to find the centroids of objects.  2) It examines how accurately proper motions of the dSphs can be measured using our methods and data.  3) Finally, the article presents a preliminary proper motion of Fornax derived from two- or three-epoch data for three distinct fields and discusses its implications. ",
        "conclusions": "This article describes our method for measuring proper motions of dSph galaxies from dithered images taken with HST at multiple epochs, each of which contains at least one previously-known QSO.  The steps in this method are: construct an ePSF from the bright objects in the images, fit the ePSF to all of the objects to determine their accurate centroids, and transform the centroids to a common coordinate system.  The measured proper motion of a dSph is the average shift of the positions of its stars with respect to the QSO.  The construction of a well-sampled ePSF is central to our method.  Ideally, the ePSF would have the freedom to vary from image to image and to vary within an image.  However, because there are relatively few stars with sufficiently high $S/N$ in our fields, constructing a well-sampled ePSF requires using a single, constant ePSF for a field at each epoch.  This limitation introduces an error that depends on the location of the centroid within a pixel and limits the accuracy of a measured centroid to 0.013~pixel for a single image. Averaging the centroids from images at eight independent dither positions yields a proper motion accurate to about 35~mas~cent$^{-1}$ for a pair of epochs separated by 1~year.  This accuracy is sufficient to measure the proper motion of the Fornax dSph galaxy.  We present four independent measurements of the proper motion of Fornax based on data from three different fields taken with two different detectors --- STIS and PC.  These measurements agree within their uncertainties, lending credibility to the method and the measured value.  The average measured proper motion of Fornax is $\\mu_{\\alpha} = 49 \\pm 13$~mas~cent$^{-1}$ and $\\mu_{\\delta} = -59 \\pm 13$~mas~cent$^{-1}$. The average proper motion vector corrected for the motion of the Sun and LSR (the proper motion in the Galactic rest frame) has a magnitude of $48 \\pm 13$~mas~cent$^{-1}$ and a position angle of $145 \\pm 15$~degrees.  This position angle places the rest-frame proper motion vector approximately along the minor axis of Fornax.  Fornax has Galactocentric radial and tangential velocities of $-40 \\pm 50$~km~s$^{-1}$ and $310 \\pm 80$~km~s$^{-1}$, respectively.  Assuming that the Galaxy has a flat rotation curve with a circular velocity of 220~km~s$^{-1}$, Fornax is near perigalacticon and, unless its space velocity is about 1$\\sigma$ or more below the measured value, it reaches such large Galactocentric radii that it is bound to the Local Group rather than to the Galaxy.  If Fornax is bound to the Galaxy, the implied lower limit on the mass of the Milky Way is $(1.6 \\pm 0.8) \\times 10^{12}~\\mathcal{M}_{\\odot}$.  Our measured proper motion for Fornax precludes its membership in the proposed Fornax-Leo-Sculptor stream or in any of the streams proposed in Lynden-Bell \\& Lynden-Bell (1995)."
    },
    "0209/astro-ph0209606_arXiv.txt": {
        "abstract": "Observations on galactic scales seem to be in contradiction with recent high resolution \\nbody\\ simulations. This so-called cold dark matter (CDM) crisis has been addressed in several ways, ranging from a change in fundamental physics by introducing self-interacting cold dark matter particles to a tuning of complex astrophysical processes such as global and/or local feedback.  All these efforts attempt to soften density profiles and reduce the abundance of satellites in simulated galaxy halos.  In this contribution we are exploring the differences between a Warm Dark Matter model and a CDM model where the power on a certain scale is reduced by introducing a narrow negative feature (''dip''). This dip is placed in a way so as to mimic the loss of power in the WDM model: both models have the same integrated power out to the scale where the power of the Dip model rises to the level of the unperturbed CDM spectrum again. Using \\nbody\\ simulations we show that that the new Dip model appears to be a viable alternative to WDM while being based on different physics: where WDM requires the introduction of a new particle species the Dip stems from a non-standard inflationary period. If we are looking for an alternative to the currently challenged standard \\LCDM\\ structure formation scenario, neither the \\LWDM\\ nor the new Dip model can be ruled out with respect to the analysis presented in this contribution. They both make very similar predictions and the degeneracy between them can only be broken with observations yet to come. ",
        "introduction": "The so-called Cold Dark Matter crisis has led to a vast number of publications trying to solve the problems which seem to be associated with an excess of power on small scales. One possibility to reduce this power is to introduce Warm Dark Matter (i.e. Knebe~\\ea 2002; Bode, Ostriker~\\& Turok 2001; Avila-Reese~\\ea 2001; Colin~\\ea 2000). But another way to decrease power on a certain scale is to introduce a negative feature (``dip'') into an otherwise unperturbed CDM power spectrum (cf. Knebe~\\ea 2001). Several mechanisms have been proposed that could generate such features in the primordial spectrum during the epoch of inflation. Among these are models with broken scale invariance (BSI) (Lesgourgues, Polarski~\\& Starobinsky 1998), and particularly BSI due to phase transitions during inflation (Barriga~\\ea 2000).  The \\LWDM\\ and the fiducial \\LCDM\\ model used in this paper are the same as those presented in Knebe~\\ea (2002) with the cosmological parameters $\\Omega_0 = 1/3$, $\\lambda_0 = 2/3$, $\\sigma_8=0.88$, $h=2/3$, and $m_{\\rm WDM} = 0.5$keV for \\LWDM. For the Dip model we are using the same prescription to introduce a Gaussian feature into an otherwise unperturbed CDM power spectrum as outlined in Knebe, Islam~\\& Silk (2001) with the parameters $A$=-0.995, $\\sigma_{\\rm mod}$=0.5, and $2\\pi/k_0$=1.8\\hMpc.   \\begin{figure} \\centerline{\\includegraphics[width=8.7cm]{aknebe.fig1.eps}} \\caption{Evolution of the power spectrum as measured on a regular 512$^3$ grid.} \\label{power} \\end{figure} ",
        "conclusions": "We conclude from the given analysis that both our non-standard models work equally well even though they are based on completely different physical processes; where WDM requires the introduction of a new particle species the Dip model is based on a non-standard inflationary period.  The only way of breaking their degeneracy might lie within the filamentary structures of the Universe and the low-mass end of the mass function. But to strengthen these findings we need more detailed and much better resolved simulations in the future."
    },
    "0209/astro-ph0209160_arXiv.txt": {
        "abstract": "Pulsar radio emission is believed to be originated from the electron-positron pairs streaming out from the polar cap region. Pair formation, an essential condition for pulsar radio emission, is believed to be sustained in active pulsars via one photon process from either the curvature radiation (CR) or the inverse Compton scattering (ICS) seed photons, or sometimes via two photon process. In pulsars with super-critical magnetic fields, some more exotic processes, such as magnetic photon splitting and bound pair formation, will also play noticeable roles. All these effects should be synthesized to discuss radio pulsar death both in the conventional long-period regime due to the turnoff of the active gap, and in the high magnetic field regime due to the possible suppression of the free pair formation. Here I briefly review some recent progress in understanding radio pulsar death. ",
        "introduction": "A pulsar drags its magnetosphere to co-rotate. In the observer's rest frame, the local charge density required for co-rotation is the ``Goldreich-Julian'' (1969) density, which when $r \\ll r_{lc}$ ($r_{lc}$ is the light cylinder radius) is approximately $\\rho_{_{\\rm GJ}} \\simeq -({\\bf \\Omega \\cdot B})/({2 \\pi c})$, where ${\\bf \\Omega}$ is the angular velocity of the star, and ${\\bf B}$ is the local magnetic field vector. In a simplest GJ magnetosphere, the positive and negative charges are separated from each other in space, and pair production is not required.  Pair production comes in for two reasons. First, it is obligatory. The most prominent feature of pulsar radio emission is its very high brightness temperature (typical value $T_B \\sim 10^{25}-10^{30}$ K). Due to self-absorption, the maximum brightness temperature for any incoherent emission is limited by the kinetic energy of the emitting electrons, i.e. $T_{incoh,max} = \\gamma_e m_e c^2/k \\sim 6\\times 10^{12} \\gamma_{e,3}$ K. The pulsar radio emission mechanism therefore must be {\\em extremely} coherent. Guided by our understanding of other coherent sources in the universe, the emission source is very likely a plasma in which various plasma instabilities can be developed to achieve coherent emission. A pair plasma is therefore required in an otherwise charge-separated magnetosphere.  Second, it is inevitable. A rotating magnetic pulsar is a unipolar generator, with a potential drop across the open field line region \\begin{equation} \\Phi=(B_p R^3 \\Omega^2/2c^2) = 6.6\\times 10^{12}~{\\rm V}~B_{p,12} P^{-2} R_6^3, \\end{equation} where $B_p$ is the surface magnetic field at the pole, $P$ is the rotation period, $R$ is the neutron star radius, and the convention $Q=10^n Q_n$ has been adopted. Such a huge potential is likely dropped along the open field lines (see \\S2 for the reasons), which accelerates a test particle up to an energy of $\\gamma_p m_e c^2 = e \\Phi \\sim 6.6$ TeV, or $\\gamma_p \\sim 10^7$ ($\\gamma_p$ is the Lorentz factor of the ``primary'' particles to be distinguished with the secondary pairs). These particles emit curvature radiation (CR) or inverse Compton scattering (ICS) photons (eqs.[2-4]) during acceleration. These primary $\\gamma$-rays inevitably materialize in a strong magnetic field or a hot thermal photon bath near the surface (see \\S2 for detailed discussions). ",
        "conclusions": "The following statements may be pertinent:  1. We now have a clear framework about the particle acceleration and photon-pair cascade in the pulsar polar cap region. Pair production from the polar cap is believed to be an essential condition for pulsar radio emission. The sufficient condition for radio emission, however, is unknown. Personally, I think that radio emission condition should be more stringent than the pair production condition. Thus, (moderate) non-dipolar fields may indeed exist at least in some pulsars. An important advance in the pulsar study in the recent years is the realization that ICS plays a crucial role in stead of CR in determining gap properties at least in some pulsars.  2. Theories are generally successful to define the conventional radio pulsar death in the long period regime, although many uncertainties prevent us from achieving a solid deathline. A death valley is more pertinent. Without introducing distortions from a star-centered-dipolar configuration, one can not include all pulsars above the deathline. The 8.5 second pulsar (Young et al. 1999) remains a challenge for any pure-dipole model after detailed numerical simulations. A best guess is that an ICS-controlled gap anchors in this pulsar with a moderate non-dipolar near-surface field configuration.  3. In the high magnetic field regime, pulsar death is not unambiguously defined. There is no strong reason against the possible radio emission from high magnetic field pulsars and magnetars. Possible reasons of apparent radio quiescence of magnetars may be due either to (a) that the coherent condition is destroyed; or to (b) that the main energy band of the coherent emission is not in radio; or to (c) the beaming effect; or else to (d) that the soft gamma-ray repeaters and the anomalous X-ray pulsars are not magnetars at all (but might be accretion-powered systems)."
    },
    "0209/astro-ph0209483_arXiv.txt": {
        "abstract": "We present axisymmetric, orbit superposition models for 12 galaxies using data taken with the {\\it Hubble Space Telescope (HST)} and ground-based observatories. In each galaxy, we detect a central black hole (BH) and measure its mass to accuracies ranging from 10\\% to 70\\%. We demonstrate that in most cases the BH detection requires {\\it both} the {\\it HST} and ground-based data.  Using the ground-based data alone does provide an unbiased measure of the BH mass (provided they are fit with fully general models), but at a greatly reduced significance. The most significant correlation with host galaxy properties is the relation between the BH mass and the velocity dispersion of the host galaxy; we find no other equally strong correlation, and no second parameter that improves the quality of the mass-dispersion relation. We are also able to measure the stellar orbital properties from these general models. The most massive galaxies are strongly biased to tangential orbits near the BH, consistent with binary BH models, while lower-mass galaxies have a range of anisotropies, consistent with an adiabatic growth of the BH. ",
        "introduction": "Most nearby galaxies contain massive compact dark objects at their centers. The number density and masses of these objects are consistent with the hypothesis that they are dead quasars: massive black holes that grew mainly by gas accretion and were once visible as quasars or other active galactic nuclei from radiation emitted during the accretion process (see Kormendy~\\& Richstone 1995 for a review).  We have obtained {\\it Hubble Space Telescope (HST)} spectra of the centers of 12 nearby galaxies, using first the square aperture of the Faint Object Spectrograph (FOS) and later the long-slit on the Space Telescope Imaging Spectrograph (STIS). Additional ground-based spectra have been obtained at the MDM Observatory. Pinkney~\\etal\\ (2002a) describe the data collected by our group for the 10 galaxies observed with STIS, and we present the data for the two galaxies observed with FOS in the Appendix of this paper. Section~2 discusses how we incorporate the data into the dynamical models.  An overall discussion of the dynamical modeling methods is given in Gebhardt~\\etal\\ (2000a) and Richstone~\\etal\\ (2002). The models are axisymmetric and based on superposition of individual stellar orbits. Section~3 provides the details of the models for these galaxies. Five other galaxies have stellar-dynamical data and models of comparable quality. Three of these are from the Leiden group: M32 (van~der~Marel~\\etal\\ 1998; Verolme~\\etal\\ 2002), NGC~4342 (Cretton~\\& van den Bosch 1999), and IC~1459 (Cappellari~\\etal\\ 2002). The remaining two are NGC~3379 (Gebhardt~\\etal\\ 2000a) and NGC~1023 (Bower~\\etal\\ 2001). Results from these five additional galaxies are included in the analysis in Section~4.   \\begin{figure*}[b] \\centerline{\\psfig{file=gebhardt.fig1.ps,width=15cm,angle=0}} \\vskip 0pt \\figcaption[gebhardt.fig1.ps]{Luminosity density profiles for the sample galaxies. These are in the $V$ band and include both {\\it HST} and ground-based data. The radii are along the semi-major axis. \\label{fig1}} \\end{figure*}   We use orbit-based models rather than parameterized models of the distribution function because parameterization can lead to biased black hole (BH) mass estimates. Parameterized models can even imply the presence of a BH when none exists. Orbit-based models do not suffer from this bias. However, we do make various assumptions whose consequences must be examined (Section~5). In particular, we model galaxies as axisymmetric. Triaxial, and worse yet, asymmetric galaxies, may be poorly represented by axisymmetric models. However, these effects are likely to be random and therefore it is reasonable to expect that the assumption of axisymmetry will not cause an overall bias in the BH mass.  In addition to measuring the BH mass ($M_{BH}$) and stellar mass-to-light ratio (\\mtl, assumed to be independent of position), our models constrain the orbital structure in the galaxy. It appears from this study and those of Verolme~\\etal\\ (2002) and Cappellari~\\etal\\ (2002) that the distribution function in axisymmetric galaxies depends on all three integrals of motion, not just the energy and angular momentum.  Preliminary BH masses for these galaxies have been reported by Gebhardt~\\etal\\ (2000b); these masses are based on a coarser grid of models (explained in Section~4) and thus have larger uncertainties than those presented here. However, the best-fit values for the BH masses are nearly the same in the two studies.  Most distances in this paper have been measured with the surface-brightness fluctuation method (SBF, Tonry~\\etal\\ 2000); for those galaxies without an SBF distance we assume the distance in an unperturbed Hubble flow and $H_0 = 80$ km s$^{-1}$ Mpc$^{-1}$. ",
        "conclusions": "The twelve galaxies in this paper all have significant BH detections, with a typical statistical significance in the masses of around 30\\%. The average significance of detection is well above 99\\% and the least significant detection (NGC~2778) has about 90\\% confidence. Thus, for this sample, every galaxy has a BH. In fact, only one nearby galaxy with high-resolution spectral data lacks any significant BH detection: the pure disk galaxy M33 (Gebhardt~\\etal\\ 2001). The most obvious difference between M33 and the galaxies with significant BH detection is that the latter have a bulge component.  For a few of these galaxies, ground-based spectra alone yield reasonably precise BH masses. The masses based on ground-based data alone are generally remarkably close to the masses based on ground-based and {\\it HST} data; there is no evidence that masses based on ground-based data alone are systematically high. The most important aspect of using ground-based data is assure that the models are fit using full generality (i.e., without assumptions about the orbital structure).  The most significant correlation with BH mass is with the velocity dispersion. The present intrinsic scatter is around 0.23 dex in BH mass (Tremaine~\\etal\\ 2002). It will be extremely illuminating to include more galaxies at both extremes, the low mass and high mass ends. The next most significant correlation is with the radial-to-tangential velocity dispersion at $R_e$/4. We do not know whether this is simply a secondary correlation due to that with the velocity dispersion, or if it represents an evolutionary pattern due to the growth of the BH. Detailed theoretical and N-body models are required to understand this. The BH mass also significantly correlates with both galaxy bulge luminosity and bulge mass, but neither of these is as strong as with dispersion.  The uncertainties in the BH masses reported here are only statistical. We have not attempted to include uncertainties from the assumptions in our models or systematic errors in our analysis outlined in \\S4.9. We believe that the increase in the uncertainties is likely to be small, but additional tests are required in order to substantiate this. We can use the $M_{BH}/\\sigma$ correlation as an approximate constraint on the uncertainties. If there is an underlying physical mechanism that causes a {\\it perfect} correlation between $M_{BH}$ and $\\sigma$, then any scatter seen in the correlation must be measurement error. Since the current scatter is comparable to the measurement error, we probably have a reasonable estimate of our uncertainties; any additional uncertainties caused by our assumptions should be smaller than 0.23 dex in BH mass. However, this argument applies only to random errors. If, for example, galaxies deviate from our assumptions systematically, then the $M_{BH}/\\sigma$ correlation may still have small scatter but incorrect BH masses. The only way to test this is to include a larger sample with general dynamical models that cover a wide variety of input configurations.  The orbit-based models provide a look into the internal orbital structure of an axisymmetric system. Based on the small sample of galaxies shown here and the limited theoretical comparisons, we are already able to place some constraints on the possible evolutionary history of the galaxy. The results in this paper suggest that core galaxies have tangentially biased orbits near their centers, while power-law galaxies show a range of tangential relative to radial motion. As suggested by Faber~\\etal\\ (1997) and Ravindranath~\\etal\\ (2002), it appears that the core galaxies are consistent with the BH/binary models, and the power-law galaxies are more consistent with adiabatic growth. This conclusion comes from analysis of the stellar surface brightness profiles, and now a similar conclusion comes from the stellar kinematics. Significant improvement in our understanding of the orbital structure will come from datasets with two-dimensional kinematics. De~Zeeuw~\\etal\\ (2002) and Bacon~\\etal\\ (2002) present examples of datasets that can be exploited for this analysis. However, in order to make progress in this area we must understand possible systematic biases that can arise from various assumptions (e.g., dark halo, different deprojections, lack of axial symmetry, etc.)"
    },
    "0209/astro-ph0209356_arXiv.txt": {
        "abstract": "We study the influence of intracluster large scale magnetic fields on the thermal Sunyaev-Zel'dovich (SZ) effect. In a macroscopic approach we complete the hydrostatic equilibrium equation with the magnetic field pressure component. Comparing the resulting mass distribution with a standard one, we derive a new electron density profile. For a spherically symmetric cluster model, this new profile can be written as the product of a standard ($\\beta$-) profile and a radius dependent function, close to unity, which takes into account the magnetic field strength. For non-cooling flow clusters we find that the observed magnetic field values can reduce the SZ signal by $\\sim 10\\%$ with respect to the value estimated from X-ray observations and the $\\beta$-model. If a cluster harbours a cooling flow, magnetic fields tend to weaken the cooling flow influence on the SZ-effect. ",
        "introduction": "The SZ-effect is rapidly turning into an important astrophysical tool thanks to the progress of the observational techniques, which allow increasingly precise measurements. In view of these developments it is thus relevant to study further corrections to it, such as relativistic effects \\citep{re95}, the shape of the galaxy cluster and its finite extension or a polytropic temperature profile (see e.g. \\citet{pu00}), corrections induced by halo rotation \\citep{co01}, Brillouin scattering \\citep{sm02}, early galactic winds \\citep{ma01} and the presence of cooling flows \\citep{sc91,ma01}. These additional effects are of different relevance and often depend on the specific cluster values.\\\\  Whereas, e.g. cooling flows are not present in every cluster of galaxies, the need for relativistic corrections due to energetic non-thermal electron populations seems to be common in most clusters. \\citep{bl00,sh02}. These relativistic electrons produce a hard X-ray component in excess of the thermal spectrum by Compton scattering off the Cosmic Microwave Background (CMB) and by non-thermal bremsstrahlung. Their emission has been quite possibly detected in \\astrobj{Coma} \\citep{re99,ff99}, \\astrobj{A2199} \\citep{ka99}, \\astrobj{A2256} \\citep{mo00} and \\astrobj{A2319} \\citep{gr02} by RXTE and BeppoSAX satellites. The main evidence for the existence of relativistic electrons comes from radio synchrotron emission of extended  intracluster regions \\citep{gi99,gi00} and is, therefore, closely related to the presence of magnetic fields. The magnetic fields in the intracluster gas lead to acceleration processes and modify the classical Maxwell-Boltzmann distribution of the electrons, which might acquire a significantly non-thermal spectrum and thus account for the observed hard X-ray spectra \\citep{en99,bl2000}. Consequently, several authors \\citep{re95,it98,ch98,bi99} derived relativistic corrections to the thermal SZ-effect up to different leading orders. Though the magnetic field is related to the relativistic electron population, its own influence - as a non-thermal cluster component - on the SZ-effect has not yet been investigated. Whereas the importance of relativistic corrections to the SZ-effect depends on the cluster temperature, the magnetic field seems to be ubiquitous with a mean field value of $5-10\\, \\mu G$ in the cluster cores \\citep{cl01}. Independently of the diffuse non-thermal radio emission, excess Faraday rotation measure of polarized radio emission in radio sources within or behind the cluster can prove the existence of magnetic fields. This method was applied to the \\astrobj{Coma} cluster, where \\citet{fe95} found a large magnetic field of $B\\ge 8.3 \\,\\mu G$. \\citet{go99} found a similar value ($5-10\\, \\mu G$) for \\astrobj{A119}. \\citet{cl99} derived magnetic field strengths of a few $\\mu G$ for their cluster sample by using a statistical Faraday rotation measure technique. Unfortunately, the structure of the magnetic field is presently poorly known. Contrary to the static situation in non-cooling flow clusters, the magnetic field is believed to become dynamically significant in the cores of cooling flow clusters \\citep{ei01}. \\citet{ta01} found a magnetic field strength of up to $40\\,\\mu G$ in the \\astrobj{Centaurus} cluster. The converging cooling flow (see e.g. \\citet{fa91,fa94}) causes a compression and enhancement in the magnetic field strength and finally reconnection might transfer the magnetic energy back to the plasma when the magnetic field pressure becomes comparable to the thermal gas pressure.\\\\ As the ratio of the magnetic pressure ($P_B$) to the gas pressure ($P_g$) reaches $\\frac{P_B}{P_g}\\sim 10^{-2}$ for the quoted values of non-cooling flow clusters and even unity for the cores of cooling flow clusters, we consider the additional magnetic field pressure term to be significant and we will examine its influence on the SZ-effect. We remark that all the quoted values refer to large scale magnetic fields with a coherence length of typically $1-10\\,kpc$. There are essentially no useful limits on the strength of any small scale magnetic fields. Our calculation is based on a macroscopic picture inferred from the existence of this additional (large scale) pressure component and we do not start our considerations at the level of the single particle movement in a magnetic field. The current cluster data reveal magnetic field values which result to be significant  for a correct SZ analysis, especially towards the cluster core where the electron density increases.\\\\  The aim of the paper is to examine the influence of large scale magnetic fields on the thermal SZ-effect. We, therefore, choose a phenomenological approach, where the addition of a magnetic field pressure to the gas pressure is well justified. For a given magnetic field model, we can then estimate the change in the electron density and the temperature profiles as compared to standard ones used in the literature in the absence of magnetic fields.   The paper is organised as follows: In section 2 we present the theoretical model for the magnetic field contribution. We derive new gas density profiles which are then used to calculate the SZ-effect. We distinguish two situations according to whether a cluster harbours a cooling flow or not. Section 3 shows our results and contains a discussion of how magnetic fields influence the SZ signal and what are the observational consequences. As an illustration we apply our results to the non-cooling flow cluster \\astrobj{A119}. Our conclusions are given in section 4. ",
        "conclusions": "In a phenomenological approach we added the magnetic field pressure to the gas pressure and for clusters without cooling flows we derived in a perturbative procedure a new gas density profile. This can be related to a standard $\\beta$-profile and a function, which takes into account the radius dependent magnetic field strength, which is assumed to be correlated to the electron density. For reasonable cluster parameters we find that magnetic fields reduce the standard SZ signal by $\\sim 10\\%$. Indeed, a reduction of up to $\\sim 15\\%$ seems possible. Our perturbative approach based on the equal mass assumption, $\\mathcal{M}_B(r)=\\mathcal{M}(r)$, turns out to be well justified by this order of magnitude correction: The corresponding decrease in the gas density is $\\le 10-15\\%$ and, therefore, the change in the dark matter dominated total cluster mass profile is less than $1\\%$. The simplifying assumption of an isothermal temperature is adequate, because the interplay between temperature and a variable magnetic field strength is not yet very clear \\citep{do01}.\\\\ Furthermore, our considerations showed that the central cluster gas density is probably overestimated by $10-20\\%$ when fitted with a standard $\\beta$-model. Better data in future might reveal the need for a modified $\\beta$-profile as discussed here.\\\\ Other possible causes of deviations from the (standard) $\\beta$-model have been discussed in the literature in the context of the determination of the Hubble constant (\\citet{bi91,in95} and references therein). Generally, the finite extension of a cluster (already adopted in our calculation) lowers the SZ signal by $5-10\\%$ compared to the case of an infinite isothermal $\\beta$-model, and requires then a larger core radius $r_c$ in the $\\beta$-model. Other departures from the standard $\\beta$-model include asphericity and small-scale clumping. The most extreme variation of geometry of the original spherical model is obtained if the unique axis of the prolate ($Z>1$) or oblate ($Z<1$) gas distribution is oriented along the line of sight. The SZ signal scales then with the factor $Z$, which is the ratio of the length of the unique axis to the major or minor axis, respectively. Typically, $0.5<Z<2$, and asphericity introduces thus a variation of a factor $\\sim 2$ in the core radius $r_c$. Whereas this effect will be averaged out in a large enough cluster sample, all cluster atmospheres will be clumpy to some degree. This might result in an overestimation of the standard SZ signal at the percentage level, but current simulations are limited in resolution and clumpiness might be more important. Whereas all these effects modify the parameters of the standard $\\beta$-model, the inclusion of magnetic fields requires an extended $\\beta$-profile, which takes explicitly into account the magnetic field strength. Unfortunately, in an integrated SZ measurement all the effects will sum up and they can hardly be separated. Precise future X-ray data might then help to reveal the inner cluster structure.  For the cooling flow clusters, following the treatment by \\citet{ss90}, we generalized the equations for the sound speed and infall velocity derived by \\citet{mb78} such as to include also a magnetic field. For typical initial magnetic field values at $r_{cool}$, $P_B/P_g\\sim 10^{-4}$, we find an equipartition radius $r_B\\sim 10\\,kpc$ and we derive profiles which are in agreement with Soker and Sarazin. Our somewhat larger sonic radius $r_s\\sim 5\\, kpc$ - though not irrealistic as it was found by \\citet{su89}, who derived sonic radii of $\\sim 10-20\\,kpc$ for plausible cluster and cooling flow parameters - results probably from different initial conditions. As the integration for the equations of sound speed and infall velocity is very delicate or even impossible for any arbitrary combination of parameters, our results can not immediately be generalized. Current discussions about the magnetic field model in the cooling flow region limit further our conclusions. Nevertheless, it turns out that the gas density in the cooling flow gets somewhat smaller in the presence of magnetic fields as compared to the case without. This translates then into a weaker influence of the order of some percent of the magnetized cooling flow on the SZ-effect. In special cases magnetic fields might almost compensate the cooling flow correction to the SZ-effect.\\\\ Indeed, a precise calculation would require more sophisticated models which are adapted to the specific cluster parameters. For all these reasons, cooling flow clusters do not seem to be ideal targets for SZ observations.     \\noindent{\\bf Acknowledgments}  This work was supported by the Swiss National Science Foundation. Part of the work of D.P. has been conducted under the auspices of the {\\it D\\raisebox{0.5ex}{r} Tomalla Foundation}."
    },
    "0209/astro-ph0209026_arXiv.txt": {
        "abstract": "{ A generic toy model of a cooling neutron star (NS) is used to analyze cooling of NSs with nucleon and exotic compositions of the cores. The model contains the parameters which specify the levels of slow and fast neutrino emission as well as the lower and upper densities of the layer where  the slow emission transforms into the fast one. The prospects to constrain these parameters from the present and future observations of isolated middle-aged NSs are discussed. ",
        "introduction": "Neutron stars (NSs) are compact stellar objects which contain the matter of supranuclear density in their cores. The equation of state (EOS) of this matter cannot be calculated unambiguously (e.g., Lattimer \\& Prakash \\cite{lp01}). Instead, there are many theoretical models which give a wide scatter of EOSs, from soft to stiff ones, with standard nucleon/hyperon or exotic compositions of matter. The nucleon matter consists mainly of neutrons (n) with an admixture of protons and electrons (p and e) and possibly muons. Some model EOSs predict also the appearance of hyperons. The exotic matter may contain pion condensates, kaon condensates, or quarks (or a mixture of these components). The indicated models are still almost not constrained by the observations of NSs.  In this paper we discuss the ability to constrain the EOS in the NS cores by confronting the observations of thermal emission from isolated NSs with the theory of NS cooling. In the last three decades the theory has been compared with the observations by many authors (e.g., Page \\cite{page98a}) but the problem is complicated and the main features are still unknown (too many factors are involved, such as neutrino emission mechanisms, superfluidity of baryon components of matter; see, e.g., Yakovlev et al.\\ \\cite{ygkp02}). Thus, at this stage it is sufficient to be general rather than accurate and use a simple toy model of cooling NSs. It will allow us to formulate the current status of the problem (Sects.\\ 4 and 5) without complicated simulations. ",
        "conclusions": "Thus, our {\\it generic cooling analysis} (5-parameter physics input leading to the families of cooling curves of NSs with different $M$) is too flexible to fix the nature of the fast neutrino emission and its density threshold. Consequently, the data {\\it can be explained by a number of physical EOSs} of dense matter. Notice that the cooling curves are more sensitive to the composition of matter than to the stiffness of the EOS (although the composition and stiffness are actually interrelated).  For instance, Kaminker et al.\\ (\\cite{khy01,kyg02}) exploit the idea of nucleon matter in the NS core with the onset of the Durca process at high densities. They assume the presence of a strong proton superfluidity at not too high densities to suppress the Murca process and broaden the Durca threshold. The suppression of Murca allows them to reduce the slow neutrino emission level $Q_{\\rm s}$ (from about $10^{21}$ to about $3 \\times 10^{19}$) and explain thus the hottest observed sources (Fig.\\ \\ref{fig1}). The broadening of the Durca threshold ensures a representative class of medium-mass NSs to interpret cooler objects.  In the phenomenological approach of Kaminker et al.\\ (\\cite{khy01,kyg02}) the crucial $^1{\\rm S}_0$ proton superfluidity is modeled by a specific density dependence of the proton critical temperature $T_{\\rm cp}$. Consider, for instance, a particular realization of this model applied by Yakovlev et al.\\ (\\cite{ykhg02}) to analyze the thermal state of PSR J0205+6449. For this specific model (model 1p with EOS A, in notations of the authors), the maximum of $T_{\\rm cp}$ ($\\sim 7 \\times 10^9$~K)  is reached at $5.75 \\times 10^{14}$ g cm$^{-3}$. The critical temperature $T_{\\rm cp}$ decreases at higher densities, but it is still $\\approx 4.1 \\times 10^9$~K at the Durca threshold, $\\rho_{\\rm D}=7.85 \\times 10^{14}$ g cm$^{-3}$, and then drops to zero at about $1.2 \\, \\rho_{\\rm D}$. Can $T_{\\rm cp}$ be as high as $\\sim 4\\times 10^9$~K at $\\rho=\\rho_{\\rm D}$ where the proton fraction is about 11\\%? In a recent paper Tsuruta et al. (\\cite{tsurutaetal02}) argue that this would contradict to the existing microscopic models of nucleon superfluidity, because the proton number density $n_{\\rm p}$ is too large at $\\rho=\\rho_{\\rm D}$ to allow for the $^1{\\rm S}_0$ proton pairing. Taking the model of Yakovlev et al.\\ (\\cite{ykhg02}) mentioned above we get $n_{\\rm p}=0.047$ fm$^{-3}\\approx 0.3 \\,n_0 $ at $\\rho=\\rho_{\\rm D}$, i.e., noticeably smaller than $n_0=0.16$ fm$^{-3}$, the nucleon number density in saturated nuclear matter. For the same number density of neutrons in neutron matter, the critical temperature of the $^1{\\rm S}_0$ neutron superfluidity is $\\sim 1.3\\times 10^{10}~$K if a realistic bare NN interaction is used (Lombardo \\& Schulze \\cite{ls01}). The medium effects (polarization effects, self-energy corrections) can decrease it to $7 \\times 10^9~$K or lower, depending on a particular model. In the case of the proton component of the neutron-star matter the np interaction can decrease the proton effective mass, lowering further $T_{\\rm cp}$. However, this does not mean that the value $T_{\\rm cp}(\\rho_{\\rm D}) \\ga 7 \\times 10^{9}~$ K, followed by a rapid drop of $T_{\\rm cp}$ with increasing $\\rho$, is ruled out by contemporary microscopic theories. In contrast to the case of neutron matter at $n_{\\rm n} < n_0$, reliable calculation of $T_{\\rm cp}(\\rho_{\\rm D})$ in the neutron-star matter, starting  from a realistic NN interaction and including the medium effects, remains still a challenge for the many-body theory. Therefore, we think that the use of a ``minimal model'' (npe matter with Durca acting at $\\rho>\\rho_{\\rm D}$, combined with the appropriate nucleon superfluidity) remains a valid first-step approach.  Notice that our general assumption on the neutrino emissivity $Q_\\nu(T,\\rho)$ would be violated in the presence of ($^3$P$_2$) neutron superfluidity in the NS core with a density dependent critical temperature $T_{\\rm cn}(\\rho)$ which has the maximum in the range from $\\sim 10^8$ to $\\sim 2 \\times 10^9$ K (see, e.g., Kaminker et al.\\ \\cite{kyg02}). The neutrino emission due to Cooper pairing of neutrons will then be so strong that it will initiate a really fast cooling even at $M < M_{\\rm s}$, violating the interpretation of relatively hot and old sources, first of all PSR 1055--52. As an example, by the dotted line in Fig.\\ \\ref{fig1} we show the cooling of 1.4 M$_\\odot$ NS with the nucleon core and forbidden Durca process in the presence of a neutron superfluidity (model of Takatsuka \\cite{takatsuka72}, with maximum $T_{\\rm c}$ of about $9 \\times 10^8$ K at $\\rho \\approx 5 \\times 10^{14}$ g cm$^{-3}$). One can see that NSs with this superfluidity would be too cold to explain the data.  Another physical model of cooling NSs was presented by Tsuruta et al.\\ (\\cite{tsurutaetal02}). It is based on pion condensation at supranuclear densities exploiting similar idea: slow cooling of low-mass NSs and faster cooling of massive NSs. Their cooling curve of low-mass NS (1.2 M$_\\odot$, nucleon core, forbidden Durca process) agrees with our basic slow-cooling curve (after translating our curve, $T_{\\rm s}^\\infty(t)$, to their format, $L_\\gamma^\\infty(t)$). Note that Tsuruta et al.\\ (\\cite{tsurutaetal02}) misplaced the positions of some observational data in their figure. Most important is RX J0822--43. They (as well as we) take the data from Zavlin et al.\\ (\\cite{ztp99}) who give $L_\\gamma^\\infty=(7.1-10.1)\\times 10^{33}$ erg s$^{-1}$ (${\\rm lg}\\, L_\\gamma^\\infty=33.85-34.00$) while Tsuruta et al.\\ present ${\\rm lg}\\,L_\\gamma^\\infty \\approx 33.57-33.97$. Thus, RX J0822--43 is sufficiently warm and cannot be explained with the basic slow-cooling model (Fig.\\ \\ref{fig1}). Moreover, according to Tsuruta et al., they employ the model neutron superfluidity of Takatsuka (\\cite{takatsuka72}). Therefore, their curve (if properly calculated) should resemble our dotted curve, and their model would then disagree with the number of observational limits. To avoid this disagreement one can change the model of nucleon superfluidity. A natural model of rather strong proton superfluidity and weak neutron superfluidity at $\\rho \\sim (3-8)\\times 10^{14}$ g cm$^{-3}$ considered by Kaminker et al.\\ (\\cite{khy01,kyg02}) seems to be most suitable."
    },
    "0209/astro-ph0209210_arXiv.txt": {
        "abstract": "{Observations of clusters of galaxies that gravitationally lens faint background galaxies can probe the amount and the equation of state, $\\5$, of the dark energy (quintessence) in the universe. Provided that the mass profile and the mass normalization of the cluster are determined, it is possible to constrain the cosmological parameters that enter the lensing equations by means of the angular diameter distances, by locating (either by observations of giant arcs and magnification bias effect) the critical lines corresponding to known redshift source populations of galaxies. This method can help to distinguish between accelerating and decelerating models of the universe. Furthermore, since the position of critical lines is affected, especially in low-matter density universes, by the properties of quintessence, the observations of a suitable number of lensing clusters at intermediate redshifts can determine the equation of state. A very preliminary application of the method to the cluster CL~0024+1654 seems to support a flat accelerating universe dominated by dark energy. ",
        "introduction": "Observational cosmology has devoted large efforts in the last years to characterize the energy content of the universe. Galaxy clustering (Bachall \\& Fan \\cite{ba+fa98}; Carlberg et al. \\cite{car&al98}) and large-scale structure (Peacock et al. \\cite{pe&al01}; Verde et al. \\cite{ver+al01}) observations favour models of a universe with a subcritical matter energy density $\\Omega_M$ (Turner \\cite{tur00}). Since, according to balloon-based measurements of the anisotropy of the Cosmic Microwave Background Radiation (de Bernardis et al. \\cite{deb&al00}; Balbi et al. \\cite{bal+al00}), the total of energy content of the universe nearly equals the critical density (Jaffe et al. \\cite{ja&al00}; Pryke et al. \\cite{pr+al02}), we expect that about $2/3$ of the critical density is in form of dark energy (also called quintessence). Furthermore, evidence coming from type Ia supernovae that the universe is accelerating its expansion (Riess et al. \\cite{ri&al98}; Perlmutter et al. \\cite{pe&al99}) demands a strongly negative pressure for the dark energy ($w_X \\equiv p_X /\\rho_X < -1/3$, where $p_X$ and $\\rho_X$ are, respectively, the pressure and energy density of the dark energy). These observations, together with other constraints coming from the age of the universe, gravitational lensing statistics and Ly$\\alpha$ forest, support a geometrically flat universe (Harun-or-Rashid \\& Roos \\cite{ha&ro01})($\\Omega_M + \\Omega_X =1$, where $\\Omega_X$ is the dark energy density parameter of the universe) with $\\Omega_M \\sim 0.3$-$0.4$ and a constant equation of state $-1 \\leq w_X \\stackrel{<}{\\sim} -0.4$ (Waga \\& Miceli \\cite{wa+mi98}; Wang et al. \\cite{wan+al00}) at the $68\\%$ confidence level or better according to a concordance analysis (Wang et al. \\cite{wan+al00}). A less conservative maximum likelihood analysis suggests a smaller range for the equation of state, $-1 \\leq w_X \\stackrel{<}{\\sim} -0.6$ (Perlmutter et al. \\cite{pe&al99b}; Wang et al. \\cite{wan+al00}; Bean \\& Melchiorri \\cite{be+me02}). According to these results, in what follows, without being explicitly stated, we will assume a flat universe.  After the first proposal of dark energy (the cosmological constant, $w_X=-1$), many other candidates have been suggested. One interesting idea is that the energy density is provided by a scalar field rolling down an almost flat potential (Caldwell et al. \\cite{ca&al98}; Ratra \\& Peebles \\cite{ra&pe98}; de Ritis et al. \\cite{rit&al00}; Rubano \\& Scudellaro \\cite{ru&sc01}). Other possibilities are represented by a fluid with a constant equation of state, called $X$-matter (Chiba et al. \\cite{ch&al97}; Turner \\& White \\cite{tu+wh97}), or by a network of light non-intercommuting topological defects (Vilenkin \\cite{vil84}; Spergel \\& Pen \\cite{sp+pe97}) ($w_X =-m/3$ where $m$ is the dimension of the defect: for a string, $m=1$; for a domain wall, $m=2$). Generally, the equation of state $w_X$ evolves with the redshift, and the feasibility of reconstructing its time evolution has been investigated (Cooray \\& Huterer \\cite{co&hu99}; Chiba \\& Nakamura \\cite{ch&na00}; Saini et al. \\cite{sa&al00}; Goliath et al. \\cite{gol&al01}; Huterer \\& Turner \\cite{hu&tu01}; Maor et al. \\cite{ma&al00}; Nakamura \\& Chiba \\cite{na&ch01}; Wang \\& Garnavich \\cite{wa+ga01}; Yamamoto \\& Futamase \\cite{ya+fu01}; Corasaniti \\& Copeland \\cite{co+co02}). Since in flat Friedmann-Lema\\^{\\i}tre-Robertson-Walker (FLRW) models the distance depends on $w_X$ only through a triple integral on the redshift (Maor et al. \\cite{ma&al00}), $w_X (z)$ can be determined only given a prior knowledge of the matter density of the universe (Goliath et al. \\cite{gol&al01}; Weller \\& Albrecht \\cite{we+al01a}; Gerke \\& Efstathiou \\cite{ge+ef02}). In what follows, we will consider only the case of a constant equation of state.  Although the listed results are really compelling, it is still useful to develop new tools for the determination of the cosmological parameters. Many of the discussed methods are affected by shortcomings, like poorly controlled systematic errors or large numbers of model parameters involved in the analysis. An independent constraint can improve the statistical significance of the statement about the geometry of the universe and can disentangle the degeneracy in the space of the cosmological parameters.  Gravitational lensing systems have been investigated as probes of dark energy. Gravitational lensing statistics (Waga \\& Miceli \\cite{wa+mi98}; Cooray \\& Huterer \\cite{co&hu99}; Wang et al. \\cite{wan+al00}; Zhu \\cite{zhu00}), effects of large-scale structure growth in weak lensing surveys (Benabed \\& Bernardeau \\cite{be+be01}) and Einstein rings in galaxy-quasar systems (Futamase \\& Yoshida \\cite{fu&yo00}; Yamamoto \\& Futamase \\cite{ya+fu01}) are very promising ways to test quintessence. Here, we propose to investigate clusters of galaxies acting as lenses on background high redshift galaxies. The feasibility of these systems to provide information on the universe is already known (Paczy\\'{n}ski \\& Gorski \\cite{pa&go81}; Breimer \\& Sanders \\cite{br+sa92}; Fort et al. \\cite{for+al97}; Link \\& Pierce \\cite{li&pi98}; Lombardi \\& Bertin \\cite{lo&be99}; Gautret et al. \\cite{ga&al00}). Provided that the modeling of the lens is constrained, once both arc positions and its redshift are measured, it is possible to gain an insight into second-order cosmological parameters contained in angular diameter distances ratios (Chiba \\& Takahashi \\cite{ch+ta01}; Golse et al. \\cite{go&al01}). In addition to observations of arcs, a statistical approach based on magnification bias (Broadhurst et al. \\cite{bro+al95}; Fort et al. \\cite{for+al97}; Mayen \\& Soucail \\cite{ma&so00}) can as well locate the critical lines (locations of maximum amplification) corresponding to background source populations.  In this paper, we will explore the feasibility of clusters of galaxies acting as lenses in probing both the amount and the equation of state of quintessence in the universe, assumed to be flat. In Sect.~2, we outline the method. Section~3 is devoted to an application to the cluster of galaxies CL 0024+1654. In Sect.~4, we discuss some systematics affecting the method. Some final considerations are presented in Sect.~5. ",
        "conclusions": ""
    },
    "0209/astro-ph0209204_arXiv.txt": {
        "abstract": "This paper is the first in a series presenting CCD multicolor photometry for 145 HII regions, selected from 369 candidate regions from Boulesteix et al. (1974), in the nearby spiral galaxy M33. The observations, which covered the whole area of M33, were carried out by the Beijing Astronomical Observatory $60/90$ cm Schmidt Telescope, in 13 intermediate-band filters, covering a range of wavelength from 3800 to 10000{\\AA}. This provides a series of maps which can be converted to a multicolor map of M33, in pixels of $1\\arcsec{\\mbox{}\\hspace{-0.15cm}.} 7 \\times 1\\arcsec{\\mbox{}\\hspace{-0.15cm}.} 7$. Using aperture photometry we obtain the spectral energy distributions (SEDs) for these HII regions. We also give their identification charts. Using the relationship between the BATC intermediate-band system used for the observations and the {\\it UBVRI} broad-band system, the magnitudes in the {\\it B} and {\\it V} bands are then derived. Histograms of the magnitudes in {\\it V} and in {\\it B$-$V} are plotted, and the color-magnitude diagram is also given. The distribution of magnitudes in the {\\it V} band shows that the apparent magnitude of almost all the regions is brighter than 18, corresponding to an absolute magnitude of $-$6.62 for an assumed distance modulus of 24.62, which corresponds to a single main sequence O5 star, while the distribution of color shows that the sample is blue, with a mode close to $-$0.05 as would be expected from a range of typical young clusters. ",
        "introduction": "HII regions provide an excellent means of studying the ongoing and accumulated star formation in a late type galaxy. In a well-resolved galaxy, HII regions can offer much useful information: about the current status of star formation, and the physical conditions in the interstellar medium near the hot young stars within the galaxy (see e.g., Wyder, Hodge, \\& Skelton 1997), and it is well established that observations of extragalactic HII regions are important for the understanding of the chemical evolution and star formation history of the parent galaxy. For these purposes HII regions in many nearby galaxies have been explored in considerable detail (Garnett \\& Shields 1987), which include M31 (Baade \\& Arp 1964; Pellet et al. 1978; Walterbos \\& Braun 1992; Walterbos 2000), and M81 (Garnett \\& Shields 1987; Petit, Sivan, \\& Karachentsev 1988; Hill et al. 1995; Miller \\& Hodge 1996).  M33, one of our two nearest spiral neighbors, is a small Scd Local Group galaxy, whose distance modulus, determined via its Cepheids by Freedman,Wilson and Madore (1991), was more recently revised to 24.62 by Freedman et al. (2001). Because it is so near, and has quite a low inclination ($\\sim 33^\\circ$), M33 is an excellent candidate for HII region studies. A good database for its star clusters has been built up from the ground-based work (Hiltner 1960; Kron \\& Mayall 1960; Melnick \\& D'Odorico 1978; Christian \\& Schommer 1982, 1988}), and from the {\\it {Hubble Space Telescope (HST)}} images (Chandar, Bianchi, \\& Ford 1999a, 1999b; \\cite{Chandar01}). Ma et al. (2001, 2002a, 2002b) obtained the spectral energy distributions (SEDs) of 144 star clusters, and estimated their ages by comparing the photometry of each object with theoretical stellar population synthesis models for different values of metallicity.  The HII regions in M33 have proved of interest to astronomers for well over half a century (see e.g., Aller 1942, Haro 1950, Court\\`{e}s \\& Cruvellier 1965).  Aller (1942) used the ratio of [OIII] $\\lambda5007$/[OII] $\\lambda3727$ to be the criterion for identifying HII regions. However, most studies searched HII regions by photographic $H\\alpha$ survey. In 1974, Boulesteix et al. (1974) compiled a major catalogue of 369 distinct HII regions by a complete photographic $H\\alpha$ survey of M33. In this study, Boulesteix et al. (1974) defined an HII region within the  $H\\alpha$ image of M33 to correspond to an emission measure of $\\rm {150~cm^{-6}pc}$, i.e. three times above the rms noise level. Later Court\\`{e}s et al. (1987) added an additional 410 objects to that catalogue, also using a photographic $H\\alpha$ survey, and defined an HII region within the  $H\\alpha$ image of M33 to correspond to an emission measure of $\\rm {40~cm^{-6}pc}$. In a paper in which they selected isolated candidates for SS433-like stellar systems, Calzetti et al. (1995) reported 432 compact regions emitting $H\\alpha$. Recently a more extensive catalogue with 2338 $H\\alpha$ emission regions of M33 was published by Hodge et al. (1999). While these authors made an explicit position and luminosity catalogue, others paid more attention to the luminosity function and size distribution of the HII regions on the basis of $H\\alpha$ surveys (e.g., Wyder et al. 1997, Cardwell et al. 2000). The latter authors claim to have identified over 10000 separate HII regions in $H\\alpha$, but so far have not published a catalogue.  Although there is already much observational information in the literature, for full physical information about the HII regions it is of great value to obtain multiband photometry, which can provide accurate SEDs. In the present study we present CCD spectrophotometry of a set of HII regions in M33 using images obtained with the Beijing-Arizona-Taiwan-Connecticut (BATC) Multicolor Sky Survey Telescope, designed to obtain SED information for galaxies (Fan et al. 1996). The BATC system uses the 60/90 cm Schmidt telescope with 15 intermediate bandwidth filters. Here we use 13 of these filters, from 3800 to 10000{\\AA}, with images covering the whole visible extent of M33. In this paper we present the SEDs of 145 HII regions selected from the sample of Boulesteix et al. (1974), and also supply identification charts for these regions. We go on to compute the {\\it B} and {\\it V} magnitudes for these regions, using previously derived relationships between the BATC intermediate band system and the {\\it UBVRI} broad-band system. Finally we plot histograms of the {\\it V} band magnitudes and also of {\\it B$-$V}, as well as plotting the corresponding color-magnitude diagram for our sample regions.  We note that the extinction for almost all of the HII regions remains undetermined. Although it has been derived for a few of the large regions (Viallefond, Donas, \\& Goss 1983; Viallefond \\& Goss 1986; Churchwell \\& Goss 1999), the extinction for the regions sampled here has not been previously presented. In M81 the extinction varies relatively little from HII region to HII region (Hill et al. 1995), but in M33 there is considerable variation (Viallefond, Donas \\& Goss 1983). We cannot use the assumption of uniform extinction to correct for reddening in the present paper, so we have not been able to infer the intrinsic fluxes of the regions. In a forthcoming article (Jiang et al. 2002) we will use the flux ratio of $H_{\\alpha}/H_{\\beta}$ to study the extinction region by region.  The outline of this paper is as follows: details of the observations and data reduction are given in section 2. In section 3 we give the SEDs and the identification charts for the 145 regions studied. In this section also we produce the histograms of the {\\it V} magnitude, and of {\\it B$-$V}, as well as the color-magnitude diagram. Finally in section 4 we give a summary. ",
        "conclusions": ""
    },
    "0209/astro-ph0209032_arXiv.txt": {
        "abstract": "{We present a deep, wide-field optical survey of the young stellar cluster Alpha Per, in which we have discovered a large population of candidate brown dwarfs. Subsequent infrared photometric follow-up shows that the majority of them are probable or possible members of the cluster, reaching to a minimum mass of 0.035 M$_\\odot$. We have used this list of members to derive the luminosity and mass functions of the substellar population of the cluster ($\\alpha$=0.59$\\pm$0.05, when expressed in the mass spectrum form $\\phi$$\\propto$$M^{-\\alpha}$) and compared its slope to the value measure for  the Pleiades. This comparison indicates that the two cluster mass functions are, indeed, very similar. ",
        "introduction": "In an ongoing effort to discover low-mass stars and brown dwarfs (BDs) belonging to  young open clusters, we have studied the association Alpha Per. This is well-known nearby cluster with (m-M)$_0$=6.23 (176 pc). The interstellar reddening is also low,  A$_V$=0.30 (Pinsonneault et al. 1998). The normally quoted age for the cluster, based on isochrone fitting of the upper main sequence (MS), is of order 50 Myr (cf. Meynet et al. 1993), though models with a larger amount of convective core overshoot can yield ages up to about 80 Myr (Ventura et al. 1998). Recently, using the data we published in a preliminary study of the BD population of the cluster (Stauffer et al. 1999), we estimated the age as 90 Myr, based on the location of the lithium depletion boundary. The theoretical background to this method can be found in  Kumar (1963) and D'Antona and Mazzitelli (1994), and when applied to the Pleiades (Basri et al. 1996, Rebolo et al. 1996; Stauffer et al. 1998a) and IC\\,2391 (Barrado y Navascu\\'es et al. 1999) has also yielded ages $\\sim$50\\% older than previously assumed. A review of these results may be found in Basri (2000).  During the last years, different clusters and star forming regions have been studied intensively  and their BD population revealed. The Pleiades, Alpha Per cluster, IC2391, M35, NGC2516, Taurus, the Trapezium cluster, Sigma Orionis cluster, Cha I dark cloud, Upper-Sco OB association, and IC348  have been investigated (Rebolo et al. 1995; Festin 1997; Bouvier et al. 1998; Stauffer et al. 1998ab, 1999; Brice\\~no et al. 1998; Barrado y Navascu\\'es et al. 1999, 2001ab, 2002; Neuh\\\"auser \\& Comeron 1999; Zapatero-Osorio et al. 1999, 2000; Lucas \\& Roche 2000; Luhman 1999, 2000; Mart\\'{\\i}n et al. 1999, 2000, 2001; B\\'ejar et al. 2001; Pinfield et al. 2000; Ardila et al. 2000; Najita et al. 2000; Moraux et al. 2001; {\\it et cetera}). All these works show that BDs are quite numerous and that the mass function (MF) usually presents an increase for very low-mass objects. In any case, the total mass below the substellar limit only contributes a few percent to the total mass of the parent cluster, at least in the case of the Pleiades  ($\\sim$3-5 \\%, Bouvier et al. 1998, Hodgkin \\& Jameson 2000). However, it is not clear whether this MF is universal and this result can be extrapolated to other young clusters.  In this paper, we present a new deep, wide-field optical survey of the Alpha Per cluster. We have followed-up the optical candidates in the near-infrared using new infrared imaging data and the 2MASS catalogue (Skrutskie et al. 1997). Using this wealth of data, we have been able to establish the presence of a substantial population of BDs in the cluster, and derived its substellar MF. ",
        "conclusions": "Optical and near infrared photometry have been used to select a list of probable and possible low-mass  members belonging to  the young cluster Alpha Per, unveiling a large population of brown dwarf candidates. This information has been used to derive the luminosity and mass function of the cluster in the substellar domain. The index of the mass function, $\\alpha = 0.59$, is very similar to that determined for the Pleiades  (120 Myr) and $\\sigma$ Orionis (2--8 Myr) clusters."
    },
    "0209/astro-ph0209497_arXiv.txt": {
        "abstract": "We consider the effect of the violation of the equivalence principle (VEP) by the massive neutrino component on the Cosmic Microwave Background angular power specrum. We show that in the presence of adiabatic and isocurvature primordial density perturbations the {\\sc Planck} surveyor can place limits on the maximal VEP by the massive neutrino component at the level of ${\\rm few}\\times 10^{-5}$, valid in the general relativity, for the case in which the gravity is the single source of VEP. This work has been performed within the framework of the {\\sc Planck}/LFI activities. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209174_arXiv.txt": {
        "abstract": "I use photometry and spectroscopy data for 24 Type II plateau supernovae to examine their observed and physical properties. This dataset shows that these objects encompass a wide range of $\\sim$5 mag in their plateau luminosities, their expansion velocities vary by $\\times$5, and the nickel masses produced in these explosions go from 0.0016 to 0.26 $M_\\odot$.  From a subset of 16 objects I find that the explosion energies vary between 0.6$\\times$ and 5.5$\\times$10$^{51}$ ergs, the ejected masses encompass the range 14-56 $M_\\odot$, and the progenitors' radii go from 80 to 600 $R_\\odot$. Despite this great diversity several regularities emerge, which reveal that there is a continuum in the properties of these objects from the faint, low-energy, nickel-poor SNe~1997D and 1999br, to the bright, high-energy, nickel-rich SN~1992am. This study provides evidence that more massive progenitors produce more energetic explosions, thus suggesting that the outcome of the core collapse is somewhat determined by the envelope mass. I find also that supernovae with greater energies produce more nickel. Similar relationships appear to hold for Type Ib/c supernovae, which suggests that both Type II and Type Ib/c supernovae share the same core physics. When the whole sample of core collapse objects is considered, there is a continous distribution of energies below 8$\\times$10$^{51}$ ergs. Far above in energy scale and nickel production lies the extreme hypernova 1998bw, the only supernova firmly associated to a GRB. ",
        "introduction": "The advent of new telescopes and better detectors is causing a rapid increase in the quality and quantity of observations obtained for supernovae (SNe, hereafter) of all types. Although the field of Type Ia SNe (exploding white dwarfs) has developed considerably faster in recent years (due to the widely acknowledged importance of such objects as cosmological probes), there is a growing body of data for core collapse SNe. In this paper I collect all of the available data on hydrogen-rich plateau Type II SNe (those undergoing little interaction with the circumstellar medium, SNe~II-P hereafter), with the purpose to better understand the nature of such objects.  I start in section \\ref{OM} by summarizing the observational material available on 24 SNe~II-P, after which (Sec. \\ref{OP}) I proceed to examine their great diversity and the correlations among the observed parameters. Using the hydrodynamic models of \\citet[hereafter LN83, LN85]{litvinova83,litvinova85} I go a step further and derive physical parameters (explosion energies, progenitor masses and radii) for 13 SNe~II-P (Sec. \\ref{PP}). Although the statistics are still poor, this study shows that progenitors with greater masses produce more energetic explosions and synthesize more nickel. These correlations provide valuable clues and a better insight on the explosion mechanisms. In section \\ref{PCCS} I combine the physical parameters of the SNe~II-P with those previously published for SNe~Ib/c. It appears that all core collapse SNe display the same correlations, which suggests that all of these objects share the same core physics. I discuss the properties of all core collapse SNe and how hypernovae fit in this group. ",
        "conclusions": "\\label{CC}  I assembled photometric and spectroscopic data for 24 SNe~II-P which allowed me to draw the following conclusions,  \\noindent 1) As previously known, I recovered the result that SNe~II-P encompass a wide range of $\\sim$5 mag in plateau luminosities and a five-fold range in expansion velocities. I recovered the luminosity-velocity relation previously reported by \\citet{hamuy02} which supports the claim that SNe~II-P have a potential utility as cosmological probes. This empirical relation is also supported by the theoretical models of LN83 and LN85.  \\noindent 2) SNe~II-P encompass a factor of 10 in nickel masses between 0.0016 (SN~1999br) and 0.26 $M_\\odot$ (SN~1992am). There is clear evidence for a correlation in the sense that SNe with brighter plateaus and greater expansion velocities produce more nickel.  \\noindent 3) There is a continuum in the properties of SNe~II-P from faint, low-velocity, nickel-poor events such as SN~1997D and SN~1999br, and bright, high-velocity, nickel-rich objects like SN~1992am. The correlations between plateau luminosities, expansion velocities, and nickel masses suggest that SNe~II-P constitute a one parameter family.  \\noindent 4) Using the theoretical models of LN83 and LN85 I derived physical parameters for a subset of 13 SNe. Including SN~1987A, SN~1997D, and SN~1999br from previous studies I found that the explosion energies vary between 0.6 (SN~1999br) and 5.5 foes (SN~1992am), the ejected masses encompass the range 14-56 $M_\\odot$, and the progenitors' radii go from 80 to 600 $R_\\odot$.  \\noindent 5) Despite the large error bars, a couple of correlations emerge from the previous analysis: (1) more massive progenitors produce more energetic explosions, which suggests that the outcome of the core collapse is somewhat determined by the envelope mass; (2) SNe with greater energies produce more nickel.  Similar relationships appear to hold for Type Ib/c SNe, which suggests that both Type II and Type Ib/c SNe share the same core physics."
    },
    "0209/astro-ph0209612_arXiv.txt": {
        "abstract": "I present high-frequency polarimetric radio observations of the radio galaxy 3C\\,171, in which depolarization is associated with an extended emission-line region. The radio hotspots, known to be depolarized at low radio frequencies, become significantly polarized above (observer's-frame) frequencies of 15 GHz. There is some evidence that some of the Faraday rotation structure associated with the emission-line regions is being resolved at the highest resolutions (0.1 arcsec, or 400 pc); however, the majority remains unresolved. Using the new radio data and a simple model for the nature of the depolarizing screen, it is possible to place some constraints on the nature of the medium responsible for the depolarization. I argue that it is most likely that the depolarization is due not to the emission-line material itself but to a second, less dense, hot phase of the shocked ISM, and derive some limits on its density and temperature. ",
        "introduction": "Extended emission-line regions (EELR) around radio galaxies are common, and often aligned with the axes of the extended radio emission (e.g.\\ McCarthy \\etal\\ 1987; McCarthy 1993; McCarthy, Baum \\& Spinrad 1996). There is considerable debate as to whether EELR are ionized by photons from the nucleus or by shocks driven by the jets (e.g.\\ Tadhunter \\etal\\ 1998). Most EELR around low-redshift sources can adequately be described by central-illumination models, while sources which show evidence for shock ionization (such as extreme emission-line kinematics) are largely at high redshift. However, there is evidence that a small minority of low-redshift radio galaxies are also more adequately described by a shock-ionization model. If the ionization is due to shocks, the physical conditions in and kinematics of the EELR may give us important information about the energetics of the radio source as a whole.  One good candidate low-redshift shock-ionization source is 3C\\,171, an unusual $z=0.2384$ radio galaxy. Its large-scale radio structure has been well studied (Heckman, van Breugel \\& Miley 1984; Blundell 1996). In its inner regions it is similar to a normal FRII, but low-surface-brightness plumes instead of normal lobes extend north and south from the hotspots. High-resolution images show knotty jets connecting the radio core with the hotspots (Hardcastle \\etal\\ 1997). 3C\\,171 is associated with a prominent EELR, elongated along the jet axis. Heckman \\etal\\ found that the emission-line gas was brightest near the hotspots, suggesting that the radio-emitting plasma is responsible for powering the emission-line regions. More recently Clark \\etal\\ (1998) have made WHT observations of the emission-line regions and shown that their ionization states are most consistent with a shock-ionization model. The unusually strong EELR in this source may be indicative of an environment unusually rich in dense, warm gas; the interactions between the jets and this gas might then be responsible for the peculiar large-scale radio structure.  Heckman \\etal\\ pointed out the depolarization of the radio source near the hotspots at frequencies up to 15 GHz. The strong spatial relationship between the EELR (or the material traced by the EELR) and depolarization was pointed out by Hardcastle \\etal\\ (1997), who superposed an 8-GHz Very Large Array (VLA) image on the {\\it Hubble Space Telescope} ({\\it HST}) snapshot (de Koff \\etal\\ 1996) in the F702W filter (which at this redshift passes several strong lines). The source has since been observed with a number of other {\\it HST} filters, and Fig.\\ 1 shows a superposition of the 8-GHz map on an {\\it HST} image in the [OIII]495.9/500.7 nm lines. Regions of low polarization in the radio map can be seen to correspond to the positions of emission-line material. The eastern hot spot is almost completely unpolarized at 8 GHz and below, while the western hot spot is partially depolarized.  The close assocation between the emission-line regions and the radio depolarization in 3C\\,171 opens up the possibility of using radio data to constrain unknown physical conditions, such as the electron number density and the magnetic field strength in the gas surrounding the source. In this paper, I present new radio observations of 3C\\,171, and discuss the constraints that they place on the nature of its environment.  Throughout the paper, I use a cosmology with $H_0 = 65$ km s$^{-1}$ Mpc$^{-1}$, $\\Omega_{\\rm m} = 0.3$ and $\\Omega_\\Lambda = 0.7$. At the redshift of 3C\\,171, 1 arcsec corresponds to 4.07 kpc.  \\begin{figure*} \\epsfxsize 12cm \\epsfbox{3C171.OIIIO.PS} \\epsfxsize 12cm \\epsfbox{3C171.OIIOP.PS} \\caption{8.1 GHz VLA image, made with VLA A-configuration data (see the text). Above is the total intensity image (contours at $300 \\times (1, 2, 4\\dots)$ $\\mu$Jy beam$^{-1}$; below is the polarized intensity image (contours at $150 \\times (1, 2, 4\\dots)$ $\\mu$Jy beam$^{-1}$. The restoring beam is an $0.29 \\times 0.20$-arcsec elliptical Gaussian in PA $-79$\\degr. The greyscale shows a 3000-s exposure in the FR680 ramp filter of the {\\it HST}'s WFPC2, kindly supplied by Clive Tadhunter, and is dominated by emission from the [OIII] 495.9/500.7 nm lines. Note the strong relationship between depolarization and emission-line regions.} \\end{figure*} ",
        "conclusions": ""
    },
    "0209/astro-ph0209048_arXiv.txt": {
        "abstract": "We present deep $UBVRI$ CCD photometry for the young open star clusters Tr 1 and Be 11. The CCD data for Be 11 is obtained for the first time. The sample consists of $\\sim$ 1500 stars reaching down to $V$ $\\sim$ 21 mag. Analysis of the radial distribution of stellar surface density indicates that radius values for Tr 1 and Be 11 are 2.3 and 1.5 pc respectively. The interstellar extinction across the face of the imaged clusters region seems to be non-uniform with a mean value of $E(B-V)$ = 0.60$\\pm$0.05 and 0.95$\\pm$0.05 mag for Tr 1 and Be 11 respectively. A random positional variation of $E(B-V)$ is present in both the clusters. In the cluster Be 11, the reason of random positional variation may be apparent association of the HII region (S 213). The 2MASS $JHK$ data in combination with the optical data in the cluster Be 11 yields $E(J-K)$ = 0.40$\\pm$0.20 mag and $E(V-K)$ = 2.20$\\pm$0.20 mag. Colour excess diagrams indicate a normal interstellar extinction law in the direction of cluster Be 11.\\\\ The distances of Tr 1 and Be 11 are estimated as 2.6$\\pm$0.10 and 2.2$\\pm$0.10 Kpc respectively, while the theoretical stellar evolutionary isochrones fitted to the bright cluster members indicate that the cluster Tr 1 and Be 11 are 40$\\pm$10 and 110$\\pm$10 Myr old. The mass functions corrected for both field star contamination and data incompleteness are derived for both the clusters. The slopes $1.50\\pm0.40$ and $1.22\\pm0.24$ for Tr 1 and Be 11 respectively are in agreement with the Salpeter's value. Observed mass segregations in both clusters may be due to the result of dynamical evolutions or imprint of star formation processes or both.\\\\ ",
        "introduction": "Young open star clusters in a galaxy provide valuable information about star formation processes and are key objects for the galactic structure and evolution. For such studies, a knowledge of cluster's parameters like distance, age, reddening and stellar content is required which can be derived from the colour-magnitude (CM) and colour-colour (CC) diagrams of star clusters. In addition to this, the distribution of stellar masses at the time of cluster formation is of fundamental importance to analysis related to evolution of galaxies. The initial mass function (IMF) also plays an important role in understanding the early dynamical evolution of star clusters, because it is a fossil record of the very complex process of star formation and provides an important link between the easily observable population of luminous stars in a stellar system and the fainter, but dynamically more important, low mass stars. Another related problem is the mass segregation in star clusters in which massive stars are more concentrated towards the cluster center compared to low mass stars. It is not clear whether the mass segregation observed in several open clusters is due to dynamical evolution or an imprint of star formation processes itself (cf. Sagar et al. 1988; Sagar 2001 and references therein). Thus one can say that the young open star clusters are the laboratories in a galaxy for providing answers to many current questions of astrophysics.\\\\  \\begin{table*} \\centering \\begin{minipage}{140mm} \\caption{General information about the clusters under study, taken from Mermilliod (1995)} \\begin{tabular}{|c|ccccccccc|} \\hline Cluster&IAU&OCL&l&b&Trumpler&Radius&Distance&$E(B-V)$&log(age)\\\\ &&&(deg)&(deg)&class&(arcmin)&(Kpc)&(mag)&(yrs)\\\\ \\hline Trumpler 1&C0132+610&328&128.22&-1.14&II 2p&1.5&2.6&0.58&7.5\\\\ Berkeley 11&C0417+448&404&157.08&-3.65&II 2m&2.5&2.2&0.95&7.7\\\\ \\hline \\end{tabular} \\end{minipage} \\end{table*}  In the light of above discussions, we performed multicolour deep CCD stellar photometry in two young open star clusters namely Trumpler 1 (Tr 1) and Berkeley 11 (Be 11). The CCD $UBVRI$ observations of Be 11 are presented for the first time. The relevant prior informations (taken from Mermilliod (1995)) of these clusters are given in Table 1. These clusters are relatively compact objects with angular radii less than $3^{\\prime}$. Previous studies, observations and data reductions are described in the next sections. The interstellar extinction, other photometric results, luminosity function, mass function and mass segregation are described in the subsequent sections. Making use of $JHK$ data with optical data, extinction law has also been studied in Be 11.  \\section[]{Previous studies} {\\bf Trumpler 1}: It is an extremely concentrated galactic open star cluster in Cassiopeia. It lies at the outer edge of the Perseus spiral arm. Oja (1966) carried out the proper motion study. $UBV$ photoelectric photometry for 43 bright stars was presented by Joshi \\& Sagar (1977) while Phelps \\& Janes (1994) published $UBV$ CCD photometry ($V$$\\sim$18 mag). McCuskey \\& Houk (1964) studied this cluster photographically in $UBV$ system while Steppe (1974) has presented three colour RGU photographic photometry. All these studies indicate that reddening across the cluster is uniform with $E(B-V)$ = 0.61 mag, distance estimate is 2630 pc and age seems to be $\\sim$ 27 Myr.\\\\  \\noindent {\\bf Berkeley 11}: It is a distant neglected compact young open cluster, apparently associated with the faint HII region S 213. Only $UBV$ photoelectric photometry of 24 bright stars has been done by Jackson et al. (1980). On the basis of this study, they found that this cluster has members earliest photometric type $\\sim$ b4 and is thus an extreme Population I object. They also found reddening $E(B-V)$ = 0.95$\\pm$0.06 mag, distance d = 2.2$\\pm$ 0.2 Kpc and age of the cluster as 3$\\times$10$^{7}$ yr.\\\\ ",
        "conclusions": ""
    },
    "0209/astro-ph0209081_arXiv.txt": {
        "abstract": "A new broad absorption line quasar (BAL) sample is derived from the first data released by the Sloan Digital Sky Survey.  With 116 objects, it is the largest BAL sample yet assembled.  Over the redshift range $1.8 \\leq z \\leq 3.8$, the crude fraction with broad absorption in the \\ion{C}{4} line is $\\simeq 15\\%$.  This fraction may be subject to small selection-efficiency adjustments.  There are also hints of redshift-dependence in the BAL fraction. The sample is large enough to permit the first estimate of the distribution of ``balnicity index\": subject to certain arbitrary parameters in the definition of this quantity, it is very broad, with (roughly) equal numbers of objects per logarithmic interval of balnicity.  BAL quasars are also found to be redder on average than non-BAL quasars.  The fraction of radio-loud BAL quasars is (weakly) consistent with the fraction of radio-loud ordinary quasars. ",
        "introduction": "Broad Absorption Line quasars (BALs) are one of the most enigmatic varieties of quasars.  Resonance lines of ordinary ions---\\ion{H}{1}, \\ion{C}{4}, \\ion{N}{5}, \\ion{O}{6}, \\ion{Mg}{2}, and others---are seen in absorption that spreads, often in highly irregular fashion, as much as 60,000~\\kms from line-center in the quasar rest-frame to the blueward.  Previous surveys (e.g. the Large Bright Quasar Survey, or LBQS: \\citet{W91}) have shown that BALs, while a minority of all quasars, are not rare; a population fraction $\\sim 10\\%$ is typically estimated.  Because few of their other properties are grossly different from ordinary quasars, it is generally thought that all quasars have BAL material, but it covers only a fraction of solid angle around the quasar nucleus \\citep{W91}.  However, subtleties of selection can complicate the inference of covering fraction from population fraction \\citep{G97,KV98}.  Numerous technical difficulties have retarded growth in our understanding of BALs.  Known cases are relatively rare, numbering less than $\\sim 100$, not solely because they are a minority of the general population but also because they are readily found only when their characteristic features are red-shifted from their rest-frame wavelengths in the ultraviolet into the visible band.  Consequently, only those quasars found in somewhat special redshift intervals can be easily searched for broad absorption.  It is hard to statistically characterize those BALs that are found because the methods used to discover them often involve some level of subjectivity that is hard to quantify.  Even if their selection were easier to articulate, there appears to be so much variation in their properties (profile shapes, relative line strengths, etc.) that it is hard to grasp which properties are generic and which are ``accidental\".  The quasar sample being compiled by the Sloan Digital Sky Survey (SDSS: \\citet{Y00}) offers a way out of this impasse.  When complete, it will be both very large ($\\sim 10^5$ in all) and selected in a uniform and quantifiable manner.  In future work, we hope to present statistical analysis of BALs in this entire sample.  Here we offer a preliminary installment on this project in the form of a more modest BAL sample drawn from the first data released from the project to public view, the Early Data Release (EDR: \\citet{Sto02}).  Several collections of BALs have already been drawn from early Sloan data \\citep{M01,H02}; these were, however, oriented toward ``by-eye\" selection of small subsamples special in some way (radio-loud in the former case, extraordinary profiles in the latter).   The work reported here differs in that it is the first attempt to create a systematically-selected sample from the SDSS.  From the EDR, \\citet{Sch02} created a quasar catalog containing 3814 quasars, selected (mostly) on the basis of their location in four-color space and on a (mostly) uniform $i$-magnitude limit.  In order to present more clearly-defined statistics, we have refined this sample so that it is almost homogeneously-selected (see \\S 2.1). Within that sub-sample (about 80\\% of the full EDR quasar catalog), roughly one-quarter (796) fall within the redshift range within which it is feasible to search for \\ion{C}{4} BAL features.  With an eye toward the homogeneity of selection to be achieved in the full SDSS, we invented an automated BAL selection algorithm that processes SDSS spectral data in a uniform way and identifies BAL quasars in a uniform manner (see \\S 2.2).  Using this algorithm, we have identified 116 BAL quasars, whose statistical properties are discussed in \\S 3.  Although the EDR represents a tiny fraction of the ultimate SDSS quasar sample, the BAL sample so derived is now the largest (as well as the most homogeneously selected) such sample known. ",
        "conclusions": ""
    },
    "0209/astro-ph0209562_arXiv.txt": {
        "abstract": "We have measured the central stellar velocity dispersion in the host galaxies of 11 BL Lac objects with redshifts $z \\leq 0.125$.  The range of velocity dispersions, $\\sim170-370$ \\kms, is similar to that of nearby radio galaxies.  Using the correlation between stellar velocity dispersion and black hole mass defined for nearby galaxies, we derive estimates of the black hole masses in the range $10^{7.9}-10^{9.2}$ \\msun.  We do not find any significant difference between the black hole masses in high-frequency-peaked and low-frequency-peaked BL Lac objects.  Combining the velocity dispersions with previously measured host galaxy structural parameters, we find that the host galaxies lie on the fundamental plane of elliptical galaxies.  This supports the conclusions of imaging studies that the majority of BL Lac hosts are normal giant ellipticals. ",
        "introduction": "One of the goals of AGN research is to develop a unified framework in which the diversity of the AGN family might be understood in terms of variations in a few fundamental parameters such as the black hole mass, the ratio of the accretion rate to the Eddington accretion rate, and the orientation relative to our line of sight \\citep[e.g.,][]{lao00, bor02}.  Determining the masses of the central black holes in different classes of AGNs is an important step toward this goal.  An key advance in this area was the recent discovery of the \\msigma\\ relation, a tight correlation between black hole mass and stellar velocity dispersion in the host galaxy bulge \\citep{fm00, geb00a}.  The \\msigma\\ relation has provided an important consistency check for black hole masses determined from reverberation mapping of Seyfert nuclei \\citep{nel00,geb00b,fer01}.  The tightness of the correlation over a wide range of host galaxy types, both elliptical and spiral, makes it tremendously useful as a means to determine the black hole masses in AGNs.  While stellar-dynamical or gas-dynamical black hole mass measurements are particularly difficult for AGNs, stellar velocity dispersions can be measured relatively easily in some classes of active galaxies.  Furthermore, black hole mass estimates based on velocity dispersions are expected to be much more accurate than estimates derived from the loose correlation between \\mbh\\ and bulge luminosity \\citep{mag98, kg01}.  We recently began a program to measure the stellar velocity dispersions in the host galaxies of low-redshift BL Lac objects in order to apply the \\msigma\\ relation and determine their black hole masses.  The first measurement of a velocity dispersion in a BL Lac object, for the TeV $\\gamma$-ray source Markarian 501, indicated a likely black hole mass of $(0.9-3.4)\\times10^9$ \\msun\\ \\citep[][Paper I]{bhs02a}.  In this paper, we report on measurements for a sample of 11 objects and discuss the results in the context of unification models for radio-loud AGNs.  We also compare our results with measurements of stellar velocity dispersions for BL Lac objects reported recently by \\citet[][hereinafter FKT]{fkt02}. ",
        "conclusions": "Our conclusions are summarized as follows:  1. Stellar velocity dispersions in low-redshift BL Lac objects are in the range $\\sim170-370$ \\kms.  Using the \\citet{tre02} fit to the \\msigma\\ relation, the corresponding black hole masses are $\\mbh \\approx 10^{7.9}$ to $10^{9.2}$ \\msun.  2. Measurement of velocity dispersions in BL Lac objects requires particular care in selecting spectral regions that are sensitive to $\\sigma$ but are not severely affected by interstellar absorption lines, telluric absorption bands, or emission lines.  Due to the dilution by nonthermal emission, high S/N (typically $\\gtrsim100$ per pixel in the extracted spectrum) is required for accurate measurements with a direct template-fitting method.  3. The distribution of velocity dispersions in BL Lac objects appears superficially similar to that of the radio galaxy sample of \\citet{bet01}, indicating a similar distribution of black hole masses for the two samples.  However, neither sample is statistically well defined or complete in any sense and we caution against drawing conclusions about radio-loud AGNs in general from this limited comparison.  4. There does not appear to be any systematic difference between the black hole masses in HBL and LBL objects.  5. The host galaxies of BL Lac objects lie on the fundamental plane of nearby elliptical galaxies.  We do not find any BL Lac objects near the top of the fundamental plane sequence in the region occupied by the most luminous hosts of FR I radio galaxies; this may be a consequence of selection effects in the identification of BL Lac objects.  6. For nearby, well-resolved BL Lac objects, the host galaxy effective radii measured in the \\hst\\ WFPC2 snapshot survey of \\citet{sca00} and \\citet{urr00} are systematically smaller than effective radii measured from deeper ground-based images.  We suggest that the measurements of \\reff\\ from the \\hst\\ snapshot survey images may be biased toward underestimates of the true radii, because they were derived from shallow exposures that did not detect the faint outer envelopes of the host galaxies that are seen in deep ground-based images."
    },
    "0209/astro-ph0209539_arXiv.txt": {
        "abstract": "The study of shapes of the images of objects is an important issue not only because it reveals its dynamical state but also it helps to understand the object's evolutionary history. We discuss a new technique in cosmological image analysis which is based on a set of non-parametric shape descriptors known as the Minkowski Functionals (MFs). These functionals are extremely versatile and under some conditions give a complete description of the geometrical properties of objects. We believe that MFs could be a useful tool to extract information about the shapes of galaxies, clusters of galaxies and superclusters. The information revealed by MFs can be utilized along with the knowledge obtained from currently popular methods and thus could improve our understanding of the true shapes of cosmological objects. ",
        "introduction": "The shape of an object provides important clues for understanding its nature, in particular, the past and  ongoing physical processes responsible for shaping the object. A mathematical branch known as the integral geometry offers an unique opportunity providing a set of simple morphological measures that can be used to characterize an isolated single objects as well as a multi-component object. These measures were suggested by H. Minkowski (Minkowski 1903) and known as the Minkowski Functionals (MFs). The parameters devised from MFs are robust, yielding local as well as global morphological information of any spatial structure. We discuss this set of measures in galaxy morphology specifically we propose a technique to restore galaxy images distorted by the background noise. The MFs was introduced into cosmology by Mecke, Buchert, \\& Wagner (1994) and later used in variety of cosmological problems (see e.g. Beisbart 2000; Beisbart, Buchert, \\& Wagner 2001; Novikov, Feldman \\& Shandarin, 1999; Schmalzing \\& Buchert 1997; Schmalzing \\& Gorski 1998; Schmalzing et al. 1999; Shandarin 2002; Shandarin et al. 2002; Shani, Sathyaprakash \\& Shandarin 1998). ",
        "conclusions": ""
    },
    "0209/astro-ph0209225_arXiv.txt": {
        "abstract": "The line profiles from rotating neutron stars are affected by a number of relativistic processes such as Doppler boosts, strong self-lensing, frame-dragging, and the differential gravitational redshift arising from the stellar oblateness. In this {\\em Letter}, we calculate line profiles taking into account the first two effects, which is accurate for rotation rates less than the breakup frequency. We show that the line profiles are not only broadened and weakened but are also significantly asymmetric, and allow for an independent measurement of both the mass and the radius of the neutron star. Furthermore, we investigate the case when a fraction of the neutron star surface contributes to the emission and find that the line profiles are typically doubly peaked. We discuss the implications of our results for searches for line features in the spectra of isolated neutron stars and X-ray bursters. We finally assess the systematic uncertainties introduced by the line asymmetry in inferring the compactness of neutron stars from the detection of redshifted lines. ",
        "introduction": "Thermal emission from the surface of a neutron star carries signatures of its strong gravitational field, which become apparent in observations of both its spectral and timing properties. Such measurements, therefore, can be used in principle to infer the masses and radii of neutron stars.  Over the past three decades, multiple attempts have been made to constrain the stellar equation of state using observations of the thermal emission from bursting (Lewin, van Paradijs, \\& Taam 1995), quiescent (e.g., Rutledge et al.\\ 1999), or isolated neutron stars (e.g., Pons et al.\\ 2002; Braje \\& Romani 2002), the X-ray pulse profiles of rotationally powered pulsars (Page 1995; Pavlov \\& Zavlin 1997), the high amplitudes of oscillations observed during thermonuclear bursts (Nath, Strohmayer, \\& Swank 2001), and the frequencies of observed quasi-periodic oscillations (Miller, Lamb, \\& Psaltis 1998).  Of all the possible methods of measuring the radius of a neutron star or a mass-radius combination, the one that suffers the least from systematic uncertainties and measurement errors is using the gravitational redshift of atomic spectral lines. For a slowly rotating, spherically symmetric neutron star, the gravitational redshift gives directly the stellar compactness (i.e., the ratio $R_{\\rm NS}/M_{\\rm NS}$). This method has received a lot of attention recently with the launch of X-ray telescopes with high spectral resolution (such as {\\em Chandra} and {\\em XMM-Newton}) and the discovery of thermal emission from nearby, isolated neutron stars (see Becker \\& Pavlov 2002 for a review). A number of observations of neutron stars have already been carried out, which yielded a potential detection of broad, redshifted absorption lines from the pulsar 1E~1207.4--5209 (Sanwal et al.\\ 2002; Mereghetti et al.\\ 2002). Other observations of isolated neutron stars have typically resulted in featureless X-ray spectra (e.g., Paerels et al.\\ 2001; Drake et al.\\ 2002; Marshall \\& Schulz 2002).  Apparently, not all targets are created equal. The detection of atomic spectral lines in X-rays requires both heavy metals to be present in the neutron-star atmosphere and the surface layers to have high temperatures for significant thermal emission to be generated.  These two requirements can be met most easily in either neutron stars that are young or in ones that are weakly magnetic and accreting steadily from a binary companion.  Young neutron stars emit thermally the heat released during their formation.  They may also possess heavy-element atmospheres if significant light-element fallback did not occur during the supernova: their short lifetimes and strong magnetic fields render unlikely a significant accumulation of hydrogen rich material from the interstellar medium that could suppress atomic lines. In the case of bursters, the heavy elements in their atmospheres are continually replenished by accretion and the thermonuclear flashes provide large amounts of thermal energy.  Both types of neutron stars that are prime candidates for the detection of spectral lines are fast rotators. (We do not consider here magnetars, for which the presence of ultrastrong magnetic fields introduces large uncertainties in calculating the rest energies of atomic lines). The spin frequencies of known pulsars with ages $<10^4$~yr is between $\\simeq 5-65$~Hz (see, e.g., Becker \\& Pavlov 2002); the inferred spin frequencies of bursters is between $\\simeq 270-620$~Hz (Strohmayer 2001). These high spin frequencies introduce several relativistic effects such as Doppler boosts, strong self-lensing, frame-dragging and differential gravitational redshift arising from the stellar oblateness. All of them alter the line profiles observed at infinity.  In this {\\em Letter}, we show the effects of relativistic Doppler boosts and strong gravitational lensing on the width and asymmetry of line profiles originating from the surfaces of rotating neutron stars. We then investigate the systematic uncertainties introduced by these effects in inferring the compactness of neutron stars. ",
        "conclusions": "We studied spectral line profiles from rotating neutron stars taking into account the effects of relativistic Doppler boosts and strong gravitational lensing. We showed that the line profiles are broad, as expected, and also significantly asymmetric. The asymmetry becomes more prominent when the surface emission is non-uniform. Our results have a number of implications for the current searches for gravitationally redshifted line features in the spectra of neutron stars.  First, the large widths and suppressed strengths of the rotationally broadened lines make their detection difficult. This may be able to account for the featureless spectra of a number of isolated neutron stars observed with {\\em Chandra} and {\\em XMM-Newton}, such as RXJ~1856--3754 (Braje \\& Romani 2002; Zavlin \\& Pavlov 2002). Correspondingly, if narrow line features are detected from rapidly rotating neutron stars, e.g., bursters, they could not have originated from the neutron star surface, unless the emission is restricted to the rotational pole. Searches for lines in such sources should take into account these relativistic effects.  \\begin{figure}[t] \\centerline{\\psfig{file=f4.eps,angle=0,width=10.5truecm}} \\figcaption{\\footnotesize Apparent ratio $R/M$ for a neutron star, when the peak of a spectral line is used to infer the magnitude of the gravitational redshift; the dotted line shows the true value of $R/M$. All the parameters are the same as in Figure~1.} \\end{figure}  Finally, the asymmetry of the line profiles introduces significant systematic uncertainties in measuring the compactness of a neutron star using gravitational redshifts. As an example, Figure~4 shows that if the peak of the line is used in measuring an apparent redshift, the resulting compactness of the neutron star will be significantly overestimated even when the entire surface is emitting. For the inferred spin frequencies of bursters, in the absence of realistic models, the systematic uncertainties can be as large as $10\\%$, which are larger than the $5\\%$ accuracy required to distinguish between the different equations of state (Prakash \\& Lattimer 2000)."
    },
    "0209/astro-ph0209155_arXiv.txt": {
        "abstract": "We report on the X-ray spectral variability of the Seyfert 1 galaxy NGC~4051 observed with the Rossi X-ray Timing Explorer ({\\em RXTE}) during a 1000 day period between May 1996 and March 1999.  The spectra were obtained as part of monitoring observations and from two long observations using the {\\em RXTE} Proportional Counter Array (PCA). During the monitoring period the 2-10 keV flux of NGC~4051 varied between $10^{-12}$ and $7\\cdot 10^{-11} {\\rm erg/(cm^2\\;s)}$.  We re-analysed {\\em RXTE} PCA observations from a distinct low state in May 1998 using the latest background and detector response models. The {\\em RXTE} and {\\em BeppoSAX} observations of NGC~4051 during the low state show a very hard spectrum with a strong unresolved fluorescence line.  This emission, probably due to reflection from a molecular torus, is likely to be constant over long time-scales and is therefore assumed as an underlying component at all flux states.  By subtracting the torus component we are able to  determine the spectral variability of the primary continuum.  In the variable component we observe a strong anti-correlation of X-ray flux and spectral hardness in the PCA energy band. We show that the changes in hardness are caused by slope variability of the primary power law spectrum rather than by changing reflection or variable photoelectric absorption. The primary spectral index varies between $\\Gamma=1.6$ for the faintest states and $\\Gamma=2.3$ during the brightest states, at which level the spectral index approaches an asympotic value. We find that the response of the flux of the 6.4 keV iron fluorescence line to changes in the continuum flux depends on the timescale of the observation. The profile of the line is very broad and indicates an origin in the innermost regions of the accretion disk. ",
        "introduction": "It is thought that the central engines of active galactic nuclei (AGN) are powered by the infall of matter onto a supermassive black hole. Theoretical arguments suggest the formation of an accretion disk which extends from the innermost stable orbit to an outer radius on the scale of light days. The popular unified scheme for Seyfert galaxies (eg Antonucci 1993) requires a dusty torus, probably coplanar with the accretion disk, which is opaque to radiation from the near infrared to soft X-rays. According to the unified scheme Seyfert type 1 and  Seyfert type 2 AGN are intrinsically identical object classes with the only difference being that in Seyfert type 2 galaxies, due to their different orientation relative to the observer, the  central source and the broad line region (BLR) are obscured by a dusty torus. The first probable direct detection of a torus was shown in our earlier SAX \\cite{Guainazzi} and RXTE \\cite{Uttley99} observations of NGC~4051. However, in general, detection of the torus and the determination of its geometry has not been possible.  X-ray observations have shed some light on the structure of the innermost regions of AGN. The soft X-ray spectrum often has a relatively steep slope, which has been explained by thermal emission from an accretion disk. However, there are alternative theoretical interpretations and due to the limited spectral resolution in this energy range the soft X-ray spectra are poorly understood.  At higher energies, a hard power law spectrum dominates  the emission. This component is probably due to Comptonization of the thermal UV photons in a hot  corona surrounding the disk.   The averaged {\\it Ginga} medium energy X-ray spectrum of several Seyfert galaxies showed the presence of an emission line at 6.4 keV, an absorption feature at 7-8 keV, and a further flattening of the spectrum beyond $\\sim 10$ keV.  These features are interpreted as iron $K_{\\alpha}$ emission, iron K edge absorption and Compton reflection and therefore are evidence for reprocessing of X-rays by relatively cold matter, The discovery of broadening due to gravitational redshift and Doppler shifts of the iron K$_{\\alpha}$ fluorescence lines in Seyfert X-ray spectra suggests that at least part of the reprocessing takes place in the inner parts of the accretion disk ( \\ncite{Tanaka}, \\ncite{Nandra97}, \\ncite{Reynolds}).  However, the detection of narrow components to the Fe K$_\\alpha$ line with {\\em ASCA} (eg \\ncite{Weaver97}) and recently {\\em Chandra} (\\ncite{Yaqoob}, \\ncite{Kaspi}) and {\\em XMM-Newton} \\cite{Reeves} suggests reprocessing of X-rays in matter at a larger distance from the black hole, eg in the BLR or in the molecular torus.  The analysis of X-ray variability is a powerful tool for the investigation of the inner regions of the AGN central engine, since the time-scales of the variability can give an indication of the geometrical sizes of the regions involved.  \\scite{Yaqoob96} reported a rapid ($<3\\cdot 10^4$ s) response of the Fe fluorescence line flux to the continuum flux in NGC~7314. From the resulting maximum distance of the reflecting material from the primary X-ray source they derive an upper limit on the mass of the black hole. However, tdhe analysis of fluorescence line variability in a number of other Seyfert galaxies have led to confusing and partly contradictory results. While in some sources a positive correlation of line and continuum flux has been observed, averaged over long timescales, (eg NGC~5506, \\ncite{Lamer2000}), other studies found no evidence for a close relationship between continuum and fluorescence line (\\ncite{Weaver2001}, \\ncite{Vaughan}).  Flux-dependent variability of the continuum spectral shape itself can be used to constrain models of the Comptonising accretion disk corona (eg \\ncite{Haardt}).  In general, the continuum slope appears to be correlated with flux (eg \\ncite{Leighly}; \\ncite{mch98} ; \\ncite{Lamer2000}), as predicted by most simple Comptonisation models.  However, it is important to determine whether the degree of slope variability, and the form of the correlation with flux (eg does the continuum slope saturate at some maximum value?) are in agreement with existing models.   NGC~4051 is a nearby (z=0.0023) low luminosity Seyfert 1 galaxy, which is among the most variable AGN in the X-ray band \\cite{Green}.  We have been monitoring NGC 4051 with RXTE since 1996.  Since our monitoring commenced, the 2-10 keV flux from the object has varied by a factor of $\\sim$100. In 1998 NGC~4051 entered a state of extremely low X-ray flux that lasted for $\\sim 150$ days (\\ncite{Guainazzi}, \\ncite{Uttley99}).  Here we report on the spectral changes that accompanied the dramatic flux variability. In section \\ref{obs} we describe the observations and the reduction of the {\\em RXTE} data. Results from observations in May 1998, during the extreme low state, are re-analysed in a consistent manner with the other observations and are discussed in section \\ref{lowstate}.  In section \\ref{specvar} we present the analysis of the continuum and fluorescence line variations observed during the long term monitoring and during an {\\em RXTE} long look in December 1996. ",
        "conclusions": "We show that the variable X-ray spectrum of NGC~4051 can be well described by a model that includes two principal components:  \\begin{itemize} \\item[1.] A hard, constant, component including a narrow iron fluorescence line as revealed during the extreme low state in May 1998.  \\item[2.] A variable component including a power law with strongly variable slope and a very broad emission line. We find a strong correlation between the power law slope and the source flux.  On the long time-scales of the monitoring observations the line flux is correlated with the continuum flux, although not in a simple manner. The reflected fraction of the disk component is less than $R=1$ at all flux levels.  \\end{itemize}  This two-component model is consistent with the model we proposed for the X-ray spectral variability of the Seyfert 2 galaxy NGC~5506 \\cite{Lamer2000}.  We note that even if the constant hard component is not included in our spectral fits, the general form of our results is not significantly affected, as the assumed constant component is relatively weak. Indeed, the lack of inclusion of a hard continuum component would result in an even larger spectral index variation with flux.  \\subsection{The Iron Line}  Even on the long timescales probed by our time-averaged monitoring data, the flux-dependent behaviour of the broad iron line is complex.  For these data, the line flux increases more-or-less proportionally with the continuum flux, resulting in a roughly constant equivalent width over a decade range of flux.  The December 1996 long-look data also show line flux increasing with continuum flux, although the relation is not directly proportionate, so that the equivalent width decreases with flux.  In fact, the line fluxes measured in December 1996 are consistently larger than the corresponding fluxes measured from the long-term monitoring data. The anti-correlation of line equivalent width and continuum flux in December 1996 timescales is also in  contrast  to the result of \\cite {Wang}, who report a positive  correlation during an {\\it ASCA} observation in 1994.  The discrepancy between the iron line behaviour in the December 1996 and long-term monitoring data might be explained if there is additional short-term variability in the iron line which is not simply related to the continuum flux. For example, Vaughan \\& Edelson (2001) show that in the Seyfert~1 MCG-6-30-15, the broad iron line flux varies significantly but independently of short term continuum variations.  One possibility is that the iron line flux tracks the long-term variations in the continuum flux (which are being probed to some extent with the long-term monitoring data), but responds only weakly to the short-term variations which are observed during the December 1996 long-look observation. A number of theoretical papers have been written to explain why the iron line flux may not vary linearly with the continuum flux (eg \\ncite{Matt}, \\ncite{Nayakshin}, \\ncite{Ballantyne}), often involving ionised discs but these models have so far been largely untroubled by data. We are aquiring more long-term monitoring data, sampling a broader range of long-term flux variations,  to determine whether the iron line does follow the continuum on long timescales and to provide some constraints for theoretical models.    \\subsection{The Reflected Component}  Our observations of NGC~4051 do not support the correlation between the photon index and the reflected fraction $R$ in the {\\sc pexrav} model as reported from {\\em ASCA} spectroscopy of a sample of Seyfert galaxies \\cite{Zdziarski} .  From Fig. \\ref{dec_cont} it is obvious that the reflected fraction remains below $R=1$ even for the softest states of the source. There is also no evidence for this correlation in the {\\em RXTE} spectra of NGC~5506 \\cite{Lamer2000}. We therefore suggest that the reported correlation does not apply to the variations of photon index and reflected fraction in a given object.   \\subsection{Continuum Spectral Variability}  During our monitoring campaign the primary continuum 2-10 keV photon index, $\\Gamma$, varies strongly with photon flux from 1.60 at the lowest flux levels to 2.35 at the highest fluxes.  The same correlation is also observed on the shorter time-scales of the December 1996 long look. Flux-$\\Gamma$ correlations have been observed before in NGC~4051 \\cite{Matsuoka} and other Seyfert galaxies (eg NGC~4151, \\ncite{Perola}) but, as in the present paper, it has only recently been possible to disentangle the effects of reflection and variations of the primary X-ray spectrum (eg \\ncite{Lamer2000}, \\ncite{Chiang}, \\ncite{Lee2000}).  The slope-luminosity correlation is often explained by stronger cooling of the accretion disk corona during episodes of high thermal seed photon flux from the accretion disk itself (eg \\ncite{Pietrini}, \\ncite{Malzac}).  \\scite{Haardt} have calculated luminosity -- spectral index relations in Compton cooled accretion disk coronae.  For a compact, pair dominated corona they predict spectral index variations of $\\Delta\\Gamma\\sim 0.3$ for luminosity variations by more than a factor of 20. However the variations seen here exceed their predictions and imply, in their scenario, a non-pair dominated corona. In certain regimes this model predicts a positve correlation of 2-10 keV flux and spectral hardness, which is not observed in NGC~4051. \\scite{Pietrini} point out that the spectral index depends almost solely on the ratio of seed photon compactness $l_s$ and hot plasma heating rate compactness $l_h$ with $\\alpha=1.6(l_s/l_h)^{1/4}$. The observed spectral indices in NGC~4051 then correspond to $l_s/l_h=0.02..0.4$.  Examination of fig~\\ref{lineflux} shows that the change of spectral index with flux is not linear. The rate of increase of spectral index with flux is very rapid at low fluxes but decreases at the highest fluxes where the spectral index approaches an asymptotic level.  This saturation of the `spectral index/flux' relationship has been known for some time; eg the saturation was clearly visible in our early RXTE monitoring observations of MCG-6-30-15 and was reported by \\scite{mch98} where it was suggested that the relationship might derive from the combination of a constant spectrum hard component, and a steeper spectrum variable component.  Saturation of the spectral index/flux relationship was again reported in MCG-6-30-15 by \\scite{Shih} from a long ASCA observation. In MCG-6-30-15 both the long term RXTE monitoring and short term ASCA observations agree that the saturation level of the spectral index is $\\sim2.1$ (see fig 7 of \\ncite{mch98} and fig 8 of \\ncite{Shih}). However the monitoring observations cover a wider flux and spectral range and show variation of the spectral index between 1.65 and 2.05 whereas the continuous ASCA observation only shows an index variation between 1.9 and 2.1. Similarly in NGC~4051 (fig~\\ref{lineflux}) we see that the monitoring observations cover a wider flux and spectral range than the December 1996 long look and, as with MCG-6-30-15, the resultant time-averaged spectral index/flux relationship is smoother. We note, however, that although the lowest spectral index so far measured in the RXTE monitoring observations is about the same in both NGC~4051 and MCG-6-30-15, the saturation level is $\\sim2.4$ in NGC~4051 compared to $\\sim2.1$ in MCG-6-30-15.  In \\scite{mch98} we suggested that the torus might be the source of the possible hard, constant, component. However the very hard spectral component found in the May 1998 very low state, which represents an upper limit to the torus contribution, was removed before producing the spectral index/flux relationship (fig~\\ref{lineflux}). Thus, if we wish to retain the two-component spectral model, we require a different location for the bulk of the hard component. However any hard component produced closer to the central source by reprocessing of primary radiation is likely to vary, although probably not as rapidly as the primary continuum. Our time-averaged spectra wash out short term variability so a hard component varying slowly in line with the primary continuum could not produce the observed spectral index flux relationship.  Unless a hard component is produced by some mechanism other than reprocessing it is therefore more likely that the change of spectral index with flux is driven by some change in the physical properties of the primary continuum source, eg by changes in the seed photon populations as discussed above.  Any such physical model of the primary continuum must be able to produce different saturation levels in different sources and must be able to account for both the wide spectral range, and the very hard spectra measured at the lowest flux levels, in our monitoring observations."
    },
    "0209/astro-ph0209363_arXiv.txt": {
        "abstract": "We study correlations betwen X-ray spectral index, strength of Compton reflection, and X-ray and radio fluxes in accreting black holes (Seyferts and black-hole binaries). We critically evaluate the evidence for the correlation of the X-ray spectral index with the strength of Compton reflection and consider in detail statistical and systematic effects that can affect it. We study patterns of spectral variability (in particular, pivoting of a power law spectrum) corresponding to the X-ray index-flux correlation. We also consider implications of the form of observed X-ray spectra and their variability for interpretation of the correlation between the radio and X-ray fluxes. Finally, we discuss accretion geometries that can account for the correlations and their overall theoretical interpretations. ",
        "introduction": "\\label{intro}  X-ray and soft \\g-ray (hereafter X$\\gamma$) spectra from luminous accreting black holes (hereafter BH), i.e., AGNs and BH binaries, commonly show a distinct component due to Compton reflection (Lightman \\& White 1988; Magdziarz \\& Zdziarski 1995) of the primary continuum from a cold medium (e.g., Pounds et al.\\ 1990; Nandra \\& Pounds 1994; Magdziarz et al.\\ 1998; Weaver, Krolik \\& Pier 1998; Zdziarski, Lubi\\'nski \\& Smith 1999, hereafter ZLS99; Done, Madejski \\& \\.Zycki 2000; Eracleous, Sambruna \\& Mushotzky 2000; Done et al.\\ 1992; Gierli\\'nski et al.\\ 1997; Zdziarski et al.\\ 1998; \\.Zycki, Done, \\& Smith 1998; 1999; Gilfanov, Churazov \\& Revnivtsev 1999, 2000, hereafter GCR00; Revnivtsev, Gilfanov \\& Churazov 1999, 2001). A very interesting property of Compton reflection with a number of potential physical implications is that its relative strength, $\\sim \\Omega/2\\upi$, where $\\Omega$ is the solid angle of the cold reflector as seen from the hot plasma, correlates with some other spectral and timing properties of many sources (e.g., ZLS99; GCR00).  Another correlation often found in both Seyferts and BH binaries is between X-ray spectral index and X-ray flux (e.g., Chiang et al.\\ 2000; Done et al.\\ 2000; Nowak, Wilms \\& Dove 2002; Zdziarski et al.\\ 2002b, hereafter Z02; Lamer et al.\\ 2003; Gliozzi, Sambruna \\& Eracleous 2003). The X-ray flux is also correlated with the level of radio emission in the hard states of BH binaries (Corbel et al.\\ 2000; Gallo, Fender \\& Pooley 2003). All these correlations appear to reflect fundamental properties of BH accretion flows. We critically study the correlations, relationships between them, their theoretical models, and the corresponding physical implications. In Appendix A, we present properties of spectral variability due to a power-law pivoting, which process is closely related to the flux-index correlation as well as it puts constraints on the interpretation of the radio--X-ray correlation. In spectral fits, we use {\\sc xspec} (Arnaud 1996). ",
        "conclusions": "We have considered correlations between various spectral properties of accreting BHs in Seyfert galaxies and X-ray binaries, with particular emphasis on the correlations between the X-ray spectral index, strength of Compton reflection and the X-ray flux. The main results can be summarized as follows.  Using published data of the observations of Seyferts with \\ginga, \\xte\\/ and \\sax, we have critically re-evaluated the evidence for presence of correlation between the X-ray spectral index and strength of Compton reflection. We conclude that when considering a large number of observations of a large sample of objects, the existence of {\\it global\\/} correlation between these two parameters is established beyond any reasonable doubts.  Smallness of the error bars in comparison with the extent of the correlation and good agreement of the results obtained by various satellites confirm that the correlation cannot be an artifact caused by statistical or systematic effects. The ratios of the spectra with different values of reflection in BH binaries demonstrate that the correlation cannot be a consequence of a trivially inadequate spectral model. We note, however, that the particular values of the spectral index and, especially, of the strength of the Compton reflection, do depend on the details of the spectral approximation. This fact should taken into consideration when comparing results obtained by different authors.  The $\\Omega$-$\\Gamma$ correlation shows significant spread, larger than the statistical uncertainties of the data. It is not clear at present to which degree this spread is intrinsic to the sources and to which degree it is due to imperfectness of the spectral model and/or difference in the details of the spectral approximation.  Distinction should be made between classes of objects and multiple observations of individual objects. In the case of luminous BH binaries, spectral variability of individual sources obeys the general $\\Omega$-$\\Gamma$ correlation obtained for BH binaries as class. In the case of Seyfert galaxies, the correlations holds for a class of objects but can be violated for repeated observations of individual objects, remaining, however, within the spread of the global correlation. In our opinion, these results are still inconclusive and require further investigation as a number of complications are involved in the case of Seyfert galaxies in comparison with X-ray binaries. The most obvious among those are lower statistics due to significantly lower brightness of Seyferts and existence of molecular tori and  broad line regions, which can give additional contribution to the reflected component, and uncorrelated with the Comptonized emission on short time scales.  The physical interpretion of the $\\Omega$-$\\Gamma$ and $\\Gamma$-$F$ correlations will advance our understanding of the geometry of the accretion flow in X-ray binaries and Seyfert galaxies and impose valuable constrains on the theoretical models. At present, the correlations appear to be a natural consequence of co-existence of cold media (e.g., an accretion disc) and a hot Comptonizing cloud in the vicinity of the BH. Their geometrical closeness results in double feedback between the two components of the accretion flow. The cold medium provides seed soft photons for the Comptonization as well as it reprocesses and reflects the hard radiation from the hot plasma. If significant fraction of soft seed photons is due to reprocessing of the hard radiation of the hot cloud, the more reprocessing results in more cooling of the hot plasma and, correspondingly, in the softer X-ray spectra. Among the several specific geometries proposed, the most promising appears to be the disc-spheroid  model with variable overlap between the hot and cold components of the accretion flow.  Since the characteristic temperature of the accretion disk in AGNs is much lower than in BH binaries, a generic prediction of this type of models is that for a given value of reflection the value of the spectral index will be higher in AGNs than in BH binaries, This prediction is in good agreement with the observed behavior.  We presented a diagnostic to test an interpretation of the $\\Omega$-$\\Gamma$ correlation as due to strong ionization of the disc surface layer. It utilizes the independence of the Klein-Nishina cutoff in the reflected spectrum of the ionization state. Applied to existing data, this diagnostic does not support that interpretation.  We found that the pattern of broad-band spectral variability of Seyfert galaxies on day-to-month time scale includes a pivoting of a power law spectrum with the pivot at a high energy, usuallly above a few hundred keV. The pivoting well explains the linear correlations between the logarithm of the X-ray flux and $\\Gamma$ seen in both BH binaries and Seyferts. Then, the observed high pivot energies rule out the interpretation of the $\\Omega$-$\\Gamma$ correlation is Seyferts as due to a time-lag effect.  We discuss also the correlation between X-ray and radio fluxes in BH binaries and its physical implications. We conclude that this correlation is most likely due to relation between the level of X-ray emission and the rate of ejection of radio-emitting clouds forming a compact jet. Although peculiar sources might exist, it seems highly unlikely that the correlation is due to synchrotron origin of the X-ray emission from BH binaries in general."
    },
    "0209/astro-ph0209405_arXiv.txt": {
        "abstract": "We have developed a code that models the formation, destruction, radiative transfer, and vibrational/rotational excitation of H$_2$ in a detailed fashion.  We discuss generally how such codes, together with FUSE observations of H$_2$ in diffuse and translucent lines of sight, may be used to infer various physical parameters.  We illustrate the effects of changes in the major physical parameters (UV radiation field, gas density, metallicity), and we point out the extent to which changes in one parameter may be mirrored by changes in another.  We provide an analytic formula for the molecular fraction, f$_{H2}$, as a function of cloud column density, radiation fields, and grain formation rate of of H$_2$.  Some diffuse and translucent lines of sight may be concatenations of multiple distinct clouds viewed together.  Such situations can give rise to observables that agree with the data, complicating the problem of uniquely identifying one set of physical parameters with a line of sight.  Finally, we illustrate the application of our code to an ensemble of data, such as our FUSE survey of H$_2$ in the Large and Small Magellanic Clouds (LMC/SMC), in order to constrain the elevated UV radiation field intensity and reduced grain formation rate of H$_2$ in those low-metallicity environments. ",
        "introduction": "\\subsection{Overview} No molecule in astrophysics is as ubiquitous or as far-reaching in its effects as molecular hydrogen (H$_2$).  Found in nearly every physical environment and in every temporal domain, it was the first neutral molecule to form after the Big Bang, it is a major constituent of giant molecular clouds, and it forms the bulk of the atmospheres of Jovian planets.  Of particular relevance to our interests is the fact that H$_2$ is the most abundant molecule in the interstellar medium (ISM); nearly every target through the Galactic disk and halo observed with the Far Ultraviolet Spectrographic Explorer (FUSE) since its launch in June 1999 has exhibited signs of H$_2$ along the line of sight (Shull et al. 2000). Studying H$_2$ is a major scientific goal of FUSE; in addition to the diffuse cloud program, nearing completion with $\\sim$ 100 targets in the Galactic disk, active FUSE campaigns exist to study H$_2$ in the LMC/SMC (Tumlinson et al. 2002) and in denser ``translucent'' clouds having a visual extinction in the range of about $A_V = 1 \\rightarrow 5$ (Snow et al. 2000; Rachford et al. 2001). Thoroughly understanding such clouds of H$_2$, which are the raw ingredients out of which giant molecular clouds and, later, stars form could lead to better comprehension of that process or the physical nature of the ISM in general.  Here, we discuss new computational models of interstellar clouds of H$_2$ and their application to FUSE data.  Information on the theory and construction of these models, a summary of their application to large FUSE datasets, and interpretations of the results form the bulk of this paper.   \\subsection{Background material}  A brief reminder of the physics of the H$_2$ molecule is in order. Molecular hydrogen has quantized electronic, vibrational, and rotational degrees of freedom, giving rise to a commensurate set of energy levels.  The ground electronic state ($X^{1} \\Sigma_{g}^{+}$) is split into a number of vibrational levels (labeled $v=0, 1,..., 14$), which are in turn split into rotational levels ($J=0, 1, ...$). Higher electronic states (the next two of singlet symmetry are labeled $B^{1}\\Sigma_{u}^{+}$ and $C^{1}\\Pi_{u}$) are also split into vibrational and rotational levels, and absorptions from the ground state into these electronic states are referred to as the Lyman ($B^{1}\\Sigma_{u}^{+} \\leftarrow X^{1} \\Sigma_{g}^{+}$) or Werner ($C^{1}\\Pi_{u} \\leftarrow X^{1} \\Sigma_{g}^{+}$) bands.  Molecular hydrogen in its ground electronic state has no permanent electric dipole moment, and hence no dipole-allowed vibrational or rotational transitions. Consequently, cold H$_2$ is most readily detected via observation of its ultraviolet absorption in these electronic bands (Shull \\& Beckwith 1982).  The Lyman and Werner bands occur in the far-ultraviolet portion of the spectrum, and so their observation is entirely the purview of space-based missions.  Early efforts in this vein (see Snow 2000, and references therein) included rocket-borne spectrographs (e.g., Carruthers 1967), the \\emph{Copernicus} mission, the Interstellar Medium Absorption Profile Spectrograph, the ORFEUS spectrograph, and the Hopkins Ultraviolet Telescope.  The data we consider here are from FUSE -- the most recent mission capable of attacking this problem -- the details of which are described in Moos et al. (2000) and Sahnow et al. (2000).  FUSE is a multi-channel Rowland circle spectrograph, with coverage of the spectral range from approximately 912 to 1187 \\AA\\, and a spectral resolution of approximately $R=20,000$. It is FUSE's ability to measure accurate column densities in rotational and vibrational levels that makes possible the detailed analysis of diffuse and translucent lines of sight described here. ",
        "conclusions": "We have presented computational models of clouds of H$_2$ and selected applications of those models to FUSE observations.  Relatively simple models can duplicate the observed properties of diffuse and translucent clouds fairly well, but there are complications.  Changes in the major physical parameters of such clouds can have many competing effects on observables, and we have demonstrated the general trends obeyed by changes in each parameter.  The effects of changes in one parameter can be mirrored by changes in others, so uniquely identifying a line of sight with one set of physical properties is in some cases not feasible.  Rather, consideration should be given to the classes of models which can fit a given ensemble of observations.  We suggest that some high-$N(H_{2})$ ``clouds'' may be separate and physically distinct absorbers viewed along a common line of sight.  This interpretation is supported by the fact that the modeled rotational excitation matches the observations, but also by what is needed if one does not allow such combinations: the high-$N(H_2)$ cloud targets can generally be matched by a single cloud only if that cloud is exposed to a high ($\\gtrsim$ 20 times Galactic mean) radiation field.  For some Galactic targets, such a radiation field is probably not likely; to explain these observations without recourse to high radiation field requires some additional pathway by which radiation may enter the system.  Multiple-component clouds represent such a pathway, if each cloud is physically distinct from the others and therefore does not filter out the incident radiation.  Finally, we have illustrated the application of the code to a large FUSE survey of H$_2$ in the LMC and SMC.  Ensembles of models, though not unique, can match the observed patterns of molecular abundance and rotational excitation.  We find evidence for an enhanced UV radiation field (10 to 100 times the Galactic mean), and a reduced grain formation rate of H$_2$ ($R \\approx 3 \\times 10^{-18}$ cm$^{3}$ s$^{-1}$, one-tenth the nominal Galactic rate)."
    },
    "0209/astro-ph0209319_arXiv.txt": {
        "abstract": "We present time-dependent numerical hydrodynamic models of line-driven accretion disk winds in cataclysmic variable systems and calculate wind mass-loss rates and terminal velocities. The models are 2.5-dimensional, include an energy balance condition with radiative heating and cooling processes, and includes local ionization equilibrium introducing time dependence and spatial dependence on the line radiation force parameters. The radiation field is assumed to originate in an optically thick accretion disk. Wind ion populations are calculated under the assumption that local ionization equilibrium is determined by photoionization and radiative recombination, similar to a photoionized nebula. We find a steady wind flowing from the accretion disk. Radiative heating tends to maintain the temperature in the higher density wind regions near the disk surface, rather than cooling adiabatically. For a disk luminosity $L_{disk}=L_{\\sun}$, white dwarf mass $M_{wd}=0.6 M_{\\sun}$, and white dwarf radii $R_{wd}=0.01 R_{\\sun}$, we obtain a wind mass-loss rate of $\\dot M_{wind}=4 \\times 10^{-12} M_{\\sun} \\, {\\rm yr}^{-1}$, and a terminal velocity of $\\sim 3000 {\\rm \\, km \\, s}^{-1}$. These results confirm the general velocity and density structures found in our earlier constant ionization equilibrium adiabatic CV wind models. Further we establish here 2.5D numerical models that can be extended to QSO/AGN winds where the local ionization equilibrium will play a crucial role in the overall dynamics. ",
        "introduction": "Ultraviolet observations of Cataclysmic Variables (CVs) show that virtually all nonmagnetic CVs with high mass accretion rates ($\\gtrsim 4\\times 10^{-10} M_{\\sun} \\, {\\rm yr}^{-1}$) exhibit P~Cygni profiles in the resonance lines \\citep[e.g.][] {cor82,gre82,gui82,kla82,vit93,pri95,fri97,gan97,kni97,pri00}. These profiles consist of blue-shifted absorption and redshifted  emission, with the emission most prominent in high inclination systems \\citep[e.g.][]{hut80,kra81,hol82,cor85,mas95,che97}, and the absorption appearing at low and intermediate inclinations. The most straightforward explanation for these observations is in terms of a wind originating from the accretion disk. The wind is driven by radiation pressure from the disk UV continuum absorbed by line transitions, analogous to the radiation driven winds in early type stars. Support for this scenario came from \\citet{dre87}, who calculated theoretical \\ion{C}{4} 1549 \\AA \\  line-profiles from one-dimensional kinematic CV wind models, for a wide variety of wind and disk parameters. \\citet{dre87} found that the  dependence on the form of the UV resonance lines with inclination angle could be accounted for by a bipolar wind emerging from an accretion disk.  One-dimensional dynamical models for line-driven winds (LDW) in CVs were developed independently by \\citet{vit88} and \\citet{kal88}. Further efforts came through two-dimensional kinematic modeling \\citep{shl93,kni95}, which succeeded in showing consistency between the assumed polar geometry of a disk wind and observed profiles; but such models were unable to self-consistently calculate the wind dynamics including the effects of rotation and the anisotropic disk radiation field.  Two-dimensional disk wind dynamical models were developed by \\citet{ick80}, but these did not take into account the radiation pressure due to line absorption. For a typical white dwarf mass of $0.6M_{\\sun}$ \\citep{lei80,sil01} the disk luminosity required to produce such a wind without line radiation pressure (assuming radiation pressure due to continuum scattering only) would require a luminosity of $\\sim 10^5 L_{\\sun}$ \\citep{ick80}, several magnitudes above the observed luminosity of CVs which vary between $0.01L_{\\sun}$ and $10 L_{\\sun}$ \\citep{pat84}. \\citet{ick80} did find that, for sufficiently high disk luminosities, a biconical disk wind would form. \\citet{ick81} suggested that biconical winds were a general property of accretion disk winds independent of the wind driving mechanism.  Two-dimensional dynamical models of isothermal line-driven disk winds were presented by us \\citet[hereafter Paper~1]{per97a}. Results from Paper~1 showed, in analogy with line-driven winds from early type stars, that terminal velocities are approximately independent of the luminosity of the disk, although increments in luminosity produce increments in mass-loss rate, and that rotational forces are important in the study of winds from accretion disks. They cause the velocity streamlines to collide, which reduce the speed and increase the density of the wind producing an enhanced density region. The highest absorption occurs in the enhanced density region where density is increased relative to a spherically diverging wind with the same mass loss rate and the velocity is roughly half the terminal velocity. This density increase is necessary in order to produce at least marginally optically thick lines.  We developed \\citep[hereafter Paper~2]{per97b,per00} an adiabatic wind model (rather than isothermal), driven by a standard accretion disk \\citep{sha73,lyn74} rather than an isothermal disk. The hydrodynamic models developed of line-driven accretion disk winds (LDADWs) in CVs included the radial structure of an optically thick accretion disk with the corresponding radiation fields and surface temperature distributions. The corresponding energy conservation equation was also implemented, including the adiabatic heating and cooling effects due to compression and expansion. From the computational models, assuming single scattering and constant ionization, we calculated theoretical line profiles for the \\ion{C}{4} 1550~\\AA \\ line, and found that the line profiles obtained were consistent with observations in their general form and strong dependence with inclination angle.  \\citet{pro98} also developed two-dimensional models for these systems, obtaining similar results to the models presented here. They assumed an isothermal wind and constant values for the line radiation parameters throughout the wind. They found similar wind velocities and wind mass loss rates as we do here, although they assumed disk luminosities about an order of magnitude greater. A difference we find with the results of the models presented by \\citet{pro98} is that they obtain unsteady flows characterized by large amplitude fluctuations in velocity and density, while for our previous models (Paper~1; Pereyra~1997; Paper~2) and in the models presented here we find a steady flow. We note that these large fluctuations (rather than the steady flow we find) could account for the difference in luminosity required by the Proga models, with respect to our models, to obtain similar wind mass loss rates. One possibility suggested by \\citet{pro98} was that the spatial resolution of the models we presented in Paper~1 was too low to adequately resolve the wind. In paper~2 we increased significantly the spatial resolution of our models and obtained similar steady results when compared to our earlier spatial resolutions used in Paper~1; thus showing that differences in spatial resolution of the models was not generating the difference between our steady disk wind models and the ``intrinsically unsteady'' disk winds reported by \\citet{pro98}. Another possibility is that the difference in the unsteady flows of the Proga models with respect to the steady flows we find is due to the boundary condition treatment in the Proga models, which appears to lead to instabilities at the base of the wind and in turn to the strong fluctuations reported.  It is well known that for early type stars, when a Sobolev treatment is used for the line radiation pressure (as we have done in our models and as has been applied by \\citet{pro98} for accretion disk winds in CVs), a steady wind solution is obtained \\citep[e.g.][]{owo99}. The treatment of boundary conditions in numerical models for line driven winds in early type stars was discussed by \\citet{owo94} and \\citet{cra96}. They find that the treatment of the lower (base of wind) boundary condition is ``somewhat problematic''. Further they find that if they used an arbitrarily high density, similar to the boundary conditions used by \\citet{pro98}, they would find oscillations at the base of the wind in both density and velocity. In both papers, in addition to other details in the boundary condition treatment \\citep{owo94,cra96}, they lowered the density at the inner boundary until they found steady conditions at the base and were simultaneously able to resolve acceleration in the subsonic region. Therefore  a conclusion in the area of numerical modeling that we derive is that by varying the treatment of boundary conditions at the base of a line driven wind, one can obtain strong fluctuations in density and velocity which are not necessarily physically intrinsic but are rather a consequence of the numerical details of the model.  \\citet{fel99a} and \\citet{fel99b} developed two-dimensional stationary models for CV disk winds where they solve the dynamics under the assumption that the wind streamlines lie on straight cones. The values of wind mass loss rates obtained through the models presented by \\citet{fel99b} for an isothermal disk (equation~[11] of their paper) are consistent with those found in Paper~1. In addition \\citet{fel99b} find that in an isothermal disk (equations~[10]-[11] of their paper) the wind mass loss rate scales as $\\dot M_{wind} \\propto L_{disk}^{1/\\alpha}$, where $L_{disk}$ is the disk luminosity and $\\alpha$ is line force multiplier parameter. This is the same scaling law reported in Paper~1. Further, the biconical geometry and wind tilt angle in the models of \\citet{fel99b} are roughly consistent with the  velocity structures we find through full 2.5D hydrodynamic modeling of our earlier models (Paper~1; Pereyra~1997; Paper~2) and in the models presented here. We further note that Feldmeier et~al. suggest that detailed non-LTE calculations of the line-force multiplier may be important in the modeling of CV disk winds. This work is an additional step in that direction.  In spite of these encouraging results, a more realistic model which would include a detailed calculation of the UV spectrum from dynamically calculated density and velocity fields is desirable in order to make specific comparisons with observations. For instance, a maximum flux  greater than $1.7$ times the continuum has been reported for the emission component of P~Cygni line profiles of \\ion{C}{4} 1550\\AA \\ from some cataclysmic variables \\citep[e.g.][]{cor82}, while from our previous models (Pereyra~1997; Paper~2) we find that the emission maximum of the \\ion{C}{4} lines stays below 1.3 times the continuum flux for those angles where the absorption component is observed. This discrepancy may be due to some of the simplifications in our previous models or it may indicate a \\ion{C}{4} line-emitting region other than the disk wind itself. Furthermore, in recent observational results, \\citet{fro01} find relatively narrow absorption lines ($\\leq 700 \\, {\\rm km \\, s}^{-1}$) in the FUV spectra of the cataclysmic variable U~Gem. As \\citet{fro01} have indicated, such lines cannot arise from the disk photosphere. In their work they suggest the possibility of an outer disk chromosphere to account for these lines, but it is also possible that some of these narrow FUV lines could be accounted for by an accretion disk wind.  An additional motivation for this work is that the models developed here can also be extended to the study of LDADWs in Quasars (QSOs), where the local ionization equilibrium will play an important in the overall dynamics \\citep{mur95,hil02}. Evidence for winds in QSOs is found in the broad absorption lines (BALs) observed in approximately 10\\% of the QSOs \\citep[e.g.][]{wey97}. If the wind in QSOs is generated from an accretion disk, then the presence of BALs in only a fraction of QSOs can be accounted for as a viewing-angle effect \\citep{tur84}; similar to the case of CVs where the accretion disk wind produces P~Cygni profiles when observed ``face-on'' and emission lines (without absorption components) when seen ``edge-on'' \\citep{per00}. The x-ray luminosity in QSOs are generally comparable to the UV/optical luminosities \\citep[e.g.][]{tan79,gri80,mus93,lao97,geo00}. \\citet{dre84} found that such high x-ray luminosities would ionize the wind to a point where the populations of ions responsible for the BAL observed in QSOs would be too low to produce observable absorption lines. \\citet{mur95} found that, with an appropriate x-ray shielding mechanism, LDADWs can produce wind densities and velocities consistent with BALs observed in QSOs. Strong changes in the ionization balance can produce observable effects in line formation in QSOs; as well as significant changes in the line radiation force parameters which will play an important role in the overall wind dynamics \\citep{hil02}.  In our previous models (Pereyra~1997; Paper~2) we had assumed constant ionization equilibrium throughout the wind and single scattering in the line profile calculations. In this work we extend our previous models to include radiative heating and cooling and ionization balance in the wind, and a calculation of the line radiation force parameters at each point in space at each time step. Our results are similar to those of Paper~2, and therefore act to confirm the general results found in our previous simpler models. They are a further step towards a model capable of self-consistently producing theoretical spectrum which may be compared in detail with observations; they can also be extended to the study of LDADWs in QSOs/AGN, where the local ionization equilibrium will play a crucial role in the overall dynamics \\citep{mur95,hil02}.  In \\S\\ref{sec_adisk} we discuss the radial structure of the accretion disk and the radiation field as implemented in our models. In \\S\\ref{sec_ioneq} we present the ionization balance calculations applied in this work. We derive the expression used for the treatment of the line radiation pressure in \\S\\ref{sec_rf}. The treatment of radiative heating and cooling in this work is discussed in \\S\\ref{sec_rhc}. In \\S\\ref{sec_hyd} we present and discuss the hydrodynamic calculations. The results of the models are presented and discussed in \\S\\ref{sec_res}. We present a summary and the conclusions of this work in \\S\\ref{sec_sum}. ",
        "conclusions": "\\label{sec_sum}  We have developed a 2.5-dimensional hydrodynamic line-driven accretion disk wind model. Our model solves a complete set of adiabatic hydrodynamic partial differential equations, using the PPM numerical scheme and implementing the radial temperature and radiation emission distributions on the surface of an accretion disk. Our models calculate the line radiation force parameters throughout the wind at each computational grid point for each time step through the calculation of line opacities from atomic data, similar to the calculations presented by \\citet{abb82} for OB stars. We treat the hydrodynamic equation of energy self consistently including radiative heating and cooling as well as adiabatic expansion and compression.  We find steady wind flows coming from the disk with similar density and velocity structures as our earlier models of Paper~2 in which we had not yet included radiative heating and cooling and in which we had assumed constant ionization equilibrium (therefore constant values for the line radiation force parameters). The steady nature of our solution contrasts with the strong fluctuations in density and velocity found by \\citet{pro98} for CV winds for physical parameters similar to the ones we used here. A possibility that could account for the difference in results from our models and those of \\citet{pro98} is the treatment of the lower boundary conditions. In future work we shall explore this possibility in detail.  We note that as our numerical models evolve towards more realistic ones, it may result that a more accurate treatment of the line radiation force (which would include line-overlapping, rather than assuming the Sobolev approximation for all lines) or a more accurate model of the accretion disk (which may result in an unsteady picture of the accretion disk rather than the standard Shakura-Sunyaev disk) may lead to physical instabilities. In our systematic approach to the accretion disk wind problem, we started with an analytic 1D model, which although being our ``crudest'' one, did allow for an analytic solution which has proven to be invaluable as a test case for our numerical algorithms. From this initial model we have included, on a step by step basis (Paper~1, Pereyra~1997, Paper~2) elements such as a more accurate treatment of the line radiation force (in which we have always applied the Sobolev approximation), radial structure of the accretion disk, adiabatic heating and cooling processes, etc. This approach has allowed us to establish clearly a relationship between the elements we have introduced (both of numerical and physical nature) and our results, as our models have developed into more realistic ones. In this sense we are confident that if intrinsic physical instabilities should appear as we continue our work, we will be able to establish their physical origin.  From our models we calculate wind mass-loss rates and terminal velocities. For typical cataclysmic variable parameter values of disk luminosity $L_{disk}=L_{\\sun}$, white dwarf mass $M_{wd}=0.6M_{\\sun}$, and white dwarf radii $R_{wd}=0.01R_{\\sun}$, we obtain a wind mass-loss rate of $\\dot M_{wind}=4 \\times 10^{-12} M_{\\sun} \\, {\\rm yr}^{-1}$, and a terminal velocity of $\\sim 3000 {\\rm \\, km \\, s}^{-1}$. We note that the mass loss rate is a approximately a factor of two below the values we found in our earlier models in Paper~2. This difference is due to the detailed calculations of the line radiation force parameters, which when the local density values at the inner region are considered, we find slightly lower values of the $k$ line radiation force parameter.  The results in this work suggest that the details of the ionization equilibrium in CV disk winds, within reasonable parameters, may not have considerable effect over the velocity fields of the wind nor in the wind mass distribution. Although these details have to be taken into account in order to derive an accurate theoretical value of the wind mass loss rate as illustrated by the difference by a factor of two between the models presented here and the earlier ones of Paper~2. The similar wind mass distribution and velocity fields found here with respect to those of Paper~2, would indicate a similar inclination angle dependence of the \\ion{C}{4} 1550 \\AA \\  line. But the decrease in total wind mass loss rate would generate weaker line features. In QSOs, where the ionization balance of the gas material may change significantly, the coupling between the ion populations and the radiative hydrodynamics will have a important effects in the overall wind dynamics; thus a model capable of representing this coupling becomes necessary.  A shortcoming of our models is that, at each spatial point for each time step we assume direction-independent values for the line radiation parameters and we also assume position independent values for the radiative heating and cooling parameters. In future work we plan to include these effects. We also plan to extend our disk wind models to low mass X-ray binaries and QSOs/AGN, where the local ionization equilibrium will play a more crucial role in the overall dynamics."
    },
    "0209/astro-ph0209543_arXiv.txt": {
        "abstract": "This paper is the fourth in a series whose purpose is to study the interstellar abundances of sulfur, chlorine, and argon in the Galaxy using a sample of 86 planetary nebulae. Here we present new high-quality spectrophotometric observations of 20 Galactic planetary nebulae with spectral coverage from 3700-9600~{\\AA}. A major feature of our observations throughout the entire study has been the inclusion of the near-infrared lines of [S~III] $\\lambda\\lambda$9069,9532, which allows us to calculate accurate S$^{+2}$ abundances and to either improve upon or convincingly confirm results of earlier sulfur abundance studies. For each of the 20 objects here we calculate ratios of S/O, Cl/O, and Ar/O and find average values of S/O=1.1E-2$\\pm$1.1E-2, Cl/O=4.2E-4$\\pm$5.3E-4, and Ar/O=5.7E-3$\\pm$4.3E-3. For six objects we are able to compare abundances of S$^{+3}$ calculated directly from available [S~IV] 10.5$\\mu$ measurements with those inferred indirectly from the values of the ionization correction factors for sulfur. In the final paper of the series, we will compile results from all 86 objects, search for and evaluate trends, and use chemical evolution models to interpret our results. ",
        "introduction": "This is the fourth paper in a project to study abundances in planetary nebulae (PNe), highlighting the elements sulfur, chlorine, and argon. The motivation to focus on these three elements has been the realization that their abundances are unaffected by nucleosynthesis in the progenitors of PNe. In contrast to elements like carbon, nitrogen, and oxygen, whose current nebular abundances can be profoundly different from the those in the original star, the sulfur, chlorine and argon abundances we measure now in a nebula are representative of the original composition of the progenitor star. In this way, PNe can provide information on both current and past chemical composition of the interstellar medium in the Galaxy, and can be used to evaluate theoretical yield predictions from stellar evolution models.  The three preceding papers in this series (Kwitter \\& Henry 2001 [Paper~I], Milingo et al. 2001 [Paper~IIA], and Milingo, Henry, \\& Kwitter 2001 [Paper~IIB]) have reported spectrophotometry and abundance analyses of a total of 56 primarily type~II (disk) PNe in our Galaxy. In the present paper we present new spectrophotometry and abundances for 20 more objects. In the final paper, we will also include results from 10 PNe originally observed for another project, bringing the total in our sample to 86.  In \\S 2 we describe the observations and data, and in \\S 3 we present our results; a summary is given in \\S 4. ",
        "conclusions": "In this, the fourth paper in a series investigating the abundances of S, Cl, and Ar in planetary nebulae and the Galactic interstellar medium, we report on spectrophotometric observations of 20 Galactic PNe, where our spectral coverage extended from 3700-9600~{\\AA}. We also calculate electron temperatures and densities, as well as ion and element abundances for each object. We find average values of S/O=1.1E-2$\\pm$1.1E-2, Cl/O=4.2E-4$\\pm$5.3E-4, and Ar/O=5.7E-3$\\pm$4.3E-3 for our sample. These numbers agree very well with our results in the previous papers in the series and with determinations made by others. One of the major features of our current work is to use the NIR lines of [S~III] $\\lambda\\lambda$9069,9532 along with [S~III] temperatures to calculate sulfur abundances. This is the first time these lines have been used to determine sulfur abundances in such a large PN sample.  In the next (final) paper we will compile results from all papers in this series and search for trends among the data. In addition, we plan to apply chemical evolution models to our data in an attempt to evaluate the quality of published stellar yields for O, S, Cl, and Ar, as well as to assess the role of Type~Ia supernovae in the cosmic buildup of these four elements."
    },
    "0209/astro-ph0209296_arXiv.txt": {
        "abstract": "QU~Car is listed in cataclysmic variable star catalogues as a nova-like variable.  This little-studied, yet bright interacting binary is re-appraised here in the light of new high-quality ultraviolet (UV) interstellar line data obtained with STIS on board the Hubble Space Telescope.  The detection of a component of interstellar absorption at a mean LSR velocity of $-$14 km s$^{-1}$ indicates that the distance to QU~Car may be $\\sim$2~kpc or more -- a considerable increase on the previous lower-limiting distance of 500~pc.  If so, the bolometric luminosity of QU~Car could exceed $10^{37}$ ergs~s$^{-1}$.  This would place this binary in the luminosity domain occupied by known compact-binary supersoft X-ray sources.  Even at a 500~pc, QU~Car appears to be the most luminous nova-like variable known.  New intermediate dispersion optical spectroscopy of QU~Car spanning 3800--7000~\\AA\\ is presented.   These data yield the discovery that C{\\sc iv}~$\\lambda\\lambda$5801,12 is present as an unusually prominent emission line in an otherwise low-contrast line spectrum.  Using measurements of this and other lines in a recombination line analysis, it is shown that the C/He abundance as proxied by the n(C$^{4+}$)/n(He$^{2+}$) ratio may be as high as 0.06 (an order of magnitude higher than the solar ratio). Furthermore, the C/O abundance ratio is estimated to be greater than 1. These findings suggest that the companion in QU Car is a carbon star.  If so, it would be the first example of a carbon star in such a binary. An early-type R star best matches the required abundance pattern and could escape detection at optical wavelengths provided the distance to QU~Car is $\\sim$2~kpc or more. ",
        "introduction": "QU~Car was first described in the astronomical literature by Schild (1969), after it had been noted as a potentially-interesting emission line object by Stephenson, Sanduleak \\& Schild (1968).  According to Schild, this variable resembled an old nova spectroscopically and exhibited irregular short-term brightness fluctuations not exceeding 0.2 amplitudes in amplitude. Over a decade later Gilliland \\& Phillips (1982, herafter GP82) presented high time resolution blue spectra of QU~Car that they were able to analyse for radial velocity variation.  They determined a period of 10.9 hours from centroid motion in He{\\sc ii}~$\\lambda$4686 and H$\\beta$ line emission, thereby confirming the binary nature of the system.  They also argued from the absence of detectable secondary star absorption lines and the implied apparent magnitude limit of the secondary star that QU Car was at a distance of at least 500~pc (see also Duerbeck 1999).  The designation of QU~Car as a nova-like variable in catalogues of cataclysmic variables (CV) originates with GP82.  Warner (1995, Table 4.1) further qualifies this by placing it in the UX~UMa sub-class.  The physical model for UX~UMa systems is that these are white-dwarf interacting binaries in which the mass transfer rate is high enough ($3\\times10^{-9} < \\mdot < 1\\times10^{-8}$ M$_{\\odot}$~yr$^{-1}$) to sustain the accretion disk in an opaque high state.  The blue spectrum of QU~Car is quite remarkable in two respects. First, the contrast of all line features against the continuum is very low. The highest contrast emission lines peak at under 1.2 times the continuum level.  Secondly, and uniquely among known CV, emission in the carbon/nitrogen blend centred near 4650~\\AA\\ is comparable in equivalent width to that of the He{\\sc ii}~$\\lambda$4686 emission.  Impressed by this unusual property, GP82 deconvolved the 4650~\\AA\\ emission and showed that it is dominated by carbon emission.  This was an important finding because it consigned fluorescent excitation via the Bowen mechanism to a minor role, and pointed instead toward significant CN abundance enhancement as the explanation for the feature's great relative strength.  \\begin{figure*} \\begin{picture}(0,270) \\put(0,0){\\special{psfile=qucar_fig1.ps angle =0 hoffset=0 voffset=0 hscale=35 vscale=35}} \\end{picture} \\caption{The HST/STIS ultraviolet spectrum of QU Car.  The data shown are the grand sum of all the echelle data obtained to date (Hartley et al. 2002).} \\label{fig:hst} \\end{figure*}  QU Car was observed a number of times at UV wavelengths with the International Ultraviolet Explorer (IUE).  In particular, a series of observations was obtained with a view to searching for orbital-phase linked variability in the stronger UV resonance lines (Knigge et al 1994), a phenomenon often apparent in high-state CV.  Nothing conclusive was found, but it was noted that O{\\sc v}~$\\lambda$1371 absorption was unusually prominent in the spectrum.  As a relatively bright source at a visual magnitude of between 11.1 and 11.5 (see Table 4.1 in Warner 1995), it was natural to include QU~Car in a recent Hubble Space Telescope (HST) programme of high time- and spectral-resolution UV spectroscopy aimed at testing a model for the accretion disk winds seen in nova-like variables (and in QU~Car -- see Knigge et al 1994).  With the advantage of both the high spectral resolution and high signal delivered by the E140M echelle of the Space Telescope Imaging Spectrograph (STIS) on this target, it nevertheless became all too plain that the UV spectrum of QU~Car deviates from the nova-like variable norm.  The main peculiarities are the higher than usual ionization signalled by bright He{\\sc ii}~$\\lambda$1640 emission and the (already noted) strong O{\\sc v}~$\\lambda$1371 absorption, together with the weak mass loss signatures in e.g. N{\\sc v}~$\\lambda$1240 and C{\\sc iv}~$\\lambda$1549. An in-depth discussion of the HST/STIS E140M data is to be found in Hartley, Drew \\& Long (2002).  The aim of this work on QU~Car is to re-open the thus-far sparse discussion on the nature of this binary.  Our conclusion will be that its bland classification as a nova-like variable has served to conceal an extraordinary binary: new evidence, derived both from the HST/STIS observations and from the first broad-band optical spectrum of QU~Car to be obtained in the digital era, suggests instead that this object is unusually luminous (with a mass transfer rate perhaps as high as $\\gtrsim 10^{-7}$ M$_{\\odot}$ yr$^{-1}$) and that the accreting material is also significantly carbon-enriched.  In Section 2, we present the interstellar line spectrum as it emerges from the co-add of all the HST/STIS E140M data available (shown as Figure 1).  This is used to argue that the distance to QU~Car could be $\\sim$2~kpc or more.  In Section 3 we estimate the bolometric luminosity both for the near distance of 500~pc and the representative longer one of 2~kpc.  Then in Section 4 we compare and contrast the optical spectrum of QU~Car with those of a selection of supersoft sources presented by Cowley et al (1998).  Finally, we begin the task of analysing the optical emission line spectrum for element abundances and find, intriguingly, that carbon is significantly enhanced (Section 5).  We close with a discussion in which we draw attention to the puzzle the apparent abundance peculiarity poses for the stellar content of this short-period binary. ",
        "conclusions": "In this study, two significant findings have emerged.  First, UV interstellar absorption line data have revealed evidence of a blueshifted diffuse cloud components at an LSR velocity of $-14$ km s$^{-1}$ (to within 2 km~s$^{-1}$).  Interpreted in terms of a standard galactic rotation model, this tells us that the minimum distance to QU~Car is about 1.8~kpc, rather than $\\sim$0.5~kpc as estimated by GP82.  Nevertheless the fact that the H{\\sc i} interstellar column towards QU Car is about a third the total for the Galaxy along the same siteline, reassures that it is located in the Galaxy.  In fact, if the distance to QU~Car actually is $\\sim 2$~kpc, it would be at much the same distance from the Sun as the Carina spiral arm at the same galactic longitude (Grabelsky et al 1987).  Second, a newly-obtained intermediate dispersion optical spectrum of QU~Car has been found to contain no O~{\\sc vi} line emission and, yet, unusually strong C~{\\sc iv}~$\\lambda\\lambda$5801,12 emission is present.  It has long been known that the $\\sim$4650~\\AA\\ emission feature is also very strong and appears to be dominated by carbon, rather than nitrogen, line emission (GP82).  Starting from the strength of the C~{\\sc iv}~$\\lambda\\lambda$5801,12 emission compared to He~{\\sc ii}~$\\lambda$5411 emission, we have shown that there is very likely significant enrichment of carbon in QU~Car's emission line region (with [C/H] in the range 4 to 40).  Furthermore, the near absence of any oxygen line emission from the UV and optical spectrum suggests the C/O abundance ratio exceeds unity.  We now consider the implications for the status of QU~Car if its distance from the Sun is $\\sim$2~kpc or more.  It was shown in section 3 that this demands a high luminosity, $L_{{\\rm bol}} \\sim 10^{37}$ ergs s$^{-1}$, and hence a high mass accretion rate, $\\mdot_a \\sim 10^{-7}$ M$_{\\odot}$ yr$^{-1}$ (assuming a white dwarf accretor).  Another way of visualising this is to scale QU~Car with respect to the most famous and brightest of dwarf novae, SS~Cyg.  This is a useful comparison also because SS~Cyg is a low-inclination binary, just as we believe QU~Car to be (see Hartley et al 2002).  Parallax observations have recently led to an upward revision of the distance to SS~Cyg placing it at a distance of 166~pc (Harrison et al 1999).  Schreiber \\& G\\\"ansicke (2002) have re-evaluated the outburst mass accretion rate, using the revised distance: at $V = 8.5$, they associate with it the distinctly high mass transfer rate of $(5.7 \\pm 2.4) \\times 10^{-8}$ M$_{\\odot}$~yr$^{-1}$ (for the widely accepted white dwarf mass of 1.19~M$_{\\odot}$).  After correcting for reddening, the apparent magnitudes of QU~Car and SS~Cyg imply that the former is 12 times fainter than the latter.  This means QU~Car is intrinsically 2.6 $\\times$ D$^2$ brighter in the $V$ band, where $D$ is the distance to QU~Car expressed in kpc.  Hence at $\\sim$2~kpc, QU Car is 10 times more luminous than SS~Cyg in the $V$ band, and possibly powered by a mass transfer rate as high as $\\sim 6 \\times 10^{-7}$ M$_{\\odot}$~yr$^{-1}$ (scaling from Schreiber \\& G\\\"ansicke's calculation).  The clear empirical restraint on relabeling QU~Car as a supersoft source is that there is only an upper limit on the soft X-ray flux (Verbunt et al 1997). A low X-ray luminosity is also consistent with the absence of O~{\\sc vi} recombination lines in the spectra we have obtained.  Nevertheless, given that `X-ray off' states have been recorded for supersoft sources (e.g. RX~J0513.9$-$6951, Reinsch et al 1996; CAL~83, Greiner \\& DiStefano 2002), this difference might diminish.  A wavelength domain in which QU~Car bears some resemblance to known supersofts is in the ultraviolet.  G\\\"ansicke et al (1998) have described the UV spectra of RX~J0513.9$-$6951 and CAL 83, and point out the presence of N~{\\sc v}~$\\lambda$1240, O~{\\sc v}~$\\lambda$1371 and He~{\\sc ii}~$\\lambda$1640 emission in both objects.  These are also prominent in QU~Car, with the difference that the N~{\\sc v} and O~{\\sc v} features are in absorption -- this might also signal less extreme ionization than in the supersoft sources.  It is unlikely that orbital inclination is implicated here given that all three objects are believed to be at low inclination (Hartley et al 2002; Cowley et al 1998).  Finally we note that for a distance of 2~kpc the absolute magnitude, $M_V = -0.4$, of QU~Car is fainter than that of these same two supersofts by 1.6 and 0.9 mags.  It remains possible that the interstellar line data mislead in indicating the longer siteline to QU~Car, leaving the earlier-proposed minimum distance of 0.5~kpc intact.  At this nearer limiting distance, QU~Car is still the most luminous nova-like variable known.  Indeed using the scaling to SS~Cyg again, at $D = 0.5$~kpc, the mass transfer rate would be expected to be in the region of $4\\times 10^{-8}$ M$_{\\odot}$~yr$^{-1}$ (see also Table 2, section 3).  Hitherto, on the rare occasions that deviations from cosmic abundances in CV have been measured and documented, they have been in the sense of enhancements due to CNO-cycle burning (i.e. mainly nitrogen enhancement, resulting from conversion of carbon and perhaps oxygen: see Long 2000, Marsh et al 1995).  QU~Car appears to be the first recognised instance of enhancement due to helium burning. This would not be remarkable if the line emission concerned could be associated with the accreting white dwarf.  We are reluctant to propose this since the character of QU~Car's UV emission line spectrum points toward an accretion disk origin (Hartley et al 2002).  The carbon enhancement has then to be located in the envelope of the companion star. Could the companion star be a CO white dwarf?  There can be two objections to this: (i) the published orbital period of 10.9 hours (GP82) is too long, (ii) a double-degenerate CV would not be expected to be so bright (even for the shorter minimum distance of 500~pc).  To counter the first objection it can be proposed that the time series sampling of GP82 was too coarse to pick up the short ($\\sim$~1000 sec) periods expected for double-degenerate systems.  Certainly, a re-examination of QU~Car's binary parameters is warranted, given the extraordinary character of this binary and the ragged character of the phase-folded radial velocity curve (see GP82).  The second objection currently has a simple empirical basis in that the absolute magnitude of the brightest of the known AM CVn systems (AM~CVn itself) is thought to be $M_V \\sim 9$ (Warner 1995; see also Nasser, Solheim \\& Semionoff 1991).  A third problem for this scenario is the undeniable presence of hydrogen line emission and absorption in QU~Car's optical spectrum.  Accepting, for the moment, that the orbital period really is 10.9~hours as determined by GP82, the implication from the period-mean density relation (e.g. Eggleton 1983) is that the companion star should have the mean density of a mid-F main sequence star (or of a star evolving away from a later-type position on the main sequence).  Near main sequence stars cannot exhibit the pronounced carbon enhancement that has been discovered here.  At higher luminosity, the abundance patterns that have been determined for barium and CH giants will not fit either, given that their carbon enrichment is against a backdrop of overall reduced metallicity, with the result that C/H or C/He is typically still less than in the Sun (see Barbuy et al 1992, Vanture 1992).  Nevertheless there is one class of not-so-bright giant that does come close to matching the abundance pattern deduced for QU~Car -- the early-type R stars.  These are carbon stars now known to have absolute magnitudes that place them on the HRD as red clump giants ($M_K \\simeq -2$, with $V-K$ in the range from 2 to 4, Knapp, Pourbaix \\& Jorissen 2001).  In the R-star sample of Dominy (1984), carbon enhancements up to $\\sim$6 times solar are found with little change or a small drop in oxygen abundance.  To accommodate such a companion within QU~Car, the binary needs to be at or beyond the larger minimum distance discussed here.  Specifically, at 2~kpc the absolute visual magnitude of QU~Car is -0.4, to be compared with the same quantity for 5 R stars from Wallerstein and Knapp (1998) ranging from -0.25 up to 4.17. It would seem that the fainter half of this range is consonant with the non-detection of the companion star in QU~Car's optical spectrum.   If further work can substantiate this concept, it could lend support to the growing opinion that the abundance patterns in early-type R stars have to do with outward mixing after the core helium flash (see discussion in Knapp et al 2001).  The alternative idea that the failure to detect binary motion in these carbon-rich stars points to their being coalesced binaries (see e.g. McClure 1997) sits awkwardly with the fact that QU~Car remains apparent as a binary in which the mass transfer is {\\em away} from the putative R star. However work will have to be done to see whether an early-type R star in a high mass transfer rate system is a plausible consequence of binary star evolution.  In this context, the 10.9~hr orbital period may prove too short for comfort -- unless the R star can have already shed much of its envelope.  If the distance to QU~Car is closer to $\\sim$500~pc, we are unable to suggest a plausible identity for the companion star.  A dwarf carbon-star companion probably can be ruled out since the overall abundance pattern in such objects is that of a reduced-metallicity Population II star, like the CH and Ba giants (see Wallerstein \\& Knapp 1998).  It is also doubtful that pollution of the companion star by carbon-rich debris from unrecorded nova explosions in the past is a viable explanation -- if the white dwarf has accreted from an unexceptional companion star, nucleosynthesis in the nova outburst will most likely yield nitrogen-enhanced ejecta with C/O $<$ 1 (see Gehrz et al 1998).   Putting all the arguments together, we conclude that the apparent abundance peculiarity in QU~Car has to be seen as favouring the picture in which this binary is distant and luminous.  We end with the remark that in carrying out this piece of work we have come to appreciate the paucity of published CV optical spectra (particularly outside the blue range).  The only comprehensive optical spectroscopic survey of CVs that we have been able to draw on is the work of Williams (1983). Among the spectra presented there it can be seen that only T~Pyx and V~Sge bear a family resemblance to QU~Car, particularly in terms of the strength of He~{\\sc ii} lines relative to those of H~{\\sc i}.  Furthermore in the case of V~Sge one may, perhaps uniquely, just make out from the plot of its spectrum the C~{\\sc iv}~$\\lambda\\lambda$5801,5812 blend in emission -- certainly Williams did not find this feature often enough to consider quoting measurements of it in tables.  Referring to the study of Herbig et al (1965) on this well-known and perplexing binary, we find a report of data on the C~{\\sc iv} transition giving an equivalent width 1.3 times that of the nearby He{\\sc ii}~$\\lambda$5411 line.  In QU~Car this ratio is close to 4! If V~Sge is bizarre, QU~Car is also.  Objects such as these are testaments to the propensity that nature has to try out all the options -- and in binary star evolution there appears to be especially many."
    },
    "0209/hep-ph0209130_arXiv.txt": {
        "abstract": "High energy neutrinos with energy typically greater than tens of thousands of GeV may originate from several astrophysical sources.  The sources may include, for instance, our galaxy, the active centers of nearby galaxies, as well as possibly the distant sites of gamma ray bursts. I briefly review some aspects of production and propagation as well as prospects for observations of these high energy astrophysical neutrinos. ",
        "introduction": "During\\footnote{Talk given at IAU 8th Asian Pacific Regional Meeting, 2$-$5 July, 2002, Tokyo, Japan.} nearly past 50 years, the empirical search for neutrinos has spanned roughly six orders of magnitude in energy, from approximately $10^{-3}$ GeV up to approximately $10^{3}$ GeV. The lower energy edge corresponds to solar neutrinos, whereas the upper energy edge corresponds to atmospheric neutrinos. The intermediate energy range include the terrestrial and supernova neutrinos. This search has already given us remarkable insight into neutrino interaction properties as well as its intrinsic properties such as mixing and mass. Here, I briefly review the possibility of having neutrinos with energy greater than $10^{3}$ GeV. The upper energy edge for high energy astrophysical neutrinos is limited only by the concerned experiments. A main motivation to search for such high energy astrophysical neutrinos is to get more accurate information about the origin of observed high energy photons (and ultra high energy cosmic rays) that is presently not possible through conventional gamma ray astronomy. For instance, the observation of high energy gamma ray flux alone from active centers of nearby galaxies (AGNs) such as M 87 and distant sites of gamma ray bursts (GRBs) does not allow us to identify its origin in purely electromagnetic or purely hadronic interactions unambiguously. Sizable high energy astrophysical neutrino flux is expected if latter interactions are to play a dominant role. Search for high energy astrophysical neutrinos will thus provide us a complementary and yet unexplored view about some of the highest energy phenomenons occurring in the known universe. For a general introduction of the subject of high energy astrophysical neutrinos, see Bahcall \\& Halzen (1996), Protheroe (1999), Bahcall (2001). See, also Battiston (2002). ",
        "conclusions": "Several astrophysical sources such  as center(s) of our as well as other nearby galaxies may produce high energy astrophysical neutrino flux with energy around or above $10^{4}$ GeV, compatible with the observed high energy gamma ray and ultra high energy cosmic ray flux above the atmospheric neutrino flux. This implies need for a more meaningful search.   The author thanks Physics Division of National Center for Theoretical Sciences for financial support."
    },
    "0209/hep-ph0209062_arXiv.txt": {
        "abstract": "We investigate the potential of a future kilometer-scale neutrino telescope such as the proposed IceCube detector in the South Pole, to measure and disentangle the yet unknown components of the cosmic neutrino flux, the prompt atmospheric neutrinos coming from the decay of charmed particles and the extra-galactic neutrinos, in the 10 TeV to 1 EeV energy range. Assuming a power law type spectra, $d\\phi_\\nu/dE_\\nu \\sim \\alpha E_\\nu^\\beta$, we quantify the discriminating power of the IceCube detector and discuss how well we can determine magnitude ($\\alpha$) as well as slope ($\\beta$) of these two components of the high energy neutrino spectrum, taking into account the background coming from the conventional atmospheric neutrinos. ",
        "introduction": "\\label{sec:intro} Large volume neutrino telescopes are being constructed to detect high-energy neutrinos primarily from cosmologically distant sources. A major challenge for these experiments will be separating the contributions  coming from the different sources in the observed flux. In this paper, we consider three different origins for high-energy neutrinos: conventional atmospheric neutrinos coming from the decay of charge pions and kaons, prompt atmospheric neutrinos from the decay of charmed particles and  neutrinos from extra-galactic sources.  Of these sources, only the conventional atmospheric neutrino flux has been observed in the energy range from sub-GeV up to $\\sim$ TeV range~\\cite{atmnuobs}. Currently, the conventional atmospheric neutrino flux is known to about 15-20\\%~\\cite{review-atm}. The other two fluxes, although anticipated by theoretical expectations, are experimentally unknown to us, and their observation will be extremely important.  Up to about $E_\\nu \\sim 100$ TeV the main source of atmospheric neutrinos is the decay of pions and kaons produced in the interactions of cosmic rays in the Earth atmosphere.  At higher energies, these mesons will interact rather than decay, making the semileptonic decay of charmed particles the dominant source of atmospheric neutrinos. This gives rise to the so called prompt atmospheric neutrino flux which is, unfortunately, subject to  large theoretical uncertainties. The uncertainties in the calculation of the prompt neutrino fluxes reflect not only our poor knowledge of the atmospheric showering parameters, which for a given model can cause a change of an order of magnitude in the fluxes, but are mostly related to the model used to describe charm production at high energies, which is responsible for a discrepancy up to two orders of magnitude in the predictions~\\cite{costa}. Typically, the energy dependence of prompt neutrino flux is $d \\phi_\\nu/dE_\\nu \\sim E_\\nu^{-3}$.  If the prompt atmospheric neutrino flux can be determined by experimental data, this can provide a unique opportunity to study heavy quark interactions at energies not accessible by terrestrial accelerators. Furthermore, the characterization of the prompt component of the neutrino flux will enhance the discriminating power of the other components at higher energies.  High energy neutrinos are also expected to be produced in astrophysical sources at cosmological distances. The most conventional source candidates are compact objects such as gamma ray bursts~\\cite{grb} and active galactic nuclei jets, called blazars~\\cite{agn}. In these sources, neutrinos may be generated via pion production in the collision between protons and photons in highly relativistic shocks. A typical energy dependence of the extra-galactic neutrino flux in these scenarios is $d \\phi_\\nu/dE_\\nu \\sim E_\\nu^{-2}$. For other possibile extra-galactic neutrino spectra, see, for example, \\cite{dmitry}.  Other possible sources of extra-galactic neutrinos include neutrinos generated in the annihilation of weakly interacting massive particles~\\cite{wimps}, the propagation of ultra-high energy protons~\\cite{cosmogenic} or in a variety of top-down scenarios including decaying or annihilating superheavy particles with GUT-scale masses~\\cite{shparticles}, decaying topological defects~\\cite{defects}, the so-called Z-burst mechanism~\\cite{z} or Hawking radiation from primordial black holes~\\cite{hawking}. The neutrino fluxes from compact sources, the propagation of ultra-high energy protons or top-down scenarios can be tied to the observed cosmic ray flux. Since a myriad of speculations exist, resolution will likely be reached only by experiment. Currently, only the upper bound on such high energy extra-galactic neutrino flux, $E_\\nu^2d\\phi_\\nu/dE_\\nu \\lsim 10^{-5}$ GeV cm$^{-2}$ s$^{-1}$ sr$^{-1}$, has been obtained~\\cite{amandacascade}. For a review of high-energy neutrino sources and detection, see~\\cite{review}.  Many important questions regarding the origin of cosmic rays can be decided by neutrino observations. The determination of an extra-galactic neutrino flux will be very important for understanding the nature of the sources of the ultra-high energy cosmic rays.  We investigate the possibility of determining the  prompt atmospheric neutrino and the extra-galactic  neutrino energy spectra (slope and magnitude)  using down-going showers~\\cite{cascade,amandacascade} induced by neutrinos in a kilometer-scale neutrino telescope conceived to detect  high-energy neutrinos at high rates, such as IceCube,  particularly in the region $ 10 \\mbox{ TeV } \\lsim E_\\nu \\lsim 1 \\mbox{ EeV}$. We demonstrate that since the energy spectra of these two neutrino fluxes are expected to be rather different, by using shower events from which one can reconstruct the initial neutrino energy with some  accuracy, IceCube will be able to determine their energy spectra separately even if they co-exist.  The organization of this paper is as follows. In Sec.~\\ref{sec:icecube}, we briefly describe the presumed detector setup as well as the type of neutrino events we will consider. In Sec.~\\ref{sec:analysis}, we describe the analysis method and in Sec.~\\ref{sec:results} we present our results. Finally, Sec.~\\ref{sec:conclusions} is devoted to discussions and conclusions. ",
        "conclusions": "\\label{sec:conclusions}  We have investigated the possibility of future neutrino telescopes to separate the various contributions to the observed neutrino flux. We have considered that high-energy neutrinos from three different origins can contribute to the measured flux: conventional atmospheric neutrinos, prompt atmospheric neutrinos from the decay of charmed particles and  neutrinos from extra-galactic sources.  We have restricted our analysis to showers induced by down-going neutrinos, not to have to worry about energy losses in the Earth and be equally sensitive to all neutrino flavors. We have also assumed the neutrino telescope will be able to reconstruct the  parent neutrino energy  from the collected shower energy within a factor of about 2-3.  Assuming the prompt atmospheric and extra-galactic neutrino fluxes can be described by a power law and parametrized by  two parameters $\\alpha$ (the magnitude) and $\\beta$ (the slope), and considering that the conventional atmospheric neutrino flux is currently known with a theoretical uncertainty $\\sigma_{\\text{atm}}=$ 15\\%, our conclusion are the following.  If extra-galactic neutrinos constitute the dominant component of the measured flux, after 10 years of observations, a detector such as IceCube will be able to determine $\\alpha_{\\text{EG}}$ within an order of magnitude and $\\beta_{\\text{EG}}$ to $\\approx$ 10\\%, assuming as input a dominant WB flux. This is independent of the conventional atmospheric neutrino contamination. We have also estimated that the maximal sensitivity of IceCube after 10 years of data taking will be $\\alpha^0_{\\text{EG}} \\approx 6 \\times 10^{-9}$ GeV$^{-1}$ cm$^{-2}$ s$^{-1}$ sr$^{-1}$.  If prompt neutrinos constitute the dominant component of the measured flux, after 10 years, IceCube can determine $\\alpha_{\\text{prompt}}$ and $\\beta_{\\text{prompt}}$ at most within 2 orders of magnitude and about 20\\%,  respectively, with the present value of $\\sigma_{\\text{atm}}=$ 15\\%. This can nevertheless be improved if this uncertainty can be substantially reduced. We also have estimated that in this case, the maximal sensitivity of IceCube will be achieved for $\\alpha^0_{\\text{prompt}}=1.5\\times 10^{-3}$ GeV$^{-1}$ cm$^{-2}$ s$^{-1}$ sr$^{-1}$.  We have also determined in which cases a complete separation of the two components can be performed if both extra-galactic and prompt neutrinos contribute to the observed flux, Fig.~\\ref{fig5} summarizes our conclusions on this. The main point here is that to clearly separate the prompt component from the extra-galactic component $\\sigma_{\\text{atm}}$ must be about 10\\% or less. If $\\sigma_{\\text{atm}}$ is much larger, a single power law will fit the data with an acceptable value of $\\chi_{\\text{min}}^2$.  Finally, let us mention that there is an additional signature that can be used to distinguish extra-galactic neutrinos from the prompt atmospheric ones. As first indicated by atmospheric neutrino data and lately confirmed by the K2K experiment~\\cite{K2K}, $\\nu_\\mu$ oscillate to $\\nu_\\tau$ implying that one third of the total original extra-galactic $\\nu$  flux will arrive at the Earth as $\\nu_\\tau$. On the other hand, prompt neutrinos are expected to have much lower $\\nu_\\tau$ than $\\nu_e$ or $\\nu_\\mu$ content~\\cite{costa}. For $E_\\nu \\gsim 1$ PeV a $\\nu_\\tau$ event can be clearly recognized through the observation a $\\tau$, produced by a $\\nu_\\tau$ charge current interaction, which will decay in the detector. This gives rise to the so-called double-bang (when the $\\tau$ is produced and decays within the detector volume) and lolly pop (when the $\\tau$ is produced outside the detector but decays inside it) events~\\cite{review,pakvasa}. We estimate that after 10 years a detector like IceCube should observe, for the WB flux, a few such events, whereas no event is expected even for the maximal value of the allowed prompt neutrino flux."
    },
    "0209/astro-ph0209557_arXiv.txt": {
        "abstract": "{We present a catalogue of 6206 stars which have proper motions exceeding 0.18$\\arcsec$yr$^{-1}$ with an R-band faint magnitude limit of 19$\\cdot$5mag. This catalogue has been produced using SuperCOSMOS digitized R-Band ESO and UK Schmidt Plates in 131 Schmidt fields covering more than 3,000 square degrees ($>$7$\\cdot$5$\\%$ of the whole sky) at the South Galactic Cap.  The survey is $\\ge$90$\\%$ complete within the nominal limits of the Luyten Two Tenths Catalogue of $m_R$$\\le$18$\\cdot$5mag and 0$\\cdot$2$\\le$$\\mu$$\\le$2$\\cdot$5$\\arcsec$yr$^{-1}$, and is $\\ge$80$\\%$ complete for $m_R$$\\le$19$\\cdot$5mag and $\\mu\\le$2$\\cdot$5$\\arcsec$yr$^{-1}$. ",
        "introduction": "\\subsection{Proper Motion Surveys}  Proper motion surveys can give us a good insight into stellar populations and dynamics in the Solar neighbourhood.  They identify nearby objects, allow analysis of their space motions and, with data in more than one passband, give population differentiation.  These three factors mean that proper motion surveys are ideal in addressing the problems of the local stellar mass function and the halo mass function, both of which are currently poorly constrained (e.g. Henry et al., 1997, Lee 1993).  The problem with the local stellar mass function is in large part due to the undersampling of nearby stars upon which it is mainly based.  Identification of objects such as the nearby brown dwarf Kelu-1 (Ruiz et al. 1997), which was discovered as part of the Calan-ESO proper motion survey (Ruiz et al. 2001) and has an astrometrically estimated distance of $\\sim$10pc, are vital in this respect. Both present day and initial halo mass functions are poorly constrained (Lee 1993) due to a number of factors.  These include the relatively low number of nearby halo subdwarf stars and the difficulty in obtaining accurate distances for them.  However,  halo stars have greater space velocity and velocity dispersion than disc stars which means that a halo star will have a higher proper motion than a disc star at the same distance.  When combined with their subluminosity this provides an excellent method of differentiating them from disc stars (seeSect. 5$\\cdot$1).  However, all of the existing surveys suffer from one or more of the following problems; \\begin{itemize} \\item internal inhomogeneities which make statistical analysis difficult and unreliable, \\item low areal coverage which makes a survey subject to low number statistics, \\item use of non-contiguous survey fields which prevents cross-checking of discovered objects and completeness levels from areas where contiguous fields overlap. \\end{itemize}  One of the main purposes of this survey is to address these problems by conducting an internally homogeneous large area survey in contiguous fields by means of scripted computer algorithms.  By using only the R-band plates we also ensured that the survey would have a high completeness relative to other surveys which require an object to be present in more than one passband, thus missing faint objects with extreme colours, such as brown dwarfs and ancient halo subdwarfs and white dwarfs.  This is important because these objects are among the most astrophysically interesting to be selected by proper motion surveys.  Using only the R-band plates also means that our completeness is easier to assess.   \\subsection{The Luyten Catalogues}  The largest catalogue of high proper motion stars to date is the New Luyten Catalogue of Stars with Proper Motions Greater Than Two Tenths of an Arcsecond (NLTT) (Luyten, 1979).  The NLTT and the Luyten Half Second Proper Motion Catalogue (LHS) (Luyten, 1979) were compiled over a period of $\\sim$60 years from two main surveys; the Bruce Proper Motion Survey (BPM), covering an area from the South Celestial Pole to a declination of $\\approx$+25$^{\\circ}$, and the Palomar Observatory Sky Surveys (POSS I and II) covering the area from the North Celestial Pole to $\\approx$--33$^{\\circ}$.  The nominal faint limit of the Bruce survey is R$\\simeq$16$\\cdot$5mag and the faint limit of the POSS is R$\\simeq$20mag (Morgan et al., 1992), the Luyten Catalogues have no stated bright limit.  The Luyten proper motion measurements were made using two different techniques; (i) manual plate blinking and (ii) automatic machine detection (Williams, 2000).  The Luyten catalogues therefore contain some internal inhomogeneities such as a brighter limiting magnitude for fields south of --33$^{\\circ}$ and variations in the completeness of the catalogues between different areas of the surveys. The stated proper motion limits of the NLTT are 0$\\cdot$18$\\arcsec$yr$^{-1}$$\\le\\mu\\le$2$\\cdot$5$\\arcsec$yr$^{-1}$ whilst the LHS comprises stars from the NLTT with $\\mu\\ge$0$\\cdot$5$\\arcsec$yr$^{-1}$.  \\subsection{Other Surveys and the Completeness of the Luyten Catalogues}  There have been a number of estimates of the completeness of the NLTT and LHS catalogues in recent years, mostly based on the results of small area proper motion surveys.  Scholz et al. (2000) conducted a survey in 40 fields using the Automatic Plate Measuring microdensitometer (APM) at Cambridge, UK.  They discuss the completeness of previous catalogues, as well as how they find far more stars in the relatively incomplete southern area of the NLTT, but beyond stating that the NLTT is incomplete they do not quantify their own completeness or, using their data, the completeness of the NLTT.  Evans (1992) compared the results from four fields of the APM proper motion project to the results for the same areas from the NLTT.  He discovered a number of NLTT stars which, when measured with the APM microdensitometer, proved to have proper motions below the 0$\\cdot$18$\\arcsec$yr$^{-1}$ cutoff limit.  He also found a larger number of stars which had proper motions above this limit which were not listed in the NLTT.  By conducting a Monte Carlo simulation he showed that these discrepancies could be almost entirely attributed to random measuring errors.  He further stated that these errors led to a contamination of between 10$\\%$ and 20$\\%$ and an incompleteness of $\\sim$16$\\%$ (ie a completeness of $\\sim$84$\\%$).  However, whilst the comparison was meticulously carried out and  extremely detailed it relied on comparison of a very small number of fields.  It is therefore possible that it suffers from selection effects and small number statistics.  Ruiz et al. (2001) compared their results from the Calan-ESO proper motion catalogue to the Luyten catalogues and concluded that for $\\mu$$\\leq$0.8$\\arcsec$yr$^{-1}$ and R$\\geq$13mag the LHS catalogue is $\\sim$60$\\%$ complete.  They also state that the Luyten Two Tenths catalogue (LTT) is $\\sim$40$\\%$ complete for R$\\geq$13mag, but this refers to the 1975 version of the Two Tenths catalogue, which was superseded by the more complete NLTT in 1979.  The Ruiz et al. comparison suffers from a similar problem to that of Evans in that it used the relatively small number of 14 fields.  In addition to this their survey was conducted by manual blink comparison which is always prone to human error, as highlighted by the larger incompleteness of the southern BPM section of the NLTT over the northern POSS section.  It was also conducted at low Galactic latitude $\\mid$b$\\mid$$\\le$40$^{\\circ}$ because fields at high Galactic latitude do not contain enough background stars for efficient blink comparison, which could be an important contributory factor in the incompleteness of the BPM section of the NLTT.  Dawson (1986) applied the V/V$_{\\rm{max}}$ method as described by Schmidt (1968, 1975) to estimate a completeness of $\\sim$90$\\%$ for $\\delta$$\\ge$-33$^{\\circ}$, $\\mid$b$\\mid$$>$10$^{\\circ}$, R$\\le$18mag and $\\mu$$\\ge$0$\\cdot$5$\\arcsec$yr$^{-1}$ (i.e. the northern areas of the LHS).  Flynn et al. (2001) applied a statistical test to the NLTT consisting of taking two concentric spheres centred around the Sun whose volumes are in the ratio 2:1 (thereby giving them a radial ratio of 1$\\cdot$259 which corresponds to an average magnitude difference of 0$\\cdot$5mag between the shells) and assuming that the completeness for stars in the inner sphere is 100$\\%$.  It is then possible to calculate the completeness of the stars in the outer `shell', and by applying this to different magnitude bins they derive a plot of completeness for the LHS and NLTT catalogues relative to the completeness at R=13 (assumed for the purposes of the plot to be 100$\\%$ complete).  This analysis shows the completeness of the NLTT dropping steadily from 80$\\%$ at R=14 to 60$\\%$ at R=18$\\cdot$5mag.  However, Monet et al. (2000) suggest that this analysis is flawed since it does not take into account the different space densities of stars at high and low Galactic Latitude. Their recomputation of the Flynn et al. test gives a completeness at low Galactic latitude (15$\\le$$\\mid$b$\\mid$$\\le$35) of between 75$\\%$ and 85$\\%$ for 14$\\le$R$\\le$18mag.  Monet et al. (2000) conducted a survey in 35 fields using POSS II plates and estimated that the LHS is $\\sim$90$\\%$ complete, as well as recalculating the result of Flynn et al. (discussed above). ",
        "conclusions": "Over the past few years there have been several papers which offer a variety of values for the completeness of the Luyten catalogues, and discuss its shortcomings (see Sect. 1$\\cdot$2).  The values offered for the completeness of the Luyten catalogues vary from 60$\\%$ for m$_R$$>$13 (Ruiz et al., 2001,comparison to Calan-ESO survey conducted at low galactic latitude) to 90$\\%$ for m$_R$$\\le$18 and $\\mu$$\\ge$0$\\cdot$5$\\arcsec$yr$^{-1}$ (Dawson 1986, V/V$\\rm{_{max}}$ method applied to northern areas of LHS)  Our completeness levels are the first to be calculated from an analysis of the search technique itself, and show that for the vast majority of stars in the catalogue (R$\\le$19mag and 0$\\cdot$2$\\le$$\\mu$$\\le$2$\\cdot$5$\\arcsec$yr$^{-1}$) the completeness is greater than 90$\\%$. Fig. 6 therefore indicates that for the Northern areas ($\\delta$$>$-33$^{\\circ}$, $\\mid$b$\\mid$$>$10$^{\\circ}$) the NLTT is $\\sim$85$\\%$ complete.  Fig. 7 shows the incompleteness of the Southern part of the NLTT for R$>$13mag ($\\sim$25$\\%$ complete for 15$\\le$R$\\le$16$\\cdot$5mag and $\\sim$15$\\%$ for 16$\\cdot$5$\\le$R$\\le$18mag).  We used the H$_B$, B-I and the B--R, R--I plots to select halo subdwarfs.  This selection yielded 1189 halo subdwarfs from a total of 6104 stars which have both B and I band magnitudes ($\\sim$19.5$\\%$). Table 6 of Reid (1984) gives four model predictions of the relative numbers of disc, halo, intermediate (thick disc) and white dwarf populations which would be discovered by a polar proper-motion survey. The models are based on different combinations of the disc and halo luminosity functions and the kinematics of the halo.  Our result gives a value which falls between the predictions of models A and D and excludes models B amd C.  Model A is based largely on the results of Wielen (1974), whilst model D uses the Luyten (1968) disc luminosity function, a `mean globular' halo luminosity function and the halo kinematics of Oort (1965).  This analysis is described in more detail in Pokorny, Jones \\& Hambly, 2002 (in preparation)."
    },
    "0209/astro-ph0209627_arXiv.txt": {
        "abstract": "We discuss the evolution of oxygen, carbon and nitrogen in galaxies of different morphological type by adopting detailed chemical evolution models with different star formation histories (continuous star formation or starbursts). In all the models detailed nucleosynthesis prescriptions from supernovae of all types and low- and intermediate-mass stars are taken into account. We start by computing chemical evolution models for the Milky Way with different stellar nucleosynthesis prescriptions. Then, a comparison between model results and ``key'' observational constraints allows us to choose the best set of stellar yields. Once the best set of yields is identified for the Milky Way, we apply the same nucleosynthesis prescriptions to other spirals (in particular M101) and dwarf irregular galaxies. We compare our model predictions with the [C,N,O/Fe] vs. [Fe/H], log(C/O) vs. 12+ log(O/H), log(N/O) vs. 12+ log(O/H) and [C/O] vs. [Fe/H] relations observed in the solar",
        "introduction": "The variation of element abundance ratios with metallicity can be used as a powerful tool for understanding the chemical enrichment of galaxies and hence, their evolution. Given the fact that different chemical elements are produced on different timescales by stars of different lifetimes, the abundance ratios of some key elements versus metallicity will depend not only on the stellar evolution processes but also on the star formation history (SFH) of a galaxy (e.g., Pagel 1997). The SFH in turn depends on important processes taking place during the galaxy evolution like, for instance, inflow and outflow of gas.  Because of their long main-sequence lifetimes and lack of deep convective zones, the lower mass stars which are still observable today, have preserved the patterns of elemental abundances generated by the initial burst of star formation. In principle, by analyzing the chemical abundances of stars of various ages, lying at different positions inside a galaxy, it is possible to constrain its star formation and enrichment histories. However, since high quality spectra of individual stars are required to determine abundances, studies of this kind are possible only in a few cases.  The CNO elements, and in particular the C/O and N/O abundance ratios, can be considered ``key'' tools for the study of chemical evolution of galaxies, as these are elements produced by different mechanisms and in different stellar mass ranges. Oxygen is almost entirely produced by massive stars and ejected into the interstellar medium via the explosion of type II SNe. The situation for C and N is more complex as these elements can be produced by stars of all masses. N is mostly a secondary element being a product of the CNO cycle and formed at expenses of the C and O already present in the star, although a primary N component originating in asymptotic giant branch (AGB) stars is also predicted by stellar nucleosynthesis studies. In fact, primary nitrogen can be produced during the third dredge-up, occurring along the AGB phase, if nuclear burning at the base of the convective envelope is efficient (hot-bottom burning, HBB, Renzini \\& Voli 1981 - hereinafter RV). Some primary N can also be produced in massive stars due to stellar rotation, according to the recent calculations of Meynet \\& Maeder (2002b). Carbon is a primary element produced during the quiescent He-burning in massive and intermediate-mass stars. However, there are many uncertainties still involved in the carbon yields.  The abundance ratio of a secondary to a primary element is predicted to increase with the abundance of its seed. This is the case for the N/O ratio as N is mostly a secondary element, especially in massive stars. On the other hand, the abundance ratio of two primary elements such as $^{12}$C and $^{16}$O, which are restored into the ISM by stars in different mass ranges, shows almost the same behavior with metallicity as the ratio between a secondary and a primary element: the C/O ratio in fact increases with metallicity. The most straightforward interpretation is that $^{12}$C is mainly restored into the ISM by intermediate-mass stars (and hence on longer timescales compared to the $^{16}$O enrichment which comes mainly from massive stars), so that the C/O ratio increases as the ISM enrichment proceeds. In other words, the production of a primary element on long timescales mimics the secondary nature. This interpretation has been recently challenged by several papers (eg. Henry et al. 2000, Carigi 2000 - see below).  The best place for studying the evolution of the CNO elements is the Milky Way (MW), where it is possible to measure the abundances of stars of different ages and hence infer the chemical composition of the ISM at various epochs. In particular, the abundances of the CNO elements in metal-poor stars provide an important constraint on the early chemical evolution of the Milky Way. The abundances of the very-metal-poor stars belonging to the halo trace the composition of the ISM at the time of their formation, i.e., many billion years ago, when most of the enrichment was due to massive stars. B stars and HII regions trace instead the ISM chemical composition in the Galactic disk at the present time. In the past years a great deal of observational work has been devoted to measure stellar abundances in the Milky Way, both in halo and disk stars (for a large compilation of the available abundance data until 1999 see Chiappini et al. 1999). More recently, careful and detailed abundance analyses have been carried out by several groups (e.g., Carretta et al. 2000; Fuhrmann 1998; Mel\\'endez et al. 2001, Mel\\'endez \\& Barbuy 2002, Depagne et al. 2002, Nissen et al. 2002). In particular, a few high quality data are now available for C and O at low metallicities (Carretta et al. 2000).  For the Milky Way, the data reveal that oxygen (as well as other $\\alpha$-elements) shows an overabundance relative to Fe in metal-poor stars ([Fe/H]$<$ $-$1.0), whereas these ratios decrease for disk stars until they reach the solar value. This trend is generally interpreted as due to the time-delay in the iron production which originates from long living white dwarfs in binary systems eventually exploding as type Ia SNe. The available data for Mg, Si, Ca and S show that for [Fe/H] $<$ $-$1.0 the [$\\alpha$/Fe] ratio is almost constant, defining a plateau, whereas for oxygen there seems to be a slight increase of the [O/Fe] ratio with decreasing metallicity (see Chiappini et al. 2001 for a discussion on this particular point - hereafter CMR2001). On the other hand, recent papers (Israelian et al. 1998, 2001; Boesgaard et al. 1999), show an even steeper increase of the [O/Fe] ratio with decreasing [Fe/H]. This trend, if real, is difficult to reconcile with our knowledge about stellar nucleosynthesis. In the present work we discuss carefully the abundance data concerning oxygen and show that when only the most reliable measurements and abundance analysis (Asplund \\& Garcia P\\'erez 2001 - see Sect. 4.1.2) are considered, a good agreement with our theoretical predictions is found.  As far as C and N are concerned, there are still many open questions. It has been shown that chemical evolution models for the MW, adopting the yields computed by RV for low- and intermediate-mass stars together with yields for massive stars as computed by Woosley and collaborators (Woosley et al. 1984; Woosley \\& Weaver 1995), could not reproduce the steep rise of C/O vs. O/H observed in the solar vicinity (Garnett et al. 1999). A way to solve this problem was suggested by Prantzos et al. (1994) who adopted the yields computed by Maeder (1992) which assume strong mass loss by stellar winds in massive stars. The C yields predicted by Maeder (1992) increase with metallicity, because mass loss itself is an increasing function of metallicity. Prantzos et al. showed that this C/O increase as well as the solar value of this ratio, could be reproduced by models adopting RV yields for low- and intermediate-mass stars and Maeder (1992) ones for massive stars. This fact, together with the uncertainty related to the $^{12}$C($\\alpha$,$\\gamma$)$^{16}$O reaction, led to the view that the main contributors to the carbon we observe today in the ISM are massive stars. Recent work in the literature seems to confirm this suggestion, even when the more recent yields of van den Hoek \\& Groenewegen (1997 - hereafter vdHG) for low- and intermediate-mass stars are used instead of those of RV (Liang et al. 2001; Henry et al. 2000; Carigi 2000). In the present paper we present strong arguments against such interpretation and we suggest that both the N and the C we observe at the present time in the ISM were mostly produced inside low- and intermediate-mass stars. Moreover we show that the N data in the solar vicinity can be explained without invoking important quantities of primary N in massive stars. We then check if our conclusions on the evolution of C, N and O in the MW are consistent with the CNO data available for other galaxies (blue compact galaxies - BCGs, other spiral galaxies and Damped Lyman-$\\alpha$ systems - DLAs), under the assumption that the stellar nucleosynthesis should be the same for all galaxies.  Abundance data are available for some extragalactic HII regions (Garnett et al. 1995a,b; 1997a,b; van Zee et al. 1997), including those measured in the outer parts of spiral disks (Ferguson et al. 1998; van Zee et al. 1998a,b; Garnett et al. 1999) and BCGs (Izotov et al. 1999; Izotov \\& Thuan 1999). Some of these objects have metallicities down to around $\\simeq$ 1/10 solar and are useful to study the behavior of chemical abundances at low metallicities. This is especially true for elements such as C and N.  The N/O vs. O/H diagram for dwarf galaxies is often interpreted in the literature (the same is true for the C/O vs. O/H diagram) as an evolutionary diagram, but instead it represents the final abundance values achieved by objects which evolved in a completely different way from each other (Diaz \\& Tosi 1986, Matteucci \\& Tosi 1985). Therefore, a meaningful comparison between model predictions relevant to dwarf irregulars and data should involve only the end points of the theoretical evolutionary tracks. Only in the case of the stars in the Milky Way we face an evolutionary diagram, where O/H can be interpreted as a time-axis.  One of the main questions nowadays is to assess the nature of nitrogen production in massive stars in order to explain the observations of N/O ratios in dwarf galaxies and DLAs which have low metallicities (Pilyugin 1999). The small ``plateau'' in the N/O ratio observed in low metallicity HII regions (e.g. Izotov and Thuan 1999) is often quoted as one indication that massive stars should produce an important quantity of primary N. This seems to be in conflict with the fact that the nitrogen over Si or S in some DLAs are well below the typical value observed in low-metallicity BCGs (e.g., Lu et al. 1998). In fact, DLAs offer another important piece of information (Pettini et al. 1995; Prochaska \\& Wolfe 2002) as in this case it is possible to probe the {\\it very early phases} of evolution of such systems. In this paper we show that the dwarf galaxy and DLA data can be understood without the necessity for primary nitrogen from massive stars.  Clearly, the best way to determine the lower limit of the N/O ratio and hence the existence of a possible primary N contribution from massive stars would be to measure the N/O ratios in Galactic halo stars at metallicities below [Fe/H] $\\simeq-$2. As we will show, this corresponds to the first 30 Myr of the evolution of the Milky Way, when the ISM was still not enriched by any intermediate-mass stars (the lifetime of an 8 M$_{\\odot}$ star is around 30 Myr). Unfortunately, the available data for N in halo stars are still too uncertain.  Finally, we discuss the problem of the abundance pattern in outer spiral disks. The CNO abundance ratios in outer disks are similar to those observed in dwarf irregular galaxies (van Zee et al. 1998b). We argue that this is consistent with the concept of ``inside-out'' formation of the disk, in which the timescale for accretion of gas onto the forming disk increases radially outwards (Matteucci \\& Fran\\c cois 1989; Chiappini et al. 1997, hereafter CMG97; CRM2001). In such a scenario, outer spiral disks have experienced slow star formation like dwarf irregulars. However, some important differences exist and we will address them by studying one spiral galaxy in particular, M101.  The paper is organized as follows. In Section 2 we discuss the different yield sets which we then use in the chemical evolution model for the MW. In Section 3 we present our chemical evolution models for the MW, dwarf galaxies and for M101. In Section 4 our results are shown and the conclusions are drawn in Section 5. ",
        "conclusions": "  \\begin{itemize}  \\item Models for the MW adopting van den Hoek and Groenewegen (1997) yields for low- and intermediate-mass stars and Thielemann et al. (1996) for massive stars are well in agreement with the abundance data on metal-poor stars, in particular the data for [O/Fe] obtained from [OI] and IR OH lines;  \\item We show that [C/Fe]$\\sim$0 over the whole [Fe/H] range clearly indicates that both C and Fe should come mainly from low- and intermediate-mass stars. This is at variance with the interpretation by several authors (e.g., Carigi 2000, Henry et al. 2000) that C should originate mainly in massive stars. This conclusion was based on the yields of Maeder (1992) which overestimate the effects of mass loss in massive stars. As shown by our good fit of [O/Fe] vs. [Fe/H], the yields of TNH are in good agreement with the observations, except for the fact that their C values should be increased by a factor of 3 for stars with M$>$40M$_{\\odot}$. This is in agreement with the new calculations of Meynet \\& Maeder (2002) taking into account rotation.  \\item A gap in the star formation rate between the thick and thin disk formation affects our model predictions for the C/O and/or N/O versus O/H plot. The existence of such a gap is already confirmed in [Fe/O] vs. [O/H] and [Fe/Mg] vs. [Mg/H] by observations (Gratton et al. 2000, Fuhrmann 1998). For the C/O and N/O versus O/H plot it is not possible to conclude the same from the few/uncertain available data.  \\item The threshold in the star formation rate is responsible for the slow chemical enrichment of the solar neighborhood in the last 4.5 Gyrs which explains the similarity between the observed abundances in Orion and the Sun.  \\item The [N/Fe] vs. [Fe/H] plot for halo stars show that N has a clearly secondary behavior at low metallicities thus implying that primary N production in massive stars is not important. Moreover, to obtain a solar [N/Fe] along the whole metallicity range, the required primary N is at least two orders of magnitudes higher than that predicted by the most detailed and up to date stellar evolution models (Meynet \\& Maeder 2002a).  \\item Our model predictions for the abundance gradients of C, N and O are in good agreement with the observations. In particular, we obtain a gradient for N/O which is flatter than the one of C/O. This is mainly due to the important contribution of primary nitrogen from intermediate-mass stars suffering HBB. More data are necessary to better constrain the C/O gradient in the Galaxy.  \\end{itemize}  When applying the ``best'' nucleosynthesis prescriptions to other galaxies we find that:  \\begin{itemize}  \\item Again, there is no need for claiming the existence of an important primary N contribution from massive stars to explain the abundance data of dwarf galaxies, outer spiral disks or DLAs;  \\item Models computed with TNH and vdHG yields can well reproduce the distribution of dwarf galaxies in plots such as log(N/O) and log(C/O) versus log(O/H)+12;  \\item The nucleosynthesis adopted for the MW is also in agreement with what is observed in other spiral galaxies (we show here the particular case of M101 for which the gradients of both N/O and C/O are available). As for the MW, there is an indication that the primary N coming from intermediate-mass stars is overestimated and that C is underestimated at solar metallicities;  \\item Invoking primary N in massive stars does not solve the problem of explaining the observed N/O gradient in M101. Instead, we suggest that the threshold in the star formation rate could explain the low abundance ratios at large galactocentric distances.  \\item Our predictions for the abundance gradients of N/O, C/O and C/N are also compatible with the idea that most of the C and N we see today in the ISM comes from low- and intermediate-mass stars.  \\item Our model for M101 suggests that the threshold in the gas density should increase with galactocentric distance.  \\end{itemize}  Finally, we stress that the N/O vs. O/H diagram for dwarf galaxies is often interpreted in the literature (the same is true for the C/O vs. O/H diagram) as an evolutionary diagram, but instead it represents the final abundance values achieved by objects which evolved in a completely different way from each other. In other words, this diagram is not the equivalent of the N/O vs. O/H for the solar neighborhood stars which is a real evolutionary plot, where for each O/H value corresponds a different galactic age. Therefore, any conclusion on the secondary/primary value of N derived from such a diagram should be taken with care."
    },
    "0209/astro-ph0209411_arXiv.txt": {
        "abstract": "In this paper, I discuss the capabilities and limitations of an 8-10\\,m ultraviolet/optical telescope in space, the proposed successor to the Hubble Space Telescope, in the context of galaxy studies.  The exquisite spatial resolution and excellent sensitivity of such a facility would open up new possibilities for the study of nearby dwarf galaxies ($z \\la 0.5$), and for studying the internal structure and kinematics of more luminous galaxies at high redshift ($z \\ga 2$). These applications are of particular importance because they would address areas in which the popular Cold Dark Matter  theory is in potential conflict with observations. ",
        "introduction": "The ``Cold Dark Matter'' (CDM) or hierarchical paradigm of structure formation provides a useful framework for attempting to understand galaxies and cosmology.  As the values of the cosmological parameters within this framework have become more and more tightly constrained by non-galaxy-based observations (like the cosmic microwave background and supernovae), most of the major uncertainties have to do with the messy ``gastrophysics'' that connects dark matter with the gas and stars that we can observe directly. Some of these issues include:  \\begin{itemize} \\item {\\bf Cooling} --- observations indicate that less of the gas in the Universe has cooled than is predicted by simulations. In particular, cooling in clusters seems to have been ``shut off'' by some unknown process.  \\item {\\bf Star formation} --- what determines how efficiently a galaxy can turn cold gas into stars? How does this efficiency scale with redshift and galaxy properties? What is the physical basis of empirical scaling laws such as the ``Kennicutt Law'' (Kennicutt 1989; 1998), and are they universal? What factors determine the duration and efficiency of the bursts of star formation seen in interacting galaxies? Is the stellar initial mass function universal in space and time, and why does it have its observed shape?  \\item {\\bf Stellar feedback} --- how does the energy from massive stars and supernovae affect the interstellar medium (ISM), intergalactic medium (IGM), and intracluster medium (ICM), and how does this impact future generations of stars?  Is the main effect predominantly thermal or kinetic?  How important are turbulence and large-scale galactic winds and outflows?  \\item {\\bf Heavy elements and dust} --- how efficiently are metals expelled from the potential wells of galaxies? How did the ICM and IGM get so uniformly polluted with heavy elements?  Is there a universal dust-to-metal ratio in all galaxies?  What determines the degree of optical/ultraviolet extinction a galaxy experiences? Do the properties of dust in galaxies (composition, temperature, etc.) differ dramatically from one galaxy to another, or as a function of redshift? \\end{itemize}  While the observations obtained with the new generation of space-based facilities and the large number of 6-10\\,m-class ground-based telescopes now coming on line will doubtless bring us a much better understanding of many of these issues, I expect that even 10--15 years from now, some of them will not be completely put to rest. In the remainder of this paper, I focus on a few specific examples of observations that would utilize the unique capabilities of the proposed Hubble Space Telescope successor (the ``Next HST'' -- NHST) to address important theoretical questions about galaxy formation. ",
        "conclusions": "I have suggested several applications of the proposed NHST 8-10\\,m ultraviolet/optical space telescope to the study of galaxies. A concise summary of these suggestions follows: \\begin{itemize} \\item Study stellar populations of dwarf galaxies (constrain their star-formation histories, and test ``squelching'' scenario). \\item Look for ``dwarf groups'' with high mass-to-light ratios (the tail of the ``squelched'' population). \\item Collect high resolution kinematic data for a complete sample of nearby galaxies, including dwarf and LSB galaxies (assess ``cusp'' crisis). \\item Constrain the dark matter around galaxies on scales of $\\sim100$ kpc using satellite galaxies, planetary nebulae, and/or globular clusters as kinematic tracers (understand disk and spheroid formation). \\item Obtain high resolution imaging and spatially resolved kinematics of $z\\ga2$ galaxies. \\item Obtain star-formation indicators and gas surface densities for a broad range of environments and redshifts, to study the universality of empirical star-formation scaling laws. \\end{itemize}"
    },
    "0209/astro-ph0209377_arXiv.txt": {
        "abstract": "We present surface photometry and stellar kinematics of a sample of 5 SB0 galaxies: ESO 139-G009, IC 874, NGC 1308, NGC 1440 and NGC 3412. We measured their bar pattern speed using the Tremaine-Weinberg method, and derived the ratio, $\\vpd$, of the corotation radius to the length of the bar semi-major axis.  For all the galaxies, $\\vpd$ is consistent with being in the range from 1.0 and 1.4, \\ie\\ that they host fast bars.  This represents the largest sample of galaxies for which $\\vpd$ has been measured this way. Taking into account the measured distribution of $\\vpd$ and our measurement uncertainties, we argue that this is probably the true distribution of $\\vpd$.  If this is the case, then the Tremaine-Weinberg method finds a distribution of $\\vpd$ which is in agreement with that obtained by hydrodynamical simulations. We compared this result with recent high-resolution $N$-body simulations of bars in cosmologically-motivated dark matter halos, and conclude that these bars are not located inside centrally concentrated dark matter halos. ",
        "introduction": "\\label{sec:introduction}  The pattern speed of a bar, $\\om$, is its main kinematic observable. When parameterized by the distance-independent ratio $\\vpd \\equiv \\lag/\\len$ (where $\\lag$ is the Lagrangian/corotation radius, at which a star is at rest in the bar's rest frame, and $\\len$ is the bar semi-major axis), it permits the classification of bars into fast ($1.0 \\leq \\vpd \\leq 1.4$) and slow ($\\vpd > 1.4$) ones. If $\\vpd < 1.0$ orbits are elongated perpendicular to the bar, so that self-consistent bars cannot exist in this regime (Contopoulos 1980).  A robust method for measuring $\\vpd$ relies on hydrodynamical simulations to model gas, particularly at shocks.  These studies find fast bars (\\eg\\ Lindblad \\etal\\ 1996; Lindblad \\& Kristen 1996; Weiner \\etal\\ 2001).  Hydrodynamical simulations can also obtain $\\vpd$ by matching morphological features in \\hi\\ (\\eg\\ Laine 1996; England \\etal\\ 1990; Hunter \\etal\\ 1989; Aguerri \\etal\\ 2001). Moreover, if the leading, offset dust lanes frequently found in bars can be identified with shocks, then fast bars seem to be the norm in late-type galaxies (van Albada \\& Sanders 1982; Athanassoula 1992). A model-independent method for measuring $\\om$ directly was obtained by Tremaine \\& Weinberg (1984).  The Tremaine-Weinberg method (hereafter, the TW method) is given by the simple expression $\\pin \\om\\sin i = \\kin$, where $\\pin$ and $\\kin$ are luminosity-weighted mean position and velocity measured along slits parallel to the line-of-nodes.  If a number of slits at different offsets from the major-axis are obtained for a galaxy, then plotting $\\kin$ versus $\\pin$ for the different slits produces a straight line with slope $\\om \\sin i$.  To date this method has been applied successfully to 3 early-type barred galaxies (Kent 1987; Merrifield \\& Kuijken 1995; Gerssen \\etal\\ 1999; Debattista \\etal\\ 2002) and preliminary results have been obtained for another 5 galaxies ranging from SB0 to SBb (Debattista \\& Williams 2001; Gerssen 2002).  Although all these galaxies are consistent with having fast bars, the sample size is still small enough that the range of $\\vpd$ recovered by the TW method is still poorly defined.  We have started a program to enlarge the sample of pattern speeds measured with the TW method, including to galaxies of various bar strength, environment, luminosity, inclination etc.  In a previous paper, (Debattista \\etal\\ 2002, hereafter Paper I), we reported on the special case of NGC 1023, a system which shows evidence of a past interaction with one or more of its satellite galaxies.  In this paper, we present results for an additional 5 SB0 galaxies. ",
        "conclusions": "\\label{sec:conclusions}  The 5 SB0's presented in this work, together with NGC 1023 studied in Paper I, which we include in our sample for this discussion, represent the largest sample of barred galaxies, with $\\om$ measured by means of the TW method.  For all of them, $\\vpd$ is consistent with being in the range 1.0 to 1.4, within the errors, \\ie\\ with each having a fast bar.  The unweighted average for the sample is $\\overline{\\vpd} = 1.1$. The apparent range of $\\vpd$ spans from 0.8 to 1.6 (0.6 to 2.1, within the 67 per cent intervals).  This spread is not related to the properties of the galaxies in any obvious way (\\eg\\ for the two galaxies at $V_{\\rm c, flat} = 277$ \\kms, NGC 1023 and NGC 1440, the measured values of $\\vpd$ are at opposite extremes of the distribution).  The fact that some of the values of $\\vpd$ are nominally less than unity leads us to suggest that the large range of $\\vpd$ is a result of random errors and/or scatter in the measurements.  The sources of random errors are largely due to measurement uncertainties in all 3 quantities used to compute $\\vpd$, \\ie\\ $V_{\\rm c,flat}$, $\\len$ and $\\om$.  Of these, the largest is in $\\om$, amounting to typical fractional uncertainties of 30 per cent, followed by $\\len$, for which the typical fractional uncertainty is 20 per cent.  These uncertainties account for the typical large (and asymmetric towards large values) errors on measurements of $\\vpd$.  A likely source of scatter is errors in the disc PA.  Debattista (2002) shows that, for PA errors of FWHM $5\\degrees$ (note that the root-mean-square PA uncertainty in our sample is $2\\fdg1$), the scatter in $\\vpd$ is of order 0.44, large enough to account for the $\\vpd < 1$ cases.  Since PA errors scatter $\\vpd$ to both larger and smaller values, then the largest measured value of $\\vpd = 1.53$ is probably an over-estimate. If this is the case, then TW measurements are finding the same range of $\\vpd$ as do hydrodynamic simulations.  The conclusion that bars are fast constrains the dark matter distribution in disk galaxies.  Debattista \\& Sellwood (1998, 2000) argued that bars this fast can only survive if the disc in which they formed is maximal.  Recent high resolution $N$-body simulations with cosmologically-motivated dark matter halos produce bars with $\\vpd$ in the range between 1.2 and 1.7 (Valenzuela \\& Klypin 2002).  Even discounting our argument above in favor of a more restricted range of $\\vpd$, Fig.  \\ref{fig:scatter} shows that $\\vpd = 1.7$ is possible only for the bars of IC 874, NGC 1440 NGC 3412 and, marginally, NGC 936, while the bars of ESO 139-G009, NGC 1023, NGC 1308 and NGC 4596 never reach this value of $\\vpd$.  Note, moreover, that 3 of the galaxies that do reach $\\vpd = 1.7$ have amongst the largest fractional errors in $\\vpd$.  Therefore we conclude that the $N$-body models of Valenzuela \\& Klypin (2002) probably produce slower bars than the observed.  The galaxy ESO 139-G009 is classified as SAB (\\ie\\ weakly barred) in RC3; the fact that it hosts a fast bar suggests that weak bars form via the same mechanism that forms the strong ones. Thus the hypothesis of Kormendy (1979), that weak bars are the end state of slowed down fast bars, already in question from $N$-body simulations (Debattista \\& Sellwood 2000), is also unsupported by the limited observational data.  Further measurements of $\\vpd$ for weak bars would be of considerable interest.    \\bigskip \\noindent {\\bf Acknowledgements.}  \\noindent VPD and JALA acknowledge support by the Schweizerischer Nationalfonds through grant 20-64856.01. JALA was partially supported by Spanish DGC (Grant AYA2001-3939). EMC acknowledges the Astronomisches Institut der Universit\\\"at Basel for the hospitality while this paper was in progress. We are indebted to R.  Bender and R. Saglia for providing us with the FCQ package which we used for measuring the stellar kinematics.  We are also grateful to A. Pizzella for the images he acquired.  This research has made use of the Lyon-Meudon Extragalactic Database (LEDA) and of the NASA/IPAC Extragalactic Database (NED). This paper is based on observations carried out with the New Technology Telescope and the Danish 1.54-m Telescope (Prop.  No. 67.B-0230 and No.  68.B-329) at the Europen Southern Observatory, La Silla (Chile), with the Italian Telescopio Nazionale Galileo (Prop. AOT-3, 3-06-119) operated on the island of La Palma by the Centro Galileo Galilei of the Consorzio Nazionale per l'Astronomia e l'Astrofisica, and with the Jacobus Kapteyn Telescope operated by the Isaac Newton group at La Palma island at the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrof\\'\\i sica de Canarias.  \\bigskip \\noindent"
    },
    "0209/astro-ph0209188_arXiv.txt": {
        "abstract": "WASP0 is a prototype for what is intended to become a collection of WASPs whose primary aim is to detect transiting extra-solar planets across the face of their parent star. The WASP0 instrument is a wide-field (9-degree) 6.3cm aperture F/2.8 Apogee 10 CCD camera (2Kx2K chip, 16-arcsec pixels). The camera is mounted piggy-back on a commercial 10-inch Meade telescope. We present some recent results from the WASP camera, including observations from La Palma of the known transiting planet around HD 209458 and preliminary analysis of other stars located in the same field. We also outline further problems which restrict the ability to achieve photon limited precision with a wide-field commercial CCD. ",
        "introduction": "Of the indirect methods, the use of transits is rapidly developing into a strong and viable means to detect extra-solar planets. A transit occurs when the apparent brightness of a star decreases temporarily due to an orbiting planet passing between the observer and the stellar disk. Since this can only be observed when the orbital plane is approximately aligned with the line of sight, this transit method clearly favours large planets orbiting their parent stars at small orbital radii (``hot Jupiters''). Hence, a large sample of stars must be monitored in order to detect transiting planets.  In this report, we briefly describe the data reduction methods and initial results from WASP0, a prototype wide-angle CCD camera currently being used to search for transiting extra-solar planet signatures. ",
        "conclusions": "The preliminary analysis of WASP0 data presented here demonstrates that this instrument is able to achieve the necessary precision required to detect transit events due to extra-solar planets. Further calibrations and refinement of the PSF model are needed to determine if WASP0 is able to detect even smaller photometric deviations. This prototype has successfully served as a proof-of-concept for SuperWASP. Further details on the SuperWASP project are available in a separate paper by Street et al. (2002) in this volume."
    },
    "0209/astro-ph0209231_arXiv.txt": {
        "abstract": "We treat the production of neutrons, photons, and neutrinos through photomeson interactions of relativistic protons with ambient photons in the compact inner jets of blazars. Internal synchrotron and external isotropic radiation due to scattered optical/UV accretion-disk radiation are considered as target photon fields. Protons are assumed to be accelerated to a maximum energy limited by the size scale and magnetic field of the jet, and by competing energy losses. We characterize the conditions when the photomeson interactions of ultrarelativistic protons become effective, and show that the presence of the external radiation field makes possible strong energy losses already for protons with energies $E_p\\gtrsim 10^{15}$ eV. Without this component, effective energy losses of protons begin at $E_p\\gtrsim 10^{18}$ eV, and would rapidly disappear with expansion of the blob.  We develop a model describing the production and escape of neutrons from a comoving spherical blob, which continue to interact with the ambient external radiation field on the parsec-scale broad line region (BLR).  Neutrons may carry $\\approx 10$\\,\\% of the overall energy of the accelerated protons with $E_p\\gtrsim 10^{15}$ eV outside the BLR. Ultra-high energy gamma rays produced by photomeson interaction of neutrons outside the blob can also escape the BLR. The escaping neutrons, gamma rays, and neutrinos form a collimated neutral beam with a characteristic opening angle $\\theta \\sim 1/\\Gamma$, where $\\Gamma$ is the bulk Lorentz factor of the inner jet.  Energy and momentum is deposited in the extended jet from the decay of neutrons at distances $l_{d}(E_n)\\approx (E_n/10^{17}\\;{\\rm eV})$ kpc, and through pair-production attenuation of gamma rays with energies $E_\\gamma\\gtrsim 10^{15}$ eV which propagate to $\\sim10$-100 kpc distances. In this scenario, neutral beams of ultra-high energy gamma rays and neutrons can be the reason for straight extended jets such as in Pictor A. Fluxes of neutrinos detectable with km-scale neutrino telescopes are predicted from flat spectrum radio quasars such as 3C 279. ",
        "introduction": "Multiwavelength observations of flares from blazars, particularly in the $\\gamma$-ray domain, have convincingly demonstrated that the compact inner jets of blazars are effective accelerators of particles to very high energies \\citep{har99,wee00}.  Analyses of correlated X-ray and TeV gamma-ray flares in BL Lac objects lend support to leptonic models \\citep{mk97,cat97,pia98,tav01,kca02}, which imply efficient acceleration of relativistic electrons in these sources, probably due to relativistic shocks.  Associated acceleration of hadrons is expected with at least the same efficiency as that of the leptons, except perhaps for electron-positron pair jet models where few hadrons are present. Comparison of the radio lobe and inner jet powers indicates that jets are composed mainly of electrons and protons \\citep{cf93}, so that a nonthermal hadronic component is expected.  Acceleration of hadrons in blazar jets could be directly confirmed with the detection of neutrinos, provided that there are significant interactions of accelerated hadrons with ambient material or photon fields. Detectable synchrotron emission at TeV energies could be radiated by ultrarelativistic protons and ions, but this requires extremely strong magnetic fields $\\sim 20$-100 G in BL Lac objects \\citep{aha00,mp01}. All other observable consequences of hadron acceleration result from the same interactions that produce neutrinos. These include X-ray and gamma-ray production from secondary leptons and gamma rays formed in hadronic interactions and, as we show here, deposition of energy transported by the escaping neutral radiations far from the inner jet.  Models invoking interactions with ambient matter \\citep{bb99,ps00,sps02} require mass-loaded jets. If the nuclear interaction energy-loss time scale $t^\\prime_{pp} = (n_p^\\prime c K_{pp} \\sigma_{pp})^{-1}$ is to be less than the variability time scale $t^\\prime_{var}$, then high plasma densities $n_p^\\prime \\gg 10^9/[t_{var}({\\rm d})(\\delta/10)]$ cm$^{-3}$ are required, where $n_p^\\prime$ is the comoving thermal proton density, $t_{var}({\\rm d})$ is the observed variability time scale in days, $\\delta$ is the Doppler factor, $\\sigma_{pp}\\cong 30\\,$mb is the nuclear interaction cross section, and $K_{pp}\\simeq 0.5$ is the inelasticity. (Henceforth primes denote quantities in the comoving frame.) Such models would, however, be inefficient if the sources are to be optically thin, as is required for nonthermal X-ray escape. For Thomson-thin jets ($\\tau_{\\rm sc} = n_p^\\prime \\sigma_{\\rm T} R^\\prime < 1$), the nuclear interaction time scale \\begin{equation} t^\\prime_{pp} = {\\sigma_{\\rm T}\\over K_{pp} \\, \\sigma_{pp}}\\;{R^\\prime\\over c\\tau_{\\rm T}} > 40 \\;t^\\prime_{dyn}\\;, \\label{tpp} \\end{equation} where $R^\\prime$ is the comoving blob radius, $t^\\prime_{dyn} = R^\\prime/c$ is the dynamical (or light crossing) time scale, and $\\sigma_{\\rm T}\\cong 665\\,$mb is the Thomson cross sections. Thus protons can only lose $\\lesssim 2$-3\\% of their energy on the dynamical time scale. Moreover, as shown in Appendix A, nuclear $pp$ interaction models require large masses and kinetic energies.   A second group of hadronic models is based upon photomeson interactions of relativistic hadrons with ambient photon fields in the jet. Most of the models of this type take into account collisions of high-energy protons with the internal synchrotron photons \\citep{mb92,man93,muc02}, while others also take into account external radiation that originates either directly from the accretion disk \\citep{bp99} or from disk radiation that is scattered by surrounding clouds to form a quasi-isotropic radiation field \\citep{ad01}. BL Lac objects have weak emission lines, so in these sources the dominant soft photon field is thought to be the internal synchrotron emission. Strong optical emission lines from the illumination of broad-line region (BLR) clouds in flat spectrum radio quasars (FSRQs) reveal bright accretion-disk and scattered disk radiation in the inner regions \\citep{net90}.  In the case of internal synchrotron radiation, the energy output of secondary particles formed in photohadronic processes is generally peaked in the energy range from $\\approx 10^{16}$-$10^{18}$ eV in either low- or high-frequency peaked BL Lac objects \\citep{muc02}, which implies that such models can only be efficient if protons are accelerated to even higher energies. This demand upon proton acceleration for efficient photomeson production on the {\\it internal} synchrotron photons also holds in FSRQs, which have similar nonthermal soft radiation spectra as low-frequency peaked BL Lac objects.  As shown by \\citet{ad01}, however, the presence of the isotropic external radiation field in the vicinity of the jets of FSRQs strongly improves the photomeson production efficiency and relaxes the very high minimum proton energies needed for efficient production of secondaries. The existence of a strong external radiation field is required to explain \\citep{ds93,ds02,sbr94,sik01,dss97,bot00} the luminous 100\\,MeV - GeV gamma-ray emission observed with the EGRET instrument on the {\\it Compton Gamma Ray Observatory} \\citep{har96}.  In the model of \\citet{ad01}, protons are assumed to be accelerated in an outflowing plasma blob moving with bulk Lorentz factor $\\Gamma$ along the symmetry axis of the accretion-disk/jet system. The relativistic protons are assumed to have an isotropic pitch-angle distribution in the comoving frame of a plasma blob, within which is entrained a tangled magnetic field. In our study, we determined the intensity of the internal radiation fields based on observations of 3C 279 during the flaring state in 1996 \\citep{weh98}, and calculated the high-energy neutrino flux expected under the assumption that the power to accelerate relativistic protons was equal to the power injected into nonthermal electrons which explains the observed gamma-ray emission.  The presence of a quasi-isotropic external radiation field enhances the neutrino detection rate by an order-of-magnitude or more over the case where the field is absent, so we predict that FSRQs can be detected with km-scale neutrino detectors, whereas BL Lac objects are not as promising for neutrino detection. The model also takes into account the effects of relativistic neutron production, which can escape from the blob unless they are converted back to protons due to decay or further photohadronic collisions inside the blob.  In this paper, we further develop this model. Details of the theory are presented that describe the evolution of relativistic protons, taking into account photohadronic energy losses and neutron escape from the relativistically moving blob. Production of neutrons has been considered earlier in the stationary cores of AGNs \\citep{ew78,sbr89, km89, brs90, gk90, a92a, a92b}, though not in the context of a jet model. The neutrons escaping from a relativistic blob form a beam with opening angle $\\theta_n \\cong \\Gamma^{-1}$. These neutrons are subject to decay and further photohadronic interactions while passing through the quasi-isotropic external radiation field on the parsec scale, which we associate with the broad emission-line region (BLR). Ultra-high energy gamma rays produced from the decay of secondary pions outside the blob can escape the BLR because they are no longer subject to strong $\\gamma\\gamma$ interactions with the synchrotron photons inside the blob when they are at a distance of a few $\\times R^\\prime$ from the blob.  This results in a neutral beam of ultra-high energy (UHE) neutrons and gamma rays that can transport the energy to large distances from the central engine without significant interactions with the ambient medium until the neutrons decay or gamma rays are converted to electron-positron pairs through pair attenuation with diffuse radiation fields. This scenario naturally explains the appearance of large-scale jets that are colinear with the inner jets and appear straight on scales of hundreds of kpc. We also consider whether the different morphologies of FRI and FRII radio galaxies are a consequence of the different neutral beam properties formed in BL Lac objects and FSRQs.  In Section 2, we describe the model, our treatment of photomeson production and the energy spectra of secondaries, and our method to estimate the radiation and magnetic fields in the blob. The results of the calculations are presented in Section 3, and a discussion and summary are given in Section 4. A comparison of neutrino production through photomeson $p\\gamma$ and nuclear $p p$ interactions in jets is given in Appendix A, and the conditions when the direct accretion-disk radiation field is less important than the scattered radiation field are derived in Appendix B. ",
        "conclusions": "Observations of powerful gamma-ray flares from blazars with EGRET and ground-based gamma-ray telescopes have confirmed that radio-loud AGNs accelerate particles to GeV and TeV energies, respectively. Particle acceleration to higher energies may occur, but this cannot be established through gamma-ray observations because of attenuation by ambient and diffuse extragalactic optical and infrared radiation fields.  Although leptonic models have been successful in fitting the spectra of blazars, models of first-order Fermi acceleration at both nonrelativistic and relativistic shocks imply that acceleration proceeds more effectively for protons than for electrons, because the Larmor radius of a particle involved into the acceleration process must be larger than the shock width, which is more easily satisfied by hadrons \\citep{gal02}. The large proton-to-electron ratio in the galactic cosmic rays supports this contention. Thus our underlying assumption that protons are injected with at least the same power as inferred from the measured gamma-ray luminosity of a blazar seems reasonable.  Evidence for the acceleration of relativistic protons and heavier nuclei in blazar jets can be provided by three lines of evidence as described in the following subsections.  First, hadronic acceleration could be established indirectly by detection of spectral features in the electromagnetic radiation produced by secondaries of the inelastic interactions of high-energy hadrons in blazar jets. The second and most compelling line of evidence will be provided through the direct detection of high-energy neutrinos by km-scale detectors such as IceCube.\\footnote{http://icecube.wisc.edu/} The third line of evidence involves consequences of the production of UHE neutrons and gamma-rays which transport significant amounts of energy from the central region of an AGN to large distances \\citep{ad01}. This last feature is sufficiently interesting that we devote a separate subsection to applying this model to explain observed features of radio-loud AGNs. Finally we summarize, considering also the possibility that ultra-high energy cosmic rays (UHECRs) are accelerated in the inner jets of blazars.  \\subsection{Electromagnetic Radiation from Nonthermal Hadrons}  Discriminating between hadronic and leptonic origins of the gamma radiation from blazars or other classes of sources is possible, although not an easy or a straightforward task.  If the emission has a hadronic origin, then the emerging gamma-ray spectrum results from a cascade induced by high-energy secondary leptons and gamma-rays. Photon spectral indices from cascade radiation tend to be rather hard, and take values between 1.5 and 2.0 in the medium-energy gamma-ray regime explored by EGRET and {\\it GLAST} \\footnote{http://glast.gsfc.nasa.gov} (see Figure 5). But this is also a common feature for gamma-ray spectra from an electron-photon cascade independently of whether the cascade is initiated by electrons and gamma-rays of secondary (pion-decay) origin or by primary (directly-accelerated) electrons.  A principal difference between the hadronic and leptonic induced cascades does, however, exist. It consists in a very significant contribution of the synchrotron radiation by the ultra-relativistic electrons (including the ones from the absorption of $\\pi^0$-decay gamma-rays) of the first cascade generation in the high energy gamma-ray flux in the case of hadronic initiation of the cascade. The Lorentz factor of these electrons $\\gamma \\gg 10^8$ is much higher than the maximum possible Lorentz factor for primary electrons, which is limited by synchrotron losses in the strong magnetic fields $B\\gtrsim 0.1$-$10\\,\\rm G$ characteristic for the inner jets of blazars.  As a result, the synchrotron radiation of these electrons can greatly exceed the characteristic maximum energy $E_{\\rm s,max}^\\prime \\sim 25$-100  MeV \\citep{gfr83,jag96} for synchrotron emission of directly accelerated electrons, as is apparent from Figure 5. Additionally taking into account the Doppler boosting of the energy $E_{\\rm s,max}$ when transforming to the observer's frame, we can say that confirmation of a significant contribution of hard-spectrum synchrotron flux in the gamma-ray flares at $\\gtrsim 0.1\\,\\rm GeV$ with {\\it GLAST} could become a strong argument in favor of UHE hadron acceleration in the jets.  A synchrotron contribution could be confirmed experimentally by the detection of a significant polarization in the total flux, but neither GLAST nor other high energy gamma-ray detectors yet have the capability for the relevant measurements.  It is possible that the synchrotron origin of the GeV radiation could be revealed by accurate modeling of the spectral and temporal behaviors of the broad-band gamma ray fluxes detected by GLAST, as well as by forthcoming ground-based Cherenkov detectors. Of particular interest in this regard could be the interpretation of flares which would {\\it decline} rapidly at 0.1-1 GeV energies at the flare fading stage. The key observation here is that in FSRQs, the Compton radiation at these energies should be due to upscattering of external UV photons by electrons with $E^\\prime \\lesssim 1\\,\\rm GeV$, which could have cooling times larger than the measured decline time, whereas the GeV synchrotron radiation is due to electrons with many orders of magnitude higher energies, and so it would have much shorter cooling times.  If radio galaxies are misdirected blazars and steep spectrum radio sources are off-axis FSRQs \\citep{bbr84,up95,ob82}, then the gamma-ray spectra of radio galaxies could also reveal hadronic acceleration in blazar jets.  Nonthermal cosmic rays that escape from the inner jets of radio galaxies can accumulate in the central gas-rich region of the AGN. Nuclear collisions of these cosmic rays with ambient gas would produce a quasi-stationary (on time scales of thousands of years) and compact gamma-ray halo with a characteristic $\\pi^0$-decay feature near 70/(1+z) MeV. Note however that the bremsstrahlung emission from $\\pi^\\pm$-decay electrons of $\\sim$0.1-1 GeV energies is likely to conceal the 70 MeV benchmark \\citep{sch82}.  Nuclear interactions will give secondary $\\pi$-decay emission that is brightest in the central BLR core of a radio galaxy where the target particle density is highest, and dimmer in the extended region due to the tenuousness of the interstellar medium on scales of hundreds of pc.  Because of the rather small size of this hadronic halo, the gamma-ray detectors would detect it as a point source.  Such a compact hadronic halo could be still resolved in the radio band, where the radio flux produced by secondary electrons of GeV energies would show an increase in brightness towards the center of the AGN on the milliarcsecond scale. Note that at very high energies, a halo on the much larger scale (multi-Megaparsec) could be formed around such sources due to absorption and cascading of multi-TeV gamma-rays in the intergalactic medium \\citep{acv94}.   Considering possible signatures of relativistic hadrons during the flares, we note that a pronounced 70 MeV $\\pi^0$-decay maximum  should be due first of all to hadrons with kinetic energies $\\sim 1$-10\\,GeV; therefore it would imply a mass-loaded jet. As we show in the Appendix A, the mass-loaded jet models are rather unlikely because they confront severe energetics difficulties. Nevertheless, the hadronic $\\pi^0$-decay gamma-ray feature, although significantly broadened and Doppler-shifted, could be expected in those particular cases when the inner jet dumps its energy in collisions with the dense target, such as a broad line cloud crossing the jet.  Observations of radio galaxies with imaging X-ray telescopes, {\\it GLAST}, and ground-based air Cherenkov telescopes with low energy thresholds will provide spectral data that should be analyzed for evidence of nonthermal hadrons in light of these considerations.  \\subsection{Neutrinos from FSRQs and BL Lacs}  Detection of neutrinos from blazars will provide the most compelling evidence for hadronic acceleration in blazar jets.  For photohadronic jet models, we have argued here and elsewhere \\citep{ad01} that the presence of a strong UV accretion-disk radiation field is required for blazar neutrino fluxes to be detectable with km-scale high-energy neutrino telescopes. The disk radiation field is needed to produce a quasi-isotropic scattered radiation component produced in the BLR of FSRQs. The existence of an external radiation field in FSRQs is suggested by luminous broad emission lines detected from these objects. For UV fields expected from optically-thick accretion disks, photomeson interactions can efficiently extract the energy of relativistic protons with (relatively modest) energies $\\gtrsim 10^{15} \\,\\rm eV$, resulting in the production of neutrinos with $E_\\nu \\gtrsim 3\\times 10^{13}\\,\\rm eV$. For flares like the one detected in February 1996 from 3C 279 with EGRET, the fluence of neutrinos shown in Figure 3 is at the level of $10^{-4} \\;\\rm erg\\, cm^{-2}$, implying $\\approx0.3\\; \\nu_{\\mu}$ per flare (i.e., a detection probability of $\\approx 30\\%$).  This would suggest a realistic possibility for IceCube to detect a few to several neutrinos from several flares over the course of a year. Given the flaring duty cycle of FSRQs estimated from EGRET observations ($\\sim 10$-20\\%), this implies that a few high-energy neutrinos will be detected from FSRQs such as 3C 279 on a time scale of 1 year. Considerably more will be detected if the efficiency to accelerate hadrons exceeds our baseline assumption given by equating proton and gamma-ray power.  The additional possibility that some neutrinos will also be detected during quiescent states strengthens our prediction that FSRQs will be detectable high-energy neutrino sources with km-scale neutrino telescopes.  We note that on a 1 year observation time scale, detection of only two neutrinos with energies above 30 TeV from the same direction pointed to a known blazar would represent a very significant detection of an astrophysical source, because at these energies the background cosmic-ray induced neutrino flux expected within one square degree (which is typical for the angular resolution of a high-energy neutrino telescope in ice or water) is $\\lesssim 0.03$ \\citep{ghs95}.  For BL Lac objects, where the accretion-disk and scattered radiation components are much weaker, and with GeV gamma-ray fluxes that are also typically not as bright as in FSRQs, the chances for detection of neutrinos even with a 1 km-scale neutrino detector are negligible. For the proton fluxes shown in Figures~6 and 7 calculated for parameters of the April 1997 flare in Mrk 501, the probability of neutrino detection during the flare is less than $10^{-6}$.  It is also important to note that unlike the case of $p\\gamma$ interactions with the internal synchrotron photons, the intensity of $p\\gamma$ interactions with the external radiation does not depend on the blob size. Therefore in FSRQs the neutrino production from photomeson interactions with the external radiation field does not decline so much in the process of expansion of the blob as it propagates through the quasi-isotropic external radiation field at $R\\lesssim R_{\\rm BLR}$ as it does in BL Lac objects.  We note that the blob expansion seems unavoidable because of the difficulty in confining the high internal pressures of the blob by the pressure of the external medium.  An important issue concerns the quenching of the jet as it sweeps up mass from the BLR.  As BLR clouds execute Keplerian rotational motion about the central supermassive black hole, some will pass through the path of the jet outflow.  This will cause energization and deceleration of the radiating plasma, as in the external shock model of gamma-ray bursts \\citep{mr93}. Suppose that the AGN jet is beamed into a fraction $f_b =0.01 f_{-2} $ of the full sky. In a simple model with a uniform BLR gas density $n_{BLR}$ within a spherical BLR with Thomson depth $\\tau_{\\rm T} = n_{BLR}\\sigma_{\\rm T} R_{BLR}$, the total mass in the beaming cone is $M_j(<R_{BLR}) \\cong 5250 f_b \\tau_{\\rm T} R_{18}^2 M_\\odot$, where $R_{18} = R_{BLR}/10^{18}\\; {\\rm cm}$ (see \\citet{bl95} for BLRs with radially dependent densities).  The time for BLR material to pass through the cone of the jet with opening half-angle $\\theta_j \\cong 2\\sqrt{f_b}$ is $t_j \\cong 1.1\\times 10^9 f_{-2}^{1/2} R_{18}^{3/2}/M_9^{1/2}$ s, where the mass of the central supermassive black hole is $M_9 = M_{bh} /10^9 M_\\odot$. Assuming that the directional power of relativistic plasma expelled in the AGN jets is comparable to the apparent isotropic gamma-ray luminosity $L_\\gamma$, then the flow will be significantly decelerated when $L_\\gamma t_j \\cong \\Gamma^2 M_j(<R_{BLR})c^2$, where $\\Gamma$ is the Lorentz factor of the plasma outflow. Thus, unless the mean apparent isotropic luminosity exceeds \\begin{equation} L_{\\gamma}({\\rm ergs~s}^{-1}) \\; \\cong \\; 10^{48}\\; (f_{-2} R_{18} M_9)^{1/2}\\;\\left({\\tau_{T}\\over 0.1}\\right)\\, \\left({ \\Gamma\\over 10}\\right)^2\\;, \\end{equation} the jet will be quenched  by sweeping up BLR gas. This quenching process could make it difficult to form extended relativistic jets on the 10-100 kpc scale, as may be required to explain enhanced X-ray emission in the knots of PKS 0637-752 \\citep{tav00}. The transport of inner jet energy to the extended jets via neutral beams avoids this difficulty. The neutrons and gamma-rays will pass through the BLR material without attenuation unless the Thomson depth of BLR is very high, with $\\tau_{T} \\gtrsim 10$. In these cases the jet and neutral beam outflow would be quenched, the nonthermal radiation would be reprocessed into lower energy emissions, and only the neutrino signal would penetrate through the dense inner region without reprocessing. If so, the neutrino emission from such buried jets in non-blazar sources could reveal these types of AGN. We return to this speculation in Section 4.4.  \\subsection{Energy Transport to the Extended Jet by Neutral Beams}  Relativistic neutrons produced by $p\\gamma$ collisions in the compact inner jet are no longer confined by the jet's magnetic field, and so can escape from the blob unless the neutron decays or interacts with another soft photon that results in conversion to a proton inside the blob. Our calculations show that in FSRQs with luminous external radiation fields, up to $\\sim 30\\%$ of the total energy injected into the blob in the form of relativistic protons with $E\\gtrsim 10^{15}\\,\\rm eV$ can leave the blob in the form of relativistic neutrons with Lorentz factors $\\gamma_n \\gtrsim 10^{6}$.  For the case of FSRQs, neutrons escaping the inner blob continue to propagate in the BLR with its quasi-isotropic scattered accretion-disk radiation up to distances $R\\sim R_{BLR} \\sim 0.1$-1 pc. Subsequent interactions of the neutron beam with the scattered disk photons will result in the production of beamed secondaries.  Neutrino secondaries will add to the overall neutrino flux. Importantly, a beam of gamma-rays with energies above $10^{15} \\,\\rm eV$ is also produced. Because these gamma-rays are made outside the blob and propagate radially outward in the blob's frame, they do not interact with the internal synchrotron photons, and therefore have a chance to escape from the central region of the quasar (see Figure 2). Depending on the extent and intensity of the isotropic external radiation field, as much as several per cent of the accelerated proton energy $W_{p}(\\geq 10^{15} \\,\\rm eV$) injected in the inner jet at distances $R\\leq R_{BLR} $ could thus be converted to gamma-rays which escape from the BLR.  Figure 8 shows the mean-free-path of gamma rays due to pair-production interactions with photons of the diffuse cosmic microwave background radiation for $z = 0$ (solid curve) and $z = 0.538$ (dashed curve). This calculation also takes into account the diffuse extragalactic infrared radiation field.  We approximate this field by a spectrum $\\nu J_\\nu \\propto const$ at the level of total energy density $u_{IR} = 0.01$ eV cm$^{-3}$ between $0.001$\\,mm and 1\\,mm \\citep{pri99}, though the precise value is not important on the size scales $\\lesssim 1$ Mpc under consideration.  Our calculations show that more energy is typically carried out from the BLR by relativistic neutrons than by gamma rays. This is especially true if the degradation of the neutron beam escaping the blob but still in the BLR is not very strong, which will always be the case for neutrons escaping the blob at distances $R \\sim R_{BLR}$. The energy carried away by neutrons with $E_n \\geq 10^{17} \\,\\rm eV$ may reach $\\sim 10\\%$ of $W_{p}(\\geq 10^{15} \\,\\rm eV$) for accelerated protons with number index $\\alpha_{p}\\simeq 2$. Adding the energy of the gamma rays to this, we find that $\\sim 5\\%$ can be used as a rough estimate for the fraction of total (starting from 1 GeV) energy of accelerated protons in the inner jet that is carried by the collimated beam of UHE neutrons and gamma rays to $>$kpc distances in FSRQs.  For a flare like the one in February 1996 from 3C 279, the gamma-ray fluence measured with EGRET corresponds to an apparent isotropic energy release of $2\\times 10^{54}\\,\\rm erg$. Assuming that the injection power in nonthermal protons in the inner jet is the same as for the parent electrons, the overall total energy in the neutral beam emerging from the inner jet per a single such flare could reach the level $\\sim 10^{51} \\delta_{10}^{-2} \\,\\rm erg$ in the stationary frame. This energy will be collimated in the cone with opening angle $\\sim 1/\\delta$, and it will be deposited at distances up to 0.1-1 Mpc, provided that the spectra of ultra-high energy neutrons extend up to $10^{19}$-$10^{20}\\,\\rm eV$, as in Figure 4b.  After neutron-decay or gamma-ray attenuation, the charged particles emerging in the direction of the parent neutral beam will interact with the ambient extragalactic medium via local magnetic and photon fields.  In the case of the neutrons, a small fraction $\\sim m_e/m_p$ of the neutron energy will appear in the form of $\\beta$-decay electrons with Lorentz-factors $\\gamma > 10^8$.  This energy will be immediately available for radiation in the synchrotron and Compton processes with $\\nu F_\\nu$ peaks at hard X-ray and multi-TeV energies, respectively, for a 1 $\\mu$G field and a dominant local CMB radiation field \\citep{der02}.  The dominant fraction of the initial neutron energy remains, however, in the protons.  Synchrotron radiation of ultrarelativistic neutron-decay protons could in principle explain the extended {\\it Chandra} X-ray jet emission if the ambient magnetic field is sufficiently large \\citep{aha02}. Proton synchrotron radiation does not, however, provide a mechanism for deposition of energy of the neutral beam into the surrounding medium to drive shocks and accelerate  $\\sim 1$-100 GeV electrons, which is required to explain the radio and optical emission from the extended jets.  In order for the transport of energy by the neutrons to be efficient from the point of view of powering the extended jets in radio galaxies, there should be a mechanism for effective transfer of the neutron-decay proton energy to the surrounding plasma. Such a mechanism is expected through interactions of the relativistic protons as well as the $e^{+}$-$e^{-}$ pairs from the gamma-ray beam with the ambient magnetic field in the surrounding medium.  In the case of a significant transverse magnetic-field component $B_\\perp$, both the protons and leptons will change their initial directions on the Larmor timescales $t_L = E/eB_\\perp c$, which implies that a significant fraction of their momentum and energy will be transferred to the ambient medium.  Another, and possibly even more effective way for transferring the initial momentum and energy of the charged particles is by generating MHD turbulence through beam instabilities excited by the highly beamed neutron-decay protons and charged secondaries emerging from the charge-neutral beam.  The continuous transfer of momentum and energy from the beam to the intergalactic medium could stretch the ambient magnetic field to form a channel facilitating beam propagation, or drive the medium into relativistic motion in the forward direction to create a powerful relativistic shock.  In the latter scenario, relativistic shocks can accelerate ambient electrons (as well as protons), thereby injecting a power-law electron spectra $Q_{e}(E) \\propto E^{-\\alpha} $ with $\\alpha \\geq 2$ in the downstream region.  Synchrotron radiation from these electrons could then explain \\citep{da02} the broad-band nonthermal radio, optical, and X-ray emission from hot spots and knots in the extended X-ray jets of extragalactic jet sources observed with the {\\it Chandra Observatory}.  The idea that the observed X-ray jets might be due to propagation of a powerful beam of gamma rays has also been considered by \\citet{nsak02} \\citep{dl01}, who suggested a ``non-acceleration\" scenario, where the observed X-rays could be due to synchrotron radiation from electrons formed in the electromagnetic cascade initiated by ultra-high energy gamma rays. Our calculations show that although some contribution to the observed X-ray fluxes from the cascade electrons is not excluded, the spectral features of the observed radiation, in particular, X-ray energy spectral indices $\\alpha_X \\geq 1$, might be difficult to explain in a pure cascade scenario. Moreover, interpretation of the radio and optical fluxes from the same X-ray knots apparently makes unavoidable the requirement of in-situ acceleration of electrons in the X-ray knots.  Given the total energy released by the neutrons and gamma rays per single flare, the energetics of these kpc-scale knots could be explained by superposition of jet activity on time scales up to thousands of yrs (since $1 \\,\\rm kpc$ translates to $\\simeq 3\\times 10^3 \\,\\rm yr$ of light travel time).  Thus, the bright knots might represent the observable signatures of the past history of activity of the central engine of the blazar when powerful acceleration of UHE cosmic rays had occurred in the inner jets of the source.  Considering now the BL Lac objects, we note that in the case of weak or negligible external radiation field components, any significant transport of the inner jet energy by ultra-high energy gamma-rays in those objects is practically absent. This is because gamma-rays are not produced through $n\\gamma$ interactions of the neutrons outside the compact blob, whereas the ultra-high energy gamma-rays produced in $p\\gamma$ collisions inside the blob will be absorbed due to $\\gamma\\gamma$ collisions with the same internal synchrotron target photons (recall that $\\sigma_{\\gamma\\gamma} \\gg \\sigma_{p\\gamma}$). Nevertheless, up to $\\sim 10\\%$ of the energy injected into the inner jet in relativistic protons with $E>1\\, \\rm PeV$ could still be taken out of the blob by neutrons with $E\\gtrsim 10^{17} \\,\\rm eV$ due to $p\\gamma$ collisions with synchrotron photons while the blob remains compact. Because of the latter limitations, and because BL Lac sources are generally much less powerful than FSRQs, the total energy that the neutron beam can extract from the accelerated protons is much less for BL Lacs than for FSRQs. Moreover, because of the smaller sizes and magnetic fields inferred for BL Lac objects, the level of neutron production and the maximum energy of accelerated protons is reduced in the former class of sources. Thus the neutral beam power is much less in BL Lac objects than in FSRQs.  \\subsection{Explanation of Observed Features of Radio-Loud Active Galactic Nuclei }  As a consequence of processes occurring in the inner jets of blazars, we predict that powerful collimated neutral beams that extend as far as hundreds of kpc from the nucleus are ejected in the axial jet directions in high luminosity FSRQs. The linearity and the spatial extent of the ejecta is a consequence of the decay properties of the neutrals.  This process can explain the origin of X-ray jets detected by {\\it Chandra}, which appear straight on scales of up to Mpc, such as the jet in Pictor A \\citep{wys01}.  The linearity of the radio jets in Cygnus A \\citep{pdc84} may also be a consequence of energetic neutral beams powering the lobes of powerful FR II galaxies. This interpretation avoids firehose instabilities in models invoking magnetic fields to collimate the extended FR II radio jets, or problems associated with jet quenching, as discussed in Section 4.2.  Our model predicts that lower luminosity blazars, in particular, the lineless BL Lac objects, have orders-of-magnitude weaker neutral beams than FSRQs. The maximum energies of the relativistic neutrons are also about an order-of-magnitude less in Mrk 501-like BL Lac objects than in 3C 279-like FSRQs. Consequently the X-ray jets driven by the high-energy neutron beams will be generally fainter (despite their relative proximity) and shorter in BL Lac objects. This could explain the faint kpc-scale X-ray jets detected from objects like 3C 371 \\citep{3C371}, which is an off-axis BL Lac object \\citep{mil75}.  We adopt the point of view, following \\citet{up95}, that FR I and FR II galaxies are the parent populations of BL Lac objects and FSRQs, respectively.  The less powerful beams of neutrons that occupy smaller spatial scales in BL Lac objects compared to FSRQs conforms to the measured relative power requirements in FR I and FR II galaxies \\citep{fr74}. If FR I/BL Lacs have smaller Lorentz factors than FR II/FSRQs, as also inferred from observations of these sources \\citep{up95}, then the weak, broad twin-jet morphologies in FR I galaxies can be qualitatively understood.  We point out that even though the inner parsec regions of a BL Lac are much less dense than the BLRs in FSRQs in view of their relative emission line strengths, the jet quenching problem (Section 4.2) can still be severe for BL Lac objects because of their weaker jets.  The transport of energy by neutral particles to the extended jet will be enhanced by an intense external radiation field in the inner regions of the jet.  The total power into the jet depends foremost on the accretion rate, which depends on the amount of gas and dust available to fuel a supermassive black hole. Thus our model is consistent with the evolutionary scenario proposed by \\citet{bd02} and \\citet{ce02} to explain the spectral sequence of FSRQs, and low- and high-frequency peaked BL Lac objects. According to this picture, FSRQs evolve into BL Lac objects due to a reduction in the accretion rate of the surrounding dust and gas that fuels the central black hole. (Note, however, that the jet power is derived primarily from mass accretion in the scenario of \\citet{bd02} and from the spin energy of the black hole in the scenario of \\citet{ce02}.) \\citet{mt02} argue that this picture accounts for the correlation between the radio jet and accretion power, where the latter is inferred from emission line strengths. High-energy neutrino observations will be crucial to test this scenario, as neutral beams of high energy neutrons and gamma-rays will be made in association with high-energy neutrinos.  Finally we speculate that if ultraluminous infrared galaxies and quasars evolve into radio-loud AGNs \\citep{san88}, then some ultraluminous infrared galaxies could harbor buried jet sources that will be revealed through their high-energy neutrino emission.   \\subsection{Summary}   Production of narrow beams of ultra-high energy neutrons with $E\\geq 10^{17} \\,\\rm eV$ becomes possible if relativistic compact jets of blazars accelerate cosmic rays to these energies. Such neutral relativistic particle beams can be especially powerful in FSRQs, where the neutrons are efficiently produced in $p\\gamma$ interactions with photons of the external quasi-isotropic radiation field on timescale of years (since the broad-line region size scale $R_{BLR}\\sim 0.1$-1 pc).  In radio-loud jet sources, the beam could contain a significant fraction of ultra-high energy gamma-rays produced in the BLR through $n\\gamma$ photomeson interactions by neutrons which escape from the compact relativistic plasma blob that forms the inner jet.  These charge-neutral beams of UHE neutrons and gamma-rays (including also neutrinos which, however, do not contribute to the process of energy transport) do not interact with the extragalactic medium until either the neutrons decay or the gamma rays collide with soft photons on scales from kpc to Mpc. The main soft-photon target for $\\gamma \\gamma$ collisions is provided by the CMB radiation and, in some sufficiently powerful sources, also by low-frequency synchrotron radiation self-produced along the jet \\citep{nsak02}).   The weaker neutral beams in FR I radio galaxies and BL Lac objects, compared to FR II radio galaxies and FSRQs, could explain the morphological differences between the two classes of radio galaxies. This scenario assumes that the inner jets of blazars are effective accelerators of high-energy cosmic rays. The limits on acceleration that we consider do not preclude the acceleration of cosmic rays to energies $\\gtrsim 10^{20}\\,\\rm eV$ in blazars, especially in the more luminous FSRQs.  This mechanism of relativistic energy transport from the central engine of AGN to large distances avoids problems of jet quenching in the central gas-rich regions of AGN, problems of adiabatic and radiative losses, and problems of long-term confinement and stability (on scales up to $10^{6} \\,\\rm yr$) of the jet against different plasma processes. Beam instabilities can, however, provide an effective method to transfer the energy of ultra-relativistic protons emerging from the neutron beam into the surrounding medium, and driving relativistic shock which ultimately accelerates nonthermal electrons to radiate observable synchrotron and Compton emission. Given also the large amount of energy that such a beam may contain, which can be of the order of a few percent of the inner jet power, our results suggest that the detection by the {\\it Chandra Observatory} of the narrow X-ray jets that remain straight on scales beyond several hundreds of kpc provides observational evidence that the relativistic inner jets of blazars, first of all the flat spectrum radio quasars, are powerful accelerators of UHE cosmic rays. Detection of high-energy neutrinos from FSRQs with IceCube or a Northern Hemisphere high-energy neutrino telescope will provide strong support for this scenario."
    },
    "0209/astro-ph0209007_arXiv.txt": {
        "abstract": "\\noindent{\\it  We first present a short overview of X-ray probes of the black hole region of active galaxies (AGN) and then concentrate on the X-ray search for supermassive black holes (SMBHs) in optically non-active galaxies.  The first part focuses on recent results from the X-ray observatories {\\sl Chandra} and {\\sl XMM-Newton} which detected a wealth of new spectral features which originate in the nuclear region of AGN.  In the last few years, giant-amplitude, non-recurrent X-ray flares have been observed from several {\\em non}-active galaxies. All of them share similar properties, namely: extreme X-ray softness in outburst, huge peak luminosity (up to $\\sim 10^{44}$ erg/s), and the absence of optical signs of Seyfert activity. Tidal disruption of a star by a supermassive black hole is the favored explanation of these unusual events. The second part provides a review of the initial X-ray observations, follow-up studies, and the relevant aspects of tidal disruption models studied in the literature. } ",
        "introduction": "\\subsection{The search for SMBHs at the centers of galaxies}  The study of supermassive black holes and their cosmological evolution is of great interest for a broad range of astrophysical topics including facets of galaxy formation and general relativity. In the last few decades, a number of different methods were developed to search for supermassive black holes (SMBHs) in external galaxies. Their detection in large numbers would clarify our understanding of the early phases of the evolution of galaxies. In {\\em active} galactic nuclei SMBHs are now generally believed to be the prime mover of the non-stellar activity. X-ray observations in the near future are expected to offer the opportunity of detecting some of the distinctive features of strong field gravity, thereby also providing the ultimate proof for the existence of black holes in AGN.  There is now strong evidence for the presence of massive dark objects at the centers of many galaxies. Does this hold for {\\em all} galaxies ? If so, why are some SMBHs `dark' ? Questions of particular interest in the context of galaxy/AGN evolution are: When and how did the first SMBHs form and how do they evolve ? What fraction of galaxies have passed through an active phase, and how many now have non-accreting and hence unseen supermassive black holes at their centers (e.g., Lynden-Bell 1969, Rees 1989)?  Several approaches were followed to study these questions. Much effort has concentrated on the determination of central object masses from measurements of the {\\sl dynamics of stars and gas} in the nuclei of nearby galaxies. Earlier (ground-based) evidence for central quiescent dark masses in galaxies (Kormendy \\& Richstone 1995) has been strengthened by recent HST results (see Kormendy \\& Gebhardt 2001 for a review).  A quite accurate determination of black hole mass was enabled by the detection of water vapor maser emission from the mildly active galaxy NGC\\,4258 (Miyoshi et al. 1995). The water masers, whose motion can be precisely mapped with VLBI, are located in a very compact disk in Keplerian rotation around the central SMBH. The fortunate geometry of the disk, nearly edge-on, allows to obtain the BH mass with high accuracy: $M_{\\rm BH} = 3.6\\,10^{7}$ M$_{\\odot}$ (Neufeld \\& Maloney 1995, Greenhill et al. 1995).  Still closer to the SMBH, in {\\em active} galaxies with with broad line region (BLR hereafter), the technique of BLR reverberation mapping (e.g., Peterson 2001) provides a powerful tool to estimate the BH mass via the clouds' distance from the center and their velocity field (e.g., Peterson \\& Wandel 2000, Ferrarese et al. 2001).   \\subsection{X-ray probes of the black hole region of AGN}  Whereas the dynamics of stars and gas probe rather large distances from the SMBH, high-energy {\\sl X-ray emission} originates from the immediate vicinity of the black hole. In {\\em active} galaxies, excellent evidence for the presence of SMBHs is provided by the detection of luminous hard power-law like X-ray emission, rapid variability, and the discovery of evidence for relativistic effects in the iron-K line profile. X-ray observations currently provide the most powerful way to explore the black hole region of AGN.  X-rays at the centers of AGN arise in the accretion-disk -- corona system (e.g, Mushotzky et al. 1993, Svensson et al. 1994, Collin et al. 2000, and references therein). On larger scales, but still within the central region, X-rays might be emitted by a hot intercloud medium at distances of the broad or narrow-line region (e.g., Elvis et al. 1990).  The X-rays which originate from the accretion-disk region are reprocessed in form of absorption and partial re-emission (e.g., George \\& Fabian 1991, Netzer 1993, Krolik \\& Kriss 1995, Collin-Souffrin et al. 1996, Komossa \\& Fink 1997b) as they make their way out of the nucleus. The reprocessing bears the disadvantage of veiling the {\\em intrinsic} X-ray spectral shape, and the spectral disentanglement of many different potentially contributing components is not always easy. However, reprocessing also offers the unique chance to study the physical conditions and dynamical states of the reprocessing material (see Komossa 2001 for a review), like: the outer parts of the accretion disk; the ionized absorber; the torus, which plays an important role in AGN unification schemes (Antonucci 1993); and the BLR and NLR. Detailed modeling of the reprocessor(s) is also necessary to recover the shape and properties of the {\\em intrinsic} X-ray spectrum.   \\begin{figure}[ht] \\caption{{Sketch of the central region of Seyfert galaxies. The black hole and accretion disk region is surrounded by two systems of gas clouds, the broad line region (BLR) and narrow-line region (NLR). These show up by their characteristic line emission in optical spectra of AGN, and their presence is usually used to identify and classify AGN. The molecular torus, and variants of it, are thought to play an important role in unification schemes of Seyfert galaxies by blocking the direct view on the BLR for certain viewing directions of the observer (Antonucci 1993). Somewhere outside the bulk of the BLR, a relatively recently discovered component of the active nucleus is located, the so-called `warm' or ionized absorber (WA). \\newline A modification of this picture was recently proposed by Elvis (2000, 2001). In his model, the BLR clouds arise from a flow of gas which rises vertically from a narrow range of radii from the accretion disk. The flow then bends and forms a conical wind moving radially outwards. The BLR clouds are identified with the cool phase of this two-phase medium. Warm absorbers appear if we view the continuum source through the wind (Elvis 2000, see his Fig.\\,1)} } \\end{figure}   Recent progress has been made based on the improved spectral resolution of the new generation of X-ray observatories, {\\sl Chandra} and {\\sl XMM-Newton}. Both missions have imaging detectors and grating spectrometers aboard. Their energy sensitivity bandpass covers $\\sim$(0.1--10) keV. Below, a short review of results from these observatories is given, starting at relatively large distances from the SMBH (NLR), and then moving further inward (warm absorber and accretion disk region).   \\paragraph{X-ray emission lines, X-ray narrow-line region.} The detection of a high-temperature, narrow-line, X-ray emitting plasma in NGC\\,4151 was reported by Ogle et al. (2000), confirming earlier evidence for extended X-ray emission from this galaxy (Elvis et al. 1983). The X-ray gas is spatially coincident with the NLR and extended narrow-line region. For the first time, numerous emission-lines were detected in the X-ray spectrum of NGC\\,4151 with the HETG (High Energy Transmission Grating Spectrometer) aboard {\\sl Chandra}.  The X-ray emission lines detected in the spectrum of NGC\\,4151 and several Seyfert\\,2 galaxies (e.g., Mrk\\,3, NGC\\,1068), contain important information on the physical conditions in the line-emitting medium, like temperature, density, and the main gas excitation/ionization mechanism - photoionization or collisional ionization. Of particular importance in determining the main power mechanism of the lines, are the Helium-like triplets (Gabriel \\& Jordan 1969; see our Fig.\\ref{wa}), the widths of the radiative recombination continua, and the strengths of the Fe-L complexes (e.g., Liedahl et al. 1990).  \\begin{figure}[t] \\hspace*{0.3cm} \\psfig{file=komossa_f2.ps,width=9.5cm,clip=} \\caption{{\\sl Chandra} LETGS X-ray spectrum of NGC\\,5548 (Kaastra et al. 2000). The inset shows a zoom of the OVII triplet to which a resonance line, two intercombination lines (unresolved), and a dipole-forbidden line contribute.} \\label{wa} \\end{figure}  \\paragraph{X-ray absorption lines, ionized absorber.} With {\\sl ROSAT}, the signatures of so-called `warm' absorbers, absorption edges of highly ionized oxygen ions at $E_{\\rm OVII}=0.74$ keV and $E_{\\rm OVIII}=0.87$ keV, were first detected in MCG$-$6-30-15 (Nandra \\& Pounds 1992), following earlier {\\sl Einstein} evidence for highly ionized absorbing material in AGN (Halpern 1984). Detailed studies of many other AGN followed, and the signatures of warm absorbers have now been seen in about 50\\% of the well-studied Seyfert galaxies (see Komossa 1999 for a review). First constraints place the bulk of the ionized material outside the BLR, and depending on its covering factor and location, the warm absorber may be one of the most massive components of the active nucleus. Evidence for ionized absorption was also found in some very high-redshift quasars, starting with observations of PKS\\,2351-154 (Schartel et al. 1997). Some (but not all) warm absorbers were suggested to contain dust, based on otherwise contradictory optical--X-ray observations (e.g., Brandt et al. 1996, Komossa \\& Fink 1997b, Komossa \\& Bade 1998). The first possible detection of Fe-L dust features in the X-ray spectra of MCG$-$6-30-15 and Mrk\\,766 was recently reported by Lee et al. (2001) and Lee (2001).  The high-resolution spectrum of the Seyfert galaxy NGC\\,5548, obtained with the {\\sl Chandra} Low Energy Transmission Grating Spectrometer (LETGS), shows many narrow absorption lines of highly ionized metal ions of ogygen, neon, iron, etc. (Fig. \\ref{wa}), confirming the presence of a warm absorber in this galaxy (Kaastra et al. 2000). Similar signatures of ionized material have been detected with {\\sl Chandra} and {\\sl XMM-Newton} in several AGN, including NGC\\,3783 (Kaspi et al. 2000), IRAS 13349+2438 (Sako et al. 2001), NGC\\,4051 (Collinge et al. 2001), Mrk\\,509 (Yaqoob et al. 2002), and MCG\\,$-$6-30-15 (see next paragraph). First results show that the ionized absorption is complex with a range in ionization states.  Assuming that the ionized absorber outflow is driven by radiation pressure of the central continuum sources, Morales \\& Fabian (2001) demonstrated that observations can then be used for an estimate of black holes masses in AGN. They derive masses of $M_{\\rm BH} \\simeq 10^{6.5-7}$ M$_{\\odot}$ for the galaxies of their sample.  \\begin{figure}[t] \\hspace*{0.6cm} \\psfig{file=komossa_f3.ps,width=9.0cm,clip=} \\caption{Soft X-ray spectrum of an AGN, plotted as log flux [a.u.] versus log Energy [keV]. The dashed line shows the input continuum spectrum. The thick solid line gives the spectrum after passage of a warm absorber. The calculation was carried out with the photoionization code {\\sl Cloudy} (Ferland 1993). Input parameters (ionization parameter $U$, column density $N_{\\rm w}$) were chosen similar to those obtained from a Beppo-SAX observation of Mrk\\,766 (Matt et al. 2000) except that $U$ was lowered. Only the absorption edges are shown, and are labeled in the graph (absorption lines were omitted). If this theoretical absorption spectrum is now re-fit, without knowledge of the intrinsic continuum (its shape and level), two fundamentally different solutions are possible: (i) a high-level steep continuum in which case the absorption-solution is recovered, or (ii) a low-level flat continuum (the horizontal thin solid line in the graph) in which case the presence of a strong soft excess (at lower energies than the OVII edge) plus some very broad emission lines are inferred. } \\label{m766} \\end{figure}  \\paragraph{Accretion-disk region, Fe-K line.}  The most direct probe of the black hole region, and particularly, of special and general relativistic effects, is emission from the inner part of the accretion disk (see Fabian 2001 for a review). Tanaka et al. (1995) reported the detection of a broadened FeK$\\alpha$ line in MCG$-$6-30-15. The line profile is well explained by the special relativistic effects of beaming and transverse Doppler effect, and the general relativistic effect of gravitational redshift. Depending on details of modeling the continuum, broad-winged Fe lines  may also be present in several XMM spectra of AGN (Nandra 2001). At certain times, the red wing of MCG$-$6-30-15 is very broad, extending down to very soft energies (Wilms et al. 2001).  With {\\sl XMM} it has also become clearer that the Fe-line profiles are complex, and the line has several sides of formation, including likely the BLR (NGC\\,5548), the torus (NGC\\,3783, Mrk\\,205), the X-ray ionization cone of NGC\\,1068, and a contribution from the outer parts of the accretion disk (MCG$-$6-30-15).   \\paragraph{The case of MCG\\,$-$6-30-15, X-ray spectral complexity.} Whereas the signature of an ionized absorber in the form of narrow absorption lines was detected in this galaxy with both, XMM-Newton (Branduardi-Raymont et al. 2001) and {\\sl Chandra} (Lee et al. 2001), new interpretations of some of the spectral features were put forward: Branduardi-Raymont et al. suggested that the dominant soft X-ray features, so far interpreted as metal absorption edges of the warm absorber, can be better understood in terms of relativistically broadened  emission lines which originate in the accretion disk. On the other hand, Matsumoto \\& Inoue (2001) noted that the {\\sl ASCA}-detected broad wing of the iron K line - so far interpreted in terms of relativistic broadening due to the line's origin in the inner parts of the accretion disk - could be successfully modeled by invoking a two-component warm absorber.  Fig. \\ref{m766} summarizes and visualizes one of the basic underlying ideas in the discussion of emission versus absorption features at soft X-ray energies (for many additional details see Branduardi-Raymont et al. 2001 and Lee et al. 2001): Depending on where the continuum is placed in Fig. \\ref{m766}, one would either infer the presence of {\\sl a huge soft excess plus broad emission lines}, or a powerlaw spectrum modified by {\\sl absorption edges}. We do not discuss these ideas further here, except for noting that a disk-line interpretation of the bulk of the soft X-ray features of MCG$-$6-30-15 would leave the puzzle of the discrepant optical and X-ray absorption of this galaxy unanswered, which could be solved by introducing a dusty warm absorber (Reynolds et al. 1997).  Deep high-resolution X-ray observations of, and a search for, variability of the spectral features will be a very important next step in disentangling all components of the complex X-ray spectrum of this galaxy.  \\subsection{The X-ray search for SMBHs in ULIRGs and \\newline LINERs}  It is now generally believed that {\\em active} galactic nuclei (AGN) are powered by accretion onto SMBHs. The search for heavily obscured SMBHs in ultraluminous infrared galaxies (ULIRGs), and for low-luminosity AGN (LLAGN) in LINER galaxies is another interesting topic. It will only be briefly touched here, since the emphasis of this review will be on recent evidence for SMBHs in non-active `normal' galaxies (next Section).  \\subsubsection{ULIRGs}  ULIRGs, characterized by their huge power-output in the infrared which exceeds 10$^{12}\\,L_\\odot$ (Sanders \\& Mirabel 1996), are powered by massive starbursts or SMBHs. The discussion, which one actually dominates received a lot of attention in recent years (e.g., Joseph 1999, Sanders 1999). In particular, only a small fraction of ULIRGs show AGN signatures in their optical and infrared spectra.  Do the remaining ones nevertheless harbor AGN ? X-ray variability and luminous hard X-ray emission are excellent indicators of obscured AGN activity.  With a redshift $z=0.024$ and a far-infrared luminosity of $\\sim 10^{12} L_{\\odot}$, NGC\\,6240 is one of the nearest members of the class of ULIRGs. Whereas X-rays from distant Hyperluminous IR galaxies, HyLIRGs, were not detected by Wilman et al. (1999), and the ULIRGs in the study of Rigopoulou et al. (1996) were X-ray weak, NGC\\,6240 turned out to be exceptionally X-ray luminous. Starburst-driven superwinds are the most likely interpretation of the extended emission (see Schulz \\& Komossa 1999 for alternatives), albeit being pushed to their limits to explain the huge power output (Schulz et al. 1998). The hard spectral component present in the {\\sl ROSAT} energy band was interpreted as scattered emission from an obscured AGN (Schulz et al. 1998, Komossa et al. 1998) which shows up more clearly at higher energies, up to 100\\,keV (e.g., Vignati et al. 1999, Mitsuda 1995, Ikebe et al. 2000). The intrinsically luminous AGN ($L_{\\rm x} \\approx 10^{44}$ erg/s), can account for at least a substantial fraction of the FIR power output of NGC\\,6240.  Using {\\sl ASCA}, Nakagawa et al. (1999) studied the hard X-ray properties of a sample of 10 ULIRGs. Among these, 50\\% have hard X-ray detections. The most stringent upper limit for the presence of any hard X-ray emission was reported for Arp\\,220. The possibility that it is a Compton-thick source cannot be excluded, though.    \\subsubsection{LINERs} LINER (Low-Ionization Nuclear Emission-Line Region)  galaxies are characterized by their optical emission line spectrum which shows a lower degree of ionization than AGN. Their major power source and line excitation mechanism have been a subject of lively debate ever since their discovery. LINERs manifest the most common type of activity in the local universe. If powered by accretion, they probably represent the low-luminosity end of the quasar phenomenon, and their presence has relevance to, e.g., the evolution of quasars, the faint end of the Seyfert luminosity function, the soft X-ray background, and the presence of SMBHs in nearby galaxies.  The X-ray properties of LINERs are inhomogeneous. Spectra of a sample of objects studied by Komossa et al. (1999) are best described by a composition of soft thermal emission and a powerlaw with varying relative contributions of the two components from object to object. Several studies of individual objects are consistent with these results (e.g., Mushotzky 1982, Koratkar et al. 1995, Cui et al. 1997, Ptak et al. 1999, Roberts et al. 1999). X-ray luminosities are in the range $\\sim$10$^{38-41}$ erg/s; below those typically observed for Seyfert galaxies. The general absence of short-time scale (hours-weeks) X-ray variability (Ptak et al. 1998, Komossa et al. 1999) is consistent with the suggestion that LINERs accrete in the advection-dominated mode (e.g, Yi \\& Boughn 1998, 1999, and references therein). However, clear positive X-ray detections of LLAGNs in LINERs are still rare. One potential problem problem is to distinguish powerlaw emission of the X-ray binary population of the host galaxy from that of a genuine LLAGN. First {\\sl Chandra} results on LINERs show that few, if any, are obscured by absorbers of high column density (Ho et al. 2001). Four out of eight LINERs of that study possess compact nuclear cores, consistent with AGNs.    \\subsection{The X-ray search for SMBHs in {\\itshape non-active} (`normal') galaxies, and tidal disruption flares as probes }  How can we find {\\em dormant} SMBHs in {\\em non-active} galaxies ? Lidskii \\& Ozernoi (1979) and Rees (1988) suggested to use the flare of electromagnetic radiation produced when a star is tidally disrupted and accreted by a SMBH as a means to detect SMBHs in nearby {\\em non-active} galaxies.  A star on a near-radial `loss-cone' orbit gets tidally disrupted once the tidal gravitational forces exerted by the black hole exceed the self-gravitational force of the star (e.g., Hills 1975, Lidskii \\& Ozernoi 1979, Diener et al. 1997). The tidal radius is given by \\begin{equation} r_{\\rm t} \\simeq 7\\,10^{12}\\,({M_{\\rm BH}\\over {10^{6} M_\\odot}})^{1 \\over 3} ({M_{\\rm *}\\over M_\\odot})^{-{1 \\over 3}} {r_* \\over r_\\odot}~{\\rm cm}\\,. \\end{equation} The star is first heavily deformed, then disrupted. About 50\\%--90\\% of the gaseous debris becomes unbound and is lost from the system (e.g., Young et al. 1977, Ayal et al. 2000). The rest will eventually be accreted by the black hole (e.g., Cannizzo et al. 1990, Loeb \\& Ulmer 1997). The stellar material, first spread over a number of orbits, quickly circularizes (e.g., Rees 1988, Cannizzo et al. 1990) due to the action of strong shocks when the most tightly bound debris interacts with other parts of the stream (e.g., Kim et al. 1999). Most orbital periods will then be within a few times the period of the most tightly bound matter (e.g., Evans \\& Kochanek 1989; see also Nolthenius \\& Katz 1982, Luminet \\& Marck 1985).  Explicit predictions of the emitted spectrum and luminosity during the disruption process and the start of the accretion phase are still rare (see Sect.\\,3 for details). The emission is likely peaked in the soft X-ray or UV portion of the spectrum, initially (e.g., Rees 1988, Kim et al. 1999, Cannizzo et al. 1990; see also Sembay \\& West 1993).   \\begin{table*}[ht] \\caption{Summary of the X-ray and optical properties of the flaring normal galaxies during outburst. $z$ gives the redshift, $T_{\\rm bb}$ is the black body temperature derived from a black body fit to the X-ray high-state spectrum (cold absorption was fixed to the Galactic value in the direction of the individual galaxies), `no emi.' means: no optical emission lines were detected. $L_{\\rm x,bb}$ gives the intrinsic luminosity in the (0.1--2.4) keV band, based on the black body fit. This is a lower limit to the actual peak luminosity, since we most likely have not caught the sources exactly at maximum light, since the spectrum may extend into the EUV, and since it was conservatively assumed that no additional X-ray absorption occurs intrinsic to the galaxies. } \\vskip0.1cm \\begin{center} \\begin{tabular}{ccccc} \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} galaxy name & $z$ & opt. type & $kT_{\\rm bb}$ [keV] & $L_{\\rm x,bb}$ [erg/s]  \\\\ \\noalign{\\smallskip} \\hline \\hline \\noalign{\\smallskip} NGC\\,5905 & 0.011 & HII & 0.06 & 3 10$^{42}$$^*$  \\\\ \\noalign{\\smallskip} &       &                 &                 \\\\ RXJ1242$-$1119  & 0.050 & no emi. & 0.06 & 9 10$^{43}$~  \\\\ &       &      &                  \\\\ \\noalign{\\smallskip} RXJ1624+7554 &  0.064 & no emi. & 0.097 & $\\sim$ 10$^{44}$~  \\\\ &       &      &                  \\\\ \\noalign{\\smallskip} RXJ1420+5334 &  0.147 & no emi. & 0.04 & 8 10$^{43}$~ \\\\ &       &      &                  \\\\ \\noalign{\\smallskip} RXJ1331$-$3243  & 0.051 & no emi. &  &  \\\\ \\noalign{\\smallskip} \\noalign{\\smallskip} \\hline \\end{tabular} \\end{center} \\noindent{\\scriptsize $^{*}$Mean luminosity during the outburst; since the flux varied by a factor $\\sim$3 during the observation, the peak luminosity is higher. } \\end{table*} ",
        "conclusions": ""
    },
    "0209/astro-ph0209312_arXiv.txt": {
        "abstract": "Whereas the total energy in zero-point fluctuations of the particle physics vacuum gives rise to the cosmological constant problem, differences in the vacuum give rise to real physical phenomena, such as the Casimir effect. Hence we consider the zero-point energy bound between two parallel conducting plates --- proxy for a solid slice of cosmological constant --- as a convenient laboratory in which to investigate the gravitation and inertia of vacuum energy.  We calculate the Casimir effect in a weak gravitational field, obtaining corrections to the vacuum stress-energy and attractive force  on the plates due to the curvature of spacetime. These results suggest that if the cosmological constant is due to zero-point energy then it is  susceptible to fluctuations induced by gravitational sources. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209254_arXiv.txt": {
        "abstract": "The host galaxies of powerful radio sources are ideal laboratories to study active galactic nuclei (AGN).  The galaxies themselves are among the most massive systems in the universe, and are believed to harbor supermassive black holes (SMBH). If large galaxies are formed in a hierarchical way by multiple merger events, radio galaxies at low redshift represent the end-products of this process. However, it is not clear why some of these massive ellipticals have associated radio emission, while others do not. Both are thought to contain SMBHs, with masses proportional to the total luminous mass in the bulge. It either implies every SMBH has recurrent radio-loud phases, and the radio-quiet galaxies happen to be in the ``low'' state, or that the radio galaxy nuclei are physically different from radio-quiet ones, i.e. by having a more massive SMBH for a given bulge mass.  Here we present the first results from our adaptive optics imaging and spectroscopy pilot program on three nearby powerful radio galaxies. Initiating a larger, more systematic AO survey of radio galaxies (preferentially with Laser Guide Star equipped AO systems) has the potential of furthering our understanding of the physical properties of radio sources, their triggering, and their subsequent evolution. ",
        "introduction": "\\label{sect:intro}  %  Powerful radio galaxies provide a convenient way to investigate the evolution of very massive galaxies over a large range in redshift. Color properties of their host galaxies have been found to be remarkably well represented by passively evolving stellar systems, with typical masses of $5 - 10$ L$_\\star$\\cite{lilly88,vanbreugel98}. Comparison with high redshift field galaxies\\cite{cowie97} confirms that radio galaxies form indeed the high-luminosity envelope. Inferred formation redshifts for these systems have been as high\\cite{spinrad97} as $z_f > 10$, implying that by a redshift of 1 to 2, emission from these galaxies is dominated by an old ($>5$ Gyr) stellar population. It is this homogeneity in population (and emission) properties that makes the near-IR Hubble K-z relation so tight.  Radio galaxy morphologies, when imaged at rest-frame optical wavelengths, often show spectacular, clumpy structures aligned with the radio source axes. This ``alignment effect'' appeared at odds with the passive evolution inferred from the near-IR K-z diagram. Its exact nature has remained unclear and evidence has been found for scattered light from hidden quasar-like AGN, nebular re-combination continuum and even jet-induced star formation\\cite{mccarthy93}.  To investigate the morphological evolution of the stellar populations of radio galaxies with redshift, it is therefore of interest to obtain high spatial resolution at {\\it infrared} wavelengths, where AGN-related emission is fainter and the old stellar population brighter. This effective isolation of just the stellar emission component also provides a useful baseline against which we can interpret the optical HST data. Again, given the rather uniform stellar population, any color-deviations will stand out in a color plot based on combining HST optical and Keck AO near-IR data (cf. Fig.~\\ref{colorMaps}). Features, such as dust-lanes, or compact regions of enhanced starformation can be clearly detected against the uniform backdrop of the underlying galaxy. Furthermore, the high resolution near-IR images allows us to extract physically meaningful luminosity profiles, which are much less affected by obscuring dust or AGN related non-stellar emission. Using HST on a large sample of (radio quiet) galaxies, Faber et al.\\cite{faber97} found the shape of the inner galaxy profile to correlate with various physical quantities such as absolute luminosity and central velocity dispersion; provided the profile can be fitted with a ``Nuker'' profile\\cite{lauer95,byun96}. Up till now, only with HST was the necessary spatial resolution attainable. With the Keck-AO system, this can be improved upon by at least a factor of 4 (cf. Fig.~\\ref{3c452nuker}). This allows us to intercompare more distant radio galaxy hosts to ``normal'' ellipticals, which could provide additional clues about why some galaxies are radio emitters and others are not.  In principle high resolution spectroscopy can not only provide us with diagnostic line ratios which constrain ionization mechanisms (i.e., starbursts vs. AGN\\cite{hill99,vanzi98,black87}), and absorption properties of circumnuclear features\\cite{rudy99,thornton99,rhee00}, but it can also provide us with accurate measurements of nuclear stellar velocity dispersions using the $^{12}$CO near-IR bandhead at 2.29$\\mu$m restframe\\cite{gaffney95,boeker99}. Given the very high spatial (and spectroscopic) resolution of the observations, these will provide nuclear black-hole (BH) mass estimates (analogous to B\\\"oker \\cite{boeker99} et al.). Values derived this way for our sample can then be compared directly to recent results on the velocity---BH-mass correlation for lower redshift galaxies\\cite{gebhardt00,ferrarese00}. However, a combination of instrumental throughput issues beyond $\\sim2.4$ $\\mu$m, the anisoplanatism inherent to off-axis correction, and just plain bad weather prevented us from achieving our planned spectroscopic goals.  Either way, both the imaging and spectroscopy parts of the program have yielded insights into the makeup of the stellar systems of the 3 powerful radio galaxies and their place in the elliptical galaxy taxonomy. This illustrates the potential AO can offer us in understanding the onset and subsequent evolution of powerful nuclear radio sources, provided this pilot project is carried out on a much larger sample and in a more complete fashion, ideally with a LGS equipped AO system. ",
        "conclusions": "We demonstrated the potential AO observations have to radio galaxy host studies. It equals, and complements, HST's optical imaging in terms of resolution, extending the realm of detailed nuclear environment imaging into the near-IR. The spectroscopic setup we used (NIRSPEC + KECK) was not ideally suited for our project, however. When working close to the diffraction limit, even 10m class telescopes are photon starved.  Based on this pilot study, AO imaging (and perhaps AO spectroscopy) on a large, and systematically set up sample can significantly further our knowledge of these large galaxies and their massive black holes."
    },
    "0209/astro-ph0209062_arXiv.txt": {
        "abstract": "{\\small SIMBOL--X is a high energy ``mini\" satellite class mission that is proposed by a French-Italian-English collaboration for a launch in 2009. SIMBOL--X is making use of a classical X--ray mirror, of $\\sim$~600~\\cm2 maximum effective area, with a 30~m focal length in order to cover energies up to several tens of keV. This focal length will be achieved through the use of two spacecrafts in a formation flying configuration. This will give to SIMBOL--X unprecedented spatial resolution (20\\arcsec HEW) and sensitivity in the hard X--ray range. By its coverage, from 0.5 to 70~keV, and sensitivity, SIMBOL--X will be an excellent instrument for the study of high energy processes in a large number of sources, as in particular accreting black-holes, extragalactic jets and AGNs.} ",
        "introduction": "The study of the non thermal component in high energy astrophysics sources is presently hampered by the large gap in spatial resolution and sensitivity between the X--ray and $\\gamma$--ray domains. Below $\\sim$~10~keV, astrophysics missions like XMM--Newton and Chandra are using X--ray mirrors based on grazing incidence reflection properties. This allows to have an extremely good spatial resolution, down to 0.5\\arcsec for Chandra, and a good signal to noise thanks to the focussing of the X--rays onto a small detector surface. This technique has however an energy limitation at $\\sim$~10 keV due to the maximum focal length that can fit in a single spacecraft. Hard X--ray and $\\gamma$--ray imaging instruments, such as those on INTEGRAL to be launched in Oct.~2002 are thus using a different technique, that of coded masks. This non focussing technique has intrinsically a much lower signal to noise ratio than that of a focussing instrument, and does not allow to reach spatial resolutions better than $\\sim$~10 arc minutes. This results for example in roughly 2 orders of magnitude of difference in point source sensitivity between X--ray and $\\gamma$--ray telescopes.  This transition of techniques unfortunately happens roughly at the energy above which the identification of a non thermal component is unambiguous with respect to thermal emission. This obviously strongly limits the interpretation of the high quality X--ray measurements, and particularly that related to the acceleration of particles. Considered from the high energy side, this renders impossible the identification of the $\\gamma$--ray emitters counterparts. As a single example of this fact, in relation with this workshop, one can cite the case of the SS~433/W50 system. The Eastern lobe is known to emit above 10~keV thanks to RXTE observations (Safi-Harb \\& Petre, 1999), but the interpretation of this emission is strongly dependent on the assumptions made on the size of the emitting region which is completely unknown because of the very poor angular resolution of RXTE ($\\sim$~1\\degree, similar to the lobe size).  The SIMBOL--X mission is basically designed to extend the X--ray focussing technique to much higher energies, up to $\\sim$~70 keV, \\ie well beyond the transition between thermal and non thermal emissions. Offering a constant spatial resolution and a ``soft X--ray type\" sensitivity over the full energy range from 0.5 to 70~keV, SIMBOL--X will be an excellent instrument to elucidate the origin of the non thermal emission in accretion / acceleration astrophysical sites, both compact and extended. ",
        "conclusions": ""
    },
    "0209/astro-ph0209581_arXiv.txt": {
        "abstract": "{ The evolution of a star of initial mass 9 $M_{\\odot}$,  and  metallicity $Z = 0.02$ in a Close Binary System (CBS) is followed in the presence of different mass companions in order to study their influence on the final evolutionary stages and, in particular, on the structure and composition of the  remnant  components.  In order to do that, we study two  extreme cases.  In the first  one the mass of the  secondary  is 8  $M_{\\odot}$, whereas in the second one the mass was assumed to be 1 $M_{\\odot}$.  For the first of those cases we have also explored the possible  outcomes of both conservative and  non-conservative  mass-loss episodes.  During the first mass  transfer  episode,  several  differences  arise  between the models.  The system with the more extreme  mass ratio  ($q=0.1$)  is not able to survive the first Roche lobe  overflow  (RLOF) as a binary,  but instead,  spiral-in of the  secondary  onto the  envelope of the primary star is most likely.  The system formed by two stars of comparable  mass undergoes two mass transfer  episodes in which the primary is the donor. We have performed two sets of calculations corresponding to this case in order to account for  conservative  and  non-conservative  mass transfer during the first mass loss episode.  One of our main results is that for the  non-conservative  case the secondary becomes a Super--AGB star in a binary system.  Such a star undergoes a final dredge-up episode, similar to that of a single star of comparable mass.  The primary  components do not undergo a Super--AGB phase, but instead a carbon-oxygen  white dwarf is formed in both  cases  (conservative  and  non-conservative),  before reversal mass transfer  occurs.  However,  given the extreme mass ratios at this stage between the  components of the binary  system,  especially for the  conservative  case, the possibility of merger episodes  remains likely.  We also discuss the presumable final outcomes of the system and possible observational counterparts. ",
        "introduction": "This is the  second of a series of papers in which we intend to  explore extensively  the  evolution of  intermediate  mass close binary  systems (IMCBS).  IMCBS  are  defined  as those  systems  in which  the  primary component  develops a partially  degenerate  carbon-oxygen  core,  after burning central helium in non-degenerate  conditions.  In particular, we focus on the evolution of  heavy-weight  intermediate  mass stars, which have primary masses between $\\sim\\,8\\,M_\\odot$ and $11\\,M_\\odot$.  These stars are thought to  ultimately  develop ONe cores  (Ritossa,  Garc\\'\\i a--Berro  \\& Iben  1996), and  populate  the  brightest  portion  of the Asymptotic  Giant Branch.  In this paper we follow the evolution of both components  of the binary  system from their main  sequence  phase until late evolutionary stages, paying special attention to the carbon burning phase, and to the final objects we can encounter afterwards.  Even  though the  evolution  of  isolated  intermediate  mass  stars has recently been analyzed --- see, for instance, Ritossa, Garc\\'\\i a--Berro \\& Iben (1999), and references therein --- the evolution of this kind of star in binary  systems has been very little  studied.  Probably  one of the reasons for the absence of this kind of model in the  literature  is that the conversion of a CO core into an ONe core involves following the evolution of an unstable  nuclear  burning flame as it propagates  first towards the center of the star and then outwards.  The  calculation is a very delicate and time-consuming task since the propagation of the flame front is not steady, but is  interrupted  by a series of rather  violent shell flashes  (Garc\\'\\i  a--Berro,  Ritossa \\& Iben 1997).  The flashes are violent  because the  nuclear  reaction  rates are very  temperature sensitive and the material is partially  degenerate.  Moreover,  because of the near discontinuity in the physical  variables at the flame front, following  the inward  motion of the flame  requires  very good  spatial resolution  and short  time-steps.  Furthermore,  the evolution of stars within this range of masses in binary systems is even more  complicated, since the  presence  of a close  companion  can  dramatically  alter the evolution  of the primary  star and its final  outcome.  As a result the only recent  calculation in which a heavy-weight  intermediate mass star is followed  from its main  sequence  phase until carbon is exhausted in the core is that of Gil--Pons \\& Garc\\'\\i  a--Berro (2001), who followed the evolution of a $10\\,M_\\odot$ star with solar metallicity in a binary system.  Moreover,  most of the  developments  in the field of  binaries frequently  disregarded  the study of AGB stars, arguing that this phase is prematurely  quenched in binary systems, due to significant mass loss in the evolution  previous to the AGB phase.  Important  exceptions  are the works of Jorissen  (1999) and Smith et al.  (1996) who  proposed the existence of AGB stars in binary systems and, more  recently, Van Eck et al.  (2001),  who found  observational  evidence  of three AGB  stars in binary systems.  In our  previous  paper  (Gil--Pons  \\&  Garc\\'{\\i}a--Berro,  2001),  we proposed a scenario  in which a $10\\,  M_{\\odot}$  star could  evolve to become a Super-AGB star and, ultimately, an ONe white dwarf, in spite of the fact  that all the  hydrogen-rich  envelope  and most of the  helium layer had been lost in previous  phases of the evolution.  In this paper we consider again such an scenario but we relax some of the  simplifying assumptions  that  were  made  there.  In  particular,  we  compute  the evolution of a $9\\, M_{\\odot}$ primary (the initially most massive star) of the  binary  system,  and we  explore  the  effect of the mass of its companion.  In doing so, we study two  extreme  cases; in the  first one the initial mass ratio ($q_0$) is close to one (scenario  1), whereas in our second  calculation  this ratio is less than 0.2  (scenario  2).  In fact, $q_0=0.2$ is a theoretical limit under which spiral-in of the less massive  star  onto  the  most  massive  and  finally,   the  merger  of components, cannot be avoided --- see, e.g., Vanbeveren et al.  (1998a).  Han et al.  (2000) did a  comprehensive  study of the evolution of IMCBS in order to  determine  the  influence  of the  initial  mass  ratio and orbital  period on the final  parameters  of close binary  systems whose primary initial mass is in the range $1\\, M_{\\odot}$ to $8\\, M_{\\odot}$. Although  they  only  consider  the  conservative  case,  a  significant variation in the primary  remnant  masses is observed.  In this paper we focus on more massive  primaries --- which have been little  studied --- and we  extend  our  previous  calculations  in order  to  consider  the evolution  of the  secondary.  Furthermore,  since  there are still many uncertainties  in relation to whether mass transfer is  conservative  or not, we have computed two  evolutionary  sequences  for the first of our scenarios, and we have  considered  the two extreme cases that may occur during the first mass loss episode:  \\begin{itemize} \\item [] 1.a :  A CBS  composed of a $9\\,  M_{\\odot}$  star plus a $8 \\, M_{\\odot}$ companion.  The initial period is of about 10 days, so that mass loss from the primary starts  shortly after  hydrogen  is ignited in a shell (a mass  transfer episode of the  $B_{\\rm  r}$  type).  Mass  transfer  is conservative during the first RLOF. \\item []  1.b : A CBS  composed of a $9\\,  M_{\\odot}$  star plus a $8 \\, M_{\\odot}$  companion.  The  initial  period is of about 150 days, so that mass loss from the primary starts when it is  climbing  the red giant  branch (a mass  transfer episode of the  $B_{\\rm  c}$  type).  Mass  transfer  is non-conservative during the first RLOF. \\item []   2 :  A CBS  composed of a $9\\,  M_{\\odot}$  star plus a $1 \\, M_{\\odot}$  companion.  The initial period is of about 5 days. \\end{itemize}  Case 1.a  corresponds to a typical  conservative  mass transfer  episode (Nelson \\&  Eggleton  2001; Han et al.  2000),  whereas  case 1.b can be highly  non-conservative.  Therefore,  we  have  studied  it  under  the assumption that a common envelope forms that removes an important amount of mass and angular momentum from the system.  It is precisely from this secondary of the non-conservative  case that a Super--AGB star develops.  The paper is  organized  as follows.  In section 2, we present in detail our  evolutionary  scenarios.  Section 3 is devoted  to the study of the evolution  of the  first of the  above  mentioned  scenarios  in which a massive  companion is assumed, whereas in section 4 we explain in detail the  evolution  in the  case  of a  low-mass  secondary.  The  phase  of reversal mass transfer in the first of our scenarios is fully  explained in section \\S 5.  Finally in  Section 6 we  discuss  and  summarize  our major findings. ",
        "conclusions": "We have  followed  the  evolution  of a 9  $M_{\\odot}$  star with  solar metallicity  in close binary  systems  with  different  initial  orbital parameters.  Our main goal has been to analyze the influence of the mass of the secondary  component on the final possible  outcomes of the close binary  systems.  In particular, we have computed a binary  evolutionary scenario that allows the formation of  Super--AGB  stars in close binary systems.  Our analysis encompassed all the relevant evolutionary phases, starting from the main sequence of both components, following the helium burning phase, and, if necessary the carbon burning  phase.  In summary, we have studied the following three cases:  \\begin{itemize}  \\item [] Case 1.a:  9 $M_{\\odot}$ + 8 $M_{\\odot}$,  with initial orbital period  $P_{\\rm orb} \\sim 10$ days, so the first mass  transfer episode is a case $B_{\\rm r}$ mass transfer episode.  \\item [] Case 1.b:  9 $M_{\\odot}$ + 8 $M_{\\odot}$,  with initial orbital period  $P_{\\rm orb} \\sim 150$ days, so the first mass transfer episode is a case $B_{\\rm c}$ mass transfer episode.  \\item [] Case 2:  9 $M_{\\odot}$ + 1  $M_{\\odot}$,  with initial  orbital period $P_{\\rm orb} \\sim 5$ days.  \\end{itemize}  For the last of these three cases the first mass transfer  episode leads to the merger of  components,  due to the  initial  mass  ratio  between components:  $q \\lesssim 0.2$.  The resulting  $10\\,  M_{\\odot}$  single star behaves very much like the normal 10  $M_{\\odot}$  star  studied by Garc\\'{\\i}a--Berro and Iben (1994), once the merger episode is over.  Cases  1.a and 1.b  are  more  complicated.  These  two  binary  systems undergo two mass  transfer  episodes.  This kind of  evolution  leads to very  similar  primary  remnants:  both of them  are  massive  CO  white dwarfs,  $M_{\\rm WD} \\sim 0.98\\,  M_{\\odot}$.  However,  the two systems are very different in orbital period and masses of the secondaries.  For case  1.a the  system  is  composed  of a  primary  remnant  and a $16\\, M_{\\odot}$  main sequence  star, with a  hydrogen-rich  envelope  highly polluted by the CNO products from the primary  remnant,  dredged--up  by convection  and expelled in its second RLOF, and also by the products of its own helium  burning  shell.  The secondary  for case 1.b is a $9.3\\, M_{\\odot}$  star.  In many aspects the evolution of this star  resembles very much that of the $9\\, M_{\\odot}$ single star previously  studied by Garc\\'{\\i}a--Berro  et al.  (1999), the only  difference  being that the second  dredge-up is not caused by carbon  burning, but, instead, by the gravothermal energy release at the end of core helium burning and before carbon is ignited off-center in the degenerate core.  Another  important  difference  between  case  1.a and  case  1.b is the evolutionary  stage of the secondary at the time at which  reversal mass transfer  occurs.  For the  case  1.a,  it  occurs  when  the  secondary component  climbs  the red giant  branch,  whereas  for case 1.b it only fills its Roche lobe when it reaches the Super--AGB phase.  However, the uncertainties  related to stellar winds, both in mass-loss  rates and in the efficiency of mass accretion, do not allow us to determine the exact orbital  parameters  at the  begining of  reversal  mass  transfer  and, ultimately,  whether merger episodes could occur.  In fact, if mass loss from the  secondary  (as a red giant or as a Super--AGB  star)  allows a significant  decrease in its mass with low enough mass  transfer  rates, the steep  change in the density  profile of the mass losing star at the border of its core could brake the spiral-in of the primary.  The system would  then  be  able  to  survive  as  a  binary,  and  the   following possibilities may arise:  \\begin{enumerate}  \\item Let us  consider  case  1.a.  If  most  of the  mass  lost  by the secondary  component during the reversal mass transfer  episode is lost by the system and the primary  accretes matter at rates lower than the critical value for hydrogen  burning, we might  encounter that the system  experiences nova outbursts.  The final outcome of the primary  component will still be a white dwarf but its massive companion will evolve to undergo a supernova  outburst and leave a neutron  star  as a  remnant.  The  highly  eccentric  pulsar  PSR B2303+46  (Van  Kerkwijk \\&  Kulkarni,  1999)  could be a possible observational  counterpart for such a system, as the companion for the neutron star is a massive ($1.2\\,  M_{\\odot}  \\lesssim  M_{\\rm WD} \\lesssim 1.4\\,  M_{\\odot}$)  white dwarf that might correspond to our primary  component  after having  accreted  some extra mass during the reversal mass transfer episode.  If, on the other hand,  accretion  onto the  primary is  efficient enough so that this  component is able to reach the  Chandrasekhar mass, it will become a neutron  star after a  supernova  explosion (Guti\\'errez  et al.  1996).  Furhermore,  if during the accretion phase of the primary, the  secondary  does not lose a  significant amount of mass, our system may help to explain the HMXB  precursor PSR J1740-3052 (Stairs et al., 2001), a binary system with $P_{\\rm orb}=234$ days, composed of a neutron star plus a massive ($\\grsim 10\\,  M_{\\odot}$)   companion.  Similarly,  the  highly  eccentric pulsar PSR B1259-63  (Johnston et al., 1992) might also be another possible   observational   counterpart,   as  the   non-degenerate companion of the neutron star has a mass $\\grsim 10\\, M_{\\odot}$.  Further evolution of the secondary might also lead it to undergo a supernova  explosion and leave a second neutron star as a remnant, and hence a binary pulsar could also be a possible  outcome.  Such a system  may  remain  bound  or be  disrupted,  depending  on the asymmetry  of the  supernova  explosion  and on the amount of mass ejected  from the  system.  A possible  observational  counterpart could be the binary pulsar PSR B1813+16 (Taylor et al., 1976).  \\item Concerning  our scenario 1.b, we have the following  possibilities during the reversal mass transfer  process.  If the  efficiency in the ejection of the common  envelope  formed during mass loss from the $9\\,  M_{\\odot}$  Super-AGB star is high, the final outcome of the system will be a very close double white dwarf (Maxted,  Marsh \\& Moran 2002), possibly after having experienced a series of nova outbursts.  The orbital  shrinkage  allowing the components to get close enough to interact would be a consequence of the last common envelope phase.  But, if the primary component is able to grow and reach  the  Chandrasekhar  mass,  the  outcome  will  be a  system composed  of  a  neutron  star  plus  a  CO  white  dwarf,   whose observational  counterpart  might be PSR  J1756-5322  (Edwards  \\& Bailes, 2001),  provided that the orbital  shrinkage due to common envelope  evolution  might  account for the short period  ($\\simeq 11^{\\rm h}$) of this system.  In this case the eccentricity of the system would be small, since the common  envelope  phase for these systems could occur after the supernova outburst.  \\end{enumerate}  Finally,  we would  like to  emphasize  that one of our  most  important findings is that Asymptotic  Giant Branch stars could indeed be found in binary  systems, in agreement with the  predictions of Jorissen  (1999). The  Super-Asymptotic  Giant  Branch star is the  secondary  of a binary system in our  scenario  1.b, and should have an evolved  (and  possibly degenerate)  companion.  The  observational  signature  of  these  stars should be an  anomalous  enhancement  of the  abundances  of carbon  and oxygen and a slight underabundance of nitrogen with respect to solar."
    },
    "0209/astro-ph0209548_arXiv.txt": {
        "abstract": "This work investigates   Bondi accretion  to a rotating magnetized star in the ``propeller\" regime using axisymmetric resistive, magnetohydrodynamic simulations. In this regime accreting matter tends to be expelled from the equatorial region of the magnetosphere where the centrifugal force on matter rotating with the star exceeds the gravitational force. The regime is predicted to occur if the magnetospheric radius larger than the corotation radius and less than the light cylinder radius. The simulations show that accreting matters is expelled from the equatorial region of the magnetosphere and that it moves away from the star in a supersonic, disk-shaped outflow. At larger radial distances the outflow slows down and becomes subsonic. The equatorial matter outflow is initially driven by the centrifugal force, but at larger distances the pressure gradient becomes significant. We find that the star is spun-down mainly by the  magnetic torques  at its surface with the rate of loss of angular momentum $\\dot{L}$ proportional to $-\\Omega_*^{1.3}\\mu^{0.8}$, where $\\Omega_*$ is the star's rotation rate and $\\mu$ is its magnetic moment. Further, we find that $\\dot{L}$ is approximately independent of the magnetic diffusivity of the plasma $\\eta_m$. The fraction of the Bondi accretion rate which accretes to the surface of the star is found to be $\\propto \\Omega_*^{-1.0}\\mu^{-1.7}\\eta_m^{0.4}$. Predictions of this work are important for the observability of isolated old neutron stars and for wind fed pulsars in X-ray binaries. ",
        "introduction": "Rotating magnetized neutron stars pass through different stages in their evolution (e.g., Shapiro \\& Teukolsky 1983;  Lipunov 1992). Initially, a rapidly rotating ($P \\lesssim 1{\\rm s}$) magnetized neutron star is expected to be active as a radiopulsar. The star spins down owing to the wind of magnetic field and relativistic particles from the region of the light cylinder $r_L$ (Goldreich and Julian 1969). However, after the neutron star spins-down sufficiently, the light cylinder radius becomes larger than magnetospheric radius $r_m$ where the ram pressure of external matter equals the magnetic pressure in the neutron star's dipole field. The relativistic wind is then suppressed by the inflowing matter (Shvartsman 1970). The external matter may come from the wind from a binary companion or from the interstellar medium for an isolated neutron star. The centrifugal force in the equatorial region at $r_m$ is much larger than gravitational force if $r_m$ is much larger than the corotation radius $r_{cor}$. In this case the  incoming matter tends to be flung away from the neutron star by its rotating magnetic field. This is the so called ``propeller\" stage of evolution (Davidson \\& Ostriker 1973; Illarionov \\& Sunyaev 1975).   The ``propeller\" stage of evolution, though important, is  still not well-understood theoretically. Different  studies have found different dependences for the spin-down rate of the star (Illarionov \\& Sunyaev 1975; Davis, Fabian, \\& Pringle 1979; Davis \\& Pringle 1981; Wang \\& Robertson 1985; Lipunov 1992). Observational signs of the propeller stage have been discussed by number of authors (e.g., Stella, White, \\& Rosner 1986; Treves \\& Colpi 1991; Cui 1997; Treves et al. 2000).  MHD simulations of  disk accretion to a rotating star in the propeller regime  were done by Wang and Robertson (1985). However, authors considered only equatorial plane and concentrated on investigation of instabilities at the boundary between the magnetosphere and surrounding medium. Thus they could not investigate accretion along the magnetic poles of the star. An analytical model of disk accretion in the propeller regime was developed by Lovelace, Romanova, and Bisnovatyi-Kogan (1999). Disk accretion at the stage of weak propeller ($r_m \\sim r_{cor}$) investigated numerically by Romanova et al. (2002).   \\begin{figure*}[t] \\centering \\caption{Geometry of the MHD simulation region, where $\\dot{M}_B$ is  the Bondi accretion rate, $\\rvecmu$ and ${\\bf \\Omega}_*$ are the magnetic moment and angular velocity of the star, $R_*$ is the radius of the star, $(R_{max}, Z_{max})$ are the limits of the computational region. In the described calculations $R_* \\ll R_{max}$. } \\label{Figure 1} \\end{figure*}    The mentioned  studies obtained  possible trends of the propeller stage of evolution. However, two and three dimensional MHD simulations are needed to obtain more definite answers to the important physical questions. The questions which need to be answered are: (1) What are the physical conditions of the matter flow around the star in the propeller regime of accretion? (2) What is the spin-down rate of the star, and how does it depend on the star's magnetic moment and rotation rate and on the inflow rate of the external matter? (3) What is the accretion rate  to the surface of the star and how does it depend on the star's rotation rate and magnetic moment? (4) What are the possible observational consequences of this stage of evolution?  This paper  discusses results of axisymmetric, two-dimensional, resistive MHD simulations of accretion a rotating magnetized star in the ``propeller\" regime. We treat the case when matter accretes spherically with the Bondi accretion rate. Bondi accretion to a non-rotating and a slowly rotating star was investigated by Toropin et al. (1999; hereafter T99). It was shown that the magnetized star accretes matter at a rate less than the Bondi rate. Toropina et al. (2002; hereafter T02) confirmed this result for stronger magnetic fields. In this paper we consider initial conditions similar to those in T02, but investigate the case of rapidly rotating stars. Section 2 gives a rough physical treatment of the propeller regime. Section 3 describes the numerical model and computations. Section 4 describes the main results from our simulations, and Section 5 gives a numerical application of our results. Section 6 gives the conclusions from this work. ",
        "conclusions": "Axisymmetric magnetohydrodynamic simulations of Bondi accretion to a rotating magnetized star in the propeller regime of accretion  have shown that: (1) A new regime of matter flow forms around a rotating star. Matter falls down along the axis, but only a small fraction of the incoming matter accretes to the surface of the star. Most of the matter is expelled radially in the equatorial plane by the rotating magnetosphere of the star. A low-density torus forms in the equatorial region which rotates with velocity significantly larger than the radial velocity. Large scale vortices form above and below the equatorial plane. (2) The star is spun-down by the  magnetic torque and to a lesser extent the matter torque at its surface. The rate of loss of angular momentum $\\dot{L}$ is proportional to $-\\Omega_*^{1.3}\\mu^{0.8}$, and it is approximately independent of $\\eta_m$. This dependence differs from the predicted dependence of equation (3) probably because the corotation radius is not much smaller than the magnetospheric radius $r_m$. The rotational energy lost by the star goes into the directed and thermal energy of plasma. (3) The accretion rate to the star is much less than the Bondi accretion rate and decreases as (a) the star's rotation rate increases ($\\propto \\Omega_*^{-1.0}$), (b) as the star's magnetic moment increases ($ \\propto \\mu^{-1.7}$), and as the magnetic diffusivity decreases [$\\propto (\\eta_m)^{0.4}$]. (4) Because the accretion rate to the star is less than the Bondi rate, a shock wave forms in our simulations and propagates outward. It has the shape of an ellipsoid flattened along the rotation axis of the star."
    },
    "0209/astro-ph0209018_arXiv.txt": {
        "abstract": "{ We develop a non-parametric inverse method to investigate the star formation rate, the metallicity evolution and the reddening properties of galaxies based on their spectral energy distributions (SEDs). This approach allows us to clarify the level of information present in the data, depending on its signal-to-noise ratio (S/N). When low resolution SEDs are available in the ultraviolet, optical and near-IR wavelength ranges together, we conclude that it is possible to constrain the star formation rate and the effective dust optical depth simultaneously with a signal-to-noise ratio of 25. With excellent signal-to-noise ratios, the age-metallicity relation can also be constrained.\\\\ We apply this method to the well-known nuclear starburst in the interacting galaxy NGC\\,7714. We focus on deriving the SFR and the reddening law. We confirm that classical extinction models cannot provide an acceptable simultaneous fit of the SED and the lines. We also confirm that, with the adopted population synthesis models and in addition to the current starburst, an episode of enhanced star formation that started more than 200\\,Myr ago is required. As the time elapsed since the last interaction with NGC\\,7715, based on dynamical studies, is about 100\\,Myr, our result reinforces the suggestion that this interaction might not have been the most important event in the life of NGC\\,7714. ",
        "introduction": "The integrated spectra of galaxies contain information about the ages of their stellar populations, the metallicity of the stars and the effects of dust extinction.  In order to infer the history of the star formation rate (SFR), one usually tries to match the spectral energy distribution (SED) as closely as possible with model populations computed with various scenarios for the SFR. The quality of the SED fit is assessed either qualitatively by visual inspection or quantitatively, e.g. by the $\\chi^2$-test or a more general expression of the likelihood. Often, the SFR is represented by a number of discrete values or by a predefined analytic form. This approach, which could be called the direct or synthetic method, has been largely used in the field of population synthesis (see Heavens et al. 2000, Reichardt et al. 2001 for efficient recent implementations). Since one usually keeps the number of free parameters as small as possible, the adopted functional forms for the SFR and age-metallicity relation (AMR) impose certain limitations on what type of model populations can be considered. Typically, these will be combinations of instantaneous bursts and episodes of constant star formation. Some inverse methods (Craig and Brown 1986, Tarantola and Valette 1982\\,a,b) deal with such a problem in an opposite way : one tries to determine the functional form of e.g. the SFR with as much freedom as possible, with a resolution in time that is dictated by the information contained in the data. Because of the latter property, these methods are called non-parametric.  This work presents a non-parametric inverse method to  estimate characteristics of galaxy evolution such as the SFR, the AMR and the intrinsic dust extinction. As for all such approaches, the method is based on a probabilistic formulation of inverse problems (in our case the formalism of Tarantola \\& Valette 1982\\,a,b). We apply the method in the framework of {\\em evolutionary} population synthesis. In other words, possible solutions are only sought among those compatible with our current understanding of star formation and evolution: the relative fraction of stars of various masses is not arbitrary but follows an initial mass function, and theoretical evolutionary tracks combined with stellar spectral libraries determine the possible emission spectra of isochrone stellar populations. A probabilistic formulation for the alternative {\\em empirical} population synthesis has been developed recently in the parametric case by Cid Fernandes et al. (2001). As noted by these authors, the exploration of solutions to an inversion problem can be tackled as a minimization problem or with an adequate sampling algorithm for the space of parameters. The second type of approach provides a complete description of the uncertainties on the estimated parameters, but may become difficult to implement in practice when some of the unknowns are non-parametric functions of time, which can take an immense variety of shapes.  Here a minimization procedure is adopted. In addition, specific tools are used in order to estimate the validity of the inverse procedure, like the a posteriori covariance and resolution, and the mean index.  The nature of the available data determines many of the capabilities and limitations of inversion procedures. Sets of equivalent widths in the optical spectrum have been used because these measurements are relatively easy to acquire (Pelat 1997, Boisson et al. 2000). Cid Fernandes et al. (2001) combined equivalent widths and colours in order also to constrain dust extinction. However, they used a classical one parameter description of extinction and reddening. The effects of dust are rarely that simple (Witt \\& Gordon 2000). The work we present here was partly motivated by previous studies of starburst galaxy spectra, which not only made it clear that average obscuration laws depend on the type of galaxy observed (Calzetti et al. 1994), but also showed that complex distributions of the stars and the dust in space can lead to significant local deviations from this average. Lan\\c{c}on et al. (2001) studied the $\\sim$330 central parsecs of the nuclear starburst galaxy NGC\\,7714, and emphasized the effects of the different optical depths of dust along various lines of sight within their small aperture. They demonstrated that in dusty objects it is necessary to combine SEDs and emission lines from the optical, the ultraviolet and the near-IR spectral ranges if one aims at recovering the SFR over a wide range of ages and useful information on the wavelength dependent attenuation by dust. With those results in mind, we chose to apply the inversion method to data sets such as those of Lan\\c{c}on et al. (2001): in this paper, the empirical constraints are combined low resolution SEDs in the three spectral ranges together with emission lines of H{\\sc ii}. Computation times were not prohibitive in the present case.  The paper is organised as follows. Section 2 \\ref{mod} presents our model assumptions for the SEDs and reddening. Section \\ref{inv} introduces the inverse method.  After applying this technique to simulated SEDs in Sect.\\,\\ref{simul}, we give new constraints on the SFR and the reddening law for the nuclear starburst of NGC 7714 in Sect.\\,\\ref{ngc}.\\\\ ",
        "conclusions": "In this paper we have presented an inversion method designed to estimate the SFR, the AMR and the reddening properties of galaxies from their spectra. This method allows us not only to derive best values for these functional parameters, but also to estimate the amount of information actually present in the data for each of the unknowns, as well as the posterior resolution in time, depending on the S/N of the studied spectra.  The main conclusions based on the inversion of simulated spectra are :  \\begin{itemize}  \\item The SFR, the AMR and the reddening law are determined with a resolution of about 0.3-0.5 in log(age) for low spectral resolution UV\\,+\\,optical\\,+\\,near infrared spectra when noise in the data is negligible (S/N=500).  \\item With S/N=25 and the same spectral coverage, the SFR is determined with a resolution of $\\sim$0.6 in log(age) and the information on the AMR is rather poor. Constraints on the reddening of the stellar continuum emission are tightest for the young stellar populations. The exact amount of recoverable information will depend on the actual star formation history of the object of study. Similar success has been reported by Cid Fernandes et al. (2001) in the framework of empirical population synthesis.  \\item Alternatively, it is possible to constrain the SFR and the wavelength dependence of the effective optical depth simultaneously with a S/N=25, if it is assumed that this optical depth is independent of age (i.e. applies to all stellar populations) and that the stellar metallicity is known. In that case, the resolution on the SFR is only about 1 in log(age).  \\end{itemize}  The inversion method applied to the spectrum of NGC 7714 confirms that the reddening properties of this galaxy cannot be modeled with simple extinction models. In order to obtain satisfactory fits to both the stellar energy distribution and the nebular emission lines, one has to allow the wavelength dependence of attenuation to deviate from ``standard\" laws (this work) or to allow for variations in the attenuation between different coexisting stellar populations (LGLG01).  The main results obtained for NGC 7714 can be compared to the previous results from non-automated studies (LGLG01):  \\begin{itemize}  \\item The SFR found is indeed multimodal, with a young star formation peak a few Myr ago and a more extended episode that formed stars at a typical rate of 0.5-1\\,M$_{\\odot}$\\,yr$^{-1}$, thus providing an important intermediate age population.  \\item No solution was found in which the episode of enhanced star formation that produced the bulk of the intermediate age stars started less than $\\sim$\\,300\\,Myr ago. The discrepancy with the dynamical results of Smith \\& Wallin (1992), who estimate that $\\sim 100$\\,Myr elapsed since closest encounter with NGC\\,7715, thus persists. New simulations tend to confirm the dynamical timescales (Struck \\& Smith 2002), although a wider exploration of initial conditions may still allow it to be stretched a little. The importance of the event in terms of the stellar mass produced, especially when compared to the most recent starburst, tends to favour the occurence of an earlier perturbation, possibly unrelated with NGC\\,7715.  \\item By solving simultaneously for the SFR and the effective optical depth one strongly increases the estimated amplitude of the youngest peak in the SFR, while producing an attenuation law that is flattened in the UV part of the spectrum. The global shape of the attenuation is similar to the average law derived by Calzetti et al. (2000), but the differences in the details are significant when the star formation rates are studied. The Br$\\gamma$ line emission is well matched, while it was underestimated by a factor of 2 with ``classical\" attenuation laws. A drawback of this solution is that it produces a suspicious bump in the optical obscuration law in order to match the shape of the Balmer jump.  \\item The solution just described mimics the effects that LGLG01 have shown to be due to a very inhomogeneous distribution of the dust within the small region observed, with one particular line of sight of very low extinction towards a cluster of hot stars. Even with UV, optical and near-IR spectroscopy combined, solving for the effective optical depths requires strong assumptions. Here we have traded freedom in the wavelength dependence of optical depths against the freedom to assign populations of different ages differing amounts of extinction. Nature might be even more complex, and require freedom in both when individual galaxies are studied.  \\end{itemize}  The precious information of the spatial distribution of stars and dust within the region observed spectroscopically is not always accessible. The detailed study of objects for which high resolution imaging is possible remains essential, and gives us insight into the fundamental uncertainties involved when only integrated light is available. Improvements will come naturally from a wider spectral coverage (if possible with overlaps between spectral ranges observed independently), and from the use of models with higher spectral resolution."
    },
    "0209/astro-ph0209532_arXiv.txt": {
        "abstract": "We use publicly available N-body simulations and semi-analytic models of galaxy formation to estimate the levels of external shear due to structure near the lens in gravitational lens systems. We also describe two selection effects, specific to four-image systems, that enhance the probability of observing systems to have higher external shear. Ignoring additional contributions from ``cosmic shear'' and assuming that lens galaxies are not significantly flattened, we find that the mean shear at the position of a quadruple lens galaxy is 0.11, the {\\em rms} shear is $\\sim$0.15, and there is a $\\sim$45\\% likelihood of external shear greater than 0.1. This is much larger than previous estimates and in good agreement with typical measured external shear. The higher shear primarily stems from the tendency of early-type galaxies, which are the majority of lenses, to reside in overdense regions. ",
        "introduction": "\\label{sec:intro}  With the benefit of hindsight it can be argued that even the second gravitational lens discovered, PG1115+080 (Weymann et al. 1980), foreshadowed what has turned into an embarrassment of riches -- a superabundance of quadruply imaged quasars.  But only with the advent of systematic lens surveys (King and Browne 1996; Rusin and Tegmark 2001) has it become clear that the high ratio of quadruple to double systems is not an artifact of observational selection and therefore presents a genuine challenge to our understanding of lensing of galaxies. \\nocite{weymann80,rusin01,king96} Furthermore, individual fits on a system-by-system basis often require large amplitude (0.1-0.3) external shear \\citep{hogg94,schechter97,kneib00,fischer98}, significantly higher than values of 0.02-0.05 expected (Keeton, Kochanek and Seljak 1997; henceforth KKS) from large scale structure or nearby galaxies.  \\nocite{blandford87,kochanek87,turner84}  There are three factors that will probably have some part to play in the ultimate resolution of these problems: galaxy ellipticities, shear due to random superpositions of mass along the line of sight, and shear due to structures that are associated with the lens galaxy. In this paper we concentrate on the shear from associated structures.  The relative importance of tides and ellipticity has been considered by KKS.  In computing the expected tidal shear they consider three contributions: a) random shear due to unassociated foreground and background structure, b) the effect of associated galaxies, through the two point correlation function, and c) the effect of clusters of galaxies, with a term proportional to the number density of clusters. \\nocite{keeton97}  In the present paper we take a different tack to estimate the expected tidal shear due to nearby structures.  We use the GIF Project recipe (Kauffmann et al. 1999) for galaxy formation within a cold dark matter (CDM) simulation to compute the shear expected along random lines of sight, in the directions of galaxies, and specifically in the directions of early-type galaxies, which appear to be the predominant type of lens galaxy. In this way, the effects of correlated structure are naturally included, without artificial distinctions between different types of structures, and we can also include the clustering properties of different types of galaxies. This is similar to recent work by \\citet{white01}, where a high resolution hydrodynamical simulation was used to find typical values of external shear at the positions of typical galaxies, rather than specifically at the positions of massive early-type galaxies.  In \\S~2 we outline our methods for using the GIF simulations to estimate the effects of correlated structure on the statistics of external shear in gravitational lens systems. In \\S~3 we describe two selection effects which will further increase the typical external shear measured in quadruple gravitational lens systems; high shear systems have a larger cross-section for quadruple lensing (which is partly offset by ``magnification bias'') and regions of high external shear are likely to be regions of higher convergence, due to large scale structure. We finish with a discussion of the consequences of our calculations for various problems associated with quadruply imaged systems. ",
        "conclusions": "\\label{sec:disc}  We have shown that the relatively high amounts of external shear found in most lens systems naturally arise from non-linear large scale structure in the vicinity of the lenses. Early-type galaxies are systematically in more overdense regions and therefore would be expected to experience higher amplitude tidal fields.  The fact that quadruple lens galaxies are mainly early-type galaxies is enough to explain the observed values of external shear, previously considered high.  The dominance of early-type galaxies in lens samples arises simply from the bias in lens samples toward more massive galaxies, which are more likely to be early-type galaxies \\citep{fukugita91,maoz93,kochanek93}.  This amount of shear expected from nearby structures is not sufficient to, by itself, solve the ``quad problem'' (the apparent relative super-abundance of four-image systems) but the effect goes in the right direction and should add with other effects, such as making density profiles in the centers of galaxies less steep. Paying careful attention to the preferred positions of galaxies relative to typical places in the universe, as we have done here, will almost certainly have an important role in understanding these observations.  Typical shear values are sufficiently high that external shear should be of comparable or greater importance in lens modeling to internal shear (galaxy ellipticity), in agreement with model fits to known lenses. We do not expect the introduction of galaxy ellipticities to significantly change our conclusions, but it would be a natural direction for future work.  Selection effects are important in gravitational lens systems. The results presented here suggest that our knowledge of structure formation and galaxy environments can be usefully applied to understanding the observed statistics and characteristics of lens systems.  As numerical simulations of large scale structure, models of galaxy formation, and gravitational lens catalogs all continue to improve we can expect the importance of an integrated approach to only increase."
    },
    "0209/astro-ph0209368_arXiv.txt": {
        "abstract": "Evolution characteristics of a Kerr black hole (BH) are investigated by considering coexistence of disc accretion with the Blandford-Znajek process (the BZ process) and magnetic coupling of the BH with the surrounding disc (MC process). (i) The rate of extracting energy from the rotating BH in the BZ process and that in MC process are expressed by a unified formula, which is derived by using an improved equivalent circuit. (ii) The mapping relation between the angular coordinate on the BH horizon and the radial coordinate on the disc is given in the context of general relativity and conservation of magnetic flux. (iii) The power and torque in the BZ process are compared with those in MC process in detail. (iv) Evolution characteristics of the BH and energy extracting efficiency are discussed by using the characteristics functions of BH evolution in the corresponding parameter space. (v) Power dissipation on the BH horizon and BH entropy increase are discussed by considering the coexistence of the above energy mechanisms. ",
        "introduction": "{It is well known that magnetized accretion disc of black hole (BH) is an effective model in astrophysics. The interaction between a Kerr BH and the surrounding magnetic field has been used not only to explain high energy radiation and jet production from quasars and active galactic nuclei, but also as a possible central engine for gamma-ray bursts (Rees 1984; Frank, King \\& Raine 1992; Lee, Wijers \\& Brown 2000, hereafter LWB). Confined by the magnetic field in the inner region of the disc, the magnetic field lines frozen previously in the disc plasma will deposit on the horizon in company with the accretion onto the BH, and the magnetized disc becomes a good environment for supporting the magnetic field on the horizon. Blandford and Znajek (1977) proposed firstly that the rotating energy and the angular momentum of a BH can be extracted by the surrounding magnetic field, and this energy mechanism has been referred to as the BZ process, in which the BH horizon and the remote astrophysical load are connected by the open magnetic field lines, and energy and angular momentum are extracted from the rotating BH and transported to the remote load. \\\\ \\indent There still remain some open problems on the BZ process, and one of them is how to estimate the ratio of the angular velocity of the magnetic field lines to that of the BH horizon. Macdonald and Thorne (1982 hereafter MT82) argued in a speculative way that the ratio will be regulated to about 0.5 by the BZ process itself, which corresponds to the optimal value of the extracting power with the impedance matching. However, as argued by Punsly and Coroniti (1990), it is hard to understand how the load can conspire with the BH to have the same resistance and satisfy the matching condition, since the load is so far from the BH that it cannot be casually connected. \\\\ \\indent Recently some authors (Blandford 1999; Li 2000a, 2000b; Li \\& Paczynski 2000) argued that with the existence of the closed field lines a fast rotating BH will exert a torque on the disc to transfer energy and angular momentum from the BH to the disc. Henceforce this energy mechanism is referred to as magnetic coupling (MC) process. Compared with the BZ process, the load in MC process is the surrounding disc, which is much better understood than the remote load though the magnetohydrodynamics (MHD) of the disc is still very complicated. \\\\ \\indent Some works have been done to discuss the influence of the BZ process on the evolution of BH accretion disc (Park \\& Vishniac 1988; Moderski \\& Sikora 1996; Lu et al. 1996; Wang et al. 1998, hereafter WLY). However MC effects involving the closed field lines were not taken into account in the previous works. In fact the closed field lines should exist on the horizon as well as the open field lines, and disc accretion should coexist with the BZ and MC process. In this paper evolution characteristics of a Kerr BH are investigated by considering coexistence of disc accretion with the BZ and MC process (henceforce DABZMC). \\\\ \\indent In order to facilitate the discussion of the evolution of the central BH surrounded by the magnetized accretion disc we make the following assumptions: \\\\ \\indent (i) The disc is perfectly conducting and the magnetic field lines are frozen in the disc. The magnetosphere is stationary, axisymmetric and force-free outside the BH and the disc (MT82); \\\\ \\indent (ii) The magnetic field is so weak that its influence on the dynamics of the particles in the disc is negligible, and the BH has an external geometry of Kerr metric; \\\\ \\indent (iii) The disc is thin and Keplerian, lies in the equatorial plane of the BH with the inner boundary being at the marginally stable orbit. \\\\ \\indent This paper is organized as follows. In Sec.II the rate of extracting energy from the rotating BH in the BZ process (henceforce the BZ power) and that in MC process (henceforce MC power) are expressed by a unified formula, which is derived by using an improved equivalent circuit based on MT82. Our result for the BZ power is consistent with that derived in LWB. In Sec.III a mapping relation between the angular coordinate on the BH horizon and the radial coordinate on the disc is given in the context of general relativity and conservation of magnetic flux. In Sec.IV The power and torque in the BZ process are compared with those in MC process in detail. In Sec.V evolution characteristics of the BH and energy extracting efficiency are discussed by using the characteristics functions in the corresponding parameter space. In Sec.VI we discuss the power dissipation on the BH horizon and the BH entropy increase, and show that the excess rate of change of BH entropy does come from the total power dissipation on the BH horizon in the BZ and MC process. Finally, in Sec.VII, we summarize our main results. } ",
        "conclusions": "{ In this paper the evolution characteristics of a Kerr BH and some related issues, such as energy extracting efficiency and entropy change on the BH horizon, are investigated by considering coexistence of DABZMC. By using an improved equivalent circuit a unified expression for the BZ power and MC power are derived. Starting from the conservation laws of energy and angular momentum, we obtain the basic evolution equations of the BH and the CFs in terms of four parameters related to our model, i.e., $a_*$, $\\theta _M$, $k$ and $n$. We find that the evolution characteristics of the Kerr BH can be well described by using the CHs and the RPs in the corresponding parameter space.  It turns out that the main difference between the BZ process and MC process lies in the two kinds of loads: the remote load with an unknown parameter $k$ for the BZ process, and the load disc with the parameter $\\beta$ for MC process. Parameter $k$ is uncertain, while $\\beta$ is thoroughly determined by the BH spin and the place where the magnetic flux penetrates. The effects of MC process on the evolution of the BH and the efficiency of BH accretion disc can be discussed in virtue of the parameters $a_*$, $\\theta _M$ and $n$, into which the parameter $\\beta$ is merged.  In this model the mapping relation between the BH horizon and the disc is derived by assuming a power-law of the magnetic field varying as the radial coordinate of the disc. It is shown that the power-law index $n$ is related to the outer boundary parameter $\\xi _{out}$. And the MC correction to the accretion rate is considered by using the conservation law of angular momentum. Only the accretion rate at the inner edge of the disc being involved in the evolution equations of the BH, the correction to the accretion rate is made by taking the corresponding values at $r_{ms}$ in our simplified model.  In summary this model provides an analytic approach to the evolution characteristics and related physical quantities of a Kerr black hole surrounded by magnetized accretion disc. }  \\ \\\\ \\\\ {\\bf ACKNOWLEDGEMENTS} \\vspace{0.2cm} \\\\This work is supported by the National Natural Science Foundation of China under Grant No. 10173004. We are very grateful to the anonymous referee for his suggestions about the MC effects on the accretion rate and the constraints to the parameters of MC process.   \\appendix"
    },
    "0209/astro-ph0209474_arXiv.txt": {
        "abstract": "{During the past few years, secure detections of cosmic shear have been obtained, manifest in the correlation of the observed ellipticities of galaxies. Constraints have already been placed on cosmological parameters, such as the normalisation of the matter power spectrum $\\sigma_{8}$. One possible systematic contaminant of the lensing correlation signal arises from intrinsic galaxy alignment, which is still poorly constrained.  Unlike lensing, intrinsic correlations only pertain to galaxies with small physical separations, the correlation length being a few Mpc. We present a new method that harnesses this property, and isolates the lensing and intrinsic components of the galaxy ellipticity correlation function using measurements between different redshift slices. The observed signal is approximated by a set of template functions, making no strong assumptions about the amplitude or correlation length of any intrinsic alignment. We also show that the near-degeneracy between the matter density parameter $\\Omega_{\\rm m}$ and $\\sigma_{8}$ can be lifted using correlation function tomography, even in the presence of an intrinsic alignment signal. ",
        "introduction": "The tidal gravitational field of mass inhomogeneities distorts the images of distant galaxies, resulting in correlations in their observed ellipticities. This cosmological weak lensing signal, or cosmic shear, depends upon cosmological parameters and the matter power spectrum (Blandford et al. 1991; Miralda-Escud\\'e 1991; Kaiser 1992). In 2000, four teams announced the first detections of cosmic shear (Bacon et al. 2000; Kaiser et al. 2000; van Waerbeke et al. 2000; Wittman et al. 2000; Maoli et al. 2001), and more recently measurements at arcminute scales have been made using the HST (H\\\"ammerle et al. 2002; Refregier et al. 2002). Interesting constraints have already been placed on the matter power spectrum normalisation $\\sigma_{8}$, and cosmic shear is also particularly sensitive to the matter density parameter $\\Omega_{\\rm m}$, the power spectrum shape parameter $\\Gamma$ and the source redshift distribution (e.g. van Waerbeke et al. 2002a; Hoekstra et al.\\ 2002).  Future multi-colour surveys will cover hundreds of square degrees, and have the potential to place tight constraints on cosmological parameters particularly when combined with results from the CMB, SNIa and galaxy surveys (Mellier et al. 2002; van Waerbeke et al. 2002b). For example, van Waerbeke et al. (2002a) compared their constraints on $\\Omega_{\\rm m}$ and $\\sigma_{8}$ from lensing with those of Lahav et al. (2002) from the CMB, noting their near orthogonality.  Various statistical measures of the cosmic shear have been suggested; here we focus on the two-point shear correlation function $\\xi_{+}$ (hereafter denoted by $\\xi$), which is convenient since it is insensitive to gaps in the data field, unlike integrated measures such as the aperture mass statistic ${\\cal M}_{\\rm ap}$ (e.g. Schneider et al. 1998).  A possible systematic contaminant of the lensing correlation function $\\xi^{\\rm L}$ is intrinsic alignment, which may arise during the galaxy formation process. This has been subject to numerical, analytic and observational studies [e.g. Croft \\& Metzler 2000; Heavens et al. 2000 (HRH); Crittenden et al. 2001; Catelan et al. 2001; Mackey et al. 2002; Brown et al. 2002; Jing 2002; Hui \\& Zhang 2002], where amplitude estimates span a few orders of magnitude due to differences in the mechanism assumed to be responsible, and the type of galaxy considered. Nevertheless, these studies agree that the intrinsic correlation signal $\\xi^{\\rm I}$ can dominate the lensing signal for surveys with ${\\ave z}\\lesssim 0.5$.  Any correlation in ellipticities due to intrinsic alignment only arises from physically close galaxy pairs, whereas $\\xi^{\\rm L}$ is sensitive to the integrated effect of the density fluctuations out to the redshift of the nearer galaxy. It has been shown that photometric redshift information could be used to suppress $\\xi^{\\rm I}$, by downweighting or ignoring galaxy pairs at approximately the same redshift (King \\& Schneider 2002), or by downweighting nearby pairs and subtracting a model of the intrinsic alignment signal from the observed ellipticity correlation function (Heymans \\& Heavens 2002).  Motivated by the fact that intrinsic galaxy alignment is not yet well understood, we present a new method to isolate the intrinsic and lensing-induced components of the galaxy ellipticity correlation function. This method assumes that photometric redshift information is available, so that the correlation function can be measured between different redshift slices. However, no specific model for intrinsic ellipticity correlation (for instance its correlation length or redshift evolution) needs to be adopted.  In the next section we outline the method and in Sect.\\,3 we present some results in the context of a possible future survey. We discuss the results in Sect.\\,4. ",
        "conclusions": "It has been suggested that the lensing correlation function may be contaminated by intrinsic galaxy alignments. Since cosmic shear probes the matter power spectrum and enables constraints to be placed on cosmological parameters such as $\\sigma_{8}$ and $\\Omega_{\\rm m}$ (e.g. van Waerbeke et al. 2002a), it is vital to have the ability to isolate the contribution from intrinsic galaxy alignments in order to remove this systematic. Of course, intrinsic alignment is interesting in its own right: its amplitude as a function of physical separation and its evolution with redshift provides clues about the galaxy formation process.  We have demonstrated that measuring galaxy ellipticity correlation functions between redshift slices would enable the intrinsic and lensing contributions to be disentangled. The total signal is decomposed into template functions, and the fact that intrinsic alignments operate over a limited physical separation enables the intrinsic component to be isolated and subracted from the total signal. Our knowledge of the amplitude of intrinsic alignments is limited, but no strong assumption about the behaviour of the intrinsic alignment signal needs to be made. Here we considered a modest number of template functions, which can easily be augmented to cover a wider range of functional forms. For example, any intrinsic alignment signal arising at the epoch of galaxy formation may be suppressed by subsequent dynamical interaction, perhaps most pertinent to galaxy pairs with extremely small physical separations. In fact, if the reduced $\\chi^{2}$ of the best fit is significantly larger than 1, this indicates that additional template functions need to be included.  \\begin{figure} \\epsfig{file=MS3114f9.ps, width=88mm} \\caption{ The ratio of partial derivatives of $\\xi$ to $\\xi$ for our fiducial $\\Lambda$CDM model, with respect to parameters (i) $\\Omega_{\\rm m}$, (ii) $\\sigma_{8}$, (iii) $\\Gamma$ and (iv) $n$.} \\label{dxp} \\end{figure}  Our choice of template functions is of course fairly arbitrary. We have taken functions which have approximately the behaviour expected from a cosmic shear measurement. Alternatively, one could consider a set of generic basis functions, which however, owing to the dependence on three variables, would require a fairly large set of functions. Another natural choice of the template functions could be the following: assuming a reasonable guess for the cosmological model, characterised by the parameters $\\vc \\pi_0$, the correlation function for neighbouring models could be written as \\begin{eqnarray} &&\\bar\\xi(\\theta,\\bar z_i,\\bar z_j; \\vc \\pi) \\approx\\\\\\nonumber &&\\bar\\xi(\\theta,\\bar z_i,\\bar z_j; \\vc \\pi_0) +(\\vc \\pi-\\vc \\pi_0) {\\partial \\bar\\xi \\over \\partial \\vc \\pi}(\\theta,\\bar z_i,\\bar z_j,\\vc \\pi_0)\\;, \\end{eqnarray} and therefore, the set consisting of $\\bar\\xi(\\vc \\pi_0)$ and its partial derivatives with respect to the relevant cosmological parameters would provide a useful set of template functions. As an illustration, in Fig.\\,\\ref{dxp} we have plotted the partial derivatives of $\\xi(\\theta)$ for the case where no redshift information is available, i.e. the derivatives of the redshift-averaged correlation function, again with ${\\ave{\\bar z}}\\sim 1$. For our cosmological model (characterised by $\\vc \\pi_0$) we take the fiducial $\\Lambda$CDM model. Derivatives are taken with respect to (i) $\\Omega_{\\rm m}$, (ii) $\\sigma_{8}$, (iii) $\\Gamma$ and (iv) the primordial spectral index $n$ (where our fiducial model is scale-invariant $n=1$). We plot the ratio $\\frac{\\partial\\xi}{\\partial\\pi_{i}}/\\xi$, where the numerator is denoted by $\\Delta(\\xi,\\pi_{i})$.  In the limiting case where one curve is a scaled version of another, $\\Delta(\\xi,\\pi_{i})\\propto\\Delta(\\xi,\\pi_{j})$, it is impossible to first order to break the degeneracy between parameters $\\pi_{i}$ and $\\pi_{j}$. We again see a nice illustration of the near degeneracy between $\\Omega_{\\rm m}$ and $\\sigma_{8}$, manifest in the similarity between $\\Delta(\\xi,\\Omega_{\\rm m})$ and $\\Delta(\\xi,\\sigma_{8})$. Since $\\sigma_{8}^{2}$ enters into the linear power spectrum just as a prefactor, on large angular scales (i.e. in the linear regime), the curve for $\\sigma_{8}$ tends to a constant, $\\Delta(\\xi,\\sigma_{8})/\\xi\\to 2 \\sigma_8^{-1}$ for large separations.  Another feature to note is that on the scale of a few arcminutes the curves for $\\Gamma$ and $n$ change sign, implying that there is less degeneracy between these parameters and either of $\\Omega_{\\rm m}$ or $\\sigma_{8}$.  There are several ways in which the method and results discussed here could be used. One way would be to consider the resulting split into intrinsic correlations and lensing signal as the final result, and to compare the resulting functions with theories of galaxies formation which predict the intrinsic alignment signal, and cosmological models predicting the shear correlation function. The resulting fits are, however, difficult to interpret statistically, i.e. the error bars on the shear correlation function are difficult to obtain. An alternative would be to consider the fitted intrinsic signal only, subtract it from the ellipticity correlation function, and consider the result as the shear correlation function, together with the corresponding error bars. Subsequently, the correlation function can then be used for the redshift-weighting method of King \\& Schneider (2002), of course yielding much smaller contributions from the intrinsic correlations than for the unsubtracted data.  Furthermore, the resulting model for the intrinsic correlation function could also be used as input for the subtraction method discussed in Heymans \\& Heavens (2002).  In addition to providing a key to the suppression of any intrinsic alignment signal, photometric redshift estimates enable much tighter constraints to be placed on cosmological parameters obtained from cosmic shear surveys, as demonstrated by Hu (1999). Although our prime goal in this paper is not the constraint of cosmological parameters, we have illustrated that the degeneracy between $\\Omega_{\\rm m}$ and $\\sigma_{8}$ can be lifted by observing correlation functions between redshift slices, even when an intrinsic alignment systematic is present."
    },
    "0209/astro-ph0209197_arXiv.txt": {
        "abstract": "We explain the physics of compressional heating of the deep interior of an accreting white dwarf (WD) at accretion rates low enough so that the accumulated hydrogen burns unstably and initiates a classical nova (CN).  In this limit, the WD core temperature ($T_c$) reaches an equilibrium value ($T_{\\rm c,eq}$) after accreting an amount of mass much less than the WD's mass. Once this equilibrium is reached, the compressional heating from within the envelope exits the surface. This equilibrium yields useful relations between the WD surface temperature, accretion rate and mass that can be employed to measure accretion rates from observed WD effective temperatures, thus testing binary evolution models for cataclysmic variables. ",
        "introduction": "Cataclysmic variables (CV) are formed when the WD made during a common envelope event finally comes into contact with its companion as a result of gravitational radiation losses over a few Gyr (see \\cite{how}). The WD will have cooled during this time; a $0.20 M_\\odot$ He WD would have $T_c=3.3\\times 10^6 \\ {\\rm K} $ at 4 Gyr \\cite{alt}, whereas a $0.6M_\\odot$ C/O WD would have $T_c=2.5\\times 10^6 {\\rm K}$ in 4 Gyr \\cite{sal}. These give effective temperatures $T_{\\rm eff}\\approx 4500-5000$ K.  A subset of the CVs, called Dwarf Novae (DN), contain a WD accreting at low time-averaged rates $\\timav<10^{-9}M_\\odot \\ {\\rm yr}^{-1}$, where the accretion disk is subject to a thermal instability which causes it to rapidly transfer matter onto the WD (at $\\dot M \\gg \\timav$) for a week once every month to year. The $\\dot M$ onto the WD is often low enough between outbursts that the UV emission is dominated by the internal luminosity of the WD, allowing for a measurement of the WD luminosity, yielding $T_{\\rm eff}>10,000\\ {\\rm K}$ \\cite{sion}.  The WD is clearly hotter than expected for its age, providing evidence of the thermal impact of prolonged accretion \\cite{sion}. The theoretical illumination of this phenomena will aid our understanding of CV evolution, ejection of material during CN events, and finally allow for theoretical seismology of the one known non-radial pulsator, GW Lib \\cite{zyl,szkody}. We summarize here both our published work \\cite{tow02,towbil} and work in progress. ",
        "conclusions": ""
    },
    "0209/astro-ph0209183_arXiv.txt": {
        "abstract": "We use simple analytic reasoning to identify physical processes that drive the evolution of the cosmic star formation rate, $\\dot\\rho_\\star$, in cold dark matter universes.  Based on our analysis, we formulate a model to characterise the redshift dependence of $\\dot\\rho_\\star$ and compare it to results obtained from a set of hydrodynamic simulations which include star formation and feedback.  We find that the cosmic star formation rate is described by two regimes.  At early times, densities are sufficiently high and cooling times sufficiently short that abundant quantities of star-forming gas are present in all dark matter halos that can cool by atomic processes.  Consequently, $\\dot\\rho_\\star$ generically rises exponentially as $z$ decreases, independent of the details of the physical model for star formation, but dependent on the normalisation and shape of the cosmological power spectrum.  This part of the evolution is dominated by gravitationally driven growth of the halo mass function.  At low redshifts, densities decline as the universe expands to the point that cooling is inhibited, limiting the amount of star-forming gas available.  We find that in this regime the star formation rate scales approximately as $\\dot\\rho_\\star\\propto H(z)^{4/3}$, in proportion to the cooling rate within halos.  We demonstrate that the existence of these two regimes leads to a peak in the star formation rate at an intermediate redshift $z = z_{\\rm peak}$.  We discuss how the location of this peak depends on our model parameters.  Only star formation efficiencies that are unrealistically low would delay the peak to $z\\simeq 3$ or below, and we show that the peak cannot occur above a limiting redshift of $z \\approx 8.7$.  For the star formation efficiency adopted in our numerical simulations, $z_{\\rm peak} \\approx 5 - 6$.  We derive analytic expressions for the full star formation history and show that they match our simulation results to better than $\\simeq$10\\%.  Using various approximations, we reduce the expressions to a simple analytic fitting function for $\\dot\\rho_\\star$ that can be used to compute global cosmological quantities that are directly related to the star formation history.  As examples, we consider the integrated stellar density, the supernova and gamma-ray burst (GRB) rates observable on Earth, the metal enrichment history of the Universe, and the density of compact objects.  We also briefly discuss the expected dependence of the star formation history on cosmological parameters and the physics of the gas. ",
        "introduction": "\\renewcommand{\\thefootnote}{\\fnsymbol{footnote}} \\footnotetext[1]{E-mail: lars@cfa.harvard.edu} \\footnotetext[3]{\\hspace{0.03cm}E-mail: volker@mpa-garching.mpg.de}  The history of cosmic star formation is of fundamental importance to cosmology, not only to galaxy formation itself, but also for ongoing efforts to determine cosmological parameters and the matter content of the Universe.  Over the past decade, various attempts have been made to directly map out the evolution of star formation observationally \\citep[e.g.][]{Gal95,Mad96,Mad98,Lil96,Cow96,Cow99,Con97,Hug98,Trey98, Tresse98,Pas98,Steid99,Flo99,Gron99,Hogg01,Bald02,Lan2002,Wil02}. Obtaining precise measurements of the star formation rate density, $\\dot\\rho_\\star$, is made challenging, however, by the difficult nature of these observations and also by uncertainties in systematic effects such as dust extinction.  Partly for this reason, there is strong motivation for predicting $\\dot\\rho_\\star$ theoretically to provide a framework for interpreting the data.  Various theoretical efforts have been made to calculate $\\dot\\rho_\\star$, using either semi-analytic models \\citep{Wh91,Cole1994,Bau98,Som00}, or numerical simulations \\citep{Weinberg99,Pea2000,Nag00,Nag01,Asc02}.  Unfortunately, due to the complexity of the physics underlying galaxy formation, the predicted behaviour for $\\dot\\rho_\\star(z)$ can be quite sensitive to the model adopted to describe star formation and associated feedback processes.  Perhaps because of this difficulty, there have been few attempts to determine whether some aspects of the expected evolution of the star formation density in cold dark matter cosmologies are relatively insensitive to the details of the physics of star formation. If such a ``generic'' behaviour exists within a reasonably broad class of physical models, it should be possible to make robust predictions for the shape of the star formation history in cold dark matter universes that could be confronted with observations to test the currently favoured paradigm of hierarchical galaxy formation.  In this paper, we examine this issue in detail.  We are motivated by the numerical results presented in \\citet{SprHerSFR}, where we used a large set of hydrodynamic simulations to infer the evolution of the cosmic star formation rate density from high redshift to the present. These simulations included a novel description for star formation and feedback processes within the interstellar medium \\citep{SprHerMultiPhase} and a novel formulation of the equations of motion \\citep{SprHerEnt}.  The broad range of scales encompassed by our set of simulations, together with extensive convergence tests, enabled us to obtain a converged prediction for $\\dot\\rho_\\star (z)$ within this model for galaxy formation.  The cosmic star formation history we inferred peaks at a redshift $z_{\\rm peak} \\sim 5.5$, declining roughly exponentially towards both low and high redshift.  Here, we establish a physical basis for the particular form of the star formation history predicted by our simulations.  This makes it possible to arrive at a clearer understanding of the physics that drives the evolution of the cosmic star formation history, and allows us to justify specific analytic fitting functions for the full star formation history.  Such closed-form descriptions are particularly useful for computing derived quantities that directly depend on the star formation history and for relating theoretical predictions to observations.  This paper is organised as follows. In Section~\\ref{secfit}, we present our analytic fitting function for the cosmic star formation history, followed in Section~\\ref{secphysbasis} by a detailed analysis of the physical basis for this particular functional form. In Section~\\ref{secderived}, we then compute a number of derived quantities based on the star formation history.  We briefly discuss the expected dependence on cosmological parameters and possible effects of metal enrichment in Section~\\ref{seccosmology}, and, finally, we summarise and conclude in Section~\\ref{secconclusions}. ",
        "conclusions": "\\label{secconclusions}  We have formulated an analytical model to identify physical processes that play an important role in determining the evolution of the cosmic star formation rate density, $\\dot\\rho_\\star (z)$.  Using this model, we obtain simple closed-form expressions for $\\dot\\rho_\\star (z)$ which match hydrodynamic simulations that include star formation and feedback to a level of $\\approx 10 \\%$.  Our model, therefore, provides a framework for interpreting both theoretical and observational estimates of $\\dot\\rho_\\star (z)$.  Our analysis shows that the evolution of the cosmic star formation rate is characterised by a number of generic features in hierarchical universes.  These properties depend on cosmological parameters but are largely insensitive to the detailed physics of star formation.  In particular, we have identified two broad regimes of star formation that are separated by a peak in $\\dot\\rho_\\star (z)$ at $z=z_{\\rm peak}$. At high redshifts, $z > z_{\\rm peak}$, cooling is very efficient and halos contain abundant quantities of star-forming gas.  In this regime, the dominant contribution to the global star formation rate comes from the highest mass halos present at any time that are not unusually rare.  Consequently, $\\dot\\rho_\\star (z)$ follows the evolution of the exponential part of the halo mass function.  The logarithmic slope of this phase of evolution depends on properties of the cosmology but not on the details of star formation, which only affect the overall normalisation.  At low redshifts, $z < z_{\\rm peak}$, cooling becomes inefficient, and the supply of star-forming gas is regulated by the cooling rate.  In this regime, $\\dot\\rho_\\star (z)$ gradually declines from its maximum at $z = z_{\\rm peak}$ to $z=0$ as a power-law function of the expansion rate, $\\dot\\rho_\\star (z) \\propto H(z)^q$.  Typically, we find $q\\approx 4/3$, weakly dependent on the gas density profiles within dark matter halos. The scaling may also be altered slightly if metal enrichment becomes important in halos at late times.  To the extent that our results apply to the real Universe, observations of $\\dot\\rho_\\star (z)$ at $z < z_{\\rm peak}$ should be well-fitted by a functional form $\\dot\\rho_\\star (z) \\propto H(z)^q$.  Thus, our prediction for the evolution of the cosmic star formation rate is, in principle, testable by accurate measurements of $\\dot\\rho_\\star (z)$ at low redshifts.  We have shown that the existence of a peak in the star formation rate at a redshift $z = z_{\\rm peak}$ is generic, but that the value of $z_{\\rm peak}$ depends on assumptions about the characteristic gas consumption timescale, as parameterised by $t_0^\\star$.  For plausible values of $t_0^\\star$ we find that $z_{\\rm peak}$ should be restricted to the range $3 \\simlt z_{\\rm peak} < 8.7$, with a firm upper limit corresponding to instantaneous gas consumption.  In our numerical simulations, in which we chose $t_0^\\star$ to reproduce the empirical Kennicutt Law, $z_{\\rm peak} \\approx 5.5$.  Overall, we broadly predict that the cosmic star formation history in hierarchical universes should have a generic form, rising exponentially at first, peaking at $z_{\\rm peak}$, and then declining to $z=0$ as a power-law function of $H(z)$.  The logarithmic slopes on either side of the peak are mainly determined by cosmology, but the overall normalisation of $\\dot\\rho_\\star (z)$ and the value of $z_{\\rm peak}$ are sensitive to assumptions about gas and star formation physics.  The generic form we propose is compactly summarised by e.g. equation~(\\ref{eqnnewfit}), where the value of $\\beta$ is fixed by cosmology and $\\dot\\rho_\\star(0)$ and $\\alpha$ determine the overall normalisation and the location of $z_{\\rm peak}$.  We note that there are implicit assumptions in our description of the star formation physics that can influence e.g. $\\alpha$ and $z_{\\rm peak}$ in fits of the form of equation~(\\ref{eqnnewfit}).  For example, for simplicity we have assumed that cooling rates are those appropriate for a H/He plasma of primordial abundance.  It is believed that at very high redshift, $z\\sim 20 - 30$, molecular cooling is an important physical mechanism for early star formation (e.g. Bromm et al. 1999, Abel et al. 2002).  Thus, our results are not applicable to Population III (e.g. Carr 1984) star formation and will not properly characterise $\\dot\\rho_\\star (z)$ until star formation globally resembles that at lower redshifts.  How and when the Universe made this transition is uncertain.  Simple estimates suggest that it may have occurred at $z\\sim 15-20$, as the Universe began to become chemically enriched (e.g. Mackey et al. 2002), but more detailed studies of this fundamental issue are clearly needed.  In our simulations, we have also ignored the contribution of metal line cooling to the overall cooling rate.  However, we have been able to estimate the possible importance of this process using our analytical model, as described in Section~\\ref{secmetalcooling}. While our conclusions depend in detail on uncertainties in how efficiently enriched gas is mixed into galactic halos and the IGM, we find that metal line cooling does not affect our results at early times and likely has only a modest influence on the behaviour of $\\dot\\rho_\\star (z)$ at low redshift.  In particular, if metals carried by galactic winds are efficiently mixed with the remaining gas in the universe, enhancements in the cooling and star formation rates at low redshift increase the stellar density only by $\\sim 20 - 30\\%$ by $z=0$, boosting the star formation rate by a similar factor for $z\\simlt 3$.  We note that when we previously compared our simulations to observations, there was an indication that our predicted $\\dot\\rho_\\star (z)$ was perhaps low at $z\\sim 1$ (e.g. figure 12 in Springel \\& Hernquist 2002b).  If we take this discrepancy seriously, given observational uncertainty, then metal cooling would boost the predicted star formation rate into the observed range, at least according to our present simple estimates, without violating constraints on the total stellar density at $z=0$.  Metal enrichment can also, in principle, influence the location of $z_{\\rm peak}$ and shift it to lower $z$.  However, our current estimate of this effect suggests that the peak will still lie at a relatively high redshift, $z_{\\rm peak}\\sim 5$, as indicated by Figure~\\ref{figmetaleffect}, and would thus not substantially alter our predictions for the evolution of the stellar density or the mean age of the stellar population with redshift.  Furthermore, we do not believe that the overall form of the star formation history we considered here would be altered significantly.  However, detailed hydrodynamical simulations of metal enrichment processes will ultimately be required to more accurately constrain the relevance of metal cooling for our modeling."
    },
    "0209/astro-ph0209460_arXiv.txt": {
        "abstract": "In this paper, we present the methodology of photometric redshift determination with the BATC 15-color system by using $hyperz$ program. Both simulated galaxies and real galaxies with known redshifts were used to estimate the accuracy of redshifts inferred from the multicolor photometry. From the test with simulated galaxies, the uncertainty in the inferred redshifts is about $0.02\\sim0.03$ for a given range of photometric uncertainty of $0\\hspace{0.1cm}{.}\\hspace{-0.15cm}^{m}05 \\sim 0\\hspace{0.1cm}{.}\\hspace{-0.15cm}^{m}10$.  The results with the 27 real galaxies are in good agreement with the simulated ones. The advantage of using BATC intermediate-band system to derive redshift is clear through the comparison with the $UBVRI$ broad-band system.  The accuracy in redshift determination with BATC system is mainly affected by the selection of filters and the photometric uncertainties in the observation. When we take the limiting magnitudes of the 15 filters into account, we find that redshift can be determined with good accuracy for galaxies with redshifts less than 0.5, using only filters with central wavelengths shorter than 6000{\\AA}. ",
        "introduction": "In multicolor photometric surveys, the redshifts of a large number of objects in a given field can readily be obtained from the color information. Although multicolor photometry does not yield redshift information as accurately as spectroscopy does, it has the virtues of deeper limiting magnitude, faster batch reduction, and better time-saving from the simultaneous determination of redshifts of many objects in a given field. With the redshifts determined for a large sample of galaxies via multicolor photometry, astronomers are able to study statistically the evolution of galaxies in number as well as in luminosity (Pascarelle et al. 1998; Volonteri et al. 2000;  Gal et al. 2000).  In a simulation using 40 bands, the efficiency of photometric redshift determination for faint objects is comparable to slitless spectroscopy (Hickson et al. 1994). The techniques of photometric redshift are thus said to be not only the ``poor person's redshift machine'' but also the only viable way so far to acquire redshift information for a large quantity of faint objects, because the majority of these objects will still remain beyond the limit of spectroscopy in the foreseeable future (Bolzonella et al. 2000).  A number of computer codes performing photometry fitting have been developed and applied to data acquired in several survey projects, such as HDF, SDSS, CADIS, etc (Sowards et al. 1999; Yahata et al. 2000; Wolf et al. 2001). Two methods have been widely used, one is the ``Empirical Training Set'' method (Connolly et al. 1995; Wang et al. 1999), the other is the ``Spectral Energy Distribution'' (hereafter SED) fitting method .  The empirical training set method determines redshifts by the empirical linear relation between magnitudes (or colors) and redshifts. Although this method requires no assumptions on galaxy spectra and their evolution, {\\bf there are still} a few shortages. For example, the empirical relation changes with the data obtained with different filter sets.  Furthermore, in high redshifts, the sample of spectroscopic templates becomes smaller and less complete, which make the redshift determination less reliable.  The SED fitting method, on the other hand, is based on the fit of the overall shape of a spectrum, i.e., it relies on the detection of apparent spectral properties such as Lyman-forest and Balmer Jump, etc. The fitting is performed by comparing the observed SEDs to the template spectra acquired using the same photometric system (Corbin et al. 2000; Fontana et al. 2000).  The BATC (Beijing-Arizona-Taipei-Connecticut) large-field sky survey in 15 intermediate-band colors commenced in 1994. Over the years, the survey has produced a database, which can be used to derive the redshifts of nearby galaxies between $z=0$ and $0.5$, providing essential information regarding the structure of the local universe and the nearby galaxy clusters, especially Abell clusters (Yuan et al. 2001). The purpose of the study in this paper is to estimate the accuracy of $z_{phot}$ using the BATC 15-color photometric system.  The content of this paper is as follows. In $\\S$ 2 we describe the BATC photometric system and the observations of two fields used as the real sample. The procedures of data reduction are given briefly in $\\S$ 3. The application of $z_{phot}$ code $hyperz$ is described in $\\S$ 4. In $\\S$ 5, the comparison between the BATC system and the $\\ UBVRI$ system using the simulation test is shown, along with the filter dependence of $z_{phot}$'s. We compare the results of $z_{phot}$'s with the spectroscopic redshifts $z_{spec}$'s in $\\S$ 6. Discussions and conclusions are presented in $\\S$ 7. ",
        "conclusions": "In this paper, with the help of $hyperz$, we examine the accuracy of redshift estimation by comparing the BATC 15 intermediate-band photometric system to the $\\ UBVRI$ broad-band photometric system using simulated spectra.  We find that with the BATC system we can obtain fairly accurate redshift estimation.  This advantage comes from the careful selection of the 15-color filter set in the begginning of the BATC survey. The $z_{phot}$ determination of the spectroscopic sample in the BATC fields is also checked. The main results are listed as follows:  1. The uncertainty in photometric redshifts comes mainly from the photometric errors.   We have made assessment of the accuracy with simulation. The dispersion can reach as low as $\\sigma_{z}=0.02\\sim 0.03$ with almost no catastrophic dropout for the typical photometric uncertainty from  $\\Delta m=0\\hspace{0.1cm}{.}\\hspace{-0.15cm}^{m}05$ to  $\\Delta m=0\\hspace{0.1cm}{.}\\hspace{-0.15cm}^{m}1$;  2. The objects that can be observed with BATC survey are generally limited to redshift range of 0 to 0.5,  hence the filters whose central wavelengths are shorter than 6000{\\AA} are especially important for the detection of the 4000{\\AA} Balmer break.  It has further been shown that we can use only the filters blueward than 6000{\\AA} for the accurate determination of redshift and save significant amount of telescope time;  3. For the 10 brightest galaxies centered in Abell 566, the results show that the accuracy of photometric redshift determination is $\\sigma_{z}=0.008$ for $z_{step}$ of 0.05 and 0.005, with systematic errors of $\\overline {\\Delta z}=-0.007$ and $-0.002$, respectively. For the 17 galaxies, which have spectroscopic measurements in NED, the accuracy is $\\sigma_{z}=0.021$."
    },
    "0209/astro-ph0209526_arXiv.txt": {
        "abstract": "{A millimetre molecular line survey of seven high mass-loss rate carbon stars in both the northern and southern skies is presented.  A total of 196 emission lines (47 transitions) from 24 molecular species were detected. The observed CO emission is used to determine mass-loss rates and the physical structure of the circumstellar envelope, such as the density and temperature structure, using a detailed radiative transfer analysis. This enables abundances for the remaining molecular species to be determined. The derived abundances generally vary between the sources by no more than a factor of five indicating that circumstellar envelopes around carbon stars with high mass-loss rates have similar chemical compositions. However, there are some notable exceptions.  The most striking difference between the abundances are reflecting the spread in the $^{12}$C/$^{13}$C-ratio of about an order of magnitude between the sample stars, which mainly shows the results of nucleosynthesis. The abundance of SiO also shows a variation of more than an order of magnitude between the sources and is on the average more than an order of magnitude more abundant than predicted from photospheric chemistry in thermal equilibrium. The over-abundance of SiO is consistent with dynamical modelling of the stellar atmosphere and the inner parts of the wind where a pulsation-driven shock has passed. This scenario is possibly further substantiated by the relatively low amount of CS present in the envelopes. The chemistry occurring in the outer envelope is consistent with current photochemical models. ",
        "introduction": "The chemistry associated with carbon stars has long been known to be rich and complex in comparison to the alternative O-rich regime (i.e., where C/O\\,$<$\\,1).  This is in part due to the favourable bonding of the carbon atom, enabling long chains and complex species to form. Most of the current understanding of carbon stars has come from both observational and theoretical work on the high-mass-losing carbon star, \\object{IRC+10216}.  This source, which lies within 200\\,pc and presents an ideal specimen for the study of carbon-rich envelopes, has been mapped interferometrically in various molecular species \\citep[e.g.][]{BiegingTafalla1993, DayalBieging1993, DayalBieging1995, Gensheimer_etal1995, Guelin_etal1993, Guelin_etal1996, Lucas_etal1995, LucasGuelin1999} and has had models of its dust \\citep[a good summary is given by][]{Menshchikov_etal2001} and chemistry \\citep[e.g.][]{Millar_etal2000} constructed.  These tools have produced groundbreaking results and have been used to set a paradigm for what has come to be known as \\textquotedblleft carbon chemistry\\textquotedblright\\ in connection with evolved stars. However, the accuracy of employing \\object{IRC+10216} chemistry to similar carbon stars has been little-tested due to the difficulties in observing them.  Much work has been done on the carbon-rich post-AGB sources \\object{CRL 618} and \\object{CRL 2688}, and the chemistry of \\object{CRL 618} in particular has been modelled by \\citet{Woods_etal2002}.  Detailed chemical studies of carbon stars on the AGB have been few in number, but examples include the molecular line survey of \\object{IRAS 15194--5115}, a peculiar $^{13}$C-rich star \\citep{Nyman_etal1993}.  Carbon star surveys which include molecular-line comparisons are fewer, and have been limited in the number of lines observed. \\citet{Olofsson_etal1993b} detected some 40 stars in a handful of species other than CO. The sample of \\citet{Bujarrabal_etal1994} included 16 carbon stars, with up to ten molecular lines observed in each.  A more recent survey by \\citet{Olofsson_etal1998} detected 22 carbon stars in up to 6 molecular lines.  \\begin{table*} \\caption{Positions, luminosities, periods and calculated distances of the sample of carbon stars.} \\label{stellardata} \\begin{flushleft} \\begin{tabular}{lllcccccc} \\hline\\hline IRAS No.    & Other cat. name & B1950 Co-ords.   & $P$    & $L$      & $D$     & $T_*$   & $T_{\\mathrm{d}}$ & $L_{\\mathrm{d}}$/$L_*$\\\\ &                 &                  & [days] & [L$_{\\odot}$]  & [pc]    & [K]     & [K]              \\\\ \\hline $07454-7112$ & \\object{AFGL\\,4078}     & 07:45:25.7 $-$71:12:18   &  ---       &  9\\,000$^a$                   & 710  & 1\\,200 &  \\phantom{0}710 & 4.3\\\\ $09452+1330$ & \\object{IRC+10216}      & 09:45:15.0 $+$13:30:45  & 630     & 9\\,600\\phantom{$^b$} & 120 & ---      & \\phantom{0}510 & ---\\\\ $10131+3049$ & \\object{CIT\\,6}         & 10:13:11.5 $+$30:49:17  & 640     & 9\\,700\\phantom{$^b$} & 440 & 1\\,300 & \\phantom{0}510 & 6.7\\\\ $15082-4808$ & \\object{AFGL\\,4211}     & 15:08:13.0 $-$48:08:43   & ---        & 9\\,000$^a$                     & 640 & --- &  \\phantom{0}590 & ---\\\\ $15194-5115$ & ---                     & 15:19:26.9 $-$51:15:19   & 580     & 8\\,800\\phantom{$^b$} & 600 &  \\phantom{0}930 & \\phantom{0}480 & 2.2\\\\ $23166+1655$ & \\object{AFGL\\,3068}     & 23.16.42.4 $+$16:55:10  & 700     & 7\\,800\\phantom{$^b$} & 820 & --- & 1\\,000 & --- \\\\ $23320+4316$ & \\object{IRC+40540}      & 23:32:00.4 $+$43:16:17  & 620     & 9\\,400\\phantom{$^b$} & 630 & 1\\,100 & \\phantom{0}610 & 6.6\\\\ \\hline \\end{tabular} {\\footnotesize \\begin{enumerate} \\renewcommand{\\labelenumi}{(\\alph{enumi})} \\item Assumed value. \\end{enumerate} } \\end{flushleft} \\end{table*}  The survey work presented here purports to be the most complete and consistent molecular-line survey in AGB carbon stars to date, covering high mass-loss rate objects in both the northern and the southern sky.  Previously unpublished spectra of five stars (\\object{IRAS 15082--4808}, \\object{IRAS 07454--7112}, \\object{CIT 6}, \\object{AFGL 3068} and \\object{IRC+40540}) are presented, and spectra taken towards \\object{IRC+10216} and \\object{IRAS 15194--5115} with the Swedish-ESO Submillimetre Telescope \\citep[SEST;][]{Nyman_etal1993} and \\object{IRC+10216} with the Onsala Space Observatory (OSO) 20\\,m telescope are used to supplement the survey.  Comparison of data from \\object{IRC+10216} taken with both telescopes affords a high degree of confidence in the relative calibration that can be derived.  Up to 51 molecular lines were observed in the sample of 7 high-mass-losing carbon stars, of which 47 were clearly detected. Mass-loss rates, dust properties and the $^{12}$CO/$^{13}$CO-ratio are calculated self-consistently using a radiative transfer method \\citep{SchoierOlofsson2000, SchoierOlofsson2001, Schoier_etal2002}.  An approach similar to that of \\citet{Nyman_etal1993} is adopted to calculate fractional abundances (including upper limits), and a detailed analysis of the comparison between the calculated abundances is carried out.  Hence, Sect.~\\ref{obs} details the observations carried out and the instrumentation used.  Section~\\ref{results} gives the observational results, including a presentation of various spectra. Section~\\ref{modelling} details the NLTE radiative transfer code used to determine the envelope parameters and Sect.~\\ref{abundances} explains the method of calculating chemical abundances.  The results and deductions are discussed in Sect.~\\ref{discuss}. ",
        "conclusions": "The seven high mass-loss rate carbon stars presented here exhibit rich spectra at millimetre wavelengths with many molecular species readily detected. A total of 47 emission lines from 24 molecular species were detected for the sample stars.  The mass-loss rate and physical structure of the circumstellar envelope, such as the density and temperature profiles, was carefully estimated based upon a detailed radiative transfer analysis of CO. The determination of the mass-loss rate enables abundances for the remaining molecular species to be calculated. The derived abundances typically agree within a factor of five indicating that circumstellar envelopes around carbon stars have similar molecular compositions.  The most striking difference between the abundances are reflecting the spread in the $^{12}$C/$^{13}$C-ratio of about an order of magnitude between the sample stars.  Also, the high abundance of SiO in the envelopes indicates that a shock has passed through the gas in the inner parts of the envelope. This is further corroborated by the relatively low amounts of CS and possibly HCN.  The abundances of species that are produced in the outer parts of the wind can be reasonably well explained by current photochemical models."
    },
    "0209/hep-th0209261_arXiv.txt": {
        "abstract": " ",
        "introduction": "It has been recently  suggested that there might exist some extra spatial dimensions, not in the traditional Kaluza-Klein sense where the extra-dimensions are compactified on a small enough radius to evade detection in the form of Kaluza-Klein modes, but in a setting where the extra dimensions could be large, under the assumption that {\\it ordinary matter is confined} onto a three-dimensional subspace, called {\\it brane} (more precisely `3-brane', referring to the  three spatial dimensions) embedded in a larger space, called {\\it bulk}.  Altough the idea in itself is not completely new \\cite{precursors}, the fact that it might be connected to recent string theory developments has suscitated a renewed interest. In this respect, an inspiring input has been the model suggested by Horava and Witten \\cite{hw}, sometimes dubbed {\\it M-theory}, which describes the low energy effective theory corresponding to the strong coupling limit of $E_8\\times E_8$ heterotic string theory. This model is associated with an eleven-dimensional  bulk spacetime with 11-dimensional supergravity, the eleventh dimension being compactified via a  $Z_2$ orbifold symmetry. The two fixed points of the orbifold symmetry define two 10-dimensional spacetime boundaries, or 9-branes, on which the gauge groups are defined. Starting from this configuration, one can distinguish three types of spatial dimensions: the orbifold dimension, three large dimensions corresponding to the ordinary spatial directions and finally six additional dimensions, which can be compactified  in the usual Kaluza-Klein way. It turns out that the orbifold dimension might be  larger than the six Kaluza-Klein extra dimensions, resulting in an intermediary picture with a five-dimensional spacetime, two boundary 3-branes, one of which could be our universe, and a ``large'' extra-dimension. This model provides the motivating framework for many of the brane cosmological models.  This concept of a 3-brane has also been used  in a purely phenomenological way by Arkani-Hamed, Dimopoulos and Dvali \\cite{add} (ADD) as a possible solution to the hierarchy problem in particle physics. Their setup is extremely simple since they consider a {\\it flat} $(4+n)$-dimensional spacetime, thus with $n$ compact extra dimensions with, for simplicity, a torus topology and a common size $R$. From the fundamental (Planck) $(4+n)$-dimensional mass $M_*$, which embodies the coupling of gravity to matter from the $(4+n)$-dimensional point of view, one can  deduce the effective four-dimensional Planck mass $M_P$, either by integrating the Einstein-Hilbert action over the extra dimensions, or by using directly Gauss' theorem. One then finds \\beq M_P^2=M_*^{2+n}R^n. \\eeq On sizes much larger than $R$, $(4+n)$-dimensional gravity behaves effectively as our usual four-dimensional gravity, and the two can be distinguished only on scales sufficiently small, of the order of $R$ or below. The simple but crucial remark of ADD is that a fundamental mass $M_*$ of the order of the electroweak scale can explain  the huge four-dimensional Planck mass $M_P$  we observe, {\\it provided the volume in the extra dimensions is very large}. Of course, such a large size for the usual extra-dimensions \\`a la Kaluza-Klein is forbidden by collider constraints, but is allowed when ordinary matter is confined to a three-brane. In contrast, the constraints on the behaviour of gravity are much weaker since the usual Newton's law has been verified experimentally only down to a fraction of millimeter\\cite{grav_exp}. This leaves room for extra dimensions as large as this millimeter experimental bound.  The treatment of extra-dimensions can be refined by allowing the spacetime to be curved, by the presence of the brane  and possibly by the bulk. In this spirit, Randall and Sundrum have proposed a very interesting model\\cite{rs99a,rs99b} with an Anti de Sitter (AdS) five-dimensional spacetime (i.e. a maximally symmetric spacetime with a negative cosmological constant), and shown that, for an appropriate tension of the brane representing our universe, one recovers effectively four-dimensional gravity even with an {\\it infinite extra-dimension}. Another model \\cite{dgp}, which will not be discussed in this review, is characterized, in contrast with the previous ones, by a gravity which becomes five-dimensional at {\\it large} scales and it could have interesting applications for the present cosmological acceleration\\cite{deffayet}.   The recent concept of extra-dimensions with branes has been explored in  a lot of aspects of particle physics, gravity, astrophysics and cosmology \\cite{maartens,rubakov}. The purpose of this review is  to present  a very specific, although very active,  facet of this vast domain, dealing with the cosmological behaviour of brane models when the curvature of spacetime along the extra-dimensions, and in particular the brane self-gravity, is explicitly taken into account. For technical reasons,  this can be easily studied in the context of codimension one spacetimes and we will therefore restrict this review to the case of {\\it five-dimensional} spacetimes. Even in this restricted area, the number of works during the last few years is so large that this introductory review is not intended to be comprehensive but will focus on some selected aspects. This also implies that the bibliography is far from exhaustive. ",
        "conclusions": ""
    },
    "0209/astro-ph0209076_arXiv.txt": {
        "abstract": "We have obtained \\emph{FUSE} spectra of $\\sigma$ Her, a nearby binary system, with a main sequence primary, that has a Vega-like infrared excess. We observe absorption in the excited fine structure lines \\ion{C}{2}$^{*}$ at 1037 \\AA, \\ion{N}{2}$^{*}$ at 1085 \\AA, and \\ion{N}{2}$^{**}$ at 1086 \\AA \\ that are blueshifted by as much as $\\sim$30 km/sec with respect to the star. Since these features are considerably narrower than the stellar lines and broader than interstellar features, the \\ion{C}{2} and \\ion{N}{2} are circumstellar. We suggest that there is a radiatively driven wind, arising from the circumstellar matter, rather than accretion as occurs around $\\beta$ Pic, because of $\\sigma$ Her's high luminosity. Assuming that the gas is liberated by collisions between parent bodies at 20 AU, the approximate distance at which blackbody grains are in radiative equilibrium with the star and at which 3-body orbits become unstable, we infer dM/dt $\\sim$ 6 $\\times$ $10^{-12}$ M$_{\\sun}$ yr$^{-1}$. This wind depletes the minimum mass of parent bodies in less than the estimated age of the system. ",
        "introduction": "Planets are believed to form within circumstellar disks around young main sequence stars (with ages $\\leq$100 Myr) composed of dust and gas (Beckwith \\& Sargent 1996). Dusty circumstellar disks, with radii comparable to the distance between the Sun and the Kuiper Belt, have been imaged around several young, nearby main sequence stars (Schneider et al. 1999, Weinberger et al. 1999, Holland et al. 1998). However, few studies have focused on the properties of gas and dust at a few AU from main sequence stars where the temperature of parent bodies may reach $\\sim$300 K (Chen \\& Jura 2001, Heap et al. 2000, Low et al. 1999). At these distances, ice sublimates and large solids may evolve into terrestrial planets. We have carried out a far ultraviolet study of gas around a main sequence star, which possess a possible 10 $\\mu$m excess attributed to dust with a grain temperature ($T_{gr}$ $\\sim$ 300 K), to learn about the physical processes in the regions where terrestrial planets may have formed or may be forming.  $\\sigma$ Herculis is a binary system, at a distance 93 pc away from the Sun (see Table 1), with a B9V primary and a companion at a projected separation of 0.07$\\arcsec$ (Hartkopf et al. 1997). The spectral type of the companion has not been well determined. Astrometric estimates for the masses, using orbital parameters determined from speckle interferometry and relative brightness measurements from \\emph{Hipparcos}, yield M$_{1}$ = 3.0 $\\pm$ 0.7 M$_{\\sun}$, M$_{2}$ = 1.5 $\\pm$ 0.5 M$_{\\sun}$, and M$_{tot}$ = 4.5 $\\pm$ 0.8 M$_{\\sun}$ (Martin et al. 1998). The $V - R$ color and the absolute $V$-band magnitude of the secondary are consistent with a classification of early-type A ($\\Delta m$ = 2.5 $\\pm$ 0.1 mag for 5000 \\AA \\ $<$ $\\lambda$ $<$ 8500 \\AA; Hummel et al. 2002). The age of $\\sigma$ Her has been estimated to be $\\sim$200 Myr based upon its uvby$\\beta$ photometry (Grosb{\\o}l 1978). If we apply a correction for the high rotational velocity of this star to the Str\\\"{o}mgren photometry (Figueras \\& Blasi 1998), we find an estimated age of $\\sim$140 $\\pm$ 100 Myr for $\\sigma$ Her.  The $\\sigma$ Her binary system possess a mid infrared excess indicative of the presence of dust ($L_{IR}/L_{*}$ = 6.6$\\times$10$^{-5}$; Fajardo-Acosta, Telesco, \\& Knacke 1998). The dust appears to be divided into two populations. The colder grains were first discovered based upon measurements of a strong \\emph{IRAS} 60 $\\mu$m excess (Sadakane \\& Nishida 1986; Cot\\'{e} 1987) which is unresolved with \\emph{ISO} (Fajardo-Acosta, Stencel, \\& Backman 1997). Recent ground based photometry of $\\sigma$ Her suggests that this star may also possess 10 $\\mu$m and 20 $\\mu$m excesses (Fajardo-Acosta et al. 1998). The spectral energy distribution of this warmer dust population is marginally fit by blackbody grains with a temperature, $T_{gr} \\sim$ 300 $\\pm$ 100 K, significantly warmer than $T_{gr}$ $\\sim$ 100 K grains observed with \\emph{IRAS} at 60 $\\mu$m. If the particles are blackbodies in radiative equilibrium with the binary, then the dust is located at a distance of between 7 AU and 30 AU, significantly closer than the 120 AU distance inferred for the cooler population.  While Vega-like systems, such as $\\sigma$ Her, possess dust, the gas:dust ratio is not well known. Recent ultraviolet searches for molecular hydrogen, using \\emph{FUSE}, have failed to detect any molecular hydrogen around the dusty main sequence stars $\\beta$ Pic (Lecavelier des Etangs et al. 2001) and 51 Oph (Roberge et al 2002). However, spectra of these systems possess time-variable, redshifted atomic gas features which are believed to be generated by the evaporation of infalling comets (Vidal-Madjar et al. 1994; Roberge et al. 2002). \\emph{IUE} observations of $\\sigma$ Her have revealed the presence of narrow time-variable \\ion{Si}{2}$^{*}$ $\\lambda$ 1533.4 absorption features which are blueshifted with respect to the heliocentric velocity of the star (-11 km/sec; Bruhweiler, Grady, \\& Chiu 1989) with a range of 2 to 21 km/sec (Grady et al. 1991). Although $\\sigma$ Her possess time-variable, velocity-shifted absorption features similar to $\\beta$ Pic, the high stellar luminosity and binary nature of the system may result in a different fate for gas liberated around $\\sigma$ Her compared with material liberated around $\\beta$ Pic. ",
        "conclusions": "We have obtained an ultraviolet spectrum (between 905 \\AA \\ and 1187 \\AA) of the nearby binary system $\\sigma$ Her using \\emph{FUSE}. We argue the following:  1. The ultraviolet spectrum of $\\sigma$ Her possess absorption in the excited fine structure lines of \\ion{C}{2} and \\ion{N}{2}. The excitation of these states and the narrow width of these absorption features suggest that the gas is circumstellar.  2. Since the Poynting-Robertson drag lifetime of dust grains at $D_{in}$ = 20 AU with $a$ = 15 $\\mu$m is $4.6\\times10^{4}$ yr, significantly shorter than the estimated age of $\\sigma$ Her, the grains must be replenished from a reservoir such as collisions between larger objects. For $\\sigma$ Her, we estimate that the minimum mass of parent bodies is 0.20 $M_{\\earth}$.  3. Because $\\sigma$ Her is a binary with a separation of 7 AU, parent body orbits become unstable at distances $\\sim$20 AU from the system. Collisions of parent bodies at this distance could liberate gas which is then blown out of this system by the high luminosity of $\\sigma$ Her.  4. If the gas is released from grains at $\\sim$20 AU from the star, then the predicted outflow velocities of \\ion{C}{2}$^{*}$ and \\ion{N}{2}$^{**}$ are $\\sim$31 km/sec and $\\sim$26 km/sec respectively, in rough agreement with the observed blueshifts of $\\sim$20 km/sec and $\\sim$28 km/sec respectively.  5. We infer a mass loss rate of dM/dt $\\sim$ 6 $\\times$ 10$^{-12}$ M$_{\\sun}$/yr, suggesting that $\\sigma$ Her depletes the mass in parent bodies in less than the estimated age of the system. This raises the liklihood that objects larger than $\\sim$1 m are in orbit in this system."
    },
    "0209/astro-ph0209240_arXiv.txt": {
        "abstract": "{In this work we present the predictions of the ``two-infall model'' concerning the evolution of D, $^{3}$He and $^{4}$He in the solar vicinity, as well as their distribution along the Galactic disk. Our results show that, when adopting detailed yields taking into account the extra-mixing process in low and intermediate mass stars, the problem of the overproduction of $^{3}$He by the chemical evolution models is solved. The predicted distribution of $^{3}$He along the disk is also in agreement with the observations. We also predict the distributions of D/H, D/O and D/N along the disk, in particular D abundances close to the primordial value are predicted in the outer regions of the Galaxy. The predicted D/H, D/O and D/N abundances in the local interstellar medium are in agreement with the mean values observed by the Far Ultraviolet Spectroscopic Explorer mission, although a large spread in the D abundance is present in the data. Finally, by means of our chemical evolution model, we can constrain the primordial value of the deuterium abundance, and we find a value of (D/H)$_{\\rm p} \\lesssim$ 4 $\\cdot$ 10$^{-5}$ which implies $\\Omega_{b}h^2 \\gtrsim $ 0.017, in agreement with the values from the Cosmic Microwave Background radiation analysis. This value in turn implies a primordial $^4$He abundance Y$_{\\rm p} \\gtrsim $ 0.244. ",
        "introduction": "Chemical evolution models are useful tools to derive the primordial abundances of light elements, such as D, $^{3}$He and $^{4}$He, and to give informations on stellar nucleosynthesis. It is well known that while the abundances of metals increase in time with galactic evolution, the abundance of D is always decreasing since this element is only destroyed in stellar nucleosynthesis (for another view see Mullan \\& Linsky 1999, who suggest possible D ejection from flares into the stellar wind). On the other hand, $^{3}$He and $^{4}$He should be partially produced, and partially burned into heavier elements. \\newline In this paper we show the predictions of the ``two-infall model'' (Chiappini et al.  2001), concerning the evolution of D, $^{3}$He and $^{4}$He in the solar vicinity and their distribution along the galactic disk. \\newline This model assumes two main infall episodes for the formation of the halo and part of the thick disk, and the thin disk, respectively. The timescale for the formation of the thin disk is much longer than that of the halo, implying that the infalling gas forming the thin disk comes not only from the halo but mainly from the intergalactic medium. The timescale for the formation of the thin disk is assumed to be a function of the galactocentric distance, leading to an inside-out picture for the Galaxy disk formation. \\newline The two-infall model differs from other models in the literature mainly in two aspects: it considers an almost independent evolution between the halo and thin disk components (see also Pagel \\& Tautvaisiene  1995), and it assumes a threshold in the star formation process (see van der Hulst et al. 1993; Kennicutt 1989, 1998; Martin \\& Kennicutt 2001). The model well reproduces the majority of observational constraints about the abundances of heavy elements both locally and in the whole disk (Chiappini et al. 1997, 2001). \\newline The main novelty of this paper concerns the $^{3}$He nucleosynthesis prescriptions. Starting from the paper of Iben (1967), stellar models have predicted that low-mass stars are strong producers of $^{3}$He. However, when stellar production of $^{3}$He is included in models for the chemical evolution of the Galaxy, no agreement between observed and predicted $^{3}$He abundances can be found. In particular, the models predict an overproduction of $^{3}$He (see Rood et al. 1976; Dearborn et al. 1996; Prantzos 1996; Chiappini et al. 1997). \\newline On the other hand, there is evidence for $^{3}$He production in stars from the observations of this isotope in Galactic Planetary nebulae, such as NGC 3242, (Balser et al. 1997; Balser et al. 1999), and also from the observations of the $^{12}$C/$^{13}$C ratio, which is related to the possible occurrence of ``extra-mixing'', with consequent $^{3}$He destruction, in low and intermediate mass stars, the main producers of $^{3}$He in the universe. The extra-mixing process has been suggested to solve the $^{3}$He overproduction problem. Recent stellar models suggested that thermal instabilities might occur during the red giant branch (RGB) phase in low mass stars and during the early asymptotic giant branch (AGB) phase in intermediate mass stars, leading to a mixing between $^{3}$He-enriched material from the external envelope and the material which suffer H-shell burning. These processes should induce $^{3}$He destruction in favor of $^{7}$Li production (Sackmann \\& Boothroyd 1999a; Palacios et al. 2001). Extra-mixing should also decrease the $^{12}$C/$^{13}$C ratio (see Sect. 3 for a detailed description). This justifies the low observed $^{12}$C/$^{13}$C ratio in RGB stars and in Li-rich giants. \\newline However, not every star in the low and intermediate mass range should suffer from extra-mixing. For example, recent Hubble Space Telescope observations of NGC 3242, (see Palla et al. 2002), suggests a lower limit $^{12}$C/$^{13}$C $>$ 38, in agreement with standard stellar models. Therefore, the lack of the $^{13}$C line and the presence of the $^{3}$He line in the spectrum of this planetary nebula suggest that the progenitor star did not undergo a phase of deep extra-mixing during the latest stages of its evolution. Charbonnel \\& do Nascimento (1998) collected the observations of the carbon isotopic ratio in field and cluster giant stars, and their conclusion was that 93$\\%$ of evolved stars undergo the extra-mixing on the Red Giant Branch, and are expected to destroy, at least partially, their $^{3}$He. Extra-mixing along the RGB is supported by recent observations of metal-poor field stars (Gratton et al. 2000) and of stars along the RGB phase in the thick disk of the Galaxy (Tautvaisiene et al. 2001). \\newline In the present work we take into account the detailed prescriptions about $^{3}$He production/destruction by Sackmann \\& Boothroyd (1999a, b) and their dependence upon stellar mass and metallicity. The $^{3}$He production is strictly correlated with the destruction of D and the production of $^{4}$He. Therefore, we consider all these elements. We compare the predictions for the time-evolution of the abundances of D, $^{3}$He and $^{4}$He and for their distribution along the disk, with the available data. The latest data on $^{3}$He/H, which are observations of the hyperfine transition of $^{3}$He$^{+}$ at 8.665 GHz (3.46cm), are from Bania et al. (2002), and show a flat distribution around a value of about 1.5$\\cdot~10^{-5}$ (with the exception of the observation of a larger $^{3}$He abundance in a few nebulae such as NGC 3242). The latest published data for D/H, D/O, D/N, are from the Far Ultraviolet Spectroscopic Explorer mission (Moos et al. 2002) and refer to the local interstellar medium (LISM). These observations (Jenkins et al. 1999; Sonneborn et al. 2000; Friedman et al. 2002) show a large dispersion in the deuterium abundances beyond a few tenth of kpc in the solar vicinity which is not easy to justify since these effects are not seen in the oxygen and nitrogen abundances. \\newline The paper is organized as follows: in Sect. 2 we describe the chemical evolution model, in Sect. 3 we discuss in detail the adopted nucleosynthesis prescriptions for D, $^{3}$He and $^{4}$He. In Sect. 4 the results are presented and discussed and finally, in Sect. 5 some conclusions are drawn. ",
        "conclusions": "In this paper we have calculated the galactic evolution of the abundances of D, $^{3}$He, and $^{4}$He. In particular, we modelled the evolution of the $^{3}$He abundance taking into account in detail the extra-mixing process in low and intermediate mass stars, and its dependence on stellar mass and metallicity. We have predicted the time-evolution of the abundances of D, $^{3}$He, and $^{4}$He, and their distribution along the Galactic disk, and studied how these abundances depend upon the ``primordial'' abundances as well as upon the percentage of extra-mixing adopted in the models. We have compared these predictions with the most recent observations, including the first published data from the FUSE mission and selected the best combination of parameters.  The adopted chemical evolution model, the ``two-infall'' one, differs from other models present in the literature because it assumes that the galactic thin-disk formed out of an infall episode completely disentangled from the formation of the galactic halo, with almost no contribution from the pre-enriched halo gas especially in the inner regions. As a consequence of this, our predicted deuterium depletion factors are lower than in previous models (e.g. Prantzos 1996), especially in the region close to the bulge. The reason for this resides in the fact that after the first halo phase, where some D depletion already occurs, the formation of the thin-disk implies mostly primordial gas and therefore the D abundance in the gas increases again and then declines (see Tosi et al. 1998). For what concerns the predicted abundances of $^{3}$He our model differs from previous ones because it includes new nucleosynthesis prescription for this element. \\newline The main conclusions are the following: \\begin{itemize} \\item[-] The redefinition of the production/destruction of $^{3}$He in low and intermediate mass stars has important consequences on the evolution of the $^{3}$He abundance and on its distribution along the Galactic disk. In particular, the new models with 93\\% of extra-mixing, show a good agreement with the PSC observations as well as with the observed gradient along the disk. We predict a $^{3}$He gradient of $-0.04<d\\log$($^{3}$He/H)/$d$R$<-0.03$ dex/kpc in the galactocentric distance range 4-18 kpc.  \\item[-] The predicted time evolution of D in the solar vicinity is well reproduced although the LISM values measured by FUSE show a large spread. Since D is only destroyed during galactic evolution, we predict a positive gradient of this element along the disk, in particular $d\\log$(D/H)$/d$R$\\simeq$ 0.02 dex/kpc, $d\\log$(D/O)$/d$R$\\simeq$ 0.13 dex/kpc, $d\\log$(D/N)$/d$R$\\simeq$ 0.14 dex/kpc, in the galactocentric distance range 4-18 kpc. Our predicted D/H gradient is less steep than the one obtained by Prantzos (1996) by means of a model with a much lower infall rate than the ``two-infall'' model. This is a consequence of the fact that D abundance is very sensitive to the infall/star formation history in the Galaxy. It should be noted that the infall/star formation history assumed here is requested to fit the majority of the observational constraints (Chiappini et al. 1997, 2001). We predict that observations towards the Galactic anticenter should provide a D abundance close to the primordial value. We predict a depletion factor for D in the solar vicinity of $\\simeq$ 1.5, whereas at 2 kpc from the Galactic center we expect a factor of $\\simeq$ 5. However, the model of the Galactic disk is probably not appropriate to describe the Galactic bulge which should have suffered a much higher star formation than we can deduce from the present models (see Matteucci et al. 1999).  \\item[-] Chemical evolution models can constrain the primordial value of the deuterium abundance and a value of (D/H)$_{\\rm p} \\lesssim$ 4 $\\cdot$ 10$^{-5}$ is suggested by the present model. This upper limit implies, according to Standard Big Bang Nucleosynthesis (Burles et al. 1999), $\\Omega_{b}h^2 \\gtrsim 0.017$, in agreement with estimates from the Cosmic Microwave Background radiation analysis, which implies $\\Omega_{b}h^{2}_{\\rm CMB} = 0.032^{+0.009}_{-0.008}$, at 95\\% Confidence Level (Jaffe et al. 2001). This limit seems to rule out high D abundance detection in DLA absorbers, and implies $Y_{\\rm p} \\geq 0.244$ from SBBN. This is at variance with the latest estimate of primordial $^{4}$He from Gruenwald et al. (2002), who suggest that the primordial $^{4}$He should be Y$_{\\rm p}$ = 0.238$\\pm$0.003 (see also  Sauer \\& Jedamzik 2002). Finally, we predict an enrichment of $^{4}$He relative to metallicity of $\\Delta Y/ \\Delta Z=1.5$ in the range $0.008\\leqslant Z\\leqslant 0.018$, and a gradient along the galactic disk $d\\log$($^{4}$He/H)$/d$R $\\approx -0.007$ dex/kpc, or $d$($^{4}$He/H)$/d$R $\\approx -0.002$ kpc$^{-1}$, in the galactocentric distance range 4-18 kpc. \\end{itemize}"
    },
    "0209/astro-ph0209289_arXiv.txt": {
        "abstract": " ",
        "introduction": "The chronology of star formation is recorded in the white dwarf luminosity function (WDLF). White dwarf (WD) structure implies a relatively simple connection between WD luminosity and age. First attempts to exploit WDs as chronometers [3,4,6] showed that the WDLF was a map of the history of star formation in the disk, and a significant shortfall of low-luminosity degenerates--the inevitable consequence of the finite age of the disk. The shortfall near log$(L/L_{\\odot})$$\\approx$$-$4.5 implies a disk age of 6.5--9.5 Gyr [2]. The WDLF from wide common proper motion binaries [5] does {\\it not} show the shortfall seen by Liebert {\\it et al.} (1988). This suggests that the disk is at least $\\sim$10.5 Gyr old. Both Liebert {\\it et al.} (1988) and Oswalt {\\it et al.} (1996) WDLFs were derived from proper motion catalogs, hence may be affected by significant kinematical bias as well as incompleteness. The simple fact is, the faintest, age-dependent end of the WDLF is not yet reliably determined. ",
        "conclusions": ""
    },
    "0209/astro-ph0209130_arXiv.txt": {
        "abstract": "Models of SN driven galactic winds for ellipticals are presented. We assume that ellipticals formed at high redshift and suffered an intense burst of star formation. The role of supernovae of type II and Ia in the chemical enrichment and in triggering galactic winds is studied. In particular, several recipes for SN feed-back together with detailed nucleosynthesis prescriptions are considered. It is shown that SNe of type II have a dominant role in enriching the interstellar medium of elliptical galaxies whereas type Ia SNe dominate the enrichment and the energetics of the intracluster medium. ",
        "introduction": "Several mechanisms have been suggested so far for the formation and evolution of elliptical galaxies. One scenario is based on an early monolithic collapse of a gas cloud or early merging of lumps of gas where dissipation plays a fundamental role (Larson 1974; Arimoto \\& Yoshii 1987; Matteucci \\& Tornamb\\`e 1987). In this scenario the star formation stops soon after a galactic wind develops and the galaxy evolves passively since then. Bursts of star formation in merging subsystems made of gas had been also suggested (Tinsley \\& Larson 1979); in this picture star formation stops after the last burst and gas is lost via stripping or wind. The alternative and more recent scenario is the so-called hierarchical clustering scenario, where merging of early formed stellar systems in a wide redshift range and preferentially at late epochs, is expected (Kauffmann et al. 1993). The main difference between the monolithic collapse and the hierarchical merging relies in the time of galaxy formation, occurring quite early in the former scenario (at reshifts $z> 3$) and continuously in the latter scenario. There are arguments either in favour of the monolithic or the hierarchical scenario, but the former one gives a more likely picture since it can reproduce the majority of the properties of stellar populations in ellipticals, in particular some fundamental facts such as that the ellipticals are dominated by old stars (K-giants) and that the [$\\alpha$/Fe]$>0$ in the dominant stellar population (Worthey et al. 1992; Weiss et al. 1995; Kuntschner et al. 2001). This high [$\\alpha$/Fe] ratio is the clear signature of the pollution from massive stars. The same occurs in the most metal poor stars in our Galaxy and is due to the fact that SNe II are the main producers of $\\alpha$-elements (O, Ne, Mg, Si. S and Ca) whereas SNe Ia, which explode with a delay relative to type II SNe, are thought to be responsible for the production of Fe. Therefore, the high [$\\alpha$/Fe] ratio in ellipticals argues strongly in favor of a short period of star formation during which type Ia SNe did not have time to substantially pollute the insterstellar medium (ISM).  In this paper we will discuss a monolithic model (Pipino et al. 2002) for the formation and evolution of ellipticals, where these objects suffer a short ($\\le$ 1Gyr) but intense star formation period halted by a SN driven galactic wind. After the onset of the wind, which devoids the galaxy of all the gas present, star formation is assumed to stop. This is because the galaxy, after the wind, contains hot and rarified gas, a situation which is unfavorable to star formation. The time for the occurrence of the wind, $t_{GW}$, is therefore crucial in determining the evolution of the galaxy and the intracluster/intergalactic medium. Therefore, the assumptions about the energy transferred from SNe into the ISM are very important. Unfortunately, very little is known about this feed-back and one has to choose the assumptions which produce a realistic model for ellipticals. In section 2 we will discuss the condition for the occurrence of a galactic wind and the chemical evolution model. In section 3 we will present the results and draw a few conclusions. ",
        "conclusions": ""
    },
    "0209/astro-ph0209306_arXiv.txt": {
        "abstract": "Photometric redshifts have proven a powerful tool in identifying galaxies over a large range of lookback times. We have been generalising this technique to incorporate the selection of candidate high redshift QSOs. We have applied this to a large optical/near-infrared imaging survey in 6 wavebands aiming to push farther in redshift (and fainter in luminosity) than previous studies. We believe that study of these very faint and distant objects provides valuable insights into galaxy formation and evolution.\\\\ Here we present work in progress and preliminary results for a catalogue of objects detected as part of the Las Campanas Infrared Survey. This is a stepping stone to the type of survey data that will become available in the next few years from projects such as UKIDSS and VISTA.\\\\ ",
        "introduction": "The technique of photometric redshifts has been extensively studied and developed in the last decade.  By comparing the flux observed from an object in a number of wavebands to that expected from templates of known  type, age and redshift, all these parameters can be found.  A publicly available application designed to perform $\\chi^2$ minimisation over such a parameter space is Hyperz \\cite{b2000}. This code has now been used and tested by several authors.\\\\ \\\\ Clearly, since objects are classified according to the best fitting template, the choice of which templates to provide has significant effects on the results obtained.  Hyperz is supplied together with a set of empirical galaxy templates \\cite{cww} and a set of evolutionary synthesis galaxy templates \\cite{bc}.\\\\ \\\\ We have been investigating the use of other template sets in Hyperz and in particular the inclusion of QSO and stellar templates.\\\\ \\\\ A model quasar spectral energy distribution (SED) has been developed from the Sloan Digital Sky Survey results and templates at a number of redshifts produced.  These model templates are then used as additional inputs for Hyperz. \\\\ \\\\ The Star/Galaxy parameter in the SExtractor software has been used to divide catalogues into `extended' and `pointlike' objects.  A local QSO template may then be included with the galaxy SED set when studying extended objects.\\\\ \\\\ Pointlike objects are considered to be quasars or stars. For an initial analysis of these objects a new set of  templates are constrained to lie at z$=$0. This template set comprises a number of stellar templates from the Bruzual 77 library as well as an array of QSO templates at different redshifts.  Objects are then classified as stellar or QSO according to the best fitting template. \\\\\\\\ The method has been tested on real and simulated data.  The reliability of the separation depends on the filter set being used but on simulated catalogues in {\\it UBVRIH} we have found that fewer than 5\\% of objects in the catalogue are misclassified in this way and fewer than 1\\% of high redshift objects are assigned stellar templates or {\\it vice versa}.\\\\ ",
        "conclusions": "This work is intended to be the basis for further study  both of the photometric redshift  technique and its applications to galaxy evolution.   We are now developing this field by  refining and applying semi-analytic galaxy models to better describe the universe at large lookback times.\\\\ \\\\ It is our belief that  galaxy redshift distributions obtained in this fashion from large optical-infrared surveys may provide valuable constraints on simulations and models of galaxy formation at large redshifts.\\\\"
    },
    "0209/astro-ph0209394_arXiv.txt": {
        "abstract": "Was the black hole in $\\xte$ ejected from a globular cluster or kicked away from the galactic disk? \\vskip -1.0truecm ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209441_arXiv.txt": {
        "abstract": "We present an exploratory \\chandra\\ ACIS-S3 study of the diffuse component of the Cosmic X-ray Background in the 0.3--7 keV band for four directions at high Galactic latitudes, with emphasis on details of the ACIS instrumental background modeling.  Observations of the dark Moon are used to model the detector background. A comparison of the Moon data and the data obtained with ACIS stowed outside the focal area showed that the dark Moon does not emit significantly in our band. Point sources down to $3\\times 10^{-16}$ \\ergscm\\ in the 0.5--2 keV band are excluded in our two deepest observations. We estimate the contribution of fainter, undetected sources to be less than 20\\% of the remaining CXB flux in this band in all four pointings. In the 0.3--1 keV band, the diffuse signal varies strongly from field to field and contributes between 55\\% and 90\\% of the total CXB signal. It is dominated by emission lines that can be modeled by a $kT=0.1-0.4$ keV plasma. In particular, the two fields located away from bright Galactic features show a prominent line blend at $E\\approx 580$ eV (O{\\small VII} + O{\\small VIII}) and a possible line feature at $E\\sim 300$ eV. The two pointings toward the North Polar Spur exhibit a brighter O blend and additional bright lines at 730--830 eV (Fe{\\small XVII}). We measure the total 1--2 keV flux of $(1.0-1.2\\,\\pm0.2)\\times 10^{-15}$ \\ergscmam\\ (mostly resolved), and the 2--7 keV flux of $(4.0-4.5\\,\\pm1.5)\\times 10^{-15}$ \\ergscmam.  At $E>2$ keV, the diffuse emission is consistent with zero, to an accuracy limited by the short Moon exposure and systematic uncertainties of the S3 background.  Assuming Galactic or local origin of the line emission, we put an upper limit of $\\sim 3 \\times 10^{-15}$ \\ergscmam\\ on the 0.3--1 keV extragalactic diffuse flux. ",
        "introduction": "The existence of a cosmic X-ray background (CXB) was one of the first discoveries of extra-solar X-ray astronomy (Giacconi et al.\\ 1962).  In the intervening four decades, observations with improving angular and spectral resolution have enhanced our understanding of the components that make up this background.  Several broad-band, all-sky surveys have been performed using proportional-counter detectors (Marshall et al.\\ 1980; McCammon et al.\\ 1983; Marshall \\& Clark 1984; Garmire et al.\\ 1992; Snowden et al.\\ 1995, 1997; for a review of pre-\\rosat\\ results, see McCammon \\& Sanders 1990). These surveys form a consistent picture of the angular distribution of X-ray emission in the various bands.  Above 2 keV, the emission is highly isotropic on large angular scales and has an extragalactic origin.  Below 2 keV, the X-ray background is a mixture of Galactic diffuse emission (e.g., Kuntz \\& Snowden 2000 and references therein), heliospheric and geocoronal diffuse component (e.g., Cravens 2000), and extragalactic flux from point sources and, possibly, from intergalactic warm gas that may contain the bulk of the present-day baryons (e.g., Cen \\& Ostriker 1999).  The earliest observations could provide only limited spectral information on the background.  Marshall et al.\\ (1980) found that the spectrum in the 3--50 keV range was well fit by a thermal bremsstrahlung model with $kT \\sim 40$ keV.  In the 3--10 keV band this can be approximated by a power law with a photon index of $-1.4$. At energies below 1 kev the background surface brightness exceeds the extrapolation of this power law (Bunner et al.\\ 1969).  Later observations with gas-scintillation proportional counters and solid-state detectors (Inoue et al.\\ 1979; Schnopper et al.\\ 1982; Rocchia et al.\\ 1984) suggested emission lines in the 0.5--1.0 keV band, most likely from oxygen.  The evidence for emission lines in this band has become more convincing in recent observations using CCDs (Gendreau et al.\\ 1995; Mendenhall \\& Burrows 2001) and in a high-resolution spectrum obtained by McCammon et al.\\ (2002) in a microcalorimetric experiment.  A definitive demonstration of spectral lines in the 0.15--0.3 keV band (at low Galactic latitude) was obtained by Sanders et al.\\ (2001) using a Bragg-crystal spectrometer.  Observations that combine high spectral and angular resolution are essential for disentangling the many CXB soft emission components.  \\begin{table*}[t] \\begin{center} \\begin{minipage}{16.5cm} \\renewcommand{\\arraystretch}{1.4} \\renewcommand{\\tabcolsep}{2.2mm} \\small \\begin{center} \\caption{Data Summary} \\begin{tabular}{p{5.8cm}ccccc} \\hline \\hline Observation Identificator (OBSID)  & 3013 & 3419 & 869    & 930  & Moon \\\\ \\hline $(l,b)$, deg\\dotfill& $(259.6, +56.9)$ & $(187.1, -31.0)$ & $(36.6, +53.0)$ & $(358.7, +64.8)$&...\\\\ Galactic $N_H$, $10^{20}$\\cmsq \\dotfill    & 4.1 & 11.4 & 4.3  & 1.8 &... \\\\ \\rosat\\ R4--R5 flux, $10^{-6}$\\ctssam\\dotfill&160 & 90 &200--250$^a$ &400&...\\\\ Observation date \\dotfill & 2001 Dec 13 & 2002 Jan 08 & 2000 Jun 24 &% 2000 Apr 19 & 2001 Jul 26\\\\ Total (uncleaned) exposure, ks      \\dotfill & 112 & 98   & 57  & 40 & 16  \\\\ Exposure for source detection, ks   \\dotfill & 101 & 92   & 52  & 28 & ... \\\\ Exposure for background spectra, ks \\dotfill & 69  & 86   & 52  & 20 & 11  \\\\ Field solid angle, arcmin$^2$       \\dotfill & 69  & 69   & 51  & 69 & 70  \\\\ \\hline \\label{table:obslist} \\end{tabular} \\end{center} \\vspace{-6mm} {\\footnotesize $^a$ Affected by an artifact in the \\rosat\\ all-sky map } \\vspace{-2mm} \\end{minipage} \\end{center} \\end{table*}  \\chandra\\ and \\xmm\\ should soon provide a wealth of new information on the CXB. Several works have already taken advantage of the \\chandra's arcsecond resolution to study the point source component of the CXB (e.g., Mushotzky et al.\\ 2000; Brandt et al.\\ 2001; Rosati et al.\\ 2002). The first \\xmm\\ results are starting to appear as well (De Luca \\& Molendi 2002; Warwick 2002; Lumb et al.\\ 2002), utilizing the large effective area of that observatory.  In this paper, we present a \\chandra\\ ACIS study of the diffuse CXB at high Galactic latitudes. The main advantage of \\chandra\\ over all other instruments is its ability to resolve point sources down to very low fluxes and probe the true diffuse background. In addition to that, compared to \\rosat\\ PSPC (which had lower detector background), ACIS has energy resolution sufficient to identify spectral lines. Compared to \\xmm\\ EPIC, ACIS appears to be less affected by instrumental background flares (although during quiescent periods, the ACIS detector background per unit sky signal is higher). To the extent that results can be compared, we confirm many of the recent findings made with other instruments.  Technical aspects of our study, especially the ACIS instrumental background modeling, are quite complex, and we discuss them here in detail. Much of our analysis procedure may be useful for studies of extended sources such as clusters of galaxies. Uncertainties are $1\\sigma$ unless specified otherwise. ",
        "conclusions": "\\subsection{Comparison with earlier results}  We can compare our total (diffuse + sources) CXB results with the recent \\xmm\\ work (Lumb et al.\\ 2002). Since the brightest sources in our four fields have fluxes similar to the lowest fluxes of the excluded sources in Lumb et al.\\ ($1-2\\times 10^{-14}$ \\ergscm\\ in the 0.5--2 keV band), our ``total'' fluxes can be directly compared to their values after the bright source exclusion. Starting from high energies, normalizations of our power-law fits in the 1--7 keV band given in Table 2 are in agreement with the \\xmm\\ average of 8.4 \\phcmskevsr\\ at 1 keV, if we exclude our bright field 930.  In the narrower 1--2 keV band, our fluxes (again, except 930) are in good agreement with Lumb et al.'s $1.0\\times 10^{-15}$ \\ergscmam\\ calculated from their best-fit model. Our 2--7 keV fluxes also agree with $3.2\\times 10^{-15}$ \\ergscmam\\ converted from their 2--10 keV flux (we note here that these three values in Table 2 are not entirely statistically independent, because the error is dominated by the same Moon dataset). Our uncertainties in the 2--7 keV band are large; future ways to reduce them are described in \\S\\ref{sec:future} below.  Our 1--7 keV power law normalizations also agree with $9-11$ \\phcmskevsr\\ at 1 keV derived by Miyaji et al.\\ (1998) from \\asca\\ GIS data for two high Galactic latitude fields, and with 11.7 \\phcmskevsr\\ derived by Vecchi et al.\\ (1999) using \\sax.  Note that those studies cover much greater solid angles and therefore are more likely to include rare, bright sources, so this comparison is only approximate.  At $E<1$ keV, the scatter between our fields (a factor of 5) is higher than that in Lumb et al.\\ (a factor of 2--3), but this, of course, is because of our specific selection of fields spanning a range of \\rosat\\ fluxes.  At the primary O line energy, the CXB is dominated by the diffuse component (Fig.~\\ref{fig:totdiff}). Thus, we can compare our line fluxes to those recently derived in a microcalorimetric experiment by McCammon et al.\\ (2002) from a 1 sr area mostly away from bright Galactic features. They report an average flux in the O{\\small VII} + O{\\small VIII} lines of $(5.4\\pm 0.8) \\times 10^{-7}$ \\phscmam. This is in the range of our values for the off-Spur observations 3013 and 3419 given in Table 3.  McCammon et al.\\ (2002) also observed lines at lower energies, some of which may explain our 300 eV feature (if it is real).  Oxygen line fluxes derived from earlier experiments (e.g., Inoue et al.\\ 1979; Gendreau et al.\\ 1995) are also within our range. Our results show, however, that even for these high Galactic latitude areas away from bright Galactic features, the line brightness varies significantly from field to field. The general shape of the North Polar Spur spectrum in our field 930 is consistent with that reported in earlier works (e.g., Schnopper et al.\\ 1982; Rocchia et al.\\ 1984; Warwick 2002).    \\subsection{Extragalactic diffuse component}  The diffuse flux in fields 869 and 930 is obviously dominated by the North Polar Spur. The O blend in 3013 and 3419 probably has a Galactic or local origin as well (extragalactic sources with redshifts $>0.05$ are excluded by our measured line energy; furthermore, the high-resolution spectrum of McCammon et al.\\ 2002 excludes {\\em any}\\/ redshift), although we cannot exclude, for example, their Local Group origin. Thus, the continuum component in the two low-brightness fields can give an approximate upper limit on the flux from the vast quantities of the putative warm intergalactic gas (e.g., Cen \\& Ostriker 1999) that should emit a mixture of lines and continua from different redshifts. For a conservatively high estimate of this continuum component, we fit the 0.3--1 keV diffuse spectra by a power law model plus the line, applying the full Galactic absorbing column.  The resulting unabsorbed continuum fluxes are given in Table 3; they correspond to the spectral density of $(2.4-2.5)\\times 10^{-15}$ \\ergscmam\\,keV$^{-1}$ at $E=0.7$ keV. This is well above the typical theoretical predictions that range between $(0.3-1)\\times 10^{-15}$ \\ergscmam\\,keV$^{-1}$ (e.g., Cen \\& Ostriker 1999; Phillips, Ostriker, \\& Cen 2001; Voit \\& Bryan 2001; but see Bryan \\& Voit 2001 for a higher predicted flux from simulations without the inclusion of cooling and preheating). However, the predicted average brightness values from the published simulations are not directly comparable to our result, because they are dominated by nearby galaxy groups and clusters that are easily detected and excluded from our and other CXB measurements.  Thus, our crude estimate does not constrain the warm intergalactic gas models. The constraint may be improved in the future by better modeling and subtraction of the Galactic emission and spatial fluctuation analysis (as in, e.g., Kuntz, Snowden \\& Mushotzky 2001); however, since the Galaxy dominates at these energies, such constraints will necessarily be model-dependent.   \\subsection{Origin of line emission}  The observed energies of the spectral lines suggest local (in the Local Group, Galaxy, or our immediate vicinity) origin of the dominant fraction of the soft diffuse CXB. Extensive literature exists that models its various components under the well-justified assumption of their thermal plasma origin (e.g., Kuntz \\& Snowden 2000 and references therein). Leaving such modeling for future work, here we mention an interesting alternative possibility.  It is likely that a significant fraction of the line flux comes from charge exchange (CX) between highly charged ions in the solar wind (primarily bare and hydrogenic O and C) and neutral gas occurring throughout the heliosphere and in the geocorona. Most of the flux that would be observed from heliospheric CX (with H and He) originates within a few tens of AU of the Sun. Geocoronal emission arises where residual atmospheric H is exposed to the solar wind, at distances of order 10 Earth radii.  In the CX process, a collision between a solar wind ion and a neutral atom leads to the transfer of an electron from the neutral species to a high-$n$ energy level in the ion, which then decays and emits an X ray. Dennerl et al.\\ (1997) and Cox (1998) were the first to suggest that these photons might contribute to the CXB, and Cravens (2000) estimated that heliospheric emission might account for roughly half of the observed soft CXB. Cravens, Robertson, \\& Snowden (2001) also argued that the excess time-variable diffuse flux often observed by \\rosat\\ was due to fluctuations in heliospheric and especially geocoronal CX emission, caused by ``gusts'' in the solar wind.  In the \\chandra\\ Moon observations, the heliospheric component will be blocked, but geocoronal emission should be present. We estimate, however, that the typical intensity of that signal, about a few $\\times 10^{-8}$ phot s$^{-1}$ cm$^{-2}$ arcmin$^{-2}$ in O K$\\alpha$ (the strongest line), is smaller than the statistical uncertainties in our measurement.  Note also that the geocoronal signature is not likely to be present in our net CXB spectra, since it is subtracted as part of the Moon spectrum.  Heliospheric CX emission should be stronger and less time-variable than geocoronal emission, and should be present in our spectra. We have constructed a numerical model, based on Cravens' (2000) work and similar to that described in Wargelin \\& Drake (2001, 2002), that predicts that roughly half the flux in the O line(s) in fields 3013 and 3419 may come from heliospheric CX. Those observations were at low ecliptic latitude, within the ``slow'' and more highly ionized solar wind. CX flux in fields 869 and 930 is expected to be much lower because those observations looked through the ``fast'' solar wind, which has a much smaller fraction of bare and H-like O ions (von Steiger et al.\\ 2000). Observations of a sample of fields selected specifically to test this possibility are required for a more quantitative analysis, which is forthcoming (B. J. Wargelin et al.\\ in preparation).   \\subsection{Future work} \\label{sec:future}  For further CXB studies with \\chandra, it is useful to look into the error budget of the present results. The errors in the 1--7 keV band are dominated by the detector background uncertainty: the statistical error of the short Moon dataset, the uncertainty on its normalization, and the possible residual flare component.  Forthcoming calibration observations with ACIS stowed but working in the full imaging mode should take care of the first component. The other two point toward the use of FI chips for studies in this band. The scatter of the quiescent background normalization probably cannot be reduced below our 2--3\\% estimate; however, the FI detector background itself is lower by a factor of 2--3, depending on the energy. The FI chips also are much less affected by the background flares. An ACIS-I study of the CXB will be presented in a forthcoming paper (S. Virani et al., in preparation)."
    },
    "0209/astro-ph0209488_arXiv.txt": {
        "abstract": "We present a short review of the current quasar (QSO) absorption line constraints on possible variation of the fine structure constant, $\\alpha \\equiv e^2/\\hbar c$. Particular attention is paid to recent optical Keck/HIRES spectra of 49 absorption systems which indicate a smaller $\\alpha$ in the past \\cite{MurphyM_01a,WebbJ_01a}. Here we present new preliminary results from 128 absorption systems: $\\da = (-0.57 \\pm 0.10)\\times 10^{-5}$ over the redshift range $0.2 < z < 3.7$, in agreement with the previous results. Known potential systematic errors cannot explain these results. We compare them with strong `local' constraints and discuss other (radio and millimetre-wave) QSO absorption line constraints on variations in $\\alpha^2g_p$ and $\\alpha^2g_pm_e/m_p$ ($g_p$ is the proton $g$-factor and $m_e/m_p$ is the electron/proton mass ratio). Finally, we discuss future efforts to rule out or confirm the current 5.7\\,$\\sigma$ optical detection. ",
        "introduction": "\\label{sec:intro}  The assumption that the constants of Nature remain constant in spacetime should be experimentally tested \\cite{BekensteinJ_79a}. Strong motivation for {\\it varying} constants comes from modern unified theories \\cite{MarcianoW_84a,BarrowJ_87a,DamourT_94a}. Here we review the QSO absorption line constraints on possible variation of the electromagnetic coupling constant, $\\alpha$. Our most recent published results are summarized in reference \\cite{WebbJ_01a} and in Section \\ref{sec:results} we present new preliminary results from a significantly extended optical sample. ",
        "conclusions": ""
    },
    "0209/astro-ph0209507_arXiv.txt": {
        "abstract": "We present the $UBVR_cI_c$ broad band optical photometry of the Type Ic supernova SN 2002ap obtained during 2002 February 06 -- March 23 in the early decline phases and also later on 2002 15 August. Combining these data with the published ones, the general light curve development is studied. The time and luminosity of the peak brightness and the peak width are estimated. There is a flattening in the optical light curve about 30 days after the $B$ maximum. The flux decline rates before flattening are 0.127$\\pm$0.005, 0.082$\\pm$0.001, 0.074$\\pm$0.001, 0.062$\\pm$0.001 and 0.040$\\pm$0.001 mag day$^{-1}$ in $U$, $B$, $V$, $R_c$ and $I_c$ passbands respectively, while the corresponding values after flattening are about 0.02 mag day$^{-1}$ in all the passbands. The maximum brightness of SN 2002ap $M_V = - 17.2$ mag, is comparable to that of the type Ic 1997ef, but fainter than that of the type Ic hypernova SN 1998bw. The peak luminosity indicates an ejection of $\\sim$ 0.06 M$_{\\odot}$ ${}^{56}$Ni mass.  We also present low-resolution optical spectra obtained during the early phases. The SiII absorption minimum indicates that the photospheric velocity decreased from $\\sim$ 21,360 km s$^{-1}$ to $\\sim$ 10,740 km s$^{-1}$ during a period of $\\sim$ 6 days. ",
        "introduction": "The supernova SN 2002ap ($\\alpha_{2000}$ = $01^{h}36^{m}23^{s}.92 $; $\\delta_{2000}$ = $+15^{\\circ}$ $45^{\\prime}$ $13^{\\prime\\prime}.3$) was discovered on 2002 January 29.4 UT by Y. Hirose at $V \\sim 14.5$ mag, in the outer region of the nearby spiral M74 (Nakano et al.\\ 2002). Low resolution spectra of SN 2002ap obtained during 2002 January 30--31 (Kinugasa et al. 2002a,b; Meikle et al. 2002; Filippenko \\& Chornock 2002) were similar to those of type Ic supernovae. However, the spectral features were found to be extremely broad resembling the type Ic `hypernovae' SN 1997ef and SN 1998bw (Filippenko 1997; Nomoto et al. 2002; Mazzali et al. 2002).  At a distance of 7.3~Mpc, SN 2002ap being the nearest hypernova discovered to date, was a good target for a detailed monitoring, and has been subject to multi-wavelength observations. In addition to optical observations, it has been observed in the X-ray (Sutaria et al.\\ 2002), radio (Berger, Kulkarni \\& Chevalier 2002, Sutaria et al.\\ 2002) and in the infrared (Mattila \\& Meikle 2002). Evidence for a high velocity asymmetric explosion has been indicated by spectropolarimetric observations (Kawabata et al.\\ 2002; Leonard et al.\\ 2002; Wang et al.\\ 2002).  The broad spectral features in the spectrum and the evolution of Si II spectral line indicate a high expansion velocity ($\\sim$ 30000 km s$^{-1}$), which supports hypernova model for the SN 2002ap (Kinugasa et al. 2002b; Meikle et al. 2002; Gal-Yam et al. 2002a,b; Filippenko \\& Chornock 2002). The spectropolarimetric observations during the early phase indicate a similarity with SN 1998bw (Leonard et al.\\ 2002). Modeling of the optical observations indicate SN 2002ap as an energetic event, with an explosion energy of $\\simeq 4-10\\times 10^{51}$~ergs (Mazzali et al. 2002). However, the radio observations seem to indicate that SN 2002ap an ordinary type Ic supernova, without a jet (Berger et al. 2002). A study of the $UBVRIH_{\\alpha}K$ images of galaxy M74 obtained several years prior to the discovery of SN 2002ap (Smartt et al. 2002) resulted in a non-detection of the progenitor.  The spectral similarity to SN 1998bw, the possible link between very energetic supernova and gamma ray bursts and the lack of substantive data on rare type Ic SN events make SN 2002ap very important object to study in detail. We have therefore carried out dense temporal multi-colour optical photometric observations during the early phases. These in combination with the published data are used to study the development of the optical light curve. Our observations are the first to indicate a flattening in the light curve of the SN 2002ap about 30 days after the B maximum. We also present low resolution optical spectra obtained during the early phases. The details of both photometric and spectroscopic observations are presented in the next section, while the development of the light curves, spectral and other properties of the SN 2002ap are discussed in the remaining part of the paper. ",
        "conclusions": "We present dense temporal optical photometric data of SN 2002ap during 2002 February 06 to March 23. $UBVR_cI_c$ photometric light curves of SN 2002ap have been studied by combining our data with those published by others. The broad band photometric observations taken up to about 40 days after the peak brightness in $B$ indicate that the flux undergoes an exponential decline similar to that observed in other SN Ic. We are the first to report a flattening in the optical light curves at JD 2452340, about 30 days after the $B$ maximum.  The supernova luminosity follows an exponential decline with a relatively faster rate up to JD 2452340. The flux decline rates are 0.127$\\pm$0.005, 0.082$\\pm$0.001, 0.074$\\pm$0.001, 0.062$\\pm$0.001 and 0.040$\\pm$0.001 mag day$^{-1}$ in $U$, $B$, $V$, $R_c$ and $I_c$ filters respectively. This indicates a clear dependence of flux decline rate on wavelength, being faster at shorter wavelengths. On the other hand the flux decline rates appear to be wavelength independent after flattening with a value of about 0.02 mag day$^{-1}$. The photospheric velocities determined by us are $\\sim$ 21,360 and $\\sim$ 10,740 km s$^{-1}$ about $-2.38$ days before and $+4.60$ days after the $B$ maxima.  We present the bolometric light curve which illustrates the decay of total luminosity of the supernova. The peak luminosity estimated yields the value of $^{56}$Ni mass ejected to be 0.06 M${_\\odot}$. The measured early decline rate of the bolometric light curve is slower than that expected from ${}^{56}$Ni $\\rightarrow$ ${}^{56}$Co decay, but not inconsistent with that expected from a mixture of iron-peaked nuclei (Colgate and McKee 1969). However as seen from the synthetic light curve in Mazzali et al. (2002), this could also represent an optical depth effect. Late time bolometric light curve decline, steeper than that of ${}^{56}$Co $\\rightarrow$ ${}^{56}$Fe decay is may be due to $\\gamma$-rays leakage from the SN envelope.  We intend to monitor SN 2002ap further, in order to study the nature of its flux decline at late phases."
    },
    "0209/astro-ph0209261_arXiv.txt": {
        "abstract": "Using simple geometrical arguments, we paint an overview of the variety of resonant structures a single planet with moderate eccentricity ($e \\lesssim 0.6$) can create in a dynamically cold, optically thin dust disk. This overview may serve as a key for interpreting images of perturbed debris disks and inferring the dynamical properties of the planets responsible for the perturbations. We compare the resonant geometries found in the solar system dust cloud with observations of dust clouds around Vega, $\\epsilon$ Eridani and Fomalhaut. ",
        "introduction": "Direct imaging of nearby stars can not yet detect light from extrasolar planets.  However, imaging can detect circumstellar dust, and when a planet orbits inside a dust cloud, the planet can reshape the cloud dynamically, as the Earth perturbs the solar dust cloud.  Several debris disks around nearby main sequence stars show structures and asymmetries which have been ascribed to planetary perturbations \\citep{burr95, holl98, schn99, koer01, fomalhaut}; perhaps these perturbed disks are signposts of extrasolar planetary systems.  Many of these disk features can be modeled as dust trapped in mean motion resonances (MMRs) with a planet. \\citet{gold75} suggested that as interplanetary dust spirals into the sun under the influence of Poynting-Robertson drag (P-R drag), planets could temporarily trap the dust in MMRs, creating ring-like density enhancements in the interplanetary cloud.  Since then, both the InfraRed Astronomical Satellite (IRAS) and the Diffuse InfraRed Background Explorer (DIRBE) on the Cosmic Background Explorer (COBE) satellite have provided evidence for a ring of dust particles trapped by the Earth \\citep{jack89,reac91,marz94,derm94,reac95}. Models of Kuiper Belt dust dynamics \\citep{liou99} suggest that Neptune may also trap dust in first-order MMRs.  Other stars may host planets like the Earth or Neptune. However, most of the known extrasolar planets do not resemble the Earth or Neptune; they have masses in the range of 0.3--15 Jupiter masses, and they often have significant orbital eccentricities  (see, e.g., the review by Marcy \\& Butler 2000). Simulations by \\citet{kuch01} show that planets as massive as these on eccentric orbits placed in a cloud of inspiraling dust often create two concentrations of dust placed asymmetrically with respect to the star. Maps of the vicinity of Vega made with the IRAM Plateau de Bure interferometer at 1.3~mm \\citep{wiln02} and with the JCMT at 850~$\\mu$m \\citep{holl98} reveal two concentrations of circumstellar emission whose asymmetries can be naturally explained by such a model, possibly indicating the presence of a few-Jupiter mass planet in an eccentric orbit around Vega \\citep{wiln02}.  Other papers have numerically explored the interactions of particular planetary system configurations with a dust disk, with a view towards developing a general key for interpreting disk structures \\citep{roqu94, leca96, liou99, quil02}.  We assemble a primitive version of such a key by mapping the geometries of the MMRs which are likely to trap the most dust near a planet embedded in a debris disk.  We illustrate the patterns formed by the libration centers of the trapped particles in an inertial frame---the frame of a distant observer. These basic patterns allow us to characterize four structures that probably span the range of high-contrast resonant structures a planet on an orbit with eccentricity up to $\\sim 0.6$ and low inclination can create in a dust disk.  Figure~\\ref{fig:patternfig} shows these structures here for reference; we discuss them throughout the paper, particularly in Section 4. Cases I and II in this figure represent the structures formed by planets on low-eccentricity orbits.  Cases III and IV represent structures created by planets on moderately eccentric orbits. Cases I and III represent structures created by planets with substantially less than 0.1\\% of the mass of the star; cases II and IV represent structures created by more massive planets.   \\begin{figure} \\epsscale{0.9} \\plotone{f1.ps} \\figcaption{Four basic resonant structures: I) low mass planet on a low eccentricity orbit, II) high mass planet on a low eccentricity orbit, III) low mass planet, moderate eccentricity orbit, and IV) high mass planet, moderate eccentricity orbit. \\label{fig:patternfig}} \\end{figure} ",
        "conclusions": "Four basic structures probably represent the range of high-contrast resonant structures a planet with eccentricity $\\lesssim 0.6$ can create in disk of dust released on low-eccentricity orbits: a ring with a gap at the location of the planet, a smooth ring, a blobby eccentric ring, and an offset ring plus a pair of clumps. Some of these structures have slowly become revealed in numerical simulations of particular debris disks; we have chased them to their dens in the resonant landscape of the 3-body problem.  The crude key we have assembled should help classify the debris disk structures seen by upcoming telescopes like SIRTF, SOFIA, ALMA, JWST and Darwin/TPF.  Observing one of these structures instantaneously should allow us to categorize the planet as high or low mass (compared to Jupiter orbiting the Sun), and low or moderate eccentricity (compared to $e_0 \\sim 0.2$). In the case of a ring with a gap or an offset ring plus a pair of clumps, the image of the face-on cloud directly indicates the current location of the planet and points to its longitude of perihelion.  In the case of a blobby eccentric ring, numerical modeling can potentially reveal these parameters."
    },
    "0209/astro-ph0209057_arXiv.txt": {
        "abstract": "We investigate the process of an inhomogeneous planetesimal disk evolution caused by the planetesimal-planetesimal  gravitational scattering. We develop a rather general approach based on the kinetic theory which self-consistently describes the evolution in time and space of both the disk surface density and its kinematic properties --- dispersions of eccentricity and inclination. The gravitational scattering of planetesimals is assumed to be in the dispersion-dominated regime which considerably simplifies analytical treatment. The resultant equations are of advection-diffusion type. Distance dependent scattering coefficients entering these equations are calculated analytically under the assumption of two-body scattering in the leading order in Coulomb logarithm. They are essentially nonlocal in nature. Our approach allows one to explore the dynamics of nonuniform planetesimal disks with arbitrary mass and random velocity distributions. It can also naturally include other physical mechanisms which are important for the evolution of such disks --- gas drag, migration, and so on. ",
        "introduction": "\\label{sect:intro}    The discovery of extrasolar giant planets around nearby stars (Mayor \\& Queloz 1995; Marcy \\etal 2000; Vogt \\etal 2000; Butler \\etal 2001) has been one of the most exciting astrophysical findings of the last decade. It has revived interest in planetary sciences and stimulated many theoretical studies. One of the areas which has received a lot of attention recently is the formation of terrestrial planets --- planets like Earth or Venus.  There are many reasons for this interest. First of all, our own Solar System hosts several terrestrial planets and we must understand their formation mechanisms if we want to know the history of our planetary system. Second, despite the huge differences in their physical properties, giant planets are probably linked to the terrestrial ones through their formation mechanism, since it is widely believed that giant planets form as a result of gas accretion on a preexisting rocky core which essentially was a massive Earth-type planet (Mizuno 1980; Stevenson 1982; Bodenheimer \\& Pollack 1986; see however Boss \\etal 2002). Thus terrestrial planets are probably an important component in the evolutionary history of the giant planets. Third, a lot of effort is currently being directed toward designing and building space missions with the ultimate goal of detecting Earth-type planets around other stars. Theoretical understanding of how terrestrial planets had come into being will help us plan these missions most effectively. Finally, observations of IR emission from the dust and debris disks around nearby stars (Heinrichsen \\etal 1999; Schneider 2001) must be interpreted in the context of the formation and evolution of such disks. They are very likely to be the outcome of the same collisional fragmentation and accumulation of massive rocky or icy bodies that form terrestrial planets.   It is widely believed that the formation of Earth-type planets proceeded via agglomeration of large number of planetesimals --- asteroid or comet-like rocky or icy bodies. The theory of this process was pioneered by Safronov (1972) who was the first to point out the importance of the evolution of the dynamical properties of the planetesimal disk for the evolution of its mass distribution. The discovery of a rapid ``runaway'' mode of planetary accretion  by Wetherill \\& Stewart (1989) has made this issue even more important since this mechanism relies on the redistribution of the energy of planetesimal epicyclic motion between populations with different masses (Wetherill \\& Stewart 1993; Kenyon \\& Luu 1998; Inaba \\etal 2001). Significant progress has been made recently in understanding the dynamical evolution of {\\it homogeneous} planetesimal disks --- disks where gradients of surface density or dynamical properties (such as the dispersions of eccentricity and inclination of various planetesimal populations) are absent (Ida 1990; Stewart \\& Ida 2000, hereafter SI; Ohtsuki \\etal 2002). This assumption should be very good during the initial stages of planetesimal growth when there are no massive bodies in the disk. However as coagulation proceeds and planetary embryos --- precursors of terrestrial planets --- emerge this assumption runs into problems. It was first demonstrated by Ida \\& Makino (1993) using N-body simulations  and then confirmed using orbit integrations (Tanaka \\& Ida 1997, 1999) that under a variety of conditions massive protoplanetary embryos would tend to repel planetesimal orbits, clearing out an annular gap in the disk around them.  This process introduces a new spatial dimension into the problem and makes it much harder to treat.   N-body simulations are not good tools to study the details of this process. The primary reason is that they become too time-consuming when one needs to follow the spatial and dynamical properties of a many-body system for many orbital periods. Moreover, the number of planetesimals which they can handle is not very large (less than $10^4$) which precludes the consideration of realistic planetesimal disks containing huge number of bodies with masses spanning an enormous range --- from one meter rocks to $100$ km planetesimals. Finally, one would need to perform a large number of such simulations to explore the whole space of physical parameters relevant for protoplanetary disk evolution. Similar problems, although less severe, plague the performance of methods based of the direct integration of planetesimal orbits using some simplifying assumptions (Tanaka \\& Ida 1997, 1999).  The large number of bodies under consideration poses no problem for the methods of kinetic theory. In this approach the disk is split into a set of planetesimal populations and each of them is characterized by three functions of position: the surface density and dispersions of eccentricity and inclination. Application of this method for studying planetesimal coagulation and the evolution of dynamical properties in homogeneous disks has proven to be very useful (Wetherill \\& Stewart 1989; Kenyon \\& Luu 1998). It can also be easily extended to the case of inhomogeneous disks by allowing the planetesimal properties to vary within the disk. This approach allows one to study many physical mechanisms important for the coagulation process --- gravitational interactions between planetesimals increasing their random motion, gas drag and inelastic collisions damping them, migration, fragmentation, etc. The use of this statistical approach for exploring inhomogeneous disks was first undertaken by Petit \\& H\\'enon (1987a, 1987b, 1988) in their studies of planetary rings. Physical conditions in planetary rings (high optical depth and frequent inelastic collisions) are very different from those which are thought to exist in protoplanetary disks. Thus we cannot directly apply the methods of Petit \\& H\\'enon to study planetesimal disks, but the spirit in which their investigation was carried out can be largely preserved.   In this paper we consider  the evolution of {\\it inhomogeneous} planetesimal disks caused by interactions between planetesimals using conventional methods of statistical mechanics (Lifshitz \\& Pitaevskii 1981). The effects of the planetary embryos on the disk evolution will be investigated later in Rafikov (2002; hereafter Paper II). We will concentrate on one of the most important processes going on in these disks --- mutual gravitational scattering of planetesimals --- although our rather general approach allows one to treat other relevant phenomena as well. This process  can proceed in two distinct regimes depending on the amplitude of planetesimal random motion. The gravitational attraction between two planetesimals with masses $m_1$ and $m_2$ becomes stronger than the tidal field of the central star of mass $M_c$ when their mutual separation is less than their Hill (or tidal, or Roche) radius $r_H$, defined as \\begin{eqnarray} r_H=a_0\\left(\\frac{m_1+m_2}{M_c}\\right)^{1/3}, \\label{eq:Hill_radius} \\end{eqnarray} where $a_0$ is the distance from the central star. When the random velocities of the epicyclic motion of interacting planetesimals are smaller than $\\sim \\Omega r_H$ ($\\Omega=\\sqrt{GM_c/a_0^3}$ is the disk orbital frequency at $a_0$) their relative approach velocities are small and close interactions can lead to a considerable change of the orbital elements of planetesimals. This velocity regime is called shear-dominated (or ``cold''). It should be contrasted with the other extreme --- the so called dispersion-dominated  (``hot'') regime which occurs when planetesimal velocity dispersions are bigger than $\\sim \\Omega r_H$. In this latter case scattering is typically weak which often allows analytical treatment of this velocity regime.    The development of planetesimal disk inhomogeneities driven by a protoplanetary embryo was explored in Rafikov (2001; hereafter Paper I) assuming that shear-dominated scattering of planetesimals prevails. It was also assumed in this study that the dynamical properties of planetesimals do not evolve as a result of scattering and that the disk always stays dynamically cold. Planetesimal-planetesimal interactions play the role of effective viscosity in the disk, and tend to homogenize it and close up any gap. Nevertheless, this study demonstrated that gap formation is the natural outcome of the embryo-planetesimal interaction when the embryo is massive enough. These interactions were local in character because in the shear-dominated case only planetesimals on orbits separated by no more than several $r_H$ (corresponding to their encounter) were able to approach each other closely.  It is however more likely that the planetesimal-planetesimal gravitational scattering in realistic protoplanetary disks proceeds in the dispersion-dominated (rather than shear-dominated) regime, at least on the late stages of disk evolution (see \\S \\ref{sect:FP_expansion}). In this case the evolution of planetesimal random motion can strongly affect the growth rate of protoplanetary embryos. It is also tightly coupled to the evolution of spatial distribution of planetesimals because any change of the energy of epicyclic motion comes at the expense of the orbital energy of planetesimals.  The treatment of the dispersion-dominated case is complicated by the fact that planetesimals in this regime can explore different regions of the disk in the course of their epicyclic motion. This makes disk evolution a nonlocal process. On the other hand, as we have said before there are natural simplifications which are valid in the dispersion-dominated regime. These include the two-body scattering approximation (relative velocities are high), Fokker-Planck type expansions (scattering is weak), and so on.   Our paper is organized as follows. In \\S \\ref{sect:master_equation} we derive general equations for the planetesimal surface density and velocity dispersion evolution. We do this using the Hill approximation which is briefly described in \\S \\ref{subsect:Hill_eq}. A Fokker-Planck expansion of the evolution equations, valid in the dispersion-dominated regime, is performed in \\S \\ref{sect:FP_expansion}. In \\S \\ref{sect:scattering_coefficients} we derive analytical expressions for the scattering coefficients used in these equations. We conclude with a brief summary of our results in \\S \\ref{sect:Summary}. Some technical details of the calculations and derivations can be found in appendices. ",
        "conclusions": "\\label{sect:Summary}    We have derived a self-consistent set of equations describing the coupled evolution of the surface density and kinematic properties of a planetesimal disk driven by gravitational encounters between planetesimals. The assumption of a dispersion-dominated velocity regime is used throughout the calculation, which is reasonable for planetesimal disks in their late evolutionary stages.  Thus this paper serves as a logical extension of Paper I which was devoted to the study of the shear-dominated velocity regime.  The evolution equations (\\ref{eq:FP_surf}) and (\\ref{eq:FP_ecc}) are of advection-diffusion type; the coefficients entering them are nonlocal which is in contrast with previous results (Paper I; Ohtsuki \\& Tanaka 2002). This is a natural outcome of the scattering in the dispersion-dominated regime since planetesimals can explore large portions of the disk (compared to their Hill radii) in the course of their epicyclic motion.  We have also derived analytical expressions for the scattering coefficients entering the different terms of evolution equations (\\ref{eq:Ups_1_surf}), (\\ref{eq:Ups_2_surf}), (\\ref{eq:Ups_0_e})-(\\ref{eq:Ups_2_e}). The analytical treatment was enabled by the use of the two-body scattering approximation which becomes valid in the dispersion-dominated regime. Our expressions are accurate up to fractional errors $\\sim (\\ln\\Lambda)^{-1}$, where $\\ln\\Lambda\\gg 1$ is a Coulomb logarithm.  Following the methods developed in SI and Ohtsuki \\etal (2002) for the case of homogeneous planetesimal disks it might be possible to improve our calculations by taking subdominant higher order terms into account (they contribute typically at the level  $\\sim 10\\%-20\\%$ and are neglected in our present consideration).   Using this system of evolution equations supplemented with the information about the behavior of the scattering coefficients one can self-consistently study the evolution of inhomogeneous planetesimal disks. Arbitrary distribution of masses of interacting planetesimals is allowed for but the evolution of mass spectrum is not considered in the present study. Thus our equations describe the disk evolution on timescales short compared with the timescale of the mass spectrum evolution. This should be a good assumption for studying effects on the disk caused by the gravity of massive protoplanetary embryos (because they can induce changes of the disk properties on rather short timescales). This deficiency can also be easily remedied in the future when the need to study very long-term disk evolution arises. Our approach can naturally incorporate physical mechanisms other than just gravitational scattering, for example gas drag or migration [see Tanaka \\& Ida (1999)] and we are going to study their effects in the future.  In the following paper (Paper II) we describe the embryo-planetesimal scattering and derive  equations governing this process in various velocity regimes. This would allow us to  provide a complete description of the disk evolution caused by both the presence of isolated massive bodies and the continuous sea of planetesimals."
    },
    "0209/astro-ph0209111_arXiv.txt": {
        "abstract": "The relationship between observed variability time and emission region geometry is explored for the case of emission by relativistic jets.  The approximate formula for the jet-frame size of the emission region, $R'=Dc\\Delta t_{\\rm obs}$ is shown to lead to large systematic errors when used together with observed luminosity and assumed or estimated Doppler factor $D$ to estimate the jet-frame photon energy density.  These results have implications for AGN models in which low-energy photons are targets for interaction of high energy particles and photons, e.g.\\ synchrotron-self Compton models and hadronic blazar models, as well as models of intra-day variable sources in which the photon energy density imposes a brightness temperature limit through Compton scattering.  The actual relationship between emission region geometry and observed variability is discussed for a variety of geometries including cylinders, spheroids, bent, helical and conical jet structures, and intrinsic variability models including shock excitation.  The effects of time delays due to finite particle acceleration and radiation time scales are also discussed. ",
        "introduction": "In the standard picture of active galactic nuclei (AGN), accretion onto a super-massive black hole is via an accretion disk, and a significant fraction of the accretion power (possibly supplemented by tapping into the rotational energy of the black hole) produces twin opposing relativistic jets moving outward along the disk axis, with typical Lorentz factors $\\Gamma \\sim 2$--10 as inferred from very long baseline interferometry (VLBI) observations.  The objects observed in high energy $\\gamma$ rays are ``blazars'', AGN in which one of the jets is closely aligned toward the observer.  It is natural that in $\\gamma$-rays we should see preferentially those AGN with aligned jets because the emission from the jet is Doppler boosted in energy and relativistically beamed along the jet direction (for a discussion of relativistic effects see Urry and Padovani 1995).  The $\\gamma$ ray emission from blazars is variable (as it is also at optical, UV and X-ray energies).  Relativistic effects also cause the observed variability time to be shorter than the time scale over which the emission changes in the jet frame.  The spectral energy distribution (SED) of blazars shows two broad peaks, the low energy peak extending from the infrared to the UV or X-ray region of the spectrum, and the high energy peak starting in the X-ray or $\\gamma$ ray range.  The usual interpretation is that relativistic electrons produce the low energy part by synchrotron emission, and that the same electrons produce the high energy part by Compton scattering the low energy part and/or external photons to higher energies.  The 3rd EGRET catalog of high-energy $\\gamma$-ray sources (Hartman et al.\\ 1999) contains around 70 high confidence identifications of AGN, and all appear to be blazars (Montigny et al.\\ 1995, Mukherjee et al.\\ 1997).  Clearly, the $\\gamma$-ray emission is associated with AGN jets.  Four BL Lac objects have been detected in the TeV energy range: Mrk~421 (Punch et al.\\ 1992), Mrk~501 (Quinn et al.\\ 1996), 1E~S2344+514 (Catanese et al.\\ 1998) and PKS~2155-304 (Chadwick et al.\\ 1999).  Recently, the spectrum of Mrk 501 has been measured up to 24 TeV by the HEGRA telescopes (Konopelko et al.\\ 1999).  Several of the EGRET AGN show $\\gamma$-ray variability with time scales of $\\sim 1$ day (Kniffen et al.\\ 1993) at GeV energies.  The TeV $\\gamma$-ray emission of two BL Lacs shows very rapid variability.  For Mrk 421, variability on a time scale as short as $\\sim 15$ minutes has been reported (Gaidos et al.\\ 1996).  In the case of Mrk 501, variability on a time scale of a few hours was observed during the 1997 high level of activity, and there is evidence of a 23 day periodicity (Protheroe et al.\\ 1998, Hayashida et al.\\ 1998) interpreted in terms of a binary black hole model for the central engine by Rieger and Mannheim (2000).  These variability timescales place important constraints on the models.  For example, the synchrotron self-Compton (SSC) model appears to be just consistent with recent multi-wavelength observations of Mrk~421 and Mrk~501 during flaring activity (Bednarek and Protheroe 1997, 1999).  However, the allowed range of physical parameters (Doppler factor and magnetic field) is rather small, and this mechanism may well be excluded by future observations.  For a recent review of TeV $\\gamma$-ray astronomy see Kifune (2002).  Rapid variability in intra-day variability (IDV) sources is a long-standing problem as it implies apparent brightness temperatures in the radio regime which may exceed 10$^{17}$ K or relativistic beaming with extremely high Doppler factors, coherent radiation mechanisms, or special geometric effects (Wagner and Witzel 1995).  The very rapid flaring observed at TeV and X--ray energies during flaring activity in blazars also presents a challenge for any model and suggests a re-examination of mechanisms which may cause very rapid variability would be worthwhile.  In this paper I concentrate on how the observed variability time is related to the geometry and motion of the emission region, and thus to the photon energy density in the emission region.  The blazar emission mechanisms to be discussed include: a shock excited emission region, bent jets, a shock propagating along a jet containing a helical structure and illuminating parts of the helix by enhanced interactions/emission of radiation such that the emission regions move along helical paths, and highly oblique conical shocks in the jet.  Together with geometry-specific time delays and variable Doppler boosting associated with relativistic motion of the emission region along a curved trajectory, it may well be possible to explain the observed flaring activity and high brightness temperatures. Another possibility briefly discussed in the context of bent jets and conical shocks is that jets may be fueled on an irregular time scale.  The observed variability time $\\Delta t_{\\rm obs}$, and some assumed or estimated Doppler factor $D$ is often used to estimate the jet-frame source radius, $R'\\approx Dc\\Delta t_{\\rm obs}$. The jet-frame photon energy density is then usually assumed to be $U'_{\\rm phot} \\approx L'/4\\pi {R'}^2 c$, where $L'$ is the jet-frame bolometric luminosity given by $L'=D^{-4} 4\\pi d_L^2 F$, $F$ being the observed bolometric flux, and $d_L$ being the luminosity distance.  However, this approach can lead to large systematic errors in the jet-frame photon energy density.  This is important because the energy density of photons of the low energy part of the SED may determine the energy losses of electrons, and the rate of up-scattering to $\\gamma$-ray energies in SSC models, and the rate of proton-photon collisions in hadronic models (see, e.g., Mannheim and Biermann 1989, Protheroe 1997, Mannheim et al.\\ 2001, M\\\"{u}cke and Protheroe 2001, M\\\"{u}cke et al.\\ 2002).  If the emission region is optically thin, it may also have consequences for IDV sources as I will show that it is quite possible that the photon energy density responsible for the so-called Compton catastrophe may actually be lower than usually estimated.  In the following sections I shall discuss how the variability time is related to the emission region geometry and intrinsic jet-frame variability, show how this can lead to large systematic errors in the jet-frame photon energy density, and discuss other scenarios which can lead to rapid variability and high fluxes. ",
        "conclusions": "Many factors can influence observed variability time.  The connection between $\\Delta t_{\\rm obs}$, Doppler factor and emission region geometry is non-trivial, and so measuring $\\Delta t_{\\rm obs}$ may give, at best, one of the dimensions of the emission region. Using $R'=c\\Delta t_{\\rm obs}D$ and $U'_{\\rm phot}=L'/4\\pi {R'}^2 c$ may then lead to over-estimation or under-estimtion of the jet-frame photon energy density by orders of magnitude.  This is clearly of importance in any AGN model in which the low energy photons produced in the jet are targets for interaction of high energy particles or radiation, such as in SSC models and hadronic blazar models.  Although not discussed in detail in the present paper, the escape of $\\gamma$-rays from the emission region depends on the optical depth to photon-photon pair production interactions.  This optical depth can be uncertain by orders of magnitude in the same way as the photon energy density, and will also depend on viewing angle.  One must therefore be careful when using the observation of apparently unattenuated gamma-rays, and an observed variability time, to place limits on the Doppler factor.  The uncertainty in the jet-frame photon energy density discussed in this paper may also have implications for the high brightness temperature/Compton catastrophe problem of IDV sources.  In this case, it is the energy density of target photons which limits the brightness temperature through the competition of inverse-Compton scattering with synchrotron radiation, and the target photon energy density may actually be lower than estimated if the emission region is non-spherical.  If the jet is bent or helical, or has some other favoured geometry (e.g. conical shocks) cusps in $t'$ vs.\\ $t_{\\rm obs}$, and/or a varying Doppler factor may cause narrow peaks in the observed lightcurve irrespective of other factors. Distinguishing between these cases from the observed lightcurve alone is likely to be difficult.  One way of distinguishing whether a flare is due to (i) an emission region moving around a bent or helical path, or (ii) a shock is exciting a curved, conical or helical structure within a jet, is that the in the first case the flare is caused by a change in viewing angle with respect to the motion of the emission region leading to a change in Doppler factor, whereas in the second case the flare is due to a pile-up in observation times with no change in Doppler factor. Hence, in case (i) not only will the observed flux increase during a flare, but the photon energies also increase -- increase in $(\\nu F_\\nu)_{\\rm peak}$ by factor $x^4$ accompanied by shift in $\\nu_{\\rm peak}$ by factor $x$ ($x$ is ratio of final to initial Doppler factor).  In case (ii), however, since there is no change in Doppler factor there should be no shift in $\\nu_{\\rm peak}$ accompanying an increase in $(\\nu F_\\nu)_{\\rm peak}$.  Distinguishing between the excitation of a conical and a helical structure by a plane shock would be almost impossible -- note the qualitative similarities between lightcurves depicted in Figs.~\\ref{agn_tvar_helix}(d) and \\ref{LC_of_coneshocks}(b). Relativistic jet simulations (Bowman 1994, Gomez et al.\\ 1997) do show the presence of conical shocks, and these shocks appear after a large perturbation (e.g. from 4 to 10 in the simulations of Gomez et al.\\ 1997) in Lorentz factor of the matter entering the jet through the nozzle.  This can result in a sequence of quasi-stationary to superluminal reverse and forward conical shocks extending from the nozzle to the perturbation as it moves along the jet (Agudo et al.\\ 2001).  For the conical shock model (Salvati et al.\\ 1998) discussed here to work, a subsequent perturbation would need to result in a plane shock, or plane thin perturbation of some kind, which could travel along the jet and excite the pre-existing conical shocks.  As far as I am aware, whether or not this could occur has not been demonstrated.  A similar uncertainty hangs over whether or not helical jet structures, which may themselves be shocks with a twisted ribbon topology, resulting from a perturbation entering the jet through the nozzle, could subsequently be excited by the passage of a plane shock or perturbation.  Nevertheless, in both cases, if the viewing angle is favourable pile-ups in $t_{\\rm obs}$, and hence flares, could occur simply as a result of the motion of the conical or helical patterns which may themselves be sites of enhanced emission.  Note that in recent 3D relativistic jet simulations, the introduction of a 1 percent helical velocity perturbation at the nozzle results in a helical pattern propagating along the jet at nearly the beam speed (Aloy et al.\\ 1999), and that the conical shocks resulting from a perturbation in jet Lorentz factor can range from being quasi-stationary to superluminal (Agudo et al.\\ 2001).  In conclusion, in models for flaring in AGN in which the emission comes from a localized region (blob) co-moving with the jet, time variability is non-trivial to interpret in terms of emission region geometry and Doppler factor.  A further complication is that flaring may arise instead due to curved or helical motion of a blob, even if the emission is constant in the istantaneous rest frame of the blob.  In this case, apparent flaring is due to the change in viewing angle, and hence Doppler factor.  Similarly, if the viewing angle is favourable, relativistic motion of curved or helical filaments or surfaces can lead to observation of flares. Excitation of curved or helical jet structures by shocks or perturbations can also lead to pile-ups in $t_{\\rm obs}$, and hence large apparent increases in flux.  Observations of time variability in AGN is therefore non-trivial to interpret and may lead to large systematic errors in estimated jet-frame photon energy density, Doppler factor and the physical parameters of the emission region."
    },
    "0209/astro-ph0209105_arXiv.txt": {
        "abstract": "We discuss the timescales for alignment of black hole and accretion disc spins in the context of binary systems.  We show that for black holes that are formed with substantial angular momentum, the alignment timescales are likely to be at least a substantial fraction of the systems' lifetimes.  This result explains the observed misalignment of the disc and the jet in the microquasar GRO J 1655-40 and in SAX J 1819-2525 as being likely due to the Bardeen-Petterson effect.  We discuss the implications of these results on the mass estimate for GRS 1915+105, which has assumed the jet is perpendicular to the orbital plane of the system and may hence be an underestimate.  We show that the timescales for the spin alignment in Cygnus X-3 are consistent with the likely misalignment of disc and jet in that system, and that this is suggested by the observational data. ",
        "introduction": "It is often stated that astrophysical black holes present one of the best laboratories available for the study of general relativity.  At the same time, relatively few concrete tests of general relativity have been made with observations of accreting black holes.  The Lense-Thirring effect is the distortion of space in the presence of a rotating compact object through the dragging of inertial frames.  When the effect is important, the central part of an accretion disc is forced to rotate in the same plane as the black hole (the Bardeen-Petterson effect - see Bardeen \\& Petterson 1975).  The geometry of the accretion flow changes significantly in the vicinity of the black hole and hence to allow for sensitive tests of general relativity.  The radius out to which this effect is observed is such that the Lense-Thirring precession period due to the warping of the disc is equal to the timescale for angular momentum to drift through the disc.  Past claims of evidence for the BP effect have come from interpretations of the quasi-periodic oscillations in X-ray binaries (Stella \\& Vietri 1998; see also Fragile, Mathews \\& Wilson 2001 for a more recent discussion).  The jets of radio galaxies are often seen to show a constant spatial direction over their $10^8$ year lifetimes (Alexander \\& Leahy 1997; Liu, Pooley, \\& Riley 1992).  Until recently, the origin for this stability had been suggested to be that the spin of the black hole provides a ``flywheel'' effect.  Assuming that the jets are powered by disc accretion, the jets should be perpendicular to the discs.  The BP effect requires that the disc plane near the black hole be perpendicular to the black hole's spin vector.  Given the suggestion that relativistic jets may be powered by extracting the spin energy of the central black hole (Blandford \\& Znajek 1977) and the expectation that the timescale for the black hole to align itself with the disc by accreting matter with net angular momentum should be at least $10^8$ years (Rees 1978), this picture became commonly accepted.  More recent observational and theoretical work challenged this picture for the case of active galactic nuclei.  The inner dust discs of some samples of nearby Seyfert galaxies are found to be perpendicular to their radio jets (van Dokkum \\& Franx 1995; but see Kinney et al. 2000 for contradictory results).  The jets in radio galaxies, on the other hand, appear not to be perpendicular to the dust discs (Schmitt et al. 2002).  If the Bardeen-Petterson effect is important and the black hole spins are not intrinsically correlated with the dust discs' angular momentum vectors, then alignment should be observed only in the case of the initial alignment of black hole and disc spins, as the size scales on which HST can resolve the discs are substantially larger than the BP radius.  The observations of alignments in the early work on Seyferts prompted theorists to explain the steady directions of AGN jets by making the accretion disc, rather than the black hole the ``flywheel'' (Natarajan \\& Pringle 1998) and showing that the black hole's spin should be much more quickly aligned with the disc's angular momentum if one took into account the efficient transfer of angular momentum in the disc plane (Papaloizou \\& Pringle 1983) as shear perpendicular to the disk place can be transported more efficiently than shear within the disk plane, so warp in a nonplanar disc may decay on a timescale much faster than the accretion timescale for a planar disk.  Additional theoretical support for the basic assumptions in Natarajan \\& Pringle (1998) has come from more detailed calculations, both analytical (Ogilvie 1999) and numerical (Torkelsson et al. 2000; Gammie, Goodman \\& Ogilvie 2000) but see also Nelson \\& Papaloizou (2000) who suggest a longer timescale; all of these papers note that the exact timescales depend on details of angular momentum transport which are currently poorly understood.  The angular momentum evolution of black holes in binary systems was studied by King \\& Kolb (1999), where it was found that black hole spins are unlikely to change substantially due to accretion, but has not been re-visited given the hypothesis of efficient warp transfer suggested by Papaloizou \\& Pringle (1983), and the specific problem of the change in the spin direction of a rapidly rotating black hole has not been considered.  In two X-ray binaries, GRO J 1655-40 and SAX J 1819-2525 (also known as V4641 Sag), measured binary orbital plane inclination angles differ substantially from the measured jet inclination angles, providing strong evidence that the Bardeen-Petterson effect may be relevant.  A third system, SS 433, shows precessing jets (Hjellming \\& Johnston 1981) which imply that either the jets are not perpendicular to the orbital plane, or that the orbital plane itself precesses, which could occur if the system is a hierarchical triple (e.g. Fabian et al. 1986).  With this motivation, we compute the timescale for a black hole in a binary system to align its spin with the angular momentum of the binary orbit, and show that in contrast to active galactic nuclei, X-ray binaries containing black holes should routinely have misaligned jets.  We consider the implications for the mass estimate of GRS 1915+105, which assumes that the jet is perpendicular to the binary plane.  We discuss the case of Cygnus X-3, whose orbital inclination angle is not well constrained, and show that its jet would have become aligned with the orbital plane only for an unlikely range of parameters.  We do not apply this model to SS 433, because the nature of its binary companion is not well known and it is often speculated that this system is undergoing thermally unstable, highly super-Eddington mass transfer (e.g. Verbunt \\& van den Heuvel 1995 and references within), for which our thin disc approximations are highly invalid. ",
        "conclusions": "We have shown that the timescale for the black hole spin to align with the accretion disc's angular momentum in binary systems is often longer than the lifetime of the binary system.  The Bardeen-Petterson effect should then be important in these systems if the black holes were formed with substantial angular momentum.  This represents a fundamental difference between black holes in binary systems and the black holes in active galactic nuclei which should align in a relatively short fraction of their lifetimes (as shown by NP).  Strong evidence of this effect is seen from GRO J 1655-40 and SAX J1819-2525.  That the timescale for spin alignment in GRO J 1655-40 is so close to the main sequence lifetime of the companion star hints at a possible explanation for the difference between radio galaxies (which are unaligned) and Seyfert galaxies (which are often aligned).  The strong jets in the radio galaxies may carry away a substantial amount of angular momentum, keeping the matter and its angular momentum from reaching the black hole and making the spin changing estimate of NP98 a severe underestimate.  The Seyfert galaxies have a much lower fraction of their total power taken away by the radio jets, so this effect will not be as severe for them.  Secondly, if the Blandford-Znajek mechanism is responsible for powering the radio jets, it is likely that the dimensionless spin parameter values of the black holes in radio galaxies are higher than those in the Seyfert galaxies, giving another reason why the timescales for the radio galaxies' black holes to change their spins might be longer.  The dynamical mass estimate for GRS 1915+105 depends upon the inclination angle of the system being such that the disc is perpendicular to the jets.  We have shown that for high mass, high spin black hole primaries in GRS 1915+105, the alignment timescale of the black hole's spin with that of the accretion flow can be longer than the characteristic age of the system.  On the other hand, the orbital/epicyclic resonance model for the quasi-periodic oscillations recently discovered in this system suggest that the black hole mass cannot be too far outside the error bars from the dynamical measurement.  As a result, the inclination angle separation between the disc and the jet is likely to be less than about 25 degrees.  Thus the system must have either been formed with a small offset between disc and jet angle, or the offset between the two angles is largely in the plane of the sky rather than in the inclination angle.  Future planned X-ray interferometry missions such as MAXIM may have the potential for measuring the position angles of systems such as GRS 1915+105 by imaging the plane of the ``hot spot'' where accretion stream hits the accretion disc.  We discuss suggestive evidence that the disc and jet are unaligned in Cygnus X-3, and show that this is a result to be expected in wind-fed accreting black holes even with very small initial angular momenta.  A neutron star origin is also consistent with the binary orbital data, since even a rapidly rotating neutron star would have too much angular momentum to be aligned in the lifetime of the donor star.  It is likely that the system will need to be observed in quiescence before polarization measurements can tell us its true orbital inclination angle.  Future observations of inclination angles in microquasars should help determine whether the results for these four systems are common."
    },
    "0209/astro-ph0209619_arXiv.txt": {
        "abstract": "We have detected the $\\hozs ~ (\\lambda = 2.1218 \\micron)$ and $\\htos ~ (\\lambda = 2.2477 \\micron)$ lines of H$_2$ in the Galactic centre, in a $90 \\times 27$ arcsec region between the northeastern boundary of the non-thermal source, Sgr A East, and the giant molecular cloud (GMC) M-0.02-0.07. The detected $\\hoz$ emission has an intensity of 1.6 -- 21 $\\intensity$ and is present over most of the region. Along with the high intensity, the broad line widths (FWHM = 40 -- 70 $\\kms$) and the $\\hto$ to $\\hozs$ line ratios (0.3 -- 0.5) can be best explained by a combination of C-type shocks and fluorescence.  The detection of shocked H$_2$ is clear evidence that Sgr A East is driving material into the surrounding adjacent cool molecular gas. The H$_2$ emission lines have two velocity components at $\\sim$ +50 $\\kms$ and $\\sim$ 0 $\\kms$, which are also present in the NH$_3$(3,3) emission mapped by \\citet*{mcg01}.  This two-velocity structure can be explained if Sgr A East is driving C-type shocks into both the GMC M-0.02-0.07 and the northern ridge of \\citet*{mcg01}. ",
        "introduction": "Sgr A East has frequently been interpreted as a supernova remnant due to its shell structure and non-thermal spectrum (\\citealt{jon74}; \\citealt{gos83} and references therein; and see the more recent references in \\citealt{mae02}). Some recent research, however, has suggested that the energetics, size, and elongated morphology ($3 \\times 4$ arcmin or $7 \\times 9$ pc at $d = 8.5$ kpc) of Sgr A East cannot have been produced by a typical supernova \\citep{yus87,mez89}. \\citet{mez89} estimate the required energy to produce Sgr A East to be more than $4 \\times 10^{52}$ ergs.  Modeling of the entire spectrum of Sgr A East by \\citet{fat99}, which fits very well with the observations of the non-thermal emission of Sgr A East and EGRET $\\gamma$-ray sources, supports the energy estimate by \\citet{mez89}.  Those authors concluded that a single supernova explosion could explain the existence of Sgr A East only if it occurred within the cavity formed by the stellar wind from a progenitor star. In that scenario, however, the formation of the cavity takes too much time ($\\sim 10^6$ yr) compared with the orbital period ($\\sim 10^5$ yr) of matter circling around the Galactic centre \\citep{mez89}.   \\begin{figure*} \\special{psfile=\"fig1.eps\" angle=0 vscale=95 hscale=95 voffset=-270 hoffset=90} \\vspace{97mm} \\caption{Central $10 \\times 15$ pc region of the Galaxy. Contours representing the velocity-integrated map of NH$_3$(3,3) emission are overlaid on a 6 cm continuum image of Sgr A complex from \\citet{mcg01}. The black dot at the centre of image is Sgr A* and the mini spiral is Sgr A West. The CND is traced by the brighter part of continuum surrounding them and Sgr A East is seen by the outer part extended to the boundary of NH$_3$ contours. The GMC M-0.02-0.07 lies to the east of this region. The dashed box at the northeastern edge of Sgr A East encloses the $90 \\times 27$ arcsec region observed in H$_2$. The two solid lines in the box are the locations from which the H$_2$ spectra shown in this paper were extracted.  Cut B is made across the 10 slits while the NE--SW cut, perpendicular to the edge of Sgr A East, is Slit 9. Letters mark the positions of OH(1720 MHz) masers with error ellipses scaled up by a factor of 15 \\citep{yus99a}. \\label{fig_target}} \\end{figure*}   \\citet{yus87} suggested that a different kind of explosive event could create Sgr A East. The energy required to make a huge shell such as Sgr A East has been associated with a `hypernova' \\citep*{woo99}. \\citet{kho96} hypothesize that Sgr A East may be the remnant of a solar mass star tidally disrupted by a massive black hole.  Their model can naturally explain the elongated shape of Sgr A East as well as the extreme energetics. However, from their observation with the {\\it Chandra X-ray Observatory}, \\citet{mae02} suggest that Sgr A East should be classified as a metal-rich `mixed morphology' supernova remnant .  They argue that the model of \\citet{kho96} cannot reproduce the metal-rich abundances observed at the centre of Sgr A East. They also conclude that a single Type II supernova explosion with an energy of $10^{51}$ ergs into an homogeneous ambient medium with a density of $10^{3} \\; \\cmv$ can most simply explain the results of both radio and X-ray observations, and thus that the extreme energy of $\\sim 10^{52}$ ergs is not required.  In principle, the energy of the explosive event can be directly measured by studying regions where Sgr A East is colliding with ambient interstellar material. By tracing the dynamics of molecular gas, an interaction between the eastern part of Sgr A East and the giant molecular cloud (GMC) M-0.02-0.07 (also known as the `$50~\\kms$ cloud') has been inferred \\citep*{gen90,ho91,ser92,mez96,nov99,coi00}. Recent observations of NH$_3$(3,3) emission in the region show that Sgr A East impacts material to the north and west as well (see Fig.~\\ref{fig_target}) \\citep*{mcg01}.  As direct evidence of this interaction, several 1720 MHz OH masers, which are a good diagnostic of the continuous, or C-type, shock excitation \\citep*{fra96,war99}, have been detected along the southern edge of Sgr A East and to the north of the circum-nuclear disc (CND) \\citep{yus96}.  \\citet{war99} and \\citet{yus99b,yus01} detected H$_2$ line emission in regions where OH-masers have been detected. In most cases H$_2$ line emission arises either from thermal excitation (e.g. by shock heating) or from non-thermal excitation by far-UV absorption \\citep{bla87,bur92,pak98}. One can in principle distinguish between these two mechanisms by comparing near-infrared (near-IR) line intensities.  The $\\hto$ / $\\hozs$ ratio has been an effective discriminant in a number of shocked regions (where the ratio should be $\\le 0.3$) and photodissociation regions (PDRs) (where it is about 0.5 -- 0.6).  However, a `thermal' line ratio can be observed in a PDR -- even though fluorescence is the dominant excitation mechanism -- if the gas density is high ($\\ge 10^5 \\; \\cmv$; \\citealt{ste89}). \\citet{gat84} observed the CND and concluded that the H$_2$ molecules are excited by collisions, while the results for larger regions (about $2 \\times 2$ deg$^2$) by \\citet*{pak96a,pak96b} are consistent with non-thermal excitation. The interpretation of \\citet{war99} and \\citet{yus99b,yus01} that the line emission in Sgr A East is thermal is supported by the presence of the 1720 MHz OH masers.  It is therefore likely that Sgr A East is indeed driving shocks into the adjacent GMCs to the south and into the CND.   \\begin{figure*} \\special{psfile=\"fig2.eps\" angle=0 vscale=73 hscale=73 voffset=-505 hoffset=5} \\vspace{134mm} \\caption{$\\hoz$ spectra at 10 positions along Cut B.  Indicated positions are relative to the centre of the cut ($\\rm \\alpha = 17^h 45^m 46\\fs1, ~ \\delta = -28\\degr59\\arcmin01\\arcsec$; J2000).  The dotted lines are Gaussian fits to the observed line profiles. The spectra are not corrected for instrumental broadening. \\label{fig_spect_bcut}} \\end{figure*}   The fields observed by \\citet{war99} and \\citet{yus99b,yus01} are restricted to the vicinity of the CND and cover only some of the regions where interaction of the Sgr A East shell with surrounding material is expected. Before one can hope to estimate the energy released in the event that created Sgr A East, it is necessary to observe additional interaction regions in diagnostic lines of H$_2$ at high spectral resolution. In this paper we present velocity-resolved, near-IR H$_2$ observations at the northeastern boundary of Sgr A East.  By measuring $\\hto$ / $\\hozs$ line ratios and line profiles simultaneously we aim to study the excitation and kinematics of the interaction between Sgr A East and M-0.02-0.07. ",
        "conclusions": "We observed the northeastern part of the Sgr A East shell in order to investigate its interaction with the GMC M-0.02-0.07. The bright $\\hoz$ emission is strong evidence that Sgr A East is physically adjacent to, and interacting with, M-0.02-0.07.  By comparing the relative intensities of $\\hoz$ and $\\hto$ emission, the distribution of the $\\hto$ / $\\hozs$ line ratio, and the radial velocities of the H$_2$ emission, we can to some extent distinguish between excitation mechanisms for the H$_2$.  The line ratios tend to support emission in either fast J-type shocks or a dense PDR.  However, on considering the bright $\\hoz$ intensity, the large line widths, and the spatial variation in the line ratio, we conclude that a combination of C-type shocks and fluorescence is required. The presence of shocks is direct evidence that Sgr A East is driving into the surrounding material, and is consistent with the detection of 1720 MHz OH masers to the north of the CND and to the south of Sgr A East \\citep{yus96}. Very recently \\citet{kar03} detected the 1720 MHz OH masers also at two positions near our target region, which is more direct evidence supporting our conclusion on the C-type shocks.  The H$_2$ emission covers most parts of our targeted region ($90 \\times 27$ arcsec).  The line profiles are made up of two velocity components both of which extend over a significant portion of the region ($15 \\times 27$ arcsec). We find that the NH$_3$(3,3) emission lines observed by \\citet{mcg01} also show a similar kinematic structure, with almost the same velocities. We suggest that the H$_2$ line emission arises at the interfaces between Sgr A East and two independent molecular clouds, with line-of-sight velocities of $\\sim$ +50 $\\kms$ (M-0.02-0.07) and $\\sim$ 0 $\\kms$ (the northern ridge). Both the observed two velocity components of the H$_2$ emission and the difference in the intensity distributions between the H$_2$ and NH$_3$ emission can be understood if the molecular clouds are composed of small dense clumps with a very small filling factor.  To study the origin and evolution of Sgr A East, it would be important to know the total H$_2$ luminosity and the total cooling rate (based on that) over the interaction region, which could be compared with those of well studied SNRs. However, it is very difficult to estimate them with the small amount of information we have at present. Given the uncertainty in the emission mechanisms, even estimating the total H$_2$ luminosity in the small mapped region would be difficult, without considering the entire interaction region. It would be premature for us to estimate the required energy to make the Sgr A East shell. We will be able to do this in the future, after observations of more of the interaction region."
    },
    "0209/astro-ph0209043_arXiv.txt": {
        "abstract": "\\noindent We present our recently developed 3-dimensional chemodynamical code for galaxy evolution. It follows the evolution of all components of a galaxy such as dark matter, stars, molecular clouds and diffuse interstellar matter (ISM). Dark matter and stars are treated as collisionless $N$-body systems. The ISM is numerically described by a smoothed particle hydrodynamics (SPH) approach for the diffuse (hot) gas and a sticky particle scheme for the (cool) molecular clouds. Additionally, the galactic components are coupled by several phase transitions like star formation, stellar death or condensation and evaporation processes within the ISM. As an example here we present the dynamical, chemical and photometric evolution of a star forming dwarf galaxy with a total baryonic mass of $2 \\times 10^9 {\\rm M}_\\odot$. \\\\ ",
        "introduction": "Since several years smoothed particle hydrodynamics (SPH, \\cite{M1992}) calculations have been applied successfully to study the formation and evolution of galaxies. Its Lagrangian nature as well as its easy implementation together with standard $N$-body codes allows for a simultaneous description of complex dark matter-gas-stellar systems \\citep{NW1993, MH1996}. Nevertheless, until now the present codes lack of processes that are based on the coexistence of different phases of the interstellar medium (ISM), mainly dissipative, dynamical and stellar feedback, element distributions, etc. We have therefore developed a 3d chemodynamical code which is based on our single phase galactic evolutionary program \\citep{Ber1999, Ber2000}.  \\begin{figure}[t!] \\centerline{% \\begin{tabular}{c@{\\hspace{0.1in}}c} \\includegraphics[width=2.35in]{100-mass.ps} & \\includegraphics[width=2.35in]{100-sigma.ps} \\end{tabular}} \\caption{The radial distribution of the cumulative mass (left) and the surface density (right) for the different components in the central region of the model galaxy after 1~Gyr.} \\label{fig:100-mass&100-sigma} \\end{figure}  This code includes many complex effects such as a multi-phase ISM, cloud-cloud collisions, a drag force between different ISM components, condensation and evaporation of clouds (CE), star formation (SF) and a stellar feedback (FB). The more detailed description of the new code and the full list of the interaction processes between all gaseous and stellar phases will be presented in a more comprehensive paper by \\cite{BHTS2002}. Here we just briefly describe some basic features and effects.  In our new (multi-phase gas) code we use a two component gas description of the ISM \\citep{TBH1992, SHT1997}. The basic idea is to add a cold (10$^2$ - 10$^4$ K) cloudy component to the smooth and hot gas (10$^4$ - 10$^7$ K) described by SPH. The cold clumps are modeled as $N$-body particles with some ``viscosity'' \\citep{TH1993} (cloud-cloud collisions and drag force between clouds and hot gas component). The cloudy component interacts with the surrounding hot gas also via condensation and evaporation processes \\citep{CMcKO1981, KTH1998}. In the code we introduce also star formation. The ``stellar'' particles are treated as a dynamically separate (collisionless) $N$-body component. Only the cloud component forms the stars. During their evolution, these stars return chemically enriched gas material and energy to both gaseous phases.  \\begin{figure}[t!] \\centerline{% \\begin{tabular}{c@{\\hspace{0.1in}}c} \\includegraphics[width=2.35in]{mass-t.ps} & \\includegraphics[width=2.35in]{dmdt-t.ps} \\end{tabular}} \\caption{The temporal evolution of the mass (left) and mass exchange rate (right) for the different components of the model galaxy.} \\label{fig:mass-t&dmdt-t} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209275_arXiv.txt": {
        "abstract": "We present high-resolution optical spectra of 15 objects near or below the sub-stellar limit in the Upper Scorpius and $\\rho$ Ophiuchus star-forming regions. These spectra, obtained with the HIRES instrument on the Keck I telescope, are used to investigate disk accretion, rotation and activity in young very low mass objects. We report the detection of a broad, asymmetric H$\\alpha$ emission line in the $\\rho$ Oph source GY 5 which is also known to harbor mid-infrared excess, consistent with the presence of an accreting disk. The H$\\alpha$ profiles of the Upper Sco objects suggest little or no on-going accretion. Our results imply that if most brown dwarfs are born with disks, their accretion rates decrease rapidly, at timescales comparable to or smaller than those for T Tauri disks. The Upper Sco brown dwarfs appear to be rotating faster than their somewhat younger counterparts in Taurus, consistent with spin-up due to contraction following disk unlocking. The H$\\alpha$ activity is comparable to saturated activity levels in field M dwarfs with similar spectral type and rotation rates. Comparison of our data with published (albeit lower-resolution) spectra of a few of the same objects from other epochs suggests possible variability in accretion/activity indicators. ",
        "introduction": "In their pre-main-sequence phase, low mass stars accrete high angular momentum material from a circumstellar disk.  The star remains a comparatively slow rotator, however, apparently due to disk-locking, i.e., magnetic braking by large-scale stellar field lines that thread the disk.  In the classical T Tauri phase, the fields may also mediate accretion by truncating the inner disk and channeling in infalling material (K\\\"onigl 1991; Ostriker \\& Shu 1995).  While many independent observations support this picture for stars with masses $>$0.2\\msun, whether it applies to objects at or below the hydrogen-burning limit is only now being explored.  The formation mechanisms of brown dwarfs are still open to debate. Most recently, Padoan \\& Nordlund (2002) have argued that brown dwarfs form in the same way as more massive stars, via `turbulent fragmentation'. Reipurth \\& Clarke (2001), on the other hand, suggested that brown dwarfs are stellar embryos ejected from newborn multiple systems before they accreted sufficient mass to eventually start hydrogen burning. In this model, dynamical interactions are expected to prune their disks. Studies of disk accretion, rotation and activity in young brown dwarfs could help distinguish between these scenarios. In particular, it is of considerable interest to investigate whether some or all young sub-stellar objects undergo a T Tauri-like phase and if so how long that phase lasts.  Accretion signatures have been seen already in high-resolution spectra of a few very low mass sources. Muzerolle et al. (2000) detected an asymmetric H$\\alpha$ line profile in the M6 object V410 Anon 13 in Taurus and used magnetospheric accretion models to show that it is accreting from a disk but at a much lower rate than that found in higher mass stars. White \\& Basri (2002) found two more Taurus M5.5-M6.5 objects with H$\\alpha$ profiles similar to other classical T Tauri stars.  Here we present high-resolution optical spectroscopy of a much larger sample of objects near or below the sub-stellar limit in the nearby ($\\sim$ 150 pc) Upper Scorpius and $\\rho$ Ophiuchus star-forming regions. Our targets are sources with M5 or later spectral types from the surveys of Ardila, Mart\\'in, \\& Basri  (2000) for Upper Sco and Wilking, Greene \\& Meyer (1999) for $\\rho$ Oph. Our goals are to search for broad, asymmetric H$\\alpha$ lines indicative of disk accretion, to measure rotational properties as denoted by {\\it v~sin~i}, and to investigate chromospheric activity, in these young very low mass objects in comparison to their older counterparts in the field and to (higher-mass) T Tauri stars.  In a subsequent paper (Mohanty et al., in prep), we will use these optical spectra, together with observed photometry and theoretical isochrones, to derive effective temperatures, gravities, masses and ages for our sample objects. For the purposes of this paper, we assume an age of $\\sim$ 1 Myr for $\\rho$ Oph and $\\sim$ 5 Myr for Upper Sco (Preibisch \\& Zinnecker 1999). If we also assume that the effective temperatures of our young M5-M8 objects are comparable to those of field M dwarfs of the same spectral type ($\\sim$ 3000-2500 K, similar to what is derived by Wilking, Greene \\& Meyer 1999 for their $\\rho$ Oph sample), then theoretical models (Chabrier et al. 2000) imply masses of $\\sim$ 0.1 -- 0.02 \\msun. Thus, our sample spans a mass range from the stellar/sub-stellar boundary well into the brown dwarf domain. ",
        "conclusions": "Table 1 lists the equivalent widths of Li I and H$\\alpha$ lines, full widths at 10\\% of the peak flux level of H$\\alpha$, and $v~sin~i$ values of our sample, along with the relevant error estimates. Of our 14 targets in Upper Sco, 11 show Lithium 6708 \\AA~ absorption and have similar radial velocities (standard deviation $\\sim$ 2 kms$^{-1}$).  The remaining three (USco 85, USco 99 and USco 121) do not show Lithium, and are likely non-members; we do not discuss them further. Except for GY 5, our $\\rho$ Oph sources do not have sufficient signal-to-noise in the blue to measure Li equivalent widths, primarily because they suffer from significant extinction. However, their membership in the $\\rho$ Oph molecular cloud has been established by Wilking, Greene \\& Meyer (1999) using near-infrared photometry and spectroscopy.  Figure 1 shows H$\\alpha$ line profiles of our sample. The $\\rho$ Oph source GY 5 exhibits a broad, asymmetric H$\\alpha$ profile while none of the Upper Sco targets do. There is evidence of a measurable change in the H$\\alpha$ emission of USco 128 between two observations separated by about an hour.  \\subsection{Disk Accretion} Accreting (``classical'') T Tauri stars exhibit strong, broad H$\\alpha$ line profiles indicative of high velocities in a nearly free-falling flow (e.g., Hartmann, Hewett, \\& Calvet 1994). Weak-line T Tauri stars, on the other hand, harbor weak, narrow H$\\alpha$ lines, presumably originating in their active chromospheres. The equivalent width of the H$\\alpha$ line is often used to distinguish between these two types of objects. However, the threshold value of H$\\alpha$ equivalent width depends on the spectral type (Mart\\'in 1998). White \\& Basri (2002) suggested full-width at 10\\% of the peak emission as a more accurate empirical indicator of accretion than either the H$\\alpha$ equivalent width or optical veiling: 10\\% widths $>$ 270 kms$^{-1}$ indicate accretion {\\it independent of the stellar spectral type}.  The $\\rho$ Oph source GY 5 has a 10\\% H$\\alpha$ width of 352 kms$^{-1}$ (and an equivalent width of 65 \\AA~): it appears to be undergoing accretion. Interestingly, it is one of the sources with a mid-infrared excess detected by the Infrared Space Observatory (ISO), consistent with the presence of a circumstellar disk (Comeron et al. 1998; Bontemps et al. 2001; Natta et al. 2002). Thus, GY 5 may be the first spectroscopically confirmed sub-stellar object with an accreting disk detected via infrared excess as well as H$\\alpha$ characteristics. Surprisingly, Wilking, Greene, \\& Meyer (1999) did not see H$\\alpha$ emission ($<$ 5 \\AA) in their low-resolution optical spectrum of GY 5 (albeit obtained ``under nonideal conditions''). It may be that the accretion onto GY 5 is highly variable (or highly asymmetric so that geometric effects are important).  According to ISO mid-infrared flux measurements, GY 141 and GY 310 also harbor excess emission but GY 37 does not (Comeron et al. 1998; Bontemps et al. 2001). H$\\alpha$ line profiles in our optical spectra do not show clear evidence of on-going accretion in any of these three objects. Given the problems of extinction, it would be interesting to obtain near-infrared spectra of all $\\rho$ Oph sources to investigate other accretion signatures such as Pa $\\beta$ and Br $\\gamma$. Given the large ISO beam, it would also be prudent to confirm via high angular resolution observations (e.g., ground-based L- and N-band imaging) that the mid-infrared sources indeed coincide with the brown dwarfs.  By the White \\& Basri (2002) criterion, none of the Upper Sco objects in our sample shows evidence of on-going accretion. The case of USco 128, however, is intriguing. In low-resolution spectra, Ardila et al. (2000) reported a `constant' H$\\alpha$ equivalent width of 130 \\AA~ in two observations separated by a month. Our two HIRES spectra, separated by just one hour, yield equivalent widths of 16 and 25 \\AA; i.e., there was a measurable change in the line over a short time interval (see Fig. 1). While it is true that low-resolution spectra of late-type objects systematically overestimate H$\\alpha$  as a result of blending with the 6569 \\AA~ TiO band-head, a factor of $\\sim$5 difference between the Ardila et al. (2000) values and ours is difficult to account for in this way. However, variation in the H$\\alpha$ width by more than a factor of 5 is not uncommon in flares. It may be that chromospheric activity in USco 128 is highly variable. In that case, the large width reported by Ardila et al. (2000) may correspond to a period of sustained high activity and flaring.  The mean age of stars in the Upper Sco region is $\\sim$5 Myr whereas $\\rho$ Oph sources are even younger at $\\sim$1 Myr (e.g., Preibisch \\& Zinnecker 1999). Our results suggest that if most brown dwarfs are born with disks, their accretion rates decrease rapidly, within the first few million years. Such a conclusion is also consistent with measurements of disk frequency as a function of age using infrared excess. While a large fraction --$\\sim$60\\%-- of brown dwarf candidates in the $\\sim$1-Myr-old Trapezium cluster show near-infrared excess (Muench et al. 2001), the fraction appears to be much lower in the somewhat older $\\sigma$ Orionis ($\\sim$3-8 Myr) and TW Hydrae ($\\sim$10 Myr) associations (Jayawardhana, Ardila \\& Stelzer 2002). Thus, it appears that at least the inner disks of brown dwarfs deplete rather quickly, at timescales comparable to or smaller than those for T Tauri stars (Jayawardhana et al. 1999). Disk dissipation timescales in Upper Sco might be even shorter than in some other star-forming regions due to strong winds and ionizing radiation from numerous luminous stars in its midst (Preibisch \\& Zinnecker 1999 and references therein). Studies of large samples of brown dwarfs in several young clusters, spanning a range of ages and environments, will provide a more definitive answer.  \\subsection{Stellar Rotation and Activity} Measures of stellar rotation and chromospheric activity in young brown dwarfs, in comparison with their older counterparts in the field and higher-mass coeval T Tauri stars, can shed light on a variety of questions, from the nature of sub-stellar magnetic fields to the efficiency of disk locking.  A large fraction of the lowest-mass stars in Orion appear to be fast rotators whereas Taurus objects of similar spectral type rotate much more slowly (Clarke \\& Bouvier 2000; Herbst et al. 2001; White \\& Basri 2002). According to a scenario advocated by Hartmann (2002), late-type Orion objects have not yet had time to slow down via disk braking whereas (somewhat older) Taurus objects have. In the standard picture, following disk unlocking, the low-mass stars (and presumably brown dwarfs) would spin up again. Eventually, braking by magnetized winds is expected to slow the stars down again, though it appears that such braking may not be efficient in the lowest mass stars and brown dwarfs (Mohanty \\& Basri 2002a; Mohanty \\& Basri 2002b). In the first few Myr, in any case, contraction timescales are expected to be much shorter than braking timescales, so spin-up is expected to dominate (once the disk locking ends) over this period (Bouvier, Forestini \\& Allain 1997).  Among the $\\rho$ Oph targets, we find $ v~sin~i \\approx $ 14 kms$^{-1}$ in GY 5, which very likely harbors a disk. It is interesting to note that GY 141 and GY 310, which also show mid-infrared excess, are relatively slow rotators ($\\sim$ 6 and 10 \\kms, respectively) while GY 37, which does not show a mid-infrared excess, is a rapid rotator ($\\sim$ 22 \\kms). This may be a hint of disk-locking in action, though further observations and a larger sample are required for verification.  Among the 11 Upper Sco members in our sample (all apparently non-accretors), six show rotational velocities $ v~sin~i >$ 15 kms$^{-1}$.  The average $v~sin~i$ of the Upper Sco sample is $\\sim$ 25 \\kms. In seven similar mass non-accretors in Taurus, on the other hand, White \\& Basri (2002) find an average $v~sin~i$ of $\\sim$ 10 \\kms, with only one object rotating above 15 \\kms. Thus, the Upper Sco brown dwarfs appear to be rotating noticeably faster than their Taurus counterparts. Since Upper Sco is believed to be somewhat older than Taurus, we may be seeing the signature of spin-up due to gravitational contraction following disk unlocking.  In our Upper Sco sample, the equivalent widths of H$\\alpha$ arising from chromospheric activity are in the range $\\sim$ 6-18 \\AA~ (except one larger width in USco 128). Without accurate effective temperature determinations, we cannot yet calculate the H$\\alpha$ fluxes these widths correspond to. However, the H$\\alpha$ widths are comparable to saturated widths in field M dwarfs of similar spectral type (Mohanty \\& Basri 2002a; Mohanty \\& Basri 2002b). Thus, if the effective temperatures of the Upper Sco objects are similar to those of field dwarfs of the same spectral type, then their H$\\alpha$ fluxes would correspond to saturated levels in the field. This issue will be investigated further in a subsequent paper (Mohanty et al., in prep)."
    },
    "0209/astro-ph0209513_arXiv.txt": {
        "abstract": "We present 47 spectroscopically-confirmed quasars discovered behind the Magellanic Clouds identified via photometric variability in the MACHO database.  Thirty-eight quasars lie behind the Large Magellanic Cloud and nine behind the Small Magellanic Cloud, more than tripling the number of quasars previously known in this region.  The quasars cover the redshift interval $0.2 < z < 2.8$ and apparent mean magnitudes $16.6\\le \\overline{V} \\le 20.1$.  We discuss the details of quasar candidate selection based on time variability in the MACHO database and present results of spectroscopic follow-up observations.  Our follow-up detection efficiency was 20\\%; the primary contaminants were emission-line Be stars in Magellanic Clouds.  For the 47 quasars discovered behind the Magellanic Clouds plus an additional 12 objects previously identified in this region, we present 7.5-year MACHO $V$- and $R$-band lightcurves with average sampling times of 2-10 days. ",
        "introduction": "Techniques to find quasars, largely successful in other regions of the sky, have had limited results towards the Magellanic Clouds \\citep{tin99,sch99,dob02}.  Crowding, recent star formation and significant dust extinction cause major quasar surveys to avoid these regions entirely, resulting in very few quasars known over the substantial sky coverage of the Magellanic Clouds.  The most successful selection method to date in this region has been at X-ray wavelengths.  Although many tens of sources background to the Magellanic Clouds have been identified in the X-rays \\citep{kah99,hab99,sas00}, counterparts to these sources at other wavelengths have been stymied by postitional uncertainties; targeted X-ray follow-up has allowed optical identification of $\\sim 20$ extragalactic sources in the Magellanic Cloud region \\citep{cra97,sch99}.  The MACHO lightcurve database \\citep{alc97,alc00} provides an opportunity to search for quasars behind the Magellanic Clouds via an alternative method: optical variability.  Optical variability has been studied by many groups as a means of constraining models of the quasar central engine \\citep[][and references therein]{hoo94,cri97,sir98, haw02}, as well as a method of quasar identification behind globular clusters \\citep{meu02}. Although a handful of gravitationally-lensed quasars have well-sampled lightcurves on the timescale of years \\citep{alc02,hjo02}, most studies have had short time baselines and poor resolution.  In one of the largest optical monitoring programs, \\citet{giv99} observed a sample of 42 quasars over 7~years with an average sampling interval of 40~days.  Long-term optical variability of quasars in this study show no strong evidence for underlying periodic structure.  This is in sharp contrast to the majority of stellar sources in the Magellanic Clouds which either do not vary or do so periodically.  We have used this difference to separate quasar candidates from the overwhelming stellar background in the Clouds.  A comprehensive search for quasars behind both the Large and Small Magellanic Clouds (LMC and SMC, respectively) is motivated in part by the lack of a suitable reference frame against which to measure the proper motion of the Clouds.  Previous proper motion estimates have suffered from an insufficient number or poorly distributed set of reference objects \\citep{jon94,kro97,ang00}.  Since the proper motion of the Clouds is expected to be only a few mas/year, a well distributed set of point-like background quasars could significantly improve the accuracy of this measurement, constraining the orbital history of these galaxies.  These objects may also prove useful as light beacons for absorption line studies of the interstellar medium in the Magellanic Clouds \\citep{gib00,pro02}, as has been done for suspected extragalactic X-ray sources in this region \\citep{kah01,hab01}.  Finally, this search was also motivated by interest in the quasars themselves, in hope that the dense time sampling of the MACHO lightcurves will provide clues to the physical mechanisms underlying quasar light variation.  We discuss the MACHO database and our optical variability quasar candidate selection methods in \\S\\,\\ref{select}.  In \\S\\,\\ref{spectra}, we describe spectroscopic follow-up observations and present 47 quasars discovered behind the Magellanic Clouds.  In \\S\\,\\ref{lightcurves}, MACHO lightcurves are presented for these quasars.  Finally, in \\S\\,\\ref{summary} we summarize our results and discuss future quasar searches in this region.  Finding charts, light curves and spectra for the quasars in this paper are available on request from the authors or on a website given at the end of \\S\\,\\ref{summary}. ",
        "conclusions": "\\label{summary}  We present 47 quasars discovered behind the Magellanic Clouds: 38 behind the LMC and 9 behind the SMC, significantly increasing the number density of known quasars in this region.  The quasars cover the redshift interval $0.2 < z < 2.8$ and apparent mean magnitudes $16.6\\le \\overline{V} \\le 20.1$.  Candidate quasars were identified based on aperiodic variability in the MACHO database.  MACHO light curves are presented for the newly discovered quasars as well as 12 quasar/AGNs identified in previous studies.  The primary contamination during spectroscopic follow-up were quasi- or aperiodic Be/Ae stars in the Magellanic Clouds.  Spectroscopic follow-up of quasar candidates in the MACHO database is not yet complete.  In the outer LMC, 52 MACHO fields have become accessible for variability-only (\\S\\,\\ref{vselect}) selection since the original candidate selection was run.  In addition, $\\sim 100$ candidates have not yet been followed-up from the search presented in this paper.  We therefore expect the MACHO database to yield many more sources in the future.  A similar photometric variability selection has been applied to another microlensing database in the Magellanic Clouds, OGLE-II \\citep{eye02}, but has not yet been followed-up spectroscopically.  We have checked our sample of quasars against the OGLE candidate list presented by \\citeauthor{eye02}.  Four of the MACHO quasars are listed as candidates in this paper, (OGLE candidates: L92, L114, L155, S12), eight OGLE candidates are within $1''$ of spectroscopically confirmed Be stars (L51, L87, L103, L121, L148, L153, S8, S25) and six OGLE candidates are common to our list which have not yet been followed-up spectroscopically.  Despite similar quasar candidate selection strategies, the majority of the OGLE candidates did not make our candidate list.  This is due in part to our final subjective rejection of Be/Bumper lightcurves: 24 objects classified as quasar candidates by \\citeauthor{eye02} were considered Be/Bumper star candidates according to MACHO photometry; the remaining objects in the OGLE list either did not pass our minimum variability cut or fell outside MACHO fields.  Continued follow-up of both MACHO and OGLE candidates is certain to increase significantly the number quasars in the Magellanic Cloud region.  Quasars behind the Magellanic Clouds are extremely useful tools with which to study the Clouds themselves.  Our motivation for this study was to uncover a robust set of reference objects against which to measure the proper motion of the Magellanic Clouds.  First epoch {\\it Hubble Space Telescope} images have been scheduled for a subset of quasars presented in this paper; second epoch imaging is expected to allow an estimation of the Cloud's orbital motion with sufficient accuracy to constraint models of the Galactic halo.  Quasars are also invaluable tools for a variety of other studies, for example as probes of the Clouds' interstellar medium.  We therefore provide finding charts, lightcurves and spectra for all quasars presented in this paper, available electronically at http://www.ucolick.org/$\\sim$mgeha/MACHO or on request from the authors."
    },
    "0209/astro-ph0209039_arXiv.txt": {
        "abstract": "Shine a flashlight on a black hole, and one is greeted with the return of a series of concentric rings of light. For a point source of light, and for perfect alignment of the lens, source, and observer, the rings are of infinite brightness (in the limit of geometric optics). In this manner, distant black holes can be revealed through their reflection of light from the Sun. Such retro-MACHO events involve photons leaving the Sun, making a $\\pi$ rotation about the black hole, and then returning to be detected at the Earth. Our calculations show that, although the light return is quite small, it may nonetheless be detectable for stellar-mass black holes at the edge of our solar system. For example, all (unobscured) black holes of mass $M$ or greater will be observable to a limiting magnitude $\\bar{m}$, at a distance given by: $0.02\\,\\mbox{pc}\\times\\sqrt[3]{10^{(\\bar{m}-30)/2.5}\\,(M/10\\,M_\\sun)^{2}}$. Discovery of a Retro-MACHO offers a way to {\\em directly}\\/ image the presence of a black hole, and would be a stunning confirmation of strong-field general relativity. ",
        "introduction": "In the discovery of MACHOs light has shown its power to reveal dark compact objects. In these events the photon from a distant source suffers a very small angular deflection, small enough to make gravitational lensing the relevant mechanism.  The bending power of a black hole is not limited, however, to small angles but reaches to $\\pi$ and odd multiples of $\\pi$. Illuminated by a powerful point source of light, the black hole therefore will shine back with a series of concentric rings (we call this retrolensing). A waterdrop, too, likewise illuminated, shines back, but for a different reason: the internal reflections experienced by the photons. That returned light shows to the air traveler flying over a fogbank as a ``glory'': a rainbow-like halo surrounding the shadow of the plane on the cloud. Each ray of light that contributes to this sensation has suffered its $\\pi$ deflection in a different waterdrop, therefore the impression of colored rings that the glory makes is impression only, built up in the last analysis out of multitudes of tiny dots of illumination. No one would be so rash as to expect a detectable backscatter from a single water droplet. It is precisely the search for such backscatter from a single black hole (the putative retro-MACHO) that is, however, the topic of this paper.  How search for retro-MACHOs, for ``$\\pi$ in the sky'', especially when dogged by the negative implications of the familiar phrase? To tell observers to go and look everywhere is to them as dismaying as to be told to look nowhere. Fortunately the successful search for MACHOs provides a helpful model, and the records from that search an immediate place of reference. Recent microlensing events appear to directly confirm that a population of stellar-mass black holes exists in our galaxy~\\citep{macho2,macho3,agol2}.  The Sun, in turn, furnishes a powerful nearby source with whose help one can hope to search a nearby region of the galaxy for black holes not otherwise revealed, taking advantage of times when the Earth interposes itself to spare the registering device from the direct glare of the light source. Although the observation of a retro-MACHO is unlikely, it nonetheless affords one of the few ways to directly image nearby black holes, and if ever observed, would make for an impressive confirmation of Einstein's theory. ",
        "conclusions": ""
    },
    "0209/astro-ph0209380_arXiv.txt": {
        "abstract": "We have determined the central velocity dispersions and surface brightness profiles for a sample of powerful radio galaxies in the redshift range $0.06<z<0.31$, which were selected on the basis of their young radio source. The optical hosts follow the fundamental plane of elliptical galaxies, showing that young radio sources reside in normal ellipticals, as do other types of radio galaxies into which these objects are believed to evolve. As young radio sources are relatively straightforward to select and the contributions of the AGN light to the optical spectra are minimal, these objects can readily be used to study the evolution of the fundamental plane of elliptical galaxies out to z=1, independently of optical selection effects.  The black hole masses, $M_{bh}$, of the objects in our sample have been determined using the tight empirical relation of $M_{bh}$ with central velocity dispersion, $\\sigma_e$, and for literature samples of classical radio galaxies and optically selected ellipticals. Only the optically selected in-active galaxies are found to exhibit a correlation between $M_{bh}$ and radio luminosity. In contrast, the radio powers of the AGN in the samples do not correlate with $M_{bh}$ at all, with objects at a given black hole mass ranging over 7 orders of magnitude in radio power.  We have been able to tie in the local population of powerful radio sources with its parent population of in-active elliptical galaxies: the local black hole mass function has been determined using the elliptical galaxy luminosity function and the Faber-Jackson and $M_{bh}-\\sigma$ relations. This was combined with the fraction of radio-loud black holes as function of $M_{bh}$, as determined from the optically selected galaxy sample, to derive the local volume-density of radio galaxies and the distribution of their black hole masses. These are shown to be consistent with the local radio luminosity function and the distribution of black hole masses as found in the radio selected samples, and confirms that elliptical galaxies comprise the large majority of the radio-loud population of active galaxies. ",
        "introduction": "\\subsection{Galaxy formation and the fundamental plane} How and when galaxies were formed and how they have evolved since, is one of the central issues of contemporary astrophysics today. Elliptical galaxies may be the best objects to study galaxy evolution, since their optical light is not dominated by relatively recent star-formation, unlike that of spiral galaxies, and their observational properties are a better reflection of their long-timescale cosmological development.  In the 1980s it was shown that the main global properties of local elliptical galaxies, velocity dispersion, surface brightness and effective radius, are strongly correlated, and that they describe a plane in this 3-dimensional parameter space (Dressler et al., 1987; Djorgovski \\& Davis, 1987). Initially, this `fundamental plane' (FP) of elliptical galaxies drew attention mainly for its use as a distant indicator, to measure the Hubble constant and peculiar motions in the local universe. However, it was also realised that it implied a strong regularity in the process of galaxy formation, in particular that the mass-to-light (M/L) ratio varied only slightly over 4 orders of magnitude in galactic mass (Djorgovski \\& Davis, 1987).  Since the M/L ratio is a strong function of the age of the stellar population, the study of the fundamental plane as function of redshift is a sensitive diagnostic of the evolutionary history of galaxies. van Dokkum, Franx and Kelson and collaborators, who observed elliptical cluster galaxies with the Keck telescope and the HST, located at z=0.39 (van Dokkum \\& Franx 1996), z=0.58 (Kelson et al. 1997), and z=0.83 (van Dokkum et al. 1998), showed that the M/L ratio evolves as $\\Delta$ log $M/L_B \\propto -0.4z$ ($\\Omega_m=0.3$,$\\Omega_\\Lambda = 0$), putting strong constraints on the formation epoch of cluster ellipticals, with $z_{form}>2.8$ for a Salpeter initial mass function.  The Mg\\,b or Mg$_2$ indices are also valuable measures of the age of the stellar population, which is often used in conjunction with the FP.  The Mg$_2$ index is found to be correlated with the central velocity dispersion in local elliptical galaxies, with only a small scatter (eg. Dressler et al., 1987; Djorgovski \\& Davis, 1987, J\\/orgensen, Franx \\& Kjaergaard 1996). Ziegler \\& Bender (1997) showed that for cluster galaxies at z=0.37, this correlation is offset by $<$Mg\\,b$>$ $\\approx -0.4$\\AA\\, which can be fully attributed to the lower luminosity-weighted ages of these stellar populations at this earlier epoch.  These studies may seem to suggest that, except for perhaps some minor details, the evolution of elliptical galaxies is well understood. However, various strong biases may still be present in the studies described above. The most serious bias may be the progenitor bias (van Dokkum \\& Franx 1996): present day ellipticals which were spirals at an earlier epoch, would not have been included in a sample at that epoch, biasing the sample to what are likely to be the oldest galaxies. Furthermore, there is also an environmental bias: present studies are focused on cluster galaxies, while it is not known how these clusters evolve with cosmological epoch. For example, it is far from clear what a local cluster such as Coma would have looked like at high redshift, and to what present-day structures the high-z clusters have evolved into. More generally, connecting populations at one redshift with their predecessor at high redshift, without making uncertain assumptions about the history of star formation is a problem.  \\subsection{The role of nuclear activity in galaxy formation and evolution}  The place of active galaxies in the general picture of galaxy evolution is unclear. Until recently, AGN were seen as rare and peculiar objects, which, although interesting for their own sake, were disconnected from galaxy evolution issues. However, the recent discovery that every galaxy seems to have a central massive black hole, the mass of which is closely related to the total mass of (the bulges of) its host (Magorrian et al, 1998; Gebhardt et al. 2000), has changed this perception dramatically. It is now clear that the formation and evolution of the central massive black hole is closely linked to that of the host galaxy (eg. Silk \\& Rees 1998). In addition, it implies that all galaxies may be capable of having phases of central activity, meaning that active galaxies are just normal galaxies, with their central black hole coincidently caught during a period of (high) accretion. This places active galaxies back in the centre of attention of galaxy formation and evolution studies, especially since they are relatively easy to select at  high redshift. Recent claims suggest that non-thermal radio emission from ellipticals is strongly correlated with the mass of their central black hole, ranging from dwarf ellipticals to powerful radio galaxies (Franceschini et al. 1998). This would make the connection between quiescent and active galaxies even stronger.  \\subsection{Radio galaxies as probes of galaxy evolution}  Since active galaxies play such a potentially important role in galaxy evolution, it is important to determine their fundamental structural parameters as function of redshift. An additional factor is that selection of AGN allows the study of galaxy evolution independently of the optical selection biases as mentioned above. Smith, Heckman \\& Illingworth (1990) have studied the fundamental plane parameters of nearby powerful radio galaxies, and showed that they indeed reside in normal ellipticals. A similar result has been found recently by Bettoni et al. (2002). Unfortunately, it is almost impossible to use similar objects at higher redshift, since their spectral energy distributions have invariably been found to be strongly influenced by AGN light, especially in the optical (eg. optical/UV alignment effect, McCarthy et al. 1987, Chambers et al. 1987).  This problem can be avoided by selecting galaxies containing a very young radio source. In the very early stages of radio source evolution, when the radio source is only 100-1000 years old and confined to the central hundred parsecs, the contribution of the AGN to the global properties of its host galaxy is still minimal (Snellen et al. 1996a,b; 1999): broadband photometry showed that their optical to near-infrared colours are consistent with passively evolving ellipticals, indicating that their light is dominated by old stellar populations. The dispersion in their $R$ and $K$ band Hubble diagrams (0.3 mags) is smaller than that for other classes of radio galaxies, indicating a homogeneous and well behaved population of hosts (Snellen et al. 1996a,b). In addition, their optical spectra typically only show weak emission lines but deep stellar absorption lines, again indicative of them being dominated by old stellar populations (Snellen et al. 1999). For example, the archetype young radio source, B0108+388, is as powerful as Cygnus A. However, it has a remarkably low OIII$_{5007}$ equivalent width of only $\\sim$4\\AA$ $(Lawrence et al. 1996), corresponding to a line luminosity about a factor 50 lower than that of Cygnus A.  That the contamination from young radio-loud AGN is only small or even absent may not come as a surprise. The young radio source may not yet have pierced through the dense and dusty central regions of the host. AGN related light is also likely to still be confined to the inner hundreds of light years, and obscured from our view. There is indeed much evidence through HI absorption studies of young radio-loud AGN that large column densities of gas are present (eg. Carilli et al. 1999; Peck, Taylor \\& Conway 1999)  In this paper, we investigate the use of the host galaxies of young radio-loud AGN as probes of galaxy evolution. The connection between radio activity and black hole mass, as infered from their central velocity dispersion, has also been investigated. We present high dispersion spectroscopic and optical CCD imaging data of a sample of 7 young radio-loud AGN located between $0.06<z<0.31$. Section 2 describes the properties of young radio-loud AGN and the selection of the sample. Section 3 describes the observations, reduction and analysis. Section 4 and 5 present the results and discussion. For consistency with previous papers on this subject, we use a cosmology with $H_0$=50 km sec$^{-1}$ Mpc$^{-1}$, $\\Omega_m$=1.0, $\\Omega_\\Lambda$=0, throughout this paper, unless stated otherwise. ",
        "conclusions": "\\begin{figure*} \\psfig{figure=figure4.ps,width=16cm} \\caption{\\label{funplane} The fundamental plane of the host galaxies of young radio-loud AGN (solid squares), with on the horizontal axis their effective radii, and on the vertical axis a combination of the velocity dispersion and their effective brightness, corrected for the cosmological evolution in the mass-to-light ratio as found by van Dokkum et al. (1998). The solid line indicates the fundamental plane of elliptical galaxies, as derived from nearby galaxies by Faber et al. (1989; small circles). The diamonds are the radio galaxies measured by Smith et al. (1990).} \\end{figure*}  The results of the imaging and spectroscopy are given in table \\ref{results}. Column 1 gives the B1950 name, columns 2 and 3 give the redshift and the velocity dispersion corrected to an aperture of $r_e/4$. Column 4 and 5 give the effective radii in arcseconds and kpc. Column 6 gives the brightness level at the effective radius for rest-frame B-band, corrected for cosmological dimming and galactic foreground extinction. Columns 7 and 8 give the rest-frame equivalent widths for the Mg b band (also corrected to an aperture of $r_e/4$) and the [OIII]5007\\AA\\ emission line. Column 9 gives the velocity shift of the [OIII] line with respect to absorption line redshift, where a positive value indicates a redshift. The uncertainties in $I_e$ are typically 0.05 magnitude, while the uncertainties in the equivalent widths are $\\sim10\\%$.  The part of the spectra used to measure the velocity dispersions of the galaxies is shown in figure \\ref{absorption}, in order of increasing redshift. The thick lines indicate the best fit to the velocity broadened K-star spectra. The measured velocity dispersions range from from 146 km sec$^{-1}$ for B0830+5813, to 228 km sec$^{-1}$ for B0019-000. Figure \\ref{r4law} show the normalised galaxy profiles with their best fits. The observed effective radii range from 1.1$''$ for B009$-$0001, to 11.9$''$ for B0830+5813. Note that the profile of B0830+5813 slightly flattens in its inner part. The best fitted velocity broadened K-star spectra were subtracted from those of the galaxies to obtain their emission line spectra. These are shown in figure \\ref{emission}. The three adjacent panels to the right zoom in on the [OIII]4959\\AA, [OIII]5007\\AA, and the [NI]5199\\AA$ $ emission lines, with the dashed lines indicating their rest-frame wavelengths, assuming the absorption line redshifts. it shows that the emission line redshifts correspond well with those infered from the absorption lines, except for B2352+4933, which emission lines are clearly redshifted by 350 km/sec.  \\subsection{The Fundamental Plane}  \\begin{figure} \\psfig{figure=figure5.ps,width=8cm} \\caption{\\label{funevol}The offset of the host galaxies to the local fundamental plane for cluster ellipticals as function of redshift. The dotted line is the evolution in mass-to-light ratio as found by van Dokkum et al. (1998). The symbols are as in figure \\ref{funplane}. There is a possible hint that the more distant galaxies are brighter, but more data are needed to confirm this.} \\end{figure} \\begin{figure} \\psfig{figure=figure6.ps,width=8cm} \\caption{\\label{sigmaMgb}The Mg\\,b-$\\sigma$ relation for the host galaxies of young radio-loud AGN (squares). The solid line indicates the Mg\\,b-$\\sigma$ relation for local ellipticals from Faber et al. (1989; small circles).} \\end{figure}  \\begin{figure} \\psfig{figure=figure7.ps,width=8cm} \\caption{\\label{OIII_Mgb}The offsets of the Mg\\,b indices to the Mg\\,b-$\\sigma$ relation for local ellipticals by Faber et al. (1989), as function of [OIII] equivalent width. The correlation indicates that the Mg\\,b index has likely been influenced by continuum contamination from AGN related light in the nuclear spectra.} \\end{figure}  The main aim of the project was to investigate whether young radio-loud AGN could be used to study the cosmological evolution of the fundamental plan for ellipticals, being selected independently of optical selection biases. Figure \\ref{funplane} shows the fundamental plane with the angle of projection in the $\\sigma$, $I_e$, $r_e$ - space chosen to be similar to that used in previous studies, with on the horizontal axis the logarithm of the effective radius, $r_e$ (in kpc), and on the vertical axis 1.2 times the logarithm of the velocity dispersion $\\sigma$ (in km sec$^{-1}$) minus 0.83 times the logarithm of the brightness level at the effective radius, $I_e$, with $I_e=10^{-0.4\\mu_e}$. The data are corrected for the cosmological evolution in mass-to-light ratio as found by van Dokkum et al. (1998), corresponding to $\\Delta$mag = 1.22$\\times z$. The solid squares show the galaxies in the sample, and the diamonds are the radio galaxies from Smith, Heckman \\& Illingworth (1990). The small circles are the local ellipticals from Faber et al. (1989) with heliocentric velocities $>$1500 km/sec, with the solid line their best linear fit. The galaxies in our sample follow the fundamental plane well, with an average offset of 0.01 and a dispersion of 0.17, similar to that of the Smith et al. sample, but with a slightly higher dispersion than the local galaxies in the Faber et al. sample (0.13).  To investigate whether evolution in M/L is visible over the small redshift range in our sample, the offsets of the datapoints to the local fundamental plane, $\\Delta$FP, as function of redshift is shown in figure \\ref{funevol}. The symbols are as in figure \\ref{funplane}, with the grey band indicating the dispersion in the local fundamental plane, and the dashed line showing the evolution as found by van Dokkum et al. (1998) for cluster ellipticals. The galaxies containing young radio sources located at z$>$0.2 seem to have on average a slightly positive offset from the local fundamental plane, with $\\Delta$FP=0.10$\\pm$0.06. Therefore evolution is only detected at a 1.7$\\sigma$ level in this small dataset. Clearly, more data are needed, preferably out to higher redshift, to confirm evolution, and to test the findings by van Dokkum et al.   \\subsection{The Mg\\,b - $\\sigma$ relation}  The absorption line strengths of a galaxy also give a good, independent insight in its stellar population. The Mg\\,b-$\\sigma$ relation, and its dependence on redshift, has been shown to be a powerful tool to study galaxy evolution (Ziegler \\& Bender 1997). The Mg\\,b-$\\sigma$ relation for the galaxies in our sample is shown in figure \\ref{sigmaMgb}. The small circles are the data from the local Faber et al. sample, where the best fit is indicated by the solid line. These datapoints were converted Mg$_2$ magnitudes, using the relation Mg\\,b\\AA$^{-1}$=15.0Mg$_2$ (Ziegler \\& Bender 1997). The Mg\\,b index of B0941-0805 is clearly lower than expected from the local relation. Since this object exhibit a strong emission line spectrum, we suspect that this Mg\\,b deficit is caused by a contamination of the continuum by AGN related light. In figure \\ref{OIII_Mgb}, the offsets of the Mg\\,b indices to the local Mg\\,b-$\\sigma$ relation is shown to be clearly correlated with OIII$_{5007}$ equivalent width, indicating that AGN contamination indeed plays a role. The Mg\\,b deficit in B0941-0805 can be explained by an AGN continuum contribution of 35\\%. For the other galaxies in the sample the AGN contribution is less than $<$5-10\\%. Note that this AGN contamination is for nuclear spectra, using a narrow slit. It is unlikely to have any effect on the determinations of the surface brightness profiles as done at and beyond $r_e$, and it is therefore improbable to affect the determination of the fundamental plane parameters. However, this AGN contribution, although generally small, makes investigating galaxy evolution using the Mg\\,b-$\\sigma$ relation not very useful for radio galaxies.  \\subsection{Black hole masses and their connection with radio power}  \\begin{figure*} \\psfig{figure=figure8.ps,width=16cm} \\caption{\\label{Mbhradio} The radio luminosity at 1.4 GHz versus black hole mass, as derived from the central velocity dispersion. The squares are those objects selected in the radio (from this paper and Smith et al. 1990). The open and closed circles are optically selected galaxies from the Faber et al. (1989) sample, with the former being point sources in the NVSS, and assumed to be radio-quiet, and the the latter being extended in the NVSS and assumed to be radio-loud AGN. The few diamonds at low black hole mass are galaxies added from the local group. The solid line is the $M_{bh}-P_{\\rm{radio}}$ relation as found by Franceschini et al. (1998).} \\end{figure*}  The work by Gebhardt et al. (2000) and Ferrarese \\& Merrit (2000) have shown that the mass of a central black hole is strongly correlated with the central velocity dispersion of its host galaxy. According to Gebhardt et al. the relation follows \\[ M_{\\rm{bh}}=1.2\\times10^8\\times \\left(\\frac{\\sigma_e}{200 \\ \\rm{km/s}}\\right)^{3.75} M_\\odot\\] which is found to have a lower dispersion than the correlation with absolute (bulge) magnitude. This $M_{bh}-\\sigma$ relation allows us to infer black hole masses from the central velocity dispersion measurements, for the galaxies in our sample, and those in Smith et al. and Faber et al. Uncertainties in the black holes masses are dominated by the scatter in the $M_{bh}-\\sigma$ relation, which is estimated to be a factor of $\\sim$2 (Gebhardt et al. 2000).  A variety of authors have discussed the relation between $M_{bh}$ and radio power. Long before the correlations between black hole mass and central velocity dispersion or galactic bulge mass were established, evidence has been provided of a relation between galactic bulge luminosity and/or velocity dispersion with radio luminosity $L_R$ (Heckman 1983, Nelson \\& Whittle 1996). Franceschini et al. (1998) showed that for a small sample of nearby galaxies, the non-thermal radio emission is strongly correlated with the black hole mass, from the dwarf elliptical M32 to the powerful radio galaxy M87, with a surprisingly small scatter, which was confirmed by McLure et al. (1999). More recently, other studies (eg. Laor 2000; Ho 2002) show that AGN are in general located above the overall $M_{bh}-L_{R}$ relation as found by Franceschini et al. Interestingly, Dunlop and McLure and collaborators (Dunlop et al. 2002; Dunlop \\& McLure 2002) have suggested that there is an $M_{bh}$-dependent lower and upper limit to the possible radio output of a black hole, offsetted from each other over several orders of magnitude.  To investigate these issues further, and in particular the relative position of radio galaxies on the $M_{bh}-L_{R}$ relation, we correlated the galaxies from Faber et al. with the NVSS-VLA 1.4 GHz survey (Condon et al. 1998) and determined their radio luminosities and sizes.  The $M_{bh}-L_{R}$ relation for the galaxies in the Faber et al. (circles), and the radio selected galaxies from Smith et al. and this paper (squares) are shown in figure \\ref{Mbhradio}. The galaxies in Faber et al. which exhibit extended radio structure, and therefore clearly identified as active galaxies, are indicated by solid circles. A few objects from the local group are added, and are indicated by diamonds.  To avoid the diagram being crowded by meaningless upperlimits, only those objects are plotted, if their radio flux would have been detectable in the NVSS if they radiate at a luminosity expected from the Franceschini relation. Although this produces a varying redshift cut-off with black hole mass, it does not bias the result towards particular radio luminosities. Clearly, the radio luminosities of the presumably inactive galaxies correlate well with black hole mass, although they seem to be slightly fainter than expected from the Franceschini relation, and have a larger scatter, with the radio power of  inactive galaxies at a given black hole mass ranging more than 2 orders of magnitude. The active galaxies (those selected in the radio, and those selected from the Faber et al. sample exhibiting extended NVSS emission), do not show any sign of a correlation with black hole mass. At a given $M_{bh}$ the radio luminosity can span over 7 orders of magnitude, making radio power unsuitable as a black hole mass estimator. We note that the black hole masses for radio galaxies as derived by Dunlop et al. (2002) are on average about a factor of 4 higher than given here (assuming $M_{bh}$=0.0013$M_{\\rm{bulge}}$). This is caused by their different method used to determine $M_{bh}$, based on the bulge luminosity $-$ bulge mass $-$ black hole mass relation: the average absolute R-band magnitude for the Smith et al. sample ($-23.71\\pm0.12$, assuming V-R=0.6) is the same as that for the radio galaxies in the Dunlop et al. sample ($-23.66\\pm0.16$), indicating that the properties of the host galaxies in the two samples are the same. Note that the differences between these methods result mainly in an uncertainty in the absolute values of $M_{bh}$, not in their relative values. All the radio galaxies are within the allowed range of radio power for their black hole masses as proposed by Dunlop et al. (2002).   \\subsection{Connecting the local populations of radio galaxies and inactive ellipticals}  Since the radio galaxies are selected virtually independently of their optical properties, and the optically selected galaxies independently of their radio properties, the samples can be used to connect the two populations. Figure \\ref{Mbhradio} shows that there is a clear difference between the $M_{bh}$ distribution for radio selected and optically selected AGN, with those selected in the radio, being biased towards lower black hole masses. We believe this is the natural consequence of the Faber et al. sample strongly favouring intrinsically more luminous galaxies, caused by its apparent magnitude limit. In contrast, the selection of objects on radio emission is practically volume based, with the parent population of potential radio-loud AGN containing relatively many more galaxies with lower velocity dispersions and black hole masses than the Faber et al. sample, which causes the sample of radio-selected AGNs to cover more galaxies with low $M_{bh}$.  \\begin{figure} \\psfig{figure=figure9.ps,width=8cm} \\caption{\\label{fraction} The fraction of radio-loud AGN as function of black hole mass, as derived from the optically selected sample of Faber et al. (1989). The distributions are explained in the text. The smooth curve is the function P(AGN|$M_{bh}$) used to determine the density and $M_{bh}$ distribution for radio-loud AGN with $P_{\\rm{1.4GHz} > 23.0}$ W/Hz.} \\end{figure}  \\begin{figure} \\psfig{figure=figure10.ps,width=8cm} \\caption{\\label{bh_dist}The distribution of black hole masses for the radio selected galaxies. The smooth curve (arbitrarily scaled) shows the predicted distribution of $M_{bh}$ derived from the local elliptical galaxy luminosity function and the fraction of black holes with powerful radio sources as function of $M_{bh}$ as derived from the optically selected sample.} \\end{figure}  These black hole mass distributions form the key to understand the connection between the populations of radio-loud and radio-quiet galaxies. First we determined the fraction of radio-loud AGN in the Faber et al. sample as a function of black hole mass. This is shown in figure \\ref{fraction}, with the solid and dotted lines indicating the fraction of galaxies with $P_{\\rm{1.4GHz}}$ $>23.0$ and $>24.0$ W/Hz respectively, the former decreasing from about 1/5 at $M_{bh}=10^9M_\\odot$ to $<1/20$ at $M_{bh}=10^{7.5}$. Note that a different subsample has been used for this than in figure \\ref{Mbhradio}, with such a distance limit that it can be determined from the NVSS whether the galaxy has a radio power above or below the cut-off. The dashed line shows the fraction of objects as function of $M_{bh}$ with a radio power more than 100 times that expected for the $M_{bh}-P_{\\rm{radio}}$ relation by Franceschini et al (1998). It shows that although the radio selected AGN do not have black holes with M$_{bh}<3\\times10^7M_{\\odot}$, black holes below this mass can still be active but only with a very low probability. Several objects with $M_{bh}\\sim10^7M_\\odot$  have been found with extended double-lobed low power radio sources in the Faber et al. sample.  With this estimate of the fraction of black holes producing powerful radio sources as a function of $M_{bh}$, the populations of radio-loud and radio-quiet galaxies can be linked. The absolute magnitude of a galaxy is connected to its central velocity dispersion by the Faber-Jackson relation (Faber \\& Jackson 1976), which in itself is related to the black hole mass via the $M_{bh}-\\sigma$ relation. In this way, the local distribution of black hole masses can be roughly deduced from the local galaxy luminosity function. The combination of the Faber-Jackson relation, which was re-determined for $\\sigma_e$ from the galaxies in the Faber et al. sample, and the $M_{bh}-\\sigma_e$ relation, results in a local black hole mass distribution of \\[ \\phi(M_{bh})dM_{bh} =0.65 \\frac{\\phi^\\star}{M^\\star_{bh}}\\rm{Exp} \\left( -\\frac{ M_{bh} }{M^\\star_{bh}} \\right)^{0.65}\\left(\\frac{M_{bh}}{M^\\star_{bh}}\\right)^{0.65(1+\\alpha)-1} \\hspace{-1.2cm} dM_{bh} \\] with $\\rm{Log} \\ M^{\\star}=-(8.818+M_B^\\star)/1.616$, where $M^\\star_B$, $\\alpha$, and $\\phi^\\star$ are the Schechter Function parameters of the local luminosity function in B band. Our estimate of the fraction of black holes producing powerful radio sources should now link this distribution of $M_{bh}$ to the local {\\it radio} luminosity function. For our calculation we assume that the fraction of radio-loud black holes with $P_{\\rm{1.4GHz}}>23.0$ W/Hz is $0.25\\times(1+(M_{bh}/1.5\\times10^8M_\\odot)^2)$, which is represented by the smooth curve in figure \\ref{fraction}. Using the Schechter parameters for the local luminosity function of {\\it elliptical} galaxies only, as derived from the 2dF Galaxy Redshift Survey (type 1 gals; Madgwick et al., 2002) we derive a local radio source density of $1.6\\times10^{-5}$ Mpc$^{-3}$ at Log P$_{\\rm{1.4GHz}}>23.0$ W/Hz. This corresponds very well to the density of radio sources above this luminosity cutoff as derived from the local radio luminosity function, which is $1.3\\times10^{-5}$ Mpc$^{-3}$ (eg. Dunlop \\& Peacock, 1990). In addition, the distribution of the black hole masses in the radio selected samples is shown in Figure \\ref{bh_dist}. Although these samples are not complete, we believe that they are roughly selected at random from the general population of radio sources with Log P$_{\\rm{1.4GHz}}>24.0$ W/Hz. The smooth curve represents the expected distribution of black hole masses from the calculations above, which is strikingly similar. These results confirm that elliptical galaxies must comprise the large majority of the radio-loud population of active galaxies."
    },
    "0209/astro-ph0209349_arXiv.txt": {
        "abstract": "We have observed three fields of the Coma cluster of galaxies with a narrow band (modified Str\\\"omgren) filter system.  Observed galaxies include 31 in the vicinity of NGC 4889, 48 near NGC 4874, and 60 near NGC 4839 complete to $M_{5500}=-18$ in all three subclusters.  Spectrophotometric classification finds all three subclusters of Coma to be dominated by red, E type (ellipticals/S0's) galaxies with a mean blue fraction, $f_B$, of 0.10.  The blue fraction increases to fainter luminosities, possible remnants of dwarf starburst population or the effects of dynamical friction removing bright, blue galaxies from the cluster population by mergers.  We find the color-magnitude (CM) relation to be well defined and linear over the range of $M_{5500} = -13$ to $-$22.  The observational error is lower than the true scatter around the CM relation indicating that galaxies achieve their final positions in the mass-metallicity plane by stochastic processes.  After calibration to multi-metallicity models, bright ellipticals are found to have luminosity weighted mean [Fe/H] values between $-$0.5 and $+$0.5, whereas low luminosity ellipticals have [Fe/H] values ranging from $-$2 to solar.  The lack of CM relation in our continuum color suggests that a systematic age effect cancels the metallicity effects in this bandpass.  This is confirmed with our age index ($\\Delta(bz-yz)$) which finds a weak correlation between luminosity and mean stellar age in ellipticals such that the stellar populations of bright ellipticals are 2 to 3 Gyrs younger than low luminosity ellipticals.  With respect to environmental effects, there is a slight decreasing metallicity gradient with respect to distance to each subcluster center, strongest around NGC 4874. Since NGC 4874 is the dynamic and x-ray center of the Coma cluster, this implies that environmental effects on low luminosity ellipticals are strongest at the cluster core compared to outlying subgroups. ",
        "introduction": "The age and star formation history of elliptical galaxies are, of course, key tests to our understanding of galaxy evolution processes.  Our early notions that ellipticals are old objects forming their stars at high redshift in a single episode has come into question by HST examination of nearby dwarf galaxies (see review by Grebel 1998).  Detailed CMD's indicate several episodes of star formation in dwarf galaxies, some within a few Gyrs of the present.  The detection of Balmer lines (Caldwell \\etal 1993) and the high fraction of blue galaxies in distant clusters (Rakos \\& Schombert 1995) both challenge the monolithic collapse scenario for ellipticals.  Observations of ellipticals from dwarf to giant in a range of environments is the first step in resolving various predictions from hierarchical models of galaxy formation.  In recent years, several spectroscopic programs have determined the age and metallicity of Coma and Fornax galaxies through the use of such indices as Mg$_2$ and H$\\beta$ (Trager \\etal 2000, Kuntschner \\etal 2001, Terlevich \\& Forbes 2001, Poggianti \\etal 2001).  Spectroscopic studies are, of course, superior to determining metallicity values directly from atomic lines.  While some comparison to models is required to interpret the spectral age indicators, there are several independent estimators of both age and metallicity.  An alternative approach is our narrow band photometry method, which focuses on the shape of a galaxy's integrated spectral energy distribution (SED) by way of near-UV and blue colors. This has the advantage of measuring the behavior of the stellar population as it reflects the temperature of the red giant branch and turnoff points, but has the disadvantage of being very model dependent since the red giant branch (RGB) must be broad even in ellipticals and requires some estimation of the luminosity weighted contribution from different metallicity populations (multi-metallicity models).  In our most recent paper using the generalized Str\\\"omgren filter system (Rakos \\etal 2001), the possibility was explored of using various color indices as age and metallicity estimators for ellipticals.  The integrated colors of Galactic globular clusters, with known ages and metallicities, were used to demonstrate and calibrate our narrow band system for single generation objects.  Multi-metallicity models were derived from the globular cluster (GC) calibration and compared to the colors of dwarf ellipticals in Fornax.  While we have been successful at predicting the global characteristics of ellipticals in clusters, such as slope of the color-metallicity relation, with our simple multi-metallicity models, our methods are by no means superior to spectroscopic survey but, in fact, serve as a complement to their programs.  In general, spectral investigations are superior in the detailed information they provide, but the goal here is to develop a system that can be used for low brightness objects or distant (i.e. faint) clusters where spectroscopy is very impractical.  In an attempt to confirm and extend the Fornax results, we have carried out photometry of three separate regions of the Abell cluster A1656 (Coma) as a first step in the investigation of the behavior of low luminosity ellipticals and dwarf galaxies in clusters of galaxies with increasing redshift.  We have selected Coma because it is the nearest rich cluster (richness class 2) with a dynamically evolved galaxy distribution (Bautz-Morgan type II).  Coma is one of the most studied clusters in the sky, and has long been regarded as the prototypical relaxed massive cluster of galaxies.  However, recent studies of Coma have suggested that the cluster is the product of a recent and an ongoing cluster-group merger. While the bright galaxies in Coma are well understood, the distribution and history of faint cluster galaxies has not yet been fully investigated.  In addition, there has been a renewal of interest in the galaxy luminosity function in rich clusters, in particular at its faint end. It is assumed that low-luminosity galaxies and their evolution almost certainly play an important cosmological role in explaining the large numbers of galaxies counted at faint magnitudes.  Interest in the faint end of the luminosity function derives from the fact that low luminosity ellipticals and dwarf galaxies serve as key tests to our understanding of galaxy formation and evolution. For example, in hierarchical models, galaxies are constructed from mergers with smaller-mass dwarfs.  The goal of this paper is threefold.  First, we will examine the cluster populations for the three regions in Coma around NGC 4889, 4874 and 4839 in terms of the spectrophotometric classifications.  Two recent papers (Rakos, Odell \\& Schombert 1997 and Rakos, Dominis \\& Steindling 2001) have shown that the ratio of the number of blue to red galaxies in clusters (A2317 and A2218 respectively) has a strong dependence on absolute magnitude, such that blue galaxies dominate at both the bright and faint end of the cluster luminosity function.  We will examine if the same effect occurs in Coma, a present-day counterpart to these intermediate redshift clusters.  Second, we will investigate the behavior of the color-magnitude (CM) relation for ellipticals.  In particular, we will map the CM relation into our metallicity and age calibrations. Lastly, we will examine the radial dependence of age and metallicity for early-type galaxies around the three dominate galaxies to check for environmental influences on the star formation history of cluster galaxies. ",
        "conclusions": "We can summarize our results in two parts, pure observational results and interpretation.  From the observational side, we have found that the Coma population is, unsurprisingly, rich in early-type galaxies based on their spectrophotometric classification (71\\%).  Morphologically, these systems are predominately ellipticals and S0's as one would expect in a dense cluster like Coma.  We summarize the analysis as the following:  \\begin{itemize}  \\item{} The fraction of blue galaxies rises slowly with decreasing galaxy luminosity, reflecting the dynamically evolved nature of the Coma system.  The blue fraction is lowest in those subclusters (NGC 4874 and NGC 4839) with cD envelopes attached to the primary galaxy.  It seems clear that the low blue fractions and the presence of a cD envelope are due to stripping effects, on one hand removing stars for the development of an expanded envelope while, on the other hand, halting star formation by gas stripping and lowering the number of blue, star-forming objects.  \\item{} What little blue population is found in Coma is unlike the blue population in distant clusters.  It is neither bright, disk galaxies nor faint starburst objects (i.e. there are no bright, blue galaxies in Coma nor are there faint S+ type systems as found in A115 and A2283, Rakos \\etal 2000).  \\item{} The color-magnitude relation for Coma (plus Fornax dE's) is extremely well defined for E type systems from $M_{5500} = -23$ to $-$13, although the scatter is larger than the observation error.  Our metallicity color, $vz-yz$, maps into mean [Fe/H] such that the brightest ellipticals have luminosity weighted mean [Fe/H] values of +0.3 while a typical low luminosity elliptical ($M_{5500}=-15$) will have a [Fe/H] value of $-$0.8.  \\item{} While our metallicity color ($vz-yz$) displays a strong CM relation, our continuum color ($bz-yz$) reveals no correlation with luminosity (i.e. stellar mass), although there is a bluer increase in scatter to fainter magnitudes.  This is unexpected since $bz-yz$ should be weakly sensitive to metallicity through changes in the temperature of turnoff stars and suggests a counteracting age effect.  \\item{} An age effect is confirmed with our mean age indicator, $\\Delta(bz-yz)$,  which finds a correlation with luminosity such that bright ellipticals are about 2 Gyrs younger than dwarf ellipticals.  This is a small and subtle difference, only detectable due to the wide range in luminosity for our combined Coma/Fornax sample.  Since spectroscopic studies are restricted to the top of the luminosity function, it is unsurprising that this effect has been missed in previous work.  This result is in agreement with the results from Terlevich \\& Forbes (2001) who also find that young galaxies (high mass) have high metallicities and that old galaxies (low mass) have a broad range of metallicities.  \\end{itemize}  Interpretation of the above observations is problematic.  A classic galactic wind model reproduces the CM relation by a simple relationship between the mass of a galaxy and its ability to retain gas by the depth of its gravitational well leading to longer, and more chemically evolved, star formation.  Observations of distant clusters strongly support the metallicity interpretation for the CM relation (Kodama \\& Arimoto 1997). The wind model also predicts a mean age difference since more massive galaxies have a longer episode of star formation, whereas dwarf systems are extinguished quickly producing a relative older mean age.  However, for traditional models (Bressan, Chiosi \\& Fagotto 1994), this age difference is extremely minor and the age effect for $bz-yz$ remains unresolved by a wind model.  It seems almost inescapable that the cluster environment plays an important role to the stellar populations of cluster ellipticals, especially since the most massive systems have evidence of a history of past mergers.  While the number of galaxies with blue populations is very low in present-day clusters, this was not the case in the recent past (i.e. Butcher-Oemler effect).  Mergers with gas-rich, star forming galaxies would have a minor effect on the metallicity color, yet would dramatically lower the mean stellar age as estimated by the $\\Delta(bz-yz)$ index. Thus, later mergers by galaxies such as spirals, with young stellar populations would explain the younger mean age of massive ellipticals, which is in agreement with the discovery of a large number of red merger systems in MS 1054-03 ($z=0.83$, van Dokkum \\etal 1999).  This also provides a cautionary tale for the study of CM relationships in distant clusters.  If an intermediate age population is visible in present-day clusters, then at moderate redshifts these same galaxies may not even be on the CM relation.  For example, the scenarios developed by Kauffmann (1996) indicate that some ellipticals could enter the red envelope by redshift of 0.4 to 0.5, even though most of the stars formed at redshift greater than 2.  Lastly, our results also favor the so-called hierarchical clustering and merging scenario of galaxy formation (Kauffmann, White \\& Guiderdoni 1993, Kauffmann \\& Charlot 1998) where the massive galaxies are assembled from older, lower mass systems.  Younger mean age and higher metallicities are achieved by star formation as massive ellipticals consume disk galaxies, and their cold gas, and efficiently turn this material into the bulk of their stellar populations.  While the scenario to convert low metallicity, gas-rich systems into present-day red, massive ellipticals seem finely tuned to produce the correct metallicities, the difference between present-day dwarf and giant elliptical ages (about 2 Gyrs, see Figure 5) does match the predictions of the hierarchical models (see Kauffmann \\& Charlot 1998 Figure 3).  However, the data only supports a conclusion that massive systems are younger, whether that is from assembly history or an extended star formation history is unclear.  The narrow range of metallicity for high mass galaxies argues for a uniform star formation history in high mass systems.  The range in metallicities for younger systems implies a more stochastic process where the weaker gravitational fields are more susceptible to environmental effects such as cluster ram pressure stripping."
    },
    "0209/astro-ph0209545_arXiv.txt": {
        "abstract": "The occultation of background stars by foreground Solar system objects, such as planets and asteroids, has been widely used as an observational probe to study physical properties associated with the foreground sample.  Similarly, the gravitational microlensing of background stellar sources by foreground mass concentrations has also been widely used to understand the foreground mass distribution. Though distinct, these two possibilities present two extreme cases during a transit; At the edge of the Solar system and beyond, the Kuiper belt and Oort cloud populations may present interesting foreground samples where combinations of occultation and lensing, and possibly both during the same transit, can be observed. To detect these events, wide-field monitoring campaigns with time sampling intervals of order tenths of seconds are required. For certain planetary occultation light curves, such as those involving Pluto, an accounting of the gravitational lensing effect may be necessary when deriving precise physical properties of the atmosphere through the associated refraction signal. ",
        "introduction": "The occultation of background stars by foreground objects has been widely used as an observational probe to study physical properties of various Solar system constituents such as planets, asteroids, comets, and rings (for an early review, see, \\cite{Ell79} 1979). In addition to basic physical properties, such as the radius of the foreground source that occulted the background star, the refraction of background stellar light by the foreground planetary atmosphere provides a well utilized probe to derive certain physical properties of the lower atmosphere (see, for example, the recent review by \\cite{EllOlk96}  1996 for further details).  On the other hand, at galactic distance scales well beyond the Solar system, foreground mass concentrations are expected to gravitationally microlens background stars (\\cite{Pac86} 1986). The microlensing of background sources by foreground objects is now well utilized to understand mass distributions in the galaxy, including potential dark matter candidates involving the so-called Massive Compact Halo Objects (MACHOs; \\cite{Gri91} 1991). A typical microlensing observational campaign now involve continuous monitoring of million or more stellar sources towards, say, the galactic bulge and the Magellanic clouds with time sampling intervals of order tens of minutes or more (e.g., \\cite{Alcetal93} 1993; \\cite{Udaetal93} 1993).  Though occultation, with a decrease in background source flux, and microlensing, with an increase in background source flux, have been mostly discussed as two separate phenomena, the two possibilities form essentially extreme cases during the transit of a foreground source across the surface of a projected background stellar surface. In general, though, one expects signatures of both occultation and gravitational lensing to be evident for a given population of foreground sources, the nature provides a simple reason why only these two extreme cases have been observed so far; the distance scale involved is such that known objects in the Solar system always occult background sources while foreground sources at galactic distances always microlens background stars. The transition between that of an occultation to a lensing event is rapid with only a limited range of parameters where both an occultation and a lensing signature will be visible during the same transit.  The favorable condition to observe both a combination of occultation and microlens in the same foreground sample involve a projected extent to the foreground object that is of the same order as the Einstein radius associated with gravitational lensing. In the case of a pure occultation, the projected foreground source radius is larger than the Einstein radius while the opposite is true for the observation of a gravitational  lensing effect. While transit events involving binary stars have been previously suggested as potential occurrences of both microlensing and occultation (\\cite{Mar01} 2001), favorable conditions may also be present with foreground sources in the Solar system, but at distance scales well beyond planets.  Here, we identify  outer Solar system populations, such as the Kuiper belt objects (KBO; \\cite{Kui51} 1951) and the Oort cloud objects (OCO; \\cite{Oor50} 1950) as interesting samples of foreground sources where both gravitational lensing and occultation, as well as a combination of the two, can be observed when they transit background stars. In fact, KBOs have been suggested as potential occulters in a previous study  where the use of transits was considered in detail to extract the small size members of this population (Roques \\& Moncuquet 2000). Here, we suggest that the consideration of KBOs as occulters may only apply to the currently observed KBO object population. If KBOs extend to much larger distances, as far as the Oort cloud, and contains massive members at large distances that currently probed, then there is some possibility that the distant members may in fact produce either a signature of lensing alone or a combination of lensing and occultation.  While no detailed data on the OCO population are observationally available, the currently cataloged KBO population is mostly at orbital distances between 40 and 50 AU with over $10^5$ objects of 100 km or more in size. The total estimated mass is of order 0.08 $M_{\\earth}$ (see, \\cite{LuuJew02} 2002 for a recent review). Extending current wide-field microlensing campaigns, which have been well executed to monitor millions of stars or more on a given night  (e.g., \\cite{Alcetal93} 1993; \\cite{Udaetal93} 1993), we suggest that small bodies of the outer Solar system can be detected and cataloged via similar continuous monitoring programs. Since the duration of transit events involving outer Solar system objects are of order a minute and less, sampling time intervals, however, must be at the level of few tenths of seconds instead of usual tens of minutes or more  time scales currently used in galactic microlensing campaigns. Such high sampling rates, while keeping the same flux threshold levels as current surveys, are within reach with the advent of dedicated large area telescopes and continuous improvements in the instrumental front. The combined occultation and lensing measurements allow the observational data on KBO and OCO samples to be significantly extended since one is no longer sensitive to individual fluxes, as in direct imaging observations, but rather on the ability to detect and extract transit events during the monitoring of a large sample of background sources.  The discussion is organized as following. In the next section, we introduce the concept of occultation  and microlensing as two extreme cases of the same transit event. We discuss potential signature of KBOs when transiting across a background stellar surface. A detailed study of the statistical signature of occultation due to KBO population is presented in Roques \\& Moncuquet (2000) to which we refer the reader to further details. Similarly, the paper by Agol (2002) considers the signature of an occultation and lensing during transit events that involve binary sources orbiting each other. A prior discussion on the occultation signature in a microlensing light curve, as applied to galactic lensing surveys, is available in \\cite{Bro96} (1996). Here, we consider the application to the Solar system as a potential way to extend our understanding of outer members which may have avoided direct detection due to low flux levels. ",
        "conclusions": "We have explored the role of gravitational lensing when outer Solar system objects, mainly members of the Kuiper belt and the Oort cloud, transit background stars. The known population of KBOs, at distances of order 40 AU with sizes of order few hundred kilometers and less, will always occult background stars. Their occultation signature can be detected in monitoring campaigns involving few-meter class telescopes with time sampling intervals of order few tenths of seconds (\\cite{RoqMon00} 2000). While we have ignored here due to our interest in large size objects, as discussed in \\cite{RoqMon00} (2000), the detection rate of few kilometer size and below KBOs is partly enhanced by the diffraction effect that appears during the occultation.  At distances much large than the currently known KBO population, the potential observability of a microlensing event significantly increases. At distances of few 10,000 AU and more, corresponding to the Oort cloud, objects with sizes of order hundreds of kilometers or more will gravitationally microlens background stars instead of simply occulting them. For certain foreground object sizes, or mass, at favorable distances, one can potentially observe a combination of an occultation and a lensing event during the same transit. The detection of an onset of a lensing event on an occultation light curve is interesting since the point at which the lensing signature enters allows one to measure the ratio of the foreground source radius to its Einstein radius accurately (\\cite{Bro96} 1996). This additional information will aid in constraining  physical parameters of the foreground population beyond what is available solely from the occultation  or the microlensing light curve.  Note that the estimated KBO optical depth near the ecliptic is of order 10$^{-6}$  (\\cite{RoqMon00} 2000); this is at the same level as the microlensing optical depth towards the galactic bulge. The Oort cloud optical depth, however, is highly uncertain due to our limited knowledge on various properties of its population. The dynamical constraints, based on orbits of long-period comets, suggest a total population of $\\sim 10^{12}$ with a total mass of 38 $M_{\\earth}$ (\\cite{Wei96} 1996); these estimates are clearly uncertain for obvious reasons. While the total mass is higher than that associated with KBOs, for the observable transit optical depth, what is required is the distribution of source sizes. If sizes are all equal, then with a mass of order $\\sim$ 10$^{14}$ g, and radii of order few tenths of kilometers, Oort cloud will remain undetectable with observations that attempt to detect transists.  The detectable transits, occultations and/or lensing, however, require the presence of objects with masses of order $\\sim$ 10$^{22}$ g or with radii of order few hundred kilometers. Note that certain constraints on the KBO population limits the size distribution  of the outer  KBOs, at distances between 50 and 70 AU, to be below few hundred kilometers (\\cite{Alletal01} 2001). Such surveys, however, are not sensitive to even massive objects with radii of thousands kilometers  at distances corresponding to Oort cloud suggesting that instead of direct detection techniques, such as through imaging data, indirect techniques such as transit signatures will be needed to constrain its population. Note that the monitoring of transits involving both KBOs and OCOs can be concurrently considered except that the detection of events in monitoring data should involve the search for both a flux decrement as well as a possible increment due to a lensing event. Note that any increment due to lensing can easily be ascribed to massive bodies at large distances such that one breaks the usual degeneracy one encounters in galactic microlensing studies involving the mass and distance of the object. On the other hand, even if no lensing events are detected, any reliable upper limit on the lensing optical depth towards the Oort cloud can be used to constrain the massive end of its population and will aid in understanding the role OCOs play in the formation and evolution of the Solar system.  As we have discussed, the role of lensing on the occultation light curves of outer Solar system bodies is likely to be only limited to distant Oort cloud members. The gravitational lensing effect, however, may already be important for objects in the inner Solar system. For Pluto, at a distance of $\\sim$ 39.5 AU and a mass of 0.002 $M_{\\earth}$, the Einstein radius is of order $\\sim$ 15 km. This is small when compared to the Pluto radius of order 1200 km and leads to the naive conclusion that any effects related to lensing by Pluto can be ignored when interpreting its data. For precision calculations and parameter estimations, however, there may be an additional consequence associated with lensing. While the magnification signature may not be dominant, gravitational lensing also induce variations in astrometry, mainly a relative change in the image position with respect to the unlensed position. When interpreting light curves to derive atmospheric parameters, as was done in \\cite{EllYou92} (1992), it may  be necessary to account for the shift in image position due to lensing along with variations arising from the atmospheric refraction effect.  If not accounted properly, one will wrongly conclude the depth to which the light curve probes the Pluto's atmosphere with an error that is of the same order as the size of the Einstein radius.  The best published data on an occultation by Pluto comes from the 9th June 1998 event involving the background star P8 (\\cite{Miletal93} 1993). At the inner most depths probed by refraction, the light curve associated with this event showed an anomalous gradient beyond a simple refractive atmosphere. This gradient has been modeled either as extinction due to a haze layer or due to an abrupt thermal gradient. A preliminary calculation of the astrometric lensing correction to the light curve depth indicated that the abrupt change in the light curve was unlikely due to gravitational lensing modifications. Recently, it was reported that several new light curves related to an occultation by Pluto has now been obtained. It'll be an interesting exercise to see if these data require an accounting of the astrometric shift in the background source image during the inner depths of the occultation due to gravitational lensing.  To summarize, minor bodies of the outer Solar system will always occult background stellar sources. There is still some possibility that a distance sample of objects, such as members of the Oort cloud, will microlens background stars. Another possibility is that there will be a combined signature of an occultation and a lensing event during the same transit. These events, and occultation and lensing only events as well, can be extracted from continuous monitoring campaigns similar to those that are currently pursued to detect galactic microlensing towards the bulge and the Magellenic clouds. The data sampling  intervals of a Solar system targeted campaign, however, should be at the order of few tenths of seconds and is within reach experimentally in the near future.  \\smallskip {\\it Acknowledgments:} This research was supported at Caltech by a senior research fellowship from the Sherman Fairchild foundation and additional support from the Department of Energy.  The author thanks Marc Kamionkowski for encouraging the author to work on topics beyond cosmology."
    },
    "0209/astro-ph0209259_arXiv.txt": {
        "abstract": "{We have detected 58 Wolf-Rayet candidates in the central region of the nearby spiral galaxy NGC~300, based on deep VLT-FORS2 narrow-band imaging. Our survey is close to complete except for heavily reddened WR stars. Of the objects in our list, 16 stars were already spectroscopically confirmed as WR stars by Schild \\& Testor and Breysacher et al., to which 4 stars are added using low resolution FORS2 datasets. The WR population of NGC~300 now totals 60, a threefold increase over previous surveys, with WC/WN$\\geq$1/3, in reasonable agreement with Local Group galaxies for a moderately sub-solar metallicity. We also discuss the WR surface density in the central region of NGC~300. Finally, analyses are presented for two apparently single WC stars -- \\#29 (alias WR3, WC5) and \\#48 (alias WR13, WC4) located close to the nucleus, and at a deprojected radius of 2.5~kpc, respectively. These are among the first models of WR stars in galaxies beyond the Local  Group, and are compared with early WC stars in our Galaxy and LMC. ",
        "introduction": "Over 500 Wolf-Rayet stars have been identified in Local Group galaxies, principally the Milky Way, M31 and M33. These stars beautifully trace young stellar populations, and their number and distribution reacts sensitively to metallicity, which varies by an order of magnitude from the Small Magellanic Cloud (SMC) to M31. Detailed studies of individual WR stars in Local Group stars have been carried out (e.g. Smartt et al. 2001; Crowther 2000; Crowther et al. 2002) using 2--4m class telescopes.  The availability of 8--10m class telescopes permits the discovery and study of individual stars at greater distances, spanning a greater range of metallicities. As a first application, we present here VLT imaging and spectroscopy of WR stars in NGC~300, located in the Sculptor group at a distance of 2 Mpc (Freedman et al. 2001). It's metallicity is bracketed by the Milky Way and Large Magellanic Cloud (LMC) and therefore we expect a similarly large number of WR stars in NGC~300. Previous surveys have however failed to identify them. A large population might also be anticipated since NGC~300 is a late type spiral, reminiscent of M33, which harbours at least 140 WR stars (Massey \\& Johnson 1998). Because of its low inclination NGC~300 is well suited to studies of its stellar content -- recent surveys include blue supergiants (Bresolin et al. 2002a), Cepheids (Pietrzy\\'nski et al. 2002b), OB associations (Pietrzy\\'nski et al. 2001) and Supernova remnants (Pannuti et al. 2001).  The first signature of WR stars in NGC\\,300 was found in spectra of H\\,{\\sc ii} regions by D'Odorico et al. (1983). They detected the broad WR feature in two out of sixteen H\\,{\\sc ii} regions. Five years later, Deharveng et al. (1988) presented a catalogue of 176 H\\,{\\sc ii} regions and found broad WR emission in four of them. Although this clearly demonstrated that WR detection was feasible at this distance, the spectroscopy could not in itself locate the individual WR stars. To achieve this, imaging in narrow band filters was necessary. First results with this technique were reported by Schild \\& Testor (1991, 1992) and Testor \\& Schild (1993). They found in total 13 WR candidates and confirmed them spectroscopically. Six additional WR stars were later identified in the same way by Breysacher et al. (1997) in stellar associations. One additional weak--lined late WN star was serendipitously found by Bresolin et al. (2002ab).    \\begin{figure*} \\mbox{\\psfig{figure=schild1a.eps,width=8.5cm,height=8.5cm,clip=}} \\hskip 0.8cm \\mbox{\\psfig{figure=schild1b.eps,width=8.5cm,height=8.5cm,clip=}}  \\caption[]{Finding chart for WR stars/candidates in NGC~300. Left: nuclear region, right: area northwest of the nucleus. The horizontal bar represents 10\\,\\arcsec. North to the top, East to the left.} \\label{nucnw.fig} \\end{figure*}   In total, there are presently 22 confirmed WR stars in NGC~300, strongly skewed towards WC subtypes. From a census of the WR distribution  in Local Group galaxies, Massey (1996) identified a rather tight correlation between the WC/WN ratio and metallicity, as characterized by oxygen content. Although there have not been any recent studies of the NGC~300 metallicity gradient, Deharveng et al. (1988) used data from Pagel et al. (1979) and  Webster \\& Smith (1983), to imply a range between log(O/H)+12=8.9 in its nucleus and  8.3 in its outer spiral arms. Similar conclusions were obtained by Zaritsky et al. (1994) from a recalibration of previous results. One would expect a WC/WN ratio of $\\sim$\\,1/2 from comparison with Local Group galaxies, yet the census of WR stars in NGC~300 indicates WC/WN\\,$\\sim$\\,2. Consequently, we might expect that the WR population of NGC~300 is highly incomplete, particularly amongst WN stars.  In this paper we  present results from a new imaging survey of the central region of NGC\\,300 with the Very Large Telescope (VLT). New WR candidates are identified, some of which are spectroscopically confirmed. Spectral types of the latter are discussed, with particular reference to the WC/WN ratio of the inner galaxy. An analysis of two apparently single WC stars is presented, one located close to its nucleus, the other at $\\sim$\\,50\\,\\% of the Holmberg radius, $\\rho_0$. Comparisons are made with recent comparable studies of WC stars in a variety of metallicity environments. ",
        "conclusions": "We have demonstrated the feasibility of using narrow-band filters to detect Wolf-Rayet candidates in NGC~300. Restricting our survey to the central 6.8\\,\\arcmin$\\times$6.8\\,\\arcmin\\ region, we have trebled the known WR content from 20 to 58 stars, within a factor of two of the global content of its northern Local Group counterpart, M33. Surveys of the outer spiral arms of NGC~300 are sought in order to determine its total WR content, which probably approaches $\\sim$\\,100, as is spectroscopic confirmation of remaining candidates. The WC/WN ratio of the central region of NGC~300 has been revised from $\\sim$\\,2 to $\\sim$\\,1/3, in reasonable accord with Local Group galaxies spanning a similar metallicity range. Modern  abundance analyses of H\\,{\\sc ii} regions and/or AB supergiants are urgently required to verify previous determinations of the metallicity gradient of NGC~300. We have purposefully not discussed the WR/O ratio in NGC~300, since it is extremely difficult to constrain  this ratio observationally,  as discussed by Massey (2003).  Using VLT-FORS2, 600 sec imaging provides over 90\\,\\% completeness for WR candidates down to an excess of 0.1\\,mag at a distance of 2\\,Mpc. The number of known WR stars can therefore be rapidly increased with a moderate investment of observing time.  This approach greatly improves our chance of witnessing a WR star undergoing a  supernova explosion in the nearby universe. Imaging surveys towards such goals are presently underway (e.g. Smartt et al.  2002), albeit based solely around broad-band filters, such that WR candidates can not easily be identified. Under favourable conditions one can reasonably expect to extend our imaging/spectroscopic  approach to WR stars within galaxies  at distances of up to at least $\\sim$\\,5\\,Mpc. Recession radial velocities, shifting WR emission lines redward of the $\\lambda$4684 filter, only become problematic in excess of 1000 km\\,s$^{-1}$,  typically corresponding to distances in excess of $\\geq$10\\,Mpc. We have recently obtained VLT-FORS2 narrow-band imaging of M83 (NGC~5236), at distance of 3.2\\,Mpc\\footnote{We assume that its distance is comparable to NGC~5253} (Freedman et al. 2001). M83 has a metallicity of up to  5\\,$Z_{\\odot}$ (Webster \\& Smith 1983), such that a large WR population is expected. Indeed, WR signatures are already known from H\\,{\\sc ii}  region studies (Bresolin \\& Kennicutt 2002).  Once the census of WR stars is reasonably complete in a galaxy or part thereof, we can obtain surface density plots. We present here the WR star distribution in the central 2\\,kpc of NGC~300. We find that the very centre of the galaxy is apparently void of WR stars, in contrast with our own Galaxy, but that a maximum of the surface density occurs at a galacto-centric distance of about 0.4\\,kpc. At 1\\,kpc the surface density drops to a minimum, beyond which it steadily increases to about half the value of the 0.5\\,kpc ring. NGC~300 compares favourably with most Local Group galaxies in its WR surface density, with the exception of IC10, and perhaps M33 (Massey \\& Johnson 1998).  We have also illustrated that single, early-type WC stars with $v\\,\\sim$\\,23 mag can be quantitatively studied using modest integration times  with VLT-FORS2. More problematic is the challenge of obtaining uncontaminated WR spectroscopy at such large distances, since a slit width of 1\\,\\arcsec\\ corresponds to a spatial scale of $\\sim$\\,10\\,pc at 2\\,Mpc. Isolated WR stars are present, although they are in the minority and will be even more problematic for still more distant galaxies. An order-of-magnitude reduction in slit size is ultimately required using ground-based telescopes, without  a corresponding loss of throughput."
    },
    "0209/astro-ph0209403_arXiv.txt": {
        "abstract": "Physical science has changed in the century since Lord Kelvin's celebrated essay on {\\it Nineteenth Century Clouds over the Dynamical Theory of Heat and Light}, but some things are the same. Analogs in what was happening in physics then and what is happening in astronomy today serve to remind us why we can be confident the Virtual Observatory of the twenty-first century will have a rich list of challenges to explore. ",
        "introduction": "Astronomy has enjoyed a very good century. Have the basic problems now been solved, leaving for the astronomers of the 21$^{\\rm st}$ century the task of working out the pesky details? The question is little discussed -- astronomers are too busy with ongoing research -- but worth considering from time to time. I shall argue that we have a useful guide to the long-term prospects for research in astronomy from analogs to the present situation in was happening in physics 100 years ago. In both cases there is a basis of fundamental concepts that are strikingly successful, apart from some stubborn clouds, or, as we would now say, challenges for research. The clouds over electromagnetism and thermal physics at the start of the 20$^{\\rm th}$ century foreshadowed relativity and quantum physics. We can't say what will be learned from the clouds over present-day astronomy -- I shall mention aspects of the dark sector, strong space curvature, and the meaning of life --  but we can be sure they will continue to drive difficult but fascinating research in astronomy for quite some time to come. ",
        "conclusions": "Our ability to explore the physical universe is limited by resources and intellectual energy: the scientific enterprise must eventually reach completion by exhaustion. But we can be sure this will not happen any time soon to astronomy and its Virtual Observatory, because the subject has a rich list of Rowland-type problems to address, and, as I have discussed, a key role to play in the exploration of clear and present Kelvin-level gaps in our understanding of the fundamental basis for physical science. There was no guarantee in 1900 that the clouds over physics would clear, with a wonderful expansion of our knowledge. It would be foolish to try to guess what the present clouds might foreshadow, but we can list the general possibilities. Maybe the clouds will resist all efforts at resolution. If so, convincing people of this certainly will generate a lot of work for astronomers. Maybe the clouds will be cleared and at last leave astronomers to tidy up the pesky details. Or maybe clearing the clouds will reveal a new set, as has happened before.  I have avoided until now commenting on a serious issue under debate in the astronomy community: is this an appropriate time to commit limited resources to an International Virtual Observatory? I respect the arguments against, but am persuaded by personal experience that the growth of the Virtual Observatory is inevitable and would benefit from intelligent design. Two years ago the walls of my office were covered by about 25 meters of journal rows, dating back to 1965. I loved the convenience of reaching for a copy of the wanted article. But I've discarded the journals; I love even more the much greater convenience and power of ADS, arXiv, and JSTOR. I notice many colleagues feel the same: we have become addicted to these Virtual Libraries. Present-day Virtual Observatories are a useful but limited counterpart. Their further development seems to me to be an inevitable part of what we see happening around us, and surely calls for the proactive community response I have observed at this meeting."
    },
    "0209/astro-ph0209129_arXiv.txt": {
        "abstract": "{We present results derived from VLT--FORS2 spectra of 24 different globular clusters associated with the lenticular galaxy NGC\\,3115. A subsample of 17 globular clusters have sufficiently high signal--to--noise to allow precision measurements of absorption line-strengths. Comparing these indices to new stellar population models by Thomas et al. we determine ages, metallicities and element abundance ratios. For the first time these stellar population models explicitly take abundance ratio biases in the Lick/IDS stellar library into account. Our data are also compared with the Lick/IDS observations of Milky Way and M\\,31 globular clusters. Unpublished higher order Balmer lines (H$\\gamma_{A,F}$ and H$\\delta_{A,F}$) from the Lick/IDS observations are given in the Appendix.  Our best age estimates show that the observed clusters which sample the bimodal colour distribution of NGC\\,3115 are coeval within our observational errors (2--3 Gyr). Our best calibrated age/metallicity diagnostic diagram (\\hb\\/ {\\em vs}\\/ [MgFe]) indicates an absolute age of 11--12 Gyr consistent with the luminosity weighted age for the central part of NGC\\,3115. We confirm with our accurate line-strength measurements that the $(V-I)$ colour is a good metallicity indicator within the probed metallicity range ($-1.5 < \\mathrm{[Fe/H]} < 0.0$).  The abundance ratios for globular clusters in NGC\\,3115 give an inhomogeneous picture. We find a range from solar to super-solar ratios for both blue and red clusters. This is similar to the data for M\\,31 while the Milky Way seems to harbour clusters which are mainly consistent with $[\\alpha / \\mathrm{Fe}] \\simeq 0.3$. From our accurate recession velocities we detect, independent of metallicity, clear rotation in the sample of globular clusters.  In order to explain the metallicity and abundance ratio pattern, particularly the range in abundance ratios for the metal rich globular clusters in NGC\\,3115, we favour a formation picture with more than two distinct formation episodes. ",
        "introduction": "The analysis of globular cluster systems in external galaxies is starting to fulfil its long-held promise as a probe of the formation of galaxies. Globular clusters (hereafter GCs) are a fossil record of this formation process, and provide one of the best tools with which to investigate the chemical enrichment and star formation history in the initial stages of galaxy formation \\citep[e.g.,][]{ash97}. Much recent interest has focused on the globular cluster systems of luminous elliptical galaxies where the combination of metallicities and kinematics can be used to distinguish between variants of the popular monolithic collapse and merger models for galaxy formation \\citep[e.g.,][]{forb97,sharp98}.  Lenticular galaxies hold a key position in the Hubble sequence of morphological types, intermediate between pure spheroidal systems like luminous ellipticals and disk-dominated spiral galaxies.  Their formation mechanism is still the subject of considerable debate with evidence both for \\citep{dres97} and against \\citep{dres80} their evolution from star-forming spirals via processes of gas stripping and exhaustion. A key question is when and how did such processes occur for S0 galaxies in a wide range of environments from rich clusters to the field. The globular cluster systems of S0 galaxies can provide independent constraints on when the major star formation episodes occurred both in the disk and halo. However, thus far they have been little studied with only NGC\\,1380 \\citep{kis97} and NGC\\,4594 \\citep[the Sombrero galaxy,][]{bri97} having received any detailed attention and only with photometric methods.  NGC\\,3115 is one of the nearest S0 galaxies ($9.7\\pm0.4$~Mpc, $M_B=-20.1$; \\citealp{ton01}) and is located in the sparse low-density environment of the Leo Spur. As such it provides an ideal test case for studying the formation mechanism of field S0's. A significant globular cluster system containing $\\sim$500 clusters was first detected by \\cite{han86} using photographic plates. The nature of the cluster system and its origin were recently thrown into question with the discovery by \\cite{els97} that the red-giant stars in the NGC\\,3115 halo $\\sim$40~kpc from the centre showed a {\\em bimodal}\\/ colour distribution. The inferred presence of two distinct halo populations of roughly equal size at metallicities of $\\mathrm{[Fe/H]} \\simeq -0.7$ and $\\mathrm{[Fe/H]} \\simeq -1.3$ suggests at least two distinct epochs of formation.  The $(V-I)$ colour distribution of the NGC\\,3115 globular cluster system has been the target of two recent independent studies using HST \\citep{kundu98} and CFHT \\citep{kav98b} data. Both studies find bimodality in the colour distributions of the GCs, with mean metallicities at $\\mathrm{[Fe/H]} \\simeq -0.37$ and $\\mathrm{[Fe/H]} \\simeq -1.36$ suggesting that the cluster and halo star systems may have formed coevally. This suggestion has gained further support from an investigation by \\citet{puz02a} who employed optical--IR colours to probe the globular cluster population close to the centre of NGC\\,3115. They also find two peaks in metallicity and an average age around $\\approx$10~Gyr. However, their age discrimination power is very limited for metallicities lower than ${\\rm [Fe/H]} = -0.4$.  One scenario in which the above observations could be understood is if the metal-poor component corresponds to a primordial $\\simeq 13$~Gyr old population, whilst the metal-rich component formed a few Gyr later from enriched gas, possibly as the result of a minor merger \\citep[e.g.,][]{bekk98}. With only broad-band colours available, however, the well-known degeneracy between metallicity and age \\citep{wor94} makes such conclusions very uncertain.  For that reason, we have started a campaign to spectroscopically study the globular cluster system in NGC\\,3115. Our precision measurements of absorption line-strengths can be used to derive age and metallicity estimates directly from the comparison with new stellar population models. Unlike photometric methods, with spectroscopy we are also able to explore element abundance ratios for the GCs. We compare our results with other very recently obtained spectroscopic samples of GCs in early-type galaxies: in the giant Fornax elliptical NGC\\,1399 \\citep{for01}, the Sa/Sb galaxy M\\,81 \\citep{schro02}, the SB0 galaxy NGC\\,1023 \\citep{lar02a} and the Sombrero galaxy NGC\\,4594 \\citep{lar02b}.  The paper is organized as follows. In Section~\\ref{sec:obs}, the observations and their reduction are discussed. Section~\\ref{sec:colour} presents the colour distribution of our sample while in Section~\\ref{sec:abrat} the treatment of abundance ratios and new stellar population models are investigated. Our results on abundance ratios, age and metallicity distributions for GCs in NGC\\,3115, the Milky Way (hereafter MW) and M\\,31 are presented in Section~\\ref{sec:results} with a general discussion in Section~\\ref{sec:discussion}. We present our conclusions in Section~\\ref{sec:conclusions}. ",
        "conclusions": "\\label{sec:conclusions} We present new, accurate measurements of absorption line-strength indices of 17 globular clusters (hereafter GCs) in the nearby S0 galaxy NGC\\,3115. Our objects span a range in colour so that the bimodal $(V-I)$ colour distribution is well sampled.  A critical comparison with Lick/IDS data \\citep{tra98} of GCs in M\\,31 and the Milky Way (hereafter MW) is presented. The Lick/IDS measurements of the H$\\gamma_{A,F}$ and H$\\delta_{A,F}$ indices for MW and M\\,31 GCs are presented for the first time in this paper. The data are analysed with new stellar population models \\citep{TMB02b} which are able to predict line-strength not only as a function of age and metallicity but also as a function of abundance ratio.  Specifically, abundance ratio biases in the stellar library, which is an essential ingredient to the model predictions, have been taken into account for the first time.  Our main results are listed in the following:  \\begin{itemize}  \\item The GCs in NGC\\,3115 show a range of abundance ratios as estimated by the strength of Mg and Fe lines. Specifically we find for both red and blue clusters solar as well as super-solar values (up to $\\mathrm{[Mg/Fe]} \\simeq +0.5$). Lick/IDS data of M\\,31 GCs show a similar distribution while MW GCs are consistent with a constant value of $\\mathrm{[Mg/Fe]} \\simeq 0.3$.  The latter is in agreement with recent studies of the resolved stellar populations in MW GCs.  \\item Our analysis of \\hb, \\hgf\\/ and \\hdf\\/ {\\em vs}\\/ [MgFe] age/metallicity diagnostic diagrams shows that the red and blue GC populations are coeval within $\\approx$2~Gyr. Our best calibrated diagram (\\hb, [MgFe]) indicates a mean age of $11-12$~Gyr. However, the higher order Balmer lines, although confirming the similar ages of blue and red clusters, indicate younger absolute ages ($\\approx$6~Gyr). We ascribe these younger age estimates to inaccurate model calibrations of the higher order Balmer lines. Evidence for the existence of young ($<5$~Gyr) GCs in the studied galaxies is scarce. Only one cluster in NGC\\,3115 and perhaps $\\sim$3 clusters in M\\,31 show a combination of Balmer and metal absorption strength which is consistent with such young ages. The strong \\hb\\/ absorption can be alternatively explained if these clusters have old stellar populations with a populous extended horizontal branch.  \\item We present a comparison of photometric and spectroscopic metallicity determinations and find a good linear relation in the metallicity range probed by our sample of NGC\\,3115 clusters ($-1.5 < \\mathrm{[Fe/H]} < 0.0$). The photometric estimates are systematically lower ($\\simeq -0.26$) in comparison with our spectroscopic measurements. We note that our observations clearly show that each colour peak has a significant spread in metallicity rather than being consistent with a narrow distribution.  \\item The existence of solar as well as elevated Mg-to-Fe ratios at a given metallicity for GCs in NGC\\,3115 indicates that a simple scenario of two distinct star-formation episodes is not sufficient to explain the formation of this galaxy. Probably a realistic model needs to incorporate more than two distinct star-formation events.  \\item We detect a clear signal of rotation in our sample of GCs independent of their metallicities.  \\end{itemize}  Larger samples of high signal-to-noise spectra of GCs in nearby galaxies are required to increase our knowledge of how these galaxies formed. Particularly the measurement of abundance ratios in extragalactic GCs is now possible and will deliver important new constraints for galaxy formation models which are not accessible with broad-band colours alone."
    },
    "0209/astro-ph0209479_arXiv.txt": {
        "abstract": "Using photometry and spectroscopy of 144,609 galaxies from the Sloan Digital Sky Survey, we present bivariate distributions of pairs of seven galaxy properties: four optical colors, surface brightness, radial profile shape as measured by the \\Sersic\\ index, and absolute magnitude. In addition, we present the dependence of local galaxy density (smoothed on 8 $h^{-1}$ Mpc scales) on all of these properties.  Several classic, well-known relations among galaxy properties are evident at extremely high signal-to-noise ratio: the color--color relations of galaxies, the color--magnitude relations, the magnitude--surface brightness relation, and the dependence of density on color and absolute magnitude. We show that most of the $i$-band luminosity density in the universe is in the absolute magnitude and surface brightness ranges used: $-23.5 < M_{\\band{0.1}{i}} < -17.0~\\mathrm{mag}$ and $17 < \\mu_{\\band{0.1}{i}} < 24~\\mathrm{mag}$ in $1~\\mathrm{arcsec^{2}}$ (the notation $\\band{z}{b}$ represents the $b$ band shifted blueward by a factor $(1+z)$). Some of the relationships between parameters, in particular the color--magnitude relations, show stronger correlations for exponential galaxies and concentrated galaxies taken separately than for all galaxies taken together. We provide a simple set of fits of the dependence of galaxy properties on luminosity for these two sets of galaxies. ",
        "introduction": "\\label{motivation}  There are strong correlations among the measurable physical properties of galaxies.  The classification of galaxies along the visual morphological sequence described by \\cite{hubble36a} correlates well with the dominance of their central bulge, their surface brightnesses, and their colors. These properties also correlate with other properties, such as metallicity, emission-line strength, luminosity in visual bands, neutral gas content, and the winding angle of the spiral structure (for a review, see \\citealt{roberts94a}). The surface brightnesses of giant galaxies classified morphologically as elliptical are known to be strongly correlated with their sizes (\\citealt{kormendy77a}). Galaxy colors (at least of morphologically elliptical galaxies) are known to be strongly correlated with galaxy luminosity (\\citealt{baum59,faber73a,visvanathan77,terlevich01a}). The gravitational mass of a galaxy is closely related to the luminosity and other galaxy properties. These galaxy relations manifest themselves in the Tully-Fisher relation for spiral galaxies (\\citealt{tully77a,burstein97a}) and the Fundamental Plane for morphologically elliptical galaxies (\\citealt{faber76a,djorgovski87,dressler87,burstein97a}); \\citet{strauss95a} review these dynamical relations. Furthermore, the environment of a galaxy is related to its type, in the sense that early-type galaxies are found in denser regions than late-type galaxies, as first noted by \\citet{hubble36a} and as found by numerous subsequent investigators (\\citealt{oemler74a,dressler80a,davis76a, giovanelli86a,santiago92a,guzzo97a,hashimoto99a,blanton00a}). In short, different physical properties of galaxies are closely related to each other.  In order to understand theoretically how galaxies formed and acquired their current properties, it is first necessary to characterize the observed distribution of galaxy properties in a comprehensive manner. Such a characterization is a primary goal of the Sloan Digital Sky Survey (SDSS; \\citealt{york00a}).  The SDSS has already created the largest sample to date (144,609 galaxies) of luminous galaxies with well-measured photometric and spectroscopic properties. As a first step in understanding the joint distribution of galaxy properties, we present here a study of the number density and luminosity density distributions of galaxy colors, profiles, luminosities, surface brightnesses, and local densities. Our purpose is to measure properties related to quantities that cosmological gasdynamical simulations (\\citealt{nagamine01a}; \\citealt{pearce01a}; \\citealt{steinmetz02a}) or semi-analytic models (\\citealt{somerville01a}; \\citealt{mathis02a}) can predict or will soon be able to predict, such as galaxy ages, sizes, stellar masses, and degrees of concentration.   The paper is organized as follows. Section \\ref{data} briefly describes the spectroscopic sample of the SDSS. Section \\ref{properties} describes how we measure the set of properties studied here.  Section \\ref{results} shows the number density and luminosity density distributions of these properties of galaxies, and provides simple fits for the dependence on luminosity of the properties of galaxies with nearly exponential radial profiles and those with more concentrated radial profiles. Section \\ref{conclusion} summarizes and outlines future lines of research. ",
        "conclusions": " \\begin{enumerate} \\item The most luminous galaxies comprise a homogeneous red, highly concentrated, high surface brightness population, and reside in locally dense regions. Underluminous galaxies are less homogeneous, but in general are bluer, less concentrated, lower surface brightness, and reside in less dense regions. These results are in qualitative agreement with previous work. \\item The relations among galaxy luminosities, surface brightnesses, and colors separate neatly once one has separated them into concentrated and unconcentrated (exponential) groups. Some of these relationships are well-known (such as the color--magnitude relation of concentrated galaxies; \\citealt{baum59}) while others are less commonly discussed (such as the color--magnitude relation for exponential galaxies). \\item Local density is a strong function of luminosity, at least for the most luminous galaxies, and on color for galaxies of all colors. This dependence of clustering on galaxy type has been quantified many times previously, early on in the work of \\citet{oemler74a}, later in the cluster studies of \\citet{dressler80a}, and also on larger scales by \\citet{davis76a,giovanelli86a,santiago92a,guzzo97a,blanton00a}, and many others. \\end{enumerate}  In addition, we make several more minor observations about the distribution of galaxy properties: \\begin{enumerate} \\item Optical galaxy colors all correlate very strongly with \\band{0.1}{(g-r)} color. The distribution of \\band{0.1}{(g-r)} is double-peaked. \\item Most of the luminosity density in the universe is contained in galaxies within our range of galaxy luminosities ($-23.5 < M_{\\band{0.1}{i}} < -17.0$). \\item The very reddest galaxies (which are small in number) are in optical colors exponential galaxies, not concentrated galaxies. \\item For all types of galaxies, size increases with luminosity. For concentrated galaxies, surface brightness decreases with luminosity for the most luminous galaxies. For exponential galaxies, surface brightness increases with luminosity. \\end{enumerate}  Simple quantitative expressions of some of these results are contained in the parametric fits to the conditional distributions of Figure \\ref{cmrplot} and Table \\ref{cmrtable} as well as the two-dimensional projections given in the associated electronic tables. However, in future work we may characterize the distribution using a full seven-dimensional model of the distribution of properties.  A potentially important effect to account for in future analyses is the passive evolution of galaxy luminosities. Measurements of this evolution in the SDSS suggest that galaxy luminosities are brighter in the past (a trend of $4.2$, $2.0$, $1.6$, $1.6$, and $0.8$ magnitudes per unit redshift in the \\band{0.1}{u}, \\band{0.1}{g}, \\band{0.1}{r}, \\band{0.1}{i}, and \\band{0.1}{z} bands). While the difference in absolute magnitude over the whole redshift range is quite small (0.3 magnitudes) compared to the dynamic range in $\\band{0.1}{i}$-band absolute magnitude, some of the colors may be affected significantly. Because the survey is flux limited, this effect can, in principle, alter the slope of our measurement of the color-magnitude relationship. For the trends quoted here, accounting for evolution would make the measured slope steeper, because the high luminosity galaxies are observed primarily at high redshift, and are observed to be bluer than they would be at the lower redshifts of the lower luminosity galaxies to which we are comparing them. For example, \\citet{bernardi02q} have found such evolution of the $g-r$ color--magnitude diagram in their early-type galaxy SDSS sample (their Figure 24). We note that one can detect the effect of not accounting for evolution by comparing the results of Blanton et al. (in preparation) for the evolution-corrected luminosity function at $z=0.1$ to the $1/V_{\\mathrm{max}}$-based estimate of the $\\band{0.1}{i}$-band luminosity function presented here (our luminosity function extends to slightly higher luminosities than that estimate).  The scope of this paper has not allowed us to delve into a more detailed discussion of the individual relationships and their relation to galaxy formation theory. However, we could not do justice to these topics in the space and time available here, and leave this for the future, encouraging readers to use the data distributed here to perform similar work. We note that \\citet{kauffmann02a} have performed a similar analysis using fits to stellar population models. They also see bimodality in the distribution of galaxy properties, particular in their measure of the 4000 \\AA\\ break. In addition, they find that because their estimated mass-to-light ratio (in their case in the $\\band{0.1}{z}$-band) is a strong (and increasing) function of mass or luminosity, the dependence of surface mass density in a galaxy on total stellar mass is monotonic, even though the dependence of surface brightness on luminosity shows a maximum surface brightness at around an absolute magnitude of $M_\\ast$. A particularly interesting application along these lines would be to consider the dependence of local density on stellar mass.  Two important photometric parameters missing from this space are color gradients and axis ratios. As it turns out, the color gradients of galaxies in our sample are usually quite small, partly because the central bulges of our spiral galaxies are blurred with the disks by seeing. Nevertheless, using a restricted sample of nearby galaxies, one can characterize this distribution. The distribution of axis ratios is related to the three-dimensional shape of the galaxy, so understanding the conditional distribution of this quantity on other parameters will yield a better understanding of galaxy geometry. Preliminary tests have shown that the axis ratios of our red, concentrated objects tend to be close to unity, whereas the axis ratios of blue, unconcentrated objects tend to differ significantly from unity, agreeing with the common understanding of the nature of elliptical and spiral galaxies. Including a characterization of the two-dimensional shape of each galaxy would be an interesting way to expand the results found here.  We began this paper by noting that many galaxy properties have been known for decades to be correlated, and in particular to be correlated with position on the Hubble Sequence. Many readers will ask how galaxies of different morpholopgical classifications are distributed in this space, a question we could address using a nearby subsample of the galaxies studied here. We have not done so here because we believe that morphological classification is not a sufficiently specified measurement to be straightforwardly interpreted. The position along the Hubble Sequence is determined by most galaxy classifiers from a consideration of a galaxy's surface brightness, smoothness, concentration, axis ratio, the prominence of dust lanes, and spiral arm pitch angle. However, astronomy has not standardized the weights to be accorded by the classifier to each of these qualities of a galaxy image when placing each galaxy along the one-dimensional Hubble Sequence. In addition, the decision about a galaxy's classification is highly dependent on the observing conditions, especially the distance of the galaxy from the observer, the dynamic range of the image, the passband of the observation, and the angle from which the galaxy is observed. Even when considering a single image and allowing only the classifier to vary, the repeatability of classification appears to be low ($\\sigma \\sim 2$ in units of Revised Hubble $T$ type; \\citealt{naim95a}). Even this level of repeatability is not clearly due to the fact that classifiers all weight the various Hubble type criteria similarly, since the properties under consideration all correlate.  Nevertheless, as reviewed by \\citet{roberts94a}, morphological classification is clearly important, since it does correlate with many physical properties of galaxies. Furthermore, we {\\it can} specify certain aspects which determine morphological classification. For example, many investigators have quantified measures of surface brightness, concentration, smoothness/blobbiness (e.g. \\citealt{naim97a}), and lopsidedness (e.g. \\citealt{rudnick00a}), among others, and investigated the dependence of these measures on observing conditions. These efforts, of which this paper is a part, are forming a new approach to quantitative morphology that we hope will have a more direct and better specified connection to theoretical predictions."
    },
    "0209/astro-ph0209153_arXiv.txt": {
        "abstract": "We have developed a new numerical technique for simulating dusty-gas flows.  Our unique code incorporates gas hydrodynamics, self-gravity and dust drag to follow the dynamical evolution of a dusty-gas medium.  We have incorporated several descriptions for the drag between gas and dust phases and can model flows with submillimetre, centimetre and metre size ``dust\".  We present calculations run on the APAC\\footnote{Australian Partnership for Advanced Computing {\\texttt{http://nf.apac.edu.au/}}} supercomputer following the evolution of the dust distribution in the pre-solar nebula. ",
        "introduction": "Up until recently we had only one observation to test our theories of planet formation against - our own Solar System. Now however with new planets and solar systems being both identified and parameterised at a rate approaching one a month, the observational constraints are much tighter and our lack of understanding of many aspects of the planet formation process is all too obvious.  At the most basic level we know that micron size grains of dust in the pre-solar nebula clump and coagulate together to form planets, objects 10$^{13}$--10$^{14}$ times larger. Planet formation is a multi-stage process, taking us from dust grain to boulder to planetesimal to planetary embryo. Analytical arguments (Goldreich \\& Ward 1973) have presented us with constraints on the time scales for each stage but little more. It is the very first stage of the process that we are concerned with in this paper -- from micron scale dust to metre sized boulders.  Theoretical models have changed much in recent years and the simple picture of a thin dust layer accumulating at the disk midplane, becoming gravitationally unstable, and breaking into planetesimals (Safronov 1969; Goldreich \\& Ward 1973) now seems unlikely. Recently Goodman \\& Pindor (2000) showed that turbulent drag causes radial instabilities in the dust layer, even if the disk self-gravity is negligible. In their steady state models, grains in a uniform dust layer experience radial drift as expected.  However, for perturbed disks they predict that over-dense rings form within an orbital period, with the ring thickness similar to the thickness of the dust layer.  These rings eventually collapse into planetesimals in the kilometre size range.  In this paper we present the first three-dimensional numerical simulations that include the effects of hydrodynamical forces, self-gravity and gas drag upon an evolving dusty gas disk.  We describe a new numerical code, based upon the smoothed particle hydrodynamics (SPH) technique which uses a collection of particles to approximate a fluid. At present we run simulations with uniform grain size and do not allow dust particles to have individual or time varying chemical and physical properties. Adding this level of sophistication to the model would require only  minor changes to the code. ",
        "conclusions": "For large (10m) and small (micron) dust sizes, we expect that the dust distribution will stay close to the initial flared disk. The largest grains (not shown) are weakly coupled to the gas, and if started in Keplerian motion, they will remain there.  On the other hand, the tiny grains are so strongly coupled to the gas that they are essentially co-moving (on the timescales we examine).  For both extremes, we see little evolution of the dust distribution. It is for the regime 0.1\\,mm $< r_{dust} <$ 1\\,m that the most significant disk evolution occurs.  In the $r-z$ plots of figure~1, significant deviation from the initially flared disk occurs in the inner regions of the 1m and 10cm plots, in the mid regions (from $r=0.6$ to $r=0.9$) of the 1cm plot, and in the outer regions of the 1mm and smaller plots. In these regions, the dust is moving at close to Keplerian speeds, whereas the gas is (as always) sub-Keplerian. As the velocity difference is sustained, the Epstein drag is optimal, and the energy loss rate is high -- allowing a thin layer of dust to form in the midplane and  migrate radially. These thin dense dust disks are those that Goodman \\& Pindor suggest have global turbulent instability modes.  While the 10cm and 1cm dust exhibits the highest surface density (top right of figure~1), the volumetric density is largest in the inner regions of the 1m and 10cm disks.  Therefore these size ranges are probably the most interesting from a planet formation viewpoint.  The Lagrangian nature of the code means that it is trivial to add empirical grain growth models, and to follow the grain temperature and density histories and hence generate chemical compositions.  The equations of state and drag term can easily be altered on a per-region or per-particle basis to account for local disk conditions."
    },
    "0209/astro-ph0209365_arXiv.txt": {
        "abstract": "{\\small We present unique radio observations of SS433, using MERLIN, the VLBA, and the VLA, which allow us to, for the first time, properly image and derive a meaningful spectral index for the `ruff' of equatorial emission which surrounds SS433's jet.  We interpret this smooth ruff as a wind-like outflow from the binary.} ",
        "introduction": "The central quarter-arcsecond of SS433's appearance at 5\\,GHz is rich in structure: both compact and smooth features may be found. To image this at radio wavelengths requires an interferometer with sufficiently long baselines to give adequate resolution.  Those long baselines will act as a spatial frequency filter which only detects compact emission; they are insensitive to larger-scale structures.  At a frequency such as 5\\,GHz, short baselines are also needed to faithfully detect smoother extended emission.  We illustrate this in Figure~\\ref{fig:ruff}: the left figure shows the central region of SS433 imaged using only the VLBA; the right figure shows the same region, with the same contour levels, on the same epoch, at the same frequency, at the same resolution, using the same VLBA data, but adding in also the shorter baselines of MERLIN.  In the left figure the only believable brightness structure is that associated with SS433's familiar jet, although hints of surrounding emission are also seen.  On the right figure, a wide smooth structure surrounding the jet appears, which we \\cite{Blu01} have termed SS433's {\\em ruff}.  Since the spatial filtering depends on the baseline length as measured in {\\em wavelengths}, it is most severe at the highest frequencies, and even the VLBA alone can detect the ruff at 1.4\\,GHz \\cite{Blu01}.  The spectral index of any extended emission may only be measured if that emission has been properly sampled at both frequencies. With VLBA data alone one can simply not detect the ruff at 5\\,GHz, while at 1.4\\,GHz it is obvious; the undersampling at high frequencies would lead to the derivation of a spuriously {\\em steep} spectral index.   Time variability is a further complication, making it essential to observe the two frequencies simultaneously. The observations we presented in \\cite{Blu01} at 1.4\\,GHz and at 5\\,GHz were taken on the same day (1998\\,Mar\\,7), and included the VLBA, MERLIN, and the VLA. This is thus a {\\em unique} dataset: there are sufficient short baselines at high frequency to adequately sample the emission, and both frequencies were observed quasi-simultaneously.  We find a flat spectral index for the anomalous emission (see below).  Paragi et al. \\cite{Par02a} claim a steep spectral index; but their high-frequency data are undersampled (as they pointed out in \\cite{Par99}), and their observations at the different frequencies are not simultaneous. Those data do not therefore usefully constrain the spectral index.  Our measurements of the distribution of the spectral index across SS433's ruff are shown in Figure\\,\\ref{fig:alpha}.  The spectral indices of the ruff were measured after convolving our images to a common beam of $10 \\times 10$\\,mas$^2$ HPBW.  The resulting total flux densities, measured in identical boxes in these images, which were chosen to avoid the jet but include the full `ruff' emission, are shown in Figure\\,\\ref{fig:alpha}$a$.  The spectral index for the combined (northern+southern) emission is $\\alpha=-0.12\\pm0.02$ ($S_{\\nu} \\propto \\nu^\\alpha$, where $S_{\\nu}$ is the flux density at frequency $\\nu$). Most resolved synchrotron sources are characterized by $\\alpha < -0.4$; indeed, $\\alpha=-0.1$ is normally considered the signature of thermal bremsstrahlung emission as is often observed in outflows from symbiotic binaries \\cite{Sea84,Mik01}.  The complication here is that the peak surface brightness corresponds to a brightness temperature of $(2-4)\\times10^7\\,\\rm K$ at 1.4\\,GHz, implying a similar {\\it lower limit} to the physical temperature of a thermally-emitting plasma. \\begin{figure}[h] \\centering \\hbox{ \\psfig{file=blundell_1.eps,width=0.5\\textwidth} \\hfill \\psfig{file=blundell_2.eps,width=0.5\\textwidth} } \\caption{Two identically-contoured images of the central quarter-arcsec of SS433 at 5\\,GHz which have been made (left) {\\bf without} and (right) {\\bf with} shorter baselines from MERLIN than the VLBA's long baselines.  The right image clearly reveals a smooth ruff of emission around SS433's jet which is only barely hinted at in the left image.  The measured flux density of the ruff is low by an order of magnitude if short baselines are missing, as quantified below: \\label{fig:ruff}} \\end{figure}  \\begin{center} \\begin{tabular}{cll} \\hline & & \\\\[-0.1cm] & Measurements from  & Measurements from  \\\\ & {\\bf VLBA} only    &   {\\bf VLBA \\& MERLIN}  \\\\[0.2cm] \\hline & & \\\\[-0.1cm] Northern ruff & \\multicolumn{1}{c}{3.5 mJy} & \\multicolumn{1}{c}{22.8 mJy}    \\\\ Southern ruff & \\multicolumn{1}{c}{3.1 mJy} & \\multicolumn{1}{c}{27.4 mJy}    \\\\[0.2cm] & \\underline{\\em without} short baselines & \\underline{\\em with} short baselines \\\\[0.2cm] \\hline \\end{tabular} \\end{center}   \\begin{figure}[h] \\centering \\hbox{ \\psfig{file=blundell_3.eps,width=0.5\\textwidth} \\hfill \\psfig{file=blundell_4.eps,width=0.5\\textwidth} } \\caption{{\\em a:} The total flux density in the ruff emission as a function of frequency (see text).  Crosses: northern emission; triangles: southern emission; filled circles: sum of northern and southern emission.  {\\em b:} The flux density integrated over 40\\,mas strips parallel to the jet, as a function of distance perpendicular to the jet, at 10\\,mas resolution.  The solid line is 18\\,cm, the dotted line the 6\\,cm data.  Note the flat spectrum of the ruff emission, compared to the core. \\label{fig:alpha}} \\end{figure}  The distribution of the flux density perpendicular to the jet is shown in Figure\\,\\ref{fig:alpha}$b$, which suggests that the spectral index is indeed almost flat throughout the ruff, and shows that the emission extends to \\gtsim\\ $40\\,\\rm mas$ at our sensitivity, or $\\sim120\\left(d/3\\,\\rm kpc\\right)\\,\\rm AU$.  Note also that the ruff is roughly symmetric about the jet. ",
        "conclusions": ""
    },
    "0209/astro-ph0209015_arXiv.txt": {
        "abstract": "A relativistic version of Pauli paramagnetism for $n-p-e$ system inside a strongly magnetized neutron star has been developed. An analytical expressions for the saturation value of magnetic field strength for each of these constituents at which they are completely polarized have been obtained. From the fully polarized configuration of electronic component, an upper limit for neutron star magnetic field is predicted. It has been concluded that indeed, magnetars, as stronly magnetized young neutron stars can not exist if the constituents are electron, proton and neutron in $\\beta$-equilibrium. An alternative model has been proposed. ",
        "introduction": "The study of the effect of strong quantizing magnetic field of neutron stars on dense nuclear matter has gotten a new dimension after the discovery of a few magnetars. These exotic objects are believed to be strongly magnetized young neutron stars of surface magnetic field $\\approx 10^{15}$G \\cite{R1,R2,R3,R4,R5}. The use of scalar virial theorem shows that the magnetic field strength at the core region may go upto $10^{18}$G \\cite{R6}. It is therefore very much advisable to study the effect of such strong magnetic field on various physical properties of dense neutron star matter as well as on various physical processes taking place inside neutron stars.  An extensive studies have already been done on the equation of state of dense neutron star matter in presence of strong magnetic field \\cite{R7,R8,R9,R10}.  Such studies are based on quantum mechanical effect of strong magnetic field. The effect of strong quantizing magnetic field on the gross properties, e.g., the mass, radius, moment of inertia etc., of neutron stars, which are strongly dependent on the equation of state of matter have also been obtained \\cite{R11}. In the second kind of studies, how the weak processes (reactions and decays) are affected by the quantum mechanical effect of strong magnetic field have been obtained \\cite{R13,R14}. As a consequence the $\\beta$-equilibrium condition also depends on the strength of magnetic field. Since the cooling of neutron stars is dominated by the emission of neutrinos produced by weak processes inside the stars, these studies also give an idea of the effect of strong magnetic field on thermal evolution of neutron stars. Not only that, the presence of strong magnetic field can change significantly, both qualitatively and quantitatively the transport coefficients (viscosity, thermal conductivity, electrical conductivity etc.) of dense neutron star matter \\cite{R15,R16,R17}. The magnetic field can change the tensorial character of transport coefficients of neutron star matter in presence of strong magnetic fields. Such qualitative change in transport coefficients can cause some significant changes in thermal evolution of neutron star matter and the evolution of its magnetic field. There are another kind of studies; the effect of strong quantizing magnetic field on quark-hadron phase transition \\cite{R18,R19}. It was shown explicitly that a first order quark-hadron phase transition is absolutely forbidden if the strength of magnetic field exceeds $10^{15}$G. However, a metal insulator type (color insulator to color metal) second order phase transition is possible unless the field strength exceeds $10^{20}$G. It has also been shown, that even if there is a first order quark-hadron phase transition for magnetic field strength  $< 10^{15}$G at the core region of a neutron star, an investigation of chemical evolution of quark matter, with various initial conditions leads to the system in $\\beta$-equilibrium, revealed that the system becomes energetically unstable in chemical equilibrium \\cite{R19}.  In some completely different type of studies, the stability and some of the gross properties of deformed stellar objects are analyzed with general relativity \\cite{R20,R21,R22,R23,R24}. The presence of strong magnetic field destroys the spherical symmetry of neutron star. Then it is possible for a deformed and rotating neutron star to emit gravity waves, which in principle may be detected.  Very recently we have critically studied the ferro-magnetism of neutron star matter which could be one of the sources of residual magnetism of old neutron stars/sources  of magnetic field of millisecond pulsars. In these studies, we have shown that a spontaneous Ferromagnetic transition in absence of external magnetic field is not possible in neutron star matter in $\\beta$-equilibrium. However in the case of neutrino trapped neutron star matter (Proto-neutron star matter), the possibility  of such a transition can not be ruled out, provided the neutrinos carry some non-zero mass \\cite{R25}. We have also analyzed the problem with the occupancy of zeroth Landau levels by electrons/protons, which occur in presence of ultra-strong magnetic fields \\cite{R26}. It has been argued in this critical analysis that in presence of a strong quantizing magnetic field the existence of neutron star matter in $\\beta$-equilibrium is questionable. Which further opens up a vital question on the possibility of magnetars as young and strongly magnetized neutron stars.  In this paper we shall present a relativistic version of Pauli paramagnetism of neutron star matter in $\\beta$-equilibrium. We shall extend the work of Shul'man \\cite{R27} to the relativistic region (i.e., in the high density regime of neutron star matter) and study the paramagnetism of relativistic nuclear matter. The aim of this paper is to show that the fully polarized configuration of electronic components puts a restriction on the upper limit of neutron star magnetic field.  We have also discussed the alternative picture of neutron star structure, if the magnetic field exceeds that limit.  The paper is organized in the following manner. In section 2, we shall develop the formalism of relativistic version of Pauli paramagnetism, in section 3 we study the Pauli paramagnetism of neutron star matter and in the last section, we shall conclude and discuss the importance of this work. ",
        "conclusions": "We have noticed that although the electronic component becomes fully polarized for $B\\sim 10^{16}$G, the nuclear matter can only become fully polarized if and only if the magnetic field strength exceeds $10^{20}$G. We have also observed that for such strong magnetic fields make $n^\\downarrow_e$ negative, which is completely unphysical.  Therefore, the fully polarized configuration of electronic component restricts the upper limit of a neutron star magnetic field to $\\approx 10^{16}$G. However the inclusion of pions $(\\pi^-)$ or kaons $(K^-)$ instead of electrons does not impose any restriction on the upper limit. At the same time the inclusion of these two components instead on electrons allow the $n-p$ system to become fully polarized even in $\\beta$-equilibrium condition. We can therefore conclude that the magnetars, if they exist at all and are assumed to be strongly magnetized young neutron stars, the constituents are possibly $n-p-\\pi^-$ or $n-p-K^-$ instead of $n-p-e$. Otherwise the existence of such objects is impossible. The incorporation of interaction in nuclear matter sector does not change the conclusion. The observed maximum polarization in the electronic sector whereas almost unpolarized configuration of nuclear matter regime is because of several orders of magnitude difference in the magnitudes of magnetic dipole moments. In the case of electron it is $\\sim$ Bohr magneton, whereas in the case of nucleons it is $\\sim$ nuclear magneton. This is independent of the type of interaction in the neutron star matter. \\begin{figure} \\psfig{figure=updown.eps,height=0.5\\linewidth} \\caption{ The variation of $n^\\uparrow$ and $n^\\downarrow$ (in terms of $n$) with $x$. } \\end{figure} \\begin{figure} \\psfig{figure=pm20.eps,height=0.5\\linewidth} \\caption{ Variation of dimensionless energy variables $x_i$'s with $B$, expressed in terms of $B_c^{(e)}$, for $n_B=2n_0$ and $\\beta=0$. } \\end{figure} \\begin{figure} \\psfig{figure=pm25.eps,height=0.5\\linewidth} \\caption{ Variation of dimensionless energy variables $x_i$'s with $B$, expressed in terms of $B_c^{(e)}$, for $n_B=2n_0$ and $\\beta=0.5$. } \\end{figure} \\begin{figure} \\psfig{figure=pm21.eps,height=0.5\\linewidth} \\caption{ Variation of dimensionless energy variables $x_i$'s with $B$, expressed in terms of $B_c^{(e)}$, for $n_B=2n_0$ and $\\beta=0.99$. } \\end{figure} \\begin{figure} \\psfig{figure=pm60.eps,height=0.5\\linewidth} \\caption{ Variation of dimensionless energy variables $x_i$'s with $B$, expressed in terms of $B_c^{(e)}$, for $n_B=6n_0$ and $\\beta=0$. } \\end{figure} \\begin{figure} \\psfig{figure=pm65.eps,height=0.5\\linewidth} \\caption{ Variation of dimensionless energy variables $x_i$'s with $B$, expressed in terms of $B_c^{(e)}$, for $n_B=6n_0$ and $\\beta=0.5$. } \\end{figure} \\begin{figure} \\psfig{figure=pm61.eps,height=0.5\\linewidth} \\caption{ Variation of dimensionless energy variables $x_i$'s with $B$, expressed in terms of $B_c^{(e)}$, for $n_B=6n_0$ and $\\beta=0.99$. } \\end{figure}"
    },
    "0209/astro-ph0209223_arXiv.txt": {
        "abstract": "As indicated by Einstein's general relativity, matter and geometry are two faces of a single nature. In our point of view, extra dimensions, as a member of the {\\em geometry face}, will be treated as a part of the {\\em matter face} when they are beyond our poor vision, thereby providing dark energy sources effectively. The geometrical structure and the evolution pattern of extra dimensions therefore may play an important role in cosmology. Various possible impacts of extra dimensions on cosmology are investigated. In one way, the evolution of homogeneous extra dimensions may contribute to dark energy, driving the accelerating expansion of the universe. In the other way, both the energy perturbations in the ordinary three-space, combined with homogeneous extra dimensions, and the inhomogeneities in the extra space may contribute to dark matter. In this paper we wish to sketch the basic idea and show how extra dimensions may lead to the dark side of our universe. ",
        "introduction": "\\label{introduction} It is strongly suggested by observational data that our universe has the critical energy density and consists of 1/3 of dark matter and 2/3 of dark energy (see e.g., Ref.~\\cite{Turner:2002zb} and references therein), where ``dark'' indicates the invisibility. Even though it is generally not an elegant way to explain data via something we cannot see, the avalanche of data, including those from type Ia supernova measurements \\cite{Perlmutter:1999np,Riess:1998cb}, cosmic microwave anisotropies \\cite{Sievers}, galactic rotation curves, and surveys of galaxies and clusters (providing the power spectrum of energy density fluctuations), make it more and more convincing. Nevertheless, we accordingly need to ask a question: {\\em Why are dark matter and dark energy so dark?}  This question reminds us another ``dark'' stuff, extra dimensions. The existence of extra dimensions is required in various theories beyond the standard model of particle physics, especially in the theories for unifying gravity and other forces, such as superstring theory. Extra dimensions should be ``hidden'' (or ``dark'') for consistency with observations. This common feature, ``invisible existence'', of dark energy, dark matter, and extra dimensions provides us a hint that there may be some deep relationship among them.  In this paper we show how extra dimensions may manifest themselves as a source of energy in the ordinary three-space and lead to the dark side of the universe. Basically homogeneous extra dimensions will contribute to dark energy and may also provide some sort of dark matter effectively if combined with the effects of inhomogeneities in the ordinary three-space, and inhomogeneities in the extra space will contribute to dark matter effectively. The basic idea is sketched in the next section, and then we discuss in Sec.\\ \\ref{homog ED to dark energy} how homogeneous extra dimensions provide ``effective'' dark energy and influence the evolution of the ordinary three-space, especially, producing the accelerating expansion of the universe. The extra dimensions employed throughout this paper are small and compact, as introduced in the Kaluza-Klein theories.\\footnote{Various scenarios for hidden extra dimensions have been proposed, for example, a brane world with large compact extra dimensions in factorizable geometry proposed by Arkani-Hamed \\emph{et al.} \\cite{Arkani-Hamed,Antoniadis:1990ew}, a brane world with extra dimensions in warped nonfactorizable geometry proposed by Randall and Sundrum \\cite{Randall&Sundrum}, and small compact extra dimensions in factorizable geometry as introduced in the Kaluza-Klein theories \\cite{Kaluza&Klein}.} ",
        "conclusions": "\\label{summary} In this paper we make a point that there may be a deep relationship between \\mbox{``hidden''} (or ``dark'') extra dimensions and the dark side of the universe, i.e., dark matter and dark energy. This conjecture is based on Einstein's general relativity, which indicates an important aspect that matter (with energy and momentum) and geometrical structures of a space-time are two faces of a single nature, to be called {\\em matter face} and {\\em geometry face}, respectively. In our point of view, if there exists a part of the {\\em geometry face} which is beyond our poor vision, this missing part will be treated as a member of the {\\em matter face}, and consequently provide mysterious, dark, ``effective'' energy sources. A possible missing part of the {\\em geometry face} we consider in this paper is the existence of extra dimensions. This idea is sketched in Sec.~\\ref{idea sketch} via analyzing the Einstein equations, including perturbations of both the metric tensor and the energy-momentum tensor, for a higher-dimensional world. We conclude that extra dimensions may manifest themselves as a source of energy in the ordinary three-space, such as ``effective'' dark energy, under the consideration of homogeneous extra dimensions, and ``effective'' dark matter, as contributed by inhomogeneities in the extra space or the ordinary three-space.   As a particular demonstration of the general idea, we consider in Sec.~\\ref{homog ED to dark energy} a (non-relativistic-) matter-dominated universe with homogeneous extra dimensions and show that the evolution of homogeneous extra dimensions can lead to ``effective'' dark energy and consequently change the evolution pattern of the universe. There are many possibilities of evolution patterns in this higher-dimensional universe, in contrast to the unique way of evolution, eternally decelerating expansion, for a matter-dominated universe in the standard cosmology without extra dimensions. It needs further detailed studies to determine which evolution pattern can appropriately describe our universe. In addition, there are various possible realizations of this idea worthy of further quests, and some are currently under our investigation.   As mentioned in Sec.\\ \\ref{introduction}, this work is motivated by a fundamental question: {\\em Why are dark matter and dark energy so dark?} Through the preliminary studies of the general idea discussed in this paper, here comes up a possible answer: {\\em Dark matter and dark energy are generated from the extra dimensions, a nature of geometry we are too blind to see.} This simple answer indicates an intriguing possibility of unifying these two kinds of dark entities, extra dimensions and dark energy sources, into one.       \\newpage"
    },
    "0209/astro-ph0209001_arXiv.txt": {
        "abstract": "The intervening large--scale structure distorts cosmic microwave background (CMB) anisotropies via gravitational lensing.  The same large--scale structure, traced by dusty star--forming galaxies, also induces anisotropies in the far--infrared background (FIRB).  We investigate the resulting inter--dependence of the FIRB and CMB with a halo model for the FIRB.  In particular, we calculate the cross--correlation between the lensing potential and the FIRB. The lensing potential can be quadratically estimated from CMB temperature and/or polarization maps.  We show that the cross--correlation can be measured with high signal--to--noise with data from the {\\it Planck Surveyor}. We discuss how such a measurement can be used to understand the nature of FIRB sources and their relation to the distribution of dark matter. ",
        "introduction": "Dusty star--forming galaxies give rise to a far--infrared background (FIRB) \\citep{puget,fixsen98,dwek,sfd,lagache,guiderdoni98,blain99}. Correlations in the large--scale structure traced by these contributing sources lead to correlated fluctuations in the FIRB \\citep{bond86,sw00,hk00,knox01,magliocchetti01}.  At arcminute scales and more, fluctuation power associated with the source distribution can potentially be detected with Planck and other planned CMB experiments with channels at frequencies around and above 300 GHz \\citep{knox01}.  The same large--scale structure that generates FIRB anisotropy also generates anisotropy in the CMB in several ways.  These include modifications due to scattering via free electrons in galaxy clusters, such as the thermal Sunyaev-Zel'dovich effect \\citep{SZ}, and modifications imposed by the time evolving gravitational field, such as the integrated Sachs-Wolfe effect \\citep{ISW}. The large--scale structure mass field also deflects CMB photons propagating to us from the last scattering surface via the gravitational lensing effect \\citep{seljak96}. Since the lensing effect on CMB anisotropies is second order in temperature fluctuations, it induces non-Gaussian signatures in the temperature data; the cross--correlation between the lensing effect and other secondary anisotropies, such as SZ or ISW, contributes to the temperature bispectrum \\citep{GS,CH00,SM99}.  We extend previous discussions on correlations between the CMB and large--scale structure and consider the cross-correlation of CMB anisotropies and FIRB fluctuations.  The FIRB contributes significantly at the high frequency end of certain CMB experiments, such as the 350 GHz, 545 GHz and 850 GHz channels of the High Frequency Instrument (HFI) of the Planck Surveyor.  Over this range of frequencies, and in regions of the sky with low galactic dust emission, the FIRB stands out as the dominant source of fluctuation power over a wide range of angular scales \\citep{knox01}.  We model the expected cross-correlation between lensing potentials and the FIRB though a physically motivated semi-analytical approach involving the halo model \\citep{scherrer,seljak,scocci,cooray02}.  Though the lensing potential-FIRB correlation leads to a bispectrum or a three-point correlation function in CMB temperature and FIRB, we suggest a direct cross-correlation between lensing potentials and FIRB. Our suggestion follows from recent discussions in the literature on how to reconstruct lensing potentials associated with CMB lensing, especially through higher order statistics such as quadratic estimates optimized for the lensing extraction in temperature and polarization data \\citep{zs99,hu011b,hu01,ck02}.  We consider this possibility for planned missions such as Planck.  Using reasonable assumptions, we show that Planck has sufficient sensitivity for a detection of the lensing-FIRB correlation.  A detection of such a correlation would allow us to test how well the sources of the FIRB trace the large--scale mass distribution.  While semi--analytic approaches such as the halo model suggest a high correlation, the exact correlation as measured will allow us to further refine details of these models and to understand certain physical properties of contributing FIRB sources.  The layout of the paper is as follows. In Section 2 we present the halo model for FIRB fluctuations.  In Section 3 we revisit the angular power spectrum of the FIRB and describe the cross correlation between the FIRB and a quadratic function of the CMB temperature map. The quadratic CMB statistic is an estimator for the lensing potential.  In section 4 we discuss associated errors and the extent to which the FIRB-lensing correlation can be detected. We conclude with a discussion of our results in Section 5. We refer the reader to \\citet{hk00} and \\citet{knox01} for initial details on our calculation of correlations in the FIRB.  More details of the halo approach to large--scale structure are available in the recent review by Cooray \\& Sheth (2002). For a review of our observational knowledge of the FIRB see \\citet{hauser01}.  While we provide a general derivation of the FIRB-lensing correlation, when illustrating our results we assume a $\\Lambda$CDM cosmological model with $\\Omega_m=0.35,\\Omega_\\Lambda=0.65,\\Omega_b=0.05,h=0.65$. To describe linear clustering, we use the transfer function given by \\citet{eisen} and normalize fluctuations to COBE \\citep{BunWhi97} such that $\\sigma_8=0.86$. ",
        "conclusions": "We now discuss the main results of our paper. We have introduced a semi-analytic approach to describe the  FIRB fluctuations. The model involves a distribution of dark matter halos populated by dusty, starforming galaxies. The basic ingredients of the calculation include properties of this dark matter distribution, such as the mass function and clustering properties of dark matter halos, and properties of the sources responsible for the FIRB. To describe these sources we have introduced a relation, the halo occupation number, that attempts to capture how FIRB sources populate dark matter halos. These model  can be used to calculate measurable properties of the sources such as their bias as a function of redshift and their spatial correlation with the underlying dark matter as a function of redshift.  The approach presented here differs from previous attempts to model the fluctuations in the FIRB.  HK00 used several biasing models, all of which were scale--independent.  The one most similar to our calculation here assumed all the FIRB sources were in $10^{12} \\msun$ halos and used the biasing prescription of \\citet{mo96}.  Because we include a range of halo masses extending below $10^{12} \\msun$ and because we take the FIRB mean to be $\\sqrt{2}$ lower than HK00 did our power spectra have smaller amplitudes.  The fluctuation power we predict is also smaller than that in \\citet{knox01} who used a constant $b=3$ amplification of the non--linear power spectrum calculated using the fitting formula of \\citet{PeaDod96}.  Even though we predict a smaller power spectrum than in the constant bias models we still expect FIRB fluctuations to be detectable in upcoming high frequency experiments such as Planck. Any multifrequency detection will allow the testing of our models and hopefully the extraction of interesting properties of our universe such as the evolution of the star--formation rate.  We do not discuss such possibilities in detail here as they have been already addressed in \\citet{knox01} and references therein.  In this paper we have gone beyond the FIRB fluctuation power and focused on the cross-correlation between FIRB sources and the dark matter as traced by the intervening  lensing potential that  deflected CMB photons propagating to us. Note that previous studies have already suggested that experiments such as Planck will detect the lensing potential power spectrum with considerable signal-to-noise ratios \\citep{zs99,zal00,hu01,ck02}. In addition to the separate detections of FIRB and projected lensing potential power we have suggested that one can also perform a combined study to measure the cross-correlation between FIRB fluctuations and the lensing potential.  We have discussed an extension of the estimators suggested for lensing reconstruction in CMB data. Using quadratic statistics one can define an estimator of the deflection angle, which can then be directly correlated with a high frequency map where FIRB fluctuations are expected to dominate. Such a direct approach avoids complications associated with other methods for extracting the lensing-FIRB cross power spectrum such as the direct measure of the three point function. As discussed in the literature, the FIRB-lensing correlation leads to a three-point correlation function, or a bispectrum in Fourier space \\citep{GS,CH00}. Note that measuring the full configuration dependence of this bispectrum is difficult and currently limited by computational methods  and measurement techniques. While improvements are expected, our suggested technique avoids these issues by combing the different three point function configurations  in a particular way.  As shown in Fig.~4, we find a considerable correlation between FIRB fluctuations and the lensing potential. As illustrated in Fig. \\ref{fig:weights}, this is due to the broad overlap between the respective radial weighting functions. Such a high correlation coefficient also leads to the  conclusion that upcoming CMB experiments will be able to detect the cross power spectrum between FIRB and lensing with high signal-to-noise.  For Planck, we expect the cumulative signal-to-noise to be of order 40 in the $\\nu=545$ GHz channel.  The high signal to noise expected for the detection  of the cross correlation and of the FIRB and lensing power spectra suggests that we will be able to  test our underlying model for the clustering of FIRB sources. As we discussed  our approach needed two main ingredients, the bias of FIRB contributing sources which was calculated in the halo model, and the  star formation rate which is needed to obtain FIRB weighting function. In turn, the halo approach depends on how starforming dusty galaxies populate dark matter halos. As shown in Fig. 3 and Fig. 4 we expect that different values of the model parameters can be distinguished with the level of noise expected for Planck.  In Fig.~\\ref{fig:rcoeff}, we summarize our results with respect to how well the projected correlation coefficient, $r_{\\rm corr}(l)$, and the bias factor, $b(l)$, can be measured with upcoming Planck data. While Planck allows a reliable detection of the correlation coefficient as well as bias out to a multipole of $\\sim$ 1000, there are further improvements one can hope to achieve in the post-Planck era. As summarized in Fig.~\\ref{fig:rcoeff}(c), in the case of the correlation coefficient, the error is dominated by the uncertainty associated with the lensing reconstruction. With improved data involving better angular resolution and noise one can hope to reach the limiting case of a perfect experiment which we have demonstrated with a dashed line Fig.~\\ref{fig:rcoeff}(c).  A few words of caution are required at this point. Note that we have described the FIRB and its fluctuations as coming only from dusty starburst galaxies. In addition to the stellar contribution, there is also some from dust associated with torii surrounding active galactic nuclei. While the fraction of AGN contribution at far-infrared wavelengths could be as high as 30\\% to 40\\% \\citep{alm01,risaliti02}, we have limited information on their redshift evolution so we have not included them in our calculation. The inclusion of AGNs within the halo model is trivial, the problems arise when trying to calculate the weighting function.  Another caveat comes from our simple  description of how submm sources populate halos. While we have described them through a  halo occupation number, the physics is likely to be more complicated especially if dusty star formation is associated with mergers. With a merger rate proportional to $N_{\\rm gal}^2(m)$, the effective occupation number will probably depend on a higher power of mass than considered here. In such a scenario, we expect source bias to be larger with an increase in clustering power compared to the results presented here. Such an increase should be  detectable and constrained using observational data.  While these issues may complicate the direct interpretation of the observations, a reliable detection of the FIRB-lensing correlation can allow one to introduce more sophisticated analysis techniques to try to understand  these complicating factors. For example, the two power spectra and the cross-power spectrum can be combined and written as $C_l^{FF} = b2(l) C_l^{\\phi\\phi}$ and $C_l^{\\rm F\\phi}=b(l)r(l)C_l^{\\phi\\phi}$, which can be inverted with appropriate techniques to obtain $r(k,z)$ and $b(k,z)$. Such an inversion requires accurate information on the FIRB radial weight function which can be obtained observationally through the redshift distribution of sources that contribute to FIRB. While unresolved surveys such as Planck will not allow such studies, in the future, targeted studies of resolved FIRB sources, over small but representative patches of sky may provide the necessary information.  Constraints on this emissivity as a function of redshift from current data are discussed in \\citet{gispert00}, \\citet{chary01} and \\citet{takeuchi01}. With an inversion of two-dimensional clustering to  three-dimensions one can eventually constrain various aspects of the halo model such as the occupation number \\citep{c02}. While previous studies have motivated the use of FIRB fluctuations alone to understand an important aspect of the large scale structure and its evolution history, we also suggest that cross-clustering aspects, such as between FIRB and dark matter as traced by gravitational lensing of CMB also plays an important role."
    },
    "0209/astro-ph0209371_arXiv.txt": {
        "abstract": " ",
        "introduction": "The formation of populous secondary star cluster systems is a widespread phenomenon in mergers of gas-rich galaxies. Many, if not most, of those clusters are massive and compact enough to be young globular clusters (GCs). GC systems in most E/S0 galaxies feature bimodal color distributions with a fairly universal blue peak similar to the blue peak of halo GCs in the Milky Way (MW) and M31, and a variable red peak. Due to the well-known age -- metallicity degeneracy of optical broad-band colors, the metallicities and ages, and, hence, the origin of the red peak GCs are not yet known. We use evolutionary synthesis models for GC {\\bf systems} of various metallicities to study the time evolution of their luminosity functions (LFs) in various bands U,..., K and of their color distributions. By comparison with the universal blue peak GC population we investigate for which combinations of age and metallicity a second GC population can or cannot be identified in typical observations of GC color distributions and we discuss implications for the GC LF as a distance indicator. ",
        "conclusions": ""
    },
    "0209/astro-ph0209147_arXiv.txt": {
        "abstract": "{We examine the impact of discrete numbers of stars in stellar populations on the results of Chemical Evolution Models. We explore the resulting dispersion in the true yields and their possible relation with the dispersion in observational data based on a Simple Closed-Box model. \\\\ In this framework we find that the dispersion is larger for the less evolved or low abundance regions. Thus, the age-metallicity relation may be a tracer of the Star Formation History of our Galaxy. This theoretical dispersion is especially high for the relative abundance log(N/O) in regions where the total number of stars created is still low. This may explain part of the scatter in the N/O ratio observed in star forming galaxies.  \\\\ We have also found a first order theoretical estimation for the {\\it goodness} of a linear fit of the Helium abundance vs. 12 + log (O/H) with values of the regression coefficient between 0.9 and 0.7 (independent of sampling effects).  \\\\ We conclude that it is necessary to include these sampling effects in a more realistic Chemical Evolution Model in order that such a model reproduces, at the same time, the mean value and the dispersion of observed abundances. ",
        "introduction": "The modelling of any observable related with the integrated contribution of stellar populations needs to assume an Initial Mass Function (IMF) and a Star Formation History (SFH). The representation of such distributions in the form of analytical expresions allows an easy implementation in any evolutionary model. However, such analytical laws are valid only under the assumption of an infinite number of stars, when sampling effects (i.e. the integer nature of the number of stars) are not relevant. Even in the case of the modelling of a system where the real physical properties are known, sampling effects must be present showing a dispersion of the data around the mean value obtained by the model \\citep[see, e.g.][]{CLC00}.  \\cite{CVGLMH02}, based on the work by \\cite{Buzz89}, have presented a method to evaluate quantitatively the intrinsic dispersion due to sampling effects in spectrophotometric synthesis models. It is important to remark that the dispersion associated to such sampling effects is indeed observed, and used as a distance indicator, in the form of surface brightness fluctuations in elliptical galaxies \\cite[see][ for a theoretical study of the subject]{Buzz93}.  In a next step, \\cite{Ceretal01} applied this formalism to some quantities which are time-integrated to be obtained, such as the ejected masses of elements by the stars when they evolve.  With that work, it was demonstrated that the estimation of these ejection rates may have a large dispersion, mostly in the first 5 Myr of the evolution of a stellar cluster.  On the other hand, the observational data used in the comparison of Galactic Chemical Evolution Models (GCE) have a very large dispersion: in the Milky Way Galaxy (MWG) there exist regions at a same Galactocentric distance with oxygen abundances differing by 0.2 dex, means dispersions as large as 50\\%. Such dispersion can be due to observational errors, but also to sampling effects.  As an example, a stellar cluster that has transformed gas into stars following a Salpeter IMF in the mass range 120 -- 0.08 M$_\\odot$ in a single burst, will produce $6.8 \\times 10^{-3}$ SN M$_\\odot^{-1}$. In the galactic context, the most massive OB association known \\citep[Cygnus OB2, c.f.][]{Kno00} has transformed into stars $6.5 \\times 10^{4}$ M$_\\odot$ of gas, assuming the presence of 120 O stars in the association (that is, in fact, an upper limit). A region of such characteristics will produce $440\\pm 20$ SN in a 63.8\\% (1 $\\sigma$) confidence interval due to the limited number of stars in the region. The formal relative dispersion is around 5\\%, which will result in a similar relative dispersion in the observed metallicity of that type of regions. So, if the chemical evolution of our Galaxy is driven by such a type of massive stellar clusters, at least, a {\\it theoretical} relative dispersion around 5\\% must be expected in the comparison of GCE models with observed data. But in fact, the dispersion must be larger as long as most of the observed OB associations have a lower amount of gas transformed into stars and the enrichment depends on the initial mass of the exploding star.  Taking into account the dispersion found for the stellar yields, we think that the same sampling effect may also produce a dispersion in the interstellar medium abundances. However, the stellar yields do not give directly the abundances in a region. It is necessary to use them as the input in a chemical evolution model.  Chemical evolution models are usually computed through numerical methods which solve a system of equations where the IMF such as the SFH are taken into account. However, before computing the dispersion of the abundances predicted in these numerical chemical evolution models, we would like to estimate the importance of these sampling effects with simpler chemical evolution models, and see if the resulting dispersion is at least of a similar order of magnitude to that observed.  Therefore, the objective of this paper is to obtain a first order estimation of how relevant these sampling effects can be in simple GCE models in order to consider whether it is necessary to include them in more complicated numerical chemical evolution models.  The structure of the paper is the following: In Sect. 2 we show the application of the formalism presented in \\cite{CVGLMH02} to the computation of yields in GCE models. In Sect. 3 we obtain the dispersion on the true yields for several metallicities and turn-off masses.  In Sect. 4 we apply the formalism to a GCE closed model and we obtain a first order approximation of the relevance of sampling effects in the results.  We show the conclusions and implications of this work in 5. ",
        "conclusions": "In this work we have shown a theoretical formalism for the evaluation of the dispersion on GCE models. Using a simple Closed-Box model, we have obtained a first order approximation to the relevance of the discreteness of the stellar population and the sampling effects in GCE models.  Despite the approximations, the effect of the discreteness of the stellar populations may be relevant for explaining the dispersion in the age--metallicity relation. Such observed dispersion may be an indicator of the SFH of our Galaxy. It may be also relevant in the study of the observed dispersion in the N/O ratio in H{\\sc ii} regions and dwarf galaxies.  We have found that the dispersion in the N/O ratio is larger for lower metallicities. This effect is indeed observed in our Galaxy.  We have obtained a first order theoretical estimation for the {\\it goodness} of a linear fit of the Helium abundance vs. 12 + log (O/H) with values of the regression coefficient between 0.9 and 0.7.  Finally, we want to note that in the case of GCE models attempting to trace the metallicity distribution of our Galaxy, the dispersion will not scale with the total mass of the Galaxy. Instead, it must be computed with the masses of the subsystems that are sampled: the dispersion of abundances for individual H{\\sc ii} regions from the global metallicity gradient observed in our Galaxy \\citep{hen99} could be understood in this framework since these regions are less massive systems.  However, no conclusion can be obtained unless this effect is included in a more sophisticated GCE.  It is therefore necessary to make a more exhaustive study on this subject. It will be performed in forthcoming papers."
    },
    "0209/astro-ph0209621_arXiv.txt": {
        "abstract": "% Using three complete, radio flux limited, blazar samples we compare the LogN-LogS and the preliminary radio luminosity function of the general population of BL Lacs to those of the subclass of high energy synchrotron peaked (HBL) BL Lacs. We also examine recent results on the cosmological evolution in different samples of BL Lacs and we investigate the controversial issue of the correlation between the synchrotron peak frequency and radio luminosity in BL Lacertae objects. We find that the fraction of HBL objects is approximately the same at all observed radio fluxes and luminosities implying that there cannot be any strong correlation between the position of the synchrotron peak and radio luminosity. The amount of cosmological evolution in BL Lacs is confirmed to be low and negative at low radio fluxes, although the large number of objects  without redshift prevents a precise estimation. At high radio fluxes the amount of cosmological evolution is zero or slightly positive but this could be induced by a possible contamination with Flat Spectrum Radio Quasars. ",
        "introduction": "The blazar class includes BL Lacertae type objects (BL Lacs) and Flat Spectrum Radio Quasars (FSRQs), both characterized by a strong and highly variable non-thermal continuum across the entire electromagnetic spectrum. The often extreme and peculiar observational properties of these sources  (e.g. Urry \\& Padovani 1995) are thought to be the signature of physical phenomena seen in rare conditions such as the electromagnetic emission arising from a jet of material moving toward the observer at relativistic speeds observed at small viewing angles. For these reasons blazars are intensively studied at all frequencies from radio to high energy gamma rays. However, their low surface density and the intrinsic difficulties in discovering them at optical frequencies where BL Lacs do not display strong emission lines nor show any Blue Bump, make blazar research a difficult business. So difficult, in fact, that it took nearly 20 years to build the first statistically complete samples such as the 1Jy and 2Jy samples at 5GHz (Stickel et al. 1991, Wall \\& Peacock 1985) and the X-ray flux-limited samples derived from the {\\it Einstein} EMSS and Slew surveys (Stocke et al. 1991, Rector et al. 2000, Perlman et al. 1996a). Most of the experimental results on which our present understanding of blazars is based come form the study of these early samples which include 30-50 objects, the brightest and most luminous in their selection band.  Despite the early difficulties many new surveys, based on classical or more recent multi-frequency selection techniques, are now becoming available. These surveys, however are characterized by different degrees of completeness and by a range of sensitivity limits in one or more energy bands; see Padovani (2002) for a recent review.  Using three recent, radio flux-limited and nearly complete surveys, which cover a portion of the blazar parameter space much wider than earlier samples, we determine some cosmological properties of BL Lacs such as the radio LogN-LogS, the radio luminosity function and its cosmological evolution and we address the controversial issue of the experimental basis of the often quoted correlation between \\nupeak and radio luminosity (Fossati et al. 1998). Throughout this paper we assume $H_0=50~$Km~s$^{-1}$Mpc$^{-1}$ and $q_0=0$. ",
        "conclusions": "Using three new radio flux limited samples that cover a large portion of the blazar parameter space we have derived the cosmological properties of BL Lac objects and we have investigated the existence of the correlation between \\nupeak and radio luminosity in BL Lac objects. We have shown that \\begin{itemize}  \\item The radio LogN-LogS of extreme HBL objects remains approximately parallel to that of all types of BL Lacs down to 50 mJy and flattens significantly at fainter fluxes.  \\item The radio luminosity function of HBL sources also remains approximately parallel to that of the general population of BL Lacs although the large number of objects without redshift leaves some uncertainties. There is no evidence for a strong increase in the fraction of HBL sources as would be expected if \\nupeak is strongly correlated with radio luminosity.  \\item The correlation found by Fossati et al. (1998) might have been induced by the comparison of samples with widely different radio flux limits and by the fact that high luminosity HBLs were under-represented in early samples since a) their space density at high radio fluxes is so low that hardly any object of this type is expected in a shallow radio survey, and b) their flat synchrotron spectrum, which extends to very high energies, easily outshines the optical emission of the host galaxy thus preventing the measurement of their redshift and hence of their luminosity.  \\item The amount of cosmological evolution in BL Lacs is confirmed to be low and probably negative. This is mainly confined to samples of low radio flux and high \\nupeak objects, although some contamination by FSRQ in radio flux limited surveys may have artificially increased the value of \\vovaave .  \\item Although at the moment we do not have direct evidence of the existence of high radio power HBL sources, such evidence could be obtained detecting high redshift absorption lines, due to intervening intergalactic material, in high quality spectra of the many featureless HBLs in the Sedentary survey. \\end{itemize}  \\setcounter{figure}{5} \\begin{figure}[!ht] \\vspace*{-2.9cm} \\centering \\epsfysize=8.0cm\\epsfbox{gal_lbl_hbl.ps} \\caption[h]{The optical emission from a typical (giant elliptical) BL Lac host galaxy is compared to the broad-band Synchro-Self-Compton emission of two hypothetical LBL and HBL sources of equal radio luminosity. While the emission of the HBL object outshines that of the host galaxy by a large factor, the optical non-thermal emission of the LBL is a small fraction of that of the galactic emission. In this case the HBL would appear as a featureless object for which no redshift would be measurable, while the LBL would be classified as a radio galaxy.} \\label{gal_lbl_hbl} \\end{figure}"
    },
    "0209/astro-ph0209284_arXiv.txt": {
        "abstract": "We define an enhanced spectral classification scheme for M dwarf stars, and use it to derive spectral classification of 104 northern stars with proper motions larger than $0.5\\arcsec$ yr$^{-1}$ which we discovered in a survey of high proper motion stars at low galactic latitudes. The final tally is as follows: 54 M dwarfs, 25 sdK and sdM subdwarfs, 14 esdK and esdM extreme subdwarfs, and 11 DA and DC white dwarfs. Among the most interesting cases, we find one star to be the coolest subdwarf ever reported (LSR2036+5059, with spectral type sdM7.5), a new M9.0 dwarf only about 6pc distant (LSR1835+3259), and a new M6.5 dwarf only 7pc from the Sun (LSR2124+4003). Spectroscopic distances suggests that 27 of the M dwarfs, 3 of the white dwarfs, and one of the subdwarfs (LSR2036+5059) are within 25pc of the Sun, making them excellent candidates for inclusion in the solar neighborhood census. Estimated sky-projected velocities suggest that most of our subdwarfs and extreme subdwarfs have halo kinematics. We find that several white dwarfs and non metal-poor M dwarfs also have kinematics consistent with the halo, and we briefly discuss their possible origin. ",
        "introduction": "The current census of stars in the Solar Neighborhood (the volume of space within $\\approx25$pc of the sun) is believed to be significantly incomplete, especially at the faint end of the luminosity function. Despite the fact that increasing numbers of substellar objects (L dwarfs, T dwarfs), and low mass stars (M7-M9 dwarfs) are now being discovered with the help of large infrared surveys \\citep{KRLCNBDMGS99, DTFEBFKT99, GR97} there remain significant numbers of red dwarfs and white dwarfs which are still unaccounted for \\citep{HWBG97}.  Nearly all the local M dwarfs and white dwarfs are expected to be brighter than magnitude $\\sim20$ in the optical bands. The majority of them should be detectable as stars with large proper motions. If they haven't been identified yet it is because existing all-sky surveys of high proper motion stars \\citep{L79, L80} are themselves significantly incomplete \\citep{SIIJM00, LSR02b}. Furthermore, the classification and characterization of even the known high proper motion stars is still under way \\citep{GR97, JSML01, RKC02}.  Recently, several new additions to the solar neighborhood have been confirmed from follow-up observations of newly discovered high proper motion stars \\citep{PGCDFBEFS01, HWBG02, SIILSS02a, RRSI02}. Follow-up observations of faint stars with large proper motions have also revealed the existence of previously unreported types of objects like a very cool extreme subdwarf \\citep{SSSIM99}, a magnetic DZ white dwarf \\citep{RLS01}, and a pair of nearby cool white dwarfs \\citep{SSAII02b}.  In a previous paper \\citep{LSR02b} we have reported the discovery of 140 new stars with proper motions larger than $0.5\\arcsec$ yr$^{-1}$ at low galactic latitudes in the northern sky. For the past two years, we have been carrying out a large spectroscopic follow-up survey of the new high proper motion stars that we are finding in the northern sky. The spectroscopy is being performed at the Lick Observatory, the MDM Observatory, and the Kitt Peak National Observatory. This first paper of a series presents spectral classification of a first set of 104 stars, all of which are listed in \\citet{LSR02b}. Observations are described in \\S2. In \\S3, we describe our classification method, which expands on previous spectral index methods. Estimation of the radial velocities and calculation of the spectroscopic distances is detailed in \\S4, where we also discuss the kinematics of the objects. Especially interesting or intriguing stars are discussed in \\S 5. Important results are briefly summarized in \\S 6. ",
        "conclusions": "We have obtained spectra for 104 new stars with large proper motions ($\\mu>0.5\\arcsec$ yr$^{-1}$) found at low galactic latitudes \\citep{LSR02b}. We find that 54 of the targets are M dwarfs (M), 25 are subdwarfs (sdK, sdM), 14 are extreme subdwarfs (esdK, esdM), and 11 are late-type DA and DC white dwarfs.  We have expanded and refined the G97 classification method for M dwarfs and subdwarfs by defining a larger set of spectral indices whose calibration with spectral type can be used for spectral classification. The new scheme can be applied to perform spectral classification of all M dwarfs (early and late-type), subdwarfs, and extreme subdwarfs, from a spectrum covering the 6000\\AA-9000\\AA\\ wavelength range. Among the M dwarfs classified in this paper, 8 have a spectral subtype M7 and later, including one new M9.0V star. We also find one subdwarf with a very late spectral type of sdM7.5, the coolest subdwarf ever reported.  We have provided a crude classification of the 11 white dwarf stars, from blackbody fits of the spectral energy distribution in the 6000\\AA-9000\\AA\\ range. All the white dwarfs are relatively cool DA and DC, with the warmest at DC8 (6500K) and the coolest at DC11 (4750K). Many of the DC spectra are relatively noisy, and the DC spectral class was assigned only by default, for lack of clear identification of H$\\alpha$ absorption. We thus suspect some of our DCs might be actual cool DAs.  We have derived a spectral-type / absolute magnitude calibration using sets of M, sdM, and esdM stars with published astrometric parallaxes. We have found the counterparts of those stars in the Guide Star Catalog-II (GSC2.2.1) and in the 2MASS Second Incremental Release, to obtain their observed B, R, and ${\\rm K_s}$ magnitudes. The spectral-type / absolute magnitude relationships for M dwarfs was modeled with a third order polynomial, while we fit a linear relationship for the sdM and esdM stars. Using the empirically determined spectral-type / absolute magnitude relationships, we have determined spectroscopic distances for all the observed red dwarfs and subdwarfs in our sample.  We have obtained crude radial velocity measurements for most of our stars using centroid measurements of \\ion{Na}{1}, \\ion{K}{1}, \\ion{Ca}{1}, and \\ion{Ca}{2} atomic absorption lines. Combining these radial velocities with the estimated distances and proper motions, we estimate the UVW velocity components (measured relative to the local standard of rest) of all the stars. The distribution of M dwarf stars in UVW space is consistent with most of the stars being components of the disk. However, some have velocities that are more consistent with halo orbits, which is quite surprising for non-metal poor stars. We suggest that these stars originate in the disk but got ejected into halo orbits through some yet undefined mechanism. Possible scenarios include 3-body gravitational interactions or massive cluster evaporation.  While we find the UVW space distribution of white dwarfs to be similar to that of the M dwarfs, the velocity space distribution of subdwarfs and extreme subdwarfs is markedly different, and is largely consistent with the distribution of known, very metal-poor, candidate halo stars. We find that a significant number of sdM and esdM stars fall outside the 2$\\sigma$ limit of halo stars in UVW space. These stars are apparently on very eccentric halo orbits, some of which appear to be at the limit of being gravitationally bound to the Galaxy.  Finally, we report the discovery of a number of new, very nearby stars, including two stars that are most likely to be well within 10pc of the Sun, and 29 more (25 M dwarfs, 3 white dwarfs, and 1 subdwarf) that are likely to be within 25pc of the Sun. The two new very nearby stars are LSR1835+3259, an M9.0 dwarf at about 6pc, and LSR2124+4003, and M6.5 dwarf at about 7pc from the Sun.  Follow-up spectroscopy of newly discovered faint stars with large proper motions proves yet again to be very productive in recovering intrinsically faint stars in the Solar neighborhood, and faint stars with extreme transverse velocity components (possibly halo stars). Our spectroscopic follow-up survey of new high proper motion stars is still in progress, and is being expanded as new northern stars with large proper motions are being discovered. Future results will be presented in upcoming papers in this series."
    },
    "0209/astro-ph0209020_arXiv.txt": {
        "abstract": "We present spin-resolved X-ray data of the neutron star binary Her X-1 taken using the EPIC detectors on {\\sl XMM-Newton}. The data were taken at three distinct epochs through the 35 day precession period. The energy dependent light curves of this source vary significantly from epoch to epoch. It is known that the relative phasing of the soft (\\ltae 1~keV) and hard (\\gtae 2~keV) X-rays varies. Here, we find that the phase shift between the soft and hard bands during the main-on state is considerably different from that observed in the past. Further, it continues to change significantly during the other two observations. This suggests that we are observing, for the first time, a {\\it substantial and continuous variation in the tilt of the disk}, as it is expected if the accretion disk is precessing in the system.  Analysis of the spin resolved data confirms that the equivalent width variation of the fluorescence Fe K line at $\\sim$6.4~keV follows that of the soft X-ray emission in the main-on state, thus suggesting a common origin for Fe K line and thermal component. The Fe K line is considerably broader when the source is brightest. ",
        "introduction": "\\label{int}  Her~X-1 is an eclipsing binary system consisting of a neutron star and an A/F secondary star, HZ~Her.  Since its discovery, it has been observed extensively in many wavebands revealing a high degree of complexity in its behaviour. Her~X-1 has a spin period of $\\sim 1.24$~s and a binary orbital period of 1.7~d (Tananbaum et al. 1972, Giacconi et~al. 1973, Deeter, Boynton \\& Pravdo 1981, Oosterbroek et~al. 2001). Eclipses are seen which recur on the timescale of the binary orbital period, indicating that the system has a high inclination. It also varies in X-rays on a period of 35~d, with a ``main-on'' state lasting $\\sim 10$~d and a secondary ``short-on'' state lasting $\\sim 5$~d. Between these states the source is fainter for $\\sim 10$~d.  The origin of the 35~d cycle is thought to be due to the precession of a tilted, warped accretion disk that periodically obscures X-rays from the central neutron star (see e.g. Gerend \\& Boynton 1976, Parmar et~al. 1999, Ketsaris et~al. 2000, Coburn et~al. 2000). This idea is strengthened by the fact that optical emission from HZ~Her persists at the same level throughout the main-on and low state, suggesting that the companion is still being irradiated by a strong X-ray source (Bahcall \\& Bahcall 1972, Bahcall 1978, Delgado et~al. 1983).  The broadband X-ray spectrum of Her~X-1 is also extremely complex. During the main-on state, the overall continuum is well described by a thermal blackbody component with temperature $kT_{bb}\\sim0.1$~keV, that dominates the spectrum at low energies (McCray et~al. 1982, Oosterbroek et~al. 1997), and a broken power-law plus an exponential cut off at higher energies. The galactic hydrogen column density in the direction of Her X-1 is low, $N_H \\sim 10^{19}-10^{20}$~atoms cm$^{-2}$.  In addition to these spectral features, at least three other components have been identified: i) a feature at $\\sim 1$~keV, often referred to as the `Fe~L line' (McCray et~al. 1982, Oosterbroek et~al. 1997; see also Mihara \\& Soong 1994, Oosterbroek et~al. 2000 who reported a discrimination in two narrow lines at 0.90 and 1.06~keV); ii) a broad Fe emission feature at $\\sim 6.4$~keV (Pravdo et~al. 1977, Choi et~al. 1994, Oosterbroek et~al. 1997, Coburn et~al. 2000); iii) a cyclotron absorption feature at $\\sim 40$~keV. The latter gives an inferred value of the magnetic field $\\sim 3\\times 10^{12}$~G (see e.g. Dal Fiume et~al. 1997).  An analysis of {\\sl Ginga}, {\\sl RXTE} and {\\sl BeppoSax} data (Mihara et~al. 1991, Oosterbroek et~al. 1997, Coburn et~al. 2000, Oosterbroek et~al. 2000) has shown that out with the main-on state the effects of a significant intervening absorption (with typical $N_H$ $> 10^{21}$ atoms cm$^{-2}$) are clearly seen as a change in the spectral shape. However, a substantial flux remains below 0.5~keV that should be completely absorbed by such material unless the associated covering factor is partial (Oosterbroek et~al. 2000).  There are two possibilities which cannot be spectrally distinguished: a) the presence of two separate ``scattering'' and ``absorbed'' regions; b) an intrinsic partial covering of the emitting region, i.e. a situation in which the emission spectrum is the same in the low and main-on states, but a fraction of the emerging radiation is highly scattered and absorbed. Both cases are referred in the literature as ``partial covering'' models, and b) is what is expected, for instance, during occultation phases of the neutron star by the accretion disk.  Another manifestation of the 35~d precession cycle is the repeating, systematic evolution of the X-ray pulse profile over this cycle. Extensive coverage of these variations have been obtained using $Ginga$ (Deeter et~al. 1998). Simultaneous X-ray and UV observations have also been carried using $ASCA$ and $HST$ by Boroson et al. (1996), showing that the pulse shape becomes smooth and sinusoidal from the soft X-ray band into the UV (see also Leahy, Marshall, \\& Scott 2000). Past attempts to model the pulse changes in various energy bands relied on a combination of free neutron star precession and disk obscuration, as well as obscuration by flaps of matter at the magnetosphere or obscuration by the tilted disk (see Deeter et~al. 1998, Scott, Leahy \\& Wilson 2000).  However, most of these studies fail in explaining more than a few aspects of the complex pulse evolution. Recently, detailed interpretations have been presented by Deeter et~al. (1998) and Scott et~al. (2000). In particular, the model by Scott et~al. (2000) is based on inner disk occultation of the X-ray beam from the neutron star. The latter, in turn, consists of two components: a pencil direct beam and a fan beam emission in the antipodal direction (see \\S \\ref{enres}). This model refines the disk and pulsar beam geometry and qualitatively accounts for {\\it both} the pulse phase {\\it and} its evolution during the main-on and short-on high states.  In this paper we present the results of a series of three observations made using the EPIC cameras on board {\\sl XMM-Newton} at different $\\Phi_{35}$ (see \\S \\ref{obs}). We discuss the spin period of the neutron star in \\S \\ref{spin} and the pulse profile in different energy bands are reported in \\S \\ref{pulse}. In \\S \\ref{spec} we report the pulse-averaged broadband spectra, and in \\S \\ref{pulsespec} we examine the pulse resolved spectra. We discuss our results in \\S \\ref{disc}. ",
        "conclusions": "\\label{disc}   The results reported in this paper show for the first time that there is a continuous variation in the relative phase offset between the soft and hard X-ray components. We also find that normalisation and equivalent width of the fluorescence Fe line at 6.4~keV are modulated in phase with the soft X-ray component. Further, we find that for energies above $\\sim$2~keV the three spectra corresponding to three different $\\Phi_{35}$ are consistent with a single emission model and an absorption component which is lowest at $\\Phi_{35}$=0.17, and highest at $\\Phi_{35}$=0.26. Both the inferred values of $N_H$ and covering factor for the intervening matter are consistent with Coburn et~al. (2000), with the former considerably higher than that reported by Oosterbroek et~al. (2000). We also require more than one partial covering component, and both these facts hint towards a fairly complex substructure of the thermal emission. On the other hand, the nature of the soft X-ray emission is not well understood and deserves a further, more detailed investigation. We now proceed by discussing the physical implications of our results.   \\subsection{The energy resolved light curves} \\label{enres}  The pulse profile of Her~X-1 is known to evolve through the main-on state; those reported here closely resemble that observed using $Ginga$ (Deeter et~al.  1998). The EPIC observations taken at $\\Phi_{35}=0.17$ and $\\Phi_{35}=0.26$ can be qualitatively compared to Figure~6 of Deeter et~al. (1998; main state D), while those taken at $\\Phi_{35}=0.60$ are similar to their Figure~7 (short state C). Many of those features are naturally explained within the scenario proposed by Scott et~al. (2000). This model was originally developed to explain $Ginga$ and $RXTE$ observations, and it is based on the obscuration of a multi-component X-ray beam by a counter-precessing, tilted, twisted disk.  For simplicity, the X-ray beam is assumed to be decoupled from the disk and is axi-symmetric; the observer's line of sight must be close to the binary plane (to explain two maxima per 35~d cycle). One of the main features of this model is that it ascribes the variations observed in the pulse profile over the 35~d cycle to occultation from the {\\it inner} part of the disk, whereas most of the previous investigations have assumed an occultation from the {\\it outer} boundary (but see also Petterson, Rothschild \\& Gruber, 1991, who argued that vertical ``flaps'' which form near the magneto-spheric radius are partly responsible for the 35~d phase behaviour).  Inner disk phenomena, in fact, are more plausible in explaining the peaks evolution of the spin profile, as well the differences in the pulse-shape of main-on and short-on. Such differences require that the size of the dominant pulse-emitting region is roughly the scale-height of the inner disk, so that the two high states are caused by progressive occultation of an {\\it extended} source.  The complex pattern of peaks observed in Her X-1 can be explained in terms of successive occultations of a beam consisting of a direct pencil beam (that originates close to the poles at the surface of the neutron star) plus a reversed fan beam that is focused in the antipodal direction. These two components are superimposed on three other constant flux components: two due to magneto-spheric emission and one due to low-state coronal emission. In the ``reversed fan beam'' geometry, the fan beam component originates at the same distance above the neutron star surface and has an opening angle $> 90^\\circ$. A similar configuration has been predicted theoretically by Brainerd \\& Meszaros (1991), who studied the backscattering of the magnetic polar cap radiation by the incoming accretion flow and its subsequent gravitational focusing around the neutron star. The accretion column is predicted to be optically thin to Thomson scattering, while the fan beam photons are produced by cyclotron resonance scattering.  The overall situation is summarized in the bottom panel of Figure 8 in Scott et~al. (2000), while their Figures 10a-10b illustrate the evolution of the pulse profiles predicted during the main-on and short-on respectively.  A number of the features observed in our {\\sl XMM-Newton} data are qualitatively reproduced.  At $\\Phi_{35}=0.17$, the hard emission shows two main peaks (`A' and `B' in Figure~\\ref{fold}), and a third, lower peak (`C' in Figure~\\ref{fold}). This situation is close to that modeled by Scott et al. (2000) at similar $\\Phi_{35}$, during the progressive occultation of leading and training peaks of the hard beam. At $\\Phi_{35} \\sim 0.27$, when the main components are occulted, we only observe the survival of a broad, underlying modulation that is attributed to the magneto-spheric emission. Since this component is emitted from a larger region at some distance from the neutron star, it is naturally expected to have a lower modulation as well as a broad maximum (Figure~\\ref{fold}).  The pulse profile close to the short-on is also similar to that presented by Scott et~al. (2000) at $\\Phi_{35}\\approx 0.58$ (see their Figure~10b).  In the EPIC data, we can in fact recognize a main peak `D' as well a less prominent peak `E' (Figure~\\ref{fold}).  A cross-correlation between different {\\sl XMM-Newton} datasets does not allow a proper phase alignment between main-on and short-on peaks based on the extrapolation of the pulse timing ephemeris. However, because the peak `E' is the hardest feature, spectral considerations suggest that this maximum is associated with the small hard peak and the feature `D' with the soft peak discussed by Scott et~al.~(2000). If this is the case, `E' is actually due to direct emission from the pencil beam, while `D' is the radiation redirected into the fan beam from the antipodal accretion column.  \\subsection{The phase shift between the soft and hard light curves} \\label{shift}  Given the complexity of the source, pulse-phase spectroscopy is of paramount importance to separate the different spectral components observed in Her X-1. Using $Einstein$ (McCray et~al. 1982) and $BeppoSax$ data (Oosterbroek et~al. 1997, 2000) it has been shown that, during the main-on state, the maximum of the thermal component and the power law components are shifted by $\\sim 250^\\circ$ and that the maximum of the unresolved feature at $\\sim 1$~keV is in phase with that of the blackbody component. The situation is less consistent as far as the 6.4~keV Fe K line is concerned: Choi et~al. (1994) have shown that its intensity is modulated in phase with the soft emission, suggesting a common origin for the two Fe lines, while Oosterbroek et al. (2000) have found it correlated with the hard (power law) emission.  The shift in phase between hard and soft emission can be explained if the latter results from re-processing of hard X-rays in the inner part of the accretion disk. If a non-tilted disk intercepts (and re-processes) a substantial fraction of the hard beam from the neutron star, the expected phase difference between direct and reflected component is $180^\\circ$.  Therefore, the value of $\\sim 250^\\circ$ determined using $Einstein$ and $BeppoSax$ data has been associated with the disk having a tilt angle. If the tilt of the disk changes with phase along the 35~d cycle (as predicted by the precessing disk models, see Gerend \\& Boyndon 1976) the shift in phase should therefore vary with $\\Phi_{35}$. However, both Einstein and Sax data were obtained at the same $\\Phi_{35}$, i.e. during the main-on state ($\\Phi_{35} =0.1$ for $Einstein$; $\\Phi_{35} =0.07-0.15$ for $BeppoSax$ in 1997 and $\\Phi_{35} =0.1-0.2$ for $BeppoSax$ in 2000). Oosterbroek et~al. (2000) also observed the source at $\\Phi_{35} =0.5$ and found that the pulse phase difference in the short-on and main-on state are consistent.  This is not surprising, since symmetry considerations allow for the same behaviour at $\\Phi_{35} =0.0$ and 0.5. A tracking of the phase difference between the two components over the entire cycle was therefore required.  Here we have found that not only the phase shift derived from {\\sl XMM-Newton} main-on data is considerably different from previous observations made in the main-on, but it continues to change dramatically during the other two observations. This suggests that we are observing, for the first time, a {\\it substantial and continuous variation in the tilt of the disk}, which is what we would expect from a system which had a precessing accretion disc. It should be noted that the interpretation of the phase shift observed at the short-on may be affected by a systematic error, depending on whether during the observation the soft peak `D' is higher than the small hard peak `E' or vice-versa (see \\S \\ref{enres}).  \\subsection{The fluorescent line variation} \\label{line}  To investigate in more detail the possible common origin of the soft component and the Fe line at 6.4~keV, we have derived the line parameters as a function of the spin phase, $\\phi_{spin}$ (see \\S \\ref{pulsespec} and Figure~\\ref{fitph}).  At $\\Phi_{35}$=0.26 and 0.60, there is little evidence for a significant variation in the line parameters.  The observation made during the main-on ($\\Phi_{35}=0.17$) clearly shows that the soft flux below 0.7~keV and the equivalent width all exhibit a common minimum at 0.2\\ltae$\\phi_{spin}$\\ltae0.4, which, in turn, is shifted with respect to that of the hard emission. This supports the idea that the 6.4~keV Fe line originates from fluorescence from the relatively cold matter of the illuminated spot where the soft emission is reprocessed.  In this case, the flux emitted in the line may provide a lower limit to the size of the spot because the observed line energy allows us to put an upper limit to the ionization degree and to the temperature of the emitting region, $T$\\ltae 0.3~keV (Kallman \\& McCray 1982). This is in principle appealing since it allows us to constrain the size of the illuminated spot in a way which is less subject to absorption (compared with the usual methods based on the value of the soft flux).  However, assuming a spherical spot and a distance to the source of 6.6~kpc (Reynolds et~al. 1997), the inferred lower limit for the radius is $R$ \\gtae 0.5~km, too low to add any significant constraints. For comparison, assuming the same distance for the source, the radius of the equivalent blackbody derived from the soft flux varies between $\\sim 140$~km and 500~km.  We have also found evidence for a variation in the Fe line parameters over the 35~d period. The line flux and the line width is greatest at the main-on state. We also show that the line energy is significantly higher in the main-on state and lowest in the low state. In the main-on state, the energy of the Fe line (6.540keV) corresponds to Fe XX-XXI, while in the low state (6.385keV) it is from low ionisation species.  This suggests two possible explanations for both the line broadening and the centroid displacement: 1) an array of Fe K fluorescence lines exists for a variety of charge states of Fe (anything from Fe I-Fe XIII to Fe XXIII); 2) Comptonization from a hot corona with a significant optical depth for a narrower range of charge states centered around Fe XX. Similar line broadening have been observed in some Low Mass X-Ray binaries observed using $ASCA$ (Asai et~al. 2000).  The Fe line broadening may also be explained in terms of Keplerian motion, if the inner disk (or some inner region) comes into view during the main-on state. If this is the case, at $\\Phi_{35}=0.17$ the Keplerian velocity will be $\\sim 13000$ km/sec. This, in turn, corresponds to a radial distance of $\\sim 4 \\times 10^8$~cm (for a neutron star of 1.4\\Msun), which is close to the magneto-spheric radius for a magnetic field of $\\sim 10^{12}$~G."
    },
    "0209/astro-ph0209216_arXiv.txt": {
        "abstract": "Hydrogen atoms inside virialized minihalos (with $T_{\\rm vir}\\leq10^4{\\rm K}$) generate a radiation background from redshifted 21-cm line emission whose angular fluctuations reflect clustering during before and during reionization.  We have shown elsewhere that this emission may be detectable with the planned Low Frequency Array (LOFAR) and Square Kilometer Array (SKA) in a flat Cold Dark Matter Universe with a cosmological constant ($\\Lambda$CDM). This is a direct probe of structure during the ``Dark Ages'' at redshifts $z\\ga6$ and down to smaller scales than have previously been constrained.  In our original calculation, we used a standard approximation known as the ``linear bias'' [e.g. Mo \\& White (1996)]. Here we improve upon that treatment by considering the effect of nonlinear clustering.  To accomplish this, we develop a new analytical method for calculating the nonlinear Eulerian bias of halos, which should be useful for other applications as well. Predictions of this method are compared with the results of $\\Lambda$CDM N-body simulations, showing significantly better agreement than the standard linear bias approximation. When applied to the 21-cm background from minihalos, our formalism predicts fluctuations that differ from our original predictions by up to 30\\% at low frequencies (high-$z$) and small scales.  However, within the range of frequencies and angular scales at which the signal could be observable by LOFAR and SKA as currently planned, the differences are small and our original predictions prove robust. Our results indicate that while a smaller frequency bandwidth of observation leads to a higher signal that is more sensitive to nonlinear effects, this effect is counteracted by the lowered sensitivity of the radio arrays. We calculate the best frequency bandwidth for these observations to be $\\Delta\\nu_{\\rm obs}\\sim2$ MHz.  Finally we combine our simulations with our previous calculations of the 21-cm emission from individual minihalos to construct illustrative radio maps at $z=9$. ",
        "introduction": "Despite striking progress over the past decade, cosmologists have still failed to see into the ``Dark Ages'' of cosmic time.  Although the details of this epoch between recombination at redshift $z\\sim10^3$ and reionization at $z\\ga6$ are crucial for understanding issues ranging from early structure formation to the process of reionization itself, no direct observations of any kind have been made in this redshift range.  Recently, we made the first proposal for such direct observations (Iliev et al. 2002, from hereafter Paper I) based on collisional excitation of the hydrogen 21-cm line in the warm, dense, neutral gas in virialized minihalos (halos with virial temperature $T_{\\rm vir}\\leq10^4{\\rm K}$), the first baryonic structures to emerge in the standard CDM universe.  We showed that collisional excitation is sufficient to increase the spin temperature of hydrogen atoms inside minihalos above the temperature the cosmic microwave background (CMB).  Minihalos should generally appear in emission with respect to the CMB, producing a background ``21-cm forest'' of redshifted emission lines, well-separated in frequency. Unlike all previous works \\citep{HR79,SR90,SP93,MMR97,SWMB99,TMMR00}, this mechanism does not require sources of Ly$\\alpha$ radiation for ``pumping'' the 21-cm line and decoupling it from the CMB.  In Paper I we calculated the 21-cm emission properties of individual minihalos in detail, along with the total background and its large-scale fluctuations due to clustering.  Although individual lines and the overall background are too weak to be readily detected, the background fluctuations should be measurable on $\\sim 10'-100'$ scales with the currently-planned radio arrays LOFAR and SKA.  We demonstrated that such observations can be used to probe the details of reionization as well as measure the power spectrum of density fluctuations at far smaller scales than have been constrained previously.  In the current paper we extend these results to smaller scales.  Our previous calculations showed that the fluctuation signal increases as the beam size and frequency bandwidth decrease, which corresponds to sampling the 21-cm emission from minihalos within smaller volumes.  As these volumes correspond in turn to length scales that are more nonlinear, nonlinear effects have the greatest impact on the angular scales and frequency bandwidths at which the signal is the strongest.  This issue is of particular importance as our previous investigations relied on the standard, simplified nonlinear bias of \\cite{MW96}, which breaks down at many of the relevant redshifts and length scales.  For example, an rms density fluctuation in a Cold Dark Matter universe with a cosmological constant ($\\Lambda$CDM) at $z = 6 (8),$ is strongly nonlinear ($\\sigma(M)\\geq 1$) for $M\\leq10^9M_\\odot$ ($6.2\\times10^7M_\\odot$), which roughly corresponds to the region sampled by beam sizes 9'' (3''), for frequency bandwidths 12 kHz (3 kHz), respectively.  Nonlinear effects can influence the predicted background fluctuations on even larger scales, up to few comoving Mpc (corresponding to few hundred kHz frequency bandwidths and 1-10 arc min beams), even if these scales are still not strongly nonlinear.  Additionally, rare halos are always more strongly clustered than the underlying density distribution (i.e. the bias is $>1$), again bringing nonlinear issues to the fore.  In order to address these important issues we have carried out a series of high-redshift N-body simulations of small scale structure formation, which we use to construct 21-cm line radio maps that illustrate the expected fluctuations in the emission.  As no current simulations are able to span the full dynamic range relevant to 21-cm emission, however, we extend our results by developing a new formalism for calculating the nonlinear Eulerian bias of halos, which is based on the Lagrangian bias formalism of Scannapieco \\& Barkana (2002, hereafter SB02). We describe this approach in detail in this paper, verify it by comparing it with the results of our N-body simulations, and apply it to to calculate improved predictions for the fluctuations in the 21-cm emission. These are then compared to the predictions given in Paper I, quantifying the impact of nonlinear effects on minihalo emission.  The structure of this work is as follows.  In \\S~\\ref{simul} we describe a set of simulations of minihalo emission at high redshift, and use these to construct simulated maps at small angular scales. In \\S~\\ref{bias_calc} we present our improved calculation of the Eulerian bias and verify it by comparison with the numerical simulations presented in \\S~\\ref{simul}. In \\S~\\ref{rms_nonlin} we modify the calculation of the radiation background from minihalos to incorporate the contribution due to nonlinear clustering of sources, according to the formalism described in \\S~\\ref{bias_calc}, and describe the results of our nonlinear formalism.  Conclusions are given in \\S~\\ref{conlucions}. ",
        "conclusions": "\\label{conlucions}  We have developed a new analytical method for estimating the nonlinear bias of halos and used it to calculate the fluctuations of the 21-cm emission from the clustering of high-$z$ minihalos.  This method is likely to be useful in tackling a much wider range of problems that lie beyond the capabilities of current numerical simulations, and will be refined and further verified in a future publication.  The minihalo bias predicted by our method at large scales reproduces the linear bias, as derived in \\cite{MW96}, and is much larger at small scales, confirming naive expectations.  At intermediate scales, however, its behavior is more complex and both mass- and redshift-dependent. For very rare halos at high-$z$ (roughly $z>15$), the standard linear bias is lower than our nonlinear prediction at all length scales. At the lower end of the redshift range we consider (down to $z=6$) the linear bias is higher at intermediate scales (few hundred kpc to few Mpc comoving) and lower at smaller scales.  We have also compared the predictions of our new method with the results of N-body numerical simulations, which we used both to verify our approach and to produce sample radio maps.  Due to the limited dynamical range of our simulations, however, these maps are only illustrative of the behavior of the fluctuations on very small scales, below the sensitivity limits of LOFAR and SKA.  On the scales at which the simulations are reliable, we find excellent agreement between these results and our analytical approach.  The most significant discrepancies occur in the calculation of cross-correlation functions of very different mass bins at higher redshifts. However, these departures are relatively modest, and in all cases our method reproduces the simulation results significantly better than the linear bias.  Despite these differences, we find our original linear bias predictions for the fluctuations of the 21-cm emission from minihalos to be robust at the scales and frequencies corresponding to observable signals.  The prediction using the flux-weighted nonlinear bias never departs from the linear prediction by more than few percent in that range, well within the other uncertainties of the calculation. This robustness is partly accidental, however, and is due to the nonlinear bias varying above and below the linear one depending on the length scale, leading to partial cancellation of the differences when the correlation function is integrated over the length scales.  For small observational bandwidths, $\\Delta\\nu_{\\rm obs}$, there is less cancellation and the differences in the two predictions grow at all beam sizes, $\\Delta\\theta_{\\rm beam}$. Similarly, if $\\Delta\\theta_{\\rm beam}$ is small, there is little cancellation, and the discrepancies are larger at all bandwidths.  In these cases the linear bias gives an overestimate of the 21-cm emission fluctuations from minihalos at the low end of the redshift range, and an underestimate at high-$z$, even at large values of $\\Delta\\nu_{\\rm obs}$. However, when both the beam size and the frequency bandwidth are large the differences in the two approaches at small scales are diluted, and the resulting 21-cm emission fluctuations become identical.  Finally, we predict that the best observational frequency bandwidth for improving the chances for detection is $\\Delta\\nu_{\\rm obs}\\sim 2$ MHz.  Thus it may be through a such frequency window that astronomers get their first glimpses of the cosmological Dark Ages."
    },
    "0209/astro-ph0209166_arXiv.txt": {
        "abstract": "X-ray binaries in the Milky Way are among the brightest objects on the X-ray sky.  With the increasing sensitivity of recent missions, it is now possible to study X-ray binaries in nearby galaxies.  We present data on six luminous sources in the nearby spiral galaxy, M101, obtained with the \\chandra\\ ACIS-S.  Of these, five appear to be similar to ultraluminous sources in other galaxies, while the brightest source, P098, shows some unique characteristics. We present our interpretation of the data in terms of an optically thick outflow, and discuss implications. ",
        "introduction": "X-ray binaries in the Milky Way, with typical intrinsic luminosities in the range 10$^{34}$--10$^{38}$ ergs\\,s$^{-1}$ at a typical distance of 8 kpc, dominate our 2--10 keV sky.  We therefore know a great deal about these Galactic X-ray binaries through many studies over the last several decades (see \\citealt{WNP95} for a review).  They are close binaries in which a neutron star or a black hole is accreting from a non-degenerate companion.  They can be divided into low-mass X-ray binaries (LMXBs) and high-mass X-ray binaries (HMXBs) depending on the spectral type of the mass donor.  The HMXBs are young objects that are concentrated in the Galactic plane, preferentially in spiral arms. The LMXBs, on the other hand, appear to belong variously to the old disk, bulge, and globular clusters.  Many neutron star LMXBs show thermonuclear flashes (i.e., type I bursts) suggesting a relatively low-magnetic field; many HMXBs are coherent X-ray pulsars, containing highly magnetized neutron stars.  The Eddington limit for a 1.4 M$_\\odot$ object is $\\sim 2 \\times 10^{38}$ ergs\\,s$^{-1}$.  Significantly more luminous X-ray binaries can be considered black hole candidates on this argument alone, although HMXB pulsars SMC X-1 and LMC X-4 appear to exceed this limit at times (however, we do not know if they actually violate the Eddington limit, since the emission geometry of X-ray pulsars is not spherically symmetric).  The definitive evidence for a black hole in X-ray binaries comes from radial velocity studies of the mass-donor in the optical.  In particular, over a dozen soft X-ray transients (SXTs; a subtype of LMXBs) have a measured mass function which exceeds 3 M$_\\odot$, with inferred compact object masses typically in the 5--15 M$_\\odot$ range \\citep{BO98}. In the X-ray regime, black hole binaries have characteristic spectral shapes, including a low/hard state dominated by a power law spectrum, and a high/soft state dominated by a $\\sim$1 keV thermal component, usually interpreted as arising from the inner disk \\citep{T95}. Although detailed studies have led to suggestions of additional spectral states \\citep[see, for example,][]{Z01}, these spectral states are different from those of typical neutron star systems, either a 5--10 keV bremsstrahlung-like spectrum (LMXBs) or a power-law with an exponential cut-off (HMXBs).  Although a great deal is known about Galactic X-ray binaries, studies of extragalactic X-ray binaries offer complementary insights.  In particular, a complete census of Galactic systems is difficult due to the extinction in the Galactic plane, and the luminosity estimates of Galactic systems are generally subject to large uncertainties.  Study of a face-on galaxy such as M101 allows a study of a luminosity-limited sample with, on average, a low absorbing column.  In recent years, we have gained an additional motivation to study extragalactic X-ray source populations, in the form of off-nuclear point sources with luminosities greater than 10$^{39}$ ergs\\,s$^{-1}$ (hereafter Ultraluminous X-ray sources, or ULXs\\footnote{Sources with luminosities in the 10$^{38}$ -- 10$^{39}$ ergs\\,s$^{-1}$ range have also been considered ULXs by some authors. They are likely to be related to ULXs, as well as to Galactic BHCs.}) that have been found in many nearby galaxies \\citep{C99}. One possible interpretation is that they are accreting intermediate mass (10$^2$--10$^4$ M$_\\odot$) black holes.  This would be exciting if confirmed, because previously known black holes could be categorized into stellar ($\\leq$10 M$_\\odot$) or supermassive ($> 10^6$ M$_\\odot$) subclasses. However, disk blackbody models of the \\asca\\ spectra of ULXs \\citep{M00} suggest they may have accretion disks as hot as several keV at their inner edges.   In the standard model, it is difficult for a disk around an intermediate mass black hole to achieve such high temperatures.  The superb angular resolution of \\chandra\\ allows the detection of point sources well below 10$^{38}$ ergs\\,s$^{-1}$ in nearby (say, closer than 10 Mpc) galaxies, thus sampling LMXBs, HMXBs, and both black hole and neutron star systems.  Consequently, many groups are studying X-ray source populations in a considerable number of nearby galaxies as summarized, for example in \\citet{Pr01}. Here we present preliminary results of our observation of M101. ",
        "conclusions": ""
    },
    "0209/astro-ph0209350_arXiv.txt": {
        "abstract": "We present a method for optimising experimental cuts in order to place the strongest constraints (upper limits) on theoretical signal models. The method relies only on signal and background expectations derived from Monte-Carlo simulations, so no bias is introduced by looking at actual data, for instance by setting a limit based on expected signal above the ``last remaining data event.'' After discussing the concept of the ``average upper limit,'' based on the expectation from an ensemble of repeated experiments with no true signal, we show how the best model rejection potential is achieved by optimising the cuts to minimise the ratio of this ``average upper limit'' to the expected signal from the model. As an example, we use this technique to determine the limit sensitivity of kilometre scale neutrino detectors to extra-terrestrial neutrino fluxes from a variety of models, e.g. active galaxies and gamma-ray bursts. We suggest that these model rejection potential optimised limits be used as a standard method of comparing the sensitivity of proposed neutrino detectors. ",
        "introduction": "In this paper, we examine the problem of choosing experimental cuts in order to place the most restrictive limits on theoretical signal models. How should one choose cuts with this in mind? If a signal is assumed, the goal is to maximise the significance of the observation, by for instance optimising signal to noise or signal to square root noise. If, however, one assumes that no signal will be observed, a different technique is required to optimise the ``model rejection potential,'' i.e. the limit setting potential of an experiment. This technique must not depend on the experimental data, since choosing cuts based on the data (e.g. cutting after the last remaining event) leads to confidence intervals that do  not have frequentist coverage.  In this paper, we describe and assess an unbiased method~\\cite{brussels,hartill}, based only on Monte Carlo signal and background expectations.      Firstly we discuss how an experimental observation is used to set a limit on a theoretical flux (section \\ref{howtolimits}). The desire to minimise the upper limit (and thus place the strongest constraint on the theoretical model) leads to the concepts of ``average upper limits'' and ``model rejection factor,'' which are discussed in section \\ref{aulamrp}. We show how choosing cuts based on optimising the ``model rejection factor'' leads to, on average, the best possible limit from the experiment. In section \\ref{km3sens}, we illustrate this technique by calculating the model rejection potential of a kilometre  scale neutrino detector with respect to predicted extra-terrestial neutrino fluxes such as active galaxies and gamma-ray bursts. ",
        "conclusions": "In this paper we have described  a method of choosing experimental cuts in order to maximise the chance of placing the strongest possible constraint on an expected signal model. Choosing the experimental cuts to optimise the model rejection factor (ratio of expected average upper limit to expected signal) is shown to yield the best average constraint on the signal model. We have demonstrated this method by determining  the sensitivity of kilometre-scale neutrino detectors to diffuse and point sources of astrophysical neutrinos, and find that such detectors will strongly constrain present models. We suggest that the optimised  average flux upper limits described here be used as a standard method of comparing the capabilities of different neutrino detectors.    \\ack{We would like to thank Albrecht Karle, Ty DeYoung, Francis Halzen and David Steele for useful discussions and comments on the manuscript, and Tom Gaisser for encouragement to bring this work to publication. We thank an anonymous referee for general comments and for reminding us of the importance of the prompt charm contribution to the atmospheric neutrino background. We acknowledge the support of the National Science Foundation, under contract number OPP-9980474.}"
    },
    "0209/astro-ph0209399_arXiv.txt": {
        "abstract": "{ In this paper we study A3560, a rich cluster at the southern periphery of the A3558 complex, a chain of interacting clusters in the central part of the Shapley Concentration supercluster. \\\\ From a ROSAT-PSPC map we find that the X-ray surface brightness distribution of A3560 is well described by two components, an elliptical King law and a more peaked and fainter structure, which has been modeled with a Gaussian. The main component, corresponding to the cluster, is elongated with the major axis pointing toward the A3558 complex. The second component, centered on the Dumb-bell galaxy which dominates the cluster, appears significantly offset (by $\\sim 0.15$ \\hmpc) from the cluster X-ray centroid. \\\\ From a Beppo-SAX observation we derive the radial temperature profile, finding that the temperature is constant (at $kT \\sim 3.7$ keV) up to 8 arcmin, corresponding to 0.3 \\hmpc: for larger distances, the temperature significantly drops to $kT \\sim 1.7$ keV. We analyze also temperature maps, dividing the cluster into 4 sectors and deriving the temperature profiles in each sector: we find that the temperature drop is more sudden in the sectors which point towards the A3558 complex. \\\\ From VLA radio data, at 20 and 6 cm, we find a peculiar bright extended radio source (J1332-3308), composed of a core (centered on the northern component of the Dumb-bell galaxy), two lobes, a ``filament\" and a diffuse component. The morphology of the source could be interpreted either by a strong interaction of the radio source with the intracluster medium or by the model of intermittency of the central engine. ",
        "introduction": "Merging has been recognized as the leading process in massive cluster formation, as a consequence of the hierarchical structure formation scenario. A large amount of numerical work has been done to study this phenomenon at all the relevant scales: among others, Ricker et al. (\\cite{ricker01}) studied in detail the physics of the plasma during the collision of two clusters under different initial conditions, while Colberg et al. (\\cite{colberg99}) analyzed the role of the cosmological environment on the merging. \\\\ From the observational point of view, the improvements of the Point Spread Function and sensitivity of the Chandra satellite led to the detailed description of the shocks (Markevitch et al. \\cite{markevitch02}) and the discovery of the so-called ``cold fronts\" (Vikhlinin et al. \\cite{vikhlinin01}), which are direct consequences of merging at an advanced state. \\\\ Little work has been done on the global, multiscale  description of this phenomenon. To this end, we are carrying on a long term project aimed at studying the merging in the particularly rich environment of the central part of the Shapley Concentration supercluster. In this region, three ``cluster complexes\" are found (Zucca et al. \\cite{zucca93}), i.e. structures of $\\sim 7$ \\hmpc (hereafter $h=H_o/100$) which represent major cluster mergings at various evolutionary stages. \\\\ The most massive structure, the A3558 complex (Figure \\ref{fig:largeview}), is probably a collision seen after the first core-core encounter (Bardelli et al. \\cite{bardelli98b}). The whole complex, formed by a chain of three ACO clusters and two poor groups, is embedded in a hot gas filament (Bardelli et al. \\cite{bardelli96}; Kull \\& B\\\"ohringer \\cite{kull99}) and in a common envelope of galaxies (Bardelli et al. \\cite{bardelli94}, \\cite{bardelli98a}). The estimated mass ranges between $10^{15}$ and $10^{16} M_{\\odot}$ (Bardelli et al. \\cite{bardelli00}, Ettori et al. \\cite{ettori97}, Reisenegger et al. \\cite{reisenegger00}). \\\\ The entire structure presents a large number of substructures, some of which have an X-ray counterpart as diffuse emission (Bardelli et al. \\cite{bardelli02}). Moreover, the merging seems to lead to a lack of radio sources with respect to ``normal\" clusters (Venturi et al. \\cite{venturi00}). The presence of a halo radio source, of a minihalo and a relic radio source (Venturi et al., in preparation) is further evidence of ``stormy weather\". \\\\ In this paper we concentrate on the cluster A3560, a rich cluster at the southern periphery of the A3558 complex. In Figure \\ref{fig:largeview} a mosaic of the available ROSAT-PSPC frames is shown and gives the large-scale distribution of the clusters in this region. The two groups labelled SC1327 and SC1329 are SC$1327-312$ and SC$1329-313$. \\\\ The distance of A3560 from the nearest X-ray clump of the A3558 complex (corresponding to SC$1329-313$) is $\\sim 3$ \\hmpc. Given the proximity of such a large mass concentration, the high predicted infall velocity ($\\sim 2000$ km s$^{-1}$, Reisenegger et al. \\cite{reisenegger00}) and the existence of the underlying overdensity of the supercluster (see Bardelli et al. \\cite{bardelli00}), a certain degree of disturbance can be expected for A3560. \\\\ The plan of the paper is the following. In Sect.2 we describe the general properties of A3560 and in Sect.3 we perform the spatial analysis of the ROSAT-PSPC map. In Sect.4 we analyze our new Beppo-SAX observations on this cluster, obtaining temperature profiles and maps, while in Sect.5 we present the properties of the central radio source of A3560. Finally in Sect.6 we discuss and summarize the results. ",
        "conclusions": "In this paper we studied A3560, a rich cluster (richness class 3) at the southern periphery of the A3558 complex, a chain of interacting clusters in the central part of the Shapley Concentration supercluster. \\\\ From the ROSAT-PSPC map we found that the X-ray surface brightness distribution of A3560 is well described by two components, an elliptical King law and a more peaked and fainter structure, which has been modeled with a Gaussian. The main component, corresponding to the cluster, is elongated with the major axis pointing toward the A3558 complex. The second component, centered on the Dumb-bell galaxy which dominates the cluster, appears significantly offset (by $\\sim 0.15$ \\hmpc) from the cluster X-ray centroid. However, the contribution of this component to the global luminosity is only $\\sim 2 \\%$. \\\\ From the Beppo-SAX observation we derived the radial temperature profile, finding that the temperature is constant (at $kT \\sim 3.7$ keV) up to 8 arcmin, corresponding to 0.3 \\hmpc: for larger distances, the temperature significantly drops to $kT \\sim 1.7$ keV. This drop questions the validity of the hydrostatic equilibrium hypothesis for regions at distances $> 0.3$ \\hmpc from the cluster center. We also analyzed temperature maps, dividing the cluster into 4 sectors and deriving the temperature profiles in each sector: we found that the temperature drop is significantly more sudden in sectors 1 and 2 (which point towards the A3558 complex), while it is smoother in the other sectors. \\\\ From the VLA radio data, at 20 and 6 cm, we found a peculiar bright extended radio source (J1332-3308), composed of a core (centered on the northern component of the Dumb-bell galaxy), two lobes, a ``filament\" and a diffuse component. The filament is not aligned with any of the two lobes and seems to point towards the diffuse component. \\\\ At a first look this source seems a Wide Angle Tail source, but the filament and the diffuse component do not fit this scenario. We suggest two possible interpretations of this: \\\\ i) All components are related to the same nuclear activity: in this case the filament and the diffuse emission are due to a strong interaction of the radio source with the intracluster medium. This interaction could originate from the offset position of the peaked X-ray component hosting the radio source with respect to the overall cluster (see Figure \\ref{fig:chi2fit}). This offset can suggest a motion of the peaked component roughly along the major axis of the cluster: the consequent ram pressure can be responsible for the peculiarity of the radio source. \\\\ ii) The components are the result of an intermittency of the nuclear engine, following the model invoked for 3C~338 by Burns et al. (\\cite{burns83}): in this case the filament and the diffuse component are the remnants of a previous activity of the radio source, while the core and the lobes are the result of the present activity of the same source. As in case i), also in this scenario a motion of the radio core in the North-South direction is required. The only difference with respect to 3C~338 is that in our case all components have flatter spectral indexes, indicating that J1332--3308 is younger. \\\\ Further investigations are needed to discriminate between these models: in particular, the interaction between the radio source and the cluster diffuse emission is a typical topic where the high resolution power of the Chandra satellite will be decisive. \\\\ As a general conclusion, the elongation in the direction of the A3558 complex, the offset of the peaked component with respect to the centroid of the cluster and its motion (suggested by the radio data) and the sudden drop in the temperature profile seem to indicate that A3560 is a dynamically disturbed cluster."
    },
    "0209/math0209047_arXiv.txt": {
        "abstract": "We describe a mechanical device which can be used as an analog computer to solve the transportation problem. In practice this device is simulated by a numerical algorithm. Tests show that this algorithm is 60 times faster than a current subroutine (NAG library) for an average $1000 \\times 1000$ problem. Its performance is even better for degenerate problems in which the weights take only a small number of integer values.  \\vspace{12pt} {\\em Key words}: transportation problem, analog computer, mechanical model. ",
        "introduction": "We describe here an algorithm for the solution of the transportation problem \\cite{Hit41a} (also known as the Hitchcock problem).  The development of this algorithm had its origin in studies of the lattice gas method for three-dimensional fluid simulations \\cite{HLF86a}.  The optimization of the collision table has generally the form of a transportation problem \\cite{Hen89a,RHF*88a}, with large cost matrices.  Classical algorithms were found to require prohibitively long computing times. Therefore an attempt was made to devise a method which would take advantage of the peculiarities of the lattice gas problem. This method then turned out to be of general applicability.  The present algorithm was developed independently of the already published studies of the transportation problem and related optimization problems.  This was not planned; it only reflects the way things happened, and the ignorance of this author who comes from a rather different field.  More will be said about this in Section~\\ref{s:conclusions}.  The paper is organized as follows. Section~\\ref{s:problem} defines the problem. In Section~\\ref{s:analog}, we describe a mechanical device which can be used as an analog computer to solve the transportation problem.  In Section~\\ref{s:num-sim} we develop an appropriate graph representation and a numerical scheme which simulates the mechanical model. This is illustrated by a detailed example in Section~\\ref{s:example}. In Section~\\ref{s:formal}, we give a rigorous definition and justification of the algorithm. Section~\\ref{s:implementation} describes some aspects of the computer implementation. In Section~\\ref{s:tests}, the algorithm is compared with the NAG library subroutine for the solution of the transportation problem. In the particular case of the assignment problem, comparisons are also made with the algorithm of Burkard and Derigs \\cite{BD80a}. A few comments are made in Section~\\ref{s:conclusions}. Finally, an Appendix derives some bounds on the number of operations. ",
        "conclusions": ""
    },
    "0209/astro-ph0209436_arXiv.txt": {
        "abstract": "We have developed a numerical model for the temporal evolution of particle and photon spectra resulting from nonthermal processes at the shock fronts formed in merging clusters of galaxies.  Fermi acceleration is approximated by injecting power-law distributions of particles during a merger event, subject to constraints on maximum particle energies. We consider synchrotron, bremsstrahlung, Compton, and Coulomb processes for the electrons, nuclear, photomeson, and Coulomb processes for the protons, and knock-on electron production during the merging process.  The broadband radio through $\\gamma$-ray emission radiated by nonthermal protons and primary and secondary electrons is calculated both during and after the merger event.  Using ROSAT observations to establish typical parameters for the matter density profile of clusters of galaxies, we find that typical merger shocks are weak and accelerate particles with relatively soft spectra. We consider the prospects for detecting nonthermal radio and $\\gamma$-ray emission from clusters of galaxies and implications for the origin of ultra-high energy cosmic rays and the diffuse $\\gamma$-ray background. Our results suggest that only a few of the isotropically-distributed unidentified EGRET sources are due to shocks formed in cluster mergers, and that only a minor contribution to the diffuse extragalactic $\\gamma$-ray background can originate from cluster merger shocks. Cluster merger shocks can accelerate protons to $\\lesssim 10^{19}$ eV for the standard parameters considered here. We predict that {\\it GLAST} will detect several cluster mergers and, depending on the mean magnetic fields in the intracluster medium, the Low Frequency Array could detect anywhere from several to several hundred. ",
        "introduction": "\\label{sec:introduction}  According to the hierarchical merging scenario, cold dark matter halos evolve to form larger structures by merging with adjacent dark matter halos.  Within dark matter halos, baryonic matter condenses to form clusters of galaxies.  A cluster merger event results from the interaction of galaxy clusters during the merger of cold dark matter halos.  The gravitational potential energy available in a cluster merger event involving halos with masses $\\sim 10^{15} \\msun$ is $\\sim\\!\\!10^{63}-10^{64}\\ergs$.  As two clusters of galaxies merge, the infall velocities can exceed the sound speed of the ICM.  As a result, a shock front will form at the interaction boundary between the clusters. First-order Fermi acceleration at the shock front produces a population of nonthermal, relativistic particles.  Relativistic electrons are detected from their synchrotron radio emission or from $\\gamma$-rays due to Compton-scattered cosmic microwave radiation. Nonthermal protons are detected through $\\gamma$ rays emitted from secondaries formed in nuclear production processes, including the $\\pi^0$ decay signature at 70\\mev.  Rich clusters contain a hot and tenuous ionized intracluster medium (ICM), with observed temperatures $T_X \\approx 5-10\\kev$, sound speeds $\\sim\\!\\!1000\\kms$, and thermal bremsstrahlung luminosities $L_{X} \\sim 10^{45}\\ergss$ between 2 and 10\\kev.  In addition to the luminous thermal component present in these clusters, there is a growing body of evidence supporting the presence of nonthermal distributions of particles in cluster mergers \\citep{eilek:99,feretti:00}.  Deep radio observations of clusters of galaxies indicate the presence of extended diffuse emission not easily associated with an optical counterpart.  These diffuse radio features are commonly classified as either {\\em radio halos}, or {\\em radio relics} (which are also called {\\em periphery halos}).  Radio halos mimic the observed X-ray profiles and are characterized by their central location in the cluster and by a highly disorganized magnetic field, and are generally thought to be a consequence of a merger event \\citep{feretti:00}.  Radio relics, usually found on the periphery of the cluster, are characterized by highly organized magnetic fields and often display filamentary structures.  We focus here on radio relics, which are thought to result from synchrotron emission emitted by electrons directly accelerated at shock fronts. By contrast, radio halos might be due to reacceleration of relic electrons by magnetic turbulence or enhanced magnetic turbulence arising from the cluster merger or motions of the galaxies \\citep{brunetti:01,ohno:02}.  The first radio halo in a cluster of galaxies was detected from the Coma Cluster \\citep{large:59}.  Many halos and relics have since been found in a number of other clusters, including A754 \\citep{kassim:01}, A2256 \\citep{berrington:02}, and others (see \\citet{govoni:01,slee:01} for recent observations). Typical radio powers from the radio halos and relics are at the level of $10^{40}$-$10^{42}\\ergss$ \\citep{giovannini:00}.  These clusters show a favorable correlation \\citep{kassim:01,berrington:02} between the existence of diffuse radio features and recent or on-going merger activity.  Detection of these features is largely due to sensitivity limitations; the number of detected diffuse radio features has increased with the improvement in sensitivity of radio telescopes.  The synchrotron evidence for nonthermal electrons in clusters of galaxies indicates that an acceleration mechanism must exist in the environment of the host cluster with sufficient power to produce the observed emission.  Several mechanisms have been proposed to explain the correlation of the radio halo and relic features and the recent or on-going merger activity.  These mechanisms often require the presence of shock waves or magnetic turbulence to accelerate particles via the Fermi acceleration process \\citep{schlickeiser:87,tribble:93,kang:97,ensslin:98b,blasi:00,miniati:01}. Other theories include adiabatic compression of fossil radio plasma by a cluster-merger shock wave \\citep{ensslin:00}.  Optical surveys show that approximately $30$-40\\% of clusters of galaxies display evidence for the presence of substructure \\citep[and others]{forman:81,gellers:82}.  This internal structure is often interpreted as a subset of galaxies merging with a larger cluster of galaxies.  These internal structures indicate velocity differences near or greater than the expected sound speed of the intracluster medium (ICM).  Observed velocities typically range from $\\sim 1000$ to 3000\\kms\\, which are consistent with values expected from parabolic orbits \\citep{oegerle:94}.  Emission from nonthermal particles will also appear at EUV and hard X-ray energies as a power-law excess.  While the EUV emission seen in the Coma cluster \\citep{lieu:99a}, and possibly also A2199 and A1795 \\citep{lieu:99b} may have a cool thermal origin, it is unlikely because of the extreme mass requirements of cool gas. It is more likely that the EUV emission has a nonthermal origin \\citep{hwang:97,ensslin:98a,mittaz:98,bc99,av00}.  Excess hard X-ray (HXR) emission has been reported from the clusters A1656 \\citep{fusco-femiano:99,rephaeli:99}, A2256 \\citep{fusco-femiano:00}, A3667 \\citep{fusco-femiano:01} and possibly A2199 \\citep{kaastra:99}.  Estimated luminosity of the HXR emission are $\\approx\\!\\!10^{43}\\ergss$.  Thermal origins of this HXR excess emission would require unrealistic temperatures greater than $40\\kev$, and so is thought to be caused by the presence of relativistic nonthermal electrons\\citep{fusco-femiano:99}.  Numerical models of merging clusters of galaxies \\citep[and others]{ricker:98,takizawa:00,ricker:01,miniati:01} have treated the development of shocks as a result of the cluster merging process.  Given the presence of thermal ionized particles in the vicinity of these shocks and a cluster magnetic field, electrons and ions will be accelerated via the first-order Fermi process \\citep{bell:87,blandford:87}.  Numerous attempts have been made to model the emissions from nonthermal particles \\citep{colafrancesco:98,fujita:01,petrosian:01,miniati:01} produced by these cluster merger shocks, but this study differs from previous attempts in that we accurately model the diffusion of particles in energy space due to Coulomb interactions, and allow for a variable injection rate that depends on the environment local to the shock front.  Recent studies by \\citet{liang:02} have addressed the balance of Coulomb losses and diffusion in energy space, but do not address the injection and diffusion of particles in energy space along with a variable source function of nonthermal particles.  In our treatment, we follow particle energies up to $10^{19}\\ev$, and calculate the bremsstrahlung, Compton, synchrotron, and $\\pi_{0}$ $\\gamma$-rays from p-p collisions. Because the shock front lifetime is a significant fraction of the age of the universe, we also follow the changing environment due, for example, to the changing CMB energy density.  In addition we accurately model nonthermal electrons and protons up to $\\sim\\!\\!10^{21}\\ev$, though we find that limitations on particle acceleration make it difficult to produce protons above $\\sim 10^{19}\\ev$ in cluster merger shocks.  From the particle distributions, we calculate nonthermal photon spectra for energies up to $\\sim\\!\\!10^{7}\\gev$.  We also include the effects of secondary production on the nonthermal photon spectra.  The physical processes and the temporal evolution of the particles are described in section \\ref{sec:models}.  The results of the simulations are presented in section \\ref{sec:results}.  A comparison of the photon spectra with observed clusters, and the potential of detecting these shocks with space-based satellite observatories is discussed in section \\ref{sec:discussion}. ",
        "conclusions": "We have presented the results of a computer code designed to calculate the nonthermal particle distributions of a shock front formed in the merger event of two clusters of galaxies.  We have calculated nonthermal particle energy spectra for primary electrons and protons, and secondary electrons.  Photon spectra were calculated for bremsstrahlung, Compton, and synchrotron processes as well as $\\pi^{0}$-decay $\\gamma$ radiation from pp collisions.  Our results apply to shocks that form at the interaction boundary between two merging clusters of galaxies, and not to cluster accretion shocks which form at the outer regions of the cluster.  Evidence for nonthermal particle acceleration in merger shocks, which penetrate into the inner region of the accreting cluster, are provided by radio halos and relics and nonthermal X-ray emission from clusters of galaxies.  The thermal X-ray bremsstrahlung was modeled from observations of luminosities, temperatures, and masses of clusters of galaxies. We modeled the nonthermal emission under the assumption that electrons and protons are accelerated with a 5\\% efficiency of the available gravitational energy that is dissipated during the course of gravitational interactions between two merging clusters of galaxies.  Particle acceleration at shocks formed between merging clusters of galaxies produces a population of nonthermal electrons that Compton-scatters CMB photons. This process naturally produces a power-law distribution of photons in the 40-80\\kev\\ energy range with luminosities consistent with HXR observations of the galaxy clusters A1656, A2256, A3667 and, possibly, A2199.  In addition, the EUV emission observed in A1656 is also consistent with a nonthermal Compton-scattered CMB origin.  Cluster magnetic field strengths $\\sim\\!\\!0.1\\mug$ are in accord with this interpretation of the observed HXR excess and EUV emission.  In contrast to the results of \\citet{totani:00}, \\citet{kawasaki:02}, and \\citet{colafrancesco:01}, we have argued that it is unlikely that more than a few of the isotropic unidentified EGRET sources can be attributed to radiation from nonthermal particles produced by cluster merger shocks.  Previous studies have assumed hard nonthermal spectra that give brighter $\\gamma$-ray emission than implied by our simulation results.  Such hard spectra can only be obtained in the infrequent merger events involving two very massive clusters, or between clusters where the dark-matter density profiles are centrally peaked.  The diffuse extragalactic $\\gamma$-ray background is a featureless power law with photon index of $2.10(\\pm 0.03)$.  The dominant nonthermal $\\gamma$ radiation components in cluster-merger shocks include Compton-scattered CMB radiation and bremsstrahlung from nonthermal electrons, and $\\pi^{0}$-decay emission from nuclear interactions involving nonthermal protons. The $\\pi^0$ decay signature will be present if the hadronic acceleration efficiency exceeds the electron acceleration efficiency, as expected in diffusive shock acceleration theory. Furthermore, spectra calculated using parameters obtained from ROSAT observations of 45 Abell clusters \\citep{wu:00} produce spectra with slopes $\\approx 2.2$-$2.4$.  Unless dark matter density profiles are centrally peaked and hadronic acceleration efficiency is low, our results imply that nonthermal emission from merging clusters of galaxies can make only a minor contribution to the diffuse extragalactic $\\gamma$-ray background. Using standard diffusive shock acceleration with mean ICM magnetic fields $\\lesssim 1 \\mu$G, our results also indicate that the merger shocks in clusters of galaxies do not accelerate $\\gtrsim 10^{19}$ eV cosmic rays, though additional effects, such as shock obliquity and the presence of preexisting particle populations, could permit higher energy acceleration by these shocks."
    },
    "0209/astro-ph0209600_arXiv.txt": {
        "abstract": "{We study the dynamics of stellar wind bubbles around hydrogen-deficient stars using numerical simulations with time- and ion dependent cooling. We consider two types of hydrogen-deficient stars, massive WR stars, producing Ring Nebulae, and low mass [WR] stars, producing Planetary Nebulae. We show that for the Planetary Nebulae, the different cooling properties of the hydrogen-deficient wind lead to a later transition from momentum- to energy-driven flow, which could explain the observed turbulence of these nebulae. We find that Ring Nebulae should all be energy-driven, and show how comparing the bubble's momentum and kinetic energy to the input wind momentum and kinetic energy, can give misleading information about the dynamics of the bubble. ",
        "introduction": "Both high and low mass stars can under certain circumstances reduce the hydrogen content of their atmospheres.  In most cases this leads to the so-called Wolf-Rayet (WR) phenomenon, i.e.\\ a dense fast wind starting below the photosphere, which produces a Wolf-Rayet spectrum, dominated by bright emission lines \\citep{AbbottConti87}.  Traditionally most attention was given to the massive WR stars, but the last ten years their lower mass cousins, the [WR] stars have been studied in more detail \\citep[see e.g.][for a review]{Koesterke01}.  The winds from [WR] and WR stars produce stellar wind bubbles (SWBs) in their environment. In the case of WR stars they are called Ring Nebulae (RNe), in the case of [WR] stars, Planetary Nebulae (PNe). PNe also form around H-rich central stars, so [WR] stars constitute a subgroup among central stars of PNe. Approximately 7\\% of central stars is estimated to be [WR], the rest being H-rich \\citep{Gorny2001} (with the exception of so-called weak emission line stars or wels which appear to be H-poor without showing the WR phenomenon). All central stars are considered to be in the same evolutionary phase, namely the post-AGB phase, where the [WR] have changed their atmospheric abundances through a timely thermal pulse \\citep{Herwig2001}. The existence of two different groups of central stars suggests that a comparison between the two could be interesting.  In contrast, WR stars are thought to be a phase in the evolution of most stars with Zero Age Main Sequence masses higher than $\\sim 25$~M$_\\odot$, and there is no class of H-rich stars in an equivalent evolutionary stage. Massive stars lose mass already on the Main Sequence, followed by a slower wind when the star moves over to the red part of the Hertzsprung--Russell diagram, and finally leading to the WR wind. This leads to a whole series of interactions between the wind phases \\citep[see e.g.][] {GuileNorbert1}. The more complicated environments and probably also the clumpiness of the actual winds make that the RNe are mostly irregular and filamentary, lacking the overall symmetries found in PNe.  In this paper we investigate whether the fact that the winds from WR and [WR] stars are H-poor changes the dynamics of their SWBs. In the case of the [WR] stars this is relevant because we can compare the PNe between the H-rich and H-poor central stars. In the case of the WR stars this is relevant because it has been suggested that even at high wind velocities their SWBs can be strongly cooling.  The layout of the paper is as follows. The effects of cooling on SWBs are outlined in Sect.~2.  We investigate the effects of WR winds using numerical hydrodynamic models with detailed cooling, described in Sect.~3. Section 4 and 5 contain the results of the simulations for PNe and RNe, respectively. We discuss these results further in Sect.~6 and sum up the conclusions in Sect.~7. ",
        "conclusions": "We simulated the effects H-deficient winds have on their SWBs, studying simplified cases for PNe and RNe. We find that the extreme abundances in the winds of [WC] stars can keep their PNe momentum-driven for a longer time. We speculate that this leads to more turbulent nebulae and would also produce more aspherical PNe if their shape was mostly due to an aspherical post-AGB wind.  For the RNe around massive WR stars, we showed that they cannot be momentum-driven, despite earlier claims to the opposite. We pointed out some of the difficulties related to the deriving the momentum- versus energy-driven character of the RNe using the comparisons of RN momentum and kinetic energy to the assumed total input of momentum and kinetic energy by the stellar wind.  The models in this article illustrate the physical effects of abundances on the structure of SWBs. To be realistic models for PNe and RNe, they should be improved in several ways. As shown for instance in \\citet{RHPNIII}, ionization fronts can play an important role in the early evolution of PNe. For the RNe adding photo-ionization to the models could help in estimating the amount of neutral material, which is one of the unknowns in the $\\pi$-$\\epsilon$ method. Letting their winds evolve according to a more realistic recipe such as in \\citet{RHPNIII} for PNe, and in \\citet{GuileNorbert2} for RNe, would be another step in the direction of more realistic models. However, given that the results in this paper show that for RNe the wind abundances play only a minor role, the models of \\citet{GuileNorbert1} and \\citet{GuileNorbert2} remain largely valid, and only for the case of the [WR]-PNe it makes sense to pursue more realistic models."
    },
    "0209/astro-ph0209570_arXiv.txt": {
        "abstract": "We present ATCA radio observations of the giant radio galaxy J0116$-$473 at 12 and 22 cm wavelengths in total intensity and polarization. The images clearly reveal a bright  inner-double structure within more extended edge-brightened lobe emission. The lack of hotspots at the ends of the outer lobes, the strong core and the inner-double structure with its edge-brightened morphology lead us to suggest that this giant radio galaxy is undergoing a renewed nuclear activity: J0116$-$473 appears to be a striking example of a radio galaxy where a young double source is evolving within older lobe material. We also report the detection of a Mpc-long linear feature which is oriented perpendicular to the radio axis and has a high fractional polarization. ",
        "introduction": "The concept of episodic activity in radio galaxies, with each phase manifesting itself as an extended radio structure, was inherent in the models suggested for sources with X-shaped structures and powerful radio galaxies with wings \\citep{leahy84}.  Restarting beams following an interruption in nuclear activity was again suggested as a cause for source structures which appeared to have partial jets \\citep{bridle86}. This idea gained support from the observations of 3C388 by \\citet{roettiger94} in which the lobe spectral index distribution revealed two distinct regions. Observational indications coupled with simulations of the development of extended radio structures have suggested that episodic activity may play an important role in the evolution of at least some categories of radio sources \\citep{baum90,clarke91}.  In a study of the morphologies of a sample of giant radio galaxies, \\citet{subrahmanyan96} drew attention to a variety of morphological features which were suggestive of interrupted nuclear activity. Recently, the WENSS discovered several giant radio galaxies exhibiting double-double morphologies which have been attributed to renewed nuclear activity \\citep{schoenmakers00b} and the study is indicative of a higher incidence rate of such inner doubles among large radio galaxies. As a consequence, systematic studies of the role of episodic nuclear activity are possible for the category of giant radio galaxies. The relatively long timescales ($\\sim 10^{8}$ yr; Komissarov and Gubanov 1994) over which radio lobes remain visible after the central activity which energizes them stops makes synchrotron lobes useful indicators of any past activity phases in radio galaxies.  Such studies have interesting implications for the fuelling of the central engine and the conditions under which a renewal of nuclear activity may occur. Moreover, these studies may address the question of the role of such recurrent activity in the attainment of the extraordinary sizes in the giant radio sources.  J0116$-$473 was previously imaged as part of a study of the morphologies in radio galaxies of megaparsec dimensions \\citep{subrahmanyan96}. This object was noted as exhibiting unusual characteristics: lack of hotspots in a source which had properties consistent with FR-{\\sc II} type \\citep{fanaroff74} radio galaxies and an elongated structure extending along a direction perpendicular to the jets. It was hypothesized that the morphological features in this giant radio galaxy --- and indeed some others in the sample --- might be a manifestation of recurrent nuclear activity.  \\citet{subrahmanyan96} concluded that the large sizes of giant radio galaxies may be a result of a restarting of their central engines in multiple phases of activity along roughly similar directions.  We are currently following up on our earlier hypothesis with case studies of the giant sources which showed evidence of recurrence in activity. In this paper, we present higher dynamic range Australia Telescope Compact Array (ATCA; see The Australia Telescope 1992) images of J0116$-$473  in total intensity and polarization. The ATCA observations presented here provide new evidence for a restarting of activity in this giant source.  In the next section we describe our observations. In later sections we discuss the role of recurrence in the creation of the unusual morphological features in this source. We postpone comparison of the source features and parameters with  other sources exhibiting similar inner double structures to a later paper. ",
        "conclusions": "We have presented 12 and 22 cm total-intensity and polarization ATCA observations of the giant radio galaxy J0116$-$473. The observations were carried out with the purpose of following up on intriguing aspects noted for this galaxy in an earlier work. Our new higher dynamic range observations have provided much support for our earlier hypothesis of interrupted nuclear activity in this source. The inner double structure located within the much larger diffuse lobes is argued to be a pair of new lobes formed as a result of renewed activity in the core. The observations show this inner double in detail revealing its edge brightened morphology, its symmetric location about the core, and a narrow jet.  The 1-Mpc long bar-like feature close to the core is seen to be highly polarized with fractional polarization as high as 50 per cent all along its length and with projected magnetic field vectors oriented along its length. We discuss a possible origin for this feature, suggesting it to be a result of earlier activity. Additionally, we note the presence of two other unusual features seen in this source, a band of low rotation measure along the length of the source and a step in the rotation measure situated towards the southern inner lobe. We briefly discuss possible causes for these features."
    },
    "0209/astro-ph0209093_arXiv.txt": {
        "abstract": "The equation of state of the hypothetical dark energy component, which constitutes about two thirds of the critical density of the universe, may be very different from that of a cosmological constant. Employing a phenomenological model, we investigate semi-analytically the constraints imposed on the scalar quintessence by supernovae observations, and by the acoustic scale extracted from recent CMB data. We show that a universe with a quintessence-dominated phase in the dark age is consistent with the current observational constraints. ",
        "introduction": "Recent astrophysical and cosmological observations such as dynamical mass, Type Ia supernovae (SNe), gravitational lensing, and cosmic microwave background (CMB) anisotropies, concordantly prevail a spatially flat universe containing a mixture of matter and a dominant smooth component, which provides a repulsive force to accelerate the cosmic expansion~\\cite{ctrig}. The simplest candidate for this invisible component carrying a sufficiently large negative pressure is a true cosmological constant. The current data, however, are consistent with a somewhat broader diversity of such a repulsive ``dark energy'' as long as its equation of state approaches that of the cosmological constant at recent epoch. A dynamically evolving scalar field $\\phi$ called ``quintessence'' (Q) is probably the most popular scenario so far to accommodate the dark energy component. It is very interesting and fundamentally important to distinguish the Q field from the true cosmological constant case.  Many efforts have been put forth to reconstruct the scalar potential $V(\\phi)$ from observational data based on various reasonable physical motivations. They include pseudo Nambu-Goldstone boson, inverse power law, exponential, tracking characteristics, oscillating feature, and others~\\cite{qmods}. Several attempts have been made to test different Q-models~\\cite{many}. Nevertheless, it proves to be premature at this stage to perform a meaningful data fitting to a particular quintessence model, or to differentiate between the variations. Reconstruction of $V(\\phi)$ would likely require next-generation observations.  Since the scalar potential of the Q-field is scarcely known, it is convenient to discuss the evolution of $\\phi$ through its equation of state, $p_\\phi = w_\\phi \\rho_\\phi$. Physically, $-1\\le w_\\phi\\le 1$, where the former equality holds for a pure vacuum state. Lately some progress has been made in constraining the behavior of Q field from observational data. A combined large scale structure, SNe, and CMB analysis has set an upper limit on Q models with a constant $w_\\phi < -0.7$~\\cite{bond,bean}, and a more recent analysis of CMB observations gives $w_\\phi=-0.82^{+0.14}_{-0.11}$~\\cite{bac}. Furthermore, the SNe data and measurements of the position of the acoustic peaks in the CMB anisotropy spectrum have been used to put a constraint on the present $w_\\phi^0 \\le -0.96$~\\cite{cope}. The apparent brightness of the farthest SN observed to date, SN1997ff at redshift $z\\sim 1.7$, is consistent with that expected in the decelerating phase of the flat $\\Lambda$CDM model with $\\Omega_\\Lambda \\sim 0.7$~\\cite{riess}, inferring $w_\\phi= -1$ for $z<1.7$.  Given the above observational constraints, one would like to know the possible role played by the Q field in the early universe. If the dark energy is a pure cosmological constant or it has a constant equation of state of $\\lesssim -0.7$, it would be quickly dominated by the matter for $z>0.3$. The situation may be very different for quintessence with a time-dependent $w_\\phi$. Here we will adopt a model-independent approach in which a phenomenological form for the time-dependent $w_\\phi$ is assumed and then used to unfold the dynamics of the Q field up to the epoch of the last scattering surface. In particular, we focus on the CMB constraints that apply to such a generic quintessence (GQ) scenario in the hope of distinguishing the Q field from the true cosmological constant and the constant equation of state. ",
        "conclusions": "We have investigated the evolution of the quintessence allowed by the observational constraints from CMB and SNe, using a semi-analytic method with a simple square-wave function for the time-varying equation of state. Although the true equation of state, if there is any, may be a complicated function of time, the square-wave should roughly capture the generic feature of the evolution of the quintessence. This generic quintessence model is sufficient for us to confront the current observational data. Future high-precision data will tighten the constraints to this model and we may even need a more sophisticated model to parametrize the physics of the Q component. Also, the present method gives more physical insights and is much simpler though less accurate than the numerically intensive maximum likelihood analysis of CMB data (see, e.g., Ref.~\\cite{hansen}).  Three extreme GQ models have been presented. Figure~\\ref{bg1} shows the maximum dynamics that the Q field can attain at low redshifts for $z>2$. The evolving Q field during the large-scale-struture formation may have interesting cosmological implications. For instance, the authors in Ref.~\\cite{qpmf02} have attempted to generate primordial magnetic fields from the dynamics of the Q field coupled to electromagnetism. This electromagnetic Q field may also be responsible for the time-varying fine structure constant ($\\alpha$)~\\cite{alpha} as it was recently claimed that the results of a search for time variability of $\\alpha$ using absorption systems in the spectra of distant quasars yield a smaller $\\alpha$ in the past~\\cite{webb}. We have also studied the constraint from the growth of large-scale matter perturbations on the GQ model. Figure~\\ref{bg3} shows that the Q component can make up about $40\\%$ of the total energy density of the universe at last scattering. This result is consistent with the upper bound $\\Omega_\\phi < 0.39$ during the radiation dominated epoch obtained by performing a maximum likelihood analysis on the CMB data~\\cite{hansen}. In general, the GQ scenario bears a salient feature that the Q component overwhelms the matter during the dark age. It is worth studying in more details about its influence on the evolution of matter perturbations and the subsequent structure formation. At last, we would like to point out that an acceleration of the universe in the past is consistent with all observations. So the fact that the universe is accelerating today would not be quite unnatural."
    },
    "0209/astro-ph0209278_arXiv.txt": {
        "abstract": "{We present the first results of a study of the expected properties  of the first stellar generations in the Universe. In particular, we consider and discuss a series of properties that, on the basis of the emission from associated HII regions,  permit one to discern {\\it bona fide} primeval stellar generations from the ones formed after pollution from supernova explosions. The expected performance of NGST for the study and the characterization of primordial sources is also discussed.}     \\addkeyword{Cosmology: early universe} \\addkeyword{Cosmology: observations} \\addkeyword{Galaxies: abundance} \\addkeyword{Galaxies: starburst} \\addkeyword{H~II regions} \\begin{document} ",
        "introduction": " ",
        "conclusions": "We have considered and discussed a series of properties that, on the basis of the emission from associated HII regions,  permit one to discern {\\it bona fide} primeval stellar generations from the ones formed after pollution from supernova explosions. We find that it is possible to discern truly primordial populations from the next generation of stars by measuring the metallicity of high-z star forming objects. The very low background of NGST will enable it to image and study first-light sources at very high redshifts, whereas its relatively small collecting area olimits its capability in obtaining spectra of z$\\sim$10--15 first-light sources to either the bright end of their luminosity function or to strongly lensed sources."
    },
    "0209/gr-qc0209009_arXiv.txt": {
        "abstract": "We have studied the problem of all sky search in reference to continuous gravitational wave particularly for such sources whose wave-form are known in advance. We have made an analysis of the number of templates required for matched filter analysis as applicable to these sources. We have employed the concept of {\\it fitting factor\\/} {\\it (FF)\\/}; treating the source location as the parameters of the signal manifold and have studied the matching of the signal with templates corresponding to different source locations. We have investigated the variation of FF with source location and have noticed a symmetry in template parameters, $\\theta_T$ and $\\phi_T$. It has been found that the two different template values in source location, each in $\\theta_T$ and $\\phi_T$, have same {\\it FF\\/}. We have also computed the number of templates required assuming the noise power spectral density $S_n(f)$ to be flat. It is observed that higher {\\it FF\\/} requires exponentially increasing large number of templates. ",
        "introduction": "\\indent Gravitational wave (GW) Laser Interferometer antennas are essentially omni - directional with their response better than 50\\% of the average over 75\\% of the whole sky (Grishchuk et al., 2000). Hence the data analysis systems will have to carry out all sky searches for its sources. We know that the amplitude of intense GW believed bathing the earth is very small, as compared to the sensitivity of GW detectors, and is further masked by the dominant noise. In these circumstances, continuous gravitational wave (CGW) sources are of prime importance because for such sources we can achieve enhanced signal-to-noise ratio (SNR) by investigating longer observation data set. However, a long observation time introduces modulation effects, arising due to the relative motion of the detector and the source. As a consequence, there results redistribution of power in the forest of side bands resulting into the reduction of the expected power due to amplitude modulation (AM). The problem of all sky search gains another dimension in view of the fact that there are reasons to believe the presence of intense GW sources whose locations and even frequencies are not known. Amongst such sources pulsars occupy an important position. Similar to all sky search one will also have to do all frequency search. All sky all frequency search is the holy grail of gravitation pulsar astronomy. In this paper we confine ourselves to the problem of all sky search.  \\par Search of CGW without a priori knowledge appears to be computationally quite demanding even by the standard computers expected to be available in the near future. For example, in the case of bandwidth $10^3$ Hz, observation time $10^7$ sec. and star's minimum decay time  of $100$ years one would require $10^{14}\\, Tflops$ computer (Frasca, 2000). Very fast computer and large memories with ample amount of disk space seems inevitable. However, a choice of optimal data processing and a clever programming is also integral part of a solution to this problem. Amongst these the pre-correction of time series due to the Doppler modulation before the data is processed may be a method, which will reduce the computational requirements. In reference to this, Schutz (1991) has introduced the concept of patch in the sky as the region of space throughout which the required Doppler correction remains the same. He has also demonstrated that the number of patches required for $10^7$ sec. observation data set and one KHz signal would be about $1.3 \\times 10^{13}$. However, the size of the patch would also depend on the data analysis technique being employed.  \\par Matched filtering is the most suitable technique for the detection of signals from sources viz., pulsars whose wave form is known. The wave forms are used to construct a bank of templates, which represent the expected signal wave form with all possible ranges of its parameters. The time of arrival, source location, frequency of the signal, ellipticity of the source and its spin down represent important parameters of GW emitted by a pulsar. For detection of GW we check if the cross correlation of the templates with the corresponding data set exceeds the preassigned threshold. We introduce in the next section the criterion of the {\\it fitting factor (FF) \\/} (Apostolatos, 1995) applicable to such analysis. We consider the source location as parameters of the signal manifold and investigate the matching of the waveforms corresponding to different source locations. In section 3 we compute the number of templates required for all sky search. A discussion of the results is provided in the section 4. ",
        "conclusions": "\\label{sec:concl5} In view of the complexity of the FT, which contains trignometric as well as Bessel functions; one has to be careful in computing $FF$. We have found useful to employ the Romberg integration using Pad\\'e approximation. We have used (i) QROMO of numerical recipes instead of QROMB as the former takes care of singularities, and (ii) RATINT routine for Pad\\'e approximation.  \\par We  have noticed marked symmetries in all sky search in both $\\theta$ and $\\phi$ space for one day observation time. It has been found that the two different template values in source location, each in $\\theta_T$ and $\\phi_T$, have same {\\it FF\\/}. Accordingly, computation burden will be reduced by a factor of four. However, it is not clear whether this symmetry property can be established analytically as well. The source location, because of these symmetries is uncertain and some other analysis is to be adopted for getting the exact location. We have computed the number of templates assuming the noise power spectral density $S_n(f)$ to be flat which is justified as the bandwidth is extremely narrow.  \\par The issues of optimum template parameterization and placement, and the related computational burden have been discussed in the literature by several authors notably by Sathyaprakash and Dhurandhar (1991), Dhurandhar and Sathyaprakash (1994), Owen (1996), Apostolatos (1995, 1996), Mohanty and Dhurandhar (1996), Mohanty (1998), Owen and Sathayaprakash (1999). The question of possible efficient interpolated representation of the correlators is a problem of current interest and remains still unsolved."
    },
    "0209/astro-ph0209087_arXiv.txt": {
        "abstract": "We present \\xmm\\ observations of the luminous star \\objectname[]{$\\eta$~Carinae}, including a high resolution soft X-ray spectrum of the surrounding nebula obtained with the Reflection Grating Spectrometer. The EPIC image of the field around \\ec\\ shows many early-type stars and diffuse emission from hot, shocked gas. The EPIC spectrum of the star is similar to that observed in previous X-ray observations, and requires two temperature components. The RGS spectrum of the surrounding nebula shows K-shell emission lines from hydrogen- and helium-like nitrogen and neon and L-shell lines from iron, but little or no emission from oxygen. The observed emission lines are not consistent with a single temperature, but the range of temperatures observed is not large, spanning $\\sim\\,0.15\\,-\\,0.6$ keV. We obtain upper limits for oxygen line emission and derive a lower limit of $\\mathrm{N/O} > 9$. This is consistent with previous abundance determinations for the ejecta of \\ec, and with theoretical models for the evolution of massive, rotating stars. ",
        "introduction": "The massive, luminous star \\objectname[]{$\\eta$~Carinae} is famous for an extended outburst beginning in 1843, during which it temporarily became the second brightest star in the sky. This outburst gave rise to a bipolar optical nebula, obscuring the star from direct observation. \\ec\\ is thought to be very massive ($M \\sim 100~\\mathrm{M_\\odot}$), and to lose mass at a rate of $\\dot{M} \\sim 10^{-3} \\, \\mathrm{M_\\odot~yr^{-1}}$. For a general review of its history and properties, see \\cite*{dav97}.  {\\it Einstein} observations of \\ec\\ showed it to be a complex X-ray source \\citep{sew79,sew82,chl84}. \\ec\\ has two X-ray emission components: hard, absorbed ($N_{\\mathrm{H}} \\sim 5 \\, \\times 10^{22}\\, \\mathrm{cm^{-2}}$), spatially unresolved emission coming from the star, and soft, extended emission coming from the nebula around the star. The {\\it Einstein} observations also showed that there are many other X-ray sources in the field around \\ec, and that there is diffuse X-ray emission with an extent of about a degree.  {\\it Ginga} observations found evidence for iron K-shell emission from \\ec\\ consistent with Fe XXV, indicating a thermal origin for the hard X-ray emission \\citep{koy90}. \\cite{cor95} used {\\it ROSAT} PSPC observations to show that the hard X-ray emission is variable.  {\\it ASCA} observations obtained much higher quality spectra, and found evidence for a very strong \\ion{N}{7} Ly~$\\alpha$ feature, which was thought to result from the supersolar abundance of nitrogen in the ejecta \\citep{tsu97,cor98}. This also was consistent with previous optical and UV spectroscopic observations of the ejecta around \\ec\\ \\citep{dav82,dav86}.  Recent \\ch\\ ACIS-I imaging observations have resolved \\ec\\ spatially at the subarcsecond scale \\citep{sew01}. The soft X-ray nebula shows complex structure with several knots of X-ray emission. \\ch\\ HETGS observations have given us the first high resolution X-ray spectrum of the star, showing that the hard emission is non-isothermal, with emission lines from H- and He-like iron, calcium, argon, sulfur, and silicon \\citep{cor01}.  In this paper, we report the results of \\xmm\\ observations of \\ec, including the high resolution soft X-ray spectrum obtained with the Reflection Grating Spectrometer (RGS) \\citep{dh01}. Until now, no X-ray observatory has been able to obtain high resolution soft X-ray spectra of extended sources. RGS has a spectral resolution of about 0.1 \\AA\\ for the $\\sim$ 1' nebula of \\ec, or $\\frac{\\lambda}{\\Delta\\lambda} \\sim 200$ at 20 \\AA. This is important in the case of \\ec, because we can study the physical state of the X-ray nebula in detail, and obtain much more accurate elemental abundance measurements than with a CCD spectrometer. We also present the EPIC image of the field and the CCD spectrum of \\ec. ",
        "conclusions": "\\label{dis}  There are two main results from the analysis of the RGS spectrum of \\ec. The first is a constraint on the range of temperatures in the nebula ($0.15-0.6\\,\\mathrm{keV}$), which allows us to infer shock velocities for the expansion of the ejecta into the surrounding medium. The second is a lower limit on the nitrogen to oxygen abundance ratio ($\\mathrm{N/O} > 9$). This allows us to constrain the evolution of \\ec.  \\subsection{Temperature distribution}  The upper end of the temperature distribution is strongly constrained by the Fe L-shell spectrum. The lack of measureable emission from charge states higher than \\ion{Fe}{18} rules out the presence of appreciable quantities of gas above $\\sim\\,0.6\\,\\mathrm{keV}$. The lower end of the distribution appears to be flat, based on the emission from \\ion{Ne}{9}, \\ion{N}{7} and \\ion{N}{6}. However, there are no other potentially observable spectral lines originating from ions that exist at lower temperatures than \\ion{N}{6}, so the emission measure distribution cannot be constrained below about 0.2 keV. \\cite*{dav82,dav86} find UV emission lines from \\ion{N}{1} through \\ion{N}{5} in the spectra of the ejecta, so there is certainly a range of temperatures present. As noted in the previous section, a substantial difference between the assumed absorption and the real absorption could change the overall shape of the emission measure distribution, especially at low temperatures.  \\cite*{wei01a} attempt to correlate the observed projected velocity of optical blobs which are spatially coincident with X-ray emission using Hubble Space Telescope and ROSAT HRI images. The velocities they find in the brightest X-ray regions would produce plasma at temperatures an order of magnitude higher than observed, assuming that the ejecta was colliding with a stationary ISM. Of course, \\ec\\ should be surrounded by a wind blown bubble out to much larger radii than 0.3~pc (the radius of the X-ray nebula), and the material inside the bubble should be streaming outward. It seems likely that the observed shock temperature reflects the velocity at which the X-ray emitting ejecta are overtaking the previously emitted stellar wind. The temperature range $0.15-0.6\\,\\mathrm{keV}$ implies a shock velocity range of $300-700\\,\\mathrm{km\\,s^{-1}}$. If the X-ray emitting ejecta date from the great eruption of 1843, then the rough expansion velocity for a free expansion is $\\sim\\,\\mathrm{0.3\\,pc\\,/\\,150\\,yr\\,=\\,2000\\,km\\,s^{-1}}$, so the velocity of the stellar wind before the great eruption was $\\sim\\,\\mathrm{1500\\,km\\,s^{-1}}$.  \\subsection{Abundance measurements}  The N/O ratio observed in the ejecta has implications for the evolution of \\ec. It is clearly a signature of CNO processing, and the degree of conversion of oxygen to nitrogen observed in the ejecta is high.  All massive main-sequence stars burn hydrogen on the CNO cycle. Its nucleosynthetic signatures are the conversion of most of the catalytic carbon and oxygen to nitrogen, and the burning of H into He. For CNO processed material to be observed on the surface of these stars, or in their ejecta, it must be transported there from the core.  Previous measurements of N/O in \\ec\\ give similar but generally less constraining results than our RGS measurements. Optical and UV spectroscopy of the S condensation (corresponding spatially roughly to the brightest X-ray knot \\citep{sew01}) shows that most CNO is nitrogen and that the helium mass fraction is $0.40\\pm0.03$ \\citep{dav82,dav86}. A quantitative measurement of the CNO abundance ratios is not made because of their dependence on ionization and thermal structure, and also because some oxygen and carbon may be in solid grains. It should be noted that the measured value of the helium mass fraction may be systematically too low if the ionization balance of helium was not properly modelled.  More recent measurements of the abundances in the S condensation have been made by \\cite{duf97} with HST-FOS. They report CNO and He abundances for the S2 and S3 ``sub-condensations'', respectively, of $\\mathrm{[N/O]}>1.72,1.75$, $\\mathrm{[N/C]}>1.95,1.85$, and $Y = 0.39,0.42$. They did detect weak oxygen and carbon lines, but treated them as upper limits due to potential contamination from the foreground \\ion{H}{2} region. However, they also find that preliminary analysis of the S1 and S4 sub-condensation spectra show much lower N and He enrichment, with correspondingly lower N/O and N/C ratios.  Previous X-ray observations \\citep{tsu97,cor98,sew01,wei01b} have shown the presence of a strong \\ion{N}{7} Ly~$\\alpha$ feature in the spectrum, but the CCD spectra lacked the resolution to strongly constrain the \\ion{O}{7} and \\ion{O}{8} features. Our measurement of N/O is not limited by the spectral resolution of RGS, but rather by source/background contamination, and the observed line strength is not influenced by the formation of dust grains or large uncertainties in the temperature distribution of the plasma.  Recent HST-STIS long-slit spectroscopy of the central star have obtained a lower limit of $\\mathrm{N/O}\\gax 1$ \\citep{hil01}. This is a conservative interpretation of the data; the lower limit could easily be taken to be an order of magnitude higher. On the other hand, UV spectra taken with HST-GHRS show evidence for moderate carbon depletion which may be inconsistent with the level of depletion found in the ejecta \\citep{lam98}. In light of recent work indicating that \\ec\\ may be a binary system \\citep{dam96,dam00}, the apparent contradiction in the stellar and nebular abundances is taken to be an indication that the star producing the carbon features is actually the secondary (assuming the star that produced the nebula is the primary). \\cite{wal99} points out that there are several difficulties with this conclusion, the most obvious being the concealment of the luminous blue variable (LBV) primary.  Spectroscopic measurements of the abundances of the central star cannot invalidate the RGS abundance measurements, but the binary scenario requires us to treat the nebular abundances with some care. It is unlikely that both members of a binary system could contribute substantially to the ejecta around \\ec, but it is possible, in principle, that the ejecta from the primary could mix with the wind of the secondary. If both stars had a high N/O ratio, but substantially different helium abundances, then it would be possible to misinterpret the significance of the nebular abundances. However, this is not a likely scenario, so we make the simplest assumption, which is that the observed nebular abundances reflect the current surface abundances of the primary.  The signatures of CNO processing have been observed in various types of hot stars, including OBN stars, blue supergiants, and LBVs \\citep{mae95}. However, CNO processed material is not observed on the surface of all hot stars, and the amount of processed material observed spans a wide range. The fact that N/O is so high in the ejecta of \\ec\\ is strongly constraining, regardless of the mechanism responsible for mixing.  The two most plausible mechanisms which could have resulted in the measured abundances in the ejecta of \\ec\\ are : 1.) \\ec\\ is on the main sequence and is rotating. This rotation has caused very thorough mixing. 2.) \\ec\\ is in a post-red-supergiant blue supergiant phase, and the CNO abundance ratios are a result of the onset of convection in the envelope during the red supergiant phase. We refer in particular to the discussion in \\cite{lam01}, which deals with the same question in the case of other LBV nebulae. We can use the measured N/O ratio in conjunction with the He abundance of \\cite{dav86} to assess the plausibility of these two mechanisms.  Using Figure 3 of \\cite{lam01} for the case of an $85\\, \\mathrm{M_{\\odot}}$ star with $\\mathrm{Z}\\,=\\,0.02$, we find that for $\\mathrm{log(N/O)}>1.0$, $\\mathrm{log(He/H)}>-0.3$, or $\\mathrm{Y}>0.67$. This simply reflects the fact that although a high surface ratio of N/O can be obtained in the red supergiant phase, this can only happen if the star lost enough of its envelope on the main sequence to allow core processed material to dominate the resulting composition. This value of Y is not consistent with the \\cite{dav86} measurement of $\\mathrm{Y}\\,=\\,0.4$, although a conservative assessment of the possible errors, particularly in measuring the helium mass fraction, does not allow us to rule out that \\ec\\ could be a post-red-supergiant object.  \\cite{mey00} make predictions for the abundances of rotating massive stars. Their $\\mathrm{Z}\\,=\\,0.02$ model with an initial rotation velocity of $300\\,\\mathrm{km\\,s^{-1}}$ and a mass of $120\\,\\mathrm{M_{\\odot}}$ predicts $\\mathrm{Y_{s}}=0.89$ and $\\mathrm{N/O}=45.4$ at the end of H-burning. While this value of Y is also not consistent with the observed value, in this case Y will clearly be lower earlier in the life of the star, whereas if the mixing is efficient enough, N/O will already be high enough to be consistent with the measured lower limit. This is an important point; the conversion of oxygen to nitrogen in CNO burning is considerably slower than the conversion of carbon to nitrogen. If rotational mixing is responsible for the observed abundances, the mixing timescale must be short compared to the evolutionary timescale. As pointed out in \\cite{mae87}, the ratio of the mixing timescale to the main sequence lifetime in rapidly rotating stars is indeed expected to decrease with increasing mass.  \\subsection{Summary}  We have analyzed XMM-Newton X-ray spectra of \\ec. The EPIC spectral data from the star are consistent with past observations by {\\it ASCA} and \\ch. The data are not consistent with an isothermal plasma, but require at least two temperatures.  The RGS spectra show that the nebula is nonisothermal and has strongly non-solar CNO abundances. The temperature range in the nebula is $0.15\\,-\\,0.6\\,\\mathrm{keV}$. If this is interpreted as a shock velocity, it corresponds to $300\\,-\\,700\\,\\mathrm{km\\,s^{-1}}$. We find a lower limit of $\\mathrm{N/O}\\,>\\,9$, which is indicative of very thorough mixing in the envelope of \\ec. Taken with previous measurements of the surface helium abundance $\\mathrm{Y}\\,=\\,0.4$, this implies that \\ec\\ is a main-sequence object with some strong mixing mechanism at work, although it does not decisively rule out the possibility that it is a post-red-supergiant object."
    },
    "0209/astro-ph0209564_arXiv.txt": {
        "abstract": "We have measured the central stellar velocity dispersions of 33 nearby spiral and elliptical galaxies, using a straightforward template-fitting algorithm operating in the pixel domain.  The spectra, obtained with the Double Spectrograph at Palomar Observatory, cover both the Ca triplet and the \\mgb\\ region, and we present a comparison of the velocity dispersion measurements from these two spectral regions.  Model fits to the Ca triplet region generally yield good results with little sensitivity to the choice of template star.  In contrast, the \\mgb\\ region is more sensitive to template mismatch and to details of the fitting procedure such as the order of a polynomial used to match the continuum shape of the template to the object.  As a consequence of the correlation of the [Mg/Fe] ratio with velocity dispersion, it is difficult to obtain a satisfactory model fit to the \\mgb\\ lines and the surrounding Fe blends simultaneously, particularly for giant elliptical galaxies with large velocity dispersions.  We demonstrate that if the metallicities of the galaxy and template star are not well matched, then direct template-fitting results are improved if the \\mgb\\ lines themselves are excluded from the fit and the velocity dispersion is determined from the surrounding weaker lines. ",
        "introduction": "The recent discovery of a tight correlation between stellar velocity dispersion and black hole mass \\citep[the \\msigma\\ relation;][]{fm00, geb00} has placed new emphasis on the importance of accurate velocity dispersion measurements for the central regions of nearby galaxies. Since black hole mass is approximately proportional to $\\sigma^4$, even modest errors in $\\sigma$ for galaxies with black hole mass measurements can have a substantial impact on the correlation \\citep{tre02}.  Use of the predictive power of the \\msigma\\ relation to obtain estimates of the masses of black holes in galaxy nuclei also relies on the accuracy of the velocity dispersion measurements.  In view of these issues, it is worthwhile to examine the level of agreement between velocity dispersion measurements obtained with different techniques and from different spectral regions, to determine the methods that are most likely to yield accurate results.  This paper presents an examination of a simple direct template-fitting technique operating in the wavelength domain.  Initially, our main goal was to measure velocity dispersions for the sample of nearby galaxies observed with \\chandra\\ by \\citet{ho01}, so that estimates of the black hole masses in these galaxies could be derived by applying the \\msigma\\ relation.  We were able to observe most of the galaxies from this sample during two observing runs.  We also observed some other nearby galaxies for which no previous velocity dispersion measurements were available, a few velocity dispersion ``standard'' galaxies from \\citet{mce95} for comparison, and several low-redshift BL Lac objects for which the results have been reported separately \\citep{bhs02a,bhs02b}.  Here, we present velocity dispersions for 33 nearby galaxies and a comparison of measurements obtained from the Ca triplet and \\mgb\\ spectral regions.  We also discuss some systematic issues relevant to the application of the direct template-fitting method.  In particular, we demonstrate that model fits to the \\mgb\\ spectral region are sensitive to the [Mg/Fe] abundance ratio, and that the results are usually improved if the \\mgb\\ lines themselves are \\emph{excluded} from the fitting region used to determine the velocity dispersion. ",
        "conclusions": "\\subsection{Comparison of blue and red side results}  Table 3 lists the velocity dispersions measured from the blue and red side spectra.  For three galaxies (NGC 404, NGC 660, and NGC 6503), the velocity dispersion was too small to be measured from the blue side data.  We did not attempt to fit models to the blue spectrum of Arp 102B as it contains a number of weak emission lines in this region.  In general, we consider the red side results to be more reliable for the reasons described above.  One simple measure of the relative degree of template mismatch between the red and blue sides is the goodness of fit as measured by \\chisqdof.  The mean value of \\chisqdof\\ for all galaxies is 1.12 on the red side, and 3.03 on the blue side.  Given that the same LOSVD model was used for the red and blue side data, this difference in \\chisqdof\\ clearly demonstrates the larger degree of template mismatch on the blue side.  In addition to template mismatch issues, the red measurements are aided by higher spectral resolution.  The low spectral resolution of the blue side data obtained with the 600 line grating is clearly reflected in the large error bars for these measurements, in comparison with the data obtained with the 1200 line grating.  To determine whether the red and blue results are in agreement within their uncertainties, we compute the statistic $\\delta = (\\sigma_B - \\sigma_R) / (\\epsilon_B^2 + \\epsilon_R^2)^{1/2}$.  If $|\\delta| \\leq 1$, then the difference between the red and blue measurements is consistent with zero, and the two measurements are considered to be in agreement.  Figure \\ref{brcompare} shows the results of this test. Out of 29 galaxies with blue and red measurements, 15 (or 52\\%) agree within the estimated $1\\sigma$ uncertainties.  The worst disagreements are at the $3\\sigma$ level.  This suggests that the measurement uncertainties may be somewhat underestimated, particularly for the blue side data.  The systematic uncertainty in the choice of the polynomial order for the blue measurements may be largely to blame for this situation.  As described above, there appears to be an additional uncertainty of roughly 5\\% in the blue side measurements due to the choice of polynomial order, in addition to the uncertainty determined by the fitting routine.  If we add this 5\\% uncertainty in quadrature to the blue side measurement uncertainties, the agreement between the red and blue sides appears more satisfactory, with $|\\delta| \\leq 1$ for 19 of 29 galaxies, or 65\\% of the sample.  Thus, we conclude that the blue side uncertainties are systematically too small and should be increased by adding $0.05\\sigma_B$ in quadrature to the uncertainties listed in Table 3.  Increasing the error bars by this amount leads to a satisfactory comparison between the red and blue measurements, giving roughly the level of agreement that would be expected from $1\\sigma$ uncertainties in the case of Gaussian statistics.  It is still somewhat surprising how large the disagreement is between the blue and red data for a few of the galaxies, but in some cases (such as NGC 4569) the discrepancy could be ascribed to a poor template match due to the presence of a young starburst component.  Figure \\ref{mg} demonstrates the results obtained if the \\mgb\\ lines are included in the fitting region.  For comparison, we performed measurements of our blue side data using the entire wavelength range 5040--5430 \\AA, except for a small 20 \\AA\\ window centered on the [\\ion{N}{1}] emission line at 5200 \\AA.  Including the Mg lines in the fit nearly always leads to a significant increase in \\chisqdof\\ because of the poor match between the [Mg/Fe] ratios of the templates and the galaxies.  The outcome is that, for all but one galaxy, the velocity dispersions measured by including \\mgb\\ in the fitting region are larger than those obtained with our default fitting region.  The discrepancy increases systematically as a function of $\\sigma$ due to the correlation between [Mg/Fe] and $\\sigma$, and is as bad as 25--30\\% for galaxies with large $\\sigma$.  The large increase in \\chisqdof\\ clearly demonstrates that template stars of near-solar metallicity should not be used to fit the Mg and Fe lines of galaxy spectra simultaneously.  This template matching problem appears to affect the measurements over the entire range in $\\sigma$, not just galaxies with large velocity dispersions.  Even for galaxies with $\\sigma\\approx100$ \\kms, the model fits are still severely degraded in quality by including \\mgb, and the velocity dispersions are affected at the $\\sim5-10\\%$ level.  The difficulty in matching the \\mgb\\ line strength is a problem for the direct-fitting method in particular, because the widths and the depths of the absorption lines are coupled together in the calculation of \\chisq.  Methods for measurement of velocity dispersions that operate in the Fourier domain may be less sensitive to variations in line strength for individual lines.  The sensitivity of the Fourier quotient method to metallicity variations in the \\mgb\\ region has been examined by \\citet{ll85}, who found that the derived dispersions have a modest dependence on [Fe/H]; it would be worthwhile to perform similar tests for various other measurement techniques, using galaxies and template stars with a range of [Mg/Fe] ratios.  The [Mg/Fe] mismatch problem can be seen in some previous kinematic studies, in cases where velocity dispersion measurements were performed using a small wavelength region containing both \\mgb\\ and Fe 5270.  For example, some of the model fits shown by \\citet{rw92} and \\citet{km93} appear to underpredict the strength of the \\mgb\\ lines relative to Fe 5270.  In kinematic studies that attempt to derive the shape of the LOSVD by methods operating in the pixel domain, it is important to be aware of this [Mg/Fe] mismatch problem, because it does affect the velocity dispersions (as shown in Figure \\ref{mg}) and there is the possibility that it could affect the shape of the derived velocity profile as well.  As Figure \\ref{bfits} demonstrates, this problem can be largely avoided by shifting the fitting region redward to cover the Fe 5270 and Fe 5335 blends, and excluding \\mgb.  \\subsection{Comparison with previous results}  Velocity dispersions have been reported previously for all but 6 of the galaxies in our sample.  Two of the most commonly used references for velocity dispersions measurements are the compilation by \\citet{mce95}, which lists averages of measurements from the literature for each galaxy, and the online Hypercat database \\citep{pru98}, which lists all previous measurements of $\\sigma$ and also computes an average value for each galaxy.\\footnote{The Hypercat database is available at http://www-obs.univ-lyon1.fr/hypercat .} Figure \\ref{litcompare} shows a comparison of our results with the average values reported by \\citet{mce95} and by Hypercat.  The measurements compiled by McElroy are corrected to a standard aperture of $2\\arcsec\\times4\\arcsec$, very close to the aperture size we used, so the results should be directly comparable.  Hypercat also applies correction factors to homogenize the measurements to a consistent aperture size prior to computing the average values.  In general, the Hypercat mean results agree more closely with our red side results than do the McElroy average values.  This is primarily due to the fact that the Hypercat catalog is more up-to-date and contains a larger number of measurements for some galaxies, so the Hypercat averages are sometimes less affected by individual discrepant measurements.  For the 20 galaxies in common between our sample, \\citet{mce95}, and Hypercat, the RMS deviation between the catalog mean results and our red side measurements is 28 \\kms\\ for McElroy, and 20 \\kms\\ for Hypercat.  Since most of the galaxies in our sample are spirals, some portion of this scatter must be due to different slit position angles, which would lead to different amounts of rotational broadening in the extracted spectrum.  The averages computed by these catalogs often include quite discrepant measurements taken from different literature sources.  The case of NGC 3627 serves as a useful example.  Hypercat lists two sources for $\\sigma$ that disagree by far more than their quoted uncertainties: $117\\pm9$ \\kms\\ \\citep{hs98}, and $184\\pm19$ \\kms\\ \\citep{wks79}.  Our result ($124 \\pm 3$ \\kms) agrees well with \\citet{hs98} but is significantly lower than that of \\citet{wks79}.  Similarly, for NGC 4826, \\citet{wks79} find $\\sigma=160$ \\kms\\ while two other sources listed in Hypercat give 90 and 113 \\kms, closer to our red side measurement of 96 \\kms.  These examples demonstrate that the average values listed by Hypercat, and by McElroy, should be viewed with a great deal of caution, because the average dispersions for galaxies with a small number of measurements can be badly influenced by a single discrepant value.  There appears to be a systematic problem with the measurements of \\citet{wks79} in particular.  There are 8 galaxies in common between our sample and \\citet{wks79}, and in only one case (NGC 4374) is their velocity dispersion smaller than our result.  On average their measurements are larger than ours by 25\\%.  The sytematic offset of this one source contributes a nonnegligible amount to the overall disagreement between our results and the average literature results.  The worst apparent disagreement between our results and the mean literature data is for NGC 1058, a late-type spiral galaxy with a compact central stellar cluster for which we find $\\sigma = 31 \\pm 6$ \\kms.  The mean velocity dispersion is given by \\citet{mce95} as 60 \\kms, and by Hypercat as 59 \\kms.  However, according to Hypercat, the sole previous measurement is from unpublished data of Whitmore \\& Rubin (1985), and according to Hypercat the original measurement was actually an \\emph{upper limit} of 60 \\kms.  This example serves as a reminder that when velocity dispersions for individual galaxies are taken from the literature, it is probably safer to consult the original sources than to rely on averaged results reported in these compilations."
    },
    "0209/astro-ph0209614_arXiv.txt": {
        "abstract": "A description is given of the samples of Low Surface Brightness galaxies (LSBs) used for comparison with models of their chemical and spectro-photometric evolution (Boissier et al., this Volume). These samples show the large variation and scatter in observed global properties of LSBs, some of which cannot be modeled without adding starbursts or truncations to their star formation history. ",
        "introduction": "In the past few decades, the existence has been shown of galaxies with a blue central disc surface brightness well below the Freeman value of $\\mu_{B,0}$=21.65 mag arcsec$^{-2})$, which is typical of the average previously catalogued ``classical'' High Surface Brightness  spirals (HSBs). Particularily in the last decade a considerable body of observational data has been collected on this class of Low Surface Brightness galaxies (LSBs), which may turn out to be crucial to studies of galaxy formation and evolution and of the `cosmic' chemical evolution,  if (as suggested by O'Neil \\& Bothun 2000) they represent the majority of galaxies.  Although there is no unambiguous definition of an LSB galaxy, we adopted as the limit between HSBs and LSBs a blue central disc surface brightness value of $\\mu_{B,0}$=22 mag arcsec$^{-2}$, a commonly used criterion.  Though LSBs show a large variety of observed properties (see below), their obvious interest has motivated various theoretical studies towards an understanding of their nature. In a recent series of papers, the chemical and spectro-photometric evolution of ``classical'' HSB spiral galaxies was computed (for various galactic masses and angular momentum) and compared quite successfully to various types of observational quantities like colours, spectrum, star formation efficiency and gas fraction as well as to their Tully-Fisher relationship (Boissier \\& Prantzos 1999,2000; Boissier et al. 2001; see also Prantzos et al.,  this Volume). These models have now been extended to cover LSBs (Boissier et al. 2002, and this Volume), based on the suggestion of Dalcanton et al. (1997) that LSBs have a larger angular momentum than HSBs, as used also by Jimenez et al. (1998) in their models. ",
        "conclusions": ""
    },
    "0209/astro-ph0209108_arXiv.txt": {
        "abstract": "We report the results of a radial velocity survey of planetary nebulae (PNe) located in the tidal features of the well-known interacting system NGC~5194/95 (M51).  We find clear kinematic evidence that M51's northwestern tidal debris consists of two discrete structures which overlap in projection -- NGC~5195's own tidal tail, and diffuse material stripped from NGC~5194.  We compare these kinematic data to a new numerical simulation of the M51 system, and show that the data are consistent with the classic ``single passage'' model for the encounter, with a parabolic satellite trajectory and a 2:1 mass ratio.   We also comment on the spectra of two unusual objects: a high-velocity PN which may be associated with NGC~5194's halo, and a possible interloping high-redshift galaxy. ",
        "introduction": "The ``Whirlpool Galaxy,'' M51 (NGC~5194, 5195)\\footnote{Throughout this paper, we use the designation M51 when considering both galaxies as a complete system; when referencing the system's individual components, we use NGC~5194 and NGC~5195.} is probably the most famous of all interacting galaxy systems.  As a nearby system \\citep[$d = 8.4 \\pm 0.6$~Mpc;][hereafter FCJ]{fcj} with grand-design spiral morphology (Rosse 1845; see Rosse 1880), distorted outer isophotes \\citep{zwicky59, burk78}, and an apparent bridge-like feature between the primary, NGC~5194, and the secondary, NGC~5195, M51 has been extensively studied as an example of tidally induced spiral structure \\citep[\\eg][]{tully74, scoville83, rots90, zaritsky93}. In fact, the wealth and detail of the observational data has made M51 a favorite target for dynamical modeling, starting with the seminal work of \\citet{tt72}. In the \\citet{tt72} study, many of the tidal features of the M51 system were explained by a parabolic encounter of two galaxies with mass ratio of 3:1 viewed shortly after the initial collision. Since then, in response to the ever-increasing amount of observational data on the system, a number of alternate scenarios have been proposed \\citep{toomre78, howard1990, hern90, barnes98, salo00}.  Despite M51's long history of dynamical modeling, significant uncertainties in the basic description of the system remain.  While the original \\citet{tt72} study proposed a very recent ($\\sim 100$ Myr) collision, the discovery of M51's long H~I tidal tail \\citep{rots90} shifted the preferred solution to somewhat later times (several hundred Myr past the initial collision) in order to give the tail more time to develop \\citep{hern90}.  More recently, \\citet{salo00} have suggested that a multiple passage model might be more appropriate for the system.  Such a scenario appears to do a better job of explaining NGC 5194's H~I velocity field, although the predicted structure for the H~I tidal tail is more complex than is observed.  Furthermore, the simulations of \\citet{salo00} used rigid halo models, which do not self-consistently follow the orbital evolution of the system. Because the multiple passage model relies on orbital decay to provide the proper second passage, the lack of a self-consistent solution remains a concern for these models.  As a result, these models have not followed the full dynamical response of the system \\citep{salo00}.  Consequently, no single scenario satisfactorily explains all of the system's observational data \\citep[see the discussion in][]{barnes98}.  One reason for the continuing uncertainty about the M51 system is the lack of kinematic information for the companion galaxy.  Unlike NGC~5194, NGC~5195 contains no neutral hydrogen, so the only kinematic data we have on NGC~5195 comes from measurements of the stellar kinematics of the system's inner disk \\citep{schweizer77}.  Since tidal kinematics provide strong constraints for dynamical models of interacting galaxies \\citep[\\eg][]{hibbard95}, this dearth of information at large radii is a significant stumbling block for unraveling the evolutionary history of the system.  In principle, there is another dynamical tracer which can reveal the kinematic structure of M51's tidal debris --- planetary nebulae (PNe).  Since PNe are a normal and common phase of stellar evolution, their spatial distribution and kinematics closely follows that of the stellar component as a whole.   As a result, surveys for PNe can trace the distribution of stars to lower surface densities than is possible with diffuse light.  More importantly, PNe are extremely luminous emission-line sources.  At the distance of M51 ($d=8.4 \\pm 0.6$~Mpc; FCJ), PNe surveys with 4-m class telescopes can reach $\\sim 2$~mag down the planetary nebula luminosity function, and the velocity of each PN can be measured to $\\sim 10$~\\kms\\ accuracy.  This makes PNe uniquely useful as a kinematical probe of diffuse tidal structures, such as those found in the M51 system.  In 1997, FCJ surveyed M51 for planetary nebulae in order to obtain a distance to the system via the planetary nebula luminosity function. This survey found a substantial number of PNe directly west-southwest of NGC~5195 in a tidal tail-like structure.  At first, this discovery was a bit of a surprise, since the deep broadband images of \\citet{burk78} place the western tail of NGC 5195 more to the northwest, and not at the location of these planetaries. However, numerical models \\citep[\\eg Toomre 1994, as reported by][]{barnes98, salo00} do predict the presence of tidal material from NGC~5194 in the region where the planetaries are located.  In order to study this feature in more detail, and to provide kinematic data on the diffuse tidal structures surrounding M51, we have conducted a radial velocity survey of a significant fraction of the FCJ planetary nebula sample. The ultimate goal of these observations is to provide a more complete description of the kinematics of the M51 system, and test whether the kinematics of NGC~5195's tidal features are consistent with the extant models. Interestingly, our PN velocities reveal significant kinematic substructure in the diffuse material to the west of NGC~5195; this fact, combined with differences in the spatial distribution of the region's PNe and diffuse light, implies that the observed tidal tail consists of two distinct but overlapping features. We interpret these data in the light of published models and our own new N-body model.  The outline of our paper is as follows: in \\S2 we detail the original imaging and follow-up spectroscopic observations.  In \\S3, we describe our data reduction and the determination of the planetary nebula velocities.  In \\S 4 and 5, we describe the kinematic structure of M51's western tidal tail, and compare the data to a new simulation which follows the response of both NGC~5194 and its companion.   In \\S 6, we describe two unusual objects whose properties are significantly different from the bulk of the planetaries, and discuss their implications. Finally, in \\S7, we summarize our results. ",
        "conclusions": "We have presented the velocities of 36 PNe in the interacting galaxy system M51.  The planetaries have a velocity structure that is clearly multi-modal and complex, especially in the tidal debris west of the companion galaxy NGC~5195. In this region, we find two distinct kinematic components, consistent with the idea that we are viewing two tidal features -- one from each galaxy -- cospatially in projection.  Such features have been predicted by numerical simulations \\citep[\\eg][]{salo00}, but the component from NGC~5194 has remained undetected until now.  The PN kinematics do not yet discriminate between the various dynamical scenarios for the M51 system: both single- and multiple-passage models predict multiple kinematic features in this region. We do, however, note that the kinematic {\\it coldness\\/} of these features may be difficult to reproduce in a multiple passage scenario, where the galaxies have experienced several perturbative events. Self-consistent dynamical models of the multiple passage scenario are needed to test this effect.  We expand on the previous simulation of \\citet{hern90} and present a new single-passage model that reasonably reproduces the morphology and kinematics of the system's tidal debris.  How can we better constrain the dynamical evolution of the M51 system? Observationally, the FCJ survey of M51 probed only the first 1.2~mag of the planetary nebulae luminosity function.  The sample of 64~PNe found in this survey could easily be increased by a factor of $\\sim 3$ by going deeper with current 4~m and 8~m class telescopes.   There are roughly 10 PNe in an 18 arcmin$^{-2}$ area in the FCJ sample of the northwestern tidal tail.   Assuming roughly $\\sim 20 \\times 10^{-9}$~PNe per bolometric solar luminosity (2.5 mag down the PNLF) for a stellar population \\citep[\\eg][]{c95}, and a distance of 8.4 Mpc, this PNe density translates to a stellar surface brightness of $\\mu_V \\sim 24.5$ mag arcsec$^{-2}$.   Allowing for a 3-fold increase in the number of PNe detected in deeper surveys, and adopting a (conservative) minimum of 10 PNe in a given region to derive useful kinematic information, PNe could be used to probe surface brightnesses as faint as $\\mu_V\\sim 26$ mag arcsec$^{-2}$.  The kinematics of such low-surface brightness features cannot be measured in any other way.  In addition to a deeper M51 PNe survey, more information on the secondary, NGC~5195, would be extremely helpful.  In particular, integral-field spectroscopy \\citep[\\eg][]{and1999} of the galaxy's stellar component could dramatically improve our constraints on galactic rotation and inclination angle.   Finally, better theoretical models are needed to predict the detailed tidal kinematics probed by the PNe. This is particularly true for the multiple-passage models, for which no fully self-consistent solution has yet been developed."
    },
    "0209/astro-ph0209422_arXiv.txt": {
        "abstract": "Using data from more than ten-years of observations with the Akeno Giant Air Shower Array (AGASA), we published a result that the energy spectrum of ultra-high energy cosmic rays extends beyond the cutoff energy predicted by Greisen \\cite{gzk_g}, and Zatsepin and Kuzmin \\cite{gzk_zk}. In this paper, we reevaluate the energy determination method used for AGASA events with respect to the lateral distribution of shower particles, their attenuation with zenith angle, shower front structure, delayed particles observed far from the core and other factors. The currently assigned energies of AGASA events have an accuracy of $\\pm$25\\% in event-reconstruction resolution and $\\pm$18\\% in systematic errors around 10$^{20}$eV. This systematic uncertainty is independent of primary energy above 10$^{19}$eV. Based on the energy spectrum from 10$^{14.5}$eV to a few times 10$^{20}$eV determined at Akeno, there are surely events above 10$^{20}$eV and the energy spectrum extends up to a few times 10$^{20}$eV without a GZK-cutoff. ",
        "introduction": "From ten-years of data collected by the Akeno Giant Air Shower Array (AGASA), we have shown that the energy spectrum of primary cosmic rays extends up to a few times 10$^{20}$eV without the expected GZK cutoff \\cite{takeda98a}. On the other hand, the HiRes collaboration has recently claimed that the GZK cutoff may be present with their exposure being similar to AGASA \\cite{HiResICRC2001}. Ave et al. \\cite{ave01a} have re-analyzed the Haverah Park events and their energies are reduced by about 30\\% using a new energy conversion formula. Although we have already published our statistical and systematic errors in the energy determination in related papers \\cite{takeda98a,hayashida94a,yoshida94a,yoshida95a}, it is now quite important to reevaluate uncertainties in the energy determination of AGASA events with the accumulated data of ten years. The uncertainties due to shower front structure and delayed particles far from a shower core are also evaluated and described in some detail.   The AGASA array is the largest operating surface array, covering an area of about 100km$^{2}$ and consisting of 111 surface detectors of 2.2m$^{2}$ area \\cite{teshima86a,chiba92a}. Each surface detector is situated with a nearest-neighbor separation of about 1km  and the detectors are sequentially connected with pairs of optical fibers. All detectors are controlled at detector sites with their own CPU and through rapid communication with a central computer. In the early stage of our experiment AGASA was divided into four sub-arrays called ``branches'' for topographical reasons, and air showers were observed independently in each branch. The data acquisition system of AGASA was improved and the four branches were unified into a single detection system in December 1995 \\cite{ohoka97a}. After this improvement the array has operated in a quite stable manner with a duty cycle of about 95\\%, while the duty cycle before unification was 89\\%.  In a widely spread surface array like AGASA, the local density of charged particles at a specific distance from the shower axis is well established as an energy estimator \\cite{hillas71a} since the local density of the electromagnetic component depends weakly on variations in interaction models, fluctuations in shower development and primary mass. In the AGASA experiment, we adopt the local density at 600m, $S(600)$, which is determined by fitting a lateral distribution function (LDF) of observed particle densities to an empirical formula \\cite{yoshida94a}. This empirical formula is found to be valid for EAS with energies up to 10$^{20}$eV and for zenith angles smaller than 45$^{\\circ}$ \\cite{doi95a,sakaki99a}. The relation for converting $S(600)$ to primary energy has been evaluated so far by Monte Carlo simulations \\cite{dai88a} up to 10$^{19}$eV and is \\begin{equation} E = 2.03 \\times 10^{17} \\cdot S_{0}(600) \\hspace{1em} \\mbox{eV} \\hspace{2em}, \\label{eq:econv} \\end{equation} where $S_{0}(600)$ is the $S(600)$ value per m$^{2}$ for a vertically incident shower. This conversion relation is derived from electron components for air showers observed 900m above sea level. In \\S \\ref{ssect:energy}, a new conversion constant is evaluated taking account of the average altitude. More modern simulation codes have been used in this simulation.  In the southeast corner of AGASA there is the Akeno 1km$^2$ array \\cite{hara79a}.  This is a densely packed array of detectors covering an area of 1km$^2$ operated since 1979.  This array was used to determine the energy spectrum between 10$^{14.5}$eV and 10$^{18.5}$eV.  In this experiment, the total number of electrons, known as the shower size $N_e$, was used as an energy estimator.  The relation between this energy spectrum and the AGASA energy spectrum is discussed in \\S \\ref{sect:spectrum}. ",
        "conclusions": "We have reevaluated the uncertainties in energy estimation using data accumulated over ten years. Table \\ref{tbl:systematics} summarizes the major systematics and uncertainties in energy estimation. Here, the symbol ``$+$'' means that currently assigned energies should be pushed up under a particular effect, and the symbol ``$-$'' represents a shift in the opposite direction. The probable overestimation of 10\\% due to  shower front structure and delayed particles may be compensated for by the probable underestimation of the energy conversion factor by 10\\%, an effect resulting from the inclusion of the average altitude of AGASA and the proper definition of what is meant by a ``single particle''. Adding uncertainties in quadrature, the systematic uncertainty in energy determination in the AGASA experiment is estimated to be $\\pm$18\\% in total. Therefore, the currently assigned energies of the AGASA events have an accuracy of $\\pm$25\\% in event-reconstruction resolution and $\\pm$18\\% in systematics.  It should be noted that the Akeno-AGASA spectra cover over five decades in energy, connecting smoothly from the {\\it knee} to a few times 10$^{20}$eV, except for a 10\\% difference in energy in the 10$^{19}$eV region. This may be due to the difference in the energy conversion relations for the experiments and is within the systematic errors evaluated here. It is concluded that there are surely events above 10$^{20}$eV and the energy spectrum extends up to a few times 10$^{20}$eV. The present highest energy event may only be limited by exposure. The next generation of experiments with much larger exposures are highly anticipated."
    },
    "0209/hep-ph0209267_arXiv.txt": {
        "abstract": "{\\normalsize We examine the prospects for using lead as a supernova-neutrino detector by considering the spectrum of electrons produced, and the number of one- and two-neutron events.  We show that the electron energy spectrum from charged-current reactions can be used to extract information about the high-temperature component of the neutrino spectrum. Some degree of electron neutrino oscillation is expected in the supernova envelope. We examine the prospects for untangling the signatures of various oscillation scenarios, including, e.g.\\ normal or inverted hierarchies, and different values for the small mixing angle $\\theta_{13}$.} ",
        "introduction": "\\label{sec:intro}  The idea of detecting supernova neutrinos is exciting because although galactic supernovae are rare, we could potentially learn a great deal from the neutrinos.  Measuring their spectra would give us information about the mass of the proto-neutron star core and its equation of state, and provide input for supernova explosion calculations.  In addition, we can add to what we've learned about neutrino oscillations from solar, atmospheric, reactor, and accelerator neutrinos.  In this paper, we analyze the capabilities of a detector based on lead \\cite{omnis,land} that can see both electrons from charged-current neutrino interactions and neutrons from charged- or neutral-current interactions, e.g.\\ a detector containing lead perchlorate \\cite{elliott}.  Supernova-neutrino detection differs from solar-neutrino detection. Supernova neutrinos have a higher average energy, and an intense flux for a very short period, $\\sim 10 \\, {\\rm s}$.  Calculations of neutrino diffusion in the proto-neutron star show that neutrinos emitted as $\\mu$ and $\\tau$ types have higher average energies, $\\langle E_{\\nu_{\\mu,\\tau}} \\rangle = 16 - 25 \\, {\\rm MeV}$, than those emitted as electron or anti-electron types, which have $\\langle E_{\\nu_e}\\rangle \\sim 11 \\, {\\rm MeV}$ and $\\langle E_{\\bar{\\nu}_e} \\rangle \\sim 13 \\, {\\rm MeV}$ respectively.  The energy distributions can be fit to Fermi-Dirac spectra, although some calculations show that they differ from Fermi-Dirac form on their tails \\cite{totani}.  This has been discussed extensively in \\cite{raffelt}. The interesting supernova physics conveyed by neutrinos lies in the details of the energy distributions, acquired as the neutrinos are emitted from the supernova core.  Neutrino oscillations will mix the spectra in the outer envelope of the supernova.  Recent data from the Sudbury Neutrino Observatory (SNO) and Superkamiokande (SuperK) have demonstrated that neutrinos oscillate, implying that they have mass. SuperK indicates that atmospheric $\\mu$ neutrinos oscillate into objects that are not electron neutrinos, with $\\delta m^2_{\\rm atmos} \\approx 3 \\times 10^{-3} {\\rm eV}^2$ and a mixing angle of $\\sin^2 2 \\theta_{\\rm atmos} \\approx 1$ \\cite{SuperK}.  Although it has long been known that the flux of electron neutrinos from the sun is smaller than expected, SNO has very recently used its sensitivity to neutral-current scattering to measure the total flux from all species of neutrinos.  The flux is approximately the same as predicted by the standard solar model \\cite{bahcall}, confirming that the electron neutrinos are oscillating into some combination of $\\mu$ and $\\tau$ neutrinos \\cite{SNO1}.  The signal favors the large-mixing-angle (LMA) solution to the solar-neutrino problem, corresponding to mixing parameters $\\delta m^2_{\\rm solar} \\sim 10^{-5} {\\rm eV}^2$ and $\\sin^2 2 \\theta_{\\rm solar} \\approx 0.8$.  The LMA will be tested by the reactor experiment KamLAND \\cite{kamland}.  If all these results are cast in the form of 3-neutrino mixing, then there is still an unknown mixing angle, which is currently limited by reactor neutrino data to be \\\\ $\\sin^2 2 \\theta_{\\rm reactor} \\raisebox{-.25ex}{$\\stackrel{<}{\\scriptstyle \\sim}$} 0.1$ \\cite{chooz,paloverde}.  The LSND experiment \\cite{lsnd} complicates the picture, requiring a singlet neutrino or CPT violation in the neutrino sector \\cite{cptv}. The oscillation parameters inferred from the solar and atmospheric results imply that neutrinos will change flavor in the outer envelope of a supernova.  The details of the transformation depend on {\\it a)} the unknown mixing angle $\\theta_{\\rm reactor}$ and {\\it b)} whether there are more than three species of neutrino, as implied by the combination of atmospheric, solar and LSND results.  Several existing facilities can detect supernova neutrinos.  If a supernova exploded 10 kpc from the earth with a luminosity of $3 \\times 10^{53}$ ergs and the energy partitioned equally among neutrino flavors, SuperK would see about 8300 events from $\\bar{\\nu}_e + p \\rightarrow n + e^-$ in its water detector \\cite{beacom1}. KamLAND would see about 330 events from the same reaction in its scintillator, and SNO would see about 360 such events in its light water component \\cite{beacom2}.  KamLAND may be able to measure the spectrum of the high-temperature neutrinos through neutral-current neutrino-proton elastic scattering \\cite{beacom3}. SNO would also be able to see 80 events from $\\nu_e$ charged-current break-up of the deuteron. This number would increase in the presence of oscillations.  An additional 500 events in SNO would come from the neutral-current break-up of the deuteron \\cite{vogel}, which has a low threshold. An analysis of the supernova signal in a water-Cherenkov detector and in a heavy-water detector has been conducted by in ref. \\cite{dutta}. An analysis of neutrino mass limits from supernova time of flight is done in ref. \\cite{beacom4}.  Here we consider a detector based on, e.g.\\ lead perchlorate (${\\rm Pb(ClO_4)_2}$), first proposed by Elliott \\cite{elliott}.  Lead had been considered previously as a supernova detector by several groups, e.g.\\ OMNIS \\cite{omnis} and LAND \\cite{land}.  Lead has an attractively large neutrino-scattering cross section per nucleon compared with other elements, and most of the scattering events produce neutrons \\cite{fhm,kl,volpe}. Lead perchlorate, which would be sensitive mainly to the higher-energy neutrinos, has the appealing ability to measure energy deposited by the electrons produced in charged-current reactions in coincidence with zero, one, two, or more neutrons, as well as to measure the number of neutrons emitted (in isolation) in neutral current reactions.  Per kt of lead perchlorate, for the supernova described above, there would be about 378 electrons produced with one neutron and about 234 with two neutrons, if the oscillations are induced only by the large solar mixing angle and the high-energy neutrinos have a Fermi-Dirac distribution with temperature $T = 8.0~{\\rm MeV},$ and effective chemical potential $\\eta = 0$ (see eqn.\\ (\\ref{phi2})). In the neutral-current channel, there would be about 105 one-neutron events and about 72 two-neutron events. These numbers come from the calculations described below.  In what follows, we investigate the supernova-neutrino signal in lead perchlorate in more detail.  We discuss how the electrons produced by the charged-current scattering on lead can be used to obtain spectral information on the high-temperature neutrinos.  We then examine the possibility that by using this information together with a comparison of the numbers of charged- and neutral-current events, one could distinguish among the oscillation scenarios discussed below.  In section \\ref{sec:oscillations} we discuss the particulars of the neutrino oscillations and what they would mean for the spectrum of electron neutrinos coming from a supernova. In section \\ref{sec:calc} we describe our calculations of the cross sections for neutrino-induced spallation of neutrons from lead via the charged and neutral currents.  We compare our results to previous calculations and discuss uncertainties.  In section \\ref{sec:results} we present the results of our calculations and show how they may be used to learn about the spectra and oscillation of supernova neutrinos. Section \\ref{sec:conclusions} is a conclusion. ",
        "conclusions": "\\label{sec:conclusions}  In this paper, we have analyzed the properties of a lead-based supernova-neutrino detector. The expected energy distribution of electrons emitted with one or two neutrons peaks somewhere in the low 20's of MeV, depending on the details of the incoming neutrino spectral shape. We expect a few hundred charged- and neutral-current events per kt of lead.  We have used the Random-Phase-Approximation with effective Skyrme forces to calculate neutrino-lead cross sections.  Our results  are in agreement with those of refs.\\ \\cite{kl} and \\cite{volpe}, but lower than the ones in ref.\\ \\cite{fhm}, for reasons we understand. We used two different Skyrme forces, SIII and a parameterization we call SkO+ based on the force SkO'; the average electron energies obtained with these two interactions are within a few percent of each other. Despite this good agreement, it is difficult to assess quantitatively the overall systematic uncertainties in our cross sections. We don't know how much forbidden strength, which is particularly important in the two-neutron channel, is quenched, or even much about how it's distributed.  It's possible, however, to improve the calculations, and additional data from neutrino scattering and electron scattering would help tremendously.  The spectra of neutrinos reaching the earth can be modified by oscillations.  We have argued that in most scenarios, the electron energy spectrum can be used to determine the temperature (and perhaps the effective chemical potential) of the hot neutrino spectrum, due to neutrinos that were originally emitted as $\\nu_\\mu$ or $\\nu_\\tau$. (We find that lead has little sensitivity to the cold neutrinos originally emitted as $\\nu_e$, or, in the charged-current channel, to antineutrinos.) With a $\\pm 5\\%$ theoretical uncertainty in the average energies of the electrons, there would be an approximately 0.5 MeV uncertainty in the temperature of the hot neutrino spectrum.  This doubles if the uncertainty is $\\pm 10\\%$. These numbers are small enough so that a lead detector would provide crucial information about the proto-neutron star.  We discussed the possibility of distinguishing between the complete and partial neutrino transformation by using the ratio of charged-current to neutral-current event numbers.  Although the idea works in principle, it would be hard to use now because of the uncertainty in calculated cross sections. However, it should be possible to distinguish between the case of no transformation and either partial or complete transformation.  Since some degree of transformation is expected, either due to the LMA or $\\theta_{13}$ this ability will provide an important check of our understanding of neutrino physics.  A measurement of neutrino cross sections on lead is very important. As discussed above, we need a reasonable degree of certainty in the average energies to extract information about the incoming neutrino spectrum. That certainty is even more important if we want to see whether some or all of the original electron neutrinos have oscillated. The ability to distinguish between the two would let us use supernovae to learn something about the size of ${\\theta_{13}}$ or discover whether nature has chosen the normal or inverted hierarchy.   \\vspace{1.cm} We wish to thank S. Elliott, J. Beacom, A. Murphy and D. Boyd for useful discussions.  Two of us (G.C.M. and C.V.) acknowledge the European Centre for Theoretical Studies in Nuclear Physics and Related Areas (ECT*). G.C.M. is supported by the U.S. Department of Energy under grant DE-FG02-02ER41216, and J.E. by the U.S. Department of Energy under grant DE-FG02-97ER41019."
    },
    "0209/astro-ph0209344_arXiv.txt": {
        "abstract": "We have observed the hot post-AGB star ZNG~1 in the globular cluster M5 with the {\\it Far Ultraviolet Spectroscopic Explorer (FUSE).} From the resulting spectrum, we derive an effective temperature $T_{\\rm eff} \\sim 45,000$ K, a rotational velocity $v_{\\rm rot} \\sim 100$ km s$^{-1}$, carbon and nitrogen abundances approximately ten times solar, a wind velocity $v_{\\infty} \\sim 1000$ km s$^{-1}$, and evidence for an expanding shell of material around the star.  The carbon and nitrogen enhancements suggest dredge-up of nuclear-processed material on the AGB.  The high rotational velocity may reflect a previous merger with a binary companion. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209491_arXiv.txt": {
        "abstract": "We have simulated the interferometric observation of the Cosmic Microwave Background (CMB) temperature and polarization fluctuations. We have constructed data pipelines from the time-ordered raw visibility samples to the CMB power spectra which utilize the methods of data compression, maximum likelihood analysis, and optimal subspace filtering. They are customized for three observational strategies, such as the single pointing, the mosaicking, and the drift-scanning. For each strategy, derived are the optimal strategy parameters that yield band power estimates with minimum uncertainty. The results are general and can be applied to any close-packed array on a single platform such as the CBI and the forthcoming AMiBA experiments.  We have also studied the effect of rotation of the array platform on the band power correlation by simulating the CBI single pointing observation. It is found that the band power anti-correlations can be reduced by rotating the platform and thus densely sampling the visibility plane. This enables us to increase the resolution of the power spectrum in the $\\ell$-space down to the limit of the sampling theorem ($\\Delta\\ell = 226 \\approx \\pi / \\theta$), which is narrower by a factor of about $\\sqrt{2}$ than the resolution limit ($\\Delta\\ell \\approx 300$) used in the recent CBI single pointing observation. The validity of this idea is demonstrated for a two-element interferometer that samples visibilities uniformly in the $uv$-annulus.  From the fact that the visibilities are the Fourier modes of the CMB field convolved with the beam, a fast unbiased estimator (FUE) of the CMB power spectra is developed and tested. It is shown that the FUE gives results very close to those from the quadratic estimator method without requiring large computer resources even though uncertainties in the results increase. ",
        "introduction": "It is theoretically expected that the CMB is polarized. The CMB quadrupole anisotropy causes CMB photons polarized by Thomson scattering with electrons at the last scattering surface ($z \\simeq 1,100$) and during the reionization epoch ($z \\lesssim 7$) (Hu \\& White 1997). The amplitude of polarization is predicted to be 1 -- 10\\% of that of the temperature anisotropies, depending on angular scales. The CMB polarization can provide useful information that is not much contained in the temperature anisotropy, such as the epoch of reionization or the tensor perturbations. Recently Keating et al. (2001) have reported the result of the POLAR experiment, giving upper limit of order 10 $\\uK$ on the large-scale CMB polarization. The PIQUE experiment (Hedman et al. 2001, 2002) have also obtained a similar upper limit at subdegree scales. De Oliveira-Costa et al. (2002) have tried to measure the cross-correlation $C^{TE}_{\\ell}$ between temperature and $E$-mode polarization power spectra by cross-correlating the PIQUE and Saskatoon data (Netterfield et al. 1997). Many single dish and interferometry experiments such as MAXIPOL, POLAR, Polatron, COMPOSAR, CMB RoPE, DASI, and CBI are on-going to detect these faint polarized signals of CMB origin. Very recently, the DASI has reported a detection of the CMB $E$-mode polarization and the $TE$ cross-correlation by differencing the CMB fluctuations in two fields in 271 days of observation (Leitch et al. 2002b; Kovac et al. 2002).  Since the Cambridge Anisotropy Telescope has detected the anisotropy on subdegree scales (CAT; O'Sullivan et al. 1995; Scott et al. 1996), the Degree Angular Scale Interferometer (DASI; Leitch et al. 2002a; Halverson et al. 2002), the Cosmic Background Imager (CBI\\footnote{http://www.astro.caltech.edu/\\~{}tjp/CBI}; Padin et al. 2001; Mason et al. 2002; Pearson et al. 2002), and the Very Small Array (VSA; Taylor et al. 2002; Scott et al. 2002) have also measured the CMB angular power spectrum down to subdegree scales. A desirable feature of the interferometer for CMB observation is that it directly measures the power spectrum, and that the polarimetry is a routine. In addition, many systematic problems that are inherent in single-dish experiments, such as the ground and near field atmospheric pickup and spurious polarization signals, can be significantly reduced.  The observation of CMB polarization using close-packed interferometers are also on-going or planned by the experiments like the DASI, CBI, and the forthcoming interim Array for Microwave Background Anisotropy (AMiBA\\footnote{http://www.asiaa.sinica.edu.tw/amiba}; Lo et al. 2001). They have full capabilities to probe the CMB temperature and polarization simultaneously. The feed horns are able to detect $T$, $Q$, $U$, and $V$ Stokes parameters using complex linear or circular polarizers, aiming to detect CMB linear ($Q$ and $U$) polarization at wave numbers $100 \\lesssim \\ell \\lesssim 4,000$.  This paper proposes the data analysis techniques for various strategies of the interferometric observation, especially when the $uv$-space beam size is larger than the structures of the CMB power spectrum in the $\\ell$-space. An attempt is made to increase the resolution of the estimated power spectrum in the $\\ell$-space. We have adopted three observational strategies, namely single pointing, mosaicking, and drift-scanning methods, and tested a few analysis methods extracting the angular power spectra from mock data for efficiency.  The outline of this paper is as follows. In $\\S2$ and $\\S3$, beginning with a summary of CMB interferometric observation, we describe a theoretical formalism for analyzing CMB interferometric data. A prescription for making mock CMB observations with interferometric array is given. Application of each observational strategy is made in $\\S4$. A fast unbiased power spectrum estimator is introduced in $\\S5$. Finally, the summary and discussion are given in $\\S6$. ",
        "conclusions": "We have simulated interferometric observation of CMB temperature and polarization fluctuations. For each observational strategy the data pipelines from the time-ordered raw visibility samples to the CMB angular power spectra ($\\pow_b^{TT}$, $\\pow_b^{EE}$, and $\\pow_b^{TE}$) have been developed. The pipelines are composed of making mock observation, data compression, and power spectrum estimation. Data compression is achieved by pixelization of time-ordered visibilities in real- and $uv$-spaces by means of a common map-making process. This method can be applied to any kind of interferometric observation (see $\\S3.2$). In estimating the band powers from the mock visibility samples, the optimal subspace filtering or signal-to-noise eigenmode analysis along with the quadratic estimator was used. By discarding the modes with low signal-to-noise ratios, we were able to reduce the data set to a manageable size. One drawback of the optimal subspace filtering is that while it conserves the information with signal-to-noise ratio higher than the limit of eigenvalue threshold, some useful information may be lost in certain cases. For instance, in measuring the CMB polarization power spectrum, if the band width is too small to keep sufficient amount of the signal compared with the noise level, the weak signal can disappear during the optimal subspace filtering. Therefore, we need to choose a wider band width to obtain a higher signal-to-noise ratio, especially at higher $\\ell$ region.  The measured band powers are found to be quite consistent with the band power expectation values $\\left< \\pow_b \\right>$ for the AMiBA 19-pointing mosaic (Fig. 5). This implies that our data pipelines are working reliably. Using the fact that the visibility contains direct information of CMB power spectrum, we have developed a fast unbiased estimator of the CMB power spectra (FUE, $\\S5$) that requires only $\\mathcal{O}(N_p^2)$ operations. This method is very similar to the power spectrum estimation method using Gabor transform (Hansen, G\\'orski, \\& Hivon 2002). The FUE also gives band power estimates that are consistent with those from the quadratic estimator (see Fig. 5). The FUE method does not require large computer resources. Given the precomputed quantities $B_{ij}(b)=\\partial S_{ij}/\\partial \\pow_b$, the computational speed is extremely fast. Even if the noise covariance matrix is highly non-diagonal, which is the usual case in real data analyses (e.g., handling constraint matrices to subtract the point source effect), the FUE method is still fast because the prewhitening transformation of $B_{ij}(b)$ and $V_i$ is needed only once.  Our main goal was to propose data analysis techniques for each observational strategy of a CMB interferometer, especially when the $uv$-beam size is larger than the scale of structures in the CMB power spectrum. Using the mock CBI single pointing observations, we have investigated the effect of rotation of the array platform on the band power correlations and the uncertainties of the band powers. Based on the results, summarized in Figure 2 and Table 2, we conclude that the band power anti-correlations can be reduced by rotating the platform and thus densely sampling the visibility plane. However, the uncertainties of the band power estimates slightly increase (when the total integration time is fixed). This is because the CMB signal is shared by the neighboring visibilities due to the finite beam size. In this way, we can increase the resolution of the power spectrum in the $\\ell$-space down to a resolution limit $\\Delta\\ell \\approx \\pi /\\theta$ given by the sampling theorem.  Using the recent CBI result of single pointing observation, Mason et al. (2002) have shown a power spectrum with band width of $\\Delta\\ell = 4\\sqrt{2} \\ln 2 / \\theta_{\\rm fwhm} \\approx 300$. This limit for $\\Delta\\ell$ is the FWHM of the visibility window function, which is proportional to the square of the Fourier transform of the primary beam with $\\theta_{\\rm fwhm}$ (Pearson et al. 2002). On the other hand, our choice for the band width is $\\Delta\\ell = 226$ ($\\Delta u = \\Delta u_{\\rm fwhm}/2 =36$). This is the limit given by the sampling theorem ($\\Delta\\ell$ $\\approx \\pi/\\theta$). It is $\\Delta\\ell = 4 \\ln 2 / \\theta_{\\rm fwhm}$ for a Gaussian primary beam, which is a factor of $\\sqrt{2}$ narrower than that adopted by CBI team. We show in Table 2 that a mock CBI observation with 30 different orientations results in about $20$\\% anti-correlations between neighboring bands at $\\ell \\lesssim 1,000$, and higher values at higher $\\ell \\gtrsim 1,000$, while they are 10 -- 15\\% in Mason et al. (2002) due to the wider band width. The band power correlation at high $\\ell$ regions can be reduced by more densely sampling visibilities with sufficient integration time. As shown in the example of $uv$-mosaicking using a two-element interferometer (Fig. 3), the band widths of power spectrum can be reduced while keeping the band correlations at a tolerable level by increasing the number of rotation steps with increasing dish separation, and by assigning longer integration time to the visibilities at low CMB signal regions. For intermediate five bands, the band width and the average band power correlation are $\\Delta\\ell = 132$ ($\\Delta u =21$) and $-24$\\%, respectively. The width is smaller than our resolution limit ($\\Delta\\ell= 161$, $\\Delta u = \\Delta u_{\\rm fwhm}/2$ where $\\Delta u_{\\rm fwhm} = 51.2$), also a factor of 1.7 narrower than the limit obtained by Pearson et al.'s formula ($\\Delta\\ell \\simeq 230$).  The recent DASI power spectrum is measured from the single pointing observation without platform rotation (Halverson et al. 2002). The DASI band powers have the resolution of $\\Delta\\ell \\approx 80$ (with 18 -- 28\\% anti-correlations), which is broader than the resolution limit $\\Delta\\ell = 4\\sqrt{2} \\ln 2 / \\theta_{\\rm fwhm}= 66$ where $\\theta_{\\rm fwhm} = 3\\fdg4$. We expect that the DASI single pointing observation with dense rotation of platform will allow higher resolution of about $\\Delta\\ell = 50$ at the similar level of anti-correlations. Since the mosaicking is the most efficient method for increasing the resolution of the power spectrum, the combination of mosaicking and dense rotation of the platform followed by the $uv$-pixelization is thought to be the most ideal observational strategy for DASI- and CBI-type CMB interferometers.  For each observational strategy, optimal parameter choices for the AMiBA experiment are discussed in $\\S4$, and summarized in Table 3. The 7-element AMiBA is expected to detect CMB polarization power spectrum near $\\ell \\approx 1,300$ at $4\\sigma$ level within 20 days by observing 43 fields. In AMiBA mosaicking with $t_{\\rm tot}=6$ months, the optimal parameter sets are ($\\theta_{\\rm opt}=7\\deg$, $n_{\\rm f}=1$) or ($\\theta_{\\rm opt}=3\\deg$, $n_{\\rm f}=5$) for a minimum uncertainty of the $E$-polarization power spectrum. In fact, the optimal parameters strongly depend on the characteristics of the interferometer (e.g., $\\eta_{s}$, $\\eta_{a}$, and $T_{sys}$) and on the $E$-polarization power spectrum $\\pow_b^{EE}$. Since we are considering the shortest baselines in deriving the parameters, the optimization is only for the sensitivity range of the shortest baselines ($\\ell < 2,000$). At higher $\\ell$-range ($\\ell > 2,000$) where the 7-element AMiBA has only a few baselines, the CMB polarized signal is expected to be very low. Therefore, we have chosen a wider band width for the last band for the $12 \\times 12$ mosaicking and drift-scanning observations (see Fig. 6). To obtain a meaningful polarization power spectrum at high $\\ell$ region with narrow band widths, we need to increase the integration time or the number of baselines. This can be seen in the simulation of the 19-pointing mosaicking by AMiBA where the integration time per pointing is almost one day (see Fig. 5). Although the band widths are quite wide in the temperature power spectra in Figure 6, we can measure temperature band powers independently with narrower band width because the signal-to-noise ratios of the $T$ visibilities are very high, compared with those of polarization ($\\Delta\\ell = 196$, see open circles in Fig. 6$d$).  Among the three observational strategies that we have studied, the single pointing is useful for a detection of the CMB polarized signal while the mosaicking or the drift-scanning of a large area of the sky is essential for measuring the polarization power spectrum with high $\\Delta\\ell$-resolution. The drift-scanning strategy is efficient for removing the ground contamination. It can also save half of the integration time when compared to the method of differencing two fields in the removal of the ground spillover adopted by the CBI experiment (Padin et al. 2001). In the drift-scanning, the survey region can have a shape for which the flat-sky approximation is inapplicable. Since the survey area drift-scanned by the AMiBA interferometer will be over 100 deg$^2$, it is necessary to take into account the curvature of the sky. Our future work will deal with important issues such as the removal of Galactic foreground emission, the identification of radio point sources, and the subtraction of unresolved point sources. It is also important to study the topology of the CMB temperature and polarization fields to test the primordial fluctuations for Gaussianity (see, e.g., Park \\& Park 2002)."
    },
    "0209/astro-ph0209172_arXiv.txt": {
        "abstract": "We give mean spectra and report orbital periods $P_{\\rm orb}$ based on radial velocities taken near minimum light for five dwarf novae, all of which prove to have $P_{\\rm orb} < 2$ hr. The stars and their periods are KX Aql, 0.06035(3) d; FT Cam, 0.07492(8) d; PU CMa, 0.05669(4) d; V660 Her, 0.07826(8) d;, and DM Lyr, 0.06546(6). The emission lines in KX Aql are notably strong and broad, and the other stars' spectra appear generally typical for short-period dwarf novae. We observed FT Cam, PU CMa, and DM Lyr on more than one observing run and constrain their periods accordingly. Differential time-series photometry of FT Cam shows strong flickering but rules out deep eclipses. Although dwarf novae in this period range generally show the superhumps and superoutbursts characteristic of the SU UMa subclass of dwarf novae, none of these objects have well-observed superhumps. ",
        "introduction": "In this paper we continue determining orbital periods $P_{\\rm orb}$ for SU-UMa type dwarf novae and candidate SU UMa stars. The SU UMa stars are dwarf novae, generally with $P_{\\rm orb} < 3$ h, which occasionally undergo bright and long-duration eruptions, called superoutbursts.  During superoutburst they show quasi-periodic photometric oscillations, called superhumps, which have periods a few per cent {\\it longer} than $P_{\\rm orb}$. \\citet{warn} gives an excellent discussion of these stars (and cataclysmic binaries in general).  At this time the most compelling explanation of the superhump clock invokes precession of an eccentric accretion disk \\citep{whit88}. Such disks are expected to develop in high mass ratio (hence short-period) cataclysmic binaries.  The stars reported on here all prove to have $P_{\\rm orb} < 2$ h, yet to our knowledge none of them have well-observed superhumps or superoutbursts.  Most of the stars in this sample should show superhumps in the future.  The superhump periods can then be combined with the independently determined orbital periods to compute the superhump period excess $\\epsilon = (P_{\\rm sh} - P{\\rm orb}) / P_{\\rm orb}$, which in turn appears to correlate well with the mass ratio $q = M_2 / M_1$ \\citep{patprecess01}. ",
        "conclusions": "None of these five stars appears particularly unusual. They all prove to have periods in the range occupied by the SU UMa stars. Superhumps are evidently detected (but not well measured) in DM Lyr, and presumably they have not yet turned up in the other four stars only because the objects have not been observed long or intensively enough.  If one or more of the other four objects proves after extensive monitoring {\\it not} to be an SU UMa star, it will present an interesting anomaly.   {\\it Acknowledgments.} We thank the NSF for support through AST 9987334.   Tim Miller obtained the direct images of FT Cam. This research made use of the Simbad database, operated at CDS, Strasbourg, France.   \\clearpage"
    },
    "0209/astro-ph0209458_arXiv.txt": {
        "abstract": "{Recent observations support the view that the universe is described by a FLRW model with $\\Omega_m^0 \\approx 0.3$, $\\Omega_{\\Lambda}^0 \\approx 0.7$, and $w \\leq -1/3$ at the present epoch. There are several theoretical suggestions for the cosmological $\\Lambda$ component and for the particular form of the energy transfer between this dark energy and matter. This gives a strong motive for a systematic study of general properties of two-fluid FLRW models. We consider a combination of one perfect fluid, which is quintessence with negative pressure ($p_Q = w\\epsilon_Q$ ), and another perfect fluid, which is a mixture of radiation and/or matter components with positive pressure ($p = \\beta \\epsilon_m$), which define the associated one-fluid model ($p = \\gamma \\epsilon$). We introduce a useful classification which contains 4 classes of models defined by the presence or absence of energy transfer and by the stationarity ($w = const.$ and $\\beta = const.$) or/and non stationarity ($w$ or $\\beta$ time dependent) of the equations of state. It is shown that, for given $w$ and $\\beta$, the energy transfer defines $\\gamma$ and, therefore, the total gravitating mass and dynamics of the model. We study important  examples of two-fluid FLRW models within the new classification. The behaviour of the energy content, gravitating mass, pressure, and the energy transfer are given as functions of the scale factor. We point out three characteristic scales, $a_E$, $a_{\\cal P}$ and $a_{\\cal M}$, which separate periods of time in which quintessence energy, pressure and gravitating mass dominate. Each sequence of the scales defines one of 6 evolution types. ",
        "introduction": "\\label{introd}  A number of recent observations reveal the cosmological $\\Lambda$ component. Type Ia Supernovae (Riess et al. \\cite{Riess98}, Perlmutter et al. \\cite{Perl}, Riess et al. \\cite{Riess01}) and the Boomerang, Maxima and Dasi measurements of the total density parameter $\\Omega$ via the first acoustic peak location in the angular power spectrum of the CBR (de Bernardis et al. \\cite{deBern00}, Balbi et al. \\cite{Balbi00}, Jaffe et al. \\cite{Jaffe00}) show that $\\Omega = \\Omega_m^0 +\\Omega_{\\Lambda}^0 = 1 \\pm 0.05$ with $\\Omega_{\\Lambda}^0 \\approx 0.7$. Existing cosmological data allow a wide range for the equation of state coefficient $w$ from $-1$ to $-1/3$ (Perlmutter et al. \\cite{PerlTur99}; Podariu \\& Ratra \\cite{Pod00}; Wang et al. \\cite{Wang00}).   Also, the smooth Hubble flow around our Local Group, inside a highly lumpy matter distribution, suggests still other evidence for a dominating $\\Lambda$ component and its variation with time (Chernin et al. \\cite{CherTerBar00}; Baryshev et al. \\cite{BarCherTer01}; Chernin \\cite{CherninUFN}; Klypin et al. \\cite{Klypin01}; Axenides \\& Perivolaropoulos \\cite{Axenides02}).  There are several theoretical models for $\\Lambda$-like cosmological components of the universe with positive energy density and negative pressure, including vacuum with a constant $\\Lambda$, decaying $\\Lambda$, and variable equation of state $w(t)$ (Peebles \\& Ratra \\cite{PeebRat88}; Lima \\& Maia \\cite{Lima94}; Wetterich \\cite{Wett95}; Ferreira \\& Joyce  \\cite{Ferr97}; Caldwell et al. \\cite{Cald98}; Steinhardt et al. \\cite{Stein98}; Zlatev et al. \\cite{Zlatev99}; Bahcall et al. \\cite{bahcall99}; Mbonye \\cite{Mbonye02}). We use the term ``quintessence'' for any kind of substance having the equation of state $p_Q = w\\,\\varepsilon_Q$ with $-1 \\le w < 0$, which may be time variable.   Thus observations and theory give strong motivation to study general properties of two-fluid Friedmann--Lema\\^{i}tre--Robertson--Walker (FRLW) models in which the $\\Lambda$ component dominates at late epochs and there is energy transfer between $\\Lambda$ and matter components.  The energy transfer between dust-like matter and radiation was first studied by Davidson (\\cite{David62}). Some examples of the physics producing the energy transfer were given in Sistero (\\cite{Sis71}) and McIntosh (\\cite{McIn1967}, \\cite{McIn1968}). The first exact solution of the equation of motion for a dust+radiation model with no energy transfer was obtained by Chernin (\\cite{Chernin65}), who applied it to the case where the radiation  component is the cosmic background radiation or neutrinos.  In the present paper we give a systematic presentation of the properties of two-fluid cosmological models, with and without energy transfer, in the frame of a new classification which naturally arises in the two-fluid problem. In section 2 we briefly summarize the standard two-fluid FLRW model. In section 3 we consider general properties of two-fluid models with matter and quintessence. Section 4 includes three examples of models from different classes of our classification. Section 5 contains the conclusions. ",
        "conclusions": "The cosmological view that the universe is described by a FLRW model with $\\Omega_m^0 \\approx 0.3$, $\\Omega_{\\Lambda}^0 \\approx 0.7$, and $w \\leq -1/3$ has initiated many studies of FLRW models with  an essential $\\Lambda$ component at late epochs. Usually one has viewed $\\Lambda$ and matter as independent substances so that the energy-momentum tensors of the partial fluids are separately conserved (Eq.(\\ref{divergence 12})). However, there are a number of suggestions in the litterature on the particular forms of energy transfer between dark energy and matter. This has motivated us to consider on a phenomenological level the general case when one-fluid converts into another and the equation of state for both components is non-stationary (Section 3). The properties of the model we  gave in terms of the coefficients of the equation of state for two partial models ($\\beta$ for a component with positive pressure and $w$ for a negative pressure one) and a coefficient $\\gamma$ for the associated one-fluid. The energy transfer, when a one-fluid converts into another, was represented via these coefficients and the scale factor (Eqs.(\\ref{classific equation}) - (\\ref{classific equation flat})).  We have analyzed four classes of models defined by the presence or absence of the energy transfer and by the stationarity ($w = const.$, and $\\beta = const.$) or/and non-stationarity ($w$ or $\\beta$ time dependent) of the equations of state (see Tables 1, 4, 5). It was shown in sect.3 that  \\begin{itemize}  \\item[*]{for given $w$ and $\\beta$, the energy transfer defines $\\gamma$ and, therefore, the total gravitating mass and the dynamics of the model.} \\item[*]{also, the model can be coherent only if there is energy transfer.}  \\end{itemize} The classification was illustrated with interesting examples of two-fluid FLRW models in sections 4.1 -- 4.3.  From the behaviour of the energy content, gravitating mass, and pressure as functions of the scale factor we have defined three characteristic scales, $a_E$, $a_{\\cal P}$ and $a_{\\cal M}$. These separate time intervals when quintessence energy, pressure and gravitating mass were dominating (Eqs.(\\ref{def a 1}) - (\\ref{character scales 1})). Any sequence of the scales defines one of 6 evolution types of the model (Eqs.(\\ref{6 nonequal})). There is a correspondence between the dynamics of a model, its evolution type and energy transfer."
    },
    "0209/astro-ph0209034_arXiv.txt": {
        "abstract": "Stars form out of molecular gas and supply dust grains during their last evolutionary stages; in turn hydrogen molecules (\\H2) are produced more efficiently on dust grains. Therefore, dust can drastically accelerate \\H2 formation, leading to an enhancement of star formation activity. In order to examine the first formation of stars and dust in galaxies, we model the evolution of galaxies in the redshift range of $5<z<20$. In particular, we focus on the interplay between dust formation in Type II supernova ejecta and H$_2$ production on dust grains. Such effect causes an enhancement of star formation rate by an order of magnitude on a timescale ($\\sim 3$--5 galactic dynamical times) shorter than the Hubble timescale. We also find that about half of the radiative energy from stars is reprocessed by dust grains and is finally radiated in the far infrared (FIR). For example, the typical star formation rate, FIR and ultraviolet (UV) luminosity of a galaxy with $\\Mvir =10^{11.5}~M_\\odot$ (virial mass) and $\\zvir =5$ (formation redshift) are 3 $M_\\odot$ yr$^{-1}$, $4\\times 10^9~L_\\odot$, and $3\\times 10^9~L_\\odot$, respectively. This object will be detected by both ALMA and {\\it NGST}. Typical star formation rates and luminosities (FIR, UV and metal-line luminosities) are calculated for a large set of $(\\Mvir ,\\,\\zvir )$. Using these results and the Press-Schechter formalism, we calculate galaxy number counts and integrated light from high-redshift ($z>5$) galaxies in sub-millimetre and near-infrared bands. We find that: i) ALMA can detect dust emission from $\\mbox{several}\\times 10^3$ galaxies per square degree, and ii) {\\it NGST} can detect the stellar emission from $10^6$ galaxies per square degree. Further observational checks of our predictions include the integrated flux of metal (oxygen and carbon) lines; these lines can be used to trace the chemical enrichment and the gas density in early galactic environments. We finally discuss possible color selection strategies for high-redshift galaxy searches. ",
        "introduction": "In order to understand the chemical and thermodynamical state of the interstellar medium (ISM) of primeval galaxies, dust formation needs to be considered. Even in metal poor galaxies, dust grains can drastically accelerate the formation rate of molecular hydrogen (H$_2$), expected to be the most abundant molecule in the ISM (Hirashita, Hunt, \\& Ferrara 2002a). Hydrogen molecules emit vibrational-rotational lines, thus cooling the gas. This process is particularly important to understand the formation of stars in metal-poor primeval galaxies (e.g., Matsuda, Sato, \\& Takeda 1969; Omukai \\& Nishi 1998; Nishi \\& Susa 1999; Bromm, Coppi, \\& Larson 2002; Abel, Bryan, \\& Norman 2002; Nakamura \\& Umemura 2002; Kamaya \\& Silk 2002; Ripamonti et al.\\ 2002). The important role of dust on the enhancement of \\H2 abundance is also suggested by observations of damped Ly$\\alpha$ systems (DLAs; Ge, Bechtold, \\& Kulkarni 2001; cf.\\ Petitjean, Srianand, \\& Ledoux 2000).  The existence of dust in young galaxies is naturally expected because Type II supernovae (SNe II) are shown to produce dust grains  (e.g., Dwek et al.\\ 1983; Moseley et al.\\ 1989; Kozasa, Hasegawa, \\& Nomoto 1991; Todini \\& Ferrara 2001). Since the lifetime of SN II progenitors (massive stars) is short, SNe II are the dominant production source for of dust grains in young ($<1$ Gyr) star-forming galaxies. The winds of evolved low-mass stars contribute to dust formation considerably in nearby galaxies (Gehrz 1989), but the cosmic time is not long enough for such stars to evolve at high redshift ($z>5$), when all galaxies should have ages smaller than $\\sim 1$ Gyr. However, dust is also destroyed by SN shocks (McKee 1989; Jones, Tielens, \\& Hollenbach 1996). The detailed modelling of dust evolution in galaxies therefore requires an accurate treatment of both types of processes (for recent modeling, see e.g., Edmunds 2001; Hirashita, Tajiri, \\& Kamaya 2002b).  Here we model the evolution of dust content in primeval galaxies. We adopt the results of Todini \\& Ferrara (2001) for the dust formation rate in SNe II. Although further discussion on their application of nucleation theory is necessary (e.g., Frenklach \\& Feigelson 1997), their results have been successfully applied to the interpretation not only of the properties of SN 1987A but also of the FIR properties of the young dwarf galaxy SBS 0335$-$052 (Hirashita et al.\\ 2002a).  One of the most direct observational constraints for the evolution of dust content in galaxies comes from the far-infrared (FIR) properties of galaxies. Dust grains absorb stellar light and reemit it in FIR. Recent observations by the Submillimetre Common-User Bolometer Array (SCUBA) and the {\\it Infrared Space Observatory} ({\\it ISO}) have made it possible to study galaxy evolution in the FIR band up to $z\\la 3$ (Smail et al.\\ 1998). The detection of the cosmic infrared--submillimetre (sub-mm) background by the {\\it COsmic Background Explorer} ({\\it COBE}\\,) (Puget et al.\\ 1996; Fixsen et al.\\ 1998) has also provided crucial information on the star formation history of galaxies in the universe (e.g., Dwek et al.\\ 1998). Some theoretical works have modelled the FIR evolution of galaxies up to $z\\sim 5$ (Tan, Silk, \\& Balland 1999; Pei, Fall, \\& Hauser 1999; Takeuchi et al.\\ 2001a; Xu et al.\\ 2001; Pearson 2001; Totani \\& Takeuchi 2002), and the FIR luminosity of galaxies per unit comoving volume seems to be much higher at $z\\sim 1$ than at $z\\sim 0$ (see also Elbaz et al.\\ 2002). However, such a strong ``evolution'' beyond $z=2$ has been excluded (Gispert, Lagache, \\& Puget 2000; Malkan \\& Stecker 2001; Takeuchi et al.\\ 2001a).  Although there is clear evidence for the existence of dust in galaxies at $z\\la 5$ (Armus et al.\\ 1998; Soifer et al.\\ 1998), few works focusing on the early dust formation in galaxies exist. Some ``semi-analytic'' works have included the dust formation in the early galaxy evolution (e.g., Devriendt \\& Guiderdoni 2000; Granato et al.\\ 2000), but there has been no study treating the dust formation, the molecular formation on grain surfaces, and the star formation history in a consistent manner. Therefore, in this paper, we model the three processes consistently so that we can obtain an observational strategy under a consistent scenario for the early evolution of galaxies.  In order to understand which physical processes govern dust formation, observations at sub-mm wavelengths ($300~\\mu{\\rm m}\\la\\lambda\\la 1$ mm) are crucial. For high-redshift objects, redshifted FIR radiation, i.e., sub-mm light, should be observed to detect the dust emission. In particular, detecting the sub-mm radiation from galaxies at $z>5$ requires more sensitive and high-resolution observations (e.g., Takeuchi et al.\\ 2001b). A future ground-based interferometric facility, the Atacama Large Millimeter Array (ALMA\\footnote{http://www.eso.org/projects/alma/}), can be used to study such high-redshift galaxies. The detected amount of metals and stars can be used to constrain the galaxy evolution through a chemical evolution model (Tinsley 1980). Redshifted sub-mm metal emission lines can also be observed with ALMA. This can directly constrain the abundance of metals formed in the early epoch of galaxy evolution. (Oh et al.\\ 2002 have also proposed to probe high-redshift intergalactic medium metallicity by metal absorption lines.) In order to detect the stellar light from the high-redshift universe, observations by the {\\it Next Generation Space Telescope} ({\\it NGST}~\\footnote{http://ngst.gsfc.nasa.gov/}) in near infrared (NIR) will be particularly suitable. At 2 $\\mu$m, for example, we can observe the $\\sim 2000$ \\AA\\ ultraviolet (UV) light radiated from a galaxy at $z\\sim 10$. Therefore, our scenario will become testable in the near future. This means that it is worth constructing a consistent model for the high-redshift galaxy evolution.  For any observational facility, statistical properties of galaxies should be discussed to obtain a general picture of galaxy evolution. Two quantities are particularly important for statistical purposes: galaxy number counts (the number of galaxies as a function of observed flux) and integrated light (the sum of the flux from all the galaxies considered; Hauser \\& Dwek 2001 for a review). In this paper, therefore, we estimate the contribution of high-redshift galaxies to these two quantities.  Throughout this paper, we assume a flat cold dark matter (CDM) cosmology with a cosmological constant. The values of quantities are the same as those in Mo \\& White (2002) ($\\Omega_{\\rm M}=0.3$, $\\Omega_\\Lambda =0.7$, and $H_0\\equiv 100 h$ km s$^{-1}$ Mpc$^{-1}=70$ km s$^{-1}$ Mpc$^{-1}$). The baryon density parameter is assumed to be $\\Omega_{\\rm b}=0.02h^{-2}$. For the power spectrum of the density fluctuation, $n=1$ and $\\sigma_8=0.9$ are adopted. We first model the physical state of gas and the content of dust and metals in a galaxy during its early evolutionary stage (\\S~\\ref{sec:each}). There we also model the luminosities of FIR, UV, and metal lines. The result of our model for fiducial galaxies are presented in \\S~\\ref{sec:result}. Based on these results, we next calculate the galaxy number counts and the integrated light at various wavelengths in \\S~\\ref{sec:nc}. We discuss the observational implications of our results in \\S~\\ref{sec:discussion}. ",
        "conclusions": "\\label{sec:discussion}  \\subsection{Summary of evolutionary properties}  In order to quantify the importance of dust on the first star formation activity in the universe, we  have solved the time evolution of dust mass in the galaxies formed in the redshift range $z>5$, when the age of the universe is $\\la 1$ Gyr. We have taken into account the importance of \\H2 abundance for the star formation rate, and the formation of molecules on dust in a consistent manner (\\S~\\ref{sec:each}). In particular, we have made the first attempt to tie the star formation efficiency to \\H2 abundance in relatively primordial environments (\\S~\\ref{subsec:sfr}). Even when this inefficient phase of star formation is included, an active phase of star formation takes place after a few $t_{\\rm cir}$ (much shorter than the Hubble timescale) because a significant amount of dust is accumulated to activate the \\H2 formation on the grain surfaces (Fig.\\ \\ref{fig:H2}). This suggests that the grains play an essential role in causing the first active phase of star formation. As a result, we have provided a robust support for some theoretical works that have implicitly assumed that stars are formed at high redshift as efficiently as at low redshift.  Radiative properties of high-redshift star-forming galaxies are also predicted. We have found that a significant amount of luminosity is radiated in the FIR range. The FIR luminosity becomes comparable to the UV luminosity in a few $t_{\\rm cir}$, when a significant amount of dust is accumulated (Fig.\\ \\ref{fig:fir_luminosity}). This efficient reprocessing of the UV light into the FIR results partly from the dense (i.e., large optical depth) environment of high-redshift galaxies. Observations of the FIR light (sub-mm light in the observer's restframe) as well as those of the stellar light are thus crucial to trace the whole stellar radiative energy from high-redshift galaxies.  In the framework of our model we have also given an approximate estimate of some sub-mm metal-line luminosities. However, this can only been seen as an upper limit to the actual luminosity because considerable uncertainty is present on the gas density (\\S~\\ref{subsubsec:metal} and Appendix \\ref{app:line}). If future sub-mm or millimetre observations detect metal lines, a density probe of ISM of the high-redshift galaxies will be possible. If more than two types of metal-lines are detected, density can be estimated more precisely.  \\subsection{Future observational tests}  In about ten years, it will become possible to detect sources at high redshift ($z>5$) in both sub-mm and NIR. The luminosity level of the galaxies at these wavelengths will put important constraints on our model. Therefore, we have calculated the number counts for both wavelengths. As a result, we have found that ALMA (450 $\\mu$m, 850 $\\mu$m, and 1.3 mm bands) and {\\it NGST} (NIR bands) can detect several times $10^3$ and $10^6$ high-redshift galaxies per square degree, respectively. These numbers can be used to test our model.  We have also calculated the integrated intensity of the metal-line emission from the galaxies from $z=5$ to 7. Although a precise determination of the line intensity requires a model for the gas density (perhaps a model for photo-dissociation region as Hollenbach \\& McKee 1979 is also necessary), we can estimate a maximum intensity for the integrated metal-line intensity. The results are listed in the last two columns of Table \\ref{tab:metal}. The contamination with the cosmic sub-mm and microwave background could represent a potential problem, but such high-redshift galaxies have a correlation scale of the order of $10''$ (Appendix \\ref{app:angular}). Therefore, a fluctuation analysis of sky brightness in (sub-)millimetre can discriminate the metal-line signal from other contaminating sources by examining the typical correlation scale. A quantitative analysis of the fluctuations using the structure formation theory is left for future work. It may be also possible to probe the gas density of high-redshift galaxies through the factor ${\\cal F}$ in Appendix \\ref{app:line}, if more than two kinds of metal-lines are detected. The relative intensities of two lines can be used to derive a probable value of $\\nH$.  Finally, we should mention how to select high-redshift galaxies efficiently. Galaxy colors (flux ratios between two bands) are often used for the selection. We have shown that ALMA can detect galaxies at $z\\la 7$. The peak of the dust emission lies roughly between 500 $\\mu$m and 800 $\\mu$m for galaxies between $z=5$ and 7. For example, 450 $\\mu$m vs. 1.3 mm flux ratio (450--1300 $\\mu$m color) gives us a useful information on the redshift, because the peak of flux lies between these bands only for the high redshift galaxies. In Figure \\ref{fig:color}, we show the flux at 850 $\\mu$m ($S_{850~\\mu{\\rm m}}$) and 450--1300 $\\mu$m color ($S_{450~\\mu{\\rm m}}/S_{1.3~{\\rm mm}}$) predicted by the modified blackbody spectra (eq.\\ \\ref{eq:firsed}). The galaxies detected by ALMA in that redshift range has typical flux levels $\\sim 10$--100 $\\mu$Jy and the 450--1300 $\\mu$m color is typically less than 3. In Figure \\ref{fig:color}, we also show the ALMA detection limit (horizontal dashed line). Galaxies with $\\Mvir\\ga 10^{11.5}~M_\\odot$ will be detected by ALMA. At $z>7$, however, the number of such massive galaxies is negligible and does not contribute to number counts. Typically, galaxies whose redshift is less than 4 fall to the right of the vertical dotted line (Takeuchi et al.\\ 2001b).  \\begin{figure} \\includegraphics[width=8cm]{fig8.eps} \\caption{Relation between the flux at $\\lambda =850~\\mu$m ($S_{\\rm 850~\\mu m}$) and the 450--1300 $\\mu$m color ($S_{\\rm 450~\\mu m}/S_{\\rm 1.3~mm}$). (We adopt the same notations as Fig.\\ 10 of Takeuchi et al.\\ 2001b in this figure.) Our model predictions are identified by filled squares for $z=5$, 6, 7, and 8 and $\\Mvir =10^{11}$, $10^{12}$ and $10^{13}~M_\\odot$. The horizontal dashed line indicates the detection limit of ALMA. Low-redshift galaxies ($z<4$) typically fall on the right of the vertical dotted line (Takeuchi et al.\\ 2001b). \\label{fig:color}} \\end{figure}  We can also select high-redshift galaxies efficiently from optical observations by using the ``dropout'' technique (Steidel et al.\\ 1996). The Lyman limit at the wavelength of 912 \\AA\\ in the restframe of a galaxy is redshifted to 5500--7300 \\AA\\ for galaxies at $z=5$--7. Therefore, optical/NIR observations of galaxies by {\\it NGST} (Mather \\& Stockman 2000) or other sensitive facilities provides us a way to sample the high-redshift candidates independent from the ALMA sample. A large sample of galaxies with $V$- or $R$-band dropout should be collected by future observations. After spatial cross identification of drop-out sample with ALMA sample, we can investigate the optical--sub-mm flux ratio as a test of our model. Galaxies with $\\Mvir\\ga 10^{11.5}~M_\\odot$ are detectable both by ALMA and {\\it NGST}. In order to see the typical luminosities for galaxies detected by ALMA, we show in Figure \\ref{fig:lum_vs_z} $\\bar{L}_{\\rm UV}$ and $\\bar{L}_{\\rm FIR}$ as a function of $\\zvir$. We see that FIR/UV flux ratios are 1.2, 2.7, and 5.8 for $\\zvir =5$, 6, and 7, respectively. In the same figure, we also present $\\bar{L}_{\\rm O146}^{\\rm max}$. As mentioned in \\S~\\ref{subsec:integ_metal}, the ratio between the observed line luminosity and $\\bar{L}_{\\rm O146}^{\\rm max}$ (i.e., ${\\cal F}$) can be used to estimate the density of ISM.  \\begin{figure} \\includegraphics[width=8cm]{fig9.eps} \\caption{Luminosities as a function of formation redshift $\\zvir$ for $\\Mvir =10^{11.5}~M_\\odot$. Such a massive galaxy will be detected by ALMA. The solid, dotted, and dashed lines represent ultraviolet, far-infrared, and O {\\sc i} 146 $\\mu$m (maximum; $\\bar{L}_{\\rm O146}^{\\rm max}$) luminosities defined at 4 circular times. \\label{fig:lum_vs_z}} \\end{figure}  \\subsection{Connection to lower redshift}  As shown in Figure \\ref{fig:cosm_sfh}, our predictions connect smoothly to the lower-redshift star formation history. Our model, however, cannot be applied to galaxy evolution at $z<5$ because after that epoch dust is supplied from late-type stars as well as SNe II. It is observationally known that mergers between giant galaxies significantly contribute to luminous infrared populations at the local universe (Sanders \\& Mirabel 1996) and even at $z\\sim 1$ (Roche \\& Eales 1999). When we apply our framework to lower redshifts, therefore, it is necessary to extend our model to include the details of the merging history of galaxies. We should note that the enhancement of molecular formation is also a key to star formation activity in mergers (e.g., Walter et al.\\ 2002).  Recent studies using the Subaru telescope (Ouchi et al.\\ 2002) have pushed observations as deep as $z\\sim 5$. Therefore, the luminosity function (or comoving star formation rate) derived from the ``Subaru Deep Field'', which is as wide as 600 arcmin$^2$ and as deep as 26 AB magnitude around 7000 \\AA will allow us to directly compare our results at $z\\sim 5$. The luminosity function at $z\\sim 5$ is also important to constrain the evolutionary scenario of Lyman break populations found at $z\\sim 3$. Is the luminosity function of galaxies at $z\\sim 5$ explained by the same population of Lyman break galaxies at $z\\sim 3$? Recently, Ferguson, Dickinson, \\& Papovich (2002) have given a negative answer to this question, but further studies are necessary to reveal the link between these two epochs.  We have stressed the importance of dust on the formation of molecular-rich environment. In the lower-redshift ($z<5$) universe, it is observationally known that there is a correlation between the abundances of dust and molecules for DLAs (Ge et al.\\ 2001). This strongly suggests the important role of dust for molecule formation (see also Levshakov et al.\\ 2002). However, the correlation is not firmly assessed and further observational sample seems to be required (Petitjean et al.\\ 2000). Petitjean et al.\\ also noted that most of DLAs may arise selectively in warm and diffuse neutral gas. Liszt (2002) has shown that even in a cool medium \\H2 formation can be suppressed because of low dust content and strong UV radiation field. Moreover, DLA trace a diverse population with various mass, surface brightness, etc.\\ (e.g., Pettini 2002). In spite of those complexities, DLAs are promising objects to study the link between the abundances of dust and molecules in the early universe."
    },
    "0209/astro-ph0209208_arXiv.txt": {
        "abstract": "The main tools in cosmology for comparing theoretical models with the observations of the galaxy distribution are statistical. We will review the applications of spatial statistics to the description of the large-scale structure of the universe. Special topics discussed in this talk will be: description of the galaxy samples, selection effects and biases, correlation functions, Fourier analysis, nearest neighbor statistics, Minkowski functionals and structure statistics. Special attention will be devoted to scaling laws and the use of the lacunarity measures in the description of the cosmic texture. ",
        "introduction": "\\label{sect:intro}  %   Cosmology is a science which is experiencing a great development in the last decades. The achievements in the observations are driven the subject into an era of precision. The two fundamental pillars upon which observational cosmology rests are the cosmic microwave background and the distribution of the galaxies. The analysis of the huge amount of data that is now being collected in both areas will provided a unified framework to explain the formation and evolution of the large-scale structure in the universe. In this paper we will review some of the aspects related with the galaxy clustering. ",
        "conclusions": ""
    },
    "0209/astro-ph0209452_arXiv.txt": {
        "abstract": "Based on the radiation hydrodynamical model for the black hole (BH) growth, incorporated with the chemical evolution of the early-type host galaxy, we construct the coevolution model of a QSO BH and the host galaxy. As a result, it is found that after a galactic wind epoch, the luminosity is shifted  from the host-dominant phase to the AGN-dominant phase (QSO phase) in the timescale of a few $10^{8}$ years. The former phase corresponds to the early stage of growing BH, and can be regarded as a ``proto-QSO'' phase. It has observable characteristic properties as follows: (1) The width of broad emission line is narrower than that of ordinary QSOs, and it is typically less than 1500km/s. (2) The BH-to-bulge mass ratio, $M_{\\rm BH}/M_{\\rm bulge}$, is in the range of $10^{-5.3}-10^{-3.9}$. (3) Host galaxies are bluer compared to QSO hosts, by about 0.5 magnitude in the colors of $({\\it B-V})$ at the rest bands  and $({\\it V-K})$ at the observed bands, with assuming galaxy formation redshifts of $z_{\\rm f}=3-5$. (4) The metallicity of gas in galactic nuclei is $\\sim 8Z_{\\odot}$, and that of stars weighted by the host luminosity is $\\sim 3Z_{\\odot}$. (5) The central massive BH ($\\simeq 10^{7}M_{\\odot}$) is surrounded by a massive dusty disk ($ > 10^{8}M_{\\odot}$), which may obscure the nucleus in the edge-on view and make a type 2 nucleus. By comparing these predictions with recent observations, radio galaxies are a possible candidate for proto-QSOs. Also, it is anticipated that the proto-QSO phase is preceded by an optically thick phase, which may correspond to ULIRGs. In this phase, $M_{\\rm BH}/M_{\\rm bulge}$ is predicted to be much less than $10^{-3}$ and grow with metallicity. Moreover, as precursors of ULIRGs, optically-thin star-forming galaxies are predicted. These may be in the assembly phase of Lyman break galaxies (LBGs) or Ly$\\alpha$ emitters. ",
        "introduction": "\\label{INTRO}  Recent X-ray and optical observations suggest the possibility that active galactic nuclei (AGNs) could be divided into two subclasses according to the rate of black hole (BH) growth; one is a rapidly growing phase and the other is a slow growing phase (Pounds et al. 1995; Boller et al. 1996; Mineshige et al. 2000; Mathur et al. 2001: Wandel 2002). This possibility has been pointed out primarily for the Seyfert 1 galaxies (Sy1s), which are divided into two subclasses according to the width of broad emission line, $V_{\\rm BLR}$. Sy1s with $V_{\\rm BLR}$ less than 2000km/s are called narrow line Sy1s (NLSy1s), whereas those with broader line width are called broad line Sy1s (BLSy1s). NLSy1s exhibit two distinctive X-ray properties, that is, rapid X-ray variability and strong soft X-ray excess. These properties can be explained in terms of the optically thick ADAF (advection-dominated accretion flow) onto a smaller BH, which is realized by higher accretion rate compared to the Eddington limit (Pounds et al. 1995; Boller et al. 1996; Mineshige et al. 2000). Also, it is pointed out that the BH-to-bulge mass ratio is noticeably smaller than that in elliptical galaxies, $M_{\\rm BH}/M_{\\rm bulge} < 10^{-3}$ (Mathur et al. 2001; Wandel 2002). All of these suggest that NLSy1s are in the rapidly growing phase of BH, in contrast to BLSy1s which are explained by conventional mild accretion onto a large BH. Additionally, Kawaguchi \\& Aoki (2000) argue that NLSy1s have a high star formation rate (SFR). According to these observations, it has been suggested that NLSy1s may be Sy1s in the early stage of their evolution (Mathur 2000). On the analogy of NLSy1s, QSOs are also expected to have rapidly growing phase of QSO BHs. But, it has not been elucidated so far what objects correspond to the early phase of QSOs.  On the other hand, recent high-resolution observations of galactic centers have revealed that the estimated mass of a central ``massive dark object''(MDO), which is the nomenclature for a supermassive BH candidate, does correlate with the mass of a galactic bulge; the mass ratio of the BH to the bulge is 0.001-0.006 as a median value (Kormendy \\& Richstone 1995; Richstone et al. 1998; Magorrian et al. 1998; Loar 1998; Gebhardt et al. 2000; Ferrarese \\& Merritt 2000; Merritt \\& Ferrarese 2001 McLure \\& Dunlop 2001; McLure \\& Dunlop 2002;  Wandel 2002). (It is noted that the bulge means a whole galaxy for an elliptical galaxy.) In addition, it has been found that QSO host galaxies are mostly luminous and well-evolved early-type galaxies (McLeod \\& Rieke 1995; Bahcall et al. 1997; Hooper, Impey \\& Foltz 1997; McLoed, Rieke \\& Storrie-Lombardi 1999; Brotherton et al. 1999; Kirhakos et al. 1999; McLure et al. 1999; McLure, Dunlop \\& Kukula 2000). These findings, combined with the BH-to-bulge relations, suggest that the formation of a supermassive BH, an elliptical galaxy, and a QSO is physically related to each other. But, the link between the formation of a supermassive BH and the evolution of a host galaxy is an open question. Also, the physical relationship among QSOs, ultraluminous infrared galaxies (ULIRGs), and radio galaxies has been an issue of long standing.  Some theoretical models of BH growth models have been considered to explain the BH-to-bulge correlations (Silk \\& Rees 1998; Ostriker 2000; Adams, Graff, \\& Richstone 2001). But, little has been elucidated regarding the physics on the angular momentum transfer, which is requisite for BH formation. Recently, as a potential mechanism to remove angular momentum, Umemura (2001) has considered the effects of radiation drag, which is equivalent to a well-known Poynting-Robertson effect. The exact expressions for the radiation drag are found in the literature (Umemura, Fukue, \\& Mineshige 1997; Fukue, Umemura, \\& Mineshige 1997). In an optically thick regime, the efficiency of radiation drag is saturated due to the conservation of the photon number (Tsuribe, \\& Umemura 1997). Thus, the angular momentum loss rate by the radiation drag is given by $d \\ln J/dt \\simeq -(L_{*}/c^{2}M_{\\rm g})$, where $J$, $L_{*}$, and $M_{\\rm g}$ are the total angular momentum of gaseous component, the total luminosity of the bulge, and the total mass of gas, respectively. Then, the maximal rate of mass accretion is given by $\\dot{M}=-M_{\\rm g}d \\ln J/dt = L_{*}/c^{2}$ (Umemura 2001). Thus, the total accreted mass on to the MDO, $M_{\\rm MDO}$, is estimated by \\begin{eqnarray} M_{\\rm MDO} &\\simeq& \\int_{0}^{\\infty}\\frac{L_{*}}{c^{2}}dt. \\nonumber \\end{eqnarray} In practice, the interstellar medium (ISM) is observed to be highly inhomogeneous in an active star-forming galaxies (Sanders et al. 1988; Gordon, Calzetti \\& Witt 1997). Kawakatu \\& Umemura (2002) have shown that the inhomogeneity of interstellar medium helps the radiation drag to sustain the maximal efficiency.  Thus, the final mass of MDO is proportional to the total radiation energy from bulge stars, and the resultant BH-to-bulge mass ratio is basically determined by the energy conversion efficiency of the nuclear fusion from hydrogen to helium, i.e. 0.007 (Umemura 2001). So far, the realistic chemical evolution of the host galaxy has not been incorporated, but a simple evolutionary model was assumed. As for the relation between a QSO BH and the host galaxy, some phenomenological models have been proposed (Haehnelt \\& Rees 1993; Haiman \\& Loeb 1998; Kauffmann \\& Haehnelt 2000; Monaco, Salucci, \\& Danese 2000; Granato et al. 2001; Hosokawa et al. 2001), but little on the physics has been known. Hence, in order to reveal the formation and evolution of QSOs and clarify what objects correspond to the early phase of QSOs, it is important to investigate the physics on the rapidly growing phase of QSO BHs. Here, based on the radiation drag model with incorporating the realistic chemical evolution, we construct a physical model for the coevolution of a QSO BH and the early-type host galaxy. The purpose of this paper is to elucidate the physical relationship between a BH growth and the evolution of host galaxy, and define a proto-QSO phase as an early stage of QSO evolution. Then, we predict the observable properties of proto-QSOs. Also, we address a unified picture for the evolution of an elliptical galaxy nucleus.  The paper is organized as follows. In section 2, we build up a theoretical model for the coevolution of a QSO BH and the early-type host galaxy. In section 3, we investigate the time-dependent relation between a QSO BH and the early-type host galaxy, and analyze the physical states of proto-QSOs which correspond to the rapidly growing phase of a QSO BH. In Section 4, we propose a unified picture for the evolution of elliptical galaxy nucleus. Section 5 is devoted to the conclusions. ",
        "conclusions": "\\label{D} Based on the radiation drag model for the BH growth, incorporated with the chemical evolution of the early-type host galaxy, we have built up the coevolution model for a QSO BH and the host galaxy. As a consequence, we have shown the possibility of the proto-QSO phase, which is optically-thin and host luminosity-dominant, and has the life-time comparable to the QSO phase timescale of a few $10^{8}$ years. We have predicted the observable properties of proto-QSOs as follows: (1) The width of broad emission line is narrower, which is less than $1500$km/s. (2) The BH-to-bulge mass ratio, $M_{\\rm BH}/M_{\\rm bulge}$, rapidly increases from $10^{-5.3}$ to $10^{-3.9}$ in $\\approx 10^{8}$ years. (3) The colors of $({\\it B-V})$ at rest bands and $({\\it V-K})$ at observed bands are about 0.5 magnitude bluer than those of QSOs. (4) In both proto-QSO and QSO phases, the metallicity of gas in galactic nuclei is $Z_{\\rm BLR}\\simeq 8Z_{\\odot}$, and that of stars weighted by the host luminosity is $Z_{*}\\simeq 3Z_{\\odot}$, which are consistent with the observations for QSOs and the elliptical galaxies. (5) A massive dusty disk ($ > 10^{8}M_{\\odot}$) surrounds a massive BH, and it may obscure the nucleus in the edge-on view to form a type 2 nucleus. The predicted properties of proto-QSOs are similar to those of radio galaxies.  The proto-QSO phase is preceded by an optically-thick phase before the galactic wind, which may correspond to ULIRGs. The present model predicts a low luminosity ratio of $L_{\\rm AGN}$ to $L_{\\rm bol}$, which is consistent with the observed ratio $L_{\\rm x}/L_{\\rm bol} \\ll 0.01$ for ULIRGs. In addition, $M_{\\rm BH}/M_{\\rm bulge}$ is anticipated to be much less than $10^{-3}$ and $M_{\\rm BH}/M_{\\rm bulge}$ grows with metallicity in the ULIRG phase.  Finally, we can predict the precursor of ULIRGs, which is optically thin and their lifetime is $\\sim 10^{7}$ years. This may correspond to the assembly phase of LBGs or Ly$\\alpha$ emitters. In this phase, the massive dusty disk of $\\approx 10^{6-7}M_{\\odot}$ exists, the metallicity is subsolar ($Z_{*} < 0.1Z_{\\odot} $), and the hard X-ray luminosity is $L_{\\rm x}\\sim 5\\times 10^{8}L_{\\odot}$ if $L_{\\rm x}=0.1L_{\\rm AGN}$. In addition, the formation of a seed black hole (BH) ($\\sim 10^{5}M_{\\odot}$) can occur due to the collapse of a rotating supermassive star (SMS) in this phase. Thus, the gravitational wave may be detectable by the Laser Interferometer Space Antenna (LISA)."
    },
    "0209/astro-ph0209178_arXiv.txt": {
        "abstract": "We report new estimates for the lower mass function of 5 young open clusters spanning an age range from 80 to 150 Myr.  In all studied clusters, the mass function across the stellar/substellar boundary ($\\sim$ 0.072 $\\msun$) and up to 0.4 $\\msun$ is consistent with a power-law with an exponent $\\alpha \\simeq -0.5 \\pm 0.1$, i.e., $dN/dM \\propto M^{-0.5}$. ",
        "introduction": "Young open clusters are ideal locations to search for isolated brown dwarfs. Their youth ensures that substellar objects have not yet cooled down to undetectable levels, and the rich stellar populations of the nearest open clusters complement the recent discoveries of cluster brown dwarfs to yield a complete mass function for coeval systems from the substellar domain up to massive stars.  Nearby clusters have been surveyed by various groups in an effort to build statistically significant samples of young brown dwarfs and derive reliable estimates of the substellar mass function. In this contribution, we present the latest results obtained from the CFHT Pleiades wide-field survey (Section 2) and estimates of the lower mass function for several other open clusters (Section 3). We then briefly discuss the potential effects of cluster dynamical evolution on the shape of the mass function (Section 4). ",
        "conclusions": "Deep wide-field photometric surveys of brown dwarfs in nearby young open clusters have yielded estimates of the mass function across the stellar/subtellar boundary. The best studied cluster so far is the Pleiades whose lower mass function can be approximated by a power-law with an exponent $\\alpha = -0.6 \\pm 0.1$ (i.e.  $dN/dM \\propto M^{-0.6}$) over the mass range 0.03-0.4 $\\msun$. Though the determination of the mass function in other Pleiades-age clusters (M35, Alpha Per, NGC 2516, Blanco 1) is not yet as precise as for the Pleiades itself, current estimates suggest that there is no appreciable differences in the shape of the lower MF between the various clusters, regardless of their precise age, metallicity or richness. This might be an indication that the currently measured mass function of these clusters at an age of about 100 Myr is representative of their initial mass function (IMF) and thus provides a quantitative constraint to the formation scenarios for stars and brown dwarfs."
    },
    "0209/astro-ph0209387_arXiv.txt": {
        "abstract": "We study the chemo-dynamical evolution of elliptical galaxies and their hot X-ray emitting gas using high-resolution cosmological simulations. Our Tree N-body/SPH code includes a self-consistent treatment of radiative cooling, star formation, supernovae feedback, and chemical enrichment. We present a series of ${\\rm \\Lambda}$CDM cosmological simulations which trace the spatial and temporal evolution of heavy element abundance patterns in both the stellar and gas components of galaxies. X-ray spectra of the hot gas are constructed via the use of the {\\tt vmekal} plasma model, and analysed using XSPEC with the XMM EPN response function. Simulation end-products are quantitatively compared with the observational data in both the X-ray and optical regime. We find that radiative cooling is important to interpret the observed X-ray luminosity, temperature, and metallicity of the interstellar medium of elliptical galaxies. However, this cooled gas also leads to excessive star formation at low redshift, and therefore results in underlying galactic stellar populations which are too blue with respect to observations. ",
        "introduction": "The hot X-ray emitting gas of elliptical galaxies represents an important interface between galaxies and the intergalactic medium (perhaps even the {\\it primary} interface).  The X-ray halos of ellipticals carry with them two fundamental mysteries: \\begin{itemize} \\item their X-ray luminosities are lower than that expected from an extrapolation of the cluster X-ray luminosity-temperature (${\\rm L_X}-{\\rm T_X}$) relation (e.g. Matsushita et~al. 2000). \\item their X-ray metallicities are lower than that of the mean stellar iron abundance (the so-called ``iron discrepancy'' - e.g. Arimoto et~al. 1997 - a ``discrepancy'' in the sense that the halo gas metallicity was expected to exceed that of the stars, since it should bear the pollution of the enrichment from earlier generation of stars - enrichment byproducts that were not locked up into subsequent stellar generations).\\footnote{X-ray iron abundances remain a controversial issue (c.f. Buote \\& Fabian 1997), although the iron discrepancy appears to hold based upon recent high-resolution XMM RGS observations (Xu et~al. 2002; Sakelliou et~al. 2002).} \\end{itemize} Conversely, the optical properties of ellipticals appear less contentious! The Colour-Magnitude Relation (CMR) and Fundamental Plane provide strong constraints for any elliptical galaxy formation paradigm.  We present here our preliminary work aimed ultimately at the construction of successful self-consistent optical $+$ X-ray cosmological chemodynamical simulations of elliptical galaxies.  \\begin{figure}[t] \\plotone{evol.ps} \\caption{Dark matter density map of a portion of the 43~Mpc (comoving) simulation volume ({\\it upper panels}), and predicted $I$-band image of the target galaxy ({\\it lower panels}), over the redshift range $z$=3.0 to $z$=0.5. } \\label{evol-fig} \\end{figure} ",
        "conclusions": "The left panel of Figure~\\ref{lxt-fig} shows the predicted ${\\rm L_X-T_X}$ relation for the three models at $z$=0; crosses with error bars represent the observational data from Matsushita et~al. (2000). The adiabatic model (Model~A) appears incompatible with the data due to its excessive luminosity and low temperature. The inclusion of radiative cooling leads to lower luminosities and higher temperatures -  as a result, models with cooling (Models B and C) are (roughly) consistent with the ${\\rm L_X-T_X}$ relation of the observed elliptical galaxies. These conclusions are consistent with the analysis of Muanwong et~al. (2001). The right panel of Figure~\\ref{lxt-fig} shows the effect of cooling more clearly. In the gas density versus temperature diagrams, the region above than the line corresponds to a parameter space in which the cooling time is shorter than the Hubble time. Cooling ensures the gas within this region is cold (ie. non X-ray emitting), and of low density and high temperature, ensuring that ultimately radiative cooling drives the observed ${\\rm L_X-T_X}$ relation.  The left panel of Figure~\\ref{lxfecmr-fig} compares the X-ray weighted iron abundance of our simulations with the observational data of Matsushita et~al. (2000).  As the adiabatic model (by construction) does not form any stars (not having any cooling!), we show the results only for Models~B and C.  Both these models show lower gas-phase iron abundance, compared to their stellar abundance, consistent with the low iron abundances observed in the X-ray emitting gas of ellipticals. We find that a large fraction of iron ejected from stars is locked into future generation of stars.  Stars preferentially enrich the gas in the central region, where cooling is efficient (right panel of Figure~\\ref{lxt-fig}). The enriched gas can then cool easily and be incorporated into future generations of stars. Consequently, the hot gaseous halo has not been enriched efficiently, leading to a lower X-ray weighted iron abundance.  In summary, our radiative cooling models explain the two X-ray ``mysteries'' alluded to in Section~1.  Having said that, any successful scenario must also explain the optical properties of the underlying stellar component. To this end, we examined the position of our simulated target galaxy in the observed Coma cluster CMR (Bower et~al. 1992). We can see immediately that the colours of the resulting stellar components of both Models~B and C are inconsistent with the data (being too blue).  This inconsistency can be traced to an excessive population of young and intermediate age stars ($<$8~Gyr) which form from successively cooled gas, regardless of the strength of SNe feedback. If the contribution of these young stars was ignored, the resulting colours would match the observed CMR.  Therefore, one exotic (if somewhat {\\it ad hoc}) solution is to ``hide'' these younger stars within a bottom-heavy initial mass function (IMF) such that they cannot be seen today even if they did exist (e.g. Fabian et~al. 1982; Mathews \\& Brighenti 1999). Another (more plausible) possibility is that extra heating sources, such as intermittent AGN activity, suppress star formation at low redshift.  Before suggesting this is the true solution though, we must re-examine the predicted X-ray properties of the simulation end-products after introducing these additional heating sources; we will be pursuing this comparison in a future paper.    Our cosmological chemodynamical code makes it possible to undertake quantitative comparisons between numerical simulations and observational data in both the X-ray and optical regime with minimal assumptions. We find that radiative cooling is required to explain the observed X-ray luminosity, temperature, and metallicity of elliptical galaxies.  Unfortunately, the resulting cooled gas also leads to unavoidable overproduction of young and intermediate age stellar populations, at odds with the observational constraints. Although a bottom-heavy IMF is one solution for this problem, extra heating by intermittent AGN activity seems to be more plausible (e.g. Brighenti \\& Mathews 2002); recent observations are consistent with this latter picture (e.g. Churazov et~al. 2001)."
    },
    "0209/hep-th0209159_arXiv.txt": {
        "abstract": "We discuss interacting quantum field theory in de~Sitter space and argue that the Mottola-Allen vacuum ambiguity is an artifact of free field theory. The nature of the nonthermality of the MA-vacua is also clarified. We propose analyticity of correlation functions as a fundamental requirement of quantum field theory in curved spacetimes. In de~Sitter space, this principle determines the vacuum unambiguously and facilitates the systematic development of perturbation theory. ",
        "introduction": "The quantum theory of de~Sitter space is important because it plays a central role in cosmology, particularly in the generation of cosmic structure from inflation and in the puzzles surrounding the cosmological constant. More generally, de~Sitter space is useful for testing theoretical ideas because, as a maximally symmetric space, it is tightly constrained. Eventually, we want to understand the full quantum gravity of de~Sitter space, but many interesting questions appear already at the level of quantum field theory (QFT) in a fixed de~Sitter background. In this paper we study the vacuum structure of the theory and develop the features of interacting QFT needed for this purpose.   The starting point of QFT in any background is a mode expansion \\begin{equation} \\phi(X) = \\sum_{n} [ a_{n} u_{n}(X) + a_{n}^{\\dagger}u_{n}^{*}(X)]~, \\label{modeexp} \\end{equation} and the corresponding specification of the vacuum \\begin{equation} a_{n}|{\\rm vac}\\rangle = 0 ~. \\label{vacdef} \\end{equation} The modes {$u_{n}$} must be chosen so that the corresponding vacuum respects the symmetries of the theory. In Minkowski space this principle determines the modes completely and, accordingly, there is a unique Poincar\\'{e} invariant vacuum. In curved spacetime, symmetries do not in general determine the vacuum state completely. Indeed, in de~Sitter space, symmetries identify the vacuum only modulo a two parameter ambiguity, corresponding to a family of distinct de~Sitter invariant vacua. The existence of this ambiguity was first emphasized by Mottola\\cite{Mottola:ar} and Allen\\cite{Allen:ux}.  One vacuum is almost universally taken as the starting point for QFT in de~Sitter space. We will refer to this vacuum as the ``Euclidean vacuum'' and reserve the term ``MA-vacua'' for the alternative, nonstandard vacua. The Euclidean vacuum is singled out by several features: \\begin{itemize} \\item The correlation functions can be obtained by continuation from the Euclidean de~Sitter space, {\\it i.e.} a sphere. This is the origin of the terminology we employ. \\item It coincides with the adiabatic vacuum in the FRW coordinates customarily employed in cosmology. This facilitates a consistent particle interpretation of the theory. In cosmology the Euclidean vacuum is often referred to as the Bunch-Davies vacuum\\cite{Bunch:yq}. \\item The correlations of the field are experienced as precisely thermal by an Unruh detector (a comoving detector linearly coupled to the quantum field). \\item The $2$-point correlation functions reduce to the standard Minkowski propagators when the de~Sitter radius is taken to infinity. \\end{itemize} These properties are desirable both technically and conceptually, but they do not by themselves identify the Euclidean vacuum as the ``right'' vacuum. It could be that the MA-vacua simply have different properties, and that the question of which vacuum is appropriate depends on additional physical input such as boundary conditions, or even observational data. This is the point of view taken in much recent work\\cite{SSV,BMS,SV,danielsson,transplanck,Balasubramanian:2002zh}.  Previous discussions have been at the level of free field theory in fixed background. In most contexts, such as effective field theory, what we are really interested in is the weakly interacting theory; so it is important that higher order interactions can be included, at least in principle. In this paper we argue that this is possible {\\it only} for the Euclidean vacuum. The problems we encounter in the general case are particularly sharp for the loop amplitudes. For these the issue {\\it is not} that amplitudes take values we deem physically unreasonable; rather, {\\it they are ill-defined in the MA-vacua}, a much worse problem. Tree level amplitudes are also problematic even though they are mathematically well-defined: they have unusual, and most likely physically unacceptable, singularities related to antipodal events which, as we discuss, cannot be hidden behind an event horizon. We conclude that {\\it the MA-vacua are artifacts of the free field limit}. Of course, we cannot actually {\\it prove} that no definition of the MA-vacua exists at the interacting level. What we argue is that it is known how to include interactions in the Euclidean vacuum, and, whenever the propagator is not the boundary value of an analytic function, as in these MA-vacua, this prescription does not generalize.  Thus, at the very least, more work is needed to establish the viability  of the MA-vacua.  Relativistic QFT is a tight structure. In addition to symmetries, the interacting theory is constrained by analyticity properties. The principle we need is \\begin{itemize} \\item Correlation functions are boundary values of analytical functions (in the sense of distributions). \\end{itemize} This principle singles out the Euclidean vacuum. Following Bros \\etal\\cite{Bros:1990cu,Bros:dn,Bros:1995js,Bros:1998ik}, it allows one to construct an interacting theory for de~Sitter space that closely mimics QFT in Minkowski space. For example, it ensures the K\\\"all\\'en-Lehman representation for the two-point Green's function with a positive spectral density. The non-existence of an S-matrix, a major confusion clouding the quantum theory of de~Sitter space, can be circumvented, at least for our purposes, by considering correlation functions directly. It is the correlation functions that are observables, measured by Unruh detectors, and it is the correlation functions that satisfy strong analyticity properties. These may be important lessons for formulating the quantum theory of de~Sitter space.  One of the motivations for studying the MA-vacua is their possible applications to cosmology. According to the inflationary paradigm, all structure in the universe ultimately originated from the fluctuations of a scalar field in a de~Sitter background. It has been proposed that physics at extremely high ``trans-Planckian'' energies determines which vacuum is appropriate for this scalar field and thus, using the MA-vacua as interlocutor, the cosmic structure could contain data pertaining to such energies\\cite{danielsson,transplanck}. In this regard our results are unfortunately negative: they indicate that this possibility is illusory, at least in its simplest form.  The remainder of the paper is organized as follows. In section 2 we review the classical de~Sitter geometry, QFT in the de~Sitter background, and the MA-vacua in the free theory. In section 3 we first discuss some general features of interacting QFT in curved spacetimes and singularities of amplitudes. Then we exhibit the problems with the MA-vacua, considering in turn tree level amplitudes and loops. In section 4 we discuss the nature of the non-thermality of the MA-vacua, and examine some of their difficulties from this point of view. Finally, in section 5, we discuss implications of our results as well as future research directions. ",
        "conclusions": "\\label{sect:conclusion}  We have argued that, of the alternate de~Sitter invariant vacua of Mottola and Allen, only the Euclidean vacuum has sufficient analyticity to admit a well-defined perturbation theory.  The singularities of the propagators in other vacua not only have ``unphysical\" singularities associated with the lightcone of antipodal points, but also appear not to permit a definition of loop diagrams in general.  Analyticity is the main ingredient in our considerations. More generally, we note that analyticity is a common denominator of sensible field theories in both flat and curved spacetime: in discussions of the adiabatic vacuum or of cosmologies having asymptotically flat regions, the preferred vacua lead to analytic correlation functions and S-matrix elements \\cite{Birrell:ix,Fulling:nb}. This suggests that {\\it analyticity itself is a unifying principle} for a sensible QFT in analytic, curved spacetime backgrounds, consistent with various other assumptions but subsuming them.  Much of the literature on quantum fields in curved spacetime has dealt with free fields\\cite{Mottola:ar,Allen:ux,Birrell:ix,Fulling:nb,Kay:mu}, and has been concerned with the vexing problem of how to choose the ``right\" no-particle state.  What we are suggesting is that the requirements of constructing a sensible, {\\it interacting} field theory in a curved background may paradoxically simplify the choice by resolving some if not all of the ambiguities that may exist for free fields.  The vacuum for which $n$-point functions obey the requisite analyticity is very likely unique, since, with Euclidean signature, the differential equations for propagators become elliptic rather than hyperbolic.  This requirement would also be consistent with a holographic principle that boundary values uniquely determine the function everywhere. Such a property seems a desirable starting point for formulating the conjectured dS/CFT duality\\cite{Strominger:2001pn} and, more generally, for holography in time-dependent backgrounds.  However, the conjectured implementation of these ideas to date are based on a particular choice of MA vacuum\\cite{BMS,SV,Strominger:2001pn,Strominger:2001gp}, not on the Euclidean vacuum.  The interpretation of singularities of Feynman diagrams in terms of ``on-shell\" particle properties is well-known in Minkowskian spacetime and it would obviously be useful to have a generalization to arbitrary spacetime. A restatement of the Landau rules by Coleman and Norton\\cite{C-N} provides a formulation  amenable to interpretation in coordinate space and potentially applicable to curved spacetime.  Their result is that ``a Feynman amplitude has singularities on the physical boundary if and only if the relevant Feynman diagram can be interpreted as a picture of an energy- and momentum-conserving process occurring in spacetime, with all internal particles real, on the mass shell, and moving forward in time.\"  For the purposes of seeking a similar theorem in curved spacetime backgrounds, we might restate this by saying that ``a Feynman diagram has singularities if and only if the internal lines can be interpreted as classical particles moving on timelike (or null) geodesics between vertices that are causally related.\"  Stated in this way, we may conjecture that it is true for the Euclidean vacuum in an arbitrary, analytic spacetime background \\footnote{The method of proof would have to be rather different from the familiar ones, relying as they do on properties of amplitudes in momentum space.}. In Minkowski space, the Landau rules are a reflection of the completeness of the particle spectrum, which is the essence of unitarity, so the generalization of the Coleman-Norton result to curved spacetime might be interpreted as an expression of the appropriateness of the particle interpretation associated with the Euclidean vacuum. Because of the work of Bros and collaborators, reviewed in section~\\ref{sect:axiomatic}, we are quite confident of the extension of the Coleman-Norton theorem to the Euclidean vacuum for the de~Sitter background. On the other hand, as we described in sections~\\ref{sect:treeampl} and \\ref{sect:loopampl}, the MA-vacua have a completely different singularity structure, requiring a novel, presently unknown, interpretation.  Assuming that string theory underlies quantum gravity and quantum field theory in curved spacetime, can one find any motivation for assumptions such as these? From its inception in the Veneziano model\\cite{veneziano}, a key element in the development of string theory has been the role of analyticity and the association of singularities in scattering amplitudes with particles in physical processes.  It is so much a part of the structure that it is scarcely remarked upon any more.  It is true that string theory to date can only describe S-matrix elements and that the relationship of superstrings to nonsupersymmetric theories is obscure.  Certainly it is not known at this time how to obtain a de~Sitter-like background from string theory.  Nevertheless, it may be anticipated that any effective field theory that comes from string theory will reflect both the analytic structure of Green's functions familiar from QFT in Minkowski space and the association of singularities of Feynman amplitudes with classically realizable processes involving particle propagation, as embodied in our conjectured generalization of the Coleman-Norton theorem. It certainly would be pleasing if this were the case, and it would be even more satisfying if analyticity resolved the thorny problem of how to choose the correct vacuum state, even if only for a large class of curved backgrounds.  These considerations clearly will have implications for cosmology in the very early universe and for physics above the scale relevant to the onset of an inflationary phase if not beyond the Planck scale.  The popular trans-Planckian scenarios\\cite{transplanck} that precede the inflationary era generally employ vacua that are mode-dependent generalizations of the Bogoliubov transformations (\\ref{bogtransf}) leading to the MA-vacua in de~Sitter space. The simplest construction\\cite{danielsson} considers the mode-independent transformation, equivalent to one particular MA-vacuum; it will therefore be beset by many of the difficulties emphasized in this paper, such as the nonthermal character of the background, the noncausal and nonlocal singularities associated with antipodal points, and the difficulties defining loop diagrams. The more general constructions will modify the short-distance behavior of the MA-propagators without changing their singularities at large distances.  Hence, they too will confront problems similar to those encountered for the MA-vacua. Additionally, removing antipodal singularities by modifying the spacetime history does not remedy the problems within the future lightcone.\\footnote{One may try to interpret the ``new'' vacua as excitations on the Euclidean vacuum, rather than truly different vacua; but then one will encounter the well-known problems of infinite rates of particle production\\cite{Fulling:nb}, which will be manifested in large amounts of energy that does not inflate away.  This energy must be accounted for at the end of the inflationary era in the transition to a radiation-dominated cosmology in the usual adiabatic background \\cite{shenker}.}  In summary, our conclusions justify the choice of vacuum made in the effective field theory description of inflation by Kaloper \\etal\\cite{kaloper}  In this paper, we have highlighted what we believe to be serious challenges to defining and interpreting quantum field theory in de~Sitter space in a non-Euclidean vacuum.  We suggest that a greater burden of proof rests on those who would adopt a vacuum in which the propagator is {\\it not} analytic in the usual way.  Their challenge is to show how correlation functions or observables are to be calculated in a well-defined, unambiguous manner consistent with QFT and with macroscopic causality.  \\medskip \\centerline{\\bf Acknowledgments}  The authors thank D.~Chung for helpful discussions during the initial stages of this work.  One of us (MBE) offers thanks to H.~ Rubinstein for stimulating his interest in trans-Planckian scenarios and appreciation to J.~Bros and U.~Moschella for helpful correspondence concerning their work.  He also thanks the theory group of LBL for its hospitality, where a portion of this work was completed.  The other of us (FL) thanks the Aspen Center for Physics for its hospitality as this work was completed, and J.~Cline, R.~Holman, M.~Kleban, A.~Lawrence, H.~Ooguri, A.~Rajaraman, and S.~Shenker for stimulating discussions. This work has been supported in part by the U.S.\\ Department of Energy.  \\bigskip \\noindent {\\bf Note added:}  As this manuscript was being completed, another appeared\\cite{Banks} that also argued that the MA-vacua are unacceptable.  Our arguments include problems at the tree level and, at the loop level, the difficulties we highlight are not specifically tied to the antipodal points. Secondly, although we have not considered the possibility of identification of antipodal sector with the causal sector,\\cite{verlinde} changing de~Sitter space to an RP(N) manifold, the problem with loops is already evident in the mixed $i\\epsilon$ prescription associated with the  singularities along the usual lightcone $Z=1.$  Therefore, we expect to find no alternatives to choosing the Euclidean vacuum in any case. \\vfill"
    },
    "0209/astro-ph0209428_arXiv.txt": {
        "abstract": "{We present direct evidence for CO freeze-out in a circumstellar disk around the edge-on class I object \\object{CRBR 2422.8-3423}, observed in the $M$ band with VLT-ISAAC at a resolving power $R\\simeq 10\\,000$.  The spectrum shows strong solid CO absorption, with a lower limit on the column density of 2.2 $\\times$ 10$^{18}$ cm$^{-2}$.  The solid CO column is the highest observed so far, including high-mass protostars and background field stars.  Absorption by foreground cloud material likely accounts for only a small fraction of the total solid CO, based on the weakness of solid CO absorption toward nearby sources and the absence of gaseous C$^{18}$O $J=2\\rightarrow 1$ emission 30\\arcsec south. Gas-phase ro-vibrational CO absorption lines are also detected with a mean temperature of 50 $\\pm$10~K. The average gas/solid CO ratio is $\\sim$~1 along the line of sight.  For an estimated inclination of 20\\degr $\\pm$ 5\\degr, the solid CO absorption originates mostly in the cold, shielded outer part of the flaring disk, consistent with the predominance of apolar solid CO in the spectrum and the non-detection of solid OCN$^-$, an indicator of thermal/ultraviolet processing of the ice mantle. By contrast, the warm gaseous CO likely originates closer to the star. ",
        "introduction": "Interstellar gas and dust form the basic ingredients from which planetary systems are built (e.g., van Dishoeck \\& Blake \\cite{vDB98}, Ehrenfreund \\& Charnley \\cite{Ehrenfreund00}). In particular, the icy grains can agglomerate in the cold midplane of circumstellar disks to form planetesimals such as comets. In the cold ($T<$~20~K) and dense ($n_{\\rm H} = 10^6-10^9$ cm$^{-3}$) regions of disks, all chemical models predict a strong freeze-out of molecules onto grain surfaces (e.g., Aikawa et al.\\ \\cite{Aikawa02}).  The low molecular abundances in disks compared to those in dense clouds as derived from (sub)millimeter lines is widely considered to be indirect evidence for freeze-out (Dutrey, Guilloteau \\& Gu\\'elin\\ \\cite{Dutrey97}; Thi et al.\\ \\cite{Thi01}).  Observations of gaseous and solid CO have been performed for a few transitional objects from class I to class II that are known to posses a disk. Boogert et al.\\ (\\cite{Boogert02a}) observed \\object{L~1489} in Taurus -- a large 2000 AU rotating disk --, but the amount of solid CO is not exceptionally high ($\\sim$7\\% of gaseous CO). This may stem from the fact that these systems are still far from edge-on (inclination $\\sim 20{\\degr}-30{\\degr}$) so that the line of sight does not intersect the midplane, the largest reservoir of solid CO. Shuping et al.\\ (\\cite{Shuping01}) found strong CO depletion toward \\object{Elias 18} in Taurus, but both the disk stucture and its viewing angle are not well constrained.  More promising targets are pre-main-sequence stars for which near-infrared images have revealed nebulosities separated by a dark lane (e.g. Padgett et al.\\ \\cite{Padgett99}).  The lane is interpreted as the cold midplane of a disk seen close to edge-on where visible and even near-infrared light are extremely extinct. Among such dark-lane objects, \\object{CRBR 2422.8-3423} is a red ($H-K$=4.7) low luminosity ($L=0.36~L_{\\sun}$, Bontemps et al.\\ \\cite{Bontemps01}) object surrounded by a near edge-on disk, discovered in images with the ESO {\\it Very Large Telescope} (VLT) at 2 $\\mu$m (Brandner et al.\\ \\cite{Brandner00}). Its spectral energy distribution (SED) is consistent with that of a class I object or an edge-on class II object with strong silicate absorption at 9.6~$\\mu$m. It is located at the edge of the \\object{$\\rho$ Oph cloud} complex in the core F, $\\sim$ 30\\arcsec west of the infrared source \\object{IRS~43} and a few arcmin south-east of \\object{Elias 29} (Motte, Andr\\'e \\& Montmerle \\cite{Motte98}).   This letter reports the detection of a large quantity of solid CO and the presence of gaseous CO in the line of sight of \\object{CRBR 2422.8-3423} using the ESO-VLT (\\S 2 and 3). Possible contamination by foreground cloud material is considered in \\S~\\ref{discussion}, followed by a discussion on the location and origin of the CO gas and dust in the disk (\\S 5). ",
        "conclusions": "\\begin{figure} \\resizebox{\\hsize}{!}{\\includegraphics[]{Eh151_f3.eps}} \\caption{Fraction of CO molecules in the solid phase at different temperatures as function of density. Adsorption onto H$_2$O and CO ice produce different results. The desorption energies are from Sandford \\& Allamandola (\\cite{Sandford88a}) for CO-H$_2$O and Sandford et al.\\  (\\cite{Sandford88b}) for CO-CO.} \\label{fig_co_fraction} \\end{figure}  A simple disk model cf.\\ Chiang \\& Goldreich (\\cite{CG99}) with $T_*$=3500~K and a disk radius of 250~AU has been adopted to investigate the location of the solid and gaseous CO seen in infrared absorption.  Because of its higher temperature, the gaseous CO is likely not co-located with the bulk of the apolar solid CO, which evaporates at $\\sim$~20~K. The polar solid CO can however reside in the same region of the disk as the CO gas at 40--60~K.  Assuming that CO is frozen out at $<$~20~K (apolar CO) and $<$~40~K (polar CO), the best fit to the column densities, gas/solid CO and polar ice/apolar ice ratio is obtained for $i$=20$\\pm 5\\degr$.  This inclination is consistent with the flux asymmetry seen in the near-infrared image of Brandner et al.\\ (\\cite{Brandner00}). For such line of sight, most of the CO ice is located above the midplane in the outer disk, whereas the CO gas is found in the warm inner disk.  Thus, the overall CO depletion could be significantly higher than the ratio of $\\sim$~1 found here.  Several time-dependent chemical models were run to quantify the gas/solid CO ratios in different density and temperature regimes.  The models simulate gas-phase chemistry, freeze-out onto grain surfaces and thermal as well as non-thermal evaporation.  Cosmic-ray induced desorption is modeled as in Hasegawa \\& Herbst (\\cite{Hasegawa93}), which may be an overestimate for large grains in disks (Shen et al., in prep.).  The sticking coefficient was set at 0.3 to account for other non-thermal mechanisms (e.g., Schutte \\& Greenberg \\cite{Schutte97}) and photodesorption is assumed ineffective.  At $T>T_{\\rm evap}$ thermal desorption dominates, whereas at $T<T_{\\rm evap}$ cosmic-ray desorption prevails.  In the model, $T_{\\rm evap}$ is 20~K for apolar CO ice ($E_{\\rm des}$=960~K) and 40~K ($E_{\\rm des}$=1740~K) for polar CO. To illustrate the effects of thermal and cosmic-ray induced desorptions, Fig.\\ \\ref{fig_co_fraction} shows the gas/solid CO ratio at chemical equilibrium. In cold ($T<$~20~K) but moderately dense (10$^4-10^5$ cm$^{-3}$) regions of the disk around \\object{CRBR 2422.8-3423}, only a small amount of CO is depleted onto grains in the apolar form, but much larger fractions $>$~50\\% can occur at higher densities ($\\sim 10^6-10^8$ cm$^{-3}$).  In summary, we detected a large amount of solid CO in the line of sight toward \\object{CRBR 2422.8-3423}. The majority of this ice is likely located in the flaring outer regions of the edge-on circumstellar disk.  Very high resolution ($R>10^5$) near-infrared spectroscopy is needed to reveal the gaseous CO line profiles and thus their origin and the gas dynamics in the inner disk.  Future submillimeter interferometer data can probe the velocity pattern and excitation conditions of the gas as functions of disk radius, whereas mid-infrared spectroscopy with, e.g., the {\\it Space Infrared Telescope Facility} (SIRTF) will allow searches for other ice components, in particular solid CO$_2$."
    },
    "0209/astro-ph0209102_arXiv.txt": {
        "abstract": "We fit the broad-band RXTE PCA and HEXTE spectrum from 3--200 keV with reflection models which calculate the vertical ionization structure of an X-ray illuminated disc. We consider two geometries corresponding to a truncated disc/inner hot flow and magnetic flares above an untruncated disc. Both models are able to fit the PCA 3--20 keV data, but with very different spectral components.  In the magnetic flare models the 3--20 keV PCA spectrum contains a large amount of highly ionized reflection while in the truncated disc models the amount of reflection is rather small. The Compton downscattering rollover in reflected emission means that the magnetic flare models predict a break in the spectrum at the high energies covered by the HEXTE bandpass which is {\\em not} seen. By contrast the weakly illuminated truncated disc models can easily fit the 3--200 keV spectra. ",
        "introduction": "The black hole binary (BHB) systems in their low/hard state have spectra dominated by a power law which rolls over at $\\sim 200$ keV. This continuum is generally well fit by thermal Comptonization models, in which low energy photons from the accretion disc are upscattered by energetic electrons (e.g. the review by Zdziarski 2000). To get these hard X-rays, a large fraction of the gravitational energy released by accretion must be dissipated in an optically thin environment i.e. not in the disc itself. However, there is no consensus on how this happens, or on the geometry of this hot region. There are currently two main models, one in which the hot electrons are confined in magnetic flares above a disc which extends down to the last stable orbit around the black hole (magnetic flares), and one in which the electrons form an quasi-spherical hot flow, replacing the inner disc (truncated disc).  The magnetic flare model is motivated by the discovery that the physical mechanism for the disc viscosity is a magneto-hydrodynamic dynamo (e.g. the review by Balbus \\& Hawley 2002). Buoyancy could cause the magnetic field loops to rise up to the surface of the disc, so they can reconnect in regions of fairly low particle density, forming a patchy corona. Numerical simulations (although these are highly incomplete as in general the simulated discs are not radiative) do show this happening (Hawley 2000), but they do not yet carry enough power to reproduce the observed low/hard state (Miller \\& Stone 2000).  The truncated disc model has its physical basis in the accretion flow equations. The standard Shakura-Sunyaev disc solution assumes that the accreting material is at one temperature (protons and electrons thermalize) and that the accretion energy released by viscosity is radiated efficiently. At low mass accretion rates neither of these are necessarily true. The thermalization timescale between the electrons and protons can be long, so the flow is intrinsically a two temperature plasma. Where the electrons radiate most of the gravitational energy through Comptonization of photons from the outer disc then the hot inner flow is given by Shapiro, Lightman \\& Eardley (1976). Alternatively, if the protons carry most of the accretion energy into the black hole then this forms the advection dominated accretion flows (Narayan \\& Yi 1995). These are related, as in general both advection and radiative cooling are important for a hot accretion flow (Chen et al. 1995; Zdziarski 1998).  These two models of the accretion flow have very different geometries. A potential way to test the geometry is with X-ray reflection.  The amount of reflection scales with the solid angle subtended by the optically thick material while the relativistic smearing of the atomic features shows how far this material extends into the gravitational potential (Fabian et al.  2000). With magnetic flares, the disc subtends a solid angle of $2\\pi$ as seen from the X-ray source, and extends down to the last stable orbit. Its reflected spectrum should be large and strongly smeared. Conversely, a truncated disc illuminated by an inner hot flow subtends a solid angle $\\le 2\\pi$, and the reflected spectrum is small and is only weakly smeared by relativistic effects.  The BHB spectra in the low/hard state show overwhelmingly that the solid angle is significantly less than $2\\pi$, and that the smearing is less than expected for a disc extending down to the last stable orbit (\\Zycki\\ Done \\& Smith 1997; 1998; 1999; \\Gierlinski\\ et al. 1997; Done \\& \\Zycki\\ 1999; Zdziarski et al. 1999; Gilfanov, Churazov \\& Revnivtsev 1999; 2000).  While this is clearly consistent with the idea that the disc is truncated in the low/hard state, the magnetic flare models can be retrieved in several ways.  Firstly, magnetic reconnection on the Sun is known to produce an outflow in the coronal mass ejection events. In the extreme conditions close to the black hole it is possible that this outflow velocity could be large so that the hard X--ray radiation is beamed away from the inner disc (Beloborodov 1999).  An alternative explanation for the lack of reflection and smearing is that the inner disc or top layer of the inner disc is completely ionized. There are then no atomic features, and the disc reflection is unobservable in the 2--20 keV range as it appears instead to be part of the power law continuum (Ross \\& Fabian 1993; Ross, Fabian \\& Young 1999).  However, these models with passive illumination of the disc require a fairly sharp transition between the extreme ionization and mainly neutral material (Done \\& \\Zycki\\ 1999; Done, Madejski \\& \\Zycki\\ 2000; Young et al. 2001). Such a transition can be produced as the disc {\\em responds} to the intense X--ray illumination. There is a thermal ionization instability which affects X--ray illuminated material in pressure balance, which can lead to a hot, extremely ionized skin forming on top of the rest of the cooler, denser, mainly neutral disc material (Field 1965; Krolik, McKee \\& Tarter 1981; Kallman \\& White 1989; Ko \\& Kallman 1994; \\Rozanska\\ \\& Czerny 1996; Nayakshin, Kazanas \\& Kallman 2000, hereafter NKK; \\Rozanska\\ \\& Czerny 2000; Nayakshin \\& Kallman 2001; Ballantyne, Ross \\& Fabian 2001). Such X-ray illuminated disc models can fit the 2--20 keV data from BHB with reflection from a disc which subtends a solid angle of $2\\pi$ and extends down to the last stable orbit around a black hole (Done \\& Nayakshin 2001b).  While the truncated disc and magnetic flare/X-ray illuminated disc models are indistinguishable with current data in the 2--20 keV range, they are very different at higher energies. In the magnetic flare models, a lot of the 2--10 keV 'continuum' is actually ionized reflection, so the true continuum level is lower than in the truncated disc models, in which reflection is small. At higher energies, where reflection is negligible, the lower continuum level for the magnetic flare models leads to a smaller predicted flux in the 100-200 keV range than for the truncated discs (Done \\& Nayakshin 2001a). Here we fit both truncated disc and magnetic flare/X-ray illuminated disc models to the 3--200 keV PCA/HEXTE spectrum of the hard/low state spectra of Cyg~X-1. We show that the truncated disc models provide an excellent fit to these data, but that the magnetic flare/X-ray illuminated disc does not, as it dramatically underpredicts the 100--200 keV spectrum.  Similar conclusions were independently reached by Maccarone \\& Coppi (2002) from fits to the Cyg X-1 broad-band spectrum. However, they approximated the reflection from magnetic flares by highly ionized, single zone reflection models rather than the full complex ionization reflection models used here. Here we are able to show explicitally that the complex ionization magnetic flare models do not fit the high energy spectrum, while the truncated disc models do. ",
        "conclusions": "The data clearly show that the 2--20 keV spectrum from the low/hard state of Cyg X-1 does {\\em not} contain a large fraction of highly ionized reflection. This rules out models which have static magnetic flares above an untruncated disc unless the flares have a spectrum which is much harder than that predicted by a single temperature Comptonization model.  One way to get some spectral hardening which is expected in a magnetic flare geometry but is neglected in our modelling is for the flares to comptonize some fraction of the reflected photons.  The reflected photons are from the disc, so are intercepted by the hot electrons in the flares in the same way as the soft seed photons, and are Compton upscattered to form a hard continuum.  With flares covering most of the disc then this can make a $50-100$ per cent increase to the flux at 200 keV (Petrucci et al. 2001). A covering fraction of unity for the flares (slab corona geometry) is normally ruled out for the low/hard spectra as this produces many non-reflected, thermalized photons from the hard X-ray illumination. These are also Comptonized by the corona, leading to spectral indices which are too soft to explain the low/hard state (Pietrini \\& Krolik 1995; Stern et al. 1995; Zdziarski et al. 1998).  However, a covering fraction of $\\sim 0.5$ might give enough Compton scattering of reflection to produce the required $\\sim 50$ per cent excess emission at 200 keV, while also allowing enough seed photons to escape to produce the required hard continuum.  The problems that the magnetic flare models have in matching the high energy flux are exacerbated by the anisotropy break which should be present in the continuum in this assumed geometry, but which has never been convincingly observed (\\Gierlinski\\ et al. 1999).  If the magnetic flares are to fit the high energy data then the anisotropy break must be hidden by having a multiple temperature Comptonised continuum. At some level there {\\em must} be a distribution of electron temperatures: it is almost inconceivable that a single temperture distribution can be maintained, especially as the sources are variable so the flare spectra probably evolve with time (e.g. Poutanen \\& Fabian 1999). A multiple temperature continuum gives another way to boost the high energy flux so that we do not necessarily need to strongly Comptonise the reflected continuum. However, it still seems somewhat contrived that a combined complex continuum plus complex ionization reflection spectrum should so precisely mimic a simple single temperature continuum, truncated disc reflection.  By contrast, a truncated disc/hot inner flow geometry at low mass accretion rates is compatible with the observed hard continuum, lack of anisotropy break, low amount of reflection and relativistic smearing.  It can also explain the low temperature and luminosity of the direct emission from the disc (e.g. Esin et al. 2000). This geometry can also give a qualitative explanation for a range of observed correlations if the truncation radius decreases with increasing (average) mass accretion rate.  The disc penetrates further into the hot flow, increasing the seed photon flux intercepted by the hot inner flow, leading to a softer continuum spectra.  This changing geometry gives a larger solid angle subtended by the disc, leading to an increasing amount of reflection (Poutanen Krolik \\& Ryde 1997; Zdziarski et al. 1999; Gilfanov et al. 1999; 2000), and relativistic smearing (\\Zycki\\ et al. 1999; Gilfanov et al. 2000; \\Lubinski\\ \\& Zdziarski 2001). The variability power spectra are also affected as they contain characteristic frequencies which are most probably linked to the inner edge of the disc, so this can explain the correlated increase in break and quasi-periodic oscillation frequency (e.g. the review by van der Klis 2000; Churazov, Gilfanov \\& Revnivtsev 2001). Lastly, the collapse of an inner hot flow when it becomes optically thick gives a physical mechanism for the state transition (Esin McClintock \\& Narayan 1997).  Thus, if one were to choose between relatively straightfoward models, then the truncated disk is definitely favored by our analysis while the magnetic flare model is ruled out. However, the straightforward solutions may be too simple to describe the complexity of accretion disk structure near the black hole. The role and magnitude of secondary effects (comptonization of the reflection component; multi-temperature nature of flares) not taken into account in our modelling needs to be clarified in the future with detailed calculations."
    },
    "0209/astro-ph0209334_arXiv.txt": {
        "abstract": "{ We present the generalization of the Sedov-Taylor self-similar strong spherical shock solution for the case of a central energy source varying in time, $E=A t^k$, where $A$ and $k$ are constants. The known Sedov-Taylor solution corresponds to a particular adiabatic case of $k=0$ or \\emph{instant shock} with an instant energy source of the shock, $E=A$. The self-similar hydrodynamic flow in the nonadiabatic $k\\neq0$ case exists only under the appropriate local entropy (energy) input which must be supported by some radiative mechanism from the central engine. The specific case of $k=1$ corresponds to a permanent energy injection into the shock, or \\emph{injection shock} with a central source of constant luminosity, $L=A$, $E=A t$.  The generalized self-similar shock solution may be applied to astrophysical objects in which the duration of central source activity is longer than the shock expansion time, e.~g. the early phase of SN explosions, strong wind from stars and young pulsars, non-steady spherical outflow from black holes and collapsing dense stellar clusters with numerous neutron star collisions. ",
        "introduction": "The well known spherical shock solution \\citep{sed46,neu47,tay50,sta69} describes the self-similar expansion of a strong spherical shock generated by instant deposition of energy $E=const$ by the central source in a homogeneous gas medium with density $\\rho_1=const$. This \\emph{instant shock} solution is commonly used in numerous astrophysics applications e.g. for the modeling of SN explosions and evolution of young SN remnants. The corresponding solution for a strong ultra-relativistic blast wave was obtained by \\citet{bla76}. For recent reviews of astrophysical shock models see \\citet{ost88} and \\citet{bis95} and references therein. The basic requirement for the realization of the instant shock solution is a short-duration injection of energy $E$ into the shock.  However there exist possible physical situations of permanent injection of energy into the expanding shock which we call the \\emph{injection shock},  when a central source has some time-varying luminosity $L=L(t)$. More exactly the instant shock solution is not applicable when the duration of energy generation by the central source $t_\\mathrm{s}$ is comparable or exceeds the shock expansion time $t_\\mathrm{sh}$, $t_\\mathrm{s}\\geq t_\\mathrm{sh}$. The condition $t_\\mathrm{s}\\geq t_\\mathrm{sh}$ is typical for the early stage of a SN explosion, powerful wind from stars and non-steady spherical accretion onto compact objects. The other example is the injection shock produced in \\emph{hidden neutrino sources} \\citep{ber01} by successive multiple fireballs after numerous neutron star collisions in the dense stellar cluster in a galactic nucleus prior to its collapse into a massive black hole. Below we derive the extension of the Sedov-Taylor self-similar spherical shock solution to the case of varying in time energy injection by the central source of power form $E=A t^k$, where $A$ and $k$ are constants. The notations and logistics of ``Fluid Dynamics'' by \\citet[Chapter X, \\S106]{ll59} are used in the self-similar solution derivation.  Let us  consider a strong expanding spherical shock in an ideal gas with polytropic index (Poisson parameter) \\begin{equation} \\gamma=c_\\mathrm{p}/c_\\mathrm{v}=const, \\end{equation} where $c_\\mathrm{p}$ and $c_\\mathrm{v}$ are the gas heat capacities under constant pressure and volume respectively.  All values on the forward side of the shock discontinuity surface (non-perturbed gas side) are designated by index 1, e.g.  $\\rho_1$, $p_1$, and behind the discontinuity surface (shock cavity side) by index 2, e.g. $\\rho_2$, $p_2$.  In a strong shock the pressure behind the shock $p_2$ far exceeds the pressure in the non-perturbed gas $p_1$.  The precise definition for a strong shock (determined from the shock adiabat) is $p_2/p_1\\gg(\\gamma+1)/(\\gamma-1)$ and is similar to condition $u_1>>c_\\mathrm{s}$, where $u_1$ is the shock expansion velocity with respect to the non-moving (non-perturbed) gas and $c_\\mathrm{s}=(\\gamma p/\\rho)^{1/2}$ is the sound speed in the non-moving gas. The following relations are valid for a strong shock discontinuity: \\begin{equation} v_2= \\frac{2}{\\gamma+1}u_1, \\quad \\rho_2=\\frac{\\gamma+1}{\\gamma-1}\\rho_1, \\quad p_2=\\frac{2}{\\gamma+1}{\\rho_1 u_1^2}. \\label{surf} \\end{equation} These relations are the (outer) boundary conditions for our problem. ",
        "conclusions": "The derived solution is the generalization of the Sedov-Taylor self-similar strong spherical shock solution for the case of an energy injection from the central source of form $E=A t^k$, where $A$ and $k$ are constants. The power-law ansatz $E=A t^k$ only is enough for deriving the scaling law for shock radius (\\ref{Rt}) and shock expansion velocity (\\ref{u1}) accurate to within the numerical constant $\\beta(\\gamma,k)\\sim1$ without knowing the exact solution. The numerical value of this constant (see Table~\\ref{Table1}) can be calculated from equation Eq.~(\\ref{Egen}) only after the complete solving of the self-similar problem. The special case of $k=0$ corresponds to the known Sedov-Taylor solution, while the case $k=1$ corresponds to permanent energy injection into the shock by a central source of constant luminosity. The cases with $k<-1$ seem to be nonphysical due to the total energy divergence at $t\\to0$.   The self-similar hydrodynamic flow in the nonadiabatic $k\\neq0$ case exists only under the self-consistency condition~(\\ref{s2}) for the local entropy input. In other words the self-similar behavior of an expanding shock in the nonadiabatic $k\\neq0$ case is realised only under the appropriate tuning of local entropy (energy) source according to Eq.~(\\ref{s2}). This is the auxiliary physical condition which supposes some radiative mechanism for sustained energy supply from the central source into the shocked gas (which depends on the detailed properties of the cental engine, radiative transfer, gas composition etc). It can be seen from Figs.~\\ref{Fig1} and \\ref{Fig4} that the main part of the energy is injected near the outer boundary of the shock at $\\xi\\geq0.8$, i.~e. at the same place where the shocked gas is mainly gathered. The similarities of the profiles for density and entropy rate are in favor of the principal realization of the required tuning of the local energy injection mechanism if the central source radiation absorption would be proportional to the gas density.  The self-similar shock solution with energy injection may be applied to the modeling of astrophysical objects in which duration of central source activity is longer than shock expansion time, such as the early phase of SN explosion, strong wind from stars and young pulsars, non-steady spherical outflow from black holes and collapsing dense stellar clusters with numerous neutron star collisions."
    },
    "0209/astro-ph0209044_arXiv.txt": {
        "abstract": "We present evolutionary population synthesis models for the study of the cool and luminous intermediate age stellar populations in resolved galaxies with particular emphasis on carbon star populations. We study the effects of the star formation history, the age and the metallicity on the populations of intermediate mass stars. In the case of instantaneous bursts, we confirm that lower metallicity results in higher contributions of carbon stars to the total star number, and in higher number ratios of carbon stars to late-type M stars.  Chemically consistent models are used to study the effect of the star formation history on the relations between carbon star population properties and global parameters of the parent galaxy (age, metallicity). Our models are able to account, for the first time, for those correlations, as observed in the galaxies of the Local Group. For stellar populations older than about 1 Gyr, the properties of carbon star populations are linked to the current metallicity in a way that is quite independent of the star formation scenario. The number ratio of carbon stars to late-type M stars forms a metallicity sequence along which stellar populations with very different star formation histories are found.  For the same populations, we find that both the mean bolometric luminosity of carbon stars and their normalized number to the luminosity of the parent galaxy are quite independent of metallicity over a large range in metallicity. This is in good agreement with the observational constraints.  The observed statistics of carbon star populations can be interpreted by two principal effects: (i) the carbon star formation efficiency is higher in metal-poor systems, (ii) the typical star formation timescale along the Hubble sequence of galaxies is much longer than the typical timescale for the production of carbon stars at any metallicity. ",
        "introduction": "\\label{intro.sec} The stellar content of galaxies has long been recognized as holding important clues to the understanding of the formation and the evolution of galaxies. Better understanding of galaxy evolution in terms of stellar populations and chemical evolution requires knowledge of the Asymptotic Giant Branch (AGB) stars. From a purely pragmatic point of view, the evolved AGB stars are easy targets to observe in external galaxies even far beyond the Local Group, and to segregate from the bulk of the stellar population: they are red and luminous sources. Stars in the AGB phase make a significant contribution to the integrated light of a stellar population (Frogel et al. 1990). Mouhcine \\& Lan\\c{c}on (2002) estimate that the thermally pulsing AGB stars (TP-AGBs) are responsible for 30--60\\% of the integrated JHK luminosities of a system whose stars are 0.2-2\\,Gyr old, with a maximum contribution at $\\sim$\\,0.8-1\\,Gyr. Those properties make AGB stars useful tools to get information on the star formation (SF) history, even through several magnitudes of absorption (Hodge 1989). Due to their large luminosities, carbon stars of the AGB are used as tracers of kinematics to probe morphological and kinematical structures (Aaronson \\& Olszewski 1987, Hardy et al. 1989, Kunkel et al. 1997, Graff et al. 2000).  Stars in the AGB phase are classified as either oxygen rich (C/O\\,$<$\\,1 by number) or carbon rich (C/O\\,$\\ge$\\,1). Spectra of oxygen rich stars are dominated by metal oxyde bands such as TiO, VO, and H$_{2}$O whereas carbon stars have bands of C$_{2}$ and CN (Barnbaum et al. 1996, Joyce 1998). Using those features, groups led by Richer (Richer et al. 1984, Richer et al. 1985, Pritchet et al. 1987, Hudon et al 1989), Aaronson and co-workers (Aaronson et al. 1982, Aaronson et al. 1984, Cook et al. 1986, Aaronson et al. 1985, Cook \\& Aaronson 1989) developed a technique to identify AGB stars and to determine their nature in crowded fields. This technique involves imaging a field though four filters. Two narrow-band filters provide spectral information on the CN and TiO bands (Wing 1971), while broad-band colours provide information on the effective temperature of the stars, and can discriminate between early type and late-type stars. Extensive observational work, using this technique and others (Frogel \\& Richer 1983, Azzopardi et al. 1985, 1986, 1998; Azzopardi \\& Lequeux 1992; Westerlund et al. 1987, 1991\\,a \\& b, 1995; Demers et al. 1993, Battinelli \\& Demers 2000) has lead to the identification and classification of AGB stars in nearby galaxies. Recent comprehensive compilations of late-type stellar contents, and of their systematic statistics, can be found in Mateo (1998), Groenewegen (1999) and Azzopardi (2000).  In order to interpret those findings, we have developed a chemically consistent population synthesis model that includes the various phenomena relevant to the production of carbon stars, and  their dependence on metallicity. We use it to evaluate, qualitatively and quantitatively, (i) the interplay between different physical processes that operate during the AGB phase and control the formation of carbon stars, (ii) how these processes lead to the observed statistics of carbon star populations in the galaxies of the Local Group, and (iii) the sensitivity of the carbon star statistics on the global properties of the host galaxies (SF history, evolutionary status). The paper is organized as follows. In Section 2 we present our current observational knowledge of the statistics of carbon star populations in the Local Group. In Section 3, we describe the calibrated semi-analytical evolution models that we use to follow the evolution of the TP-AGB stars. We also describe the population synthesis and the chemical evolution models. In Section 4 and 5, we present the predicted statistics of carbon stars for single stellar populations (SSPs) and address the effect of opaque dust envelopes on those statistics. In Section 6 we calculate the statistics for continuous SF histories representative of various morphological types, and compare our results with the observational data. In Section 7, we come back to the use of the relative number of carbon stars as an abundance indicator. The conclusions are summarized in Section 8. ",
        "conclusions": "\\label{Concl.sec}  In this paper we have provided a theoretical investigation of carbon star populations and of related statistical properties as a function of metallicity, and we have confronted these models with available observations in Local Group galaxies. To achieve this goal we have constructed evolutionary synthesis models that use a large grid of stellar evolution tracks, including the TP-AGB. The underlying TP-AGB models take into account the processes that determine the evolution of those stars in the HR diagram and establish the duration of the TP-AGB phase. To provide predictions for the nature of the cool and luminous intermediate age stellar content of stellar populations (carbon rich or oxygen rich stars, optically visible or dust-enshrouded ones), our models account explicitly for the metallicity dependence of the evolution of TP-AGB stars. We thus derive estimates of the carbon star formation efficiency and its evolution as a function of metallicity for large grid of metallicity from Z/Z$_{\\odot}$\\,=\\,1/50 to Z/Z$_{\\odot}$\\,=\\,2.5.b Note that because fundamental physical processes that govern TP-AGB evolution remain poorly understood and are calibrated in the solar neighbourood or the Megellanic Clouds, uncertainties grow when moving away from this metallicity domain.  The evolution of carbon star properties as a function of metallicity for stellar systems with continuous star formation was modeled via new chemically consistent population synthesis models. Those models use metallicity-dependent stellar yields available in the literature for massive stars, and our synthetic TP-AGB evolution models for the new metallicity-dependent yields of intermediate mass stars. The effect of SNe Ia on the chemical evolution is also included. Effect of gas infall and of the star formation history were explored.  Comparisons between predicted carbon star population statistics as a function of the metallicity of the interstellar medium and the available data show that our models are, for the first time, able to reproduce qualitatively and quantitatively the observations in the Local Group galaxies. This success supports the choices made in the inputs of the evolutionary models. The models show that the evolution of the properties of carbon star populations are established by a combination of the following effects: \\begin{itemize} \\item The temperature of the giant branch varies with metallicity, leading to more late type M stars with increasing metallicity. \\item In metal-poor systems, dredge-up events are more efficient in producing carbon stars and carbon rich lifetimes are longer. This is due to (i) higher core masses at the onset of the TP-AGB phase and the resulting earlier onset of third dredge-up events, and (ii) longer interpulse periods leading to more violent third dredge-up events. \\item The time needed for a stellar population to form the bulk of its carbon stars ($\\sim 1$\\,Gyr) is significantly shorter than the typical evolutionary timescale along the Hubble sequence. This means that for systems older than $\\sim 1$\\,Gyr the evolution of the statistical properties of carbon star populations will merely reflect the sensitivity of the evolution of individual TP-AGB stars to metallicity. \\end{itemize}  The last effect implies that the observed statistics are mainly determined by the current metallicity. Star formation history or any other process which has a timescale longer than about 1\\,Gyr have no significant effects on the evolution of the statistics of carbon stars as function of metallicity. This means that the observed N$_{C}$/N$_{M5+}$ vs. [Fe/H] correlation is a metallicity sequence rather than an age sequence. Consequently, we predict that, at least for unbarred spirals or noninteracting galaxies, the radial profile of N$_{C}$/N$_{M5+}$ ratio will have the same slope as the radial abundance profile.  We have also given estimates on the fraction of carbon stars that may be missed in optical surveys, showing that at the age when the bulk of carbon stars are formed ($\\sim$0.8 Gyr after the burst), only 10\\%-20\\% are dust-enshrouded. We argue that the statistics of carbon star populations will not be affected dramatically by the missed high mass-loss rate stars.  The models show that, for stellar systems older than $\\sim\\,0.8-1\\,$Gyr, carbon star populations have the following properties:  \\begin{itemize} \\item The mean bolometric luminosity of the carbon stars is independent of metallicity. Such a behavior over a wide range of metallicity is consistent with a long record of observational constraints discussed in the literature (Aaronson \\& Mould 1985, Richer et al. 1985). The value derived from our calculations is $<\\!M_{Bol,C}\\!>\\,=\\,-4.7$. Carbon stars can be considered as potential distance indicators. \\item The number of carbon stars normalized to the luminosity of the parent stellar system is independent of metallicity over a wide range in abundance. This behavior supports the interpretation of the anti-correlation between N$_{C}$/N$_{M5+}$ and [Fe/H] as due, partially, to more efficient carbon star formation at lower metallicity. The value derived from our calculations is $\\log(N_{C,L})\\,\\simeq\\,-3$. This is consistent with the observational constraints discussed recently by Azzopardi et al. (1999), where the authors report that the observed value is independent of metallicity and dispersed around -3.3. \\end{itemize} We note that our models were able to reproduce the observed statistics considering only the carbon stars formed via the third dredge-up channel, which means that faint, dwarf carbon stars will represent a small number fraction of the whole population of carbon stars."
    },
    "0209/astro-ph0209272_arXiv.txt": {
        "abstract": "We find that the infrared excess around HD 233517, a first ascent red giant, can be naturally explained if the star possesses an orbiting, flared dusty disk.  We estimate that the outer radius of this disk is ${\\sim}$45 AU and that the total mass within the disk is ${\\sim}$0.01 M$_{\\odot}$.   We speculate that this disk is the result of the engulfment of a low mass companion star that occurred when HD 233517 became a red giant. ",
        "introduction": "The  evolution of dust disks into large solid bodies such as asteroids, planets and comets is of central importance in astronomy.  While it is usually assumed that such disks are primordial and associated with young stars,  some binary systems such as the Red Rectangle apparently create orbiting  reservoirs of gas and dust during their post main sequence evolution (see Waters et al. 1993, Jura \\& Kahane 1999).  To date, there are only a few  good candidate systems  for orbiting dusty circumstellar material around evolved stars, and therefore every example of this phenomenon is worth studying.  Here, we propose that the puzzling infrared excess around HD 233517, a first ascent red giant, can be naturally understood as resulting from a flared, orbiting disk.  Attention was originally drawn to HD 233517 because it is a K type star (m$_{V}$ = 9.72 mag) that is an IRAS source with an unusually large fractional infrared excess ($L_{IR}/L_{*}$ ${\\sim}$ 0.06, Sylvester, Dunkin \\& Barlow 2001) produced by  cold grains ($T$ ${\\sim}$ 100 K), and it was thought to be an example of the Vega phenomenon -- a main sequence star with dust (Walker \\& Wolstencroft 1988). However, detailed spectroscopic studies strongly suggest that HD 233517 is a first ascent red giant with a luminosity near 100 L$_{\\odot}$ (Fekel et al. 1996, Balachandran et al. 2000, Zuckerman 2001).  While infrared excesses are common around second ascent red giants on the Asymptotic Giant Branch  with luminosities in excess of 1000 L$_{\\odot}$ (see, for example, Habing 1996),   relatively few  first ascent red giant stars with their lower luminosities  have detectable infrared excesses  (Judge, Jordan \\& Rowan-Robinson 1987, Zuckerman, Kim \\& Liu 1995). HD 233517 has a uniquely high value of $L_{IR}/L_{*}$ among first ascent red giants where this ratio is generally less than 10$^{-3}$ (see Zuckerman et al. 1995).  It is unlikely that HD 233517 is a pre-main sequence star since it does not lie near any known region of star formation and since it lies far from the location of young stars in the H-R diagram  (Fekel et al. 1996, Balachandran et al. 2000).  Also, HD 233517 appears to be similar to a few other K giants  which rotate rapidly, have strong lithium lines, a detectable far-infrared excess and are definitely post main sequence since at least in the case of PDS 365, the carbon isotope ratio, $^{12}$C/$^{13}$C, is approximately 12 (Drake et al. 2002).  Three models to explain infrared excesses around first ascent red giants are (1) dust is being produced by mass loss, (2) nearby interstellar dust is illuminated accidentally by the red giant, and (3) the  dust is orbiting (see Jura 1999, Kalas et al. 2002).  The fraction of red giants that display true dust excesses is uncertain and controversial (Jura 1990, Plets et al. 1997, Jura 1999, Kim, Zuckerman \\& Silverstone 2001). In any case, since ground-based  imaging shows that the 10 ${\\mu}$m excess is   physically associated with HD 233517 (Skinner et al. 1995, Fisher et al. 2000), at least for this particular star, the dust producing the infrared excess  either is orbiting or is expanding away from the star.  For some other stars, the infrared excess has been resolved with ISO and in these objects, the dust is probably not orbiting (Kim et al. 2001).  There are difficulties with the model that HD 233517 is currently losing enough mass to produce the observed infrared excess. As with other first ascent red giants with dust, the infrared spectrum of HD 233517 peaks near 60 ${\\mu}$m.  This spectral energy distribution is characteristic of cool material and is very different from that associated with a continuous mass loss rate where there is a substantial amount of dust near a temperature of 1000 K (see, for example, Sopka et al. 1985).  One possibility is that the mass loss rate from HD 233517 is episodic and  currently the star is not expelling much matter.  However, as noted by Jura (1999), the characteristic time scale for the dust to expand  around a first ascent red giant to its inferred location may be less than 20 years.  There is no evidence for recent infrared variability of this star, although the star does exhibit a full amplitude optical variation of 0.02 mag with a 47.9 day period  which may be caused by star spots rotating in and out of the view (Balachandran et al. 2000).  Since the dust around HD 233517 may not be carried in a wind, we consider models where the dust is orbiting the star. One scenario is that the dust around HD 233517 is simply ``left over\" from the main sequence phase.  However, a major difficulty with this model is that the fractional infrared excess around HD 233517, $L_{IR}/L_{*}$, of 0.06  is much larger than that found for even the ``dustiest'' main sequence stars such as ${\\beta}$ Pic and HR 4796 where $L_{IR}/L_{*}$ is 2-4 ${\\times}$ 10$^{-3}$ (see, for example, Zuckerman 2001).  Here, we  speculate that HD 233517 had a low mass companion which was engulfed when it became a red giant.  Although uncertain, the engulfment of this companion could have led to the ejection of an equatorial ring of orbiting material (see Taam \\& Sandquist 2000, Spruit \\& Taam 2001).   As discussed by Pringle (1991), after this ring is created, the matter expands under the action of tidal torques which transfer angular momentum from the binary into the circumstellar material which thus evolves into an extended disk.    Although not required in Pringle's model, it is possible that in this dense equatorial ring, dust grains formed. We suggest that the structure of the circumstellar system is described by the models of  Chiang \\& Goldreich (1997) for passive, orbiting disks which flare in response to the illuminating radiation. Below, we present details of this  idealized model. ",
        "conclusions": "We have modeled the currently available infrared data for HD 233517 and find that the observations can be naturally explained by a flared, orbiting disk of mass  ${\\sim}$0.01 M$_{\\odot}$ and   outer radius of  ${\\sim}$45 AU.  We speculate that this disk was created by the engulfment of a low mass stellar companion.  This work has been partly supported by NASA."
    },
    "0209/astro-ph0209266_arXiv.txt": {
        "abstract": "We present CO(1-0) and CO(2-1)  maps of the Seyfert galaxy NGC 7217, obtained with the IRAM interferometer, at 3\" and 1.5\" resolution respectively. The nuclear ring (at r=12\"=0.8kpc) is predominant in the CO maps, with a remarkable surface density gradient between the depleted region inside the ring and the inner border of the ring. The CO nuclear ring is significantly broader (500-600pc) than the dust lane ring. The CO(2-1)/CO(1-0) ratio is around 1, typical of optically thick gas with high density. The overall morphology of the ring is quite circular, with no evidence of non-circular velocities. In the CO(2-1) map, a central concentration might be associated with the circumnuclear ring of ionised gas detected inside r=3\" and interpreted as a polar ring by Sil'chenko and Afanasiev (2000). Our interpretation is more in terms of a bar/spiral structure, in the same plane as the global galaxy but affected by non-circular motions, which results in a characteristic S-shape of the isovels. This nuclear bar/spiral structure, clearly seen in a V-I HST colour image, is essentially gaseous and might be explained with acoustic waves. ",
        "introduction": "NGC 7217 is one of the first galaxies observed from the NUGA (NUclei of GAlaxies) sample of about 20 spirals with AGN, which are being mapped at high resolution with the IRAM interferometer to explore their molecular content, and to determine their possible fueling mechanisms (primary or nuclear bars, spiral waves, warps, or m=1 perturbations).  NGC 7217 is an (R)SA(r)ab galaxy classified as a LINER/Seyfert. It is particularly axisymmetric, and possesses nuclear, inner and outer rings, at 8, 31 and 77\" (Buta et al. 1995). Dominated by a massive and extended bulge, the spiral structure is flocculent, and a possible oval distortion might be the vestige of an ancient bar, which was responsible for the formation of the three rings.  We have observed in 2001 this galaxy with the IRAM interferometer in the CO(1-0) and CO(2-1) lines, with resolution of 3.2''x2.8'' and 1.6''x1.4'' respectively.  \\vspace{-0.3cm} ",
        "conclusions": ""
    },
    "0209/astro-ph0209099_arXiv.txt": {
        "abstract": "In an earlier paper we presented nuclear X-ray flux densities, measured with {\\it ROSAT}, for the B2 bright sample of nearby low-luminosity radio galaxies. In this paper we construct a nuclear X-ray luminosity function for the B2 radio galaxies, and discuss the consequences of our results for models in which such radio galaxies are the parent population of BL Lac objects. Based on our observations of the B2 sample, we use Monte Carlo techniques to simulate samples of beamed radio galaxies, and use the selection criteria of existing samples of BL Lac objects to compare our simulated results to what is observed. We find that previous analytical results are not applicable since the BL Lac samples are selected on beamed flux density. A simple model in which BL Lacs are the moderately beamed ($\\gamma \\sim 3$) counterparts of radio galaxies, with some random dispersion ($\\sim 0.4$ decades) in the intrinsic radio-X-ray relationship, can reproduce many of the features of the radio-selected and X-ray-selected BL Lac samples, including their radio and X-ray luminosity functions and the distributions of their radio-to-X-ray spectral indices. In contrast, models in which the X-ray and radio emission have systematically different beaming parameters cannot reproduce important features of the radio-galaxy and BL Lac populations, and recently proposed models in which the radio-to-X-ray spectral index is a function of source luminosity cannot in themselves account for the differences in the slopes of the radio and X-ray-selected BL Lac luminosity functions. The redshift distribution and number counts of the X-ray-selected EMSS sample are well reproduced by our best models, supporting a picture in which these objects are beamed FRI radio galaxies with intrinsic luminosities similar to those of the B2 sample. However, we cannot match the redshift distribution of the radio-selected 1-Jy sample, and it is likely that a population of FRII radio galaxies is responsible for the high-redshift objects in this sample, in agreement with previously reported results on the sample's radio and optical-emission-line properties. ",
        "introduction": "\\label{intro} The extreme properties of BL Lac objects are explained in terms of relativistic beaming of the emission from a jet oriented close to the line of sight. This model implies the existence of a substantial `parent population' of sources whose jets are less favourably aligned, and it is widely accepted that this is the population of low-power radio galaxies (Browne 1983; Urry \\& Padovani 1995). Properties which are isotropic and unaffected by beaming should be similar in BL Lac objects and low-power radio galaxies. This is broadly supported by observations of extended radio emission (e.g.,\\ Antonucci \\& Ulvestad 1985, Kollgaard \\etal\\ 1992, Perlman \\& Stocke 1993, 1994), host galaxies (e.g., Ulrich 1989, Abraham, McHardy \\& Crawford 1991; Wurtz, Stocke \\& Yee 1996; Falomo 1996) and cluster environments (Pesce, Falomo \\& Treves 1995; Smith, O'Dea \\& Baum 1995; Wurtz, Stocke \\& Ellingson 1997).  In a series of papers (Padovani \\& Urry 1990; Padovani \\& Urry 1991; Urry, Padovani \\& Stickel 1991), Urry and co-workers made this model quantitative by predicting the luminosity function of BL Lac objects based on that of radio galaxies. They used the analysis of Urry \\& Schafer (1984) and Urry \\& Padovani (1991) to calculate the expected luminosity function of a population of beamed objects given a parent (unbeamed) luminosity function. With the data on radio-galaxy and BL Lac populations then available they were able to show reasonable agreement between the predictions of the model and the observed luminosity functions. They found that the luminosity function of radio-selected BL Lac objects from the 1-Jy sample (Stickel \\etal\\ 1991) was consistent with that of radio galaxies from the 2-Jy sample (Wall \\& Peacock 1985) if the radio-emitting plasma in the cores had a bulk Lorentz factor $\\gamma_{\\rm radio} > 5$. The X-ray luminosity function and number counts of X-ray selected BL Lac objects were consistent with the luminosity function of FRI radio galaxies observed with {\\it Einstein} by Fabbiano \\etal\\ (1984) if the X-ray emitting plasma has a somewhat lower bulk Lorentz factor, $\\gamma_{\\rm Xray} \\approx 3$.  The issue of the relationship between radio galaxies and X-ray and radio-selected BL Lac objects is only partially resolved by this work. The best-studied sample of X-ray selected BL Lacs is the EMSS sample (Wolter \\etal\\ 1991, Rector \\etal\\ 2000) and the best radio-selected sample is the 1-Jy sample (Stickel \\etal\\ 1991, Rector \\& Stocke 2001). When these samples are compared, a number of differences emerge. Some -- perhaps as many as half -- of the 1-Jy objects have radio structures, luminosities and emission-line properties similar to those of FRII radio galaxies (Antonucci \\& Ulvestad 1984, Kollgaard \\etal\\ 1992, Rector \\& Stocke 2001) while the EMSS BL Lacs are all FRI-like (Perlman \\& Stocke 1993). The luminosity functions of the two samples are also different; the radio-selected objects show evidence for positive evolution (i.e. sources were more numerous or more powerful at higher redshift) while X-ray-selected sources appear to be {\\it negatively} evolving. Urry \\etal\\ were forced by the data then available to use an X-ray-selected sample of BL Lacs for their comparison with the X-ray luminosity function of radio galaxies, and a radio-selected sample when considering the radio luminosity function. They were thus unable to say anything about the relation between radio galaxies and the two BL Lac populations.  Radio-selected BL Lacs are often considered to be more extreme than X-ray selected objects; they have more prominent radio cores (Perlman \\& Stocke 1993, Laurent-Muehleisen \\etal\\ 1993, Rector \\& Stocke 2001), higher optical polarization with less stable position angle (Jannuzi, Smith \\& Elston 1994) and a lower optical starlight fraction (Stocke \\etal\\ 1985). On the other hand, their environments and host galaxies are similar to X-ray selected sources (Wurtz, Stocke \\& Yee 1996, Wurtz \\etal\\ 1997). This has led to suggestions that radio-selected objects are more strongly affected by Doppler boosting than X-ray selected objects, and so are being observed at smaller angles to the line of sight. Since radio-selected and X-ray selected objects have similar X-ray luminosities, this requires that the X-ray-emitting regions be less strongly beamed, or even isotropic (Maraschi \\etal\\ 1986; Celotti \\etal\\ 1993); the idea of weaker beaming is consistent with the difference in the Doppler beaming factors found by Urry \\etal\\ (1991). This in turn implies that X-ray-selected BL Lacs should be the more numerous population, since they can be seen at larger angles to the line of sight, which Celotti \\etal\\ (1993) argue is consistent with the X-ray luminosity functions of the two populations. Models of this kind can either rely on differences in velocity between X-ray and radio-emitting regions (e.g., Ghisellini \\& Maraschi 1989) or differences in opening angle of the jet (Celotti \\etal\\ 1993). However, these models do not account for the differences in evolutionary properties between the two samples. In section \\ref{beaming} of this paper we shall discuss whether they are consistent with observations of radio galaxies and with the X-ray and radio luminosity functions of the two classes.  A description of BL Lac objects in terms of the selection waveband does not necessarily reflect the underlying physics. An alternative approach is to refer to high-energy peaked BL Lacs (HBL) and low-energy peaked BL Lacs (LBL) (e.g., Giommi \\& Padovani 1994, Padovani \\& Giommi 1995), distinguishing between the two classes by their radio/X-ray flux ratios; a typical dividing line is that HBL have $\\log_{10}(F_{\\rm 1\\ kev}/F_{\\rm 5\\ GHz}) >-5.5$, or, equivalently, $\\alpha_{\\rm RX}<0.72$, where $\\alpha$ is defined here and throughout the paper in the sense $F \\propto \\nu^{-\\alpha}$ and $\\alpha_{\\rm RX}$ denotes the radio-to-X-ray two-point spectral index. By selecting bright sources at one or the other waveband we may simply be picking up objects with extreme radio/X-ray flux ratios, which suggests various possible schemes for unifying the two populations (e.g., Giommi \\& Padovani 1994, Fossati \\etal\\ 1997) and is consistent with the discovery of `intermediate' BL Lacs in deeper surveys (Laurent-Muehleisen \\etal\\ 1999). From multi-wavelength observations it has been found that the spectral energy distributions (SEDs) of BL Lacs can be represented by a low-frequency peak (assumed to be synchrotron radiation) and a high-frequency peak (perhaps inverse-Compton radiation). There is evidence that the positions of these peaks shift to higher frequencies at lower bolometric luminosities (Sambruna, Maraschi \\& Urry 1996, Fossati \\etal\\ 1998). However, Giommi, Menna \\& Padovani (1999) do not find the increase in the numbers of HBL at fainter radio fluxes that is expected in such a model.  To try to resolve some of the outstanding issues in BL Lac unification it is productive to learn more about the presumed parent population of many or all of the BL Lac objects, namely the low-power radio galaxies. Work to date has been hampered by the lack of a well-defined, low-frequency-selected sample of radio galaxies with well-known radio and X-ray properties. In a previous paper (Canosa \\etal\\ 1999) we presented results for the 40 members ($\\approx 80$ per cent) of the B2 bright sample of radio galaxies which had been observed in pointed {\\it ROSAT} observations. The high spatial resolution of {\\it ROSAT} allowed us to separate the nuclear emission from extended emission due to the hot-gas environment of the radio source. In this paper we use these data to derive a new X-ray luminosity function for the nuclei of B2 radio sources, largely free from contamination from thermal emission from the sources' hot-gas environments. We use numerical techniques to extend earlier work, first asking how well a simple beaming model can be used to link the X-ray and radio luminosity functions of radio galaxies, and then investigating the expected relationship between radio galaxies and BL Lac objects when selection bias is taken into account.  Throughout the paper we use a cosmology in which $H_0 = 50{\\rm\\ km\\,s^{-1}\\,Mpc^{-1}}$ and $q_0 = 0$. ",
        "conclusions": "\\label{conclusion}  In this paper we have been trying to determine to what extent and under what assumptions B2 radio galaxies (or, more accurately, the radio galaxy population represented by the B2 radio galaxies) can be the parent population of two well-studied samples of BL Lac objects (the EMSS and 1-Jy samples). Following earlier work, we have characterized the properties of the populations using their luminosity functions: luminosity functions for the B2 and BL Lac samples were constructed in sections \\ref{b2lum} and \\ref{bllum}. In section \\ref{usmodl} we discussed the analytical results of Urry \\& Schafer (1984); we argued that their models cannot properly be applied to luminosity functions of samples with different selection criteria.  Instead, in sections \\ref{mc} and \\ref{sim}, we used Monte Carlo simulations to set up a parent population of beamed radio galaxies with B2-like properties and drew objects from them with selection criteria which as closely as possible matched those actually used in generating the EMSS and 1-Jy samples. By adjusting the unknown parameters of the simulation (section \\ref{matching}) we were able to show that our simulated observations of a beamed population of radio galaxies matched to the B2 sample were able to reproduce reasonably well the observed numbers of BL Lac objects and the slopes of the radio- and X-ray-selected BL Lac luminosity functions, if low values ($\\sim 3$) of the beaming Lorentz factor $\\gamma$ are adopted. In section \\ref{dispersion} we found that the introduction of a moderate dispersion in the rest-frame ratio of core X-ray to core radio luminosity (for which there is some direct evidence in the observations of the B2 radio galaxies) can help to explain the observed differences in the luminosity functions and radio-to-X-ray spectral indices of radio-selected and X-ray-selected BL Lacs. This approach gives better results than other proposed ways of generating the differences between BL Lac populations (sections \\ref{asymm} -- \\ref{ldep}). These results therefore support a model in which the two apparently different BL Lac populations are simply extreme objects drawn from a single parent population (cf.\\ Laurent-Muehleisen \\etal\\ 1999), although we have not attempted to reproduce all of the observed differences between radio-selected and X-ray-selected BL Lacs, such as the apparent differences in cosmological evolution (section \\ref{intro}).  The beaming Lorentz factors used here are much smaller than those inferred from superluminal motion or $\\gamma$-ray transparency in some BL Lac objects. These typically require $\\gamma \\ga 10$, which would grossly overpredict the number of BL Lac objects that should be observed above the flux limit (cf.\\ Table \\ref{numbers}). Chiaberge \\etal\\ (2000) have previously pointed out that the nuclear luminosities of FRI radio galaxies are a factor 10--10$^4$ too bright to be consistent with the `de-beaming' of BL Lacs using these high Lorentz factors. The solution they adopt, velocity structure in the nuclear jet, seems plausible in view of what we know about the existence of velocity-structures in the {\\it kiloparsec}-scale jets (e.g., Laing 1996). Velocity structure in the jets should not significantly affect our general conclusions here. If jets have velocity structure, our values of $\\gamma$ parametrize the relationship between the observed luminosity and the angle to the line of sight, rather than describing real physical bulk velocities. However, the details of this parameterization make a difference to the predicted numbers of BL Lac objects and the range in their luminosities. Although attempting to determine a typical jet emissivity/velocity profile is beyond the scope of this paper, it is certainly the case that luminosity functions can be used to help to constrain it.  The simulated observations fail to reproduce the real data in two important ways. Firstly, although the simulated EMSS sample is a good match in redshift distribution to the observed objects (Fig.\\ \\ref{redshifts}, the simulated 1-Jy sample contains far too few high-redshift objects. As discussed in section \\ref{intro}, the EMSS objects seem all to be FRI-like in their radio structure and luminosity, while it is known that some 1-Jy objects are FRII-like; since FRIIs have higher luminosities than the parent population we use, it is not surprising that some of the 1-Jy sample appear at higher redshifts. Our models predict that essentially all the 1-Jy objects whose parent population are FRIs should have redshifts less than $\\sim 0.55$ (Fig. \\ref{redshifts}), which agrees well with the observations of Rector \\& Stocke (2001). The fact that some of the 1-Jy objects may have a different parent population may also help to explain the anomalously steep 1-Jy X-ray luminosity function.  Secondly, the simulations predict a considerably larger number of BL Lac objects than is observed (by factors $\\ga 2$) for $\\gamma \\ga 4$ or if a significant dispersion in intrinsic $\\alpha_{\\rm RX}$ is introduced; even the models which best reproduce other features of the population, such as the differences between the slopes of the X-ray and radio luminosity functions, overpredict the numbers of 1-Jy BL Lacs by a factor $\\sim 1.5$. The mismatch in numbers becomes even greater if some of the observed high-redshift 1-Jy objects are not drawn from an FRI parent population, as discussed above. We can attribute at least some of this effect to the optical selection criteria applied when defining BL Lac samples. As pointed out by March\\~a \\etal\\ (1996), a definition in terms of the strength of the 4000-\\AA\\ break measures the strength of the optical non-thermal emission in terms of the starlight in the host galaxy, which may have very little to do with the AGN, while a condition on the equivalent width of the strongest emission lines has little physical justification (and may exclude objects which are in all other ways BL Lac objects; cf.\\ Vermeulen \\etal\\ 1995). So some of the predicted objects may be present in the surveys, identified as something other than BL Lacs. However, a search in the EMSS survey for objects intermediate between FRIs and traditional EMSS BL Lacs (Rector \\etal\\ 1999) found relatively few candidates, and only a handful of EMSS sources are directly identified with radio galaxies. It remains possible that some of our simulated sources are identified as groups or clusters in the EMSS survey. If, as new {\\it Chandra} results (section \\ref{chandra-caveat}) suggest, we are overestimating the radio galaxies' nuclear X-ray fluxes (section \\ref{chandra-caveat}) then we will also have overestimated the intrinsic X-ray core prominence; correcting for this would result in a (probably small) reduction in the predicted total number of EMSS sources. In the 1-Jy sample, there are additional optical magnitude and radio spectral selection criteria which are not modelled in our simulations, and so there is more scope for `hiding' the excess sources. On the other hand, we are extrapolating our B2 luminosity function to high redshifts to generate BL Lac objects without including the effects of cosmological evolution of the FRI population, now reasonably well-established (e.g., Waddington \\etal\\ 2001). Because there is little evidence for evolution below $z<0.5$, where most of our BL Lac candidates are generated (Fig. \\ref{redshifts}), this does not have a strong effect on our models, but there may be up to an order of magnitude increase in the numbers of the most luminous sources at $z \\sim 1$. Taking this into account would lead our models to produce $\\sim 10$ additional high-redshift 1-Jy objects. Without a detailed model for source evolution, we cannot make this more quantitative.  Combining these factors with the large statistical uncertainties on the predicted and actual numbers of objects, we regard the degree of agreement between the simulations and observations as encouraging. It supports a model in which FRI radio galaxies in the luminosity range of the B2 bright sample are the parent population of all the EMSS X-ray-selected BL Lac objects, and of $\\sim 50$ per cent of the radio-selected 1-Jy objects. A higher-luminosity population, probably beamed FRIIs, must be responsible for the remaining, higher-redshift, 1-Jy objects."
    },
    "0209/astro-ph0209050_arXiv.txt": {
        "abstract": "{ Using a sample of  unprecedented size (about 400 objects) of \\hii\\ galaxies in which the oxygen abundances have been obtained using the temperature derived from the \\rOiii\\ line ratio, we confirm that the \\hii\\ galaxies form a very narrow sequence in many  diagrams relating line ratios and \\Hb\\ equivalent width. We  divide our sample in three metallicity bins, each of which is compared with sequences of photoionization models for evolving starbursts with corresponding metallicity. Our aim is to find under what conditions a theoretical sequence can reproduce all the observed trends. Taking into account the presence of an older, non-ionizing stellar population, for which independent indications exist, we find that the simple model of an adiabatic expanding bubble reproduces the observational diagrams very well if  account is taken of an aperture correction and the covering factor is assumed to decrease with time exponentially with  an e-folding time of 3~Myr. We find that the \\Heii\\ nebular line emission occurs too frequently and in too wide a range of $EW$(\\Hb) to be attributable to either the hard radiation field from Wolf-Rayet stars or the X-rays produced by the latest stellar generation. Assuming that the \\Heii\\  line is due to photoionization by a hot plasma at a temperature of $10^{6}$~K, a total X-ray luminosity of $10^{40} - 4\\times10^{40}$\\ergs\\ is required for at least half of the sources. We find evidence for self-enrichment in nitrogen on a time scale of several Myr, and argue for a possible self-enrichment in oxygen  as well. ",
        "introduction": "It has been known for a long time that giant \\hii\\ regions form a well defined sequence in emission line diagnostic diagrams (McCall et al. 1985; Veilleux \\& Osterbrock 1987). The first interpretations of this sequence were based on single star photoionization models, and it was concluded that the driving parameter of the sequence is the metallicity, and that variations of an additional parameter -- either the effective temperature of the ionizing stars or the ionization parameter -- are required in order to reproduce the observed sequence (McCall et al. 1985; Dopita \\& Evans 1986).  However,  giant \\hii\\ regions are powered by intensive bursts of star formation (e.g. Sargent \\& Searle 1970; Mas-Hesse \\& Kunth 1991). Stellar evolution produces a gradual change with time of the integrated stellar energy distribution, which has to be taken into account in the modeling of giant \\hii\\ regions. Numerous studies have produced grids of photoionization models for \\hii\\ regions considering that aspect (e.g. Terlevich \\& Melnick 1985; Olofsson 1989; McGaugh 1991; Cid Fernandes et al. 1992; Bernl\\\"ohr 1993; Cervi\\~{n}o \\& Mas-Hesse 1994; Garc\\'{\\i}a-Vargas \\& Diaz 1994; Garc\\'{\\i}a-Vargas et al. 1995; Stasi\\'{n}ska \\& Leitherer 1996 (hereafter SL96); Bresolin et al. 1999; Dopita et al. 2000; Moy et al. 2001; Charlot \\& Longhetti 2001; Stasi\\'{nska} et al. 2001 (hereafter SSL01); Zackrisson et al. 2001). These studies used different prescriptions for the modeling of the stellar population as well as of the nebular emission. Comparisons with observations are difficult because there are at least two independent parameters that drive the emission line properties of extragalactic \\hii\\ regions: age and metallicity. The general conclusion from these studies is that the Salpeter initial mass function with an upper stellar mass limit around 100\\Ms\\ is consistent with the observed emission line ratios (although Bresolin et al. (1999) argue for a lower cut-off mass at high metallicity). However, several problems are noted. All the models for instantaneous starbursts predict a strong drop in \\Oiii/\\Hb\\ at about 5~Myr. This corresponds to the maximum lifetime of O stars so the prediction is very robust. The age of a starburst in an isolated giant \\hii\\ region can be estimated, to a first approximation, from the equivalent width of \\Hb, $EW$(\\Hb) (this is not feasible in giant \\hii\\ regions belonging to spiral galaxies, where old stellar populations strongly contribute to the continuum at \\Hb). Using samples of isolated extragalactic \\hii\\ regions, hereinafter referred to as \\hii\\ galaxies, SSL01 and Zackrisson et al. (2001) showed that the observed drop was rather mild and displaced towards lower values of $EW$(\\Hb). SSL01 attributed this to the effect of underlying old populations of stars, shown by Raimann et al. (2000) to be present in \\hii\\ galaxies. Other causes such as leakage of ionizing photons or selective dust extinction were also mentioned. The classical diagnostic diagram \\Oiii/\\Hb\\ vs. \\Oii/\\Hb\\ does not seem to be completely understood in terms of pure photoionization models. One way out is to advocate some additional heating (SSL01) or the contribution of zones of low ionization parameter (Moy et al. 2001). A very significant trend of \\Oi/\\Hb\\ increasing with decreasing $EW$(\\Hb), discovered by  SL96 and SSL01 remains to be quantitatively explained.  The purpose of the present paper is to quantify the conditions needed to reproduce the observed emission line sequence of giant \\hii\\ regions, including the very small dispersion seen in many emission line diagrams. For a more meaningful comparison of models with observed data, our observational sample is composed of objects for which the age of the ionizing star cluster and the metallicity can be estimated in an independent way. This limits the sample to \\hii\\ galaxies with measured  \\Oiiit\\ line intensities. It is only when such a sample is fully understood that one has a chance to better understand the entire extragalactic \\hii\\ region sequence, including  more metal-rich \\hii\\ regions such as those found in the inner parts of disk galaxies.  In Sect. 2 we describe the observational sample to which we compare the models and the preliminary treatments of the data. In Sect. 3 we  outline the modeling strategy, and in Sect. 4 we present our results. The main findings of our paper are summarized and discussed in Sect. 5. ",
        "conclusions": "Using a sample of \\hii\\ galaxies of unprecedented size (about 400 objects) in which the oxygen abundances have been obtained using the temperature derived from the \\rOiii\\ line ratio, we have confirmed the impressive trends and extremely strong correlations found in diagrams relating line ratios and \\Hb\\ equivalent width that we already found in previous studies (SL96 and SSL01). The \\Heii\\ line is present at a level of 1 -- 3 \\% of \\Hb\\ in a large number of \\hii\\ galaxies, roughly half of the galaxies where the signal-to-noise was sufficient to detect and measure this line (which often appears on top of a Wolf-Rayet bump in the spectra). This line is present in the entire range of \\Hb\\ equivalent widths, and its intensities show no correlations neither with $EW$(\\Hb) nor with metallicity. The increase of \\Nii/\\Oii\\ with decreasing   $EW$(\\Hb), already noted by SSL01, is clearly seen (although more dispersed that some other trends).  So far, analysis of emission line trends in \\hii\\ galaxies made use of grids of photoionization models, but without attempting to propose a sequence of evolutionary models reproducing the observed trends.  Taking advantage of the large size of our sample and of the fact that the metallicities were determined in a direct and  independent way, we divided our sample in three metallicity bins. We considered each bin separetely, trying to find under what conditions a sequence of photoionization models, taking into account the time evolution of the ionizing cluster,  reproduces all the observational diagrams adequately. In such an exercise, we used only very simple prescriptions, justified if possible by arguments related with what is known of the physics of these objects.  We have found that the simple model of an adiabatic expanding bubble gives rise to an increase of the \\Oi/\\Hb\\ ratio with time, reaching the highest values observed in about 5~Myr. The presence of an older, non ionizing stellar population that contributes to the continuum at \\Hb\\ has been attested by direct studies of the stellar features in the continuum of  \\hii\\ galaxies. The characteristics of such an old population should of course be independent of the age of the most recent burst of star formation. Since the contribution of this old population to the stellar continuum at \\Hb\\ is proportionally larger for younger starbursts, such an underlying population is not sufficient to reproduce quantitatively the observed trends in the \\Oiii/\\Hb\\ vs. $EW$(\\Hb), \\Oiii/\\Oii\\ vs. $EW$(\\Hb), and \\Hei/\\Hb\\ vs. $EW$(\\Hb) diagrams. The observational data require that the covering factor $f$ of the ionizing source by the emitting gas decreases with time. A reasonable fit to the observations is provided assuming an aperture correction of 0.5 to the models and an exponentially decreasing covering factor, with an e-folding time of 3~Myr, associated with an old population contributing at a level equal to the \\Hb\\ continuum produced by the young stars at zero age. That the covering factor of giant \\hii\\ regions is smaller than unity and depends on the evolutionary stage of the region has already been suggested by Castellanos et al. (2002) from the study of a limited number of objects in spiral galaxies. What we find here is that there seems to be an universal law for the temporal variation of the covering factor. The physical reason underlying such a law could be linked to departure from spherical symmetry and blowout or to shell fragmentation (see Tenorio-Tagle et al. 1999).  We find that the \\Heii\\ nebular line emission in \\hii\\ galaxies occurs too frequently and in too wide a range of $EW$(\\Hb) to be attributable to either the hard radiation field from Wolf-Rayet stars, as was suggested by Schaerer (1996, 1998), or the X-rays produced by current evolutionary synthesis models of  young starbursts (Cervi\\~{no} et al. 2002). Assuming that the \\Heii\\  line is due to photoionization by a hot plasma at a temperature of $10^{6}$~K, a total X-ray luminosity of $10^{40} - 4\\times10^{40}$\\ergs\\ is required for at least half of the sources. In some of the objects, the observed \\Heii\\ nebular emission could actually be due to harder X-rays produced  either by massive binaries or by supernova remnants from previous generations of stars in the age range of 10 -- 50~Myr (Van Bever \\& Vanbeveren 2000). Upcoming X-ray spectroscopy and imaging will help clarify the matter in the near future.  Our evolutionary sequences of models fail to reproduce perfectly the \\Oiii/\\Hb\\ vs. \\Oii/\\Hb\\   and \\Oiii/\\Hb\\ vs.  \\Oi/\\Hb\\ diagrams for the intermediate and low metallicity bins, if the chemical composition is homogeneous and identical to the one of the gas which gave birth to the young stars. The collisionally excited lines are slightly -- but significantly -- too weak with respect to the observations. This problem had been noted before (SSL01), but here it is more visible due to the fact that we compare models with objects of similar metallicity. One way to understand the discrepancy is to postulate the existence of an additional heating mechanism. It is unlikely that this heating agent could be dust, or shocks produced by the energy released by winds and supernovae from the latest burst of star formation. Planetary nebulae and hot white dwarfs from earlier generations of stars would also be far from sufficient. The hypothesis of shock heating due either to stellar winds from previous generations of stars, or to cloud-cloud collisions needs to be investigated. However, another option to explain these emission line diagnostic diagrams is to invoke chemical inhomogeneities. Chemical inhomogeneities are expected in regions experiencing mass loss and supernova explosions from massive stars. The question is in what form is the newly synthesized matter, and how much of it has escaped the nebula. The problem is far from being settled at present (Roy \\& Kunth 1995; Tenorio-Tagle 1996; Silich \\& Tenorio-Tagle 1998; Mac Low \\& Ferrara 1999; Ferrara \\& Tolstoy 2000). So far direct evidence for self-enrichment has been scarce (see Kobulnicky (1999) for a review), but the fact that \\Nii/\\Oii\\ increases as $EW$(\\Hb) decreases, especially in the intermediate and low metallicity bins of our sample, argue for the existence of self-enrichment in nitrogen on timescales of a few Myr. Of course, this in itself does not argue for any self-enrichment in oxygen, since the nitrogen and oxygen donors are different (see Matteucci \\& D'Ercole (1999) for a review). Given that the fate of the elements is still not well understood, we have felt it instructive to investigate the effect of inhomogeneous metallicity on the emission line diagnostic diagrams. We found that for the intermediate metallicity bin, a combination of two models with abundances $Z_{neb}$ = 0.2\\Zs\\ and 0.05\\Zs\\ leads to good agreement with the observations, but other solutions are likely possible as well. For the low metallicity bin our models indicate that a qualitatively similar composite model would be acceptable. The scatter of observational points in that bin is larger than in the other  bins. This can be attributed to the small number of points in that bin and also to the fact that the effects of self-enrichment become more drastic at low metallicity. It is only for our high metallicity bin that there is a priori no need to invoke self-enrichment in oxygen, since the diagrams are satisfactorily reproduced  for models with homogeneous chemical composition. However, chemical inhomogeneities, if present, are expected to have a smaller impact, due to the fact that  \\Oiii/\\Hb\\ and \\Oii/\\Hb\\ are not so dependent on metallicity for $Z_{neb}$ between 0.2 -- 1 \\Zs\\  (see e.g. Stasi\\'{n}ska 2002).  Finally, it is worth of noting that our prescriptions for the time dependence of the covering factor and of the degree of inhonomogeneity were tailored to reproduce the emission line sequence of \\hii\\ galaxies. On the other hand, the time dependence of the radius of the expanding bubble is exactly the one predicted by the theory of supernova- and wind-blown bubbles. The size of the bubble required to reproduce the observational diagrams corresponds rather well with the prediction using the energy available from stellar winds and supernovae computed in population synthesis models. It is most encouraging that emission line diagnostics of \\hii\\ galaxies confirm, on a statistical basis, the predictions of theories on stellar evolution, stellar populations and dynamical interaction with the interstellar medium. However, our interpretation should be compared to detailed modeling of selected objects in the entire sequence of $EW$(\\Hb). In particular, it would be important to compare the ages derived directly from the analysis of the stellar UV light with those implied by our interpretation of the emission line diagrams, and to use Wolf-Rayet stars as independent clocks. So far, only a handful of galaxies from our sample have been studied in such a way (Mas-Hesse \\& Kunth 1999), and the comparison is not conclusive."
    },
    "0209/astro-ph0209320_arXiv.txt": {
        "abstract": " ",
        "introduction": "Recently,  Chandra X-ray Observations reveals a new aspect of a compact star. The isolated neutron star candidate RXJ1856.5-3754, which may be the closest to us, has the radiation radius 3.8-8.2 km  \\cite{det}. The distance is estimated as  60-130pc from the parallax  \\cite{wal}\\cite{kka} and the interstellar column density \\cite{det}. The ambiguity leads to the range of the estimated size, but the value is evidently much smaller than the canonical radius of a neutron star $\\sim 10$ km \\cite{lp}.   The radiation radius $ R_\\infty$ is determined by the observed energy flux and blackbody temperature. Assuming spherical symmetric equilibrium, $ R_\\infty$ is expressed by the radius $R$ in the Schwarzshild coordinate as $R_\\infty = R(1-2GM/Rc^2)^{-1/2}$, where $M$ is the gravitational mass. It is clear that  $ R \\le R_\\infty $. After some algebra, we also have $ GM/c^2 \\le R_\\infty /(3\\sqrt{3}) . $ The gravitational mass must be less than 0.49-1.05 $M_\\odot$ corresponding to the observed value of $ R_\\infty$. This limit of mass is significantly less than the canonical mass of a neutron star $ \\sim 1.4 M_\\odot $ \\cite{lp}. In this way, the characteristic size and mass are unusual for any neutron star models.  There are however several attempts to reconcile with the standard neutron star models. For example, the atmosphere is metal dominated one, and/or the surface temperature distribution is inhomogeneous. At present, there is no positive evidence to support them even in the long term and high-resolution observation \\cite{det}.  Taken at face value, RXJ1856.5-3754 is not a neutron star, but a different species of a compact star, which is likely to be  a quark star. We would definitely judge the state of the compact star, if there is different information such as mass, which is not easily determined by the single star. Nakamura \\cite{nak} proposed a possible formation scenario of the quark star. In the standard scenario to neutron star after supernova, it is difficult to produce such light remnant $\\sim 0.7 M_\\odot$ leading to the quark star. If the progenitor has significant rotation, small core is left through the centrifugal breakup. In that case, huge amount of energy can  be radiated by gravitational waves. The quadrupole oscillation is likely to be driven by such a collapse, and the asymmetry of matter distribution is efficiently smoothed out by the gravitational radiation. The detection of such gravitational waves significantly depends on  the number of the events. There is a big uncertainty about it, because we know just one candidate, RXJ1856.5-3754.   Motivating the observational suggestion, we will study the property of the quark star models. Our concern is not the equilibrium structure, which was already calculated in the literature \\cite{lp}\\cite{afo}\\cite{hzs}\\cite{gle}~, but the dynamical aspect, that is, non-radial oscillations resulting from small disturbances. In particular, we compute the characteristic frequency of the quadrupole f-mode oscillations, which is the most important for the gravitational radiation. Such calculations were extensively performed for several non-rotating neutron stars \\cite{lindet}\\cite{cls}  and slowly rotating polytropic models \\cite{koj92}\\cite{koj93apj} . By extending to the quark star models, the 'catalog' is enriched for the future gravitational wave astronomy. In addition to direct gravitational wave signal, oscillatory phenomena after some kinds of sudden bursts or flares may appear in X-ray and/or gamma-ray observation. The global properties such as mass and radius should be relevant to dynamical overall oscillations on the star. Non-spherical distortion efficiently decays due to gravitational wave emission within a few second as shown later. The typical oscillation frequency is kHz. These properties, irrespective of observational bands or tools, provide useful diagnosis for the interior of compact stars. Other variability except f-mode may also appear in different frequency range, but they are much reflected by details of the interior structure, which are not so clear at moment. Therefore, we do not consider them here. Microscopic equation of state for high-density matter is not yet established in neutron stars and quark stars. We use a simplest one, which may be helpful within our present knowledge of high-density matter to understand the contrasts between neutron stars and quark stars. The model and the numerical method are summarized in \\S 2.  The results are shown in \\S 3. In \\S 4, we compare the self-bound quark stars with neutron stars in  their oscillation frequencies and the decay times. Finally in \\S 4, concluding remarks are given. ",
        "conclusions": "In this paper, we have presented suggestive results for the pulsations associated with gravitational radiation. The gravitational waves carry key information of the sources. We can in principle use them as probes of compact stars. The oscillation frequency scales with the average density of a star $\\nu \\propto \\sqrt{ M/R^3}$, since the f-mode has no radial node. The decay time due to gravitational radiation significantly depends on the relativistic factor $ M/R.$ We may utilize these properties to discriminate self-bound quark stars from gravitationally bound neutron stars. If self-bound quark stars are realized in less relativistic regime, the oscillation frequency is high enough due to high density, but decay time is longer than that of neutron star. In that case, the mass and size are quite different in the equilibrium sequences of two compact stars. It is clear that present self-bound quark model is  too crude like the bag model. The model is used to calculate explicitly, but the resultant properties of the oscillations roughly depend on the macroscopic  values such as the mass and radius. We therefore expect that such discriminated properties should be involved in any detailed models.    Finally, we will comment on the astrophysical relevance of the quadrupole f-mode oscillation. The driving mechanism like Cepheid variables is not yet known, so that the persistent oscillations would not be observed. Rather, abrupt changes of the structure like the formation\\cite{nak} and/or some kinds of bursts may be relevant. The origin and the nature of them are also unclear at moment. However, it is evident that the gravitational wave emission is crucial in such dynamical process. The most efficient mode is the quadrupole ($l=2$) f-mode, which was calculated here. We should await for the fully relativistic numerical calculation to simulate the violent events, but the oscillation frequency may be used even for checking the constructing numerical code \\cite{shb}."
    },
    "0209/astro-ph0209393_arXiv.txt": {
        "abstract": "NGC6752 hosts in its halo  PSR J1911-5958A, a newly discovered binary millisecond pulsar which is the most distant pulsar ever known from the core of a globular cluster. Interestingly, its recycling history seems in conflict with  a scenario of ejection resulting from ordinary stellar dynamical encounters. A scattering event off a binary system of two black holes with masses in the range of $3-50\\,M_{\\odot}$ that propelled PSR~J1911-5958A into its current peripheral orbit seems more likely. It is still an observational challenge to unveil the imprint(s) left  from such a dark massive binary on cluster's stars: PSR J1911-5958A may be the first case. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209500_arXiv.txt": {
        "abstract": "{We report the identification of the optical and radio counterparts of the ultraluminous X-ray (ULX) source XMMU~J124825.9+083020 (NGC4698-ULX1). The optical spectrum taken with the VLT yields a redshift of $z=0.43$, which implies that the ULX is not associated with the nearby galaxy NGC~4698.  The spectral energy distribution calculated from the available data indicates that the source is likely to be a BL Lac object.  The possible synchrotron peak at X-ray energies suggests that this source may be a $\\gamma$-ray emitter. ",
        "introduction": "In recent years, the improved imaging capabilities and increased sensitivity of \\emph{ROSAT}, \\emph{Chandra} and \\emph{XMM-Newton} have allowed us to effectively study discrete sources in nearby galaxies beyond the Local Group. Particularly intriguing is the discovery of off-nuclear X-ray sources with luminosities well above the Eddington limit for a typical neutron star, $\\sim 10^{38}$ \\lum, and up to $2\\times 10^{40}$ \\lum\\ (e.g., Read et al. 1997; Colbert \\& Mushotzky 1999; Roberts \\& Warwick 2000; Makishima et al. 2000; Fabbiano et al. 2001).  These sources are typically called ultraluminous X-ray sources (ULXs).  Despite much effort, little is presently known about these sources. The identification of their optical counterparts is often problematical when the sources are superposed against regions of high surface brightness in the host galaxy.  To date, only two ULXs appear to have a clear optical identification (Roberts et al. 2001; Wu et al. 2002), while for others it has been possible to study only the nearby environment (Pakull \\& Mirioni 2002; Wang 2002).  We have started a search for ULXs in a sample of nearby galaxies (Foschini et al.  2002).  This paper concerns follow-up observations of the ULX in NGC~4698.  We discuss the radio and optical counterparts and give a redshift estimate.  We identify the source with a background source, most likely a BL Lac object at $z = 0.43$.  \\begin{figure*}[ht] \\begin{center} \\includegraphics[width=13cm]{H3901F1.eps} \\end{center} \\caption{$R$-band image from VLT-U3/FORS1, superimposed in contours the smoothed image from the \\emph{XMM-Newton} EPIC-MOS2 data in the $0.5-10$~keV energy band (white over NGC~4698, black in the remaining field). North is up and East to the left. The $D_{25}$ ellipse (3\\farcm8) is shown for comparison.} \\label{fig1} \\end{figure*} ",
        "conclusions": "\\subsection{Classification} The absorption features detected in the VLT spectrum clearly places XMMU~J124825.9+083020 at $z=0.43$.  We identify  the source as a BL Lac object, for the following reasons.  (1) We detect no emission lines; the upper limit to any emission feature is $\\sim$5 \\AA. (2) The stellar features superposed on the featureless continuum have strengths generally consistent with those of other BL Lac objects.  We demonstrate this concretely by comparing a redshifted spectrum of the BL Lac object $0548-322$ (Fig.~\\ref{fig3}).  The break contrast at 4000 \\AA\\ is $\\sim$29\\%, again similar to other BL Lac objects (Laurent-Muehleisen et al. 1998).  (3) The optical continuum slope measured in the Magellan spectrum is consistent with those of high-frequency peaked BL Lac objects (e.g., Stickel et al. 1993).  And (4) the multiwavelength spectral energy distribution (Fig.~\\ref{fig4}), although not assembled from simultaneous observations, is highly reminiscent of those of BL Lac objects.  By using two-point spectral indices, namely the radio-to-optical $\\alpha_{\\mathrm{ro}}$ and optical-to-X-ray $\\alpha_{\\mathrm{ox}}$, it is possible to show that different objects populate different regions of the $\\alpha_{\\mathrm{ro}} - \\alpha_{\\mathrm{ox}}$ plane (e.g., Brinkmann et al. 1997; Laurent-Muehleisen et al. 1999). Another method has been suggested by Maccacaro et al. (1988), who proposed a nomograph to link the X-ray flux in the energy band $0.3-3.5$~keV and the visual magnitude.  The values of $\\alpha_{\\mathrm{ro}}$ and $\\alpha_{\\mathrm{ox}}$ for the present source are $0.42$ and $0.95$, respectively, thus placing it in the region of X-ray selected BL Lacs (Brinkmann et al. 1997), or high-energy peaked BL Lacs in the $\\alpha_{\\mathrm{ro}}-\\alpha_{\\mathrm{ox}}$ diagram of Laurent-Muehleisen et al.~(1999). The nomograph of Maccacaro et al. (1988) gives a ratio $f_{\\mathrm{x}}/f_{\\mathrm{v}}$ between $1.3$ and $3.8$ (depending on whether we use the {\\it HST}\\ optical magnitudes through the F606W or F450W filter, respectively), in the regime of AGNs and BL Lacs.  The spectral indices can be used to deduce some general properties of the dominant radiation mechanism.  If $\\alpha_{\\mathrm{x}} \\leq \\alpha_{\\mathrm{r}}$, the source may exhibit relativistic beaming, while if $\\alpha_{\\mathrm{x}} > \\alpha_{\\mathrm{r}}$, it may fulfill the conditions of the homogeneous synchrotron model (Harris \\& Krawczynski 2002). In our case, we have $\\alpha_{\\mathrm{x}} \\approx \\alpha_{\\mathrm{r}} \\approx 1$, so that we cannot clearly discriminate between these two cases.  If the source is a high-frequency peaked BL Lac, however, it is likely that the homogeneous synchrotron model is more applicable. This is confirmed by the $\\alpha_{\\mathrm{xox}}$ test of Sambruna et al. (1996): in this case, the difference by $\\alpha_{\\mathrm{ox}}-\\alpha_{\\mathrm{x}}=\\alpha_{\\mathrm{xox}}$ is approximately equal to zero, so avoiding a clear discrimination between the physical mechanism of the source.  The spectral energy distribution (Fig.~\\ref{fig4}) is comparable to those of BL Lac objects peaked in the $\\gamma$-ray domain (e.g., Fossati et al. 1998), but the third EGRET catalog (Hartman et al. 1999) does not have any source within several degrees of XMMU~J$124825.9+083020$.  \\subsection{Search for ULXs and Contamination with Background Objects} This research is a useful demonstration of just how difficult it is to identify the physical nature of ULXs.  It shows the vital importance of redshift determinations.  Although we had a lot of photometric data on XMMU~J$124825.9+083020$, by themselves they were insufficient to clearly establish whether the source belongs to NGC~4698 or is a background object. In the first case, the source would have been something similar to a microquasar (e.g., Mirabel \\& Rodr\\'{\\i}guez 1999), probably located in a globular cluster. Indeed, the optical image of the source appeared to be slightly extended (angular extent $\\sim 3\\arcsec$), and its position with respect to NGC~4698 suggested that it could be a globular cluster, albeit an unusually large one. The possibility of an accreting black hole in a globular cluster is not so remote: \\emph{Chandra} observations of the globular cluster system of NGC~4472 show that about $40$\\% of the bright low mass X-ray binaries are associated with optically identified globular clusters (Kundu et al. 2002). In addition, the X-ray luminosity function shows a break near $3\\times 10^{38}$ \\lum, suggesting that the brightest X-ray binaries are accreting black holes (Kundu et al. 2002), perhaps microquasars.  \\emph{XMM-Newton} observations of the Lockman Hole (Hasinger et al. 2001) show that the number of background sources in the energy band $0.5-2$~keV with flux greater than $4.0\\times 10^{-14}$~erg cm$^{-2}$ s$^{-1}$ (the best fit value from Tab.~\\ref{tab:xdata}) is $15$ deg$^{-2}$.  In the energy band $2-10$~keV, there are 40 sources deg$^{-2}$ with flux higher than $4.6\\times 10^{-14}$~erg cm$^{-2}$ s$^{-1}$.  Assuming the same $\\log N - \\log S$ relation, and considering that the $D_{25}$ area of NGC~4698 is about $7.9$~arcmin$^2$, we expect $0.08$ background objects in the $2-10$~keV energy band and $0.03$ in the energy band $0.5-2$~keV. However, despite these low values, we have found, in the present case, that the only ULX is a background AGN.  The above calculations could be underestimated in the present case because the VLT images show an unknown concentration of galaxies north-east of NGC~4698 (see Fig.~\\ref{fig1}), thus suggesting the possibility of a statistically meaningful excess of background sources. However, no additional X-ray sources is seen in the present \\emph{XMM-Newton} observation. Perhaps, a longer exposure may reveal soft X-ray emission or additional sources, if NGC 4698 lies along the line of sight to a galaxy cluster.  The three X-ray sources identified to date, however, have three redshifts: XMMU~J$124825.9+083020$ has $z=0.43$, NGC~4698 has $z=0.0033$, and the {\\it ROSAT}\\ source 1RXS~J$124828.1+083103$ has been recently identified with a Seyfert nucleus at $z=0.12$ (Xu et al. 2001). Therefore, these three sources are not members of a single cluster."
    },
    "0209/hep-ph0209203_arXiv.txt": {
        "abstract": "We show that various scalar field models of dark energy predict degenerate luminosity distance history of the Universe and thus cannot be distinguished by supernovae measurements alone. In particular, models with a vanishing cosmological constant (the value of the potential at its minimum) are degenerate with models with a positive or negative cosmological constant whose magnitude can be as large as the critical density. Adding information from CMB anisotropy measurements does reduce the degeneracy somewhat but not significantly. Our results indicate that a theoretical prior on the preferred form of the potential and the field's initial conditions may allow to quantitatively estimate model parameters from data. Without such a theoretical prior only limited qualitative information on the form and parameters of the potential can be extracted even from very accurate data. ",
        "introduction": "One of the standard methods of interpreting the growing body of evidence  from supernovae \\cite{data} and other measurements  that the expansion of the Universe is accelerating, is to assume the existence of a dark energy component and to model it using scalar fields (for a recent review see \\cite{fund}). This links the expansion history of the Universe to theories of fundamental physics. For example, from this perspective the value of the potential at its minimum is the cosmological constant (CC). Since at this point there are many theoretical ideas about the form of the potential but none that particularly stands out, it would have been helpful if the data from cosmological measurements, such as supernovae Ia, CMB  and various others, could provide hints about some generic features of the potential.  Viable scalar field models of dark energy need to have potentials whose energy scale is about the critical density $\\sim 10^{-12}{\\rm eV}^4$, and typical scalar field masses about the Hubble mass $m\\sim 10^{-33}$ eV. In such models typical scalar field variations are about Planck scale $m_p\\sim 10^{19}$ GeV, and typical time scales for such variations are about the Hubble time $1/H_0\\sim 10^{18}$ sec. Whether, and how well, it is possible to determine the parameters and form of potentials and the field's dynamical history and future from data beyond such qualitative estimates has been addressed previously \\cite{Chiba,Barger,weller,Ng,Bludman,Eriksson,hut,star1,efst}. Weller and Albrecht \\cite{weller} concluded that some potentials could be differentiated using SNAP-like data \\cite{snap}. They approximated the equation of state (EOS) of each of the models, and showed by likelihood analysis that some of the models are distinguishable. Another approach is ``reconstruction\" \\cite{hut,star1,efst}. Here one rewrites the potential as a function of the luminosity-distance ($d_L$) and its derivatives, which are in turn functions of the redshift. The potential (and not the EOS) is approximated by a fitting function, and statistically tested against a set of accurate $d_L$ measurements. The efficiency of reconstruction depends on the accuracy of knowing the values of $\\Omega_m$ (this is needed also when one fits the EOS), and $H_0$ (this is needed only for reconstruction).  On general grounds we expect that scalar field potentials are less distinguishable than their corresponding EOS, because different potentials, with properly adjusted initial conditions for the field can produce very similar EOS.  In previous papers \\cite{m1,m2} we have found that supernovae (SN) measurements are limited as a probe of the dark energy EOS $w_Q$, due to degeneracies. Specifically, it was shown that $d_L$'s corresponding to two different $w_Q$'s are degenerate if both EOS coincide at some point at a relatively low red shift, $z^*$ (see also \\cite{sweetspot} and \\cite{astier}). The purpose of the present analysis is to explore the implications of this degeneracy on the possibility to determine the scalar field potential. For a given functional form of potential, we would like to quantify the amount by which the parameters of the potential can be varied, and still be indistinguishable by accurate SN measurements. Our criteria for indistinguishability between two models is that their resulting $d_L$'s differ at most by 1\\% up to redshift $z=2$, in accordance with the anticipated accuracy of future SN measurements.  In addition, we would like to determine whether the functional form of the potential can be distinguished or constrained by data.    We look for degeneracies among potentials using the following procedure: For a given class of potentials, we change the parameters as well as the initial values of the field, with the constraint that $w_Q$ at $z^*$ remains unchanged. This results in models whose $w_Q$ cross at $z^*$. We know from our previous analysis that in this case the models tend to be degenerate. Then we evaluate numerically the differences in the $d_L$'s of the models to verify this.  There are additional sources of degeneracy that we do not consider here. In our procedure  the value of the potential energy and the value of the kinetic energy remain unchanged. Allowing changes in the potential that are compensated by changes in the initial conditions for the kinetic energy will give another dimension of degeneracy. Variation in the value of $\\Omega_m$ is yet another degree of freedom, as is relaxing the assumption of a flat Universe and considering the effects of a clumpy Universe \\cite{sps}. Additionally, the value of $z^*$ in different models can be shifted. We have found that $z^*$ varies slowly with the red shift depth of the data set, $z_{max}$. The dependence is approximately linear $z^*=\\alpha z_{max}+\\beta$ (see also \\cite{linder}), $\\alpha$ and $\\beta$ are model dependent but $\\alpha$ is typically small, about 0.2. And, finally, we have looked only at a class of simple potentials that have two independent parameters. Additional parameters in the potential yet again open up new degrees of freedom, each of which increases the degeneracy of each of the parameters. Since we have found that this class of simple potentials suffers from large degeneracies, we see no phenomenological justification for using more complex potentials. If a theoretical prior about the form of the potential and initial conditions can be motivated then some of this degeneracy can be removed. ",
        "conclusions": "We have found that it is not possible to obtain  precise quantitative estimates  for parameters of scalar field models of dark energy from data alone beyond the obvious order of magnitude estimates. This is  due to theoretical degeneracies, which would persist even with expected future data from the most accurate SN and CMB measurements.  Theoretical prior knowledge or assumptions on the form of the potential and the field's initial conditions (preferably leaving a total of just two free parameters) may allow a more quantitative determination. For example, assuming that the field is at rest at the bottom of the potential is equivalent to having a pure cosmological constant. In this case the magnitude of the cosmological constant can be determined with accurate data to within a few percent."
    },
    "0209/astro-ph0209592_arXiv.txt": {
        "abstract": "{\\small The results of photometric CCD monitoring and spectral observations of the black hole binary and microquasar V4641 Sgr in the quiet state are presented. The ellipsoidal light curve with large amplitude of 0$^m$.36 in $R$ band suggests the influence of a massive object orbiting around a normal B9 star. In the spectra taken with the 6-m telescope one hour before black hole inferior conjunction, an absorption component in the red wing of \\ha line is visible. It is formed by gaseous stream moving in the direction to the normal star. That suggests the grazing conjunction in this system. Maximum velocity of the stream is of 650 km/s. Assuming that the stream is moving through the circular Keplerian orbit around black hole, the mass of the black hole is determined to be $M_{BH}$=7.1$-$9.5\\msun, what confirms the model by Orosz \\etal } ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209071_arXiv.txt": {
        "abstract": "A new exact solution of the Einstein-Maxwell equations for the gravitational collapse of a shell of matter in an already formed black hole is given. Both the shell and the black hole are endowed with electromagnetic structure and are assumed spherically symmetric. Implications for current research are outlined. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209247_arXiv.txt": {
        "abstract": "We present self-consistent dynamical models for dust driven winds of carbon-rich AGB stars. The models are based on the coupled system of frequency-dependent radiation hydrodynamics and time-dependent dust formation. We investigate in detail how the wind properties of the models are influenced by the micro-physical properties of the dust grains that enter as parameters. The models are now at a level where it is necessary to be quantitatively consistent when choosing the dust properties that enters as input into the models. At our current level of sophistication the choice of dust parameters is significant for the derived outflow velocity, the degree of condensation and the estimated mass loss rates of the models. In the transition between models with and without mass-loss the choice of micro-physical parameters turns out to be very significant for whether a particular set of stellar parameters will give rise to a dust-driven mass loss or not. ",
        "introduction": "Mass loss by dust driven winds of asymptotic giant branch (AGB) stars probably is one of the major mechanism which recycle material in the Galaxy (e.g.\\ Sedlmayr 1994). Most stars ($M_{\\star} < 8 M_{\\odot}$) will eventually become AGB stars and subsequently end their life as white dwarfs surrounded by planetary nebulae.  AGB stars are cool ($T_{\\star} < 3500$~K) and luminous ($L_{\\star}$ of a few $10^{3}$ to a few $10^{4}$~L$_{\\odot}$) and a majority of them are pulsating long-period variables (LPVs). The outer layers of many AGB stars provide favorable conditions for the formation of molecules and dust grains. Dust grains play an important role for the heavy mass loss of these stars (up to ($\\dot{M} \\sim 10^{-4} M_{\\odot}$/yr).  Pulsation causes an extended atmosphere where the dust is condensing. The dust absorbs the light of the central star and re-radiates it at longer wavelengths in the infrared rage ($\\lambda > 2 \\mu$m). The density of material in the circumstellar envelope may be so large that the star becomes completely obscured in the optical rage. Due to observational difficulties in determining the mass loss rates of these objects considerable effort has been put into a theoretical description of the mass loss of AGB stars (e.g.\\ Fleischer et al.\\ 1992; H\\\"ofner \\& Dorfi 1997).  The overall goal is to develop a mass loss description that can be used as input to evolutionary models and thereby probe the chemical evolution of the Galaxy.  We present here how the wind properties of carbon-rich models are influenced by the micro-physical properties of the dust grains.  \\begin{figure} \\plotfiddle{AGB_star.eps}{10 cm}{0}{45}{45}{-150}{-0} \\caption{A schematic drawing of the atmosphere of an AGB star and the most important physical processes. (Courtesy of Christer Sandin).} \\end{figure} ",
        "conclusions": "We have investigated in detail how the predicted wind properties of the carbon-rich AGB models are influenced by the choice of micro-physical dust parameters.  For the mass loss models it is important to know how the uncertainty in the chosen dust parameters affects the obtained results. The choice of the micro-physical parameters can change the predictions for the mean outflow velocity of the gas and dust by a factor of 4, the predicted degree of dust condensation by a factor of 10 and the predicted mass loss by a factor of 2. In the transition between models with and without mass-loss the choice of micro-physical parameters is vital for whether a particular set of stellar parameters will give rise to a dust-driven mass loss or not."
    },
    "0209/astro-ph0209137_arXiv.txt": {
        "abstract": "Improving angular resolution is one of X-ray astronomy's big challenges. While X-ray interferometry should eventually vastly improve broad-band angular resolution, in the near-term, X-ray telescopes will sacrifice angular resolution for increased collecting area and energy resolution. Natural occultations have long been used to study sources on small angular scales, but are limited by short transit times and the rarity of transits.  We describe here how one can make use of an {\\it X-ray Occulting Steerable Satellite\\/} ({\\it XOSS\\/}) to achieve very-high resolution of X-ray sources, conventional X-ray telescopes. Similar occulting satellites could also be deployed in conjunction with future space observatories in other wavebands. ",
        "introduction": "One challenge of X-ray astronomy is the relatively low photon fluxes from target sources. X-ray telescope builders are forced to trade angular resolution against increased collecting area. Thus, while the diffraction limit of a $1.2\\meter$ aperture X-ray telescope (such as Chandra) is $0.3\\mas$ at $1\\keV$, the $0.5\\as$ reality is worse than what is routinely achieved at longer wavelengths. While development proceeds on space-based X-ray interferometers, near-future X-ray missions such as {\\it Constellation X\\/} plan to increase effective area at the price of reduced angular resolution. {\\it MAXIM Pathfinder}, which will afford $100\\;\\mu$as resolution , hopes to launch between 2008 and 2013; however, it will have an effective area of only $100\\;{\\rm cm}^2$.  In the interim, angular resolution can be substantially improved for bright, relatively stable sources. The general technique to do this is well-known---eclipse mapping. When a body such as the moon transits a telescope's field-of-view (FOV), it occults different sources within the FOV at different times.  By carefully measuring the photon count rate as a function of time during the transit, one can reconstruct the projection of the surface brightness in the FOV onto the path of the occulter.  Deployment of large occulting satellites has been discussed for the optical and near infra-red \\citep{Adams,Schneider,CS98,CS00}, for both planet finding and high-resolution astronomy.  However, occulting satellites are particularly well-suited for use in the X-ray.  In the X-ray, the diffraction of the photons by the satellite can generally be neglected (see section \\ref{sec:resolve} below).  The optimal size and placement of the satellite are therefore governed chiefly by one's ability to accurately position the satellite and station-keep with respect to the line-of-sight to the source. The achievable resolution is then determined primarily by the telescope collecting area and by the level to which one can zero the {\\it XOSS\\/}-telescope relative velocity.  We consider an X-ray telescope at the the Earth-Sun system's second Lagrange point (L2) (such as {\\it Constellation X\\/}). We discuss the size and steering of a {\\it XOSS\\/} in section \\ref{sec:buildanduse}.  In section \\ref{sec:resolve} we present the two-source angular resolution as a function of the {\\it XOSS}--telescope relative angular velocity and of the photon count rate. In section \\ref{sec:reconstruction} we reconstruct some test sources. Specific classes of sources are discussed briefly in section~\\ref{sec:results}.  Section~\\ref{sec:conclude} contains our conclusions. ",
        "conclusions": "\\label{sec:conclude}  We have found that a {\\it XOSS\\/} can lead to tremendous improvements in angular resolution.  The current trend of increasing the effective area of X-ray telescopes at the expense of angular resolution until X-ray interferometry becomes viable, meshes well with the benefits gained by including a {\\it XOSS\\/} in the mission.  For the high Earth-orbit X-ray telescope a moderate improvement in angular resolution over an appreciable fraction of the sky can be achieved through the use of a {\\it XOSS}, however the orbital mechanics are challenging and the scientific pay-off may be modest.  A technology demonstration mission along these lines may nevertheless be worthwhile.  {\\it XOSS\\/} is potentially most interesting for telescopes at L2 and other low-acceleration environments (e.g.~Earth-trailing or drift-away orbits) where $1$--$10\\mas$ resolution should be achievable for a wide range of sources.  The ideal telescope would have an angular resolution equal to the angle subtended by the XOSS.  Further investigations are necessary to determine whether XOSS is well suited to {\\it Constellation X\\/}, or whether a dedicated telescope with better angular resolution would be merited.  Either way, it seems likely that XOSS could provide a natural scientific precursor to interferometric missions like {\\it MAXIM Pathfinder\\/}, increasing the angular resolution of X-ray telescopes by one-to-two orders-of-magnitude at relatively modest incremental expense.   The authors thank A. Chmielewski for support, A. Babul and M. Dragovan for many useful comments and suggestions, N. Choudhuri for helpful suggestions on statistical tests, P. Gorenstein and W. Zhang for comments on a preliminary version of the manuscript and C. Covault and S. Rodney for recent input. This work was supported by a DOE grant to the theoretical particle-astrophysics group at CWRU, and a phase I grant from NIAC."
    },
    "0209/astro-ph0209184_arXiv.txt": {
        "abstract": "{  This paper contains the general data reduction methods used in processing the data from the Carlsberg Meridian Telescope CCD Drift Scan Survey. An efficient method to calibrate the fluctuations in the positions of the images caused by atmospheric turbulence is described. The external accuracy achieved is 36 mas in right ascension and declination. A description of the recently released catalogue is given. ",
        "introduction": "The Carlsberg Meridian Telescope (CMT) has recently undergone a major upgrade. A 2k by 2k CCD camera has been installed with a Sloan $r'$ filter operating in a drift scan mode. With the new system, the effective exposure time is about 90s, the magnitude limit is $r'_{\\rm CMT}=$17 and the initial positional accuracy is in the range 0.05$''$ to 0.10$''$.  The main task of the CMT is to map the sky in the declination range $-$3$\\degr$ to $+$30$\\degr$ with the aim of providing an astrometric, and photometric, catalogue that can accurately transfer the Hipparcos/Tycho reference frame to Schmidt plates. A secondary survey is also planned that extends the declination range covered to $-$15$\\degr$ in the South and $+$50$\\degr$ in the North. Projects similar to the CMT (UCAC, \\cite{Zacharias} and CMASF, \\cite{CMASF}), which include the Southern hemisphere, will also be able to provide astrometric calibration for VISTA and other deep wide-field surveys.  The two main systematic errors affecting the data are caused by image motions due to long timescale atmospheric turbulence and charge transfer efficiency (CTE) problems linked to the CCD. Methods are described on how to calibrate these errors. ",
        "conclusions": "By upgrading the Carlsberg Meridian Telescope to have a CCD operating in drift-scan mode, a new lease of life has been breathed into the telescope. It should be pointed out that this will only be useful over the next ten years or so. Then, data from astrometric satellites such as DIVA and GAIA will become generally available and supersede the astrometric accuracy of what can be achieved from the ground. It is thus important that planned upgrades of meridian telescopes are carried out as soon as possible and that the results are published promptly so that the maximum use can be made of the data.  The results shown here demonstrate that using transfer functions it is possible to calibrate the fluctuations caused by atmospheric turbulence using just the Tycho 2 stars. This is a more efficient method than using a subcatalogue since multiple measurements of the sky are not required.  After this calibration, the external accuracy achieved for the brightest stars in the survey is 36 mas in right ascension and declination and 0.025 magnitudes in $r'_{\\rm CMT}$ photometry.  The web site of the telescope is at:\\\\ http://www.ast.cam.ac.uk/\\verb+~+dwe/SRF/camc.html"
    },
    "0209/astro-ph0209467_arXiv.txt": {
        "abstract": "We follow the timing properties of the neutron star low-mass X-ray binary system \\fouru~in different spectral states, as monitored by the Rossi X-ray Timing Explorer over about a month. We fit the power density spectra using multiple Lorentzians. We show that the characteristic frequencies of these Lorentzians, when properly identified, fit within the correlations previously reported. The time evolution of these frequencies and their relation with the parameters of the energy spectra reported in Barret \\& Olive (2002) are used to constrain the accretion geometry changes. The spectral data were fitted by the sum of a blackbody and a Comptonized component and were interpreted in the framework of a truncated accretion disk geometry, with a varying truncation radius. If one assumes that the characteristic frequencies of the Lorentzians are some measure of this truncation radius, as in most theoretical models, then the timing data presented here strengthen the above interpretation. The soft to hard and hard to soft transitions are clearly associated with the disk receding from and approaching the neutron star respectively. During the transitions, correlations are found between the Lorentzian frequencies and the flux and temperature of the blackbody, which is thus likely to be coming from the disk. On the other hand, in the hard state, the characteristic Lorentzians frequencies which are at the lowest, remained nearly constant despite significant evolution of the spectra parameters. The disk no longer contributes to the X-ray emission, and the blackbody is now likely to be emitted by the neutron star surface which is providing the seed photons for the Comptonization. ",
        "introduction": "Rapid X-ray variability is a powerful probe of physics of accretion flows around neutron stars and black holes. Significant progress has been accomplished recently with the advent of the {\\it Rossi X-ray Timing Explorer} (\\rxte, Bradt, Rothschild \\& Swank 1993). Its large collecting area and micro-second time resolution allowed us to study with exquisite details the rapid variability of a wide variety of accreting X-ray sources. In particular, the power density spectra (PDS) of Galactic black hole candidates and neutron star low-mass X-ray binaries (LMXBs) exhibit a variety of features ranging from narrow quasi-periodic oscillations (QPOs) to broad noise components (for a review see Wijnands 2001 and references therein).  It has been shown recently that these PDS can be well represented as a superposition of a few Lorentzians (e.g.\\ Olive et al.\\ 1998; Van Straaten et al.\\ 2002; Belloni, Psaltis \\& Van der Klis 2002). The multi-Lorentzian approach gives a simple and universal phenomenological description of the PDS. Each noise component (Lorentzian) is then described by its characteristic frequency, width and amplitude, usually expressed as a fractional root-mean-square (RMS). It has been demonstrated that these characteristic frequencies are correlated for a given source and between sources, both for neutron star and black hole binaries (Wijnands \\& Van der Klis 1999; Psaltis, Belloni \\& van der Klis 1999; Belloni et al.\\ 2002). In most models of rapid X-ray variability, these frequencies are related to some kind of an abrupt transition in the accretion disk (e.g.\\ Stella, Vietri \\& Morsink 1999; Titarchuk \\& Osherovitch 1999; Psaltis \\& Norman 2002). This could be a truncation of a Shakura-Sunyaev disk into a hot inner flow (e.g.\\ R{\\'o}{\\.z}a{\\'n}ska \\& Czerny 2000). With the truncation radius decreasing (the disk coming in) the characteristic frequencies increase.  \\fouru~is a neutron star LMXB system which belongs to the class of atoll sources (Hasinger \\& Van der Klis 1989). The source shows variability on all time scales, from months down to milliseconds (Langmeier et al.\\ 1987; Berger \\& Van der Klis 1998, Ford, van der Klis \\& Kaaret 1998, Liu, van Paradijs \\& van den Heuvel 2001 and references therein). On long time scales, it displays clear luminosity--related spectral changes. This is illustrated in the observations reported by Barret \\& Olive (2002; hereafter Paper I) who followed the spectral evolution of the source during a transition between soft and hard X-ray spectral states. The data were interpreted in the framework of a truncated accretion disk geometry for which the spectral transitions observed were associated with changes in the disk truncation radius.  In this paper, we follow the multi-Lorentzian approach to study the rapid X-ray variability of the source using the same set of observations as presented in Paper I. The aim of this work is to look for correlations between the timing and spectral parameters to get further insights on the accretion flow changes associated with the state transitions. ",
        "conclusions": "The spectral and timing data of the state transitions observed from \\fouru~can be explained in the framework of a model, in which the critical parameter is the position of the truncation radius between the disk and a hotter inner flow. This parameter sets both the frequencies of the timing features and the spectral shape. What sets the value of the truncation radius is unclear at the moment, but both the timing and spectral data indicate that it cannot be the mass accretion rate derived from the bolometric source luminosity. One possibility could be this radius is set by some long-time averaging process over time scales of days.  \\newpage"
    },
    "0209/astro-ph0209521_arXiv.txt": {
        "abstract": "An adaptive multi grid approach to simulating the formation of structure from collisionless dark matter is described. {\\tt MLAPM} (Multi-Level Adaptive Particle Mesh) is one of the most efficient serial codes available on the cosmological ``market'' today.  As part of Swinburne University's role in the development of the Square Kilometer Array, we are implementing hydrodynamics, feedback, and radiative transfer within the {\\tt MLAPM} adaptive mesh, in order to simulate baryonic processes relevant to the interstellar and intergalactic media at high redshift.  We will outline our progress to date in applying the existing {\\tt MLAPM} to a study of the decay of satellite galaxies within massive host potentials. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209229_arXiv.txt": {
        "abstract": "Using archival Far Ultraviolet Spectroscopic Explorer ({\\it FUSE}) and Hubble Space Telescope ({\\it HST}) data, we have assembled a survey of eight sightlines through high-velocity cloud Complex C.  Abundances of the observed ion species vary significantly for these sightlines, indicating that Complex C is not well characterized by a single metallicity. Reliable metallicities based on [\\ion{O}{1}/\\ion{H}{1}] range from 0.1-0.25 solar. Metallicities based on [\\ion{S}{2}/\\ion{H}{1}] range from 0.1-0.6 solar, but the trend of decreasing abundance with \\ion{H}{1}\\ column density indicates that photoionization corrections may affect the conversion to [S/H].  We present models of the dependence of the ionization correction on \\ion{H}{1}\\ column density;  these ionization corrections are significant when converting ion abundances to elemental abundances for S, Si, and Fe. The measured abundances in this survey indicate that parts of the cloud have a higher metallicity than previously thought and that Complex C may represent a mixture of ``Galactic fountain'' gas with infalling low-metallicity gas.  We find that [S/O] and [Si/O] have a solar ratio, suggesting little dust depletion. Further, the measured abundances suggest an over-abundance of O, S, and Si relative to N and Fe.  The enhancement of these $\\alpha$-elements suggests that the bulk of the metals in Complex C were produced by Type II supernovae and then removed from the star-forming region, possibly via supernovae-driven winds or tidal stripping, before the ISM could be enriched by N and Fe. ",
        "introduction": "Almost four decades after their discovery (Muller, Oort, \\& Raimond 1963), the origin of high-velocity clouds (HVCs; Wakker \\& van Woerden 1997) remains a mystery. Defined by their incompatibility with a simple model of differential Galactic rotation, HVCs exhibit sub-solar metallicities and are currently favored to reside in the Galactic halo.  Within this picture, it is not clear whether the source of HVCs is predominantly the accretion of low-metallicity gas onto the Galaxy (Wakker et al. 1999) or whether the material traces condensed outflows of hot enriched gas in a ``Galactic fountain'' (Shapiro \\& Field 1976; Bregman 1980). Intermediate-velocity clouds (IVCs), in contrast, exhibit solar metallicities and are thought to be of Galactic origin (Richter et al. 2001). Alternative hypotheses for the HVCs identify the majority of them as extragalactic objects within the Local Group at Galactocentric distances up to 500 kpc (Blitz et al. 1999), or posit that at least the subset known as Compact High Velocity Clouds (CHVCs) are extragalactic (Braun \\& Burton 1999).  These objects would then be remnants of the formation of the Local Group. The statistics of moderate-redshift Mg II and Lyman limit absorbers (Charlton, Churchill, \\& Rigby 2000), would seem to rule out this hypothesis as a general scenario for group formation. In any event, the simulations which underpin the arguments for an extragalactic origin for the HVCs generally do not require such an identification for the very large complexes of HVCs, such as Complex C. However, within their model it is possible that Complex C might represent one of these building blocks viewed at a very close distance.  Complex C occupies nearly the same angular extent ($\\sim2000$ deg$^{2}$) as the Magellanic Stream and consists of a number of separate HVC cores in the northern Galactic hemisphere.  The Magellanic Stream has been linked to the Small Magellanic Cloud (SMC) through modeling of a tidal interaction with the Milky Way (Gardiner \\& Noguchi 1996) and through the similarity in metallicity of the Stream (0.2-0.3 solar; Lu et al. 1998; Gibson et al. 2000; Sembach et al. 2001) and SMC.  In contrast, Complex C shows no obvious connection to the Galaxy.  The distance to Complex C is at least 1.5 kpc (de Boer et al. 1994) from the non-detection of absorption lines toward high-latitude stars, although Wakker (2001) has recently put a new, albeit weak, lower limit at 6 kpc.  The lack of a strong distance constraint or evidence for interaction with the Galaxy make absorption-line studies of abundances in quasar sightlines the best option for gaining insight into the origin of Complex C.  The first such study for Complex C, from {\\it HST} Goddard High Resolution Spectrograph (GHRS) data along the Markarian 290 sightline by Wakker et al. (1999), inferred a metallicity of 0.1 solar based on the H$\\alpha$-corrected [\\ion{S}{2}/\\ion{H}{1}] abundance. Those authors interpreted this abundance as evidence for infalling low-metallicity gas from the intergalactic medium (IGM) posited to explain the ``G-dwarf problem.'' Recently, however, other studies have demonstrated that the metallicity of Complex C varies from sightline to sightline.  Gibson et al. (2001; hereafter G01) looked at all available archival GHRS and Space Telescope Imaging Spectrograph (STIS) data for five Complex C sightlines (including Mrk 290) and found that the metallicity inferred from the \\ion{S}{2} abundance varies between 0.1-0.4 solar, implying that Complex C may be contaminated by enriched gas from the Galactic disk. Other recent studies complicate the picture and further illustrate the point that Complex C may not be well described by a single metallicity. Richter et al. (2001a) combine {\\it FUSE} and STIS data for the PG 1259+593 sightline and find an \\ion{O}{1}\\ abundance of 0.1 solar, while Murphy et al. (2000) find an abundance of \\ion{Fe}{2}\\ near 0.5 solar from {\\it FUSE} observations of the Mrk 876 sightline.  The increasing availability of high-quality, far-ultraviolet {\\it FUSE} and {\\it HST}/STIS spectra has allowed the study of multiple transitions of the dominant ionization stages of N, O, Si, Fe, Ar, P, and S.  In addition to estimates of the absolute abundances of these elements, the relative abundances of these elements can be inferred, sometimes more precisely than their absolute abundances. This allows a preliminary study of the depletion of refractory elements onto dust grains and the nucleosynthetic history of the gas.  Using {\\it HST}-GHRS spectra, Savage \\& Sembach (1996) find that gas-phase abundances of refractory elements in the Galactic halo can be significantly depleted.  For five halo sightlines, those authors find that the mean depletion from solar abundance is $\\sim$ 0.3 dex for Si and $\\sim$ 0.6 dex for Fe, while the S abundance is essentially solar. However, it is difficult to completely separate the effects of depletion from ionization effects. Although each element has a dominant ionization stage, no element is completely in one stage, and the fraction remaining in other stages varies with the element.  As we show below, with enough measurements of different elements, it is possible to separate these two effects and assess the degree of depletion along sightlines toward Complex C.  Recent reviews have discussed the long history of exploiting relative abundances to reveal information about the chemical evolution of the gas (e.g., Pettini 2002; Matteucci 2002). Alpha-process elements (such as O, Ne, Si, and S) are produced predominantly by Type II supernovae (SNe), while Fe is produced primarily by Type Ia SNe.  Nitrogen, however, is primarily produced by lower mass stars in the asymptotic giant branch (AGB) stage.  With the bulk of $\\alpha$-element production occurring relatively soon after star-formation is initiated, the release of N and Fe into the ISM can thus lag S or O enrichment by $\\sim250$ Myr (Henry et al. 2000).  If the gas is somehow stripped from star-forming regions, either by a tidal interaction or in an outflow induced by combined SN-activity, then the resulting metal-content will be significantly N-poor.  G01 and Wakker (2001) each interpret the low relative abundance of N in Complex C as evidence for primary enrichment by Type II SNe.    The cited Complex C abundance studies have all used different methods to infer metallicities of Complex C.  Some studies fit the Complex C metal-line profiles with a Gaussian to determine the equivalent width, $W_{\\lambda}$, while others simply integrate the profile over the Complex C velocity range established by the \\ion{H}{1}\\ profile.  These studies have also used different methods to determine species' column densities, such as curve-of-growth (CoG) fitting, an apparent optical depth (AOD) calculation (Savage \\& Sembach 1991), or an assumption that these profiles are optically thin.  In addition, except for the Richter et al. (2001a) work, these studies have typically concentrated on either {\\it FUSE} or {\\it HST} data alone. The {\\it FUSE} data have the advantage that the covered spectral range includes numerous lines of \\ion{O}{1}, whose abundance is not as strongly influenced as \\ion{S}{2}\\ by ionization effects (charge exchange keeps \\ion{O}{1}\\ closely coupled to \\ion{H}{1}).  The goal of this work is to study all archival {\\it FUSE} and {\\it HST} data for sightlines through Complex C. Using a consistent method of analysis, we then determine abundances for the various metal species to infer information regarding the origin of HVC Complex C. The positions of the eight sightlines are shown in Figure 1, overlayed on \\ion{H}{1}\\ emission data for Complex C from the Leiden-Dwingeloo Survey (LDS; Hartmann \\& Burton 1997). The {\\it FUSE} and {\\it HST} observations are discussed in \\S\\ 2. Results for the analysis for the eight sightlines are presented in \\S\\ 3.  The results are discussed in \\S\\ 4. ",
        "conclusions": "\\subsection{Ionization Effects on Elemental Abundances}  For the eight sightlines, we have calculated abundances of various ion species, \\ion{O}{1}, \\ion{S}{2}, \\ion{Fe}{2}, \\ion{Si}{2}, and \\ion{N}{1}, which are the dominant ionization states in warm neutral clouds.  However, there are certain to be regions in these clouds with ionized hydrogen.  As a result, the measured column of \\ion{H}{1}\\ does not fully reflect the total column of H along the line of sight.  This is not a problem for elements such as N or O, where the ionization state is coupled to that of H through charge-exchange interactions.  Elements such as S and Si, however, are certain to be singly-ionized in both \\ion{H}{1}\\ and \\ion{H}{2}\\ regions.  Therefore, in order to make definitive statements regarding {\\it elemental} abundances, one must take into account possible photoionization corrections to ion abundances.  In Figure 15, we have plotted the runs of \\ion{S}{2}, \\ion{Si}{2}, \\ion{Fe}{2}, and \\ion{O}{1}\\ abundance versus $N$(\\ion{H}{1}).  The abundance [\\ion{S}{2}/\\ion{H}{1}], and to a lesser extent [\\ion{Si}{2}/\\ion{H}{1}], shows a trend of decreasing abundance with column density.  If one assumes that the elemental abundance does not vary significantly as a function of column density, then such a trend can be explained by an ionization effect.  In order to model the effect of column density on ion abundances, we have generated a grid of photoionization models using the code CLOUDY (Ferland 1996).  We make the simplified assumption that the absorbing gas can be treated  as plane-parallel slabs illuminated by incident radiation dominated by OB associations.  The models assume a $T=35,000$ K Kurucz model atmosphere and a cloud metallicity of $Z_{\\sun}=0.1$. While the extent of the incident ionizing field on Complex C is not well known, H$\\alpha$ emission measures of Complex C have been argued (Bland-Hawthorn \\& Putman 2001; Weiner et al. 2002) to be consistent with radiation from our Galaxy at the level of log $\\phi \\approx 5.5$ (photons cm$^{-2}$ s$^{-1}$), which we assume as the normally-incident ionizing flux in the models.  For three assumed gas volume densities, $n_{\\rm H}$, we calculate the logarithmic difference between [X$^{ion}$/\\ion{H}{1}] and [X$^{element}$/H] versus $N$(\\ion{H}{1}), shown in Figure 16 for S, Si, and Fe. We find that the ionization corrections necessary for O and N are negligible, and that [\\ion{O}{1}/\\ion{H}{1}]$_{\\rm C}$ and [\\ion{N}{1}/\\ion{H}{1}]$_{\\rm C}$ should reflect actual values of [O/H] and [N/H].  It would be useful to check this for N by measuring Complex C \\ion{N}{2}\\ $\\lambda1084$. However, this absorption feature cannot be measured in these sightlines since it is saturated and blended with saturated Galactic absorption.   Given the simplified nature of these models and the uncertainty in the actual gas densities, these models are for illustrative purposes only and are designed mainly to explain the general trend observed in the ion abundances versus \\ion{H}{1} column.  The models show that the spread in values of [\\ion{S}{2}/\\ion{H}{1}]$_{\\rm C}$ observed in these sightlines is more a reflection of ionization conditions in the cloud than actual variations in [S/H]. A similar statement can be made for the [\\ion{Si}{2}] trend. For the intermediate density model ($n_{\\rm H}=0.03$ cm$^{-3}$), the ionization correction to the [\\ion{S}{2}/\\ion{H}{1}]$_{\\rm C}$ puts those  values more in line with the [\\ion{O}{1}/\\ion{H}{1}]$_{\\rm C}$ measurements. Fe is not subject to as large a photoionization correction, which may explain the lack of a strong trend in [\\ion{Fe}{2}/\\ion{H}{1}]$_{\\rm C}$ versus $N$(\\ion{H}{1}).   \\subsection{The Origin of Complex C}  In Table 12, we show the measured abundances, relative to solar, of \\ion{O}{1}, \\ion{S}{2}, \\ion{Fe}{2}, \\ion{Si}{2}, and \\ion{N}{1}. As demonstrated in \\S\\ 4.1, many of these ion species' abundances are subject to ionization corrections in order to convert to elemental abundances.  Oxygen and nitrogen are particularly insensitive to such effects, and the values listed should accurately reflect the true values of [O/H]$_{\\rm C}$ and [N/H]$_{\\rm C}$.  \\placetable{t12}  We find that the metallicity of Complex C ranges from 0.1-0.25 solar for the three sightlines in which \\ion{O}{1}\\ could be measured.  The upper limits on [\\ion{O}{1}/\\ion{H}{1}]$_{\\rm C}$ established for the other sightlines are consistent with such a range.   As stated by G01, the moderate abundances imply that Complex C is unlikely to be representative of purely infalling extragalactic gas. Further, we find that the metallicity varies from sightline to sightline; the [\\ion{O}{1}/\\ion{H}{1}]$_{\\rm C}$ and ionization-corrected [\\ion{S}{2}/\\ion{H}{1}]$_{\\rm C}$, within their $1\\sigma$ error bars, suggest variations by factors of 2-3. Although N is produced through different mechanisms than O, the two cases for which \\ion{N}{1}\\ could be measured differ in abundance by 0.7 dex, further suggesting that Complex C cannot be characterized by a single metallicity. Metallicities in the range 0.1-0.3 solar are characteristic of tidally-disrupted dwarf-satellites (LMC and SMC) or gas in the outer disk ($>1.5R_{\\sun}$) of the Galaxy (Gibson 2002). While the calculated metallicities are still somewhat low to assign a Galactic origin, it is tempting to assign a mixed origin to Complex C, with gas of Galactic (and/or satellite) origin blending with infalling low-metallicity clouds. If this is the case, one might expect, with better data, to see a trend of lower metallicity with higher $N$(\\ion{H}{1}), since the mixing of infalling gas with enriched Galactic disk gas may take considerable time.  The interpretation of Complex C metallicity may therefore depend on the alignment of the background AGN with the cloud core or halo. We suggest that further {\\it FUSE} and {\\it HST} observations of many of these sightlines, together with higher-resolution 21-cm \\ion{H}{1}\\ maps, will more completely address these issues.  In order to further explore issues concerning the origin of Complex C, we have calculated elemental abundances relative to \\ion{O}{1}.  We consider data only from the three sightlines (Mrk 279, Mrk 817, and PG 1259+593) for which an \\ion{O}{1}\\ column density could be measured.  Figure 17 shows measured values of [\\ion{S}{2}/\\ion{O}{1}], [\\ion{Si}{2}/\\ion{O}{1}], [\\ion{Fe}{2}/\\ion{O}{1}], and [\\ion{N}{1}/\\ion{O}{1}], where we have plotted the column density weighted mean values for the three sightlines.  The upper limit for [\\ion{N}{1}/\\ion{O}{1}] is taken as the least restrictive of the limits from the various sightlines.  Using the ionization corrections discussed in \\S\\ 4.1 for the intermediate density model ($n_{\\rm H}=0.03$ cm$^{-3}$) at log $N$(\\ion{H}{1})=19.5, we can determine relative elemental abundances by adding an offset to the values illustrated in Figure 17.  With these corrections, we find that [S/O]$_{\\rm C}$ and [Si/O]$_{\\rm C}$ are consistent with a solar relative abundance.  The fact that Si is not depleted implies that dust is not a significant constituent of Complex C.  The possible absence of dust in Complex C is reinforced by the non-detection of H$_{2}$, whose formation is facilitated by surfaces of interstellar dust grains (Shull \\& Beckwith 1982). The corrected [Fe/O] is sub-solar by $\\sim0.2$ dex, which in the absence of dust depletions suggests a slight enhancement of the $\\alpha$-elements O, Si, and S. We note that [\\ion{Fe}{2}/\\ion{S}{2}] and [\\ion{Fe}{2}/\\ion{Si}{2}] are sub-solar. As a result, we can be reasonably confident in the corrected [Fe/O] given appropriate corrections to \\ion{S}{2}\\ and \\ion{Si}{2}. In addition, N is depleted by at least 0.5 dex relative to O for these three sightlines.  Evidence for under-abundant N is also apparent for the Mrk 876 sightline, where \\ion{N}{1} is reduced by 0.8 dex from solar with respect to the $\\alpha$-element ion \\ion{Si}{2}, much greater than any possible ionization correction along this sightline.  The enrichment of $\\alpha$-elements relative to Fe and the significant depletion of N suggest that the metals were produced primarily by massive stars and injected into the ISM by Type II SNe.  The gas could have then been removed from the star-forming regions before Fe and N production and delivery could pollute the ISM. If the gas originated in a star-forming region with sufficiently correlated SN activity, then it could have been driven into the halo through chimneys (Norman \\& Ikeuchi 1989) in a ``Galactic fountain'' scenario.  If the gas originated in the inner disk, then additional mixing with infalling extragalactic gas would be necessary to explain the metallicities. Alternatively, the similarity in metallicity to the LMC and SMC suggests that the gas may have been tidally stripped from the outer disk of the Milky Way or an unknown dwarf irregular galaxy. These scenarios are all speculative.  Our study suggests, however, that the data are not consistent with a simple model of an infalling low-metallicity cloud of extragalactic origin."
    },
    "0209/astro-ph0209535_arXiv.txt": {
        "abstract": "We address the classical stellar-atmosphere problem and describe our method of numerical solution in detail. The problem consists of the solution of the radiation transfer equation under the constraints of hydrostatic, radiative and statistical equilibrium (non-LTE). We employ the Accelerated Lambda Iteration (ALI) technique, and use statistical methods to construct non-LTE metal-line-blanketed model-atmospheres. ",
        "introduction": "Essentially all our knowledge about the structure and evolution of stars, hence about galactic evolution in general, rests on the interpretation of their electromagnetic spectrum. Therefore quantitative analysis of stellar spectra is one of the most important tools of modern astrophysics. The formation of the observed spectrum is confined to the atmosphere, a very thin layer on top of the stellar core. Spectral analysis is performed by modeling the temperature- and pressure-stratification of the atmosphere and computing synthetic spectra which are then compared to observation. Fitting synthetic spectra to the data yields basic photospheric parameters: effective temperature, surface gravity, and chemical composition. Comparison with theoretical evolutionary calculations allows one to derive stellar parameters: mass, radius, and total luminosity.  \\mbox{\\qquad}The classical stellar-atmosphere problem considers radiation transfer through the outermost layers of a star into free space under three assumptions. First it is assumed that the atmosphere is in hydrostatic equilibrium, thus, the matter which interacts with photons is at rest. Second, the transfer of energy through the atmosphere is assumed to be done entirely by photons, i.e.\\ heat-conduction and convection are regarded as negligible (radiative equilibrium). But the effectiveness of photon-transfer depends on the opacity and emissivity of the matter, which are strongly state- and frequency-dependent quantities. They depend in detail on the occupation numbers of atomic/ionic levels, which, in turn, are determined by the local temperature and electron density, and the radiation field, whose nature is {\\it non-local}. The occupation number of any atomic level is determined by a balance among radiative and collisional population and de-population processes (statistical equilibrium; our third assumption), i.e.\\ the interaction of atoms with other particles and photons. Mathematically, the whole problem requires the solution of the radiation transfer equations simultaneously with the equations for hydrostatic and radiative equilibrium, together with the statistical equilibrium (rate equations).  A stellar atmosphere radiates into circumstellar space and thus is an {\\it open thermodynamic system}; hence it cannot be in thermodynamic equilibrium (TE), and cannot be characterized by a single temperature. The idea of ``Local Thermodynamic Equilibrium'' (LTE) is a hypothesis which assumes that while TE does not hold for the atmosphere as a whole, it can be applied in small volume elements. In this case the atomic occupation numbers depend only on the local electron temperature and electron density via the Saha-Boltzmann equations. But this approximation can be valid only in the limit that collision rates dominate radiative rates, and photon-mean-free-paths are small. Models in which the Saha-Boltzmann equations are replaced by the physically more accurate rate equations are called non-LTE (or NLTE) models. This designation is somewhat imprecise because the velocity distribution of particles is still assumed to be Maxwellian, i.e.\\ we can still define a local temperature. NLTE calculations are more costly than LTE calculations, however, and it is hard to predict a priori whether NLTE effects will be important in a specific problem. Generally, NLTE effects are large at high temperatures and low densities, which imply intense radiation fields, hence frequent radiative processes (which have a non-local nature and, in particular, respond to the presence of the open boundary), and less-frequent particle collisions, which tend to enforce LTE conditions.  Abandonment of the LTE assumption leads to a much more difficult model-atmosphere problem, because of subtle couplings between the radiation transfer and statistical equilibrium equations. The pioneering work by Auer \\& Mihalas (1969) provided a basic tool for making such models. But the numerical problem going from LTE to {\\it realistic} NLTE models has been solved only recently, and is the topic of this paper. We now have much more powerful tools to compute non-classical models, which account for very complex opacities, and solve the radiation transfer equations in more general environments, for example in expanding stellar atmospheres. These are the topics of other papers in this volume.  Stellar-atmosphere modeling has made huge progress in recent years. This advance was achieved by the development of new numerical techniques for model-construction, and by the availability of atomic data for many species. And these achievements became possible only with an enormous increase of computing power. Model atmospheres assuming LTE have been highly refined by the inclusion of many more atomic and molecular opacity sources, and effective numerical techniques for LTE model-computation have been available for years. The progress is most remarkable in the field of NLTE model atmospheres. Replacement of the Saha-Boltzmann equations by atomic rate-equations requires radically different numerical-solution techniques, otherwise metal-opacities cannot be taken into account at all. Such techniques have been developed with great success during the last decade, inspired by important papers by Cannon (1973) and Scharmer (1981). The {\\it Accelerated Lambda-Iteration Method} (ALI) is at the heart of this development. Combined with statistical representations of line-opacities, we are finally able to compute metal-line-blanketed NLTE models, including many millions of spectral lines, with a very high level of sophistication.  In this paper we discuss the basic ideas behind the new numerical methods for NLTE modeling. We begin by presenting the classical model-atmosphere problem and introduce the basic equations. Then we focus on the ALI solution-method, and our numerical implementation of it. We then briefly describe the solution of the NLTE metal-line-blanketing problem. ",
        "conclusions": "We have presented in detail our technique for numerical solution of the classical model-atmosphere problem. The construction of metal-line-blanketed models in hydrostatic and radiative equilibrium under NLTE conditions was the last and longest-standing problem of classical model atmosphere theory. It has finally been solved with a high degree of sophistication. The essential milestones for this development, starting from the pioneering work of Auer \\& Mihalas (1969) are: \\begin{itemize} \\item Introduction of Accelerated Lambda Iteration (ALI, or ``operator splitting'') methods, based upon early work by Cannon (1973) and Scharmer (1981). The first ALI model atmospheres were constructed by Werner (1986). \\item Introduction of statistical approaches to treat the iron-group elements in NLTE by Anderson (1989, 1991). \\item Linear formulation of the statistical-equilibrium equations (Rybicki \\& Hummer 1991, Hauschildt 1993). \\item Computation of atomic data by Kurucz (1991), by the Opacity Project (Seaton \\etal 1994) and subsequent improvements, and by the Iron Project (Hummer \\etal 1993). \\end{itemize}"
    },
    "0209/astro-ph0209159_arXiv.txt": {
        "abstract": "We present VLT-UVES echelle spectroscopy of the \\HI~ and \\civ~absorption in the spatially-extended \\Lya~emission around two high-redshift radio galaxies 0200+015 (z=2.23) and 0943-242 (z=2.92).  The absorbers in 0943-242 exhibit little additional structure compared with previous low-resolution spectroscopy and the main absorber is still consistent with H{\\footnotesize{I}}~column density of $\\sim 10^{19}$\\pcmsq. This is consistent with a picture in which the absorbing gas has low density and low metallicity and is distributed in a smooth absorbing shell located beyond the emission-line gas. However, the main absorbers in 0200+015 are very different. The previous single absorber fit of \\HI~ column density $\\simeq 10^{19}$\\pcmsq, now splits into two $\\sim 4 \\times 10^{14}$\\pcmsq~absorbers which extend more than 15\\kpc~to obscure additional \\Lya~emission coincident with a radio lobe in these high-resolution observations. Although consistent with the shell-like distribution for the absorption systems, 0200+015 requires a much higher metal enrichment than 0943-242. The metallicity, inferred from the \\civ~ absorption, is considerably lower in 0943-242 than in 0200+015.  We explain these differences with an evolutionary scenario based on the size of the radio source.  In both sources the \\HI~ absorption gas originates from either a gas-rich merger or pristine cluster gas which cools and collapses towards the centre of the dark matter halo.  The higher metallicity in the larger radio source (0200+015) may be a result of a starburst driven superwind (concurrent with the triggering of the radio emission) which has engulfed the outer halo in this older source.  We also find a significant blue asymmetry in the HeII$\\lambda 1640$ emission line, suggesting that the line emitting gas is outflowing from the central regions. Dust obscuration toward the central engine, presumably due to the dusty torus invoked in Unified Scheme, prevents us from seeing outflow away from our line-of-sight. ",
        "introduction": "The existence of powerful radio galaxies at high redshift ($z>2$) demonstrates that a population of supermassive black holes ($> 10^{9}~\\Msun$; Dunlop et al. 2002) was in place just a few billion years after the Big Bang. Like their counterparts at low redshift, the high-redshift radio galaxies (HzRGs) appear to reside in the most massive elliptical galaxies at their epoch (e.g. Jarvis et al.~2001b), in line with the established correlation between black hole and spheroid mass (Magorrian et al.~1998). Whilst the discovery of radio quiet galaxies at high redshift via Lyman drop-out techniques means that they are no longer our only probe of high redshift galaxy formation, the HzRGs still provide the most important insights into the early formation of the most massive bound structures. Indeed, the depth of the gravitational potential wells in which they form renders them prime targets for searching for high-redshift protoclusters, as confirmed by the recent discoveries of over-densities of \\Lya~ emitters around such objects (see e.g. Kurk et al.~2000; Pentericci et al.~2000; Venemans et al.~2002).  One of the most prominent characteristics of HzRGs are their extended emission line regions (EELRs), which are luminous ($>10^{37}$~W in \\Lya), tens to several hundred kpc in size (often aligned with the radio axis) and kinematically active (FWHM~$\\gtrsim 1000$\\kmps). Their dominant ionisation mechanism seems to change from shocks to photoionisation by the central engine as the radio source expands beyond the confines of the host galaxy (Best et al.~2000; De Breuck et al.~2000; Jarvis et al. 2001a). Likewise the kinematics of the gas may reflect the competing influences of gravity, and energy input via shocks from the radio source and starburst-driven superwinds. However, there are many unresolved issues which remain concerning the structure, origin and fate of the emission line gas, which may hold important clues to many aspects of massive galaxy formation. Did the gas originate in a cooling flow or a merger, or was it expelled from the central galaxy during a violent starburst? Is it in the form of cloudlets, filaments or expanding shells of material, and what is their composition? What is the ultimate fate of the gas, will it form stars or escape from the galaxy to enrich the surrounding intracluster and intergalactic media?  A new perspective on many of these issues was opened up with the discovery by R\\\"{o}ttgering et al.~(1995) and van Ojik et al.~(1997; hereafter vO97) that most of the smaller high redshift radio sources (those with projected linear size $D <50$\\kpc) exhibit spatially resolved \\HI~ absorption in their \\Lya~ emission-line profiles, with column densities in the range $10^{18}$--$10^{19.5}$\\pcmsq. Binette et al.~(2000; hereafter B00) found, for the $z=2.92$ radio galaxy 0943$-$242, \\civ$\\lambda\\lambda$1548,1551 absorption lines superimposed on the \\civ~ emission, at the same redshift as the main \\Lya~ absorption system. They could not reconcile the observed \\civ/\\Lya~ emission- and absorption-line ratios with a model in which the absorption- and emission-line gas are co-spatial. Instead they proposed that the absorbing gas is of lower metallicity ($Z \\sim 0.01 Z_{\\rm{\\sun}}$) and located further away from the host galaxy than the emission line gas, beyond the high pressure radio source cocoon. This material, they claimed, is thus a relic reservoir of low metallicity, low density ($\\sim 10^{-2.5}$\\pcm) gas, similar to that from which the parent galaxy may have formed.  This shell-like distribution of the absorbing gas was also invoked by Jarvis et al. (2001a) and De Breuck et al. (2000) to explain various correlations found in a complete sample of high-redshift ($z > 2$) radio galaxies.  In this paper we probe the H{\\footnotesize{I}} and \\civ~ absorbers in greater detail using the UVES spectrograph on the VLT to obtain echelle spectra of two high-redshift radio galaxies, the aforementioned 0943-242 and 0200+015 at z=2.23. The new observations offer a factor of 10 improvement in spectral resolution over previous investigations and thus have the potential (i) to reveal whether the main absorbers are genuinely in the form of a single component with ordered global kinematics, or whether they fragment into a number of weaker absorbers; (ii) to probe the emission-line kinematics with high velocity resolution.  In section~\\ref{sec:obs} we describe the observations and data reduction of our VLT-UVES spectroscopic observations and in section~\\ref{sec:analysis} we briefly discuss the emission- and absorption-line fitting procedure. In section~\\ref{sec:0943} we use our new high-resolution spectra to investigate the absorption halo of the $z = 2.92$ radio galaxy 0943-242. Section~\\ref{sec:0200} is a detailed discussion of the absorbing halo around the $z = 2.23$ radio galaxy 0200+015 and we also investigate the kinematics of the narrow-emission lines via the unabsorbed HeII emission line and briefly discuss the effect of emission-line asymmetry on our absorption line fitting. The origin and fate of the absorbing halos is discussed in section~\\ref{sec:discussion} and we provide a summary of our conclusions in section~\\ref{sec:conc}.  All physical distances are calculated assuming $H_{\\rm{0}}=70$\\kmpspMpc, $\\Omega_{\\rm{M}}=0.3$ and $\\Omega_{\\rm{\\Lambda}}=0.7$. ",
        "conclusions": "\\label{sec:0943}  Fig.~\\ref{fig:Lya0943} shows the 1-d spectrum of the \\Lya~ and \\civ~emission lines in 0943-242, extracted over the central 5~arcsec of the slit which covers all of the emission. The fit, with a series of Voigt profile absorbers superimposed upon a Gaussian emission envelope is shown, and the parameters tabulated in Table~\\ref{tab:Lya0943}. The errors were calculated by assuming a $\\chi^{2}$ distribution of $\\Delta \\chi^{2}$ ($\\equiv \\chi^{2} - \\chi^{2}_{\\rm min}$) (Lampton, Margon \\& Bowyer 1976). The error on each parameter was calculated by setting $\\Delta \\chi^{2} = 1$ and allowing the other parameters to float. The errors quoted are $1\\sigma$.  The most notable feature of this spectrum is that the main absorber remains as a single system of column density $\\sim 10^{19}$\\psqcm, and is completely black at its base with no evidence for substructure. No new absorption systems are identified in addition to those found by R\\\"{o}ttgering et al.~(1995) at a factor of $\\sim 10$ lower resolution.  This provides strong evidence that these absorption systems are physically distinct from the extended narrow-emission-line region. If the absorption was caused by gas which is mixed with the emission-line clouds, as postulated by vO97 then we would expect a series of narrow-absorption troughs at various wavelengths across the \\Lya~ emission profile. The fact that we see one main absorber, with no detectable substructure, shifted blueward of the emission-line peak provides compelling evidence that the absorption gas encompasses the whole of the emission-line regions with a covering factor of unity along our line-of-sight. This is in line with the argument proposed by B00 who fitted the emission- and absorption-line profile of the \\civ$\\lambda\\lambda$1548,1551 doublet in 0943-242 to determine the metallicity of the absorption systems compared to the emission-line gas. Their results show that the emission and absorption line ratios of \\civ~ and \\Lya~ are incompatible with photoionisation or collisional ionisation of cloudlets with uniform properties and the possibility that the absorption and emission phases are co-spatial is rejected. A different model was preferred in which the absorption gas has low metallicity and is located further away from the host galaxy than the emission line gas.  \\begin{table} \\begin{tabular}{|ccll|} \\hline Absorber  & $z$ & $b$ & log $N(\\rm{HI})$ \\\\ & & (\\kmps) & (\\psqcm) \\\\  \\hline 1  & 2.9066 $\\pm$ 0.0062 & 88 $\\pm$ 45  & 14.02 $\\pm$ 0.30 \\\\ 2  & 2.9185 $\\pm$ 0.0001 & 58 $\\pm$ 3  & 19.08 $\\pm$ 0.06 \\\\ 3  & 2.9261 $\\pm$ 0.0005 & 109 $\\pm$ 35 & 13.55 $\\pm$ 0.16 \\\\ 4  & 2.9324 $\\pm$ 0.0001 & 23 $\\pm$ 17 & 13.35 $\\pm$ 0.30 \\\\ \\hline &  &  & log $N(\\rm{CIV})$ \\\\ & & & (\\psqcm) \\\\ \\hline 2  & 2.9192 $\\pm$ 0.0001 & 119 $\\pm$ 2 & 14.58 $\\pm$ $\\pm$ 0.04 \\\\ \\hline \\end{tabular} \\caption{\\label{tab:Lya0943} Parameters of the Voigt absorption profile fits for the main component of 0943-242. The final row corresponds to the fit of one absorber to the CIV profile.} \\end{table}  \\begin{figure} \\includegraphics[width=0.48\\textwidth,angle=0]{0943Lya_mjj2.ps}  \\includegraphics[width=0.48\\textwidth,angle=0]{0943CIV_mjj4.ps} {\\caption{\\label{fig:Lya0943}\\normalsize ({\\it top}) The \\Lya~profile of 0943-242, with the absorption model overlaid. ({\\it bottom}) The CIV profile of 0943-242, with the absorption model overlaid.  }} \\end{figure}  \\label{sec:conc} We have obtained high-resolution spectra of two high-redshift radio galaxies to determine the structure of the absorption gas which appears to be ubiquitous in small ($D < 50$~kpc) radio sources. The main conclusions of this study can be summarized as follows: \\begin{itemize}  \\item  The higher resolution observations do not uncover significant structure in the absorbers of both Ly$\\alpha$ and C{\\footnotesize{IV}} in 0943-242, however the $N_{\\rm HI} \\sim 10^{19}$~cm$^{-2}$ absorber in 0200+015 split into two absorber of $N_{\\rm HI} \\sim 4 \\times 10^{14}$~cm$^{-2}$. The absorption gas present in our two chosen high-redshift radio sources is most likely distributed in shell-like structures which encompass the radio source.  \\item The metallicity of these absorption shells can vary between sources and also between the shells in the same sources, implying that the shells are enriched by a secondary process.  \\item We speculate that the most plausible origin of the absorbing shells is two fold. The pristine shells are the relics from a gas-rich merger or (proto-)cluster gas which has cooled and collapsed towards the centre of the dark matter halo. Whereas the enrichment comes at a later stage when the AGN is triggered simultaneously with a major episode of star formation, presumably due to a large inflow of gas into the nucleus. This major burst of star formation initiates starburst driven superwinds which drive the gas toward the outer halo of the galaxy and may explain the very high metallicity of the absorption shells in 0200+015.   \\item The passage of the radio emitting plasma disrupts and fragments the absorption shells in the larger, and thus older radio sources. Evidence of this is seen in 0200+015, where the absorption screen along our line-of-sight to the radio lobe appears to be more fragmented than the absorbing screen between us and the nuclear emission.  \\item We have found direct evidence that the narrow-emission lines in high-redshift radio sources may be outflowing from the central regions. This is in agreement with previous studies of the narrow-emission line dynamics in nearby Seyfert galaxies and is in line with orientation based Unified Schemes in which the outflow from the far side of the source is obscured by dust.   \\end{itemize}"
    },
    "0209/astro-ph0209473_arXiv.txt": {
        "abstract": "{On the 13th of May 2002, \\object{supernova 2002cv} was discovered using a near-infrared camera working at the AZT-24 1.1m telescope at Campo Imperatore (AQ-Italy). After the infrared detection a simultaneous photometric follow-up was started at optical wavelengths. The preliminary results confirm a heavily obscured object with a $V-K$ color not lower than 6 magnitudes, making \\object{SN 2002cv} the most reddened supernova ever observed. This finding, along with the recent discovery of another obscured supernova, suggests a critical revision of the rates known to date. The estimate of the visual extinction and the light curves are provided here. These latter indicate that our SN 2002cv observations are the earliest available for a type-Ia supernova at IR wavelengths. ",
        "introduction": "The AZT-24 telescope of the Campo Imperatore Observatory \\footnote{see {\\it http://www.mporzio.astro.it /cimperatore/WWW/}} (a cooperation among Rome, Teramo and Pulkovo Observatories) is mainly used for photometric studies of variable sources at near infrared (NIR) wavelengths. During the follow-up of the \\object{supernova 2002bo} (Cacella et al. \\cite{iauc7847}), discovered on 2002 March 9 in the spiral galaxy \\object{NGC3190}, a new source appeared $\\sim28\\arcsec$ to the North and $\\sim10\\arcsec$ to the West of the galactic nucleus (Larionov et al. \\cite{iauc7901}) at $10^h18^m03\\fs68$ and $+21^\\circ50\\arcmin06\\farcs2$ (J2000). Our discovery images of this supernova, called \\object{SN 2002cv}, are presented in Fig.~\\ref{FigKV}.  \\begin{figure*}[tbh] \\centering \\psfig{figure=Ef172_f1.eps, clip=} \\caption{The $K$-band images obtained with AZT-24 telescope before (a) and after (b) \\object{SN 2002cv} discovery. Panel (c) contains the $V$-band image obtained with the Schmidt telescope on May 15th (after the outburst) with no detectable object at the SN location.} \\label{FigKV} \\end{figure*}  A more detailed analysis of the images obtained during the days preceding May 13, that is the discovery date, has shown that the outburst became visible between May 6 and May 9, though at the limit of detectivity, thus preventing us from a prompt detection of the new supernova. The maximum in our NIR light curves was observed between the 20th and the 22nd of May ($J=14.77\\pm 0.05$ on $MJD 52414.5\\pm 0.5$, $K=13.92\\pm 0.07$ on $MJD 52416.8\\pm 0.5$).  As soon as the object was discovered, a simultaneous follow-up was started at optical wavelengths using the Schmidt telescope of the Campo Imperatore Observatory and the TNT telescope of the Teramo Observatory.  Since our first discovery IAU Circular and the preparation of this text, 4 additional IAU Circulars were published about \\object{SN 2002cv}. In the first of them, Li (\\cite{iauc7903}) reports the failed localization of \\object{SN 2002cv} using the KAIT telescope\\footnote{see {\\it http://astro.berkeley.edu/$\\sim$bait/kait.html}} that observed \\object{NGC3190} at optical wavelengths ($B$, $V$, $R$ and $I$) both before and after the discovery. In the second Meikle and Mattila (\\cite{iauc7911}) provide a preliminary classification of the supernova as Type-Ia by means of the UKIRT telescope NIR spectroscopy; this classification is confirmed in the fourth Circular by Filippenko et al. (\\cite{iauc7917}). ",
        "conclusions": "\\begin{enumerate} \\item We present preliminary light curves of the \\object{supernova 2002cv} we have discovered in \\object{NGC3190} galaxy. \\item The supernova outburst appeared above magnitude $J=18$ between the 6th and 9th of May, and reached its maximum on May 20, making \\object{2002cv} the earliest supernova Ia observed at NIR wavelengths. \\item The color index $V-K\\ga 6$ magnitudes makes this supernova the most reddened ever observed. \\end{enumerate}"
    },
    "0209/astro-ph0209190_arXiv.txt": {
        "abstract": "Abell 2029 is one of the most studied clusters due to its proximity (z=0.07), its strong X-ray brightness and its giant cD galaxy which is one of the biggest stellar aggregates we know. We present here the first weak lensing mass reconstruction of this cluster made from a deep I-band image of $28.5'\\times 28.5'$ centered on the cluster cD galaxy. This preliminary result allows us already to show the shape similarities between the cD galaxy and the cluster itself, suggesting that they form actually a single structure.\\\\ We find a lower estimate of the total mass of $1.8\\times10^{14} h^{-1}\\;M_\\odot$ within a radius of $0.3\\;h^{-1}$Mpc.  We finally compute the mass-to-cD-light ratio and its evolution as a function of scale. ",
        "introduction": "The low redshift of this cluster is welcome for X-ray and optical analyses; however the geometry of a very close lens and very distant sources is in general a disadvantage for the lensing efficiency. Therefore, a noisy mass reconstruction has to be expected in our study compared to more distant clusters.  Furthermore, the cluster galaxies in low redshift clusters have a large angular extend so that the number density of background galaxies close to the cluster center is small.  On the other hand, the mass computed for low redshift lenses is not very sensitive to the exact source redshifts of the background galaxies used for the reconstruction.  \\begin{figure} \\begin{center} \\plottwo{figures/kappa_xray.eps}{figures/centre.eps} \\caption{Left panel : projected-mass map from weak lensing ($\\kappa=0.12$, 0.15, 0.2 and 0.3 ; in white) superposed on X-ray contours from ROSAT (in black). The scale of this cutout is 25 arcmin on each side. Right panel : close up near the center of the cD in the I-band image. We show the innermost 1.5 arcmin of the mosaic.} \\vspace{-0.5cm} \\end{center} \\label{mass_map} \\end{figure} Our data were obtained in the I-band under subarcsec seeing conditions with the UH8K mosaic camera at the Canada-France-Hawaii telescope. The UH8K consists of eight 2k$\\times$4k chips arranged in a mosaic of two rows each containing 4 CCDs. The total field-of-view is 28$'\\times$28$'$ with a pixel scale of 0.206$''$. For our analysis we did not use the upper right chip of the mosaic suffering from charge transfer efficiency problems.  The total integration time for A2029 was 9600 sec. consisting of 8 individual exposures taken with a dither pattern spanning about 15 arcsec to allow us to fill the gaps between individual CCDs.  The exact details of the data processing will be published elsewhere.  From the final coadded image we extract an object catalogue with positions, sizes, ellipticities and magnitude.  Since we do not have any color information we select the background galaxies by comparing the magnitude and size distributions of the galaxies in the center of the image (at the cluster-center location) and at the borders. The overdensity in the former distribution indicates in a statistical sense the magnitude and size of the cluster members.  We then define our background galaxy catalogue by considering the corresponding smaller and fainter galaxies. As a result of this procedure, a number of background galaxies will not be included in our sample and a number of foreground galaxies might be included as background ones. This will not affect the shape of the mass contours but will lower the amplitude of the measured gravitational shear and allow us to estimate only a lower bound of the cluster mass.  The final catalogue contains $9\\;314$ background galaxies ($\\sim$15/arcmin$^2$).  To obtain final estimates of the galaxy shapes, the raw galaxy shapes were corrected for PSF effects with the method described by Kaiser, et al. (1995) and moficiations described in Hoekstra et al. (1998) and Erben et al (2001).  This correction was applied separately on each CCD and the density of approximately one hundred stars per chip allowed an accurate PSF-correction mapping.    \\subsection{Mass distribution}   The catalogue allows us to estimate a smoothed shear map from that we can perform a weak lensing mass reconstruction. The left panel of Fig. 1 shows the resulting $\\kappa$-map, i.e. the projected mass map, from the shear smoothed with a Gaussian filter with $\\sigma=68\\arcsec$.  The map clearly shows the presence of the cluster as well as several additional peaks of lower amplitude.  In order to quantify the significance of each peak in our field we used the $M_{\\rm ap}$ statistics introduced by Schneider (1996).  With a filter scale of 4.12 arcmin, the cluster shows up as a 6.5-$\\sigma$ detection. Besides the main cluster, we find the secondary peaks to be likely noise features given their detection levels (Van Waerbeke 1999).  The mass reconstruction clearly shows that the center of the mass coincides with the center of the cD galaxy and that the mass is elongated in the same direction as the light.  We clearly see the strong correlation between the cD light and the mass distribution at the cluster center.  Next, we have computed the mean tangential shear around the cluster center in independent circular bins (see Fig. 2).  Hereby we chose as center the position of the peak in the $M_{\\rm ap}$ statistics with the smoothing scale of 4.12 arcmin where the cluster is detected at the highest significance. The massive cluster clearly causes a significant shear signal up to the borders of our data field.  The error bars are estimated from the dispersion of the cross shear component that is not influenced by gravitational lensing.  The profile is well fitted by an isothermal sphere model giving a best fit velocity dispersion of 842 km s$^{-1}$.  As we have seen before, this value has to be interpreted as a lower limit of the real one.  The X-ray contours that are also shown in Fig. 1 are centered on the cD galaxy and show the same orientation as the mass and the light distribution.  The significant X-ray emission of the cluster goes even beyond the field of our optical image.  The similar distributions of the cD light, the X-ray gas and the total mass strongly suggest that the cD is not an isolated object, but an aggregate of stars orbiting in the cluster potential.  \\begin{figure} \\label{shear_kappa} \\plottwo{figures/tang.eps}{figures/integrated_mass.eps} % \\caption{Left panel : shear profile measured in independent circular bins and fitted by an isothermal sphere profile. Right panel : projected-mass measured in ellipses of 1:2 axis ratio and aligned with the cD light.} \\end{figure}  The quantitative estimation of the mass requires breaking the so-called mass-sheet degeneracy by knowing or assuming the value of the mass at a given angular position. In order to do so, we assume no mass at the borders of the field, i.e. far from the cluster center and force the average $\\kappa$-value at these locations to be zero. However, since the X-ray map and the shear profile still indicate the presence of some mass at a distance of $17\\arcmin$ from the cluster center, this further constrains our mass estimate to be a lower limit only.  Using an elliptical bin with the same orientation and shape as the cD light, we find a mass of $1.8\\times10^{14} h^{-1}\\;M_\\odot$ within a semi-minor axis of $0.3\\;h^{-1}$Mpc. ",
        "conclusions": "\\label{Conclusion}  We have presented the first weak lensing analysis of Abell 2029.  The very low redshift of the cluster imposes a low lensing efficiency and therefore a noisy mass reconstruction compared to more distant clusters.  The available data allowed only lower mass estimates of the cluster since we can not properly select the background galaxies. However we can compute the shape of the mass contours of the cluster. They indicate that the cD galaxy lies in the center of the gravitational potential of the cluster and we find similar shape and orientation for the distribution of the mass and cD light.  We have seen that despite the large size of our field ($28.5' \\times 28.5'$) the mass can be traced up out to the border of the CCD which prevents us to break the mass-sheet degeneracy.  Using the light distribution of the cD galaxy studied in detail by Uson et al. (1991), we have computed the profile of the mass-to-cD-light ratio, which reaches a value of 300 M$_\\odot/$L$_\\odot$ in R-band at 4 arcminutes.  These preliminary results of a more detailed future study can already be used for comparing estimates made with other techniques.  Next we plan to study this interesting cluster by using deeper images in two colours with Megacam@CFHT.  Color information will allow us to obtain a fair separation of the cluster galaxies and therefore a non biased estimate of the cluster mass.  It will then be possible to compare the light and mass distributions in great detail.  With a deep, 6 hour I-band observation and the expected high number of background galaxies we will be able to investigate the cluster mass out to larger scales and thus allow a detailed comparison with existing X-ray maps and a properly break the mass sheet degeneracy.  This work was supported by the TMR Network ``Gravitational Lensing: New Constraints on Cosmology and the Distribution of Dark Matter'' of the EC under contract No. ERBFMRX-CT97-0172, and by the Deutsche Forschungsgemeinschaft under the project SCHN 342/3--1.     \\small"
    },
    "0209/astro-ph0209315_arXiv.txt": {
        "abstract": "We have used the Space Telescope Imaging Spectrograph (STIS) on the Hubble Space Telescope (HST) to obtain high spatial resolution spectroscopy of the central region of the dense globular cluster M15. The observational strategy and data reduction were described in Paper~I (van der Marel \\etal 2002). Here we analyze the extracted spectra with a cross-correlation technique to determine the line-of-sight velocities of individual stars. Our final STIS velocity sample contains 64 stars, two-thirds of which have their velocity measured for the first time. The new data set triples the number of stars with measured velocities in the central projected $R \\leq 1''$ of M15 and doubles the number in the central $R \\leq 2''$. We combine our data with existing ground-based data to obtain non-parametric estimates of the radial profiles of the projected rotation velocity, velocity dispersion, and RMS velocity $\\sigma_{\\rm RMS}$. The results differ from earlier work in the central few arcsec in that we find that $\\sigma_{\\rm RMS}$ rises to $\\sim 14 \\kms$, somewhat higher than the values of $10$--$12 \\kms$ inferred previously from ground-based data.  To interpret the results we construct dynamical models based on the Jeans equation for a spherical system. If the velocity distribution is isotropic, then M15 must have a central concentration of non-luminous material. If this is due to a single black hole, then a fit to the full velocity information as function of radius implies that its mass is $M_{\\rm BH} = (3.9 \\pm 2.2) \\times 10^3 \\Msun$. The existence of intermediate-mass black holes in globular clusters is consistent with several scenarios for globular cluster evolution proposed in the literature. The inferred mass for M15 is consistent with the extrapolation of the relation between $M_{\\rm BH}$ and $\\sigma_{\\rm RMS}$ that has been established for galaxies. Therefore, these results may have important implications for our understanding of the evolution of globular clusters, the growth of black holes, the connection between globular cluster and galaxy formation, and the nature of the recently discovered `ultra-luminous' X-ray sources in nearby galaxies. Instead of a single intermediate-mass black hole, M15 could have a central concentration of dark remnants (e.g., neutron stars) due to mass segregation. However, we argue that the best-fitting Fokker-Planck models that have previously been constructed for M15 do not predict a central mass concentration that is sufficient to explain the observed kinematics. To fit the M15 data without any central dark mass concentration one must assume that the velocity distribution is significantly radially anisotropic near the center, which contradicts predictions from both Fokker-Planck and $N$-body calculations. ",
        "introduction": "\\label{s:intro}  The globular cluster M15 (NGC 7078) has one of the highest central densities of any globular cluster in our Galaxy. As a result, it has been one of the globular clusters for which the structure and dynamics have been most intensively studied in the past decade (as reviewed in van der Marel 2001). The present paper is the second in a series of two in which we present the results of a study with the Hubble Space Telescope (HST) of the line-of-sight velocities of stars in the central few arcsec of M15. Paper~I (van der Marel \\etal 2002) discussed the observations, and the extraction and calibration of the stellar spectra. In the present paper we determine the stellar line-of-sight velocities from the spectra, and we use the results to study the dynamics and structure of M15.  M15 is a proto-typical core-collapsed cluster (Djorgovski \\& King 1986; Lugger \\etal 1987; Trager, King \\& Djorgovski 1995), with a stellar surface density profile that rises all the way into the center. Such clusters make up $\\sim\\!  20$\\% of all globular clusters in our Galaxy, and stand in marked contrast to King-model clusters, which show flat central cores and are modeled as tidally-truncated isothermal systems. Even imaging studies with HST have not provided any evidence for a homogeneous core in M15 (despite early claims to the contrary; Lauer \\etal 1991). Guhathakurta \\etal (1996) used the Second Wide Field and Planetary Camera (WFPC2) and found the projected surface number density profile inside $6''$ ($0.34$ pc) to be consistent with a power law $N(R) \\propto R^{-0.82 \\pm 0.12}$. Sosin \\& King (1997) used the Faint Object Camera (FOC) and obtained $N(R) \\propto R^{-0.70 \\pm 0.05}$ for turnoff stars. They also showed that the distributions for stars of different masses have slightly different power-law slopes, which is qualitatively consistent with the mass segregation predicted in a cluster in which two-body relaxation has been important.  Bahcall \\& Wolf (1976, 1977) constructed detailed models for the equilibrium stellar density distribution of a globular cluster in which a central black hole (BH) has been present for much longer than the two-body relaxation time. For a cluster of equal-mass stars one expects $N(R) \\propto R^{-3/4}$, in surprisingly good agreement with the observed star count profile for M15. While BHs have been convincingly detected in the centers of galaxies (e.g., Kormendy \\& Gebhardt 2001), no convincing detections exist for globular clusters. On the other hand, few, if any, previous studies have had sufficient sensitivity to unambiguously detect BHs in globular clusters with masses $M_{\\rm BH} \\lta 5 \\times 10^3 \\Msun$. There are many ways in which globular cluster evolution at high densities can lead to the formation of a massive BH in the center (Rees 1984). For example, core collapse induced by two-body relaxation may lead to sufficiently high densities for individual stars or stellar-mass black holes to interact or collide, with a single massive BH as the likely end product (Sanders 1970; Quinlan \\& Shapiro 1987, 1990; Lee 1987, 1993, 1995). Studies of such scenarios have gained much interest lately (Miller \\& Hamilton 2002; Mouri \\& Taniguchi 2002; Portegies Zwart \\& McMillan 2002) after the discovery of intermediate luminosity X-ray objects in external galaxies (e.g., Colbert \\& Mushotzky 1999). The emission of these objects may be due to accretion onto intermediate mass BHs. However, this interpretation is not uniquely implied by the data and there is no unique association of these objects with star clusters (e.g., Zezas \\& Fabbiano 2002).  While the observed star count profile of M15 is consistent with the presence of a BH, it can be explained equally well as a result of core-collapse (Grabhorn \\etal 1992). Hence, the star count profile by itself yields only limited insight. An additional problem is that photometric studies cannot determine whether light follows mass, and what the abundance and distribution of dark remnants are. Kinematical studies are therefore essential to gain further insight. Integrated light measurements of M15 initially suggested a very high central dispersion, $\\sigma = 25 \\pm 7 \\kms$. This was a sharp increase from the dispersions of 5--$15 \\kms$ found at larger radii from the radial velocities of individual stars, which was interpreted as evidence for the presence of a $10^3 \\Msun$ central BH (Peterson, Seitzer \\& Cudworth 1989). This pioneering work spurred a lot of interest in globular cluster dynamics. However, all subsequent studies were unable to confirm the high central velocity dispersion measurement. In particular, it became clear that that the weighting of stars by their brightness in an integrated-light spectrum produces a large systematic uncertainty in the velocity dispersion deduced from the broadening of lines (Zaggia, Cappaccioli, \\& Piotto 1993; Dubath \\etal 1994). Velocity measurements of individual stars are therefore called for.  Line-of-sight velocities are now known from ground-based studies for $\\sim\\! 1800$ M15 stars, as compiled by Gebhardt \\etal (2000a). Many different studies contributed to this dataset, as reviewed in the introduction of Paper~I.  The projected velocity dispersion profile increases monotonically inwards from $\\sigma = 3 \\pm 1 \\kms$ at $R=7$ arcmin (Drukier \\etal 1998), to $\\sigma = 11 \\pm 1 \\kms$ at $R=24''$. The analysis of Gebhardt \\etal (2000a) suggested that the velocity dispersion is approximately constant at smaller radii, and is $\\sigma = 11.7 \\pm 2.8 \\kms$ at the innermost available radius $R \\approx 1''$. Outside of the very center, the velocity dispersion profile is well fitted by a spherical dynamical model with an isotropic velocity distribution and a constant mass-to-light ratio $\\Upsilon = 1.7$ (in solar V-band units). However, this model underpredicts the velocity dispersion in the central $2$ arcsec. The fit can be improved by addition of a central black hole, which causes the velocity dispersion to rise in Keplerian fashion as $\\sigma \\propto R^{-1/2}$ towards the center of the cluster. The best fit was obtained with a mass $M_{\\rm BH} \\approx 2 \\times 10^3 \\Msun$ (Gebhardt \\etal 2000a). However, the ground-based M15 velocity dispersion data can be fitted equally well with a model in which the mass-to-light ratio $\\Upsilon(r)$ of the stellar population increases inwards to a value of $\\sim 3$ in the center. This would not {\\it a priori} be implausible, since mass segregation would tend to concentrate heavy dark remnants to the center of the cluster. Models with an anisotropic velocity distribution may even be able to fit the data with constant mass-to-light ratio and without a central black hole. Higher spatial resolution data and more detailed modeling are necessary to decide amongst these scenarios; this is the focus of the present series of papers.  It has been known for some time (Gebhardt \\etal 1994) that M15 has a net projected rotation amplitude of $V_{\\rm rot} \\approx 2 \\kms$ at radii comparable to the half-light radius (about 1~arcmin).  More recent work (Gebhardt \\etal 2000a; these results were also suggested by the integrated-light measurements of Peterson 1993) has revealed that the rotation amplitude is larger at small radii: $V_{\\rm rot} = 10.4 \\pm 2.7 \\kms$ for $R \\leq 3.4''$, implying that $V_{\\rm rot}/\\sigma \\approx 1$ in this region. This large amplitude is surprising because two-body relaxation should rapidly transfer net angular momentum outward from such small radii (see the discussion in Gebhardt \\etal 2000a).  Even more surprising is that the position angle of the projected rotation axis at small radii is $\\sim 100^{\\circ}$ different from that near the half-light radius.  Although the large increase in the rotation amplitude at small radii may have something to do with the presence of a central BH (Gebhardt \\etal 2000a), the increase and change in position angle are not predicted by any current theory of globular cluster dynamical evolution.  Phinney (1993) used an alternative argument to constrain the mass distribution of M15. There are two millisecond pulsars in M15 at a distance $R = 1.1''$ from the cluster center that have a negative period derivative ${\\dot P}$.  This must be due to acceleration by the mean gravitational field of the cluster, since the pulsars are expected to be spinning down intrinsically (positive ${\\dot P}$). The observed ${\\dot P}$ values place a strict lower limit on the mass enclosed within a projected radius of $R = 1.1''$. Combined with the observed light profile this implies that the mass-to-light ratio must increase centrally inwards. A similar pulsar acceleration study was recently performed by d'Amico \\etal (2002) for the cluster NGC 6752, which suggests a central increase in mass-to-light ratio in this cluster as well. For M15, Phinney (1993) obtained $\\Upsilon > 2.1$ for the total mass-to-light ratio within $R \\leq 1.1''$, with a statistically most likely value of $\\Upsilon \\approx 3.0$. These results are consistent with the analysis of stellar kinematics (see also Dull \\etal 1997). Unfortunately, the pulsar data, like the ground-based observations of the kinematics, does not constrain the distribution of mass tightly enough to discriminate between the effects of mass segregation and a central BH.  Tighter constraints on the distribution of mass near the center of M15 need observations of the kinematics with better angular resolution than previous studies and observations from space can supply these. So we started a project to use HST to determine more stellar velocities close to the center of M15 (HST program GO-8262, PI: van der Marel). As described in Paper I, we used the Space Telescope Imaging Spectrograph (STIS) to obtain observations with the $0.1''$-wide slit at 18 adjacent positions near the cluster center. All spectra cover the wavelength range from 5073--5359{\\AA}, which includes the Mg b triplet at $\\sim 5175${\\AA}. The resolution is $0.276${\\AA} per pixel, which corresponds to $15.86 \\kms$. Extensive reductions and calibrations were performed to extract spectra with signal-to-noise ratio $S/N > 5.5$ per pixel for a total of 131 stars. The velocity calibration of the spectra was the most crucial and difficult aspect of the data reduction. Corrections were necessary for: drifts in the wavelength scale during an orbit; changes in the velocity of HST as it orbits the Earth; and wavelength shifts induced by the offsets of stars from the center of the slit. The analyses in Paper~I indicate that the uncertainty in velocity scale caused by residual calibration errors in the final spectra is $\\sim 2.5 \\kms$.  Here we analyze the 131 stellar spectra from Paper~I, and we show that for 64 of them the quality is sufficient to obtain an accurate line-of-sight velocity measurement. We use the results to obtain new constraints on the dynamics and structure of M15. The paper is organized as follows. In Section~\\ref{s:extraction} we describe the cross-correlation algorithm that we have used for the extraction of line-of-sight velocities, including the choice of spectral templates. We discuss the reliability of the results based on an analysis of our STIS observations of a calibration star. In Section~\\ref{s:kinresults} we describe the application of the cross-correlation algorithm to the STIS spectra of M15.  We describe how we have corrected the inferred velocities for the effects of crowding and blending. The reliability of the inferred velocities is verified by comparison to ground-based data, for those stars for which the latter are available. In Section~\\ref{s:profiles} we infer the velocity dispersion and rotation velocity profiles of M15 from the combined HST and ground-based line-of-sight velocity samples. In Section~\\ref{s:dynamics} we present dynamical models to interpret the results, and we discuss the implications for the dynamical structure and mass distribution of M15. Section~\\ref{s:conc} discusses and summarizes the main conclusions. ",
        "conclusions": "\\label{s:conc}  We have obtained high spatial resolution spectroscopy of the central region of the globular cluster M15 with the STIS spectrograph on board HST. The observational setup, calibration and spectral extraction were discussed in Paper~I. Here we have analyzed the spectra with a cross-correlation technique to determine the line-of-sight velocities of individual stars. Our final STIS velocity sample contains 64 stars. Two-thirds of the stars in this sample have their velocity measured for the first time. Half of the stars reside within a projected radius $R = 2.4''$ from the center of M15. The new data set triples the number of stars with measured velocities in the central $R \\leq 1''$ of M15 and doubles the number in the central $R \\leq 2''$. Our analysis includes the necessary (small) corrections for the effects of blending with neighboring stars. Detailed tests on a calibration star and comparison to ground-based M15 data demonstrate that our velocities are accurate and trustworthy. We combined the STIS results with existing ground-based data to obtain a total sample of 1797 stars in M15 with known line-of-sight velocities. We use the combined sample to determine the radial profiles of the most important projected kinematical quantities: the rotation velocity $V_{\\rm rot}$; the position angle of the kinematical major axis, ${\\rm PA}_{\\rm kin}$; the velocity dispersion, $\\sigma$; and the RMS velocity averaged over rings on the projected plane of the sky, $\\sigma_{\\rm RMS}$. Our results differ from earlier work only in the central few arcsec. In particular, we find that $\\sigma_{\\rm RMS}$ rises to $\\sim 14 \\kms$ at the innermost radii. This is somewhat higher than the value of $10$--$12 \\kms$ inferred previously from ground-based data (Dull \\etal 1997; Gebhardt \\etal 2000a).  To interpret the results we constructed dynamical models based on the Jeans equation for a spherical system. We compared the model predictions to the data using a maximum-likelihood technique to obtain the best-fitting model parameters and their confidence regions. If the velocity distribution is isotropic, then M15 must have a central concentration of non-luminous material. This could be due to an intermediate-mass black hole. If one were reluctant to invoke such an object, then one alternative may be that M15 has a central collection of dark remnants (neutron stars and/or stellar mass black holes). This arises naturally in a globular cluster due to the mass segregation that occurs as two-body relaxation drives the system to equipartition. However, we argued that the best-fitting Fokker-Planck models that have previously been constructed for M15 (Dull \\etal 1997) do not predict a large enough concentration of dark remnants to fit the data. It remains to be seen whether alternative Fokker-Planck models can be constructed that generate a more massive concentration of dark remnants from plausible initial conditions. It is useful to note in this context that an important uncertainty in Fokker-Planck models comes from the assumptions used for the retention factor of neutron stars. Dull \\etal assumed that all neutron stars are retained, in contrast with most recent work which predicts that only $\\lta 10$\\% will be retained. So Dull \\etal may actually have overestimated the central concentration of dark remnants in their models. Another alternative scenario is to assume that deviations from isotropy in the velocity distribution may be responsible for the observed kinematics. However, to fit the kinematical data without any mass concentration one must assume that the velocity distribution is radially anisotropic near the center, $\\beta_0 = 0.65 \\pm 0.2$. This contradicts the predictions of both Fokker-Planck models and $N$-body calculations, which suggest that the velocity distribution in the central region of a globular clusters remains close to isotropic at all times during the cluster evolution.  In view of the results that we have presented, the presence of an intermediate mass black hole in M15 appears to be the most plausible explanation of the data. As noted in Section~\\ref{s:intro}, there are several mechanisms by which such a black hole could plausibly have formed. For the best-fit black hole mass we adopt the average of the values which were inferred in Sections~\\ref{ss:isoconstant} and~\\ref{ss:isovarying} using a constant mass-to-light ratio and the Dull \\etal (1997) mass-to-light ratio profile, respectively. This yields: $M_{\\rm BH} = (3.9 \\pm 2.2) \\times 10^3 \\Msun$. This mass is consistent with the constraints on the central mass distribution of M15 implied by observations of pulsar accelerations (Phinney 1993). The black hole mass inferred for M15 matches remarkably well with the understanding that has been developed for the presence of black holes in the centers of galaxies. For these black holes, there is a strong correlation between the black hole mass and the velocity dispersion of the bulge component (Gebhardt \\etal 2000b; Ferrarese \\& Merritt 2000). Figure~\\ref{f:bhcorr} shows the available data points and the best fit from the recent compilation of Tremaine \\etal (2002). For M15, the luminosity weighted mean velocity dispersion within the half light radius ($1.06$ arcmin, Harris 1996) is $12.1 \\kms$ (this quantity was defined and calculated similarly as in Gebhardt \\etal 2000b). At this dispersion, the estimated black hole mass fits perfectly on the extrapolation of the relation established for galaxies (Figure~\\ref{f:bhcorr}).  Interestingly, a study of the globular cluster G1 in the Andromeda galaxy, performed simultaneously with the present study, has also provided evidence for a central black hole (Gebhardt, Rich \\& Ho 2002). Like M15, this globular cluster fits perfectly on the relation shown in Figure~\\ref{f:bhcorr}. This independent research strengthens the interpretation of the M15 data in terms of an intermediate-mass black hole. It has generally been believed that globular clusters and galaxies form and evolve quite differently, so it could be that it is a mere coincidence that they fall on the same $M_{\\rm bh}$---$\\sigma$ relationship. However, it could also have some deep physical significance. For example, it may point to a new link between galaxy formation and globular cluster formation. Or it may point to a link between the black holes in these systems. For example, it could be that the massive black holes in galaxies grew from seed black holes that arose in clusters. There may also be a link with the intermediate luminosity X-ray objects that are known to exist in external galaxies, and which have been argued to be intermediate mass black holes. These issues will need to be explored with future observational and theoretical studies.  Despite the interesting evidence for the presence of an intermediate mass black hole in M15, some words of caution are justified. All of the dynamical models that have been constructed for M15 remain somewhat idealized. This is true both for the Jeans models presented here and for the Fokker-Planck models presented elsewhere. For example, the Jeans models assume exact hydrostatic equilibrium, which is generally expected to be a good assumption (see Section~\\ref{ss:modtechnique}). Nonetheless, during periods of particularly rapid evolution in the cluster structure this assumption could yield results that are biased. Fokker-Planck models can address the cluster evolution directly and do not need to rely on the assumption of hydrostatic equilibrium. On the other hand, the results of Fokker-Planck models depend strongly on the processes of stellar evolution and binary heating, both of which are generally modeled only in rudimentary ways. More generally, globular clusters are complicated systems from a theoretical viewpoint, much more so than galaxies, and not all of the essential physics may yet have been fully understood. There is some evidence from observations that this may indeed be the case. For example, M15 rotates quite rapidly in the central regions (cf.~Figure~\\ref{f:kinprofile}), and this is not naturally explained by any theoretical model (Gebhardt \\etal 2000a). Also, the $\\sigma_{\\rm RMS}$ of M15 appears to have a small dip at intermediate radii (at $R \\approx 13''$), which is not naturally explained by any of the models constructed here (Figures~\\ref{f:models}--\\ref{f:anisotropy}). Further studies to test the observational reality of these features would be valuable, as would further theoretical work to address their origin.  We have assumed throughout our study, as has previous work on M15, that the observed kinematics are characteristic of the cluster, and are not contaminated by possible orbital motion of stars in binary systems. There are several reasons that make this a reasonable assumption (Hut \\etal 1992). The binary fraction of globular clusters is believed to be only of order 10\\%. Also, most of the stars in the velocity sample are red giants. Their relatively large radii imply that any binaries must have large separations and orbital velocities $\\lta 25 \\kms$. Inclination, phase and ellipticity effects imply that for an average binary at a random epoch only a fraction of the velocity amplitude will be observed along the line of sight. These issues conspire to make it extremely challenging to identify even a few binaries in globular clusters from large line-of-sight velocity studies, even with high quality multi-epoch data (Pryor \\etal 1989). One can turn this around to argue that the average observed kinematics of large samples of stars should not be influenced significantly by any orbital motion in binaries. This assumption is supported by the fact that the mass distribution inferred here from stellar kinematics agrees with that inferred from pulsar studies. Nonetheless, it would be useful for future studies to attempt a detailed quantitative assessment of the potential contamination of the line-of-sight kinematics of globular clusters by binaries.  In the future it may be possible to strengthen the observational constraints on the central structure of M15 through proper motion studies. With two additional velocity components it will be possible to directly establish the (an)isotropy of the stellar velocity distribution. It has been demonstrated that such studies are feasible with HST (e.g., Anderson \\& King 2000). However, the severe crowding in the central few arcsec of M15 may provide a significant hurdle to overcome. An alternative way to strengthen the observational constraints would be to increase the sample of radial velocities. This would require inclusion of fainter stars near the turnoff magnitude ($V \\approx 19$ in M15), which were inaccessible to our STIS study because of limited $S/N$. In principle, a 4m class telescope has sufficient light gathering power to perform such a study in a reasonable amount of time. By centering the spectra around the CO band-head ($2.3{\\mu}{\\rm m}$), it may be possible to take full advantage of adaptive optics to attain a spatial resolution comparable to that of HST."
    },
    "0209/astro-ph0209123_arXiv.txt": {
        "abstract": "We conducted a Very Large Array survey of eleven low luminosity active galactic nuclei for linear and circular polarization at 8.4 GHz.  We detected circular polarization in one source (M81*) and linear polarization in 3 sources. Sensitivity limits were $\\sim 0.1\\%$ for both modes of polarization in 9 of 11 sources. The detections confirm the importance of nonthermal emission in LLAGN.  However, detection rates for circular and linear polarization are lower for these sources than for more powerful AGN.  Fractional linear polarization in detected sources is also lower than in more powerful AGN. The weak linear polarization in the survey sources indicates their overall similarity to Sgr A*. Confusion with thermal sources, depolarization and weaker, less extended jets may contribute to these differences.  We detect a rotation measure $\\ga 7 \\times 10^4 \\rdm$ for NGC 4579.  This may arise from magnetized plasma in the accretion, outflow or interstellar regions.  Inverted spectra are present in both M81* and Sagittarius A* and absent from all sources in which circular polarization is not detected.  This suggests that optical depth effects are important in the creation of circular polarization. ",
        "introduction": "Circular and linear polarization are important diagnostics of nonthermal radio emission and its environment in extragalactic radio sources and galactic micro-quasars.  In active galactic nuclei (AGN), linear polarization (LP) has been used to demonstrate that the emission mechanism is synchrotron radiation, that magnetic fields are present and that shocks propagate in jets, compressing the magnetic fields \\citep[e.g.,][]{1985ApJ...298..114M}.   LP is typically greater than 1\\% and is present in a very high fraction of powerful extragalactic radio sources \\citep{1992ApJ...399...16A}.  Rotation measures are used to study the plasma density and magnetic field in accretion and outflow regions of AGN \\citep{2000ApJ...533...95T}.  The significance of circular polarization (CP) is less clear.  A variety of mechanisms can produce the small levels of CP seen in these sources, including intrinsic and extrinsic origins \\citep{1977ApJ...214..522J,2000ApJ...545..798M,2000AAS...197.8315B, 2002A&A...388.1106B,2002ApJ...573..485R}. Synchrotron and cyclotron emission produce CP in varying degrees depending on the the electron population, magnetic field strength and orientation.  Coherent emission processes may also lead to CP.  In addition, the presence of electrons with Lorentz factors $\\sim 1$ in or near a synchrotron source can significantly alter both CP and LP through Faraday effects.  The combination of these effects can lead to complex polarized spectra. Finally, a magnetized scattering region can produce scintillating CP.  Recently, there has been a sharp increase in interest in CP observations. CP has been detected in Sagittarius A* \\citep{1999ApJ...523L..29B,1999ApJ...526L..85S,2002ApJ...571..843B}, a large number of powerful extragalactic radio sources \\citep{2000MNRAS.319..484R,2001ApJ...556..113H} including some IDV sources \\citep{2000ApJ...538..623M}, and galactic microquasars \\citep{2000ApJ...530L..29F,2002MNRASinpressfender}.  VLBI imaging of powerful extragalactic radio sources indicate that the CP is confined to the cores.  The spectral and variability  properties in these sources vary widely indicating potentially diverse origins. In most sources, LP dominates CP, as expected for simple synchrotron sources.  A notable exception to this is Sgr A*, which exhibits no LP.  The polarization properties of low luminosity AGN (LLAGN) have not been previously investigated. \\citet{1986AJ.....91.1011R} observed the LP  of a sample of ``weak'' radio sources ($L \\sim 10^{23} - 10^{26} {\\rm\\ W\\ Hz^{-1}}$) with flat spectra.  Many of these sources were unpolarized at a level of 1\\% at 15 GHz, suggesting that they are not simply scaled down versions of powerful AGN.  However, there exists a significant gap in luminosity between these sources and Sgr A*, which has a radio luminosity $\\sim 10^{16} {\\rm\\ W\\ Hz^{-1}}$.   The intrinsic nature of the LLAGN is still in doubt \\citep[for recent reviews see][]{2001ApJ...562L.133U,2002AApNagarinpress}. These sources display emission line luminosities that may be produced through a variety of mechanisms, including stellar photoionization, collisional ionization through shocks or in a starburst. A variety of methods have demonstrated that some of the sources display AGN-like characteristics.  However, as in the case of Sgr A*, the luminosities are substantially sub-Eddington and there is dispute over the roles of advection, convection and outflow in these sources.  The detection of high brightness temperature radio components has made a strong case for the role of jets in these sources. LP and CP can act as probes of the radio-emitting region as well as of the accretion and outflow regions through which the emission propagates.  We present here a radio polarization survey of 11 nearby LLAGN with luminosities in the range $10^{20}$ to $10^{23}$ ${\\rm\\ W\\ Hz^{-1}}$.  In \\S 2, we describe the observations, our error analysis and the results.  CP is only detected in one source, NGC 3031 (M81*).  We have discussed the significance of this detection in another paper \\citep{2001ApJ...560L.123B}.  LP is detected in 4 sources. We discuss our results in \\S 3 and summarize in \\S 4. ",
        "conclusions": "We have observed linear and circular polarization in a sample of low luminosity AGN.  The polarization properties of these sources differ from those of more powerful AGN.  Linear polarization is less frequently detected and less strong when detected.  Circular polarization is less frequently detected.  The conclusion is not as simple as claiming that these sources differ intrinsically from the more powerful AGN, however.  While increased field disorder in the sources could explain the results, environmental depolarization effects could also play a role.  These polarization measurements probe regions on a scale of parsecs and smaller in these sources.  Detection of rotation measures in these sources is a direct measure of the environment of a supermassive black hole, which may be used to constrain models for accretion and outflow.  A broader sample of sources is necessary to demonstrate that the trends we see are statistically significant.  VLBI polarimetric imaging and spectropolarimetry may also test some of these conclusions."
    },
    "0209/astro-ph0209065_arXiv.txt": {
        "abstract": "By collecting optical and infrared photometry, as well as medium resolution spectroscopy, we have discovered a sample of low mass stars and brwon dwarfs in the young cluster IC2391. Using the lithium depletion boundary near the substellar limit, we have estimated the age of the cluster as $\\sim$50 Myr. We have also estudied the H$\\alpha$ emission in this sample. ",
        "introduction": "During the last few years, a large amount of knowledge has been gained regarding the nature and  properties of low mass stars and brown dwarfs. We have studied several nearby, young clusters, which, due to these characteristics, are excellent targets. One of these clusters is IC2391, located at 155 pc --(m-M)$_O$=5.95$\\pm$0.1--, and with a low interstellar reddening --E($B-V$)=0.06 (Patten \\& Simon 1996). In previous papers, we have estimated the age of the cluster based on the lithium depletion technique ($\\tau$=53$\\pm$5 Myr, Barrado y Navascu\\'es et al. 1999) and identified  a large number of low mass candidate members (Barrado y Navascu\\'es et al. 2001). In this paper, we present new spectroscopy of cluster members, improve the lithium  age, and analyze other properties such as the H$\\alpha$ activity and the mass function (MF). ",
        "conclusions": "We have derived membership for a sample of IC2391 low mass stars and BD candidate members using medium resolution spectroscopy. In addition, by combining these data with previous spectroscopy, we have re-evaluated the LDB of the cluster and its age, which is about 50 Myr. The distribution of the W(H$\\alpha$) is very similar to those characteristics of younger and older clusters, as well as the cluster MF."
    },
    "0209/astro-ph0209586_arXiv.txt": {
        "abstract": "Using the new 23~GHz receivers at the Very Large Array (VLA), we have detected NH\\3(6,6) emission ($\\nu=25.056025$~GHz) from hot ($>150$~K) molecular clouds in the central 10~pc of the Galaxy.  This is the first successful detection of NH\\3(6,6) with the VLA.  The brightest emission comes from a region interior to the ``circumnuclear disk'' (CND), less than 1.5~pc in projected distance from Sgr A*.  This region does not show molecular emission from lower energy transitions such as NH\\3(1,1) and (2,2), HCN(1-0) and HCO$^+$(1-0).  Line ratios of NH\\3(6,6) and (3,3) emission as well as NH\\3(6,6) line widths have peak values within 1.5~pc of Sgr A*, indicating that the gas is physically close to the nucleus.  NH\\3(6,6) is also detected towards many features outside the CND observed in NH\\3(1,1), (2,2), and (3,3). These features tend to lie along ridges of gas associated with Sgr A East or the massive ``molecular ridge'' that connects the ``20~\\kms'' and ``50~\\kms'' giant molecular clouds (GMCs). ",
        "introduction": "} At a distance of only $8.0\\pm0.5$~kpc \\citep{rei93}, the Galactic center provides a unique opportunity to study in detail the environment around a supermassive black hole.  It is now generally accepted that a black hole of $2.6\\times10^6$~$M_\\odot$ is located at the dynamical center of the Galaxy \\citep{eck97,ghe98}. In the radio, emission from just outside the black hole is observed as the strong ($\\sim1$~Jy) source, Sgr~A*.  Sgr A* is surrounded by arcs of ionized gas (Sgr A West) that appear to be feeding the nucleus \\citep{lo83,rob93}.  These arcs are, in turn, surrounded by an apparent ``ring'' of molecular material called the circumnuclear disk (CND, \\citet{gus87}).   Sgr A West and the CND are located in front of or just inside the front edge of the expanding supernova remnant (SNR), Sgr A East \\citep{ped89}.  The expansion of Sgr A East appears to be moving large amounts of material away from the nucleus, forming ridges of material on all sides \\firstcite.  However, some material may move towards the nucleus after being disrupted by the passing front and three filamentary ``streamers'' possibly feeding the nucleus have been detected in NH\\3 (Okumura et al. 1989; Ho et al. 1991; Coil \\& Ho 1999, 2000; MCH).  The metastable (J=K) NH\\3(J,K) rotation inversion transitions at $\\sim23$ GHz have proven to be useful probes of dense (10$^4$--10$^5$ cm$^{-3}$) molecular material near the Galactic center.  They tend to have a low optical depth and a high excitation temperature at the Galactic center, making them almost impervious to absorption effects. Satellite hyperfine lines separated by 10--30~\\kms ~on either side of the main line enable a direct calculation of the optical depth of the NH\\3 emission and line ratios of different transitions can be used to calculate the rotational temperature, $T_R$, of the gas.    We recently observed NH\\3(1,1), (2,2) and (3,3) emission from the central 10~pc of the Galaxy.  An important result from these studies is the apparent increase in line width and $T_R$ as gas approaches the nucleus (MCH; McGary \\& Ho in press).  However, the reality of this effect has remained in doubt because the emission also becomes fainter near Sgr A*.  We suspected that even the NH\\3(3,3) line becomes less sensitive to the extreme environment near Sgr A* and observed the central 4$'$ (10~pc) of the Galaxy in NH\\3(6,6), at 412~K above ground, using the new 23~GHz receivers at the Very Large Array\\footnote{The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.} (VLA) in order to detect the hottest molecular gas.  These data represent the first successful observations of NH\\3(6,6) with the VLA. NH\\3(6,6) is detected in many of the features seen in lower NH\\3 transitions, but the velocity integrated map is dominated by emission less than 1.5~pc (40$''$) in projected distance from the nucleus. Line ratios and extremely broad line widths (50--90~\\kms) indicate this molecular gas is physically close to the supermassive black hole. The remaining features lie predominantly along the edge of Sgr A East and have line widths greater than 20~\\kms ~indicating interaction with the shell. ",
        "conclusions": "With 25~GHz observations now practical at the VLA, NH\\3(6,6) is an excellent tracer of hot, dense molecular material.  Velocity integrated NH\\3(6,6) images are dominated by emission less than 1.5~pc in projected distance from Sgr A*.  High line ratios and line widths indicate that this molecular material is likely close to the supermassive black hole and is interior to the CND.  In addition, the hottest emission to the southeast of Sgr A* appears to connect the southern streamer to the nucleus. The remaining NH\\3(6,6) emission tends to lie along the ridges that surround the Sgr A East shell, tracing features that show a high line width in earlier NH\\3(1,1), (2,2), and (3,3) data."
    },
    "0209/astro-ph0209343_arXiv.txt": {
        "abstract": "The average polarization properties of conal single and double profiles directly reflect the polarization-modal structure of the emission beams which produce them.  Conal component pairs exhibit large fractional linear polarization on their inside edges and virtually complete depolarization on their outside edges; whereas profiles resulting from sightline encounters with the outer conal edge are usually very depolarized.  The polarization-modal character of subbeam circulation produces conditions whereby both angular and temporal averaging contribute to this polarization and depolarization.  These circumstances combine to require that the circulating subbeam systems which produce conal beams entail paired PPM and SPM emission elements which are offset from each other in both magnetic azimuth and magnetic colatitude.  Or, as rotating subbeam systems produce (on average) conal beams, one modal subcone has a little larger (or smaller) radius than the other.  However, these PPM and SPM ``beamlets'' cannot be in azimuthal phase, because both alternately dominate the emission on the extreme outer edge of the conal beam. While this configuration can be deduced from the observations, simulation of this rotating, modal subbeam system reiterates these these conclusions.  These circumstances are also probably responsible, along with the usual wavelength dependence of emission height, for the observed spectral decline in aggregate polarization.  A clear delineation of the modal polarization topology of the conal beam promises to address fundamental questions about the nature and origin of this modal emission---and the modal parity at the outer beam edges is a fact of considerable significance.  The different angular dependences of the modal ``beamlets'' suggests that the polarization modes are generated via propagation effects.  This argument may prove much stronger if the modal emission is fundamentally only partially polarized.  Several theories now promise quantitative comparison with the observations. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209496_arXiv.txt": {
        "abstract": "A phenomenological model of X--ray variability of accreting black holes is considered, where the variable emission is attributed to multiple active regions/perturbations moving radially towards the central black hole. The hard X--rays are produced by inverse Compton upscattering of soft photons coming from reprocessing/thermalization of the same hard X--rays. The heating rate of the Comptonizing plasma is assumed to scale with the rate of dissipation of gravitational energy while the supply of soft photons is assumed to diminish towards the center. Two scenarios are considered: (1) an inner hot flow with outer truncated standard accretion disc and (2) an accretion disc with an active corona and a thick hot ionized skin. A variant of the model is also considered, which is compatible with the currently discussed multi-Lorentzian description of power spectral densities of X--ray lightcurves.  In the inner hot flow scenario the model can reproduce the observed Fourier frequency resolved spectra observed in X--ray binaries, in particular the properties of the reprocessed component as functions of Fourier frequency. In the accretion disc with ionized skin scenario the reduction of soft photons due to the ionized skin  is insufficient to produce the observed characteristics. ",
        "introduction": "X--ray emission from accreting compact objects carries information about geometry and physical conditions of the accreting plasma. Spectral and timing analyses of the data from sources in low/hard state have enabled formulating a number of phenomenological scenarios for accretion onto black holes. The scenarios belong to two main groups: (1) models invoking standard optically thick accretion disc with an active corona and (2) models assuming that the optically thick disc is truncated at a radius larger than the last stable orbit, and is replaced by inner hot, optically thin flow (see Done 2002, Czerny 2002 for reviews). Models of both groups have some physical foundations. Magnetic dynamos in the disc may give rise to the active corona, as suggested by recent magneto-hydrodynamical simulations (Miller \\& Stone 2000), with plasma outflow and/or hot ionized disc skin responsible for the observed correlations between spectral parameters: the power law slope and amplitude of the reflected component (Zdziarski, Lubi\\'{n}ski \\& Smith 1999; Gilfanov, Churazov \\&  Revnivtsev 2000); On the other hand, disappearance of the cold disc due to evaporation may create the inner hot flow (R\\'{o}\\.{z}a\\'{n}ska \\& Czerny 2000). Both classes of models can explain time average energy spectra, as well as correlations between spectral parameters (see discussion of models and more references in \\.{Z}ycki 2002; hereafter Z02).  Specific models of X--ray variability were constructed for both geometries (see  Poutanen 2001 for review). In the active corona geometry the model of magnetic flare avalanches of Poutanen \\& Fabian (1999; hereafter PF99) is very successful in reproducing a number of observed properties (e.g.\\ power spectra, hard X--ray time lags). However, its simple extension to accretion disc with the ionized skin fails to reproduce the  Fourier-frequency resolved spectra (Revnivtsev, Gilfanov \\& Churazov 1999, 2001) observed from black hole binaries (Z02). In the inner hot flow geometry the drifting blob model (B\\\"{o}ttcher \\& Liang 1999) reproduces certain observed characteristics. However,  Maccarone, Coppi \\& Poutanen (2000) argue that flares in hard (Comptonized) X-rays are the primary driver of variability, the soft (thermal) emission being merely reprocessed hard X-rays (see also Malzac \\& Jourdain  2000).  The observed logarithmic energy dependence of the hard X--ray time lags was shown to be compatible with a simple propagation model, where X--ray emitting perturbations move radially towards the central black hole, the emitted spectrum hardening as the perturbations approach the center (Miyamoto \\& Kitamoto 1989; Nowak et al.\\ 1999; Kotov, Churazov \\& Gilfanov 2001). Moreover, the observed characteristic 'dip' in the energy dependence of time lags at the energy of the Fe \\Ka line can be explained in the propagation model, if the line emission from inner disc is suppressed (Kotov et al.\\ 2001).  Therefore in this paper we consider a phenomenological propagation model, where the variable X--ray emission is attributed to active regions and/or perturbations moving radially towards the central compact object. X--rays are assumed to be produced by  inverse Compton upscattering of soft photons coming from a cool, optically thick disc. X--ray luminosity is assumed to increase as the emission progresses, thus creating a flare of radiation. The Comptonized spectrum is assumed to evolve from softer to harder during the flare due to diminishing supply of seed photons. This may be due to either the optically thick disc being absent at small radii, or the thickness of the ionized disc skin increasing towards the center. From the simulated event files a number of statistics will be computed, which may be compared to available data. In particular, the Fourier-frequency resolved spectral properties are studied: both the continuum slope and Fe \\Ka line strength as functions of Fourier frequency, $f$, will be computed. The observed Fourier-frequency resolved spectra, calculated from the {\\it RXTE\\/} data of black hole binaries in the low/hard state, show two important characteristics: the continuum spectra get harder with increasing $f$, while the amplitude of reflection decreases with $f$ (Revnivtsev et al.\\ 1999, 2001).  We emphasize that the main feature of the considered models is the radial propagation of the X-ray emitting structures, with the related spectral evolution. This is meant to be compared with the models assuming radially localized flares (e.g.\\ the model of PF99). However, we do not hypothesize here about possible correlations between flares (although the idea of flare avalanches will be implemented in Sec~\\ref{sec:aval}), hence this model is not meant to explain in detail the X--ray power spectra and higher order statistics of the variability (Maccarone \\& Coppi 2002).  Plan of the paper is as follows: the model is described  in Sec.~\\ref{sec:model}, results for the hot inner flow geometry are presented in Sec.~\\ref{sec:results}, while results for the case of an accretion disc with hot ionized skin are presented in Sec.~\\ref{sec:hotskin}. In Sec.~\\ref{sec:multilor} a variant of the model is considered, which is compatible with the multi-Lorentzian description of power spectra (Nowak 2000). ",
        "conclusions": "\\label{sec:discuss}  We have considered a propagation model of X--ray variability in application to high energy emission from accreting compact objects. The motivation for our considerations comes from two directions: (1) problems faced by the models of localized flares (PF99) in explaining the $f$-resolved spectra (Z02), and (2) demonstration by Kotov et al.\\ (2001) that a propagation model with spectral evolution can naturally explain the hard X--ray time lags. The model can be formulated in both considered geometries: truncated standard disc with inner hot flow and a standard disc with an active corona and the hot ionized skin. It should be emphasized, however, that the main parameters entering the spectral computations, $\\lh/\\lsoft(t)$ and $R(t)$ can be {\\em predicted\\/} only in the latter model, while they have to be assumed rather arbitrarily in the former. It is the greater predictive power of the hot skin scenario what enables to find unique model predictions and demonstrate the problems which the model faces.  Similar computations of variability could be performed in the context of the plasma  outflow model of Beloborodov (1999a,b). Here the main parameter is the outflow speed. Qualitatively, the requirement for the model to work is that the speed increases inwards. A good physical model predicting the speed as a function of distance, X--ray flux and other parameters is necessary, in order to make quantitative predictions.  Let us concentrate now on the geometry of an inner hot X--ray producing flow with an outer standard optically thick disc. This is one of the possibilities of the geometrical arrangement of accreting multi-phase plasma. It naturally accounts for hard spectra of black hole X--ray binaries, small amplitude of the reprocessed component, and observed correlations between these spectral parameters (Done 2002 and references therein). Emission in the form of flares (shots) associated with active regions and/or perturbations traveling in the hot plasma might then be responsible for observed time and spectral variability, considering that X--ray variability is generally of stochastic character (Czerny \\& Lehto 1997). Such an extension of models of steady emission from hot flows (e.g.\\ Esin, McClintock \\& Narayan 1997) is necessary, considering the many observable characteristics related to variability. The situation is fully analogous to models invoking  active regions above an accretion disc (e.g.\\ Stern et al.\\ 1995). These models could explain the steady Comptonized emission by basically adjusting one parameter,  the ratio of heating to cooling compactnesses. When supplemented with the idea of plasma outflow (Beloborodov 1999a,b) or a hot  ionized skin (Nayakshin \\& Dove 2001), they cloud explain the diversity of spectra and correlations between spectral parameters. However, in order to explain the complex time variability, a number of additional features had to be added: correlations between flares (avalanches) to explain the PSD and spectral evolution during a flare as a result of, e.g., rising the active region above the disc (PF99). Similar developments can be expected in the hot flow scenario.   Precise geometry of the X--ray emitting structures in the hot flow is obviously uncertain, but certain constraints can be obtained. For example, compact active regions (size much smaller than the inner disc radius) moving through hot plasma totally devoid of optically thick plasma (i.e.\\ the cold disc sharply disrupted at certain radius), would intercept so few soft photons at $r<\\rtr$ that the continuum slope would be much harder than observed. One possibility would then be that the active regions are compact but the disappearance of the optically thick phase is gradual, moving from $\\rtr$ inwards. This could mean that small optically thick clouds form inside the hot plasma at $r<\\rtr$ and their filling factor gradually decreases inwards from $\\approx 1$ to 0. Detailed simulations would be necessary to predict spectra from such a configuration (e.g.\\ Malzac \\& Celotti 2002). At a rather speculative level one might point out two possibilities of formation of such compact traveling active regions. The accreting plasma might form multiple shocks where the dissipation of energy would preferentially take place, or the active regions might have something to do with the dissipation of magnetic energy inside the hot flow, which dissipation is necessary to maintain the equipartition between the magnetic and thermal energy (Bisnovatyi-Kogan \\& Lovelace 2000).  Another possibility for the geometry is to assume that the disc is rather sharply truncated but the X--ray emission comes from global coherent perturbations propagating through the entire volume of the hot plasma. Then the decrease of $\\lsoft/\\lh$ would roughly correspond to the decrease  of the solid angle subtended by the cold disc, as the perturbation approaches the center.  Temporal profiles of the flares considered in this paper were rather simple, therefore to produce a broad band PSD a distribution of parameters (e.g.\\ flare timescale) was required. One might envision more complex profiles based on, e.g., recent considerations of Maccarone \\& Coppi (2002). They suggest correlations between launch times of short flares ($\\trise\\approx 0.5$ sec), such that the short flares fill in a specific temporal envelope function. We note that their flares do not explain the whole range of time-scales, since the PSD above $\\sim 1$ Hz is under-predicted. Even shorter flares would be necessary to generate enough power at sub-second time-scales, perhaps filling in the short flares considered by Maccarone \\& Coppi (2002) as envelopes. The result might then be equivalent to a hierarchical scenario proposed by Uttley \\& McHardy (2001), where a long time-scales perturbation propagate from outside, breaking down into smaller, faster structures in a fractal-like manner."
    },
    "0209/astro-ph0209175_arXiv.txt": {
        "abstract": "{We have used the SEST 15-metre and Onsala 20-metre telescopes to perform deep (r.m.s. $\\ga30$ mJy) integrations of various molecular rotational transitions towards damped Lyman-alpha absorption systems (DLAs) known to occult millimetre-loud quasars. We have observed 6 new systems and improved the existing limits for 11 transitions.  These limits may be approaching the sensitivities required to detect new systems and we present a small number of candidate systems which we believe warrant further observation. ",
        "introduction": "\\label{sec:intro}  Millimetre-band molecular absorption systems along the line-of-sight toward quasars can provide a powerful probe of cold gas in the early Universe. Wiklind \\& Combes (1994; 1995; 1996b) have used such absorption lines to study a variety of properties of the absorbers themselves (e.g. relative column densities, kinetic and excitation temperatures, filling factors). Constraints on the cosmic microwave background temperature can also be obtained by comparing the optical depths of different rotational transitions (e.g. Wiklind \\& Combes 1996a). If the background quasar is gravitationally lensed, time delay studies can yield constraints on the Hubble constant (e.g. Wiklind \\& Combes 2001). However, these studies have so far been limited by the paucity of mm-band molecular absorbers: only 4 such systems are currently known--towards TXS 0218+357 ({Wiklind} \\& {Combes} 1995), PKS 1413+135 ({Wiklind} \\& {Combes} 1997), TXS 1504+377 ({Wiklind} \\& {Combes} 1996c) and PKS 1830--211 ({Wiklind} \\& {Combes} 1998).  Recent attention has focused on using molecular absorption lines as a probe for possible changes in the fundamental constants on cosmological time-scales. Detailed studies of the relative positions of heavy element optical transitions compared with laboratory spectra favour a smaller fine structure constant ($\\alpha \\equiv e^2/\\hbar c$) at redshifts $0.5 < z < 3.5$ at the 4.1$\\sigma$ significance level (Murphy et al. 2001b, Webb et al. 2001). The observed fractional change in $\\alpha$ [$\\Delta\\alpha/\\alpha = (-0.72 \\pm 0.18)\\times 10^{-5}$] is very small, so systematic errors have to be carefully considered. However, a thorough search for systematics has not revealed a simpler explanation of the optical results (Murphy et al. 2001c) and so independent constraints at similar redshifts are required.  Comparison of molecular rotational (i.e. mm-band) and corresponding H{\\sc \\,i}\\,21-cm absorption line frequencies has the potential to constrain changes in $\\alpha$ with a fractional precision $\\sim 10^{-6}$ per absorption system. The ratio of the hyperfine (\\HI) transition frequency to that of a molecular rotational line is $\\propto y \\equiv \\alpha^2g_p$ and so any variation in this will be observed as a difference in the apparent redshifts of these lines (Drinkwater et al. 1998).  From spectra of two of the known high redshift absorbers, PKS\\,1413+135 and TXS\\,0218+357, Carilli et al. (2000) and Murphy et al. (2001a) have \\begin{table*} \\begin{center} \\caption[]{{\\bf Top}: The northern sample. The radio flux densities at various GHz frequencies of the background quasars are given (see Curran {\\rm et~al.} 2002 for details). Note that a ``$\\approx$'' denotes a variable flux density. $z_{\\rm abs}$ is the DLA redshift and the final column gives the molecules most commonly detected in the four known absorbers (see Wiklind \\& Combes references), which fall into the Onsala 84--116 GHz band.  ($^*$HCO$^+$ $0\\rightarrow1$ in B 08279+5255 and $^*$CS $2\\rightarrow3$ in B 1017+1055 are 7 mm observations). Note that CS has so far only been detected in PKS 1830--211 ({Wiklind} \\& {Combes} 1996c) but since none of the commonly detected molecules fell into the band at the DLA redshift we tuned to this dense gas tracer. {\\bf Bottom}: The southern sample. The final column lists the molecules most commonly detected which fall into the SEST 78--116 GHz and 128--170 GHz bands. } \\label{t1} \\begin{tabular}{l c c c  c r r r c c c c} \\hline\\noalign{\\smallskip} Quasar & \\multicolumn{2}{c}{Coordinates (J2000)} & \\multicolumn{6}{c}{Radio flux densities (Jy)} &$z_{\\rm abs}$ & Transition \\\\ & h ~m ~s & d~~~$'$~~~$''$\t&  $S_{1.4}$&  $S_{5.0}$ & $S_{22}$ & $S_{90}$&  $S_{230}$ &  $S_{350}$ &  & \\\\ \\noalign{\\smallskip} \\hline\\noalign{\\smallskip} B 0235+1624 & 02 38 38.9 & 16 36 59 & 2.36& 1.64 & 2.48& $\\approx1-6$& $\\approx1-4$& --& 0.523869&  CS $2\\rightarrow3$\\\\ B 0248+430 & 02 51 34.5& 43 15 16& 1.43& 0.66&  $\\approx0.7$& 0.32& 0.08& --& 0.3939& CS $2\\rightarrow3$\\\\ B 0738+313&  07 41 10.7& 31 12 00 & 2.05& --& $\\approx1.3$& 0.27& 0.10&--& 0.2212&  CO $0\\rightarrow1$\\\\ B 0827+243 &  08 30 52.1 & 24 11 00 & 0.84 &0.89& $\\approx1.3$& 2.4&  $\\approx1.3$&   --& 0.5247& CS $2\\rightarrow3$ \\\\ B 08279+5255 & 08 31 41.6 & 52 45 18 &$0.001$  & --& --& --& --& 0.08& 2.97364& $^*$HCO$^+$ $0\\rightarrow1$ \\\\ ... &... & ...& ...& ...& ...&... & ...&... &... & CO $2\\rightarrow3$  \\\\ B 1017+1055 & 10 20 08.8 & 10 40 03 & 0.22 & -- &--& --& 0.004& --& 2.380& $^*$CS $2\\rightarrow3$\\\\ ... &... & ...& ...& ...& ...&... & ...&... &... &CO $2\\rightarrow3$\\\\ B 1328+307 & 13 31 08.3 & 30 30 33& 14.7& 1.41& 2.7& 0.75& 0.33& --& 0.69218&  HCO$^+$ $1\\rightarrow2$ \\\\ ... &... & ...& ...& ...& ...&... & ...&... &... &  CS $2\\rightarrow3$\\\\ \\noalign{\\smallskip} \\hline \\noalign{\\smallskip} B 0458--020 & 05 01 12.8 & -01 59 14 & 2.2 & $\\approx3$ & -- & 0.8 & $\\approx0.5$ & -- &2.0399&  HCO$^+$  $2\\rightarrow3$ \\\\ ... &... & ...& ...& ...& ...&... & ...&... &... &CO $3\\rightarrow4$  \\\\ B 0834--201 & 08 36 39.2 & -20 16 59 & 1.97 & 1.5 & --& 0.85 & 0.28 & -- & 1.715 & HCO$^+$  $2\\rightarrow3$ \\\\ ... &... & ...& ...& ...& ...&... & ...&... &... &CO $3\\rightarrow4$  \\\\ B 1229--0207 & 12 32 00.0 & -02 24 05 & 1.90 & 0.90 & $\\approx1$ & $\\approx0.4$ & $\\approx0.2$ & -- & 0.3950 & CO $0\\rightarrow1$ \\\\ ... &... & ...& ...& ...& ...&... &... &... &... &CO $1\\rightarrow2$ \\\\ B 1451--375 & 14 54 27.4 & -37 47 33 & 1.57 & 1.84 & -- & $\\approx1.5$ & $\\approx0.5$ & -- & 0.2761 & CO $0\\rightarrow1$ \\\\ ... &... & ...& ...& ...& ...&... &... &... &... &HCO$^+$ $1\\rightarrow2$ \\\\ \\noalign{\\smallskip} \\hline\\end{tabular} \\end{center} \\end{table*} \\begin{table*} \\begin{center} \\caption[]{Molecular lines previously searched for in radio-illuminated DLAs. All of the observations were performed with the IRAM 30-m telescope, except for HCO$^+$ $2\\rightarrow3$ towards B 0458--020 which was observed with the SEST. Again, apart from those marked ``*'' which are from {Wiklind} \\& {Combes} (1995), the radio flux densities are from the various sources cited in Curran {\\rm et~al.} (2002). $\\sigma$ is the lowest r.m.s. noise obtained for this frequency (with the reference given) recalculated for a channel width of 10 \\kms (see footnote 3). Note that {Wiklind} \\& {Combes} (1994b) is a search for CO emission, although any absorption features at the appropriate redshift should be apparent.} \\label{wc} \\begin{tabular}{l cl cccc c c l} \\hline\\noalign{\\smallskip} Quasar & $z_{\\rm abs}$ & Transition & \\multicolumn{5}{c}{Radio flux densities (Jy)}&  $\\sigma$ [mK] & Reference \\\\ & & &  $S_{1.4}$&  $S_{5.0}$ & $S_{22}$ & $S_{90}$&  $S_{230}$ & & \\\\ \\noalign{\\smallskip} \\hline\\noalign{\\smallskip} B 0235+1624 &0.5238  &  CO $1\\rightarrow2$ & 2.36 & 1.64& 2.48& 1.2$^*$& 1.3$^*$& 2.3& {Wiklind} \\& {Combes} (1995)\\\\ &       & HCO$^+$ $3\\rightarrow4$ & ... &... & ...& ...& ...& 3.6&  {Wiklind} \\& {Combes} (1995)\\\\ B 0458--020 & 2.0397 & CO $2\\rightarrow3$ & 2.2 & $\\approx3$& --& 0.8& $\\approx0.5$& 3.0&{Wiklind} \\& {Combes} (1994b) \\\\ & & HCO$^+$ $2\\rightarrow3$ & ... & ...& ...& ...&... & 7.1&{Wiklind} \\& {Combes} (1996b) \\\\ B 0528--2505 & 2.1408 & CO $2\\rightarrow3$ &1.50 &1.13   & --& --& --& 1.3&{Wiklind} \\& {Combes} (1994b)\\\\ B 0834--201 & 1.715 & HCO$^+$ $2\\rightarrow3$ & --& 1.5& --& 0.85& 0.28& 3.6 &{Wiklind} \\& {Combes} (1996b)\\\\ & & HCO$^+$ $3\\rightarrow4$ & ... &... &... &... &... & 3.9&  {Wiklind} \\& {Combes} (1996b)\\\\ B 1215+333 & 1.9984 & CO $2\\rightarrow3$ &0.18 & 0.08 & --& --& --& 4.4 &{Wiklind} \\& {Combes} (1994b) \\\\ B 1229--0207 & 0.39498 & CO $1\\rightarrow2$ & 1.90 & 0.90 & $\\approx1$ & $\\approx0.4$ & $\\approx0.2$ & 12 & {Wiklind} \\& {Combes} (1995)\\\\ B 1328+307 & 0.69215 &  CO $1\\rightarrow2$ & 14.7 & 1.41& 2.7& 0.6$^*$&0.24$^*$ & 6.1&{Wiklind} \\& {Combes} (1995) \\\\ & & CO $2\\rightarrow3$ & ... & ...&... &... &... & 6.5 &{Wiklind} \\& {Combes} (1995) \\\\ & & HCO$^+$ $1\\rightarrow2$ &...  &... &... &... &... & 5.8&{Wiklind} \\& {Combes} (1995) \\\\ B 2136+142 & 2.1346 &  CO $2\\rightarrow3$ & -- & 1.11& 1.6& 0.59$^*$& 0.25& 2.8& {Wiklind} \\& {Combes} (1996b) \\\\ & & CO $3\\rightarrow4$ & ... & ...& ...& ...& ...& 2.3& {Wiklind} \\& {Combes} (1996b)\\\\ & & CO $5\\rightarrow6$ & ...  &...&... &... &... & 3.9&   {Wiklind} \\& {Combes} (1996b)\\\\ & & HCO$^+$ $2\\rightarrow3$ & ... & ...&... &... & ...& 3.4& {Wiklind} \\& {Combes} (1995) \\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\end{center} \\end{table*} obtained constraints on $\\Delta y/y$ consistent with zero $y$-variation at redshifts $z_{\\rm abs} = 0.6847$ and 0.24671, respectively. However, the major uncertainty in the mm/H{\\sc \\,i} comparison is that intrinsic velocity differences between the mm and H{\\sc \\,i} absorption lines can be introduced if the lines-of-sight to the millimetre wave and radio continuum emission regions of the quasar differ, as is certainly the case for PKS\\,1413+135 and TXS\\,0218+357 (Carilli et al. 2000). Thus, a {\\it statistical} sample of mm/H{\\sc \\,i} comparisons is required to independently check the optical results.  One systematic approach to finding new high redshift molecular absorbers is to target high column density systems with known redshifts. A convenient sample is therefore the damped Lyman-alpha absorbers defined to have neutral hydrogen column densities $N_{\\rm HI}\\ga10^{20}$ cm$^{-2}$. We therefore compiled a catalogue of all known DLAs and shortlisted those which are illuminated by radio-loud quasars (Curran {\\rm et~al.} 2002)\\footnote{Available from http://www.phys.unsw.edu.au/$\\sim$sjc/dla} Of these, seven in the northern sky have measured millimetre fluxes, and in the south there are around a dozen DLAs illuminated by millimetre-loud quasars, of which we observed the four loudest (Table \\ref{t1}).  We note that Wiklind \\& Combes (1994b; 1995; 1996b) searched with null results for redshifted molecular emission and absorption towards 12 and 46 quasars, respectively. However, only 11 are occulted by DLAs and not all of these are radio-loud (Table \\ref{wc}). This motivates a more systematic search for millimetre absorption systems associated with DLAs. In this paper we present the results of our first search: the DLAs which occult known millimetre-loud quasars with the SEST\\footnote{The Swedish-ESO Sub-millimetre Telescope is operated jointly by ESO and the Swedish National Facility for Radio Astronomy, Onsala Space Observatory, Chalmers University of Technology.} and Onsala 20-m telescopes. ",
        "conclusions": "Upon the removal of a low order baseline and subsequent averaging of the data for each quasar, no absorption features of $\\geq3\\sigma$/channel were found\\footnote{See http://www.phys.unsw.edu.au/$\\sim$sjc/dla-fig1.ps.gz for the spectra and corresponding r.m.s. noise levels.}. In Table \\ref{sum} we summarise the derived upper limits together with the previously published results. \\begin{table*} \\caption[]{Summary of our (Table \\ref{t1}) and the previously published results (Table \\ref{wc}). $V$ is the visual magnitude of the background quasar, $N_{\\rm HI}$ [\\scm] is the DLA column density from the Lyman-alpha line and $\\tau_{\\rm 21~cm}$ is the optical depth of the redshifted 21 cm \\HI ~line (see Curran {\\rm et~al.} 2002). The optical depth of the relevant millimetre line is calculated from $\\tau=-\\ln(1-3\\sigma_{{\\rm rms}}/S_{{\\rm cont}})$, where $\\sigma_{{\\rm rms}}$ is the r.m.s. noise level at a given resolution and $S_{{\\rm cont}}$ is the continuum flux density, estimated from the values at the neighbouring frequencies (Tables \\ref{t1} and \\ref{wc}). This is done for resolutions of 10 \\kms~ [$\\tau_{\\rm mm}(10)$] and 1 \\kms ~($\\tau_{\\rm mm}$), where for the latter we quote only the best existing limit. For all optical depths, $3\\sigma$ upper limits are quoted and ``--'' designates where $3\\sigma>S_{{\\rm cont}}$, thus not giving a meaningful value for this limit. Blanks in the $\\tau_{\\rm 21~cm}$ field signify that there are no published \\HI ~absorption data for these DLAs. The final column gives the best existing limit of the column density per unit velocity estimated for the transition [\\scm (\\kms)$^{-1}$] (see main text).  } \\label{sum} \\begin{center} \\begin{tabular}{l c c c c c c c c } \\hline\\noalign{\\smallskip} DLA & $z_{\\rm abs}$ & Transition &  V & $N_{\\rm HI}$ & $\\tau_{\\rm 21~cm}$ & $\\tau_{\\rm mm} (10)$ & $\\tau_{\\rm mm}$ & $N_{\\rm mm}/\\Delta v$\\\\ \\noalign{\\smallskip} \\hline\\noalign{\\smallskip} B 0235+1624 & 0.52398$^*$ & CO $1\\rightarrow2$ & 15.5 & $4\\times10^{21}$ & 0.05--0.5$^a$ &  $<0.03$ & $<0.09$& $<3\\times10^{14}$\\\\ ...\t&  ...& HCO$^+$ $3\\rightarrow4$ & ... &... &...  & $<0.03$ & $<0.3$ & $<4\\times10^{12}$\\\\ ...& 0.523869& CS $2\\rightarrow3$ & ... & ...&  ...&  $<0.2$& $<0.9$ & $<2\\times10^{13}$\\\\ B 0248+430 & 0.3939& CS $2\\rightarrow3$ & 17.7 &$4\\times10^{21}$ &  0.20& $<3$ & --&-- \\\\ B 0458--020 & 2.0397$^*$/9  & HCO$^+$  $2\\rightarrow3$  & 18.4 & $5\\times10^{21}$ &0.30$^b$  & $<0.3$/$<0.2$ & $<0.4$&  $<2\\times10^{12}$\\\\\\ ...\t&  2.0397$^*$ & CO $2\\rightarrow3$ & ... & ...&  ... & $<0.4$ & --& --\\\\ ...\t&   2.0399 & CO $3\\rightarrow4$& ... & ...& ... & $<0.1$ & $<1$& $<2\\times10^{16}$\\\\ B 0528--2505 & 2.1408 & CO $2\\rightarrow3$ &19.0  &$4\\times10^{20}$ & $<0.2$ & \\multicolumn{3}{c}{{\\it No published millimetre fluxes}} \\\\ B 0738+313 &  0.2212 & CO $0\\rightarrow1$ & 16.1 &$2\\times10^{21}$ & 0.07& $<1$ &--& --  \\\\ B 0827+243 & 0.5247 & CS $2\\rightarrow3$ & 17.3  &$2\\times10^{20}$ &0.007  &$<0.1$  & $<0.4$& $<7\\times10^{12}$\\\\ B 08279+5255  & 2.97364& HCO$^+$ $0\\rightarrow1$ & 15.2  &$1\\times10^{20}$ &  &  --$^c$ &--& --\\\\ ...\t&  ... & CO $2\\rightarrow3$ & ... &... & ... & --$^c$ &--& --\\\\ B 0834--201 & 1.715 & HCO$^+$  $2\\rightarrow3$ &  18.5& $3\\times10^{20}$ & & $<0.1$/$<0.2$ & $<0.4$& $<2\\times10^{12}$ \\\\ ...\t&  ... &HCO$^+$  $3\\rightarrow4$& ... & ...&  ...&   $<0.2$ &$<0.6$ &  $<8\\times10^{12}$\\\\ ...\t&  ... &  CO $3\\rightarrow4$& ... & ...&  ...&  --&--& -- \\\\ B 1017+1055 & 2.380& CS $2\\rightarrow3$ & 17.2 & $8\\times10^{19}$ &  &--$^d$ &--& --  \\\\ ...\t&  ... & CO $2\\rightarrow3$ & ... & ..&  ...&  --$^d$&--& -- \\\\ B 1215+333 & 1.9984 & CO $2\\rightarrow3$ & 18.1 & $1\\times10^{21}$ &   & --$^{I,e}$ &--& --  \\\\ B 1229--0207 & 0.3950 & CO $0\\rightarrow1$  & 16.8   &$1\\times10^{21}$ &  &  $<0.3$& $<1$& $<6\\times10^{15}$\\\\ ...\t&  ... & CO $1\\rightarrow2$& ... & ...&  ...& $<1$  &--& --\\\\ ...\t& 0.39498$^*$  & CO $1\\rightarrow2$& ... & ...&  ...&-- & --& --\\\\ B 1328+307 & 0.69215 & HCO$^+$  $1\\rightarrow2$ & 17.3  &$2\\times10^{21}$ &0.11  & $<0.2$/$<0.4$ &$<0.7$ & $<2\\times10^{12}$\\\\ ...\t&  ... & CS $2\\rightarrow3$ & ... &... & .... & $<0.3$ & $<2$& $<4\\times10^{13}$\\\\ ...\t&  ... & CO $1\\rightarrow2$ & ... & ...& ... & $<0.3$ &$<1$ &$<3\\times10^{15}$\\\\ ...\t&  ... & CO $2\\rightarrow3$ & ... & ...& ... & $<0.6$ & --&--\\\\ B 1451--375 &0.2761 & HCO$^+$ $1\\rightarrow2$ & 16.7 &$1\\times10^{20}$ & $<0.006$ &$<0.2$&$<0.6$ & $<2\\times10^{12}$ \\\\ ...\t&  ... &CO $0\\rightarrow1$  & ... & ...& ... & $<0.09$ &$<0.3$ &$<2\\times10^{15}$\\\\ B 2136+141 & 2.1346 & HCO$^+$ $2\\rightarrow3$&18.9  & $6\\times10^{19}$&   &  $<0.08$ &$<0.3$ &$<1\\times10^{12}$\\\\ ...\t&  ... &  CO $2\\rightarrow3$ & ... &  ...&   ...&  $<0.2$ & $<0.6$& $<4\\times10^{15}$\\\\ ...\t&  ... &  CO $3\\rightarrow4$ & ... &  ...&   ...& $<0.2$ &$<0.8$ & $<2\\times10^{16}$\\\\ ...\t&  ...&  CO $5\\rightarrow6$ & ... &  ...&  ... &  $<0.6$ & --& --\\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\end{center} {Notes: Where we have observed towards the same quasar $^*$denotes Wiklind \\& Combes results (see Table \\ref{wc} for details). $^a$\\HI ~absorption at $z=0.52385$ (Briggs \\& Wolfe 1983). $^b$\\HI ~absorption at $z=2.03945$ (Wolfe et~al. 1985; Briggs et~al. 1989). $^c$Assumed flat spectrum, i.e. $S_{45}=S_{87}=S_{350}=0.08$ Jy. $^d$Assumed $S_{44}\\approx0.015$ Jy and $S_{102}\\approx0.008$ Jy; $^e$$S_{115}\\approx0.011$ Jy, based on the two available measured values (Table \\ref{t1}).} \\end{table*} Comparing the optical depth limits with those in the four known absorbers, the strongest absorber, towards PKS 1830--211, has $\\tau_{\\rm mm}\\approx1$ for HCO$^+$ $1\\rightarrow2$ and $\\tau_{\\rm mm}\\approx0.5$ for HCO$^+$ $2\\rightarrow3$. The weakest, towards PKS 1413+135, has $\\tau_{\\rm mm}\\approx0.1$ for both CO $0\\rightarrow1$ and HCO$^+$ $2\\rightarrow3$. The two remaining systems have optical depths similar to that towards PKS 1830--211 and so our new observations and the previous surveys should have been sensitive to (at least) CO and HCO$^+$ at similar strengths to those in 3 of the 4 known systems. Only for the observations of B 0235+1624, HCO$^+$ $2\\rightarrow3$ towards B 2136+141 (IRAM) and CO $0\\rightarrow1$ towards B 1451--375 are the surveys sensitive enough, particularly for resolutions of $\\ga10$ \\kms, to detect an absorption system of a strength similar to that towards PKS 1413+135 (See Table 4).  From the optical depths we may estimate column density limits for each transition.  In Table \\ref{sum} we give the $3\\sigma$ optical depth limits according to a velocity resolution of 10 \\kms ~as well as at a resolution of 1 \\kms. We give the former in order to show the limits corresponding to a visual inspection of the spectra$^3$ and facilitate a more direct comparison with the previous surveys. By multiplying the latter with the expected width of a line, the velocity integrated optical depth for a $3\\sigma$ detection at $\\Delta v =1$ \\kms ~is obtained.  For all of the limits, assuming LTE conditions, we can estimate the total column density of each transition from \\begin{equation} N_{\\rm mm}=\\frac{8\\pi}{c^3}\\frac{\\nu^{3}.f}{g_{\\mu}A_{\\mu}}\\left.\\int\\right.\\tau dv, \\end{equation} where $\\nu$ is the rest frequency of the $J\\rightarrow J+1$ transition, $g_{\\mu}$ and $A_{\\mu}$ are the statistical weight and the Einstein A-coefficient of the transition, $\\int\\tau dv$ is the velocity integrated optical depth and $f$ is the product of the partition function for an excitation temperature, $T_x$ (assumed $\\approx10$ K), and $e^{E_J/kT_x}/(1-e^{-h\\nu/kT_x})$ (see {Wiklind} \\& {Combes} 1995, 1996b, 1999 for details). From the optical depth limits, we derive limits on the total column density per unit velocity, thus normalising the limits which depend on the spectral resolution to which the data have been smoothed.  Since we have only upper limits, and thus no knowledge of the width of any line which may be hidden in the noise, as well as uncertainties in the conversion to H$_2$ column densities for molecules other than CO, it is difficult to draw meaningful comparisons between $N_{\\rm HI}$ and $N_{\\rm mm}$. However, by assuming an absorption line of FWHM  $\\sim20$ \\kms ~(as in the case of 3 of the 4 known absorbers), we estimate for the lowest optical depth limit (CO $1\\rightarrow2$ in B 0235+1624) a value of $N_{\\rm CO}\\lapp6\\times10^{15}$ \\scm ~at the $3\\sigma$ level and a resolution of 1 \\kms, i.e. $N_{\\rm CO}/N_{\\rm HI}\\lapp2\\times10^{-6}$. According to $N_{{\\rm H}_{2}}\\sim10^{4}N_{\\rm CO}$ (e.g. {Wiklind} \\& {Combes} 1998), the molecular to atomic hydrogen column density ratio is $N_{{\\rm H}_{2}}\\lapp2\\% ~N_{\\rm HI}$, which is similar to the lower limit estimated from the optical detection of molecular hydrogen\\footnote{In the case of $z>1.8$ sources, the ultra-violet lines of H$_2$ are redshifted into the optical window, making molecular hydrogen readily observable at these frequencies. As well as towards B 0528--2505 ({Foltz}, {Chaffee} \\& {Black} 1988) molecular hydrogen has also been detected in the DLAs occulting the radio-quiet quasars B 0000--2620 ({Levshakov} {\\rm et~al.} 2000), B 0013--0029 ({Ge} \\& {Bechtold} 1997; {Petitjean}, {Srianand} \\& {Ledoux} 2002), B 0347--3819 (Levshakov {\\rm et~al.} 2002), B 1232+0815 ({Ge} \\& {Bechtold} 1997; {Srianand}, {Petitjean} \\& {Ledoux} 2000) and the inferred ({Wolfe} {\\rm et~al.} 1995) DLA towards B 0551--3637 (Ledoux, Srianand \\& Petitjean 2002).} in the DLA towards B 0528--2505 ({Carilli} et al. 1996). That is, although we have no clear detections of any molecular absorption lines, the current limits may be approaching those sufficiently low in order to detect molecular absorption in damped Lyman-alpha systems.  Bearing this in mind, we then analysed the spectra for tentative features within a range corresponding to the uncertainty in the DLA redshift\\footnote{Although we have searched for possible absorption lines close to the expected redshifts, it may possible that the line-of-sight to the mm-band and optical emission regions of the quasars may differ ($\\delta_{\\rm LOS} >0$), giving rise to intrinsic offsets between any millimetre absorption and the known DLA redshifts. Drinkwater et al. (1998) compared millimetre and H{\\sc\\,i}\\,21-cm Galactic absorption profiles and found that the individual velocity components in the profiles corresponded to within $\\delta_{\\rm LOS} = 1.2$ \\kms. Carilli et al. (2000) argued that $\\delta_{\\rm LOS} \\sim 10$ \\kms ~is typical of the velocity dispersion of the interstellar medium in galaxies, and that it may even be as high as $\\delta_{\\rm LOS} \\sim 100$ \\kms. In order to account for this, we searched for possible absorption lines in a velocity region $[-\\delta_{\\rm LOS}-\\sigma_{\\rm DLA}, \\delta_{\\rm LOS}+\\sigma_{\\rm DLA}]$ centered on the expected DLA frequency with $\\delta_{\\rm LOS}=100$ \\kms~and $\\sigma_{\\rm DLA}$ the uncertainty in the DLA redshift.}. In each case the continuum level was defined using a polynomial baseline fit and for each channel we generated a 1$\\sigma$ error from the r.m.s noise in a window of $2N_{\\rm win}+1$ channels centered on that channel. Using this error array, we identified absorption features as series of $n$ channels over which a deviation from the continuum level was observed with significance $>\\sigma_{\\rm lim}$ standard deviations. Note that the significance of an absorption feature, $\\sigma_n$, is an overestimate because the flux in adjacent channels is (positively) correlated [each spectral resolution element is sampled at approximately the Nyquist rate, see Murphy, Curran \\& Webb (2002) for details].  According to the ``expected'' $\\delta_{\\rm LOS}=100$ \\kms ~velocity differences in the optical and millimetre wave lines-of-sight$^5$, the only candidate features identified in our analysis are HCO$^+$ $2\\rightarrow3$ and CO $3\\rightarrow4$ towards B 0458--020 as well as the HCO$^+$ $2\\rightarrow3$ line towards B 0834--201. As seen in Table 4, however, increasing the line-of-sight velocity difference to $\\delta_{\\rm LOS}=200$ \\kms~ significantly increases the number of candidates.  This suggests that either the authenticity of the candidate features is questionable and/or we have \\begin{table*}[hbt] \\caption[]{Candidate molecular absorption systems located within $\\approx\\pm200$ \\kms ~of the DLA redshift. $\\nu_{obs}$ is the observed frequency, $\\sigma_n$ is the significance of the putative feature, $z_{\\rm mm}$ is the absorption redshift and $z_{\\rm 21~cm}$ is the redshift of the \\HI ~line (see Curran {\\rm et~al.} 2002). The final column gives the offset in \\kms ~between the $z_{\\rm mm}$ and the nominal value of $z_{\\rm DLA}$, with our estimate of uncertainty in the DLA redshift$^3$ quoted.} \\label{cand} \\begin{center} \\begin{tabular}{l c  c c r c r} \\hline\\noalign{\\smallskip} DLA & Transition & $\\nu_{obs}$ [GHz] &$\\sigma_n$  & $z_{\\rm mm}$ & $z_{\\rm 21~cm}$ & $z_{\\rm mm}-z_{\\rm DLA}$\\\\ \\noalign{\\smallskip} \\hline\\noalign{\\smallskip} B 0248+430 & CS $2\\rightarrow3$ & 105.3701 & 4.2 & 0.39479 & 0.3941 & $190\\pm20$\\\\ B 0458--020 & HCO$^+$  $2\\rightarrow3$ & 87.9994 & 3.8 & 2.04045 & 2.03945 &$50\\pm10$\\\\ ...& ...& 88.0229 & 3.2 & 2.03964 & ...&$-30\\pm10$\\\\ ...& CO $3\\rightarrow4$& 151.7144 & 4.3 & 2.03888 & ...& $-100\\pm10$\\\\ B 0738+313 &  CO $0\\rightarrow1$ & 94.3116 & 4.4 & 0.22224 & 0.2212 &$240\\pm20$\\\\ B 0834--201 & HCO$^+$  $2\\rightarrow3$ & 98.5135 & 4.4 & 1.71595 & --&$100\\pm100$\\\\ \\noalign{\\smallskip} \\hline\\noalign{\\smallskip} PKS 1413+135 & HCO$^+$  $2\\rightarrow3$ & 214.6110 & 4.6 & 0.24671 & 0.24671 &--\\\\ \\noalign{\\smallskip} \\hline \\end{tabular} \\end{center} {Note: The last entry is a known absorption system included as a guide. A spectrum was obtained after $\\approx 30$ hours with SEST (August 2001) which may well have been rejected upon visual inspection$^3$.} \\end{table*} underestimated the uncertainties in the DLA redshifts. It is interesting to note that all but one of the candidate lines occur in DLAs with high column densities ($N_{\\rm HI}\\ga10^{21}$ \\scm). The one exception is B 0834--201 which has the highest visual magnitude of these candidates. We emphasise that Table 4 is only intended as a shortlist for follow up observations, which may or may not yield absorption systems at the listed redshifts.  Of all the candidate systems, the most promising and visually striking example of a possible absorption line is the 151.71 GHz feature towards B~0458--020 (Fig. \\ref{d1}), which has a high visual magnitude as well as the highest neutral hydrogen column density of the sample. \\begin{figure}[h] \\vspace{6.7 cm} \\setlength{\\unitlength}{1in} \\begin{picture}(0,0) \\put(-0.3,3.05){\\special{psfile=0458-151-high.ps hscale=39 vscale=39 angle=270}} \\end{picture} \\caption[]{CO $3\\rightarrow4$ at $z=2.039$ towards B 0458--020 at the full spectral resolution of 1.38 \\kms. The antenna temperature of the peak channel is $8$ mK (c.f. $\\sigma=3$ mK) and the integrated main-beam intensity of the line is $-0.10\\pm0.02$ K \\kms. The expected frequencies of the line for the (approximate) DLA redshift of 2.0399 and \\HI~ redshift of $2.03945$ ({Turnshek} {\\rm et~al.} 1989, {Wolfe} {\\rm et~al.} 1995) are shown.} \\label{d1} \\end{figure} Comparing this ``detection'' of -8 mK (0.24 Jy at SEST) with the estimated flux density at 152 GHz (0.62 Jy, estimated from the neighbouring values in Table \\ref{t1}) gives an optical depth of $\\tau_{\\rm mm}\\approx0.5$. The $\\approx9$ \\kms ~FWHM of the line gives a column density estimate of $N_{\\rm CO}\\approx8\\times10^{16}$ \\scm ~or $N_{{\\rm H}_{2}}\\sim10^{21}$ \\scm, i.e. $\\sim20$\\% of $N_{\\rm HI}$. This compares well with the column densities and their ratios for the known absorbers, i.e. $10-40\\%$ ({Wiklind} \\& {Combes} 1994, 1995, 1996c, 1999, {Carilli} {\\rm et~al.} 1998), although it is considerably higher than the rest of the DLA sample."
    },
    "0209/astro-ph0209205_arXiv.txt": {
        "abstract": "We have used a high spatial resolution \\chandra{} observation to examine the core mass distribution of the unusually regular cD cluster Abell 2029. This bright, nearby system is especially well-suited for analysis of its mass distribution under the assumption of hydrostatic equilibrium: it  exhibits an undisturbed, symmetric X-ray morphology, and a single-phase intracluster medium (ICM).  From the deprojected temperature and density profiles we estimate the total mass, and the contributions of the gas and dark matter (DM) components from $<3\\arcsec$ to $\\sim3\\arcmin$ ($<4.4-260$\\hkpc, $0.001-0.1\\rvir$). The gas density profile is not adequately described by a single $\\beta$-model fit, due to an increase in the density at the center ($r<17$\\hkpc, $<12\\arcsec$), but it is well fitted by either a double $\\beta$-model, or a ``cusped'' $\\beta$-model. The temperature data increase as a function of radius and are well-fitted by a \\citeauthor{ber86} profile and approximately by a power-law $T(r)\\propto r^{\\alpha_T}$, with $\\alpha_T=0.27\\pm0.01$.  Using the fitted profiles to obtain smooth functions of density and temperature, we employed the equation of hydrostatic equilibrium to compute the total enclosed mass as a function of radius. We report a total mass of $9.15\\pm0.25 \\times 10^{13} h_{70}^{-1}~\\msun$ within $260$\\hkpc, using the chosen parameterization of gas density and temperature. The mass profile is remarkably well described down to $0.002\\rvir$ by the \\citeauthor*{nav97} (NFW) profile, or a \\citeauthor{her90} profile, over 2 decades of radius and 3 decades of mass. For the NFW model, we measure a scale radius $r_s = 540\\pm90$\\hkpc{} ($\\approx0.2\\rvir$) and concentration parameter $c=4.4\\pm 0.9$. The mass profile is also well-approximated by a simple power-law fit ($M(<r)\\propto r^{\\alpha_m}$), with $\\alpha_m=1.81\\pm0.04$ (corresponding to a logarithmic density profile slope of $-1.19\\pm 0.04$). The density profile is too shallow to be fitted with the profile described by \\citeauthor{moo99}. The consistency with NFW down to $<0.01\\rvir$ is incompatible with the flattened core DM profiles predicted for self-interacting DM (e.g., \\citeauthor{spe00}), and thus contrasts with the strong deviations from CDM predictions observed in the rotation curves of low surface brightness galaxies and dwarf galaxies. This suggests that while CDM simulations may adequately describe objects of cluster mass, they do not currently account properly for the formation and evolution of smaller halos.  Assuming that the cD dominates the optical cluster light within its effective radius ($R_e=52\\arcsec$, 76\\hkpc), we observe a total mass-to-light ratio $M/L_V \\approx12 \\msun/\\lsun$ at $r<20$\\hkpc, rising rapidly to $>100 \\msun/\\lsun$ beyond 200\\hkpc. The consistency with a single NFW mass component, and the large $M/L$ suggest the cluster is DM-dominated down to very small radii ($\\lesssim 0.005\\rvir$). We observe the ICM gas mass to rise from $3 \\pm 1\\%$ of the total mass in the center to $13.9 \\pm 0.4\\%$ at the limit of our observations. This provides an upper limit to the current matter density of the Universe, $\\Omega_{m}\\leq 0.29\\pm0.03~h_{70}^{-1/2}$. ",
        "introduction": "}  The large mass-to-light ratios ($M/L$) of galaxy clusters \\citep[$M/L_B \\approx 200-300 h_{70}~\\msun/\\lsun$, e.g.,][]{gir02} indicate that they contain large quantities of dark matter (DM hereafter). The nature and distribution of DM is the current subject of much theoretical work in cosmology, with detailed simulations yielding different expectations for the amount and distribution of DM in cluster cores \\citep[e.g.,][]{nav97,moo99,dave01}. X-ray observations of the density and temperature of the hot intracluster medium (ICM) gas probe the mass of a galaxy cluster, under the assumption of hydrostatic equilibrium . Such data potentially provide constraints on cluster DM simulations, and thus test DM theory \\citep[e.g.,][]{evr96,ara02,san02}.  Prior generations of X-ray satellites have provided a wealth of cluster observations, from which we have begun to build a picture of large-scale radial temperature variations \\citep[e.g.,][]{irw01,ett02b}, as well as two dimensional temperature maps of disturbed systems \\citep[e.g.,][]{bri94,mar98,mar99,deg99b,joh02}. However, detailed temperature and density profiles at the smallest scales ($r<0.1 \\rvir$) exist for only a few systems, such as Virgo \\citep{nul95}, which exhibit irregularities in their cores. While gravitational lensing studies provide a unique and important probe of DM in cluster cores \\citep{dah02,nat02,san02}, they generally cannot be applied to nearby systems, and they may also be contaminated by other sources of mass along the line of sight. The advent of the \\chandra{} and \\xmmn{} satellites now allows us to measure the ICM properties with simultaneous spatial resolution comparable to optical studies, thus providing a completely independent mass estimator over the same spatial scales. The main criticism levelled at X-ray studies is the veracity of the hydrostatic equilibrium assumption in the actual clusters under study. Several groups have now obtained X-ray constraints on the DM profiles of clusters of galaxies, which are apparently consistent with the CDM paradigm \\citep[e.g.,][]{dav01,pra02,ett02a,sch01,all01,mat02,ara02}. Such systems are either too distant for a detailed analysis at $<0.1\\rvir$, or they contain morphological disturbances indicating possible departures from hydrostatic equilibrium, especially at $r\\lesssim 100$\\hkpc. Although simulations suggest that the X-ray analysis of such clusters is generally unaffected \\citep{tsa94,evr96,mat99}, some authors argue that mass estimates will be biased at small radii in such systems, which may reflect the majority of clusters \\citep[see especially][]{mar02}.  Abell 2029 (A2029) is a nearby ($z=0.0767$), well-studied cluster of galaxies which provides an excellent opportunity to probe the dark matter content of a massive object. It has a very high X-ray flux and luminosity, as well as a hot ICM, indicating a massive system. We have previously analyzed the \\chandra{} observations of A2029 in \\citet*[][Paper 1 hereafter]{P5}, presenting the temperature and Fe abundance data. It exhibits a very regular optical and X-ray morphology, and is an excellent example of a relaxed system with no evidence of disturbances (e.g., shock fronts, filaments, or ``cold fronts'') present in other systems. In a morphological analysis of 59 clusters, \\citet{buo96b} found the global X-ray morphology of A2029 to be among the most regular in the sample. Though it contains a wide angle tail (WAT) radio source, there are no coincident X-ray ``holes'', such as those found in the clusters Hydra A \\citep{nul02}, and Perseus \\citep{fab02}. There is no optical or X-ray spectroscopic evidence for a cooling flow, though the X-ray temperature drops to $2-3$keV in the central $5\\arcsec$ (7\\hkpc). Thus, this system is almost uniquely well-suited for analysis of its mass distribution since the hydrostatic equilibrium assumption should apply with high accuracy.  In the current paper we present estimates of the gas density and temperature profiles (\\S \\ref{sec_obs}), the total mass and the DM density profile (\\S \\ref{sec_mtot}), as well as the gas mass (\\S \\ref{sec_gas}). In \\S \\ref{sec_dis} we make a comparison with the stellar mass profile. We discuss the implications of our analysis and present our conclusions in \\S \\ref{sec_conc}. Throughout this paper, we assume a cosmology of H$_0=70$~$h_{70}$~km~s$^{-1}$~Mpc$^{-1}$, $\\Omega_{m}=0.3$, and $\\Lambda=0.7$, implying a luminosity distance to A2029 of 347~\\hmpc{} and an angular scale of 1.45 kpc~arcsec$^{-1}$. ",
        "conclusions": "}  We have analyzed high spatial resolution \\chandra{} data of the A2029 cluster of galaxies, obtaining well-constrained ICM gas density and temperature profiles on scales of $0.001-0.1\\rvir$ ($\\approx3-260$\\hkpc). Fitting smooth functions to these profiles, we obtain mass profiles for A2029, measuring the shape of the total mass profile at unprecedented accuracy to a very small fraction of the virial radius. Our results are insensitive to most details of the data reduction, but are closely tied to the well-constrained temperature and density distributions.  We find that the shape of the inferred dark matter density at $<0.1\\rvir$ is consistent with the NFW parameterization of CDM halos, but apparently incompatible with that of M99, though we note that individual objects may be expected to show significant scatter from a mean DM halo profile \\citep{bul01}.    The consistency of the NFW model with the mass profile all the way down to $<0.01\\rvir$ indicates that there is no need to modify the standard CDM paradigm to fit the DM distribution in this cluster, consistent with X-ray observations of other clusters at larger radii (see \\S \\ref{subsec_dm}). This result contrasts with the strong deviations from the CDM predictions observed in the rotation curves of low surface brightness galaxies \\citep[e.g.,][]{swa00}, and dwarf galaxies \\citep[e.g.,][]{moo99} which inspired the self-interacting DM model \\citep{spe00} to explain the relatively flat core density distributions in these galaxies.  Hence in light of the good agreement with the NFW profile in clusters, the deviations observed on small galaxy scales do not seem to imply a fundamental problem with the general CDM paradigm. Instead, it is likely that the numerical simulations do not currently account properly for the effects of feedback processes on the formation and evolution of small halos.  Our analysis suggests that A2029 is dominated by a single mass (i.e., DM) component at all measured scales below $0.1\\rvir$, or that any transition from a stellar mass dominated component in the cD and a DM component is quite gradual. We also observe a rising gas fraction from $<3\\%$ to $>14\\%$ in A2029, obtaining an upper limit to $\\Omega_m \\leq 0.29\\pm0.03 h_{70}^{-1/2}$, consistent with other current studies."
    },
    "0209/astro-ph0209536.txt": {
        "abstract": "{ We study the internal dynamics of the rich galaxy cluster ABGC\\,209 on the basis of new spectroscopic and photometric data. The distribution in redshift shows that ABCG\\,209 is a well isolated peak of 112 detected member galaxies at $\\mathrm{z=0.209}$, characterised by a high value of the line--of--sight velocity dispersion, $\\mathrm{\\sigma_v=1250}$--$1400$ \\kss, on the whole observed area (1 \\h from the cluster center), that leads to a virial mass of $\\mathrm{M=1.6}$--$2.2\\times 10^{15}$\\msun within the virial radius, assuming the dynamical equilibrium. The presence of a velocity gradient in the velocity field, the elongation in the spatial distribution of the colour--selected likely cluster members, the elongation of the X--ray contour levels in the Chandra image, and the elongation of cD galaxy show that ABCG\\,209 is characterised by a preferential NW--SE direction. We also find a significant deviation of the velocity distribution from a Gaussian, and relevant evidence of substructure and dynamical segregation. All these facts show that ABCG\\,209 is a strongly evolving cluster, possibly in an advanced phase of merging. ",
        "introduction": "The investigation of clusters of galaxies offers a rare possibility to link many aspects of astrophysics and cosmology and, in particular, to understand the processes that lead to the formation of structures.  In hierarchical clustering cosmological scenarios galaxy clusters form from the accretions of subunits. Numerical simulations show that clusters form preferentially through anisotropic accretion of subclusters along filaments (West et al. \\cite{wes91}; Katz \\& White \\cite{kat93}; Cen \\& Ostriker \\cite{cen94}; Colberg et al \\cite{col98}, \\cite{col99}). The signature of this anisotropic cluster formation is the cluster elongation along the main accretion filaments (e.g., Roettiger et al. \\cite{roe97}). Therefore the knowledge of the properties of galaxy clusters, plays an important role in the study of large--scale structure (LSS) formation and in constraining cosmological models.  By studying the structure of galaxy clusters it is possible to discriminate between different cosmological models (e.g., Richstone et al. \\cite{ric92}; Lacey \\& Cole \\cite{lac94}; Thomas et al. \\cite{tho98}).  In fact, in a low--density universe the clustering tends to freeze at (z+1) = $\\mathrm{\\Omega_{M}^{-1}}$, while in a high--density universe it continues to grow to the present day. This implies that clusters in a low--density universe are expected to be dynamically more relaxed and to have less subsystems, called substructures.  On small scales, clusters appear as complex systems involving a variety of interacting components (galaxies, X--ray emitting gas, dark matter). A large fraction of clusters (30\\%-40\\%) contain substructures, as shown by optical and X--ray studies (e.g.,  Baier \\& Ziener \\cite{bai77}; Geller \\& Beers \\cite{gel82}; Girardi et al. \\cite{gir97}; Jones \\& Forman \\cite{jon99}) and by recent results coming from the gravitational lensing effect (e.g.,  Athreya et al. \\cite{ath02}; Dahle et al. \\cite{dah02}), suggesting that they are still in the dynamical relaxation phase. Indeed, there is growing evidence that these subsystems arise from merging of groups and/or clusters (cf. Buote \\cite{buo02}; and Girardi \\& Biviano \\cite{gir02} for reviews).  Very recently, it was also suggested that the presence of radio halos and relics in clusters is indicative of a cluster merger. Merger shocks, with velocities larger than 10$^3$ km s$^{-1}$, convert a fraction of the shock energy into acceleration of pre--existing relativistic particles and provide the large amount of energy necessary for magnetic field amplification (Feretti \\cite{fer00}). This mechanism has been proposed to explain the radio halos and relics in clusters (Brunetti et al. \\cite{bru01}).  The properties of the brightest cluster members (BCMs) are related to the cluster merger.  Most BCMs are located very close to the center of the parent cluster. In many cases the major axis of the BCM is aligned along the major axis of the cluster and of the surrounding LSS (e.g., Binggeli \\cite{bin82}; Dantas et al. \\cite{dan97}; Durret et al. \\cite{dur98}). These properties can be explained if BCMs form by coalescence of the central brightest galaxies of the merging subclusters (Johnstone et al. \\cite{joh91}).  The optical spectroscopy of member galaxies is the most powerful tool to investigate the dynamics of cluster mergers, since it provides direct information on the cluster velocity field.  However this is often an ardue investigation due to the limited number of galaxies usually available to trace the internal cluster velocity.  To date, at medium and high redshifts (z $\\gtrsim$ 0.2), only few clusters are really well sampled in the velocity space (with $>$ 100 members; e.g., Carlberg et al \\cite{car96}; Czoske et al. \\cite{czo02}).  In order to gain insight into the physics of the cluster formation, we carried out a spectroscopic and photometric study of the cluster ABCG\\,209, at z$\\sim$0.2 (Kristian et al. \\cite{kri78}; Wilkinson \\& Oke \\cite{wil78}; Fetisova \\cite{fet81}), which is a rich, very X--ray luminous and hot cluster (richness class $\\mathrm{R=3}$, Abell et al. \\cite{abe89}; $\\mathrm{L_X(0.1-2.4\\ keV) \\sim 14\\;h_{50}^{-2}\\;10^{44}}$ erg $\\mathrm{s^{-1}}$, Ebeling et al. \\cite{ebe96}; $\\mathrm{T_X\\sim10}$ keV, Rizza et al. \\cite{riz98} ) . The first evidence for its complex dynamical status came from the significant irregularity in the X--ray emission with a trimodal peak (Rizza et al. \\cite{riz98}).  Moreover, Giovannini et al. (\\cite{gio99}) have recently found evidence for the presence of a possible extended Radio--emission.  The paper is organised as follows. In Sect.~2 we present the new spectroscopic data and the data reduction. The derivation of the redshifts is presented in Sect.~3. In Sect.~4 we analyse the dynamics of the cluster, and in Sect.~5 we complete the dynamical analysis with the information coming either from optical imaging or from X--ray data. In Sect.~6 we discuss the results in terms of two pictures of the dynamical status of ABCG\\,209. Finally, a summary of the main results is given in Sect.~7.  Throughout the paper, we assume a flat cosmology with $\\mathrm{\\Omega_M=0.3}$ and $\\mathrm{\\Omega_{\\Lambda}=0.7}$.  For the sake of simplicity in rescaling, we adopt a Hubble constant of 100 h \\ks Mpc$^{-1}$. In this assumption, 1 arcmin corresponds to $\\sim$ 0.144 Mpc.  Unless otherwise stated, we give errors at the 68\\% confidence level (hereafter c.l.). ",
        "conclusions": "In order to study the internal dynamics of the rich galaxy cluster ABCG\\,209, we obtained spectra for 159 objects in the cluster region based on MOS observations carried out at the ESO New Technology Telescope.  Out of these spectra, we analysed 119 galaxies: 112 turn out to be cluster members, 1 is foreground and 6 are background galaxies.  ABCG\\,209 appears as a well isolated peak in the redshift distribution centered at $z=0.209$, characterized by a very high value of the LOS velocity dispersion: $\\mathrm{\\sigma_v=1250}$--$1400$ \\kss, that results in a virial mass of $\\mathrm{M=1.6}$--$2.2\\times 10^{15}$\\msun within R$\\mathrm{_{vir}}$.  The analysis of the velocity dispersion profile show that such high value of $\\mathrm{\\sigma_v}$ is already reached in the central cluster region ($<0.2$--0.3 \\hh).  The main results of the present study may be summarised as follows.  \\begin{itemize}  \\item ABCG\\,209 is characterised by a preferential SE--NW direction as indicated by: a) the presence of a velocity gradient in the velocity field; b) the elongation in the spatial distribution of colour--selected likely cluster members; c) the elongation of the X--ray contour levels in the Chandra image; d) the elongation of the cD galaxy.  \\item We find significant deviations of velocity distribution from Gaussian.  \\item Red and blue members are spatially and kinematically segregated.  \\item There is significant evidence of substructure, as shown by the Dressler \\& Schectman test.  \\item The two--dimensional distribution of the colour--selected likely members shows a strong luminosity segregation: bright galaxies $\\mathrm{R<19.5}$ are centered around the cD galaxy, while faint galaxies $\\mathrm{R>19.5}$ show some clumps. The main one, Eastern with respect to the cD galaxy, is well coincident with the secondary X--ray peak.  \\end{itemize}  This observational scenario suggests that ABCG\\,209 is presently undergoing strong dynamical evolution. Present results suggest the merging of two or more subclumps along the SE--NW direction in a plane which is not parallel to the plane of sky, but cannot discriminate between two alternative pictures.  The merging might be in a very early dynamical status, where clumps are still in the pre--merging phase, or in a more advanced status, where luminous galaxies trace the remnant of the core--halo structure of a pre--merging clump hosting the cD galaxy.   The connection between the dynamics and the properties of galaxy populations will be discussed in a forthcoming paper."
    },
    "0209/astro-ph0209033_arXiv.txt": {
        "abstract": "{ We present an analysis of ten cool stars (Algol, Capella, Procyon, $\\epsilon$~Eri, $\\alpha$~Cen A\\&B, UX~Ari, AD~Leo, YY~Gem, and HR\\,1099) observed with the Low Energy Transmission Grating Spectrometer (LETGS) on board the {\\it Chandra} X-ray Observatory. This sample contains all cools stars observed with the LETGS presently available to us with integration times sufficiently long to warrant a meaningful spectral analysis. Our sample comprises inactive, moderately active, and hyperactive stars and samples the bulk part of activity levels encountered in coronal X-ray sources. We use the LETGS spectra to carry out density and temperature diagnostics with an emphasis on the H-like and the He-like ions. We find a correlation between line flux ratios of the Ly$_\\alpha$ and He-like resonance lines with the mean X-ray surface flux. We determine densities using the He-like triplets. For active stars we find no significant deviations from the low-density limit for the ions of Ne, Mg, and Si, while the measured line ratios for the ions of C, N, and O do show evidence for departures from the low-density limit in the active stars, but not in the inactive stars. Best measurements can be made for the O\\,{\\sc vii} triplet where we find significant deviations from the low-density limit for the stars Algol, Procyon, YY~Gem, $\\epsilon$~Eri, and HR\\,1099. We discuss the influence of radiation fields on the interpretation of the He-like triplet line ratios in the low-Z ions, which is relevant for Algol, and the influence of dielectronic satellite lines, which is relevant for Procyon. For the active stars YY~Gem, $\\epsilon$~Eri, and HR\\,1099 the low f/i ratios can unambiguously be attributed to high densties in the range 1--3\\,10$^{10}$\\,cm$^{-3}$ at O\\,{\\sc vii} temperatures. We find our LETGS spectra to be an extremely useful tool for plasma diagnostics of stellar coronae. \\keywords {Atomic data -- Atomic processes -- Techniques: spectroscopic -- Stars: individual: Algol -- Stars: individual: Procyon -- Stars: individual: Capella -- Stars: individual: Alpha Cen -- Stars: individual: UX~Ari -- Stars: individual: Eps Eri -- Stars: individual: AD~Leo -- Stars: individual: HR1099 -- Stars: individual: YY~Gem -- stars: coronae --  stars: late-type -- stars: activity -- X-rays: stars} } ",
        "introduction": "\\label{intro}  The coronal X-ray emission from the Sun is spatially correlated with photospheric regions exhibiting magnetic field concentrations. Therefore spatially unresolved X-ray emission from other solar-like stars is commonly used as a tracer for stellar magnetic activity. The specific advantage of X-ray measurements is that any stellar X-ray emission exclusively comes from the corona, and unlike other activity tracers, is not affected by photospheric emissions, rapid rotation, turbulent broadening etc. X-ray observations of stars carried out with the {\\it Einstein Observatory} (cf. Vaiana et al. \\cite{vai81}) and with ROSAT (Schmitt \\cite{schm97}) revealed the ubiquity of stellar X-ray emission among stars placed in the Hertzsprung-Russell diagram. The extensive ROSAT surveys showed that essentially all late-type solar-like stars with outer convection are surrounded by hot (T $\\ge$ 1\\,MK) coronae (Schmitt \\cite{schm97}). The X-ray luminosity of a given type of star can vary over four orders of magnitude over the sample, and rotation appears to be the most important parameter characterizing the level of X-ray emission of cool stars (cf. Pallavicini et al. \\cite{pal81}). The solar corona as observed with modern X-ray and XUV telescopes on board YOHKOH, SOHO, or TRACE is found to be extremely structured, and even in the high angular resolution TRACE images there appears to be spatially unresolved fine structure. Yet spatially resolved X-ray observations of stellar coronae are currently not feasible. Information on the spatial structure of stellar coronae can in principle be derived from X-ray light curves of suitably chosen stars such as eclipsing binaries, yet the actual information derivable from such data is rather limited; a discussion of the pre-XMM and pre-{\\it Chandra} results together with the difficulties and limitations of light curve inversions is given by Schmitt (\\cite{schm98}).  How do stars manage to produce far more X-ray output than the Sun? One of the basic assumptions used in the interpretation of stellar X-ray emission is that the building blocks, the solar corona is composed of, are also those that make up the X-ray emission of stars. This assumption can only be tested by spectroscopic investigations. The {\\it Einstein Observatory} and ROSAT provided data with rather modest spectral resolution. A few measurements of selected stars were carried out with the transmission gratings available on board of the {\\it Einstein} and EXOSAT satellites. Higher spectral resolution information, albeit only at wavelengths longward of 90\\,\\AA, but not in the X-ray range, was provided by the EUVE spectrometers. The low resolution proportional counter spectra obtained with the {\\it Einstein} and ROSAT satellites were fitted with plasma emission models in order to derive plasma temperatures and emission measures (see, for example, Schmitt et al. \\cite{schm90}); later also elemental abundances were included as fit parameters (see, for example, Antunes et al. \\cite{ant94}). However, the previous measurements did not allow to measure densities $n_e$ and emission measures $EM$ independently, such that no emitting volumes $V$ could be derived from $EM= n_e^2V$. Therefore no information about loop sizes was accessible. With the new high resolution spectra obtained with the {\\it Chandra} and XMM-Newton grating spectrometers it is possible for the first time to measure individual lines in the X-ray range with reasonable effective areas over a wide bandpass for a larger sample of stars in the same fashion as X-ray emission lines from the solar corona have been obtained and analyzed for many years (e.g., Doyle \\cite{doyle80}, McKenzie \\& Landecker \\cite{mck82}, Gabriel et al. \\cite{gbf88}). Specifically, He-like triplets can be observed for a variety of elements (carbon, nitrogen, oxygen, neon, magnesium, and silicon) and stars. The theory of He-like triplets can be tested in active and inactive stars and density information can be derived; this information can be supplemented by other line ratios, e.g., Fe\\,{\\sc xxi} lines (measured with the LETGS), which also yield density constraints. The density-sensitive line ratios are the missing link relating emission measure and volumes. Further, emission measure weighted ``effective'' temperatures can be derived from suitable line ratios, in particular the ratios between the Ly$_{\\alpha}$ and He-like resonance lines.  First results from XMM-RGS measurements were presented by, e.g., Audard et al. (\\cite{aud1}, \\cite{aud2}) and G\\\"udel et al. (\\cite{gued1}, \\cite{gued2}) with special focus on global modeling techniques to derive temperature distributions and abundances for the Castor system, Capella, AB~Dor, and HR\\,1099. Measurements with the MEG on board {\\it Chandra} were presented by, e.g., Canizares et al. (\\cite{caniz00}), Ayres et al. (\\cite{ayres01}), and Phillips et al. (\\cite{phil01}). The MEG and HEG provide the highest resolution at energies $>$ 1 keV and Brickhouse et al. (\\cite{brick01}) report the measurements of orbit related line shifts in 44~Boo, thus eventually opening up the road to X-Ray Doppler imaging of stellar coronae. Plasma diagnostics carried out with LETGS spectra were presented by, e.g., Mewe et al. (\\cite{mewe_cap}) on Capella, Ness et al. (\\cite{ness_cap}) on Capella and Procyon, and Ness et al. (\\cite{ness_alg}) on Algol; these authors use line ratios in order to derive plasma densities and temperatures.\\\\ The purpose of this paper is to summarize results from recent LETGS measurements of a set of ten cool stars with a special focus on the density diagnostics with He-like triplets. Albeit some results on He-like triplets were published earlier (e.g., Ness et al. \\cite{ness_cap}, Mewe et al. \\cite{mewe_cap}, Ness et al. \\cite{ness_alg}, Stelzer et al. \\cite{stelz02}, Raassen et al. \\cite{raa}, Audard et al. \\cite{aud3}), we will use these already published results for comparison with our new results on UX~Ari, $\\epsilon$~Eri, $\\alpha$~Cen A and B, and AD~Leo, which are presented here for the first time. \\\\ The specific information that can be derived from He-like line ratios is the plasma density. The theory of such triplets is described extensively in the literature; we refer to Gabriel \\& Jordan (\\cite{gj69}), who developed the theory, and to Blumenthal et al. (\\cite{bl72}), Mewe \\& Schrijver (\\cite{ms78}), Pradhan et al. (\\cite{prad81}), Pradhan \\& Shull (\\cite{ps81}), and recently Porquet et al. (\\cite{por01}), who refined and revised the theory. In this paper we use the relation \\begin{equation} \\frac{\\rm f}{\\rm i} \\ = \\frac {R_{\\rm 0}} {1 + \\phi/\\phi_c + n_e/N_c}\\,, \\end{equation} denoting the forbidden line with f and the intercombination line with i with the low density limit $R_{\\rm 0}$, the radiation term $\\phi/\\phi_c$ (the values derived in Sect.~\\ref{radfield} are listed in Tables~\\ref{restab1} to \\ref{restab3}), and the electron density $n_e$; $N_c$ is the so-called critical density, which leads to an f/i-ratio just in between the high- and low-density limits. Thus, given an X-ray measurement of the f/i-ratio and knowledge of $R_{\\rm 0}$, $N_c$, and $\\phi/\\phi_c$ from other sources, the plasma electron density $n_e$ can be inferred.\\\\ Our paper will be structured as follows: We first give a short description of the instrument with its capabilities and provide an overview of our sample of stars. We then introduce the new spectra and give a detailed description of the analysis presenting the measured line counts and describing the methods applied to analyze the obtained results. Our results of total X-ray luminosities, temperatures, and He-like densities for the ions Si\\,{\\sc xiii}, Mg\\,{\\sc xi}, Ne\\,{\\sc ix}, O\\,{\\sc vii}, N\\,{\\sc vi}, and C\\,{\\sc v} will be presented in Sect.~\\ref{result}. Our conclusions are given in Sect.~\\ref{conclusio}. ",
        "conclusions": "\\label{conclusio}  With the data from the new generation of X-ray telescopes we can measure He-like triplets for a wide range of temperatures and for other stars than the Sun. In our conclusions we consider only the cases as significant where deviations from the low-density limit are at least 2\\,$\\sigma$. We point out that in all our stars the intercombination line is the weakest line (except for Algol), such that under exposed spectra suffer the most from large errors in the intercombination line leading to large uncertainties in the f/i ratios.  The theory of He-like triplets makes specific predictions for the ratio (f+i)/r, which for collisional dominated plasmas is near unity with some temperature dependence and for the density-dependent f/i-ratio. For the ``hot'' ions (neon, magnesium, silicon) the measured (f+i)/r-ratios are near but below unity, the same applies for the ``cool'' ions (carbon, nitrogen, and oxygen) for the ``hot'', i.e., active stars. For the less active stars (i.e., Procyon, $\\alpha$ Cen A and B) the (f+i)/r-ratios are somewhat larger, so we conclude that our measurements are consistent with theoretical expectations, but unfortunately the accuracy of the measurements is such that temperature determinations with the help of the (f+i)/r-ratios are in many cases not possible.  As to densities, an inspection of Table~\\ref{restab3} shows that for the elements neon, magnesium, and silicon no particularly convincing examples of density sensitive line ratios were found. Most measurements are consistent with the low-density limit and with the exception of Capella all ``deviations'' from the low-density limit are found only in one element at a significance between two and three $\\sigma$. We thus conclude that for none of our sample stars the magnesium and silicon He-like triplets deviate from the low-density limit. For Capella we derive a neon f/i-ratio of 2.08 $\\pm$ 0.14, significantly below the low-density limit. Interestingly, this value is confirmed by higher resolution measurements with the {\\it Chandra} HEG (cf. Ness et al.~\\cite{nessb02}).  Clearly, the best (in terms of signal-to-noise and spectral resolution) measurements are available for oxygen. For all ten stars the He-like resonance and forbidden lines were detected. The intercombination line, which is the weakest triplet line in a plasma with low to intermediate density, was detected for nine stars. In four cases (i.e., Algol, $\\epsilon$~Eri, YY~Gem, and HR\\,1099) do we find significant deviations from the expected low-density limit, in one case (Procyon) the measured f/i-ratio is within 2-3 $\\sigma$ from the low density limit. For the stars $\\epsilon$~Eri, YY~Gem, and HR\\,1099 we find f/i-ratios between 2.22 - 2.96, but the errors are so large that in principle those stars could have the same f/i-ratio. The by far lowest f/i-ratio is found for Algol, but in that particular case radiation fields can be important even for the oxygen triplet lines. In four cases ($\\alpha$\\,Cen~A and B, UX~Ari, and AD~Leo) the measurement errors are so large that no meaningful conclusions can be drawn, while in one case (Capella) our measurement agrees perfectly with the low-density limit. This low-density for Capella is consistent with the XMM-Newton RGS results, but inconsistent with the measurements obtained with the {\\it Chandra} MEG (cf., e.g., Canizares et al.~\\cite{caniz00}, Ayres et al.~\\cite{ayres01}, Phillips et al.~\\cite{phil01}).  For our analysis we attribute all measured line counts to the expected He-like lines, which may be contaminated by coincidental lines or by dielectronic satellite lines. Such effects can be checked in each individual case. Specfically in the case of Procyon inspection of the MEKAL tables (Mewe, Kaastra, \\& Liedahl \\cite{mewe95}) shows that a contribution from dielectronic satellite lines originating from O\\,{\\sc vi} to the measured line flux of the intercombination line of O\\,{\\sc vii} on the level of 10\\% can easily be accomplished. In this case the measured f/i ratio can well be consistent with the low-density limit. For the hotter coronae our measurement errors are larger and further the contribution from dielectronic satellite lines is weaker. Therefore low f/i ratios cannot be attributed to contamination of the intercombination line.  What implications can we deduce for the sizes of the underlying coronae? For the sake of argument, let us focus on the LETGS measurements of YY~Gem. The available line ratios between Ly$_{\\alpha}$ and He-like resonance lines can be described by a power-law differential emission measure distribution of the form $n^2_e \\ \\frac{dV} {dT} dT = \\xi(T) dT = \\frac {EM_0} {T_0} (T/T_0)^{\\alpha} dT.$ With this functional form we find acceptable fits for the values $T_0$ = 17\\,MK and $\\alpha$ = 0. This results in values of 0.84, 1.69, 2.71, and 3.58 for the line ratios between the Ly$_{\\alpha}$ and He-like resonance lines for Si, Mg, Ne, and O, respectively, which compare well (except for Ne) with our measurements of 0.99 $\\pm$ 0.22, 1.46 $\\pm$ 0.55, 2.0 $\\pm$ 0.19, and 3.67 $\\pm$ 0.32, respectively. Calculating the fluxes of the resonance, forbidden, and intercombination lines of the oxygen triplet with this emission measure distribution, we find for (f+i)/r a value of 0.89, which must be compared to the measured value of 0.88 $\\pm$ 0.14. The calculated f/i-ratio depends on the assumed pressure. If we assume a constant pressure scenario, we find an almost linear dependence of pressure and f/i-ratio, with the measured value of 2.29 corresponding to 17\\,dyne\\,cm$^{-2}$. Unfortunately, the large errors in f/i make pressures as low as 10\\,dyne\\,cm$^{-2}$ and high as 28\\,dyne\\,cm$^{-2}$ also possible; at any rate, we can deduce a pressure of 20 $\\pm$ 10\\,dyne\\,cm$^{-2}$ in the corona(e) of YY~Gem. We note that this value is independent of the precise emission measure distribution adopted. Fixing then the normalization $EM_0$ from the requirement to match the observed He-like r flux in oxygen, we can compute the pressure dependent total coronal volume from the formula $V_{tot} = \\frac {4 k^2 EM_0} {p^2 T_0^2 (\\alpha +3)} $ and find $V_{tot} = 4^{+7}_{-2} \\times 10^{32}$\\,cm$^{3}$; clearly, because of the quadratic dependence of the coronal volume on pressure the errors are still substantial. Nevertheless, the nominal value of $V_{tot} = 4 \\times 10^{32}$\\,cm$^{3}$ leads (for an assumed stellar radius of 0.5 R$_{\\odot}$, two identical stars and filling factor of unity) to a height of $\\approx 10^{10}$\\,cm, i.e., about one third of the stellar radius and hence much larger than typical solar coronal loops.  Using the line ratio f/i of the He-like triplets of oxygen and nitrogen one recognizes a trend, indicating lower f/i-ratios and thus lower densities for inactive stars. This conclusion has, however, to be treated with some care. The available data support it only for main sequence stars, i.e., for YY~Gem, AD~Leo, and $\\epsilon$~Eri do we measure f/i-ratios significantly below the low-density limit. For Algol the measured low f/i-ratio does not necessarily imply high densities, in principle it can also be attributed to a special geometrical configuration immersing its corona in the radiation field of its B-type companion (cf. Sect.~\\ref{radfield}). For UX~Ari the data are consistent with the low density limit, but because of the large errors significant f/i ratios can also not be ruled out. Only for Capella the measurement errors are small enough that we can state with confidence that the measured f/i-ratio places Capella very close to the low-density limit. For Capella one can definitely exclude densities as high as 10$^{13}$\\,cm$^{-3}$, which were reported by Dupree et al. (\\cite{dup93}) from EUVE measurements of Fe\\,{\\sc xxi} lines, which are, however, formed at a much higher temperature than O\\,{\\sc vii}. The analysis of the long wavelength portion of the {\\it Chandra} LETGS spectrum by Mewe et al. (\\cite{mewe_cap}) yielded only upper limits of $\\approx$ 2 $\\times 10^{12}$\\,cm$^{-3}$ for Fe\\,{\\sc xxi}, and similarly for their He-like line ratios of Mg\\,{\\sc xi} and Si\\,{\\sc xiii} no significant densities could be derived. The densities derived from N\\,{\\sc vi} show the same trend as those derived from oxygen. Again we emphasize that the peak of the emission measure distribution for the most active stars lies above the peak formation temperatures of those specific ions. For the lowest temperature ion, C\\,{\\sc v}, the f/i-ratios for the hotter stars are dominated by radiation effects; also higher order line blending limits the accuracy with which the C\\,{\\sc v}-triplet can be determined. In all cases where the C\\,{\\sc v}-triplet could be detected f/i-ratios of approximately unity were found. The higher Z ions Ne to Si are more difficult to study with the LETGS, since they are not fully resolved and line blending occurs especially for Ne\\,{\\sc ix}. These line blends can still be modelled and thus de-blended, but our results should be confirmed with higher resolution measurements available with, e.g., the MEG. The mean X-ray surface flux appears to be a useful parameter segregating stars into different levels of activity; another useful parameter appears to be the ratio between Ly$_{\\alpha}$ line flux and the He-like triplet resonance line flux, especially for oxygen. A grouping of stars is apparent when mean X-ray surface flux and the ratio between Ly$_{\\alpha}$ and He-like triplet flux are considered. Larger data samples are required to assess any physical significance of such groupings and open the road to an X-ray ``HR-diagram''."
    },
    "0209/astro-ph0209519_arXiv.txt": {
        "abstract": "Using a very deep observation with \\hst/STIS, we have searched for an optical counterpart to the nearby radio-quiet isolated neutron star \\rxj\\ (RBS~1223).  We have identified a single object in the 90\\% \\chandra\\ error circle that we believe to be the optical counterpart. This object has $m_{\\rm 50CCD}=28.56\\pm0.13$~mag, which translates approximately to an unabsorbed flux of $F_{\\lambda}=\\expnt{(1.7 \\pm 0.3)}{-20}\\mbox{ ergs s}^{-1}\\mbox{ cm}^{-2}\\mbox{ \\AA}^{-1}$ at 5150~\\AA\\ or an X-ray-to-optical flux ratio of $\\lfxo=4.9$.  This flux is a factor of $\\approx 5$ above the extrapolation of the black-body fit to the X-ray spectrum, consistent with the optical spectra of other isolated neutron stars.  Without color information we cannot conclude that this source is indeed the counterpart of \\rxj.  If not, then the counterpart must have $m_{\\rm 50CCD} > 29.6$~mag, corresponding to a flux that is barely consistent with the extrapolation of the black-body fit to the X-ray spectrum. ",
        "introduction": "\\label{sec:intro} Neutron stars have been regarded as natural laboratories for matter denser than can be obtained by heavy-ion accelerators.  The basic physics is summarized by the mass and radius, with larger radii for a given mass favoring stiffer equations-of-state (EOS; \\citealt{lp00}). It is against this backdrop that one recognizes that one of the major outcomes of the all-sky survey undertaken by the X-ray satellite \\textit{ROSAT} was the systematic identification of the nearest neutron stars (see reviews by \\citealt{motch01} and \\citealt{ttzc00}).  \\rxj\\ (also known as RBS~1223) was identified as a candidate neutron star from the \\textit{ROSAT} Bright Survey by \\citet{shs+99} on the basis of its soft X-ray spectrum (blackbody with $kT\\approx 100$~eV), constant X-ray flux, and lack of  optical counterpart. It now joins six other similar objects (RX~J1856.5$-$3754, RX~J0720.4$-$3125, RX~J1605.3+3249, RX~J2143.0 +0654, RX~J0806.4$-$4123, and RX~J0420.0$-$5022; \\citealt*{wwn96}; \\citealt{hmb+97,mhz+99}; \\citealt*{hpm99}; \\citealt{zct+01}) and three previously known pulsars (Geminga, PSR~B0656+14, and PSR~B1055$-$52) in the sample of nearby $10^{6}$-yr neutron stars detected by the \\rosat\\ Bright Survey.  Of these objects, the five brightest (in terms of soft X-ray count-rate)  have been well studied. PSR~0656+14 and PSR~B1055$-$52 are well known radio pulsars, not particularly remarkable in any other way. Geminga, first identified via is $\\gamma$-ray emission (and thereby dramatically demonstrating that radio pulsars can lose a large fraction of their energy via $\\gamma$-rays) is now generally considered to be an ordinary pulsar whose radio beam we happen to miss.  In contrast, \\rxjw\\ and \\rxjk\\ are mysterious. Both sources have (as expected) faint, blue, optical counterparts \\citep{wm97,kvk98}, with X-ray-to-optical flux ratios of $\\lfxo\\sim 5$. \\rxjw\\ shows no significant pulsations \\citep*{rgs02} and despite significant investment of {\\em Chandra} time, the X-ray spectrum is featureless \\citep{dmd+02}. There is no evidence for any non-thermal emission \\citep{vkk01}.  Conventional models for this source include a weakly-magnetized cooling neutron star \\citep{vkk01b} or an off-beam radio pulsar \\citep[like Geminga but without the $\\gamma$-ray emission;][]{br02}.  In contrast, \\rxjk\\ shows 8.4-s pulsations. It too exhibits a featureless X-ray spectrum (largely thermal; \\citealt{pmm+01}).  Again, conventional possibilities include an off-beam radio pulsar but the long period would require that the neutron star was born with either an unusually long period or an unusually large magnetic field \\citep{kkvkm02,zhc+02}.  Thus the five brightest (in soft X-rays) and presumably the nearest neutron stars show a stunning diversity. Our understanding of the nature of two (or perhaps even three) of these sources is quite incomplete.  In this {\\it Letter}, we  re-determine the position of \\rxj\\ from archival \\chandra\\ analysis. We present radio observations of \\rxj, and we then discuss very deep optical observations aimed at detecting its optical counterpart. This source exhibits long-period pulsations with $P=5.16$~s. However, unlike \\rxjk\\, a large period derivative has been measured \\citep{hhss02}.  If this is ascribed to magentic braking then the implied dipole field strength is $B \\gsim 10^{14}$~G, and \\rxj\\ is a magnetar. \\citet{kvk98} and \\citet{hk98} advocated the magnetar model for nearby long period pulsators because magnetars have an additional source of heat (their magnetic fields) and thus are warmer than ordinary neutron stars for a longer duration. ",
        "conclusions": "In the following, we use the results of the spectroscopic fits of \\citet{hhss02} to the \\chandra\\ data.  Specifically, we take $N_{H}=\\expnt{(2.4\\pm 1.1)}{20}\\mbox{ cm}^{-2}$, $kT=91\\pm 1$~eV and $R=(6.5\\pm 0.3) d_{700}$~km, where the normalization comes from the \\chandra\\ count-rate. This spectrum implies an unabsorbed flux of $\\expnt{(3.5\\pm 0.3)}{-21}\\mbox{ ergs s}^{-1}\\mbox{ cm}^{-2}\\mbox{ \\AA}^{-1}$ at 5150~\\AA.    \\begin{figure}[t] \\plotone{figure2.ps} \\caption{\\hst/STIS image of the field around \\rxj.  The image is $\\approx 15\\arcsec$ on a side, with North up and East to the left. The $1\\farcs0$-radius \\chandra\\ error circle is shown.  Source X, the likely counterpart of \\rxj, and the unrelated sources A and B, are also indicated.  Source A is extended.} \\label{fig:stiszoom} \\end{figure}    There is only one optical source inside the \\chandra\\ error circle.  This source, marked X in Figure~\\ref{fig:stiszoom}, is a possible counterpart to \\rxj.  There are no other potential counterparts visible in Figure~\\ref{fig:stiszoom}, the next closest unresolved source being $\\approx 4\\arcsec$ from the \\chandra\\ position (source B in Figure~\\ref{fig:stiszoom}).  Without color information, it is difficult to accurately photometer the 50CCD data.  This is because its wide bandwidth makes the aperture corrections and zero-point fluxes color dependent, leading to uncertainties of greater than a factor of 2 for the flux coming from stars ranging from type M to type O.  In what follows, we follow the analysis of \\citet{kkvkf+02} for \\rxjk.  We assumed that X is the counterpart and therefore has blue colors (similar to \\rxjw\\ and \\rxjk; \\citealt{vkk01,kvk98}).  Then we used the bluest of the available aperture corrections: $0.183$~mag at $0\\farcs254$ radius (T.~Brown 2002, personal communication).  This correction is for a star with $B-V=-0.09$~mag, compared with an expected $B-V=-0.3$~mag for \\rxj, and is therefore not quite right.  However, the scattered light that contributes to the color-dependence of the STIS aperture corrections is predominantly red.  For blue sources, the dependence of the correction on color is relatively small: for a star with $B-V=0.05$~mag, the correction changes by about 0.01~mag from that for a source with $B-V=-0.09$~mag. So the aperture correction used here should be reasonably appropriate, and to account for any remaining differences we have added a 0.02~mag systematic uncertainty into the photometry for \\rxj.  With this correction, we find a magnitude of $m=28.56 \\pm 0.13$~mag for X at infinite aperture.  The 3-$\\sigma$ limiting magnitude is $\\approx 29.6$~mag.  These magnitudes are in the STMAG system, where $F_{\\lambda} = 10^{-(m+21.1)/2.5}\\mbox{ ergs s}^{-1}\\mbox{ cm}^{-2}\\mbox{ \\AA}^{-1}$.  Assuming a spectrum similar to a Rayleigh-Jeans tail, this relation holds at $\\lambda\\approx 5148$~\\AA\\ (this is the wavelength at which a Rayleigh-Jeans spectrum has the same flux as a flat spectrum that produces the same number of counts in the 50CCD band; see Appendix~A of \\citealt{vkk01}).   From this we find $F_{\\lambda}({\\rm X})=\\expnt{(1.4 \\pm 0.2)}{-20}\\mbox{ ergs s}^{-1}\\mbox{ cm}^{-2}\\mbox{ \\AA}^{-1}$ at 5148~\\AA.  We estimate $A_{V}=0.14\\pm 0.06$~mag, using the hydrogen column from above and the relation from \\citet{ps95}.  Again assuming a Rayleigh-Jeans spectrum, we convert $A_{V}$ to the extinction appropriate for the 50CCD bandpass (again see \\citealt{vkk01}) and find $A_{\\rm 50CCD}=0.22\\pm 0.09$~mag.  This gives us an unabsorbed flux of $\\expnt{(1.7\\pm 0.3)}{-20}\\mbox{ ergs s}^{-1}\\mbox{ cm}^{-2}\\mbox{ \\AA}^{-1}$.  The optical flux of X is then a factor of $\\approx 5$ higher than the extrapolation of the X-ray blackbody of \\rxj, smaller than the value of 16 found for \\rxjw\\ \\citep{vkk01}, but very similar to the values found for \\rxjk\\ and PSR~B0656+14 \\citep{kkvkf+02,kpz+01} .  Likewise, the unabsorbed X-ray-to-optical flux ratio is $\\lfxo=4.9$ (where the X-ray flux has been integrated over the entirety of the blackbody spectrum).  The similarity of these values to those for other isolated neutron stars suggests that source X is the optical counterpart of \\rxj.  While a blue color would assure us that X is the counterpart of \\rxj, without color information we cannot be certain.  Source X is very similar to the counterparts of \\rxjk\\ and \\rxjw, but it is possible that it is an unrelated source and that no counterpart was detected. If that is the case, then any counterpart would have $m_{\\rm 50CCD} > 29.6$~mag ($\\lfxo>5.3$), or an optical flux just consistent with the extrapolation of the X-ray black-body fit.   Aside from color information (difficult to obtain given its faintness), another good test for the nature of source X is proper motion.  Neutron stars have significantly higher proper motions than the stellar population, with velocities of $\\sim 100\\mbox{ km s}^{-1}$ typical for the general population of neutron stars \\citep*{acc02}. Such high velocities have been found for the local neutron star population as well \\citep[e.g.,][]{mdlc00,wal01}.  Assuming a velocity of $100\\mbox{ km s}^{-1}$, the proper motion of \\rxj\\ would be $30 d_{700}^{-1}\\mbox{ mas yr}^{-1}$.  While the absolute astrometry from the STIS image does not have this precision, we expect to be able to perform relative astrometry with at least $\\sim 20$~mas precision (the limiting factors are distortion correction and modeling of the point-spread-function, which is color-dependent), although this has not been tested for STIS.  If this is the case, then in the next few years proper motion of source X may be detectable, and if so source X would almost certainly be a neutron star (if X were instead a star, it would have to be many kpc away and would therefore have negligible proper motion and be out of the galaxy, given its galactic latitude of $b=83\\degr$).   In the $P$-$\\dot P$ plane, \\rxj\\ appears very similar to the Anomalous X-ray Pulsars (AXPs; \\citealt{m99}).  However, whether or not we have detected the counterpart of \\rxj, the X-ray-to-optical flux ratio is considerably higher than those found for AXPs (\\citealt*{hvkk00}; \\citealt{htvk+01,wc02}): for 4U~0142+61, $\\lfxo\\approx 4.1$ (where the X-ray flux is measured from 0.5--10~keV; \\citealt{jmcs02}).  The optical emission from AXPs, which has a non-thermal spectrum, is thought to arise from the magnetosphere.  Therefore the lack of an active magnetosphere would significantly decrease the optical flux. Scaling the non-thermal X-ray emission of 4U~0142+61 by the optical flux of \\rxj, we would predict an X-ray power-law for \\rxj\\ that would have been easily visible with \\chandra\\ ($\\expnt{2}{-3}\\mbox{ photons s}^{-1}\\mbox{ cm}^{-2}\\mbox{ keV}^{-1}$ at 1~keV).  As this power-law is not seen \\citep{hhss02}, it appears that despite its rapid spin-down \\rxj\\ does not have an active magnetosphere.  Without an active magnetosphere, the optical emission from \\rxj\\ would likely be similar to those of \\rxjw\\ and \\rxjk, suggesting that we have indeed found the counterpart to \\rxj."
    },
    "0209/astro-ph0209049_arXiv.txt": {
        "abstract": "The association between the Pencil nebula (RCW 37, NGC 2736), the Vela X-ray fragment D/D' and the recently discovered new X-ray supernova remnant (\\velasnr) in Vela is investigated. Recently published Chandra images of D/D' are compared with optical images of RCW 37 and confirm the close association of the two objects. New optical line profiles of RCW 37 from an extended slit position passing through this unusual optical nebula are presented. They reveal a partial velocity ellipse with expansion velocities of around $120~{\\rm km~s^{-1}}$. Various scenarios for the origin of the nebula are considered and the evidence of a link with \\velasnr\\ is reviewed. A funnel of gas similar to those in the Crab and DEM34a SNRs is not ruled out but a more plausible explanation may be that a `wavy sheet' is responsible. We suggest that \\velasnr\\ is located within the older larger Vela SNR and that some of the X-ray gas from \\velasnr\\ has collided with the dense H~{\\sc i} wall of the older remnant. This gives rise to the morphology and velocity structure of the optical emission and explains the unusual X-ray emission from this portion of the supernova remnant. If our hypothesis is correct, a distance prediction of $250\\pm 30~{\\rm pc}$ can be made, based on recent measurements of the distance to the old Vela SNR. This is at the lower end of the range of distances quoted in the literature would confirm unusual nature of this young nearby supernova remnant. ",
        "introduction": "\\velasnr\\ is a young nearby supernova remnant (SNR) recently discovered \\citep{aschenbach98,iyudin.et.al98} near the southeastern perimeter of the well known old Vela SNR. \\velasnr\\ has generated much interest since the distance and age could be as low as $200~{\\rm pc}$ and $700~{\\rm yr}$ respectively and thus it could have been generated by the nearest supernova explosion in recent human history.  The SNR was discovered in Rosat hard X-ray data (shown as contours in Fig.\\@~\\ref{esorass}) and at these energies has a shell-like morphology. There is no obvious optical counterpart to the main body of the SNR but \\citet{redman.et.al00} showed that a fragment of X-ray emission (labelled D/D' by \\citealt{aschenbach.et.al95}) coincides closely with a bright optical nebula, RCW 37. The X-ray fragment is located just beyond the main circular body of the remnant and is clearly visible in hard X-ray images. The main X-ray shell is not complete and there is a break in the emission in a direction coincident with that of the X-ray fragment and with RCW 37 (see Fig.\\@~\\ref{esorass}). \\citet{redman.et.al00} suggested that RCW 37 is physically associated with \\velasnr\\ and represents a venting of hot gas from the interior of the remnant to beyond the roughly circular shell as delimited in the X-ray.  RCW 37 (NGC 2736) was discovered in the 1840s by Sir John Herschel and is a bright optical nebula known to amateur astronomers as the Pencil Nebula (Fig.\\@~\\ref{eso}). The unusual, intricate morphology of the nebula and its large size and brightness make it surprising that there have been few studies of this object. Its apparent location at the dust-obscured eastern side of the six degree diameter old Vela SNR, away from the photogenic western filaments may be one reason. \\citet{blair.et.al95} used the Ultraviolet Spectrometer on the Voyager 2 spacecraft to investigate O~{\\sc vi} emission from RCW 37 and inferred that shock speeds of $160-300~{\\rm km~s^{-1}}$ are present. This is consistent with the limited line profiles obtained by \\citet{redman.et.al00} at the edge of the nebula. The morphology of the nebula is striking and led \\citet{redman.et.al00} to suggest that its funnel-like appearance could be taken to indicate that there is indeed a collimated flow of hot gas taking place from the remnant interior. Alternatively, a pre-existing cloud could be being shocked by escaping hot gas. The kinematics were not sufficient to establish clearly the dynamics of the nebula.  In this paper, new optical forbidden line profiles from RCW 37 from an extended slit position across the filamentary bulk of RCW 37 are presented in order to determine the structure and dynamics of the nebula. The kinematics are discussed and used to describe a model for the origin of RCW 37 that places it in the context of \\velasnr\\ and the main Vela SNR. It is argued that \\velasnr\\ is embedded within the older larger Vela SNR.  \\begin{figure} \\psfig{file=mc283fig1scantiny.eps,width=250pt,bbllx=78pt,bblly=213pt,bburx=517pt,bbury=629pt} \\caption{Contour map of \\velasnr\\ from the RASS hard X-ray discovery data of \\citet{aschenbach98} overlaid on ESO IIIaJ optical images. The Pencil nebula RCW 37 and X-ray fragment D/D' coincide to the upper left of the picture.} \\label{esorass} \\end{figure}  \\begin{figure} \\psfig{file=mc283fig2scantiny.eps,width=250pt,bbllx=64pt,bblly=201pt,bburx=530pt,bbury=641pt} \\caption{ESO image of RCW 37 with the five overlapping MES slit positions marked with a white line. The sloping thin dark line is a satellite trail.} \\label{eso} \\end{figure}  \\begin{figure} \\psfig{file=mc283fig3scantiny.eps,width=250pt,bbllx=65pt,bblly=202pt,bburx=531pt,bbury=640pt} \\caption{ESO image of RCW 37 with contours of the Chandra ACIS X-ray data of \\citet{plucinsky.et.al02} overlaid. The thick dotted line indicates the edge of the Chandra data. The sloping thin dark line is a satellite trail.} \\label{esochandra} \\end{figure} ",
        "conclusions": "New \\oiii\\ line profiles of RCW 37 have been presented that show the kinematics of this nebula for the first time. A partial velocity ellipse is discovered in the pv array of line profiles. The kinematics and morphology could suggest that the structure of RCW 37 is that of a thin wavy sheet of optical emission that overlaps itself towards the eastern edge and is undergoing a systematic expansion. The western edge curves towards the line of sight but does not appear to form a complete tube or funnel of emission. We compared this feature with those found towards other SNRs. The evidence that the RCW 37 optical nebula and associated X-ray feature, D/D' are in fact part of \\velasnr\\ has been discussed and we conclude that it is likely that this is the case. A simple explanation for the origin of the morphology of RCW 37 is that \\velasnr\\ has occured within the older, larger Vela SNR and that a portion of the supernova ejecta from \\velasnr\\ has impacted the pre-existing cold dense wall of the Vela SNR. The thin sheet of optical emission then traces out the inside edge of this shocked wall while the X-ray emission marks shock-heated gas. This model predicts that the distance to \\velasnr\\ will be similar to that of the main Vela SNR which has been recently measured to lie at $250\\pm 30~{\\rm pc}$."
    },
    "0209/astro-ph0209080_arXiv.txt": {
        "abstract": "I will summarize in four slides the $40$ years of development of the standard solar model that is used to predict solar neutrino fluxes and then describe the current uncertainties in the predictions. I will dispel the misconception that the p-p neutrino flux is determined by the solar luminosity and present a related formula that gives, in terms of the p-p and $^7$Be neutrino fluxes, the ratio of the rates of the two primary ways of terminating the p-p fusion chain. I will also attempt to explain why it took so long, about three and a half decades, to reach a consensus view that new physics is being learned from solar neutrino experiments. Finally, I close with a personal confession. \\vspace{1pc} ",
        "introduction": "\\label{sec:introduction}  I will follow in this text the content of my talk at Neutrino2002, which occurred  in  Munich, May 25-30, 2002.  I begin in Section~\\ref{sec:ray}, as I did in Munich, with a tribute to Ray Davis. In Section~\\ref{sec:development}, I present a concise history of the development of the standard solar model that is used today to predict solar neutrino fluxes. This section is based upon four slides that I used to summarize the development and is broken up into four subsections, each one of which describes what was written on one of the four slides. I describe in Section~\\ref{sec:uncertainties} the currently-estimated uncertainties in the solar neutrino predictions\\footnote{Where contemporary numbers are required in this review, I use the results from the BP00 solar model, ApJ 555 (2001) 990, astro-ph/0010346.}, a critical issue for existing and future solar neutrino experiments. I show in Section~\\ref{sec:ppflux} that the solar luminosity does not determine the p-p flux, although there are many claims in the literature that it does. I also present a formula that gives the ratio of the rates of the $^3$He-$^3$He and the $^3$He-$^4$He reactions as a function of the p-p and $^7$Be neutrino fluxes. These reactions are the principal terminating fusion reactions of the p-p chain. In Section~\\ref{sec:why},I give my explanation of why it took so long for physicists to reach a consensus that new particle physics was being learned from solar neutrino experiments. I close with a personal confession in Section~\\ref{sec:confession}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209563_arXiv.txt": {
        "abstract": "We use hydrodynamic cosmological simulations to predict correlations between \\lya\\ forest absorption and the galaxy distribution at redshift $z\\approx 3$.  The probability distribution function (PDF) of \\lya\\ flux decrements shifts systematically towards higher values in the vicinity of galaxies, reflecting the overdense environments in which these galaxies reside.  The predicted signal remains strong in spectra smoothed over $50-200\\kms$, allowing tests with moderate resolution quasar spectra.  The strong bias of high redshift galaxies towards high density regions imprints a clear signature on the flux PDF, but the predictions are not sensitive to galaxy baryon mass or star formation rate, and they are similar for galaxies and for dark matter halos.  The dependence of the flux PDF on galaxy proximity is sensitive to redshift determination errors, with rms errors of $150-300 \\kms$ substantially weakening the predicted trends.  On larger scales, the mean galaxy overdensity in a cube of 5 or $10\\hmpc$ (comoving) is strongly correlated with the mean \\lya\\ flux decrement on a line of sight through the cube center.  The slope of the correlation is $\\sim 3$ times steeper for galaxies than for dark matter as a result of galaxy bias.  The predicted large scale correlation is in qualitative agreement with recently reported observational results.  However, observations also show a drop in the average absorption in the immediate vicinity of galaxies, which our models do not predict even if we allow the galaxies or AGNs within them to be ionizing sources.  This decreased absorption could be a signature of galaxy feedback on the surrounding intergalactic medium, perhaps via galactic winds.  We find that a simplified ``wind'' model that eliminates neutral hydrogen in spheres around the galaxies can marginally explain the data.  However, because peculiar velocities allow gas at large distances to produce saturated absorption at the galaxy redshift, these winds (or any other feedback mechanism) must extend to comoving radii of $\\sim 1.5 \\hmpc$ to reproduce the observations.  We also discuss the possibility that extended \\lya\\ emission from the target galaxies ``fills in'' the expected \\lya\\ forest absorption at small angular separations. ",
        "introduction": "\\label{sec:intro}  The strong clustering of Lyman-break galaxies (LBGs) at $z\\approx 3$, comparable to that of present-day, optically selected galaxies, suggests that they are highly biased tracers of the underlying dark matter distribution \\citep{adelberger98,adelberger02}.  This bias appears to arise naturally in semi-analytic models and hydrodynamic numerical simulations \\citep[e.g.,][]{baugh98,governato98,katz99, kauffmann99,cen00,benson01,pearce01,yoshikawa01,weinberg02a}, which predict that the luminous members of the high redshift galaxy population reside in massive halos, which in turn reside in regions of high background density \\citep{kaiser84,mof96,mow96}.  However, the bias of Lyman-break galaxies is inferred by comparing their observed clustering to the predicted clustering of dark matter, which depends on the assumed cosmological model.  The \\lya\\ forest offers a tracer of structure whose relation to the underlying dark matter distribution appears to be well understood on theoretical grounds, probing the same redshifts as the LBG population. Correlations between LBGs and \\lya\\ forest absorption therefore offer a natural and potentially powerful probe of the relation between high redshift galaxies and the dark matter distribution, and perhaps for the influence of high redshift galaxies on the surrounding intergalactic medium (IGM).  In this paper we use smoothed particle hydrodynamics (SPH) simulations to predict the correlations between $z=3$ galaxies and \\lya\\ forest absorption.  Hydrodynamic simulations are ideal for this purpose, since they simultaneously predict the locations and properties of the galaxies \\citep[e.g.][]{katz99,nagamine01,weinberg02b} and the structure and ionization state of the intergalactic gas that produces the \\lya\\ forest \\citep[e.g.][]{cen94,zhang95,hernquist96,theuns98}. However, the basic expectations can be understood in simple terms, using the ``Fluctuating Gunn-Peterson Approximation'' \\citep{bi97,croft97,rauch97,weinberg98,croft98}, which provides a fairly accurate description of the \\lya\\ forest results from full hydrodynamic calculations.  In this approximation, the \\lya\\ flux decrement at wavelength $\\lambda$ is related to the overdensity at a redshift $z=\\lambda/\\lambda_{\\alpha,{\\rm rest}}-1$ along the line of sight by \\begin{equation} D ~=~ 1 - \\frac{F}{F_c} = 1-e^{-\\tau} ~=~ 1 - \\exp\\left[ -A (\\rho/\\bar{\\rho})^\\beta\\right] ~, \\label{eqn:fgpa} \\end{equation} with \\begin{equation} A = 0.694 \\left(\\frac{1+z}{4.0}\\right)^6 \\left(\\frac{\\Omega_b h^2}{0.02}\\right)^2 \\left(\\frac{T_0}{6000\\K}\\right)^{-0.7} \\left(\\frac{h}{0.65}\\right)^{-1} \\left(\\frac{H(z)/H_0}{5.12}\\right)^{-1} \\left(\\frac{\\Gamma}{1.5\\times10^{12}\\;\\sec^{-1}}\\right)^{-1} ~. \\end{equation} Here $h\\equiv H_0/100\\hubunits$, $\\Omega_b$ is the baryon density parameter, and $\\Gamma$ is the HI photoionization rate due to the diffuse UV background at redshift $z$. This approximation assumes that all gas obeys a temperature-density relation $T=T_0(\\rho_b/\\bar{\\rho}_b)^\\alpha$, which emerges from the interplay between photoionization heating and adiabatic cooling caused by cosmic expansion.  For typical reionization histories, one expects $T_0 \\sim 5000-20,000\\K$ and $\\beta = 2-0.7\\alpha \\approx 1.6-1.8$ at $z\\sim 3$, with the higher $T_0$ and $\\beta$ values arising if helium reionization occurs close to this redshift \\citep{hui97}. Strictly speaking, the overdensity in equation~(\\ref{eqn:fgpa}) is the gas overdensity, but in the moderate density regions that contribute most of the \\lya\\ forest opacity, the gas pressure is low, and the gas and dark matter trace each other fairly well. This approximation ignores the effects of collisional ionization and peculiar velocities.  From equation~(\\ref{eqn:fgpa}), one can see that the probability distribution function (PDF) of \\lya\\ forest flux decrements, $P(D)$, is closely related to the PDF of the underlying density field, $P(\\rho/\\bar{\\rho})$, with an effective smoothing scale determined by a combination of pressure support and thermal broadening along the line of sight.  The flux decrement PDF can therefore provide a diagnostic for the amplitude of mass fluctuations and for departures from Gaussianity in primordial fluctuations \\citep{weinberg99a,croft99,mcdonald00}.  The statistical measure that we focus on in this paper is the {\\it conditional} flux decrement PDF, the probability $P(D)$ that a pixel in a \\lya\\ forest spectrum has a flux decrement in the range $D\\rightarrow D+dD$, conditioned on the presence of a galaxy at the same redshift close to the line of sight. (Since $D=1-F/F_c$, we will use the slightly less cumbersome term ``conditional flux PDF,'' but we treat $D$ as the observable of interest because it is an increasing function of density.) This statistic can be measured by obtaining \\lya\\ forest spectra in fields covered by LBG surveys.  The bias of galaxies towards high density regions should reveal itself as a systematic shift of $P(D)$ towards higher flux decrements in pixels close to galaxies, with closer proximity yielding stronger shifts.  Conversely, feedback of galaxies on the local IGM via ionization or galactic winds might shift the PDF towards lower flux decrements at small separations.  A complete measurement of the conditional flux PDF will require ambitious, time-consuming surveys on large telescopes, since each LBG-quasar pair contributes only a single point to the distribution.  However, characteristics of the conditional PDF like the mean decrement or the fraction of saturated and ``transparent'' pixels can be measured with fewer data points, and we present predictions for these quantities in addition to the full PDF.  We also show that the signature of galaxy proximity should persist in spectra smoothed over $50-200\\kms$, allowing theoretical predictions to be tested using relatively low resolution quasar spectra. With higher resolution spectra the dependence of the conditional PDF on spectral smoothing offers a further test of the models.  K.\\ Adelberger and collaborators have been carrying out a survey along the lines envisioned here (\\citealt{adelberger02}, hereafter ASSP; \\citealt{adelberger02b}, hereafter ASPS). While we have not designed our study primarily for comparison with their results, we will discuss the comparison between our predictions and some of their measurements in \\S\\ref{sec:discussion} below.  Inspired by their work, we also consider an additional statistic, the mean galaxy overdensity in $5\\,h^{-1}$ and $10\\hmpc$ (comoving) cubes as a function of the mean decrement on a line of sight through the middle of the cube. This statistic characterizes the large scale correlations between galaxies and IGM overdensity, while the conditional flux PDF characterizes correlations on smaller scales. McDonald, Miralda-Escud\\'e, \\& Cen (\\citeyear{mcdonald02}) have presented predictions for this statistic using dark matter in N-body simulations, and here we show predictions for the biased galaxy population of a hydrodynamic simulation.  There is substantial overlap between our investigation and the recent paper of \\cite{croft02a}. As discussed in \\S\\ref{sec:discussion}, our results generally agree well with theirs where we examine similar quantities. \\cite{croft02a} devote considerably more attention to models that incorporate galactic winds, which they add to their simulations using the post-processing approach of \\cite{aguirre01}.  Here we concentrate on the direct predictions of our simulations, which incorporate thermal feedback in the local ISM that is usually radiated away before it can drive a galactic wind.  However, we do estimate the possible effects of galaxy photoionization, and of photoionization by recurrent AGN activity associated with galaxies, and we briefly discuss the possible influence of stronger winds. We describe our simulations, galaxy identification, and spectral extraction in the next section.  We present our results for the conditional flux PDF in \\S\\ref{sec:cpdf} and for the correlation between mean decrement and galaxy counts in cubic cells in \\S\\ref{sec:cube}. In \\S\\ref{sec:discussion}, we briefly review our results, then discuss them in relation to the theoretical study of \\cite{croft02a} and the observational analyses of ASSP, concluding with some remarks about future directions. ",
        "conclusions": "The dependence of the mean flux decrement on the galaxy proximity condition provides a convenient summary of our results. The mean decrement does not capture the full information in the PDF; we have tried, for example, scaling the optical depths of the unconditional flux PDF by a constant factor to see whether we can recover the conditional flux PDF simply by matching its mean, and we cannot.  However, measurement of the full conditional flux PDF requires many galaxy-quasar pairs, so measurements of the conditional mean decrement are likely to come much sooner (e.g., ASSP).  Figure~\\ref{fig:mean} plots the mean flux decrement as a function of maximum angular separation $\\tmax$ for galaxy populations of different space density, summarizing and reinforcing the results in Figures~\\ref{fig:cpdf} and~\\ref{fig:luminosity}. The spectral smoothing length is $R_s=10\\kms$. Hexagons and pentagons show results for the full populations of resolved galaxies in the L22n128 and L50n144 simulations, with space densities $\\Sigma=48\\dunits$ and $3\\dunits$, respectively. Squares and triangles show results for the 150 and 50 ``brightest'' (i.e., highest SFR) galaxies in L50n144.  For $\\Delta\\theta < \\tmax = 4'$ ($4.8\\hmpc$ comoving) the mean decrement is already well above the global mean of 0.36, and it rises steadily with decreasing $\\tmax$ as the \\lya\\ spectra probe the denser IGM in the immediate vicinity of the galaxies, with $\\meand > 0.9$ for $\\tmax=0.25'$.  The mean decrement varies only weakly with SFR, with the rare, high SFR galaxies having slightly higher mean decrement. The $\\times$'s show, for the full population of L50n144, the conditional mean decrement computed in bins of angular separation $0.8\\tmax < \\Delta\\theta < 1.25\\tmax$, instead of all separations $\\Delta\\theta < \\tmax$.  This ``differential'' form of the statistic shows a similar trend, but the increase of $\\meand$ at a given $\\tmax$ is, of course, smaller, since closer separation pairs have been eliminated.  Figure~\\ref{fig:mean2} plots the mean decrement against $\\tmax$ for random dark matter particles and for halos weighted in various ways, as in Figures~\\ref{fig:bias} and~\\ref{fig:hod}. The effect of bias stands out clearly as an offset between the matter and halo results.  Results for halos are similar to those for galaxies in Figure~\\ref{fig:mean}, though the mean decrements near halos are typically slightly lower.  The weighting of halos has only a small impact on the conditional mean decrement, with a modest increase in $\\meand$ for a stronger weighting of high mass halos.  \\subsection{Local Photoionization} \\label{sec:gpi}  We have so far assumed that the photoionizing background is uniform, and \\cite{croft02b} show that this assumption should be adequate for most computations of \\lya\\ forest statistics.  However, the conditional flux PDF is derived from the small fraction of pixels that lie close to galaxies, and if the galaxies make a significant contribution to the ionizing background, then the flux decrement in these pixels may be depressed by the ionizing flux of their nearest galaxy neighbors. With the ionizing background predicted by \\cite{haardt96} based on the observed quasar population, these simulations (and similar ones by other groups) already match the observed mean decrement of the \\lya\\ forest given a baryon density $\\Omega_b=0.02h^{-2}$.  However, estimates of the background based on the proximity effect usually yield a higher intensity than predicted from the quasar population alone \\citep[][and references therein]{scott00}, and observations of LBGs suggest that the escape fraction of ionizing photons is high enough to make them an important contributor to the UV background \\citep{steidel01}.  Within the theoretical and observational uncertainties, there is probably room for galaxies to make a contribution equal to that of quasars without making the predicted mean flux decrement too low at $z\\sim 3$ (see \\citealt{schirber03} for a recent discussion of these issues).  The intensity of the radiation from an individual galaxy or quasar falls off as $1/r^2$, but it is further attenuated by redshifting and, more importantly for the case of ionizing radiation, by IGM absorption.  Haardt \\& Madau (\\citeyear{haardt96}; see also \\citealt{madau99}) estimate that a path length $\\Delta z \\sim 0.17$ produces an optical depth of $\\tau(912\\hbox{\\AA})\\sim 1$ at $z=3$, corresponding to $100\\hmpc$ (comoving) for our cosmology. \\cite{fardal93} conclude that the effective attenuation length for ``average'' ionizing photons is $\\sim 2.4$ times that at 912\\AA, so a reasonable estimate (with probably a factor of two uncertainty) is $r_{\\rm att} \\sim 240\\hmpc$ (comoving). Here we will make the simple approximation that intensity falls as $1/r^2$ until $\\rmax=r_{\\rm att}$ and is sharply truncated beyond $\\rmax$. If a population of galaxies with comoving space density $n$ and mean ionizing luminosity $\\meanl$ contributes a fraction $f$ of the ionizing background intensity $I$, we then have \\begin{equation} fI ~=~ \\int_0^{\\rmax} 4\\pi r^2 dr \\,n \\,{\\meanl \\over 4\\pi r^2} ~=~ n\\meanl \\rmax ~, \\label{eqn:fI} \\end{equation} implying \\begin{equation} I ~=~ {n\\meanl\\rmax \\over f} ~. \\label{eqn:I} \\end{equation} The intensity from a galaxy with luminosity $L_i$ is equal to the mean background intensity at an ``influence radius'' $r_i$ given by \\begin{equation} {L_i \\over 4\\pi r_i^2} = I \\qquad \\Longrightarrow \\qquad r_i = \\rmax \\times \\left[ {L_i \\over \\meanl} {f\\over 4\\pi r^3_{\\rm max} n}\\right]^{1/2} ~. \\label{eqn:ri} \\end{equation} Note that $r_i$ decreases as either $\\rmax$ or $n$ increases, since either change makes the contribution of a nearby galaxy less important relative to that of the numerous, distant galaxies.  For $f=0.5$, $L_i=\\meanl$, $\\rmax=240\\hmpc$, and $n=1.2\\times 10^{-3} h^3 {\\rm Mpc}^{-3}$ ($\\Sigma = 1\\dunits$), equation~(\\ref{eqn:ri}) yields $r_i=0.37\\hmpc$ (all length scales comoving). For our cosmological model, the corresponding angular and velocity scales are $\\Delta\\theta = 0.3'$, $\\Delta V = 47\\kms$, respectively. This simple estimate suggests that galaxy photoionization could have a significant impact on the conditional flux PDF at separations of $\\Delta\\theta \\sim 0.5'$, though its impact at $\\Delta\\theta \\sim 2'$ is likely to be small. The optical depth of gas at temperature $T$, with electron density $n_e$, and at a distance $r$ from a galaxy is reduced relative to the optical depth $\\tau_u$ of the uniform background case by a factor \\begin{equation} \\tau/\\tau_u = \\left[1+\\fgamma\\left({r_i\\over r}\\right)^2\\right]^{-1} ~, \\label{eqn:tau} \\end{equation} where \\begin{equation} {1 \\over \\fgamma} = {\\Gamma_u + \\Gamma_c(T) n_e \\over \\Gamma_u} \\approx 1 + 0.29(\\rho/\\bar{\\rho})e^{-1.578/T_5} T_5^{1/2} (1+T_5^{1/2})^{-1} \\label{eqn:fgamma} \\end{equation} is the ratio of the total (photo $+$ collisional) ionization rate to the uniform photoionization rate $\\Gamma_u$. The second part of equation~(\\ref{eqn:fgamma}) incorporates the \\cite{haardt96} value of $\\Gamma_u=8.3\\times 10^{-13}\\, {\\rm s}^{-1}$ and the \\cite{cen92} expression for collisional ionization at temperature $T=10^5 T_5\\K$ (which is also used in TreeSPH and TIPSY), assuming fully ionized gas to relate $n_e$ to $\\rho/\\bar{\\rho}$.  Figure~\\ref{fig:gpi} illustrates the impact of galaxy photoionization on the conditional mean decrement (left panels) and saturated fraction (right panels) for the $\\Sigma=1\\dunits$ sample from the L50n144 simulation. We compute the influence radii via equation~(\\ref{eqn:ri}) assuming that each galaxy's ionizing flux is proportional to its SFR and that galaxies of this space density collectively produce a fraction $f=0.5$ of the ionizing background, with fainter galaxies having no significant ionizing flux. Asterisks show the result of a simplified calculation in which we multiply the optical depths of the extracted spectra by the factor in equation~(\\ref{eqn:tau}) assuming $\\fgamma=1$ and a pixel-galaxy distance $r$ corresponding to the transverse separation between the galaxy and the line of sight. As expected from the order-of-magnitude estimate above, the suppression relative to the uniform UV background case is small at $\\Delta\\theta \\geq 1'$, but the mean decrement and saturated pixel fraction are noticeably reduced for $\\Delta\\theta \\leq 0.5'$. Triangles show the results of a complete calculation, in which we put individual galaxy UV sources into the simulation before extracting the \\lya\\ forest spectra with TIPSY, applying equation~(\\ref{eqn:tau}) to each SPH particle with $r$ equal to the galaxy-particle separation and $\\fgamma$ computed from the particle's temperature and density. Here the reduction of $\\meand$ and $\\fsat$ is much smaller, and the predicted trends no longer turn over at the smallest angular separations.  Why does the approximate calculation drastically overestimate the impact of galaxy photoionization?  One possibility is that collisional ionization in the hotter, denser gas within $r\\sim 0.5\\hmpc$ of galaxies dilutes the effect of photoionization by making $\\fgamma \\ll 1$. However, if we repeat the TIPSY calculation with $\\fgamma=1$ for all particles, we get a nearly identical result for the conditional mean decrement.  Collisional ionization {\\it is} important along some lines of sight (compare equation~[\\ref{eqn:fgamma}] to Figure~\\ref{fig:profiles}), but these are cases where dense gas produces heavily saturated absorption, which remains saturated even if the optical depth is reduced by a factor of several. Thus, overestimating photoionization by setting $\\fgamma=1$ still leaves a flux decrement $D\\approx 1$. The second possibility is that the gas producing absorption at the galaxy redshift is further away than the transverse separation $r_t$, and is shifted to the galaxy redshift by its peculiar infall velocity. Figure~\\ref{fig:dlos} shows the mean line-of-sight distance, weighted by optical depth, of gas that produces absorption within $20\\kms$ of the galaxy redshift.  We consider only galaxy-spectrum pairs with $\\Delta\\theta<0.5'$, implying a transverse separation $r_t<0.6\\hmpc$ (comoving); results are similar for $\\Delta\\theta<0.25'$. Figure~\\ref{fig:dlos} shows that absorption for these close pairs arises mainly in gas whose line-of-sight distance substantially exceeds the influence radius $r_i$ and is therefore little affected by galaxy photoionization. When the optical depth is high ($\\tau>10$, left-hand panel), the mean line-of-sight distance is usually smaller, but in these cases even a factor of several reduction in $\\tau$ does not move the flux decrement significantly below $D=1$. Thus, even if galaxies contribute a large fraction of the ionizing background, local photoionization has little effect on the conditional mean decrement or saturated fraction because the real-space distance to absorbing gas is larger than the redshift-space distance, except for lines of sight that are heavily saturated.  \\cite{steidel02} find that $\\sim 3\\%$ of the galaxies in their LBG sample exhibit detectable AGN activity. Given the apparent ubiquity of supermassive black holes in local galaxies, it seems plausible that most LBGs go through AGN phases, with $\\fq \\sim 0.03$ being the ``duty cycle'', i.e. the fraction of time that these black holes are active at $z\\sim 3$. Could photoionization by these low luminosity AGN have a larger impact on the local IGM than the galaxies themselves? If the interval between periods of AGN activity is longer than the time \\begin{equation} \\teq = {n^{\\rm eq}_{\\rm HI} \\over \\alpha(T) n_e n_{\\rm HII}} = [\\Gamma_u + \\Gamma_c(T) n_e]^{-1} \\approx 4\\times 10^4 \\fgamma\\,{\\rm yrs} \\label{eqn:teq} \\end{equation} that it takes gas to return to its equilibrium neutral hydrogen fraction after being fully ionized,\\footnote{We assume that the equilibrium state is already highly ionized, so that the addition of $n^{\\rm eq}_{\\rm HI}$ hydrogen ions does not significantly increase $n_{\\rm HII}$ or $n_e$.  The second equality follows from the equilibrium condition $\\alpha(T) n_e n_{\\rm HII} = \\Gamma_u n^{\\rm eq}_{\\rm HI} + \\Gamma_c(T) n_e n^{\\rm eq}_{\\rm HI}.$} then we are essentially back to the uniform ionization case: the few LBGs that currently host AGN have large ``proximity zones,'' but the rest have their normal complement of associated \\lya\\ optical depth.  However, if the process of feeding gas to the central black hole is sufficiently stochastic, then AGN might ``flicker'' on and off on timescales shorter than $\\teq$, with individual activity cycles lasting $\\tactive \\la \\fq \\teq \\la 10^3\\,$years. In this case, gas around a large fraction of LBGs could be out of ionization equilibrium, with the neutral hydrogen fraction depressed by the memory of the most recent AGN outburst.  Suppose we assume that the AGN associated with the galaxies under study collectively produce 50\\% of the ionizing background. Since the time-averaged flux of this population equals, by assumption, the flux that we previously ascribed to the galaxies themselves, the photoionization rate at distance $r$ during an active phase must exceed the previous value by a factor $\\fq^{-1}$, i.e., $\\Gammaq = \\Gamma_u(r_i/r)^2\\fq^{-1}$. During an active phase, the neutral hydrogen density at distance $r$ drops exponentially in time, $n_{\\rm HI} \\approx n^{\\rm eq}_{\\rm HI} \\exp(-\\Gammaq t)$, until it reaches a new equilibrium value with \\lya\\ optical depth \\begin{equation} \\tau/\\tau_u = \\left[1+\\fq^{-1}\\fgamma\\left({r_i\\over r}\\right)^2\\right]^{-1} ~. \\label{eqn:tau2} \\end{equation} However, at most radii the duration of the active phase will not be long enough to achieve this equilibrium.  If we assume that the activity during a recombination interval $\\teq$ occurs in a single ``outburst'' of duration $\\tactive=\\fq\\teq$, then the optical depth at the end of this phase is \\begin{equation} \\tau/\\tau_u \\approx \\exp(-\\Gammaq \\tactive) = \\exp\\left[-\\fgamma\\left({r_i \\over r}\\right)^2\\right]~, \\label{eqn:tau3} \\end{equation} except that it never falls below the equilibrium value of equation~(\\ref{eqn:tau2}). Note that the duty cycle $\\fq$ cancels out of equation~(\\ref{eqn:tau3}) because we have fixed the time-averaged flux of the AGN population relative to $\\Gamma_u$.  If the AGN are more luminous while they are on, then they must be active for less time, producing the same number of ionizing photons.  We see from equation~(\\ref{eqn:tau3}) that the scale over which recurrent AGN activity can affect the conditional mean flux decrement is the same influence radius that we found previously, but that departures from photoionization equilibrium allow the impact within this radius to be much stronger.  Since we have already found that much of the absorption for small $\\Delta\\theta$ comes from gas beyond the influence radius, we can guess from this analysis that local AGN photoionization will have little impact on the conditional mean decrement. Filled circles in Figure~\\ref{fig:gpi} confirm this expectation, showing a calculation based on equation~(\\ref{eqn:tau3}), with sources embedded in the TIPSY spectral extraction and the same influence radii used previously.  The minimum optical depth is given by equation~(\\ref{eqn:tau2}) with $\\fq^{-1}=30$.  While the absorption is slightly suppressed relative to the galaxy photoionization case, the mean decrement and saturated fraction continue to increase as $\\tmax$ decreases. These results can be considered a conservative upper limit on the effects of recurrent AGN activity, since we have ignored the partial return to equilibrium with the uniform background that will occur following any given active cycle. If we attributed 100\\% of the ionizing background to the AGN associated with the observed LBGs, then the influence radii would be larger by $2^{1/2}$, but the impact on the conditional mean decrement would be only slightly stronger.  We conclude that local photoionization by galaxies or by the AGN that they host will not reverse the trend of increasing \\lya\\ forest absorption with decreasing angular separation. Figure~\\ref{fig:dlos} suggests that {\\it any} feedback mechanism must have a strong influence out to comoving distances $\\sim 1\\hmpc$ or more if it is to effect such a reversal. We have examined simple models in which we completely eliminate neutral hydrogen out to a fixed radius around target galaxies or, alternatively, reduce the neutral fraction within this radius by a factor of three. If we eliminate all neutral hydrogen to $r=1\\hmpc$, then the conditional mean decrement drops to $\\meand \\approx 0.6$ for $\\tmax=0.25'$ and $0.5'$, and the saturated pixel fraction falls to $\\fsat \\approx 0.4$. However, a factor of three reduction out to $1\\hmpc$, or complete elimination out to $0.5\\hmpc$, has only small impact, comparable to that of the AGN ionization model shown in Figure~\\ref{fig:gpi}. Elimination of neutral hydrogen out to $0.75\\hmpc$ has an intermediate effect, reducing $\\meand$ to $\\sim 0.75$ and $\\fsat$ to $\\sim 0.5$.  \\subsection{Redshift Errors} \\label{sec:zerror}  We showed in Figure~\\ref{fig:deltav} that the signature of the overdense environments of galaxies in the conditional flux PDF is substantially weakened if one looks at pixels $\\sim 200\\kms$ away from the galaxy redshift instead of at the galaxy redshift itself.  (The angular separation corresponding to $Hr=200\\kms$ is $1.3'$.)  Since the redshifts of LBGs are usually estimated from nebular emission or absorption lines, which could be offset relative to the mean systemic velocities, this result suggests that redshift measurement errors could have a significant impact on practical determinations of the conditional flux PDF. To address this point quantitatively, we show in Figure~\\ref{fig:zerror} the conditional mean decrement and saturated fraction when pixel redshifts are drawn from a Gaussian of mean zero and dispersion $\\sigma_{\\Delta V}=150\\kms$ (squares) or $300\\kms$ (triangles). We use the $\\Sigma=1\\dunits$ sample of the L50n144 simulation, with no galaxy photoionization.  Redshift errors of $\\sigma_{\\Delta V}=150\\kms$, similar to those estimated by ASSP, depress $\\meand$ and $\\fsat$ by $\\sim 0.05$ at $\\tmax=2'$ and by $\\sim 0.1-0.2$ at $\\tmax \\leq 1'$.  Errors of $\\sigma_{\\Delta V}=300\\kms$ depress $\\meand$ and $\\fsat$ more severely, and they largely (though not entirely) remove the trend of increasing absorption with decreasing $\\tmax$. Redshift errors also reduce the already small differences between models with and without galaxy photoionization, since the absorption is now usually measured at a pixel further away from the galaxy.   \\label{sec:discussion}  The conditional flux PDF considered in \\S\\ref{sec:cpdf} and the $\\dgal-\\meands$ correlation considered in \\S\\ref{sec:cube} offer empirical tools with which to test the prediction that high redshift galaxy formation is strongly biased, and to search for the influence of star-forming galaxies on the surrounding IGM. The predicted flux PDF shifts systematically towards higher absorption in pixels close to galaxies --- the smaller the angular separation, the higher the mean decrement and saturated fraction and the lower the transparent fraction. The shift of the flux PDF remains strong in spectra smoothed over 50, 100, or $200\\kms$, allowing further tests of the predicted scale of correlations and demonstrating that moderate resolution spectra of relatively faint AGN may profitably be used in such investigations. The influence of galaxy proximity is much stronger than that of random mass elements (dark matter particles), a clear signature of the preferential formation of galaxies in overdense environments.  The conditional flux PDF is only weakly sensitive to the baryonic mass or SFR of the sample galaxies over the range that we have investigated, spanning more than two orders of magnitude in space density.  It is also insensitive to the occupation distribution of dark matter halos.  While this insensitivity reduces the utility of this statistic as a diagnostic of galaxy formation, it increases its power as a diagnostic of galaxy feedback, since the no-feedback prediction is robust. We have investigated the potential impact of one feedback process, galaxy photoionization, and find that it has only a small impact on the conditional mean flux decrement, even if the bright galaxies used in the conditional PDF measurement contribute a large fraction of the ionizing background.  The local flux from these galaxies is comparable to the uniform background at ``influence radii'' $r_i \\sim 0.4\\hmpc$ (comoving), corresponding to angular separations $\\Delta\\theta \\sim 0.3'$, and in the absence of peculiar velocities this flux would drive down the average absorption at smaller separations. However, the gas producing absorption at the galaxy redshift is typically $\\ga 1\\hmpc$ away along the line of sight, where it is minimally affected. Recurrent AGN activity, with associated departures from photoionization equilibrium, can have a stronger impact within $r_i$, but it does not substantially increase the typical influence radius, so it still has little effect on the flux decrement close to galaxies. The predicted dependence of $\\meand$ on $\\Delta\\theta$ {\\it is} sensitive to redshift measurement errors, with rms errors $\\sigma_{\\Delta V} \\ga 150\\kms$ significantly depressing the absorption signature.  There is significant overlap between our study of the conditional flux PDF and the recent paper by \\cite{croft02a}, who independently investigate the conditional mean decrement statistic and other measures of LBG-\\lya\\ forest correlations, using similar numerical techniques. As discussed in \\S\\ref{sec:cpdf}, \\cite{croft02a} and ASSP define the conditional mean decrement in a slightly different way from us, averaging over all galaxy-pixel pairs in bins of redshift-space separation $\\Delta_r$ instead of considering only pixels at the galaxy redshift.  Averaging in redshift-space separation bins instead of angular separation bins at $\\Delta V=0$ (like the ``differential'' points of Figures~\\ref{fig:mean} and~\\ref{fig:zerror}) has a small but noticeable impact, yielding slightly less absorption at a given comoving distance. To facilitate comparison to the \\cite{croft02a} and ASSP results, we have calculated their version of the mean decrement statistic for the 150 highest SFR galaxies in the L50n144 simulation ($\\Sigma=1\\dunits$).  Results are shown in Figure~\\ref{fig:croft}, assuming rms redshift determination errors of 0, 150, and $300\\kms$ (triangles, squares, and pentagons, respectively).  \\cite{croft02a} adopt slightly different cosmological parameters, with $\\Omega_m=0.3$, $\\sigma_8=0.9$ instead of $\\Omega_m=0.4$, $\\sigma_8=0.8$.  Relative to our L50n144 run, their impressive, $2\\times 300^3$ particle simulation has a volume 3.3 times smaller but mass resolution 30 times higher.  Solid and dotted curves in the left hand panel of Figure~\\ref{fig:croft} show their results for galaxies with $M_b>10^{10}M_\\odot$ and $M_b>10^{11}M_\\odot$ (Croft et al., fig. 8), kindly provided by R.\\ Croft.  Given the differences in cosmological and numerical parameters, the two independent calculations agree remarkably well.  The comoving space densities of the two Croft et al.\\ samples, 0.028 and 0.0008 $h^3\\,{\\rm Mpc}^{-3}$, correspond to $\\Sigma=24\\dunits$ and $\\Sigma=0.67\\dunits$ for our cosmology, so they are comparable to our $\\Sigma=48\\dunits$ and $\\Sigma=1\\dunits$ samples, respectively.  As shown in Figure~\\ref{fig:mean}, we find no significant difference in the conditional mean decrement for these two samples, while Croft et al.\\ find stronger absorption for the more massive galaxies.  Since their $M_b>10^{11} M_\\odot$ sample comprises only 30 galaxies, this difference in the results does not seem too serious at present.  Croft et al.\\ also consider higher density samples, with mass resolution thresholds below the resolution limit of our L22n128 simulation, and they find a continuing though modest trend of decreasing conditional mean decrement with decreasing galaxy mass threshold.  In agreement with Croft et al., we find that redshift determination errors of $\\sigma_{\\Delta V} \\sim 150-300\\kms$ have a substantial impact on the predicted trend of $\\meand$ with galaxy separation, and that galaxy photoionization does not have a large impact.  ASSP have investigated LBG-\\lya\\ forest correlations observationally, carrying out an LBG spectroscopic survey in fields around six quasar lines of sight. Their estimates of the conditional mean decrement, kindly provided by K.\\ Adelberger, are shown by the filled points in Figure~\\ref{fig:croft}.  These are somewhat different from the points shown by \\cite{croft02a} because of changes to ASSP's analysis procedures, in particular a scaling of optical depths with redshift that should make the results more directly comparable to predictions like the ones presented here. The predicted and observed trends agree reasonably well down to comoving separations $\\sim 3\\hmpc$, with a smooth increase of $\\sim 0.1$ in $\\meand$.  However, ASSP's trend flattens towards smaller separations, while our predicted trend rises and steepens.  Most strikingly, ASSP find a {\\it decrease} in $\\meand$ at separations $r < 1\\hmpc$, where the predicted absorption is strongest. If the rms redshift errors are $\\sigma_{\\Delta V} \\sim 300\\kms$ rather than the $\\sim 150\\kms$ that ASSP estimate, then the discrepancy between the predictions and observations is reduced, but the two data points at $r < 1\\hmpc$ remain in serious conflict.  ASSP suggest that the flattening and eventual decline of the mean decrement in close proximity to LBGs is a signature of galaxy feedback on the surrounding IGM, perhaps caused by the galactic scale winds suggested by the nebular line profiles of LBGs \\citep{pettini01,pettini02}.  \\cite{croft02a} consider a variety of galaxy feedback models and show that they can roughly reproduce the ASSP mean decrement results if $\\sim 10\\%$ of the available supernova energy is converted to kinetic energy of a coherent, galactic scale outflow that sweeps up all of the IGM in its path.  While this wind model is {\\it ad hoc} and idealized, it does show that it is energetically possible for supernova feedback to influence the conditional mean decrement out distances of a few $\\hmpc$ (comoving). The range of the effect is larger than that of individual wind bubbles because the galaxies themselves are correlated, and a line of sight $\\sim 2\\hmpc$ from one galaxy may pass through the wind-blown bubble of another. We have shown (in agreement with \\citealt{croft02a} and ASSP) that galaxy photoionization cannot explain the observed downturn in absorption, and we have shown that this remains the case even if one considers departures from ionization equilibrium that could be caused by recurrent AGN activity.   The results of \\S\\ref{sec:gpi} (Figure~\\ref{fig:dlos} in particular) have an important implication for wind models: peculiar velocities allow gas at comoving distances $1-2\\hmpc$ to produce strong absorption at redshift-space separations $0-0.5\\hmpc$, so the required wind radius $R_w$ is larger than one would first guess from the scale of the observed downturn.  Figure~\\ref{fig:croftalt} demonstrates this point explicitly, using the $\\Sigma=3\\dunits$ sample of the L22n128 sample (the top 40 galaxies in the box).  Squares, pentagons, and hexagons show the conditional mean decrement computed after eliminating all neutral hydrogen within a comoving radius $R_w=0.75$, $1.0$, or $1.5\\hmpc$ around each of the target galaxies (see \\citealt{kollmeier02} for further discussion). Only the $R_w=1.5\\hmpc$ case produces a downturn comparable to the observed one, and even it does not reproduce ASSP's innermost data point (as we discuss further below). Stars show a case in which we put winds around all 641 resolved galaxies in the box and scale the wind volume in proportion to each galaxy's baryon mass, $R_w\\propto M_b^{1/3}$.  This model has $R_w \\approx 1\\hmpc$ for the 40th-ranked galaxy in the box and a maximum $R_w \\approx 2\\hmpc$. The impact is similar to, though slightly weaker than, that of the constant-radius $R_w=1.5\\hmpc$ model. All of these models are, of course, highly simplified, but they show that galactic winds must efficiently eliminate neutral hydrogen out to large distances to explain ASSP's results. The energetic requirements for such winds are not impossible to meet (see \\citealt{croft02a}), but they do require sustained outflow speeds of several hundred $\\kms$ for $t \\ga 1$ Gyr and effective entrainment of surrounding gas. Winds with large filling factor can have a noticeable impact on the flux power spectrum of the \\lya\\ forest, though for $R_w \\leq 1\\hmpc$ the effect is small \\citep{weinberg03}.  The innermost ASSP data point is based on only three galaxy-QSO pairs, and the error estimation procedure is somewhat {\\it ad hoc} (see ASSP), so it is natural to ask whether the discrepancy with the data is statistically significant.  To address this point, we randomly selected 3-tuples of simulated galaxy-sightline pairs with $\\Delta\\theta < 0.4'$ and computed the mean decrement in the redshift separation bin $\\Delta_r = 0-0.5\\hmpc$, repeating the experiment 500 times. Figure~\\ref{fig:monte} shows the cumulative distribution of these 3-tuple mean decrements, for our standard model with no redshift measurement errors (solid line), for redshift errors $\\sigma_{\\Delta V}=300\\kms$ (dashed line), and for the $R_w=1.5\\hmpc$ ``wind'' model from Figure~\\ref{fig:croftalt}. Vertical lines mark the unconditional mean flux decrement, $\\meand=0.36$, and the ASSP measurement for this separation, $\\meand=0.11$. With $\\sigma_{\\Delta V}=300\\kms$, the 3-tuple mean decrement is below $0.36$ only $\\sim 5\\%$ of the time, and none of our 500 trials have a mean decrement as low as the ASSP value.  Even for the $R_w=1.5\\hmpc$ wind model, only $\\sim 5\\%$ of the trials have a 3-tuple mean decrement as low as ASSP's.  We did not incorporate redshift errors in the wind model, but (in contrast to the no-wind case) they tend to raise the predicted mean decrement by moving the measurement point further from the influence of the galaxy. We have only 150 simulated galaxies contributing to the solid and dotted curves and only 40 contributing to the dashed curve, so our 3-tuples are not independent and we may therefore underestimate the incidence of rare, low absorption cases.  Nonetheless, it appears that the ASSP results cannot easily be explained away by appealing to small number statistics.  Given the stringent demands on wind models, it is worth considering alternative explanations.  One possibility is that the IGM at these distances from galaxies is multi-phase, with processes that the simulations cannot resolve causing the geometrical cross section of the neutral phase to drop sharply.  However, the typical physical conditions at these distances from galaxies are not extreme (see Figure~\\ref{fig:profiles}), so it would be difficult to reconcile such an explanation with the overall success of simulations in reproducing (unconditional) statistical properties of the \\lya\\ forest.  A final possibility worth considering is that extended \\lya\\ {\\it emission} from the target galaxies ``fills in'' the \\lya\\ forest at the redshift of emission.  Extended \\lya\\ ``blobs'' have been observed to be associated with LBG's \\citep{steidel00} having angular extents $\\sim 15''$ and AB apparent magnitudes 21.02 and 21.14 in an 80\\AA\\ \\lya\\ band.  Cooling radiation from gas settling into massive galaxies at $z=3$ naturally produces \\lya\\ flux of this order, and numerical simulations predict that such blobs should be present around typical LBGs \\citep{fardal01}. Thus, a galaxy's \\lya\\ cooling radiation could plausibly replace the absorbed continuum flux of a 21st-magnitude background quasar. There are two substantial and partly cancelling corrections to this estimate.  First, the \\lya\\ emission extends over $\\sim 1000\\kms$ (16 \\AA) rather than the 80\\AA\\ bandpass used by \\cite{steidel00}, so the flux density at the \\lya\\ wavelength is a factor of five higher ($-1.75$ mag). Second, at a separation $\\Delta\\theta \\sim 15-20\"$ from a target galaxy, the $\\sim 0.8\"$ slits used by ASSP should intercept only $\\sim 0.8/2\\Delta\\theta \\sim 0.02-0.03$ of the galaxy's extended \\lya\\ flux ($\\sim +4$ mag), assuming constant surface brightness out to $\\Delta\\theta$. The three quasars that contribute to ASSP's innermost data point have $G$-band AB magnitudes of 20.1, 21.6, and 23.4, so with the $+2.25$ mag net correction, it appears that \\lya\\ emission could replace the absorbed quasar flux only for the faintest of the three quasar targets, unless the galaxies in question are even brighter than the \\cite{steidel00} blobs. Furthermore, a fourth pair involving the $G=17.8$ quasar Q0302-0019 shows no sign of absorption near the galaxy redshift, and in this case the quasar is clearly too bright for galaxy emission to compete with it. (This pair and two others are dropped from ASSP's $\\meand$ calculation because of possible Ly$\\beta$ contamination.) At this point, therefore, the \\lya\\ emission explanation seems unlikely, but future observations that have more close pairs involving bright quasars will be able to test it conclusively.  On the larger scales probed by the $\\dgal-\\meands$ correlation, our predictions are in qualitative agreement with the observational results reported by ASSP and ASPS.  As the Fluctuating Gunn-Peterson Approximation suggests, higher mean flux decrements tend to arise in cells of higher mass density, and these in turn have higher average galaxy overdensity.  The limited size of our largest volume simulation, a single $50\\hmpc$ cube, leaves statistical and systematic uncertainties in our predictions.  Nonetheless, our results for the $\\ddark-\\meands$ correlation at $S=10\\hmpc$ are similar to those found by \\cite{mcdonald02} using hydro-PM simulations (with $40\\hmpc$ cubes but multiple realizations, and significantly higher mass resolution). The $\\dgal-\\meands$ correlation is much steeper than the $\\ddark-\\meands$ correlation because of the strong bias of the high redshift galaxy population, and the suppression of galaxy formation in underdense environments is especially clear. The ratio of correlation slopes is approximately equal to the ratio of rms fluctuation amplitudes, $\\siggal/\\sigdm$, but a ``linear bias'' model does not fully capture the behavior of the $\\dgal-\\meands$ correlation, in particular the curvature at low $\\meands$ imposed by the constraint that $\\dgal \\geq -1$.  Our $50\\hmpc$ simulation predicts a correlation coefficient $\\rgalD=0.49$ between galaxy density contrast and averaged flux decrement in $5\\hmpc$ or $10\\hmpc$ cells.  This is smaller than the value $r=0.68 \\pm 0.06$ that ASPS measure for $S\\approx 10\\hmpc$, with a $1\\sigma$ error bar derived from the dispersion among five lines of sight. However, we cannot assign a reliable ``theoretical error bar'' to our predicted $\\rgalD$ without modeling ASPS's procedures in greater detail.  Even for an ideal, volume-limited sample, there are subtle issues associated with shot noise subtraction and estimator biases (Hui \\& Sheth, in preparation), and corrections for redshift selection functions or sample incompleteness could also have a significant impact on estimates of $r$ and its uncertainty. Accurate assessment of the discrepancy between predicted and observed correlation coefficients probably requires larger simulation volumes, which would allow creation of multiple independent ``surveys'' comparable to the observed one.  Fortunately, we find that the $\\delta-D_S$ correlations for galaxies are similar to those for halos of the same space density, so it should be possible to carry out such assessments using large N-body simulations and the PM or hydro-PM approximation \\citep{croft98,gnedin98} for the \\lya\\ forest. The absolute trend (i.e., not normalized by standard deviations) of $\\dgal$ with $\\meands$ is the best test for the strong bias of high redshift galaxies predicted by theoretical models.  As the size of LBG spectroscopic samples grows, moderate resolution spectra of fainter quasars can greatly increase the number of lines of sight available for measuring LBG-\\lya\\ forest correlations. These larger samples will shrink the error bars of the existing measurements, and they will eventually allow measurement of the full conditional flux PDF for different subsets of the LBG population. The potential impact of redshift determination errors can be investigated empirically by comparing results for galaxies whose redshifts are estimated by different methods (e.g., emission vs.\\ absorption lines). Models with galactic winds or other forms of feedback can be tested using the full PDF rather than the mean decrement alone. Variations in feedback from one galaxy to another should widen the range of flux decrement values in addition to reducing the mean, and winds that sweep up the IGM should produce many lines of sight with essentially zero flux decrement. Perhaps the best diagnostic of galaxy feedback will be separate measurements for galaxy subsets with high and low star formation rates, or older and younger stellar populations, since one would expect the young/high SFR galaxies to have had more vigorous activity in the recent past.  Systematic differences in the activity of different galaxy populations should produce systematic differences in the corresponding conditional flux PDFs, and our results show that such differences would be unlikely to arise from differences in the spatial clustering of these populations. A full accounting of the relation between LBGs and the \\lya\\ forest will test our basic expectations for the way that dark matter shapes the large scale distribution of high redshift galaxies, and it will inform our picture of the ways that galaxies can affect the high redshift IGM."
    },
    "0209/astro-ph0209339_arXiv.txt": {
        "abstract": "s{As it was recently shown by Derishev\\cite{De}, initial GRB ejecta are likely to contain  comparable number of protons and neutrons.  During acceleration process, and/or later, due to interaction with external medium, such ejecta are likely to be split into spatially distinct shells of neutron and proton-electron plasmas. This leads to dynamical effects which can affect the afterglow light curves. In this paper we study these effects including, for the first time, radiation drag imposed on the neutron rich ejecta. In the presence of efficient radiation drag and for typical ejecta Lorentz factor (below 400), this is the neutron shell which moves faster and, after conversion to proton-electron plasma, powers the afterglow. After certain amount of deceleration, this shell is hit by the second one. Such collision will lead to reflaring, as observed in some afterglow light curves.} ",
        "introduction": "Recent observational data points toward the hypernova model  for long gamma-ray bursts. Within the framework of the hypernova model Lazzati, Ghisellini, Celotti and Rees~\\cite{La} show that  the effects of radiation drag may  be very important for the  dynamics and radiation of the GRB. The ejecta  encountering   a dense radiation field can be decelerated and  can produce very hard X-ray spectra by  upscattering the external radiation field.   Derishev, Kocharovsky, and Kocharovsky~\\cite{De} noticed that  the relativistic shock must contain neutrons.  Neutrons are coupled to protons when the density is high enough, and if the density becomes low enough already in the acceleration phase then the neutron and proton ejecta  will separate.    The latter condition can be satisfied only if $\\Gamma > 400$. In this paper we point out that the GRB ejecta can split into the neutron and the proton-electron shells even for $\\Gamma < 400$, provided the radiation drag is efficient like suggested by Lazzati et~al.~\\cite{La}.  \\begin{figure} \\psfig{figure=acc0.ps,angle=-90,width=0.45\\columnwidth} \\psfig{figure=acc1.ps,angle=-90,width=0.45\\columnwidth} \\caption{The acceleration phase in the presence of neutrons. If $t_{pn}> t_{acc}$ then the protons and neutrons move together. In the opposite case  when $t_{pn}<t_{acc}$ protons and neutrons will separate.} \\end{figure} ",
        "conclusions": "The inclusion of both the effects of the radiation drag and of the neutron content in the ejecta inevitably leads to a collision of the two shells in the afterglow phase. Such collision will cause reflaring in  the afterglow.  If the external medium is dense enough and the afterglow starts at the distance shorter than the distance of the neutron decay, then in addition to the reflaring  discussed above the early afterglow will consist of two components: the first from the proton shell and the second due to the shock formed when neutrons decay to protons. This is illustrated in the right panel of Figure~2 and in the right panel of Figure~4. One component of the afterglow is emitted between $r_a$ and $r_n$ by the proton wave and the second component is formed between $r_n$ and $r_{coll}$ by the deceleration of the protons formed in neutron decays. The two components will very likely overlap in the observers frame."
    },
    "0209/astro-ph0209613_arXiv.txt": {
        "abstract": "{% Formation of planets in the 47 UMa system is followed in an evolving protoplanetary disk composed of gas and solids. The evolution of the disk is calculated from an early stage, when all solids, assumed to be high-temperature silicates,  are in the dust form, to the stage when most solids are locked in planetesimals. The simulation of planetary evolution starts with a solid  embryo of $\\sim$1 Earth mass, and proceeds according to the core accretion -- gas capture model. Orbital parameters are kept constant, and it is assumed that the environment of each planet is not perturbed by the second planet. It is found that conditions suitable for both planets to form within several Myr are easily created, and maintained throughout the formation time, in disks with $\\alpha\\approx0.01$. In such disks, a planet of 2.6 Jupiter masses (the minimum for the inner planet of the 47 UMa system) may be formed at 2.1 AU from the star in $\\sim$3 Myr, while a planet of 0.89 Jupiter masses (the minimum for the outer planet) may be formed at 3.95  AU from the star in about the same time. The formation of planets is possible as a result of a significant enhancement of the surface density of solids between 1.0 and 4.0 AU, which results from the evolution of a disk with an initially uniform gas-to-dust ratio of 167 and an initial radius of 40 AU. ",
        "introduction": "Planets in the mass range 0.1 to 10 Jupiter masses (M$_J$), separated from their central stars by 0.04 to 5 AU, have been discovered around $\\sim$ 100 main-sequence stars with masses in the range 0.3 -- 1.1 M$\\odot$. The general observed properties of these planets, several of which are in fact in planetary systems, are reviewed by Marcy, et. al, Perryman (\\cite{Marcy00}), and Bodenheimer \\& Lin (\\cite{Bodenheimer02}). In comparison to the solar system, many of these systems are unusual in two respects: they contain Jupiter-mass planets at close distances (down to 0.04 AU) from their star, and many of them are in orbits with substantial (0.4 to 0.7) eccentricity. However one particular system, belonging to the star 47 Ursae Majoris (47 UMa), stands out as being more solar-system like than most of the other extrasolar planetary systems. This paper examines a possible mechanism for the origin of the two planets in the 47 UMa system.   The solar-type star 47 UMa has spectral type G0V, a mass of 1.03 M$_\\odot$, and a metallicity [Fe/H] = $ - 0.08$.  One of the first extrasolar planets to be discovered was a companion to 47 UMa (Butler \\& Marcy \\cite{Butler}) at 2.09 AU, with a current minimum mass of 2.62 M$_J$ and an eccentricity of 0.04.  Later (Fischer et al. \\cite{Fischer02}) a second planet was discovered; current orbital parameters (Fischer \\cite{Fischer_pr}, private communication) give a semimajor axis of 3.95 AU, a minimum mass of 0.89 M$_J$ and an eccentricity which is not well determined but is probably less than 0.1.  The mass ratio of the two planets (0.34) is close to that of Saturn/Jupiter (0.30), and the ratio of semimajor axes (1.90) is also close to that of Saturn/Jupiter (1.83).  A dynamical analysis of the system (Laughlin et al. \\cite{Laughlin01}) shows that if the two planets are in the same orbital plane, Earth-mass planets in the habitable zone would have stable orbits. However in the presence of the two fully formed giant planets, the formation of Earth-mass planets in the inner regions of the system is possible only interior to the habitable zone.  Three main theories have been proposed regarding the origin of planetary-mass objects.  The first is dynamical fragmentation of a rotating collapsing protostar, the mechanism thought to be responsible for multiple stellar systems (reviewed by Bodenheimer et al. \\cite{Bodenheimer00}) and possibly the isolated planetary-mass objects observed in the young cluster $\\sigma$ Orionis (Zapatero Osorio et al. \\cite{Zapatero}).  The second is gravitational instability in a disk (Kuiper \\cite{Kuiper}; Boss \\cite{Boss00}; Boss et al. \\cite{Boss02}), in which, on a few dynamical time scales, a gravitationally bound subcondensation forms in a disk that at some location has a Toomre $Q$ value on the order unity. The third mechanism, known as the core accretion -- gas capture process, involves the relatively slow gradual accretion of small condensed particles in a disk, eventually resulting in a solid core of a few M$_\\oplus$ which is able to gravitationally capture gas from the surrounding nebular disk (Safronov \\cite{Safronov}; Perri \\& Cameron \\cite{Perri}; Mizuno \\cite{Mizuno}). The strengths and weaknesses of the second and third processes are reviewed by Bodenheimer \\& Lin (\\cite{Bodenheimer02}).  The formation of the inner giant planet in the 47 UMa system has been studied by Bodenheimer et. al (\\cite{BHL}) under the assumption that it formed {\\it in situ} by the core accretion -- gas capture process. The evolutionary calculations they performed are based on the earlier work by Bodenheimer \\& Pollack (\\cite{Bodenheimer86}), who assumed a constant solid accretion rate for the buildup of the core, and by Pollack et al.  (\\cite{Pollack96}) who employed more detailed physics, including a (non-constant) solid accretion rate calculated from three-body accretion cross sections. The aim of the calculation was to determine the disk properties needed to form the planet on a reasonable time scale at 2.1 AU. Two calculations with variable solid accretion rate were performed (their cases U1 and U2). The two most important parameters in such calculations are (1) the initial surface density of solid material $\\Sigma_s$, to which the formation time is highly sensitive, and (2) the grain opacity $\\kappa_g$ in the envelope of the protoplanet, to which the formation time is moderately sensitive (Pollack et al. \\cite{Pollack96}). In both cases presented by Bodenheimer et al. (\\cite{BHL}) the values of $\\kappa_g$ were based on interstellar grain properties as calculated, for example, by Pollack et al. (\\cite{Pollack85}). In the temperature range 100 -- 1500 K, those opacities are typically in the range 1--8 cm$^2$ g$^{-1}$.  The values of $\\Sigma_s$ were set to 50 and 90 g cm$^{-2}$ in the two calculations, and the formation times turned out to be 18.6 Myr and 1.9 Myr, respectively. The final solid core masses were 38 and 69 M$_\\oplus$, respectively, compared to a total assumed final mass of 2.5 M$_J$ in both cases.  The lifetimes of disks around young stars, which constrain the formation times for giant planets, fall in the range 1--8 Myr (Haisch et al. \\cite{Haisch}), with half of the disks in young clusters already gone after times of 3--4 Myr. The mass ratio of gas to solids in a solar--composition disk is expected to be about 200 at 2.1 AU, since the ice component is expected to be evaporated. Thus the total surface density of the disk, $\\Sigma_{total}$, would have to be 1--2 $\\times 10^4$ g cm$^{-2}$, in order for the planet to form in a reasonable time.  If one examines the steady-state disk models of Bell et al. (\\cite{Bell}) with viscosity parameter $\\alpha = 10^{-2}$, one sees that a disk with that high a $\\Sigma_{total}$ at 2.1 AU would have a temperature of about 1000 K (consistent with the evaporation of ices) and an accretion rate onto the star of $\\dot M \\approx 10^{-5}$ M$_\\odot$ yr$^{-1}$.  Thus some difficulties with the model include (1) the required disk $\\dot M$ is much higher than the values of $\\sim 10^{-8}$ M$_\\odot$ yr$^{-1}$ in typical observed disks around young stars (Calvet et al.  \\cite{Calvet}), (2) such a disk is likely to be gravitationally unstable at larger radii (Bell et al. \\cite{Bell}) which could result in the formation of a much more massive planet at a distance of about 10 AU, and (3) $\\Sigma_{total}$ is about 20 times that in the `minimum mass' solar nebula (Hayashi et al.  \\cite{Hayashi}).  However a basic assumption of such disk models is that the ratio of gas to solids is constant at each radius as the disk evolves in time, except as modified by evaporation or condensation of solids. Here we consider alternate disk models, in which the evolution of $\\Sigma_{total}$ and $\\Sigma_s$ is not necessarily coupled. It has long been recognized that the solid particles in a disk evolve differently than the gas; for a review of the physical processes involved see Weidenschilling \\& Cuzzi (\\cite{Weidenschilling}). However global disk models in which the evolution of the solids and of the gas were followed consistently over timescales comparable to disk lifetimes (a few Myr) did not become available until Stepinski \\& Valageas (\\cite{SV96}, \\cite{SV97}) published some numerical solutions, based on a number of approximations in the physics.  The typical result of such calculations was a decoupling of the evolution of the solid component from that of the gas, once the particle size had become large enough. At the end of a simulation, a typical disk had a region, say from 1 to 10 AU, where the ratio of surface densities of solid and gas was considerably higher than that suggested by solar composition, and also regions where solids were practically absent. For some disk models, however, all of the solid material accreted onto the star. Results for a larger regime of parameter space for disk models were presented by Kornet et. al (\\cite{Kornet}), based on the methods of Stepinski \\& Valageas (\\cite{SV96}, \\cite{SV97}) but with further simplifications. These models start with a uniform, solar ratio of solids to gas, and evolve for $10^7$ yr, following the buildup of the initially small particles up to the size range 1--10 km.  The final stages of planet formation are not considered; a wide variety of possible distributions of solid material in evolved disks is found. The goal of the present paper is to investigate whether the formation, on a time scale of a few Myr, of both planets in the 47 UMa system can be explained with reasonable disk models, based on the calculations of Kornet et al. (\\cite{Kornet}). ",
        "conclusions": " (1) There exists a disk model which allows the formation of both of the planets in the 47 UMa system in about 3 Myr at their present distances from the star. (2) The initial disk may be significantly less massive than the one required by Bodenheimer et al.  (2000). At 2.1 AU their gas density had to be as high as $2.1 \\times 10^4$ g cm$^{-2}$, while the disk used in the present calculation had $\\Sigma_g = 10^{3}$ g cm$^{-2}$ at 2.1 AU at the beginning of the disk evolution. (3) The solid cores of both planets are relatively small, 21 and 16 M$_\\oplus$ for the inner and outer planet, respectively. In comparison, the core mass for the inner planet in model U2 of Bodenheimer et al. (2000), which formed in about 2 Myr, was 69 M$_\\oplus$.  (4) The planet at 2.1 AU formed in a much shorter time (2.7 Myr vs. 18.6 Myr) than that with the same assumed solid surface density in the the calculations of Bodenheimer et al. (2000b). The main reason is that the grain opacity in the present calculation is up to a factor 100 lower than that assumed by Bodenheimer et al.(2000b).  That effect is known to lead to shorter formation times. A test calculation for the inner planet was made in which the grain opacity was reset to interstellar values, as used by Bodenheimer et al. (2000b), all other effects remaining the same. The formation time turned out to be $1.2 \\times 10^7$ yr, more than a factor 4 longer than the 2.7 Myr obtained with the reduced opacity, but about 30\\% shorter than in Bodenheimer et. al (\\cite{BHL}). The latter difference is explained by the fact that the earlier calculation had different surface boundary conditions, a different procedure for calculating $\\dot M_Z$, and a higher core density (5 g cm$^{-3}$). (5) In the presence of the fully formed planet at 2.1 AU, there is still some solid material left in the disk between 1 and 1.5 AU, with $\\Sigma_s \\approx 40$ g cm$^{-2}$ and a mass of about 6 M$_\\oplus$. Formation of another giant planet in this region is not possible; a numerical calculation shows that the formation time would be much longer than the lifetime of the gas disk. The minimum solid surface density required to form Jupiter size planet in this region in 3 Myr is $\\sim 100$ g cm$^{-2}$ (see below).  These results lead to the question: in what kinds of disks is it possible to form Jupiter size planets on a 3 Myr time scale? We calculated approximate planet formation models, fitted to the results of Section 3, to determine, at various distances from a 1 M$_\\odot$ star, the minimum solid surface density $\\Sigma_{s,min} $ needed to form a giant planet in 3 Myr.  The results for $\\Sigma_{s,min} $ range from 100 g cm$^{-2}$ at 1.0 AU to 9 g cm$^{-2}$ at 5 AU to a minimum of 3 g cm$^{-2}$ at 15 AU. Then they increase slowly outward to 4.6 g cm$^{-2}$ at 50 AU. This result can be explained as follows. At small $R$ the protoplanet spends most of the evolutionary time in phase 2. The time scale for phase 2 depends strongly on the core mass (see Pollack et al. 1996 for a detailed explanation), so that in order to form the planet in a given time, say 3 Myr, the core mass must exceed some critical mass $M_{crit} \\sim 15 M_{\\oplus}$. Thus the necessary condition to form  the planet at a given location is that the isolation mass, $M_{iso}$ is not smaller then $M_{crit}$. According to Pollack et al. (1996) \\begin{equation} M_{iso}=C_1 (R^2 \\Sigma_s)^{3/2} \\label{miso} \\end{equation} Setting $M_{iso}=M_{crit}=\\mathrm{const}$ we recover the rapid increase of $\\Sigma_{s,min}$  toward the center of the disk as observed in numerical results.On the other hand, at large $R$ the evolutionary time scale is determined by the length of phase 1 in which the core is assembled. Here, the necessary condition for the planet to form within a prescribed time is that the mass of the core reaches $M_{crit}$. Setting $M_{crit}=\\mathrm{const}$ and employing formula (\\ref{mdot2}) we get \\[ \\Sigma_{s,min} \\sim R^{1/2}, \\] again in a rough agreement with the numerical results.   We then examined all the disk models shown in Fig.~\\ref{mass_rout} and \\ref{rin_rout} to find the range of distances $R_{min} \\leq R \\leq R_{max}$ at which the final $\\Sigma_s$ exceeds the local $\\Sigma_{s,min}$. Figure~\\ref{crit_2} shows the results for disk models with $\\alpha = 10^{-2}$, as a function of the fundamental disk parameters $m_0$ and $j_0$. The grey scale gives the values of $R_{max}$ and the contours give the values of $R_{min}$.  Note that the region of possible planet formation at less than 5 AU is very limited.  In fact we can estimate, consistent with this model, that the minimum radius at which giant planet formation is possible is about 2 AU.  Also, the maximum radius is estimated at 23 AU. In general, if $R_{min}$ is small, $R_{max}$ is also relatively small.  In models marked by a white circle, there is solid material present at the end of the evolution, but $\\Sigma_s$ is everywhere too low to form planets. In models marked by a diamond, all of the solid material accretes onto the star. Note that the disk models do not include ice grains, which will be included in future calculations. However the minimum radius for planet formation is not expected to change if ice is included.  Figure~\\ref{crit_3} shows the same results as Fig.~\\ref{crit_2} except for a disk viscosity parameter $\\alpha = 10^{-3}$. The minimum radius for planet formation in this case is somewhat smaller, about 1 AU, and the maximum is somewhat larger, about 25 AU. The main effect of the reduction in $\\alpha$ is a more extended region in which the dust survives (Stepinski \\& Valageas \\cite{SV97}). The inner radii of the final dust disk are smaller because the lower-$\\alpha$ disks have lower temperatures as a result of a smaller energy generation rate by viscosity at a given surface density. Thus the evaporation radius tends to be at smaller distances.  On the other hand, the outer radii of the dust disks are larger because the inward drift of particles is slower.  Another effect of reducing $\\alpha$ is to reduce the maximum $j_0$ for which planet formation is possible.  Generally as $j_0$ is increased for a given $m_0$, the gas and dust disks become more extended and therefore have lower average surface density.  For the same disk parameters except for a lower $\\alpha$, the dust disk is more extended, so the $\\Sigma_s$ falls everywhere below the critical value at smaller $j_0$.  In summary, the planets in the 47 UMa system can be explained by the core accretion -- gas capture process in a disk with reasonable parameters (mass 0.16 M$_\\odot$ and outer initial radius 40 AU) in which the dust component consists of high-temperature silicates.  The inclusion of ice grains in the model would probably make little difference in the planet formation process: a corresponding disk model calculated with only ice grains initially present resulted in the accretion of all of the ice onto the star.  The planets can reach their observed minimum masses in 3 Myr.  The main factor enabling their formation is that the surface density of solids in the region 2--4 AU is considerably higher than in the `minimum mass' solar nebula.  Migration of the planets as a result of gravitational interactions with the disk was not included.  However migration is still a possibility since with slightly changed disk parameters  the planets might be formed at much larger distances from the star (note that  in the present model the inner planet is located at about the minimum distance from the star where it is possible to form a giant planet in an $\\alpha = 10^{-2}$ disk).   Future improvements of the model should include (1) interactions between different types of solid particles, (2) particle models in which a range of particle sizes at each position in the disk is considered, (3) improvements in the dust opacity in the envelopes of the protoplanets, which influences the time scale of Phase 2,  and (4) a  better description of the boundary between the envelope of the planet and the disk; in particular, the presence of the secondary (circumplanetary) disk described by Ciecielag  et al. (\\cite{Ciecielag}) and D'Angelo  et al.  (\\cite{DAngelo02}). Finally, the effect of increased or reduced metal abundance on the region of parameter space in which planet formation is possible should also be investigated."
    },
    "0209/astro-ph0209425_arXiv.txt": {
        "abstract": "GK Per is a unique cataclysmic variable star which has showed a nova explosion as well as dwarf nova-type outbursts, and has the intermediate-polar nature.  We carried out $V$-band time-resolved photometry and $B$-band monitoring during the 1996 outburst.  This outburst lasted about 60 d and is divided into three parts: the slow rise branch for 35 d, the gradual decay branch with a decay rate of 20.0 d mag$^{-1}$ for $\\sim$16 d, and the rapid decline branch with a rate of 5.6 d mag$^{-1}$ for $\\sim$10 d.  The $B-V$ color became bluest ($B-V\\sim0.18$) about 10 d before the outburst maximum, which supports an idea that the outburst in GK Per is of the inside-out type.  The spin pulse, $440$-s quasi periodic oscillations (QPOs), and $\\sim$5,000-s QPOs were detected in our light curve, as previously seen in X-ray and optical observations.  In addition, we report the discovery of $\\sim$300-s periodicity, which is shorter than the spin period. ",
        "introduction": "Cataclysmic variable stars (CVs) are a group of binary stars consisting of a white dwarf (primary star) and a late-type secondary star (for a thorough review, \\cite{war95book}).  Surface gas of the secondary pouring out from its Roche lobe is transferred via the inner Lagrange point to the primary.  While the gas is accreted by the white dwarf through an accretion disk in usual CVs, the inner region of the disk is truncated or the accretion disk is completely obstructed to be formed in CVs containing a white dwarf with strong magnetic fields.  In this case, the gas is forced to move along the magnetic fields and is accreted into the magnetic poles on the white dwarf.  These magnetic CVs are classified into two subgroups of polars having a white dwarf with a spin period equal to the orbital period and intermediate polars (IPs) containing a white dwarf asynchronously rotating (for a basic review, e.g. chapter 8 and 9 in \\cite{hel01book}).  CVs show two kinds of large-amplitude sudden brightenings, namely, nova eruptions and dwarf nova-type outbursts.  The former are driven by thermonuclear runaway reaction of the material accreted on the white dwarf surface (\\cite{sta99novareview}, and references therein).  The amplitude is 8--15 mag.  The recurrence cycle is believed to be typically 10$^4$ yr or more, but this cycle highly depends on the primary mass, the mass transfer rate, the chemical abundance, and so on. For example, the shortest recurrence period of nova eruption in U Sco is 8 yr, with recorded outbursts in 1863, 1906, 1936, 1979, 1987 and 1999. On the contrary, the thermal instability in the accretion disk is generally accepted as the driving mechanism for the dwarf nova-type outburst (for a review of the disk instability model, e.g. \\cite{osa96review}).  The amplitude is 2--8 mag, and the recurrence cycle is several days--a few years.  GK Per is the only star which has the IP nature and has shown a nova eruption and dwarf nova-type outbursts.  On 1996 February 26, a new outburst was caught in this enigmatic CV by J. McKenna (vsnet-obs 2211). Following his notice, we started $V$-band time-resolved photometry and accompanying $B$-band monitoring.  The results are reported in this paper.  We briefly summarize the history of the GK Per study in the literature in section 2.  Section 3 and 4 are devoted to describe the observation and the result.  Discussion is made in section 4, and we summarize the paper in section 5. ",
        "conclusions": "Our $V$-band time-resolved photometry revealed the spin pulse ($P_{\\rm spin}$ = 351 s), and QPOs with three typical timescales of 300 s, 440 s, and 5,000 s.  The discovery of 300-s QPOs means that the inner edge of the accretion disk shrinks over the co-rotation period during outburst. The beat period of 440 s and 351 s is $\\sim$1,700 s, which would be against the model for the kilo-second QPOs.  Further check of contemporaneousness of hundred-second and kilo-second QPOs is required. The $B-V$ color became bluest ($B-V\\sim0.18$) about 10 d before the outburst maximum.  A similar color variation was observed during an anomalous outburst in SS Cyg, and this agrees with the idea that GK Per shows inside-out-type outbursts.  The decline rate during the rapid decline phase is smaller in GK Per, as well as DO Dra, and HT Cam, than expected from the $P_{\\rm orb}$.  This is qualitatively explained by truncation of the inner part of the accretion disk, but more detailed modeling taking into account the effect of the magnetic fields, 3D structures of the accretion disk, the mass stream, the accretion curtain, and the accretion pole is needed to reproduce the whole outburst properties of IPs.  \\vskip 3mm  We thank the anonymous referee for the useful comments.  The authors are very grateful to amateur observers for continuous reporting their valuable observations to VSNET.  Thanks are also to Emi Nakaoka for her help in searching for the literature.  This work is partly supported by a grant-in-aid (13640239) from the Japanese Ministry of Education, Culture, Sports, Science and Technology (TK)."
    },
    "0209/astro-ph0209431_arXiv.txt": {
        "abstract": "In this paper, we present the results of Monte Carlo simulations of atmospheric showers induced by diffuse $\\gamma$-rays as detected by the high-energy gamma-ray astronomy (HEGRA) system of five imaging atmospheric \\v{C}erenkov telescopes (IACTs). We have investigated the sensitivity of observations on extended $\\gamma$-ray emission over the entire field of view of the instrument. We discuss a technique to search for extended $\\gamma$-ray sources within the field of view of the instrument. We give estimates for HEGRA sensitivity of observations on extended TeV $\\gamma$-ray sources.  \\vspace{3 mm}  \\noindent {\\it PACS:} 95.55.Ka; 95.55.Vj; 96.40.Pq  \\noindent {\\it Keywords:} Imaging atmospheric \\v{C}erenkov technique, very-high-energy gamma-ray astronomy ",
        "introduction": "During the last decade, very-high-energy (VHE) $\\gamma$-ray astronomy ($E_\\gamma > 100$~GeV), which utilizes ground-based imaging atmospheric \\v{C}erenkov telescopes (IACTs), has made a substantial contribution to the $\\gamma$-ray astrophysics of a number of extra-galactic and galactic objects (for a review, see \\cite{rene_ong,weeks_catanese}). One of the reasons has been the tremendous progress in the observational technique. One can point out two major trends in this direction. The first is the use of imaging cameras with very fine pixellation (a pixel size of about 0.1$^\\circ$), equipped with fast electronics and an intelligent trigger, for a single stand-alone telescope, accomplished by Whipple \\cite{whipple1}, CAT \\cite{CAT1} and CANGAROO \\cite{cangaroo1}. Secondly, there has been the development of the stereoscopic observational technique with a number of IACTs with imaging cameras of relatively coarse pixellation (a pixel size of 0.25$^\\circ$), primarily by the high-energy gamma-ray astronomy (HEGRA) collaboration \\cite{HEGRA1}. Both trends have finally converged in three future experiments---H.E.S.S. \\cite{hess1}, CANGAROO III \\cite{cang4} and VERITAS \\cite{veritas1}, which are the systems of a 10~m class of telescope. Another major project, called MAGIC \\cite{magic1}, is a single telescope with a very large reflector of 17~m. Aspects of stereoscopic observations with three such telescopes have been discussed in \\cite{duo}.  For $\\gamma$-ray point sources, the sensitivity of the imaging atmospheric \\v{C}erenkov technique substantially relies on the angular resolution of the instrument. Note that the methods of cosmic-ray rejection based on the analysis of image shape are still not effective enough to reduce entirely the background of hadronic air showers. Thus, in addition, a good angular resolution significantly reduces the background contamination induced by the isotropic cosmic rays.  The observations of $\\gamma$-ray showers with a single telescope do not allow a complete geometrical reconstruction of the shower axis in space, because only one projection of a shower is available per event. A full reconstruction becomes possible in observations with two or more telescopes offering simultaneously a number of views of an individual shower. However, using advanced methods based on the strong correlation between the shape of the \\v{C}erenkov light images and their angular distances to the source position, one can achieve for a single IACT an angular resolution for an individual $\\gamma$-ray shower of about $0.12^\\circ$ \\cite{ulrich,lebohec1998,lessard2001} using a fine pixellation camera.  A substantial improvement of the angular resolution has been achieved by the HEGRA collaboration using a stereoscopic system of five IACTs with rather coarse camera pixellation. The stereoscopic observations with such a system allow us to reach, with good quality data, an angular resolution for an individual $\\gamma$-ray as good as 0.06$^\\circ$ \\cite{hofmann1}. The rich HEGRA data sample of $\\gamma$-rays from the Crab~Nebula \\cite{hegra_crab}, Mkn~501 \\cite{hegra_mkn501} and Mkn~421 \\cite{hegra_mkn421} have allowed us to prove in great detail this angular resolution, which is in good agreement with the predictions based on the Monte Carlo simulations (for details, see \\cite{HEGRA1}). Such an angular resolution has allowed us to perform a systematic search for point-like $\\gamma$-ray sources at the flux level of $10^{-11}\\, \\rm erg\\, cm^{-2} s^{-1}$. Such high sensitivity was confirmed by the detection of a very faint $\\gamma$-ray source, the supernova remnant (SNR) Cas~A, which steadily emits $\\gamma$-rays at the flux level of about $5.9\\times 10^{-13}\\, \\rm ph \\, cm^{-2} \\, s^{-1}$ above 1~TeV \\cite{hegra_casa}.  The HEGRA system of five IACTs has proved a very effective tool to search for TeV $\\gamma$-ray emission and to study the energy spectra of point-like sources, which are well established in observations at other wavelengths. However, the potential $\\gamma$-ray source may appear anywhere within the entire field of view (FoV) of the instrument, or it may have a rather large angular size as compared to the angular resolution of the instrument, for a certain number of important tasks, such as: (i) to search for $\\gamma$-ray emitters with poorly known position (such as the EGRET unidentified sources); ii) to perform sky surveys; (iii) to study extended $\\gamma$-ray emission from supernova remnants; (iv) to investigate diffuse emission from the Galactic plane; and finally, (v) to detect the primordial $\\gamma$-ray bursts. In the following, we investigate the sensitivity in detecting $\\gamma$-ray emission with the HEGRA system of IACTs in observations of this type. For this purpose, we have performed Monte Carlo simulations of diffuse $\\gamma$-rays as well as isotropic cosmic rays (see sections~2 and 3). The simulations are compared with real data taken with the HEGRA system of IACTs (section~4). An important issue is the efficiency of applying the orientational and shape cuts in order to distinguish the $\\gamma$-rays from the background isotropic cosmic rays. Here we study how this efficiency depends on the angular distance of a $\\gamma$-ray source to the centre of the FoV (section~5). In section~6, we discuss the analysis techniques to search for point-like and extended $\\gamma$-ray sources with the HEGRA system of IACTs, and finally in section~7 we give sensitivity estimates for shell-type supernova remnants. ",
        "conclusions": "The response of the HEGRA system of IACTs to the diffuse and extended $\\gamma$-ray emission over the FoV of the instrument was studied by means of detailed Monte Carlo simulations. Within the angular region limited by 1$^\\circ$ from the centre of the FoV the detection rate of the $\\gamma$-rays as well as the quality factor, characterizing the efficacy of the $\\gamma$-ray selection, are constant. Further extension of the observational window up to 1.5$^\\circ$ still allows us to have the same sensitivity to the $\\gamma$-ray fluxes but with noticeably reduced $\\gamma$-ray detection rate. An analysis of different trigger multiplicities reveals an improvement in the sensitivity, whereas higher multiplicities lead to the substantial decrease in the $\\gamma$-ray detection rate.  We have modelled the response of the HEGRA system of five IACTs for observations of nearby SNRs. Even though the final sensitivity estimate depends on the actual morphology of the TeV $\\gamma$-ray emitting region, in a simple case of a `circular' emission region the sensitivity might be derived by rescaling the sensitivity for a point-like source.  The studies discussed here could have a general use for the forthcoming arrays of IACTs such as CANGAROO, H.E.S.S.\\ and VERITAS.  \\noindent {\\it Acknowledgments.} This work was supported by CICYT (Spain) and BMBF (Germany).  \\newpage"
    },
    "0209/astro-ph0209607_arXiv.txt": {
        "abstract": "The results of {\\chandra} snapshot observations of 11 LINERs (Low-Ionization Nuclear Emission-line Regions), three low-luminosity Seyfert galaxies, and one \\ion{H}{2}-LINER transition object are presented. Our sample consists of all the objects with a flat or inverted spectrum compact radio core in the VLA survey of 48 low-luminosity AGNs (LLAGNs) by Nagar et al. (2000). An X-ray nucleus is detected in all galaxies except one and their X-ray luminosities are in the range $5\\times10^{38}$ to $8\\times10^{41}$ {\\eps}.  The X-ray spectra are generally steeper than expected from thermal bremsstrahlung emission from an advection-dominated accretion flow (ADAF). The X-ray to H$\\alpha$ luminosity ratios for 11 out of 14 objects are in good agreement with the value characteristic of LLAGNs and more luminous AGNs, and indicate that their optical emission lines are predominantly powered by a LLAGN. For three objects, this ratio is less than expected.  Comparing with properties in other wavelengths, we find that these three galaxies are most likely to be heavily obscured AGN.  We use the ratio $R_{\\rm X} = \\nu L_\\nu$(5 GHz)/$L_{\\rm X}$, where {\\LX} is the luminosity in the 2--10 keV band, as a measure of radio loudness. In contrast to the usual definition of radio loudness ($R_{\\rm O} = L_{\\nu}$(5 GHz)/$L_{\\nu}$(B)), $R_{\\rm X}$ can be used for heavily obscured ({\\NH} \\simgt $10^{23}$ {\\pcm}, $A_{\\rm V}>50$ mag) nuclei. Further, with the high spatial resolution of {\\chandra}, the nuclear X-ray emission of LLAGNs is often easier to measure than the nuclear optical emission. We investigate the values of $R_{\\rm X}$ for LLAGNs, luminous Seyfert galaxies, quasars and radio galaxies and confirm the suggestion that a large fraction of LLAGNs are radio loud. ",
        "introduction": "Low-Ionization nuclear emission-line regions (LINERs; Heckman 1980) are found in many nearby bright galaxies (e.g., Ho, Filippenko, \\& Sargent 1997a). Extensive studies at various wavelengths have shown that type 1 LINERs (LINER 1s, i.e., those galaxies having broad H$\\alpha$ and possibly other broad Balmer lines in their nuclear optical spectra) are powered by a low-luminosity AGN (LLAGN) with a bolometric luminosity less than $\\sim10^{42}$ {\\eps} (Ho et al. 2001; Terashima, Ho, \\& Ptak 2000a; Ho et al. 1997b). On the other hand, the energy source of LINER 2s is likely to be heterogeneous.  Some LINER 2s show clear signatures of the presence of an AGN, while others are most probably powered by stellar processes, and the luminosity ratio {\\LX}/{\\LHa} can be used to discriminate between these different power sources (e.g., P\\'erez-Olea \\& Colina 1996; Maoz et al. 1998; Terashima et al. 2000b).  It is interesting to note that currently there are only a few LINER 2s known to host an obscured AGN (e.g., Turner et al. 2001). This paucity of obscured AGN in LINERs may indicate that LINER 2s are not simply a low-luminosity extension of luminous Seyfert 2s, which generally show heavy obscuration with a column density averaging {\\NH} $\\sim$ $10^{23}$ {\\pcm} (e.g., Turner et al. 1997). Alternatively, biases against finding heavily obscured LLAGNs may be important. For example, objects selected through optical emission lines or X-ray fluxes are probably biased in favor of less absorbed ones, even if one uses the X-ray band above 2 keV.   In contrast, radio observations, particularly at high frequency, are much less affected by absorption. Although an optical spectroscopic survey must first be done to find the emission lines characteristic of a LLAGN, follow up radio observations can clarify the nature of the activity. For example, VLBI observations of some LLAGNs have revealed a compact nuclear radio source with $T_{\\rm b}>10^8$ K, which is an unambiguous indicator of the presence of an active nucleus and cannot be produced by starburst activity (e.g., Falcke et al. 2000; Ulvestad \\& Ho 2001). A number of surveys of Seyfert galaxies at sub arcsecond resolution have been made with the VLA (Ulvestad \\& Wilson 1989 and references therein; Kukula et al. 1995; Nagar et al. 1999; Thean et al. 2000; Schmitt et al.  2001; Ho \\& Ulvestad 2001) and other interferometers (Roy et al. 1994; Morganti et al. 1999), but much less work has been done on the nuclear radio emission of LINERs. Nagar et al. (2002) have reported a VLA 2 cm radio survey of all 96 LLAGNs within a distance of 19 Mpc. These LLAGNs come from the Palomar spectroscopic survey of bright galaxies (Ho et al. 1997a). As a pilot study of the X-ray properties of LLAGNs, we report here a {\\chandra} survey of a subset, comprising 15 galaxies, of Nagar et al's (2002) sample. Fourteen of these galaxies have a compact nuclear radio core with a flat or inverted radio spectrum (Nagar et al. 2000). We have detected 13 of the galactic nuclei with {\\chandra}. We also examine the ``radio loudness'' of our sample and compare it with other classes of AGN. A new measure of ``radio loudness'' is developed, in which the 5 GHz radio luminosity is compared with the 2--10 keV X-ray luminosity ($R_{\\rm X}$=$\\nu L_{\\nu}$ (5 GHz)/$L$(2--10 keV)) rather than with the B-band optical luminosity ($R_{\\rm O}= L_{\\nu}$(5 GHz)/$L_{\\nu}$(B)), as is usually done. $R_{\\rm X}$ has the advantage that it can be measured for highly absorbed nuclei ({\\NH} up to several times $10^{23}$ {\\pcm}) which would be totally obscured ($A_{\\rm V}$ up to a few hundred mag for the Galactic gas to dust ratio) at optical wavelengths, and that the compact, hard X-ray source in a LLAGN is less likely to be confused with emission from stellar-powered processes than is an optical nucleus.   This paper is organized as follows. The sample, observations, and data reduction are described in section 2. Imaging results and X-ray source detections are given in section 3. Section 4 presents spectral results. The power source, obscuration in LLAGNs, and radio loudness of LLAGNs are discussed in section 5. Section 6 summarizes the findings. We use a Hubble constant of $H_0=75$ km s$^{-1}$ Mpc$^{-1}$ and a deceleration parameter of $q_0=0.5$ throughout this paper. ",
        "conclusions": "\\subsection{Power Source of LINERs}  An X-ray nucleus is detected in all the objects except for NGC 4550 and NGC 5866. We test whether the detected X-ray sources are the high energy extension of the continuum source which powers the optical emission lines by examining the luminosity ratio {\\LX}/{\\LHa}. The H$\\alpha$ luminosities ({\\LHa}) were taken from Ho et al. (1997a) and the reddening was estimated from the Balmer decrement for narrow lines and corrected using the reddening curve of Cardelli, Clayton, \\& Mathis (1989), assuming the intrinsic H$\\alpha$/H$\\beta$ flux ratio = 3.1. The X-ray luminosities (corrected for absorption) in the $2-10$ keV band are used.  The H$\\alpha$ luminosities and logarithm of the luminosity ratios {\\LX}/{\\LHa} are shown in Table 5. The {\\LX}/{\\LHa} ratios of most objects are in the range of AGNs ($\\log$ {\\LX}/{\\LHa} $\\sim$ 1--2) and in good agreement with the strong correlation between {\\LX} and {\\LHa} for LLAGNs, luminous Seyferts, and quasars presented in Terashima et al. (2000a) and Ho et al. (2001).  This indicates that their optical emission lines are predominantly powered by a LLAGN. Note that this correlation is not an artifact of distance effects, as shown in Terashima et al. (2000a).  The four objects NGC 2787, NGC 4550, NGC 5866, and NGC 6500, however, have much lower {\\LX}/{\\LHa} ratios ($\\log$ {\\LX}/{\\LHa} \\simlt 0) than expected from the correlation ($\\log$ {\\LX}/{\\LHa} $\\sim1-2$), and their X-ray luminosities are insufficient to power the H$\\alpha$ emission (Terashima et al. 2000a). This X-ray faintness could indicate one or more of several possibilities such as (1) an AGN is the power source, but is heavily absorbed at energies above 2 keV, (2) an AGN is the power source, but is currently switched-off or in a faint state, and (3) the optical narrow emission lines are powered by some source(s) other than an AGN. We briefly discuss these three possibilities in turn.   If an AGN is present in these X-ray faint objects and absorbed in the hard energy band above 2 keV, only scattered and/or highly absorbed X-rays would be observed, and then the intrinsic luminosity would be much higher than that observed. This can account for the low {\\LX}/{\\LHa} ratios and high radio to X-ray luminosity ratios ($\\nu L_{\\nu}$(5 GHz)/{\\LX}; Table 5 and section 5.3). If the intrinsic X-ray luminosities are about one or two orders of magnitude higher than those observed, as is often inferred for Seyfert 2 galaxies (Turner et al. 1997, Awaki et al. 2000), {\\LX}/{\\LHa} and $\\nu L_{\\nu}$(5 GHz)/{\\LX} become typical of LLAGNs.   Alternatively, the AGN might be turned off or in a faint state, with a higher activity in the past being inferred from the optical emission lines, whose emitting region is far from the nucleus (e.g., Eracleous et al. 1995). Also, the radio observations were made a few years before the {\\chandra} ones.  This scenario might thus explain their relatively low {\\LX}/{\\LHa} ratios and their relatively high $L_{\\rm 5 GHz}$/{\\LX} ratios.  If this is the case, the size of the radio core can be used to constrain the era of the active phase in the recent past. The upper limits on the size of the core estimated from the beam size ($\\approx$ 2.5 mas) are 0.16, 0.19, and 0.48 pc for NGC 2787, NGC 5866, and NGC 6500, respectively (Falcke et al. 2000). Therefore, the AGN must have been active until $<$0.52, $<$0.60, and $<$1.6 years, respectively, before the VLBA observations (made in 1997 June) and inactive at the epochs (2000 Jan -- 2002 Jan, see Table 1) of the X-ray observations.  This is an ad hoc proposal and such abrupt declines of activity are quite unusual, but it cannot be completely excluded.    It may also be possible that the ionized gas inferred from the optical emission lines is ionized by some sources other than an AGN, such as hot stars.  If the observed X-rays reflect the intrinsic luminosities of the AGN, a problem with the AGN scenario for the three objects NGC 2787, NGC 5866, and NGC 6500 is that these galaxies have very large $\\nu L_{\\nu}$(5 GHz)/{\\LX} ratios, and would thus be among the radio loudest LLAGNs. The presence of hot stars in the nuclear region of NGC 6500 is suggested by UV spectroscopy (Maoz et al. 1998). Maoz et al. (1998) studied the energy budget for NGC 6500 by using the H$\\alpha$ and UV luminosity at 1300 A and showed that the observed UV luminosity is insufficient to power the H$\\alpha$ luminosity even if a stellar population with the Salpeter initial mass function and a high mass cutoff of 120$M_{\\odot}$ are assumed. This result indicates that a power source in addition to hot stars must contribute significantly, and supports the obscured AGN interpretation discussed above.    The first possibility, i.e., an obscured low-luminosity AGN as the source of the X-ray emission, seems preferable for NGC 2787, NGC 5866 and NGC 6500, although some other source(s) may contribute to the optical emission lines.  Additional lines of evidence which support the presence of an AGN include the fact that all three of these galaxies (NGC 2787, NGC 5866, and NGC 6500) have VLBI-detected, sub-pc scale, nuclear radio core sources (Falcke et al. 2000), a broad H$\\alpha$ component (in NGC 2787, and an ambiguous detection in NGC 5866; Ho et al. 1997b), a variable radio core in NGC 2787, and a jet-like linear structure in a high-resolution radio map of NGC 6500 with the VLBA (Falcke et al. 2000). Only an upper limit to the X-ray flux is obtained for NGC 5866. If an X-ray nucleus is present in this galaxy and its luminosity is only slightly below the upper limit, this source could be an AGN obscured by a column density {\\NH}$\\sim10^{23}$ {\\pcm} or larger. If the apparent X-ray luminosity of the nucleus of NGC 5866 is {\\it much} lower than the observed upper limit, and the intrinsic X-ray luminosity conforms to the typical {\\LX}/{\\LHa} ratio for LLAGN ($\\log$ {\\LX}/{\\LHa} $\\approx 1-2$), then the X-ray source must be almost completely obscured. The optical classification (transition object) suggests the presence of an ionizing source other than an AGN, so the low observed {\\LX}/{\\LHa} ratio could alternatively be a result of enhanced H$\\alpha$ emission powered by this other ionizing source.  The X-ray results presented above show that the presence of a flat (or inverted) spectrum compact radio core is a very good indicator of the presence of an AGN even if its luminosity is very low. On the other hand, NGC 4550, which does not possess a radio core, shows no evidence for the presence of an AGN and all the three possibilities discussed above are viable. If the {\\rosat} detection is real (Halderson et al. 2001), the time variability between the {\\rosat} and {\\chandra} fluxes may indicate the presence of an AGN (see Appendix).  It is notable that type 2 LINERs without a flat spectrum compact radio core may be heterogeneous in nature.  For instance, some LINER 2s without a compact radio core (e.g., NGC 404 and transition 2 object NGC 4569) are most probably driven by stellar processes (Maoz et al. 1998; Terashima et al. 2000b; Eracleous et al. 2002).   \\subsection{Obscured LLAGNs}  In our sample, we found at least three highly absorbed LLAGNs (NGC 3169, NGC 3226, and NGC 4548). In addition, if the X-ray faint objects discussed in section 5.1 are indeed AGNs, they are most probably highly absorbed with {\\NH}$>10^{23}$ {\\pcm}. Among these absorbed objects, NGC 2787 is classified as a LINER 1.9, NGC 3169, NGC 4548, and NGC 6500 as LINER 2s, and NGC 5866 as a transition 2 object. Thus, heavily absorbed LINER 2s, of which few are known, are found in the present observations demonstrating that radio selection is a valuable technique for finding obscured AGNs. Along with heavily obscured LLAGNs known in low-luminosity Seyfert 2s (e.g., NGC 2273, NGC 2655, NGC 3079, NGC 4941, and NGC 5194; Terashima et al. 2002a), our observations show that at least some type 2 LLAGNs are simply low-luminosity counterparts of luminous Seyferts in which heavy absorption is often observed (e.g., Risaliti, Maiolino, \\& Salvati 1999). However, some LINER 2s (e.g., NGC 4594, Terashima et al. 2002a; NGC 4374, Finoguenov \\& Jones 2001; NGC 4486, Wilson \\& Yang 2002) and low-luminosity Seyfert 2s (NGC 3147; section 4 and Appendix) show no strong absorption.  Therefore, the orientation-dependent unified scheme (e.g., Antonucci 1993) does not always apply to AGNs in the low-luminosity regime, as suggested by Terashima et al. (2002a).    \\subsection{Radio Loudness of LLAGNs}   Combination of X-ray and radio observations is valuable for investigating a number of areas of AGN physics, including the ``radio loudness'', the origin of jets, and the structure of accretion disks. Low-luminosity AGNs (LINERs and low-luminosity Seyfert galaxies) are thought to be radiating at very low Eddington ratios ($L_{\\rm bol}$/$L_{\\rm Edd}$) and may possess an advection-dominated accretion flow (ADAF; see e.g., Quataert 2002 for a recent review). A study of radio loudness in LLAGNs can constrain the jet production efficiency by an ADAF-type disk.  Earlier studies have suggested that LLAGNs tend to be radio loud compared to more luminous Seyferts based on the spectral energy distributions of seven LLAGNs (Ho 1999) and, for a larger sample, on the conventional definition of radio loudness $R_{\\rm O}=L_{\\nu}$(5 GHz)/$L_{\\nu}$(B) (the subscript ``O'', which stands for optical, is usually omitted but we use it here to distinguish from $R_{\\rm X}$ --- see below), with $R_{\\rm O}>10$ being radio loud (Kellermann et al. 1989, 1994; Visnovsky et al. 1992; Stocke et al. 1992; Ho \\& Peng 2001).  Ho \\& Peng (2001) measured the luminosities of the nuclei by spatial analysis of optical images obtained with {\\HST} to reduce the contribution from stellar light. A caveat in the use of optical measurements for the definition of radio loudness is extinction, which will lead to an overestimate of $R_{\\rm O}$ if not properly allowed for.  Although Ho \\& Peng (2001) used only type 1--1.9 objects, some objects of these types show high absorption columns in their X-ray spectra. In this subsection, we study radio loudness by comparing radio and hard X-ray luminosities. Since the unabsorbed luminosity for objects with {\\NH} \\simgt $10^{23}$ {\\pcm} can be reliably measured in the 2--10 keV band, which is accessible to {\\asca}, {\\it XMM-Newton}, and {\\chandra}, and such columns correspond to $A_{\\rm V}$ \\simgt 50 mag, it is clear that replacement of optical by hard X-ray luminosity potentially yields considerable advantages. In addition, the high spatial resolutions of {\\it XMM-Newton} and especially {\\chandra} usually allow the nuclear X-ray emission to be identified unambiguously, while the optical emission of LLAGN can be confused by surrounding starlight.  In the following analysis, radio data at 5 GHz taken from the literature are used since fluxes at this frequency are widely available for various classes of objects.  We used primarily radio luminosities obtained with the VLA at \\simlt $1^{\\prime\\prime}$ resolution for the present sample. High resolution VLA data at 5 GHz are not available for several objects. For four such cases, VLBA observations at 5 GHz with 150 mas resolution are published in the literature (Falcke et al. 2000) and are used here. For two objects, we estimated 5 GHz fluxes from 15 GHz data by assuming a spectral slope of $\\alpha=0$ (cf. Nagar et al. 2001). The radio luminosities used in the following analysis are summarized in Table 5.  Since our sample is selected based on the presence of a compact radio core, the sample could be biased to more radio loud objects. Therefore, we constructed a larger sample by adding objects taken from the literature for which 5 GHz radio, 2--10 keV X-ray, and $R_{\\rm O}$ measurements are available.   First, we introduce the ratio $R_{\\rm X} = \\nu L_{\\nu}$(5 GHz)/{\\LX} as a measure of radio loudness and compare the ratio with the conventional $R_{\\rm O}$ parameter. The X-ray luminosity {\\LX} in the 2--10 keV band (source rest frame), corrected for absorption, is used \\footnote{Monochromatic X-ray luminosities can also be used to define the radio loudness instead of luminosities in the 2--10 keV band; such a definition would be analogous to $R_{\\rm O}$, which utilizes monochromatic B-band luminosities.  This alternative provides completely identical results if the X-ray spectral shape is known and the range of spectral slopes is not large. For example, the conversion factor $L_{\\nu}$(2 keV)/{\\LX} is 0.31, 0.26, and 0.22 keV$^{-1}$ for photon indices of 2, 1.8, and 1.6, respectively, and no absorption.}. We examine the behavior of $R_{\\rm X}$ using samples of AGN over a wide range of luminosity, including LLAGN, the Seyfert sample of Ho \\& Peng (2001) and PG quasars which are also used in their analysis. $R_{\\rm O}$ parameters and radio luminosities were taken from Ho \\& Peng (2001) for the Seyferts and Kellermann et al. (1989) for the PG sample.  The values of $R_{\\rm O}$ in Kellermann et al. (1989) have been recalculated by using only the core component of the radio luminosities.  The optical and radio luminosities of the PG quasars were calculated assuming $\\alpha_{\\rm r}=-0.5$ and $\\alpha_{\\rm o}=-1.0$ ($S_{\\nu}\\propto \\nu^\\alpha$). The X-ray luminosities (mostly measured with {\\asca}) were compiled from Terashima et al. (2002b), Weaver, Gelbord, \\& Yaqoob (2001), George et al. (2000), Reeves \\& Turner (2000), Iwasawa et al. (1997, 2000), Sambruna, Eracleous, \\& Mushotzky (1999), Nandra et al. (1997), Smith \\& Done (1996), and Cappi et al. (1996). Note that only a few objects (NGC 4565, NGC 4579, and NGC 5033) in our radio selected sample have reliable measurements of nuclear $L_{\\nu}$(B).   Fig. 3. compares the parameters $R_{\\rm O}$ and $R_{\\rm X}$ for the Seyferts and PG sample. These two parameters correlate well for most Seyferts.  Some Seyferts have higher $R_{\\rm O}$ values than indicated by most Seyferts.  This could be a result of extinction in the optical band. Seyferts showing X-ray spectra absorbed by a column greater than $10^{22}$ cm$^{-2}$ (NGC 2639, 4151, 4258, 4388, 4395, 5252, and 5674) are shown as open circles in Fig. 3. At least four of them have a value of $R_{\\rm O}$ larger than indicated by the correlation. The correlation between $\\log R_{\\rm O}$ and $\\log R_{\\rm X}$ for the less absorbed Seyferts can be described as $\\log R_{\\rm O}$ = 0.88 $\\log R_{\\rm X}$ + 5.0.  According to this relation, the boundary between radio loud and radio quiet object ($\\log R_{\\rm O}$ = 1) corresponds to $\\log R_{\\rm X} = -4.5$.  The values of $R_{\\rm O}$ and $R_{\\rm X}$ for a few obscured Seyferts are consistent with the correlation, indicating that optical extinction is not perfectly correlated with the absorption column density inferred from X-ray spectra.  The PG quasars show systematically lower $R_{\\rm O}$ values than those of Seyferts at a given $\\log R_{\\rm X}$.  For the former objects, $\\log R_{\\rm O}=1$ corresponds to $\\log R_{\\rm X}=-3.5$.  This apparently reflects a luminosity dependence of the shape of the SED: luminous objects have steeper optical-X-ray slopes $\\alpha _{\\rm ox} = 1.4-1.7$ ($S\\propto\\nu^{-\\alpha}$; e.g., Elvis et al. 1994, Brandt, Laor, \\& Wills 2000), where $\\alpha _{\\rm ox}$ is often measured as the spectral index between 2200 A and 2 keV, while less luminous AGNs have $\\alpha _{\\rm ox} = 1.0-1.2$ (Ho 1999).  This is related to the fact that luminous objects show a more prominent ``big blue bump'' in their spectra.  Fig. 8 of Ho (1999) demonstrates that low-luminosity objects are typically 1--1.5 orders of magnitude fainter in the optical band than luminous quasars for an given X-ray luminosity. Note that none of the PG quasars used here shows a high absorption column in its X-ray spectrum below 10 keV.  The definition of radio loudness using the hard X-ray flux ($R_{\\rm X}$) appears to be more robust than that using the optical flux because X-rays are less affected by both extinction at optical wavelengths and the detailed shape of the blue bump, as noted above. Further, measurements of nuclear X-ray fluxes of Seyferts and LLAGNs with {\\chandra} are easier than measurements of nuclear optical fluxes, since in the latter case the nuclear light must be separated from the surrounding starlight, a difficult process for LLAGNs.   Fig. 4 shows the X-ray luminosity dependence of $R_{\\rm X}$. In this plot, the LLAGN sample discussed in the present paper is   \\begin{figure*}[t] \\includegraphics[scale=0.55,angle=-90]{fig3.ps} \\figcaption[] { Relation between $R_{\\rm O}=L_{\\nu}$(5 GHz)/$L_{\\nu}$(B) and $R_{\\rm X}=\\nu L_{\\nu}$(5 GHz)/{\\LX} for Seyferts and PG quasars.  The radio luminosity, $L_\\nu$(5 GHz), includes only the nuclear core ($<1^{\\prime\\prime}$ size) component of the radio emission. The conventional boundary between ``radio loud'' and ``radio quiet'' objects ($\\log R_{\\rm O}=1$) is shown as a horizontal dashed line.  The two open circles connected with a line correspond to X-ray observations of NGC 4258 at two different epochs. \\label{fig-3} } \\end{figure*}   \\begin{figure*}[t] \\includegraphics[scale=0.55,angle=-90]{fig4.ps} \\includegraphics[scale=0.55,angle=-90]{fig5.ps} \\figcaption[] { Dependence of $R_{\\rm X}=\\nu L_{\\nu}$(5 GHz)/{\\LX} on {\\LX} for the present LLAGN sample, Seyfert galaxies, radio galaxies, and PG quasars.  The radio luminosity, $L_\\nu$(5 GHz), includes only the nuclear core ($<1^{\\prime\\prime}$ size) component of the radio emission. The approximate boundary between ``radio loud'' and ``radio quiet'' objects ($\\log R_{\\rm x}=-4.5$) is shown as a horizontal dashed line. The two asterisks connected with a line correspond to X-ray observations of NGC 4258 at two different epochs. \\label{fig-4} } \\figcaption[]{ Dependence of $R_{\\rm X}=\\nu L_{\\nu}$(5 GHz)/{\\LX} on {\\LX} for the present LLAGN sample, Seyfert galaxies, radio galaxies, PG quasars, and other quasars, in which {\\it total} radio luminosities (instead of the nucleus-only radio luminosities used in Fig. 4) were used to calculated $R_{\\rm X}$.  The approximate boundary between ``radio loud'' and ``radio quiet'' objects ($\\log R_{\\rm x}=-4.5$) is shown as a horizontal dashed line. The two asterisks connected with a line correspond to X-ray observations of NGC 4258 at two different epochs. \\label{fig-5} } \\end{figure*}   \\noindent shown in addition to the Seyfert and PG samples used above. This is an ``X-ray version'' of the $\\log R_{\\rm O}$-$M_{B}^{\\rm nuc}$ plot (Fig. 4 in Ho \\& Peng 2001). Radio galaxies taken from Sambruna et al. (1999) are also added and we use radio luminosities from the core only. Our plot shows that a large fraction ($\\sim70$\\%) of LLAGNs ({\\LX}$<10^{42}$ {\\eps}) are ``radio loud''. This is a confirmation of Ho \\& Peng's (2001) finding. Note, however, that our sample is not complete in any sense, and this radio-loud fraction should be measured using a more complete sample. Since radio emission in LLAGNs is likely to be dominated by emission from jets (Nagar et al. 2001; Ulvestad \\& Ho 2001), these results suggest that, in LLAGN, the fraction of the accretion energy that powers a jet, as opposed to electromagnetic radiation, is larger than in more luminous Seyfert galaxies and quasars. Since LLAGNs are thought to have an ADAF-type accretion flow, such might indicate that an ADAF can produce jets more efficiently than the geometrically thin disk believed present in more luminous Seyferts.  The three LLAGNs with the largest $R_{\\rm X}$ in Fig. 4 are the three X-ray faint objects discussed in section 5.1 (NGC 2787, NGC 5866, and NGC 6500) and which are most probably obscured AGNs.  If their intrinsic X-ray luminosities are 1--2 orders of magnitude higher than those observed, their values of $R_{\\rm X}$ become smaller by this factor and are then in the range of other LLAGNs. Even if we exclude these three LLAGNs, the radio loudness of LLAGNs is distributed over a wide range: the radio-loudest LLAGNs have $R_{\\rm X}$ values similar to radio galaxies and radio-loud quasars, while some LLAGNs are as radio quiet as radio-quiet quasars.   A comparison with blazars is of interest to compare our sample with objects for which the nuclear emission is known to be dominated by a relativistic jet and thus strongly beamed.  The average $\\log R_{\\rm X}$ for high-energy peaked BL Lac objects (HBLs), low-energy peaked BL Lac objects (LBLs), and flat spectrum radio quasars (FSRQs) are --3.10, --1.27, and --0.95, respectively, where we used the average radio and X-ray luminosities for a large sample of blazars given in Table 3 of Donato et al. (2001). The average $\\log R_{\\rm X}$ for HBLs is similar to that for LLAGNs in our sample, while the latter two classes are about two orders of magnitude more radio loud than LLAGNs. Although LLAGNs and HBLs have similar values of $\\log R_{\\rm X}$, the spectral slope in the X-ray band is different: LLAGNs have a photon index in the range 1.7--2.0 (see also Terashima et al. 2002), while HBLs usually show steeper spectra (photon index $>$ 2, e.g., Fig. 1 in Donato et al. 2001), and the X-ray emission is believed to be dominated by synchrotron radiation. Furthermore, blazars with a lower bolometric luminosity tend to have a synchrotron peak at a higher frequency and a steeper X-ray spectral slope than higher bolometric luminosity blazars (Donato et al. 2001).   We also constructed an $R_{\\rm X}$-{\\LX} plot (Fig. 5) using the {\\it total} radio luminosities of the radio source (i.e. including the core, jets, lobes, and hot spots, if present). The radio data were compiled from V\\'eron-Cetty \\& V\\'eron (2001), Kellermann et al. (1989), and Sambruna et al. (1999). The PG sample and other quasars are shown with different symbols.  This plot appears similar to Fig. 4 for LLAGNs, Seyferts, and radio-quiet quasars since these objects do not possess powerful jets or lobes and off-nuclear radio emission associated with the AGN is generally of low luminosity (Ulvestad \\& Wilson 1989, Nagar et al. 2001, Ho \\& Ulvestad 2001, Kellermann et al. 1989). On the other hand, radio galaxies have powerful extended radio emission and consequently the $R_{\\rm X}$ values calculated using the total radio luminosities become higher than if only nuclear luminosities are used. We used the same X-ray luminosities as in Fig. 4, because jets, lobes, and hot spots are almost always much weaker than the nucleus in X-rays.  In fact, in our observations of LLAGNs, we found no extended emission directly related to the AGN. Thus, the differences between Fig. 4 and Fig. 5 result from the extended radio emission."
    },
    "0209/astro-ph0209577_arXiv.txt": {
        "abstract": "{We have observed the nearby isolated globule Barnard~68 (B68) in the \\onetwo\\ inversion lines of ammonia. The gas kinetic temperature derived from these is \\mbox{$T=10\\pm1.2~\\K$}. The observed line-widths are almost thermal: \\mbox{$\\DV=0.181\\pm0.003~\\kmps$} (\\mbox{$\\DVtherm= 0.164\\pm0.010~\\kmps$}), supporting the earlier hypothesis that B68 is in hydrostatic equilibrium. The kinetic temperature is an input parameter to the physical cloud model put forward recently, and we discuss the impact of the new value in this context. ",
        "introduction": "}  Discovered by \\citet{barnard19}, the isolated, starless globule Barnard~68 (B68) received increased attention recently, after \\citet{alves_01} presented a high resolution extinction map and a cloud model, suggesting that B68 has the physical structure of a so-called Bonnor-Ebert sphere (BES, \\ie\\ a pressure bound, isothermal sphere in hydrostatic equilibrium). With its column density and thus number density profiles revealed, B68 became an ideal object to study molecular depletion in dark core interiors \\citep[][hereafter Paper~I]{bergin_02,hotzel_02}, molecular abundances for a number of species \\citep{difrancesco_02} and the gas-to-dust ratio (Paper~I).  The BES cloud model as based on the measurements of \\citet{alves_01} fixes the normalised profiles of the density ($n/\\nc$) and the column density ($N/\\Nc$) as functions of the normalised radius ($r/R$), while for the absolute values of the central density \\nc, the central column density \\Nc\\ and the radius $R$ additional measurements are necessary. The knowledge on any two of the parameters \\nc, \\Nc, $D$ (distance) and $T$ (kinetic temperature) settles the others (Paper~I). The column density is linked to the extinction profile also via the \\emph{gas-to-dust ratio} (we use this term as short form for the more precise \\emph{hydrogen column density per unit reddening by dust}). Hence, if the BES model holds, one can determine the gas-to-dust ratio if $D$ and $T$ are known, or one can determine the distance to the cloud by measuring $T$ and applying a ``standard'' gas-to-dust ratio. In any case, the kinetic temperature remains the key parameter to measure. A reliable temperature measurement in cold, dense cores is possible using the 1.3~cm lines of ammonia \\citep{walmsley_83,danby_88}. However, previous ammonia measurements \\citep[\\mbox{$T=16$~K},][]{bourke_95b} are in contradiction to other temperature derivations \\citep[][Paper~I]{avery_87}.  Here we present new ammonia measurements, carried out with the Effelsberg 100-m telescope. Apart from the temperature derivation and an assessment of the inherent uncertainties, we compare the line-width to the width expected from purely thermal motion, which is a crucial test of the assumption that B68 is in hydrostatic equilibrium. We calculate the distance, gas-to-dust ratio and fractional ammonia abundance that follow from the BES model and the new temperature value. ",
        "conclusions": "}  Our observations support the hypothesis that B68 is in a state of isothermal hydrostatic equilibrium. Its kinetic temperature is \\mbox{$10\\pm1.2$~K} and its turbulent support is negligible. The ammonia abundance is close to the values found in other dark cores, but from the BES scaling relations we conclude that either the distance or the gas-to-dust ratio of B68 is smaller than previously expected."
    },
    "0209/astro-ph0209094_arXiv.txt": {
        "abstract": "{The intergalactic medium at redshifts 2--6 can be studied observationally through the absorption features it produces in the spectra of background quasars. Most of the UV-absorption lines arise in mildly overdense regions, which can be simulated reliably with current hydrodynamical simulations. Comparison of observed and simulated spectra allows one to put contraints on the model's parameters.}  \\resumen{El medio intergal\\'actico a redshifts 2--6 se puede estudiar observacionalmente gracias a los rasgos de absorci\\'on que produce en el espectro de los cuasares de fondo de campo. La mayori\\'a de las lineas de absorci\\'on UV aparecen en regiones ligeramente sobrepobladas, que pueden ser simuladas de manera fiable con las simulaciones hidrodin\\'amicas actuales. La comparaci\\'on del espectro observado con el simulado nos permite poner limites a los par\\'ametros del modelo.}  \\addkeyword{Cosmology: observations} \\addkeyword{Cosmology: theory} \\addkeyword{Galaxies: formation} \\addkeyword{Intergalactic medium} \\addkeyword{quasars: absorption lines}   \\begin{document} ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209027_arXiv.txt": {
        "abstract": "{We study thermal states of transiently accreting neutron stars (with mean accretion rates $\\dot{M} \\sim 10^{-14}-10^{-9}$ M$_\\odot$ yr$^{-1}$) determined by the deep crustal heating of accreted matter sinking into stellar interiors. We formalize a direct correspondence of this problem to the problem of cooling neutron stars. Using a simple toy model we analyze the most important factors which affect the thermal states of accreting stars: a strong superfluidity in the cores of low-mass stars and a fast neutrino emission (in nucleon, pion-condensed, kaon-condensed, or quark phases of dense matter) in the cores of high-mass stars. We briefly compare the results with the observations of soft X-ray transients in quiescence. If the upper limit on the quiescent thermal luminosity of the neutron star in SAX J1808.4--3658 (Campana et al.\\ \\cite{campanaetal02}) is associated with the deep crustal heating, it favors the model of nucleon neutron-star cores with switched-on direct Urca process. ",
        "introduction": "Great progress in observations of soft X-ray transients (SXRTs) in quiescence has attracted attention to these objects. We consider the SXRTs containing neutron stars (NSs) in binary systems with low-mass companions (low-mass X-ray binaries); see Chen et al.\\ (\\cite{csl97}) for a review. They undergo the periods of outburst activity (days--months, sometimes years) superimposed with the periods of quiescence (months--decades). This transient activity is regulated most probably by the regime of accretion from the disks around the NSs. An active period is associated with a switched-on accretion; the accretion energy released at the NS surface is high enough for a transient to look like a bright X-ray source ($L_X \\sim 10^{36}-10^{38}$ erg s$^{-1}$). The accretion is switched off or strongly suppressed during quiescence intervals when the NS luminosity drops by several orders of magnitude ($L_X \\la 10^{34}$ erg s$^{-1}$).  The nature of the quiescent emission is still a subject of debates. The hypothesis that this emission is produced by the thermal flux emergent from the NS interior has been rejected initially due to two reasons. First, the radiation spectra fitted by the blackbody model have given unreasonably small NS radii. Second, NSs in SXRTs have been expected to be internally cold; their quiescent emission should have been much lower than the observed one. These arguments were questioned by Brown et al.\\ (\\cite{bbr98}). They suggested that the NSs can be warmed up to the required level by {\\it the deep crustal heating} associated with nuclear transformations in accreted matter sinking in the NS interiors (Haensel \\& Zdunik \\cite{hz90}), while the radiation spectra can be fitted with the NS hydrogen atmosphere models (with realistic NS radii). It turned out that the emergent radiation flux may depend on the NS internal structure which opens an attractive possibility to explore the internal structure and the equation of state of dense matter by comparing the observations of SXRTs in quiescence with theoretical models (e.g., Ushomirsky \\& Rutledge \\cite{ur01}, Colpi et al.\\, \\cite{cgpp01}, Rutledge et al.\\ \\cite{rbbpzu02}, Brown et al.\\ \\cite{bbc02}, and references therein).  In this article, we discuss this possibility in more detail by making use of the close relationship between the theory of thermal states of transiently accreting NSs and the theory of cooling isolated NSs. Since the equation of state of NS matter is still poorly known we will make a general analysis of the problem with a simple toy model of NS thermal structure (described by Yakovlev \\& Haensel \\cite{yh02}). It will enable us to confront the theory with the observations without performing complicated calculations. ",
        "conclusions": "Our numerical calculations with the toy model confirm this qualitative analysis. The results are presented in Fig.\\ \\ref{fig3} which shows the surface photon luminosity versus accretion rate for low-mass and high-mass NSs.  The thick solid curve presents the deep heating power, $L_{\\rm dh}^\\infty$, which is the upper limit of $L_\\gamma^\\infty$ for any accreting source.  Moving from top to bottom, the next two lines refer to low-mass NSs with two types of slow neutrino emission appropriate either to neutron-neutron bremsstrahlung in the NS cores with a strong proton superfluidity ($Q_{\\rm s}= 3 \\times 10^{19}$) or to the Murca process in nonsuperfluid cores ($Q_{\\rm s}=10^{21}$). The strong proton superfluidity damps the Murca process and enables us to obtain hotter NSs, just as in the theory of cooling NSs (e.g., Kaminker et al.\\ \\cite{kyg02}).  The three next lines refer to high-mass NSs with three types of fast neutrino emission appropriate to (from top to bottom) kaon-condensed matter ($Q_{\\rm f}=10^{23}$), pion-condensed matter ($Q_{\\rm f}=10^{25}$), or nucleon matter with open Durca process ($Q_{\\rm f}=10^{27}$). NSs with hyperons cores are expected to cool at about the same rate as NSs with nucleon cores.  The heating curves of low-mass NSs provide the upper limit of $L_\\gamma^\\infty$, while the curves of high-mass stars give the lower limit of $L_\\gamma^\\infty$, for any particular equation of state (EOS) of NS interiors (for any set of the four neutrino emission parameters, in our case). Accordingly, the upper limits are shaded below the curves, and the lower limits are shaded above the curves. As in the cooling theory (Yakovlev \\& Haensel \\cite{yh02}), the upper and lower heating curves are almost insensitive to the position and width of the transition zone if $\\rho_{\\rm f}$ and $\\rho_{\\rm s}$ are located anywhere between $8 \\times 10^{14}$ and $1.2 \\times 10^{15}$ g cm$^{-3}$. The upper curve is determined by the parameter $Q_{\\rm s}$, while the lower one is determined by $Q_{\\rm f}$.  Varying the NS mass from the lowest values to the highest we obtain a family of heating curves which fill in the space in Fig.\\ \\ref{fig3} between the upper and lower curve for a given EOS of dense matter. The group of NSs whose heating curves lie essentially between the upper and lower curves will be called {\\it medium-mass} stars. Their central density falls within the transition layer between the slow and fast neutrino emission zones ($\\rho_{\\rm s} \\la \\rho \\la \\rho_{\\rm f}$, Sect.\\ 2), and their mass range is sensitive to the position and width of this layer.  The same situation occurs in the theory of cooling NSs (e.g., Yakovlev \\& Haensel \\cite{yh02}). Thus, for a given EOS of dense matter we obtain its own upper and lower heating curves, and intermediate heating curves of medium-mass stars.  In analogy with the cooling theory, the existence of a representative class of medium-mass stars (a smooth transition from the upper to the lower heating curves with increasing $M$) depends on the relative width of the transition zone, $\\Delta \\rho / \\rho_{\\rm s} \\equiv (\\rho_{\\rm f}-\\rho_{\\rm s})/\\rho_{\\rm s}$, and on the contrast between the fast and slow neutrino emissivities, $Q_{\\rm f}/Q_{\\rm s}$. To ensure the existence of this NS class for a sharp emissivity contrast, $Q_{\\rm f}/Q_{\\rm s} \\gg 10^3$, we need a rather wide transition zone, $\\Delta \\rho/\\rho_{\\rm s} \\ga 0.1$. On the other hand, this class will be available even for a negligibly narrow zone if the emissivity contrast is lower, $Q_{\\rm f}/Q_{\\rm s} \\la 10^3$.  These results can be confronted with the observations of SXRTs containing NSs. Although our theoretical toy model is oversimplified, in Fig.\\ \\ref{fig3} we present an example of such an analysis for five SXRTs: Aql X-1, Cen X-4, 4U 1608--522, KS 1731--26, and SAX J1808.4--3658. The data are rather uncertain. Thus we plot the observational points as thick dots.  \\begin{table*}[ht] \\caption{Parameters of NSs in SXRTs % } \\begin{center} \\begin{tabular}{|l|l||r|c|c|ll|} \\cline{1-2} \\cline{3-7} Source & $\\dot{M}$, ${\\rm M}_\\odot$ yr$^{-1}$ & $L_\\gamma^\\infty$, erg s$^{-1}$ & $T_s^\\infty$, eV & $R^\\infty$, km & \\multicolumn{2}{|c|}{ Reference} \\\\ \\cline{1-2} \\cline{3-7} Aql X-1 & $1.0\\times10^{-10}$ & $5.3\\times 10^{33}$ & 113 & 15.9 & Rutledge et al.\\ (\\cite{rbbpz02}) : & Table 6, fit 10 \\\\ Cen X-4 & $1.4\\times 10^{-10}$ & $3.1\\times 10^{32}$ & $\\;$76$^{\\ast )}$ & 12.9 & Rutledge et al.\\ (\\cite{rbbpz01}) : & Table 4 \\\\ 4U 1608--522 & $4.2\\times 10^{-10} $ & $4.1\\times 10^{33} $ & $\\;\\:$170$^{\\ast )}$ & 9.4 & Rutledge et al.\\ (\\cite{rbbpz99}) : & Table 2 \\\\ KS 1731--260 & $5.1\\times 10^{-9}$ & $4.3\\times 10^{33}$ & 110 & 15 & Wijnands et al.\\ (\\cite{wgvm02}) : & Table 1, fit 2 \\\\ SAX J1808.4--3658 & $5.0\\times 10^{-12}$ & $\\la 1.0\\times 10^{31}$ & --- & --- & Campana et al.\\ (\\cite{campanaetal02}) : & Sect.\\ 2.2 \\\\ \\cline{1-2} \\cline{3-7} \\multicolumn{7}{l}{$\\!\\!\\!^{\\ast )}$ nonredshifted }\\\\ \\end{tabular} \\end{center} \\label{tab:SXRTs} \\end{table*}  The parameters of the selected sources are collected in Table \\ref{tab:SXRTs}. The mean mass accretion rate $\\dot{M}$ is evaluated as $\\dot{M}=\\Delta M/\\Delta t$, where $\\Delta M$ is a total mass accreted over a representative period of time $\\Delta t$. Both, $\\Delta M$ and $\\Delta t$, should include active and quiescent periods, although $\\Delta M$ is mainly accumulated in the outburst states. In principle, we need $\\dot{M}$ averaged over thermal relaxation time scales, $\\sim 10^4$ yrs, while the observations provide us with sparse data over periods not longer than several decades. For Aql X-1, the mean $\\dot{M}$ has been estimated by Rutledge et al.\\ (\\cite{rbbpz00}) (their Sect.\\ 5, an estimate from the RXTE/ASM light curve history). For Cen X-4 and 4U 1608--522, we obtain $\\dot{M}$ from Table 9 in Chen et al.\\ (\\cite{csl97}). Following these authors, we take $\\Delta t=8.67$ yrs for 4U 1608--522, which is a frequently bursting source ($\\Delta M$ is estimated for 6 outbursts in the period from 1970 to 1979). We take $\\Delta t=33.16$ yrs for Cen X-4 (with $\\Delta M$ given for the only two outbursts in 1969 and 1979), adding the period from 1979 till now when no outbursts were observed. Since the active states may be very rare for this source, it is safer to consider the obtained $\\dot{M}$ as an upper limit. KS 1731--260 recently (about 1.5 years ago) returned into quiescence after having actively accreted for $\\approx 11.5$ years. For this source, we take $\\Delta t$=11.5+1.5=13 yrs. We estimate the mass $\\Delta M$ accreted during the long outburst state from the value of the mean outburst flux given by Rutledge et al.\\ (\\cite{rbbpzu02}, Sect.\\ 3.1). The estimation is made in the same manner as in Chen et al.\\ (\\cite{csl97}, Sect.\\ 5.1.4). Since the recurrence time is unknown, our value of $\\dot{M}$ is most probably an upper limit. Finally, $\\dot{M}$ for SAX J1808.4--3658 was estimated by Bildstein \\& Chakrabarty (\\cite{bc01}) and Campana et al.\\ (\\cite{campanaetal02}).  The values of $L_\\gamma^\\infty$ in Fig.\\ \\ref{fig3} are meant to refer to the quiescent thermal luminosity from the NS surfaces. For the first four sources, these values are obtained from the values of $T_{\\rm s}^\\infty$ and $R^\\infty$ given in Table \\ref{tab:SXRTs}. The values of $T_{\\rm s}^\\infty$ and $R^\\infty$ were evaluated by the authors cited in Table \\ref{tab:SXRTs} by fitting the observed spectra with the hydrogen atmosphere models. Note that the values of the surface temperature for Cen X-4 and 4U 1608--522 are nonredshifted. We have redshifted them assuming $M=1.4\\, {\\rm M}_\\odot$ and $R=12$ km. The spectrum of SAX J1808.4--3658 in quiescence is well fitted by a power law, i.e., no surface thermal emission has been detected. The estimates of the upper limit of $L_\\gamma^\\infty$ are model dependent and range from about $10^{30}$ erg s$^{-1}$ to $2.5 \\times 10^{31}$ erg s$^{-1}$ (Campana et al.\\ \\cite{campanaetal02}). We take $L_\\gamma^\\infty = 10^{31}$ erg s$^{-1}$ with the notice that the actual surface luminosity may be much lower.  If the interpretation of quiescent emission as the thermal emission from the NS surfaces is correct, then all five NSs are heated to the neutrino emission stage ($L_{\\rm dh}^\\infty > L_\\gamma^\\infty$). Since $L_{\\rm dh}^\\infty$ is reliably determined by the theory (Haensel \\& Zdunik \\cite{hz90,hz03}) for a known $\\dot{M}$, and $L_\\gamma^\\infty$ is measured, one can immediately estimate the neutrino luminosity of any source from the thermal balance, Eq.\\ (\\ref{Main}): $L_\\nu^\\infty=L_{\\rm dh}^\\infty-L_\\gamma^\\infty$. In all our cases $L_\\nu^\\infty$ is comparable with $L_{\\rm dh}^\\infty$ (Fig.\\ \\ref{fig3}).  As seen from Fig.\\ \\ref{fig3}, we can treat NSs in 4U 1608--52 and Aql X-1 as low-mass NSs with very weak neutrino emission from their cores (suppressed by strong nucleon superfluidity). The NSs in Cen X-4 and SAX J1808.4--3658 seem to require the enhanced neutrino emission and are thus more massive. The status of the NS in KS 1731--26 is less certain because of poorly determined $\\dot{M}$ (see above). If the real value of $\\dot{M}$ is close to that in Table \\ref{tab:SXRTs} it may also require some enhanced neutrino emission. Similar conclusions have been made by several authors (particularly, by Ushomirsky \\& Rutledge \\cite{ur01}; Colpi et al.\\ \\cite{cgpp01}; Rutledge et al.\\ \\cite{rbbpz01,rbbpzu02,rbbpz02}; Brown et al.\\ \\cite{bbc02}; Wijnands et al.\\ \\cite{wgvm02}) with respect to some of these sources or selected groups. Colpi et al.\\ (\\cite{cgpp01}) presented also the heating curves for specific models of low-mass and high-mass NSs with superfluid nucleon cores and suggested that by tuning nucleon superfluidity and NS masses one can explain the data. Using the toy-model, we can present a general analysis of the problem for different EOSs of NS interiors (assuming, of course, that all the sources have to be interpreted in terms of one EOS). In this way we can quantify the assumptions on enhanced neutrino emission in terms of pion condensed, kaon condensed, and Durca-allowed nucleon models of matter.  Disregarding the SAX source for the moment, we can treat the NS in Cen X-4 either as a high-mass NS (with a kaon-condensed or quark core) or as a medium-mass NS (with a pion-condensed, quark, or Durca-allowed nucleon core); thus we cannot determine the nature of superdense matter. If the data on SAX J1808.4--3658 are really relevant for our analysis we have the only choice to treat the NS as a high-mass NS with the nucleon core (and the NS in Cen X-4 as the medium-mass NS with the nucleon core). This would mean that NS cores do not contain exotic phases of matter.  Our toy model is too flexible and does not allow us to fix the position of the transition layer in the stellar cores where the slow neutrino emission transforms into the fast one. Adopting a specific EOS of NS interiors (with this position determined by microphysics input) we would be able to construct the sequences of heating curves for the stars with different $M$, and attribute certain values of $M$ to any source (``weigh'' NSs in SXRTs, as proposed by Colpi et al.\\ \\cite{cgpp01}, just as in the case of cooling isolated NSs considered, e.g., by Kaminker et al.\\ \\cite{kyg02}). We intend to do this in our future publications, using an exact cooling code and taking into account some effects neglected in our simplified approach.  In particular, we will account for the presence of light elements on the NS surfaces: they change the thermal conductivity of the NS heat-blanketing envelope and the relation between the surface and internal temperature of NSs. The effect is well known for cooling NSs (Potekhin et al.\\ \\cite{pcy97}) and has been applied recently to accreting NSs (Brown et al.\\ \\cite{bbc02}).  We will also carefully treat the effects of baryon superfluidity in NS interiors (just as for cooling NSs, see, e.g., Kaminker et al.\\ \\cite{kyg02}). Particularly, we will study the effects of Cooper-pairing neutrino emission of baryons neglected in the present analysis. Under certain conditions, this neutrino emission would violate our general assumptions on the neutrino emissivity $Q_\\nu(T,\\rho)$ and complicate our study. For instance, according to our estimates, the $^3$P$_2$ neutron superfluidity with the maximum values of the density dependent critical temperature $T_{\\rm cn}(\\rho)$ from $\\sim 10^8$ K to $\\sim 2 \\times 10^{9}$ K in the nucleon NS core would produce a powerful Cooper pairing neutrino emission and strongly affect the thermal states of accreting NSs. However, the same effect would initiate a really fast cooling of not too massive isolated NSs in contradiction with the observations of isolated cooling NSs (e.g., Kaminker et al.\\ \\cite{kyg02}). Thus the presence of the indicated neutron superfluidity can be rejected on these grounds.  Because of the similarity between the heating and cooling curves, the observations of cooling isolated NSs and accreting NSs in SXRTs can be analyzed together employing the same EOSs of NS interiors. This increases the statistics of the sources and the confidence of the results. The theory of cooling NSs has recently been confronted with observations by Yakovlev \\& Haensel (\\cite{yh02}). Some cooling NSs (first of all, RX J0822--43 and PSR 1055--52) can be interpreted (Kaminker et al.\\ \\cite{kyg02}) as low-mass NSs with strong proton superfluidity in their cores. Other sources (first of all, Vela and Geminga) seem to require a fast neutrino emission but the nature of this emission (a choice of fast-cooling model from Table 2) is uncertain, just as for SXRTs disregarding the data on SAX J1808.4--3658. In this context, the latter source is now the only one which indicates the absence of exotic phases of matter in NS cores.  The assumption that the observed X-ray emission of SXRTs in quiescence emerges from the NS interior is still an attractive hypothesis. In any case the theory of deep crustal heating is solid and leaves no doubts that this heating does occur in accreting NSs leading to observational consequences."
    },
    "0209/astro-ph0209161_arXiv.txt": {
        "abstract": "We discuss the flight calibration of the spectral response of the {\\em Advanced CCD Imaging Spectrometer} (ACIS) on-board the {\\em Chandra X-ray Observatory} (CXO).  The spectral resolution and sensitivity of the ACIS instrument have both been evolving over the course of the mission. The spectral resolution of the frontside-illuminated (FI) CCDs changed dramatically in the first month of the mission due to radiation damage.  Since that time, the spectral resolution of the FI CCDs and the backside-illuminated (BI) CCDs have evolved gradually with time.  We demonstrate the efficacy of charge-transfer inefficiency (CTI) correction algorithms which recover some of the lost performance.  The detection efficiency of the ACIS instrument has been declining throughout the mission, presumably due to a layer of contamination building up on the filter and/or CCDs.  We present a characterization of the energy dependence of the excess absorption and demonstrate software which models the time dependence of the absorption from energies of 0.4~keV and up. The spectral redistribution function and the detection efficiency are well-characterized at energies from 1.5 to 8.0~keV primarily due to the existence of strong lines in the ACIS calibration source in that energy range.  The calibration at energies below 1.5 keV is challenging because of the lack of strong lines in the calibration source and also because of the inherent non-linear dependence with energy of the CTI and the absorption by the contamination layer. We have been using data from celestial sources with relatively simple spectra to determine the quality of the calibration below 1.5 keV.  We have used observations of 1E0102.2-7219 (the brightest supernova remnant in the SMC), PKS2155-304 (a bright blazar), and the pulsar PSR~0656+14 (nearby pulsar with a soft spectrum), since the spectra of these objects have been well-characterized by the gratings on the CXO.  The analysis of these observations demonstrate that the CTI correction recovers a significant fraction of the spectral resolution of the FI CCDs and the models of the time-dependent absorption result in consistent measurements of the flux at low energies for data from a BI~(S3) CCD. ",
        "introduction": "\\label{sect:intro}  %  The Chandra X-ray Observatory (CXO) is the third of NASA's great observatories in space\\cite{weiss00,weiss02}.  The CXO was launched just past midnight on July 23, 1999 aboard the space shuttle \\textit{Columbia} on the STS-93 mission.  The CXO was placed into a higher orbit by an Inertial Upper Stage (IUS) booster and then used its own propulsion system to achieve a final orbit with a perigee of 10,000~km, an apogee of 140,000~km, an inclination of $28.5^\\circ$ and a period of $\\sim64$~hr. The CXO is controlled and operated by the Smithsonian Astrophysical Observatory (SAO) from Cambridge, Massachusetts.  The Chandra X-ray Center (CXC), also run by the Smithsonian Astrophysical Observatory, processes and distributes Chandra data and provides analysis software and calibration products to the astronomical community.    The CXO carries two focal plane science instruments: the \\textit{Advanced CCD Imaging Spectrometer} (ACIS) and the \\textit{High Resolution Camera} (HRC). The Observatory also possesses two objective transmission gratings: a \\textit{Low Energy Transmission Grating} (LETG) that is primarily used with the HRC, and the \\textit{High Energy Transmission Grating} (HETG) that is primarily used with the ACIS. ACIS was developed by a team from the Massachusetts Institute of Technology\\cite{bautz98} and the Pennsylvania State University \\cite{garmire92} and is the primary scientific instrument aboard CXO, currently conducting $\\sim90\\%$ of the observations. It contains two arrays of CCDs, one optimized for imaging and the other for spectroscopy as the readout detector for the HETG.  The ACIS imaging array contains 4 Front-Illuminated (FI) CCDs configured in a $2\\times2$ array and the spectroscopy array contains 2 Back-Illuminated (BI) CCDs and 4~FI~CCDs configured in a $1\\times6$ array. ",
        "conclusions": ""
    },
    "0209/astro-ph0209482_arXiv.txt": {
        "abstract": "Growing observational evidence supports the proposition that gamma-ray bursts (GRBs) are powered by relativistic jets from massive helium stars whose cores have collapsed to black holes and an accretion disk (collapsars).  We model the propagation of relativistic jets through the stellar progenitor and its wind using a two-dimensional special relativistic hydrodynamics code based on the PPM formalism.  The jet emerges from the star with a plug in front and a cocoon surrounding it.  During its propagation outside the star, the jet gains high Lorentz factor as its internal energy is converted into kinetic energy while the cocoon expands both outwards and sideways.  External shocks between the cocoon and the stellar wind can produce $\\gamma$-ray and hard x-ray transients.  The interaction of the jet beam and the plug will also affect both of them substantially, and may lead to short-hard GRBs. Internal shocks in the jet itself may make long-soft GRBs. ",
        "introduction": "Although there is no universally agreed upon central engine powering gamma-ray bursts (GRBs), growing evidence supports the association of least the long-soft GRBs (all those whose counterparts have been localized) with the death of massive stars.  This evidence includes the association of GRBs with regions of massive star formation (Bloom, Kulkarni, \\& Djorgovski 2002) and ``bumps'' in the optical afterglows of several GRBs that have been related to the light curves of Type I supernovae (e.g., Bloom et al. 2002; Garnavich et al. 2002).  In addition, GRB 980425 has been associated with SN 1998bw (e.g., Iwamoto et al. 1998; Woosley, Eastman, \\& Schmidt 1999).  Frail et al. (2001), Panaitescu \\& Kumar (2001) have studied beaming angles and energies of a number of GRBs and have found that the central engines of GRBs release supernova-like energies.  Given these discoveries, the currently favored models are all based upon the collapse of massive stars and their byproducts, one of which is a relativistic jet.  Among those models involving massive stars, the collapsar model (Woosley 1993; MacFadyen \\& Woosley 1999) has become a favorite.  A ``collapsar'' is a rotating massive star whose iron core has collapsed and formed a black hole and an accretion disk.  In this model, the central black hole and the disk use neutrinos or magnetic fields to extract part of the gravitational potential or rotational energy and form powerful relativistic jets along the polar axes.  Jets from collapsars have been studied numerically in both Newtonian (MacFadyen \\& Woosley 1999; MacFadyen, Woosley, \\& Heger 2001) and relativistic simulations (Aloy et al. 2000; Zhang, Woosley, \\& MacFadyen 2002) and it has been shown that the collapsar model is able to explain many of the observed characteristics of GRBs.  These previous studies have also raised issues which require further examination, especially with higher resolution.  For instance, the emergence of the jet and its interaction with the material at the stellar surface and the stellar wind will definitely lead to some sort of ``precursor'' activity.  The long term dynamics of the jet is critical in producing the observed gamma-rays and afterglows.  We have recently carried out multi-dimensional calculations to address some of these issues. ",
        "conclusions": "Our special relativistic calculations show that a jet originating near the center of a collapsing massive will emerge while still carrying almost all the energy injected at the center.  After it breaks out of the star, the jet will move forward almost freely, no more cocoon will be generated.  The standard fireball model requires not only high Lorentz factor, but also the Lorentz factor to vary rapidly.  Our simulations show promise in satisfying these constraints.  However, due to the low resolution ($\\Delta z = 1.6 \\times 10^{9}\\,{\\mathrm cm}$) outside the star, it is hard to resolve those small scale variabilities and the numerical viscosity gradually smooths the time structure in the jet.  We have found that the cocoon can expand sideways up to $\\sim30{\\deg}$ after it emerges from the star.  The deceleration of the cocoon by external shocks will produce $\\gamma$-rays and x-rays (see also Ramirez-Ruiz et al. 2002).  The ``anomalous'' GRB 980425 might be a cocoon viewed from a large angle.  It might also contribute to the afterglows and precursors of a normal GRB and may be the origin of recently discovered hard x-ray flashes (Heise et al. 2001; Kippen et al 2002). These flashes have many properties of GRBs (energy, isotropy, approximate duration, distribution with redshift), but have a much softer spectrum (and so far no optical afterglow). The association of this kind of softer transient with gamma-ray bursts seen off-axis has been predicted by our group for many years (Woosley et al 1999; Woosley \\& MacFadyen 1999; Woosley 2000, 2001).  Behind the bow shock, there is a plug in front of the jet beam.  After breakout, the main jet beam continues to push and accelerate this plug. Meanwhile, the jet beam is slowed down as its energy is transfered to the plug and part of its kinetic energy is converted into its internal energy.  The basic picture of a fast moving jet beam pushing a ``slowly'' moving plug is that there is a reverse shock, which moves backward and slows down the jet beam, a contact discontinuity, and a forward shock.  Since the plug has a limited length, the forward shock will reflect at the end of the plug. This further complicates the dynamics (e.g., Waxman \\& M\\'{e}sz\\'{a}ros 2002).  At the end of our simulation, the jet and the plug are moving relativistically and the stellar wind has not been able to decelerate the jet, so we can make some analytic estimates treating the jet and the plug as a spherical symmetric fireball.  Where will the jet and the plug become optically thin?  In our case, the opacity in the jet and the plug is dominated by Thomson scattering.  The optical depth is $\\tau = \\kappa\\,\\Sigma = \\kappa\\,\\rho\\,\\Delta r = \\kappa\\,(M / 4 \\pi r^2) = \\tau_0\\,(r_0/r)^2$, here, $\\kappa \\sim 0.2\\,{\\mathrm cm}^2\\,{\\mathrm g}^{-1}$, and $\\tau_0$ is the initial optical depth at $r_0$.  From our results at $t = 70\\,{\\mathrm s}$, the plug will become optically thin at $r_{\\mathrm th,p} = 2.0\\times 10^{14}$, $2.6 \\times 10^{14}$, and $1.7 \\times 10^{14}\\,{\\mathrm cm}$, for Models A, B, and C, respectively. And the jet will become optically thin at $r_{\\mathrm th,j} = 5.8\\times 10^{14}$, $7.6 \\times 10^{14}$, and $2.9 \\times 10^{14}\\,{\\mathrm cm}$, for Models A, B, and C, respectively.  When it becomes optically thin, the interaction of the plug with the jet, will produce hard emission.  In fact, it is possible that this emission could be a short-hard GRB (as defined in Fishman \\& Meegan 1995).  Where will the jet catch up the plug and be decelerated by the plug? This is a very critical question. During the catch-up period, some of the kinetic energy of the jet is converted into internal energy.  If the deceleration happens in the optically thick regime, the increased internal energy can still be converted back into kinetic energy.  If, however, it happens in the optically thin regime, the increased internal energy will become radiation energy via this special kind of ``internal shock'' and escape from the jet.  There may not be enough kinetic energy left and high enough Lorentz factor to make $\\gamma$-rays via ``normal'' internal shocks.  In this case, the short hard precursors from the plug may dominate at X-ray and $\\gamma$-ray wavelengths, and a short hard GRB is likely to be seen.  Numerical simulations on the catch-up process are currently underway.  We analytically estimate the radius, $r_{\\mathrm cat}$ , at which the catch-up process will happen.  For simplicity we assume that the jet moves at a Lorentz factor of 100, the plug moves at a Lorentz factor of 20, and the length of the jet is about $5 \\times 10^{11}\\,{\\mathrm cm}$, then we get $r_{\\mathrm cat} \\sim 5 \\times 10^{14}\\,{\\mathrm cm}$.  This radius is comparable to the radius where the jet becomes optically thin.  Our simulations show a tendency for the energy in the plug relative to that in the jet behind the plug to be larger when the total energy is less (see \\S~3.2).  This suggests that less energetic jets may be more likely to make short-hard GRBs. The near coincidence of the masses of the plug and jet and of the radii where the two share their energy with the gamma-ray photosphere suggests that there may be cases where the most prominent display comes from the plug and others where it still comes from internal shocks in the jet (or at the jet-plug interface).  Where will the jet be decelerated by the stellar wind?  The deceleration by the stellar wind happens when the jet sweeps up $1/\\Gamma$ of its rest mass, here $\\Gamma$ is the Lorentz factor of the jet.  Assuming a stellar wind that has a mass loss rate of $\\sim 10^{-5}\\,M_{\\sun}\\,{\\mathrm yr}^{-1}$ and a velocity of $\\sim 1000\\,{\\mathrm km}\\,{\\mathrm s}^{-1}$ at $10^{11}\\,{\\mathrm cm}$, the deceleration radius, $r_{\\mathrm dec}$, is $3 \\times 10^{16}\\,{\\mathrm cm}$ for Model A.  And they are $4 \\times 10^{16}$ and $8 \\times 10^{15}\\,{\\mathrm cm}$, for Models B and C, respectively.  The deceleration by external shocks with the stellar wind always happens after the above events. This justifies our one-dimensional analytic calculations because the effects of the sideways expansion are negligible when the jets are highly relativistic.  In our numerical simulations, We have followed the propagation of jets to $r=2 \\times 10^{12}\\,{\\mathrm cm}$.  However, many interesting events which are directly related to observations happen at large radius. So the long time evolution of the jets in the stellar wind is very critical.  It will help us make many testable predictions. Numerical calculations which follow the long time evolution of the jets are under way.  Hopefully, they will further strengthen our knowledge on relativistic outflows of GRBs.  It is also very important to repeat our simulations in three-dimensional Cartesian coordinates."
    },
    "0209/astro-ph0209481.txt": {
        "abstract": "We have generated a set of far-ultraviolet stellar libraries using spectra of OB and Wolf-Rayet stars in the Galaxy and the Large and Small Magellanic Cloud. The spectra were collected with the {\\em Far Ultraviolet Spectroscopic Explorer} and  cover a wavelength range from 1003.1 to 1182.7~\\AA\\ at a resolution of 0.127~\\AA. The libraries extend from the earliest O- to late-O and early-B stars for the Magellanic Cloud and Galactic libraries, respectively. Attention is paid to the complex blending of stellar and interstellar lines, which can be significant, especially in models using  Galactic stars. The most severe contamination is due to molecular hydrogen. Using a simple model for the H$_2$ line strength, we were able to remove the molecular hydrogen lines in a subset of Magellanic Cloud stars. Variations of the photospheric and wind features of \\ciii\\,$\\lambda$1176, \\ovi\\,$\\lambda\\lambda$1032,~1038, \\pv\\,$\\lambda\\lambda$1118,~1128, and \\siv\\,$\\lambda\\lambda$1063,~1073,~1074 are discussed as a function of temperature and luminosity class. The spectral libraries were implemented into the LavalSB and Starburst99 packages and used to compute a standard set of synthetic spectra of star-forming galaxies. Representative spectra are presented for various initial mass functions and star formation histories. The valid parameter space is confined to the youngest ages of less than $\\simeq$10~Myr for an instantaneous burst, prior to the age when incompleteness of spectral types in the libraries sets in. For a continuous burst at solar metallicity, the parameter space is not limited.  The suite of models is useful for interpreting the restframe far-ultraviolet in local and high-redshift galaxies. ",
        "introduction": "Modeling of synthetic line profiles in stellar populations has traditionally focused on the optical, near-infrared (IR), and satellite-ultraviolet (UV) wavelength regions. Since most of the models are empirical, i.e., they utilize observational stellar libraries, or they are theoretical with the need for observational calibrations, their wavelength coverage is determined by the available body of observations.  Examples of this type of work are in the compilation of Leitherer et al. (1996). As a result of past and ongoing efforts, line-profile synthesis models cover the wavelength range from 1200~\\AA\\ to 2.3~$\\mu$m. The lower cut-off is due to the transmission of the entrance windows and the optical coatings of most UV spectrographs, such as {\\it IUE} and those of {\\it HST}, and the upper limit defines the regime where near-IR observations become background-, rather than detector-limited.  Few strong features exist in the spectra of young stellar populations longward of the K-band, and those that would in principle be observable are normally totally diluted by galactic gas and dust emission longward of the L-band (e.g., Genzel \\& Cesarsky 2000). Consequently there is little practical interest in pushing the profile synthesis work to longer wavelengths. The situation is different at short wavelengths. The spectral region of a young population between 912 and 1200~\\AA\\ contains a rich absorption- and emission-line spectrum which provides important clues on the star-formation history (e.g., Gonz\\'alez Delgado, Leitherer, \\& Heckman 1997; Leitherer et al. 2002). For instance, the strong resonance doublet of \\ovi\\,$\\lambda$1032, 1038 is often stronger than the commonly used star-formation indicators \\siiv\\,$\\lambda$1400 and \\civ\\,$\\lambda$1550 (Robert, Leitherer, \\& Heckman 1993).  Previously, the usefulness of the wavelength range 912~--~1200~\\AA\\ was limited due to the lack of observational data. Pioneering work by the Copernicus satellite (Rogerson et al. 1973) dramatically increased our understanding of stellar spectra but extragalactic sources were too faint for observation.  The Voyager~1 and 2 spacecraft (Broadfoot et al. 1977) had similar limitations.  More recently, the {\\it ORFEUS} mission (Grewing et al. 1991) explored the wavelength region below 1200~\\AA, but again, few faint objects could be observed due to brightness limitations. The Hopkins Ultraviolet Telescope ({\\it HUT}; Davidsen 1993) was the first instrument sensitive enough to collect astrophysically useful spectra of faint galaxies in the wavelength range below L$\\alpha$. The short mission duration of two weeks precluded the built-up of a significant database. Nevertheless, the spectra of star-forming galaxies obtained with {\\it HUT} demonstrated the potential of this wavelength for our understanding of star-forming galaxies (Leitherer et al. 2002).  Renewed interest in the far-UV region of star-forming galaxies is generated by two major science drivers. First, the launch of the multi-year Far Ultraviolet Spectroscopic Explorer (\\fuse) mission (Moos et al. 1998) is producing high-quality far-UV spectra of numerous star-forming galaxies which await interpretation with spectral synthesis modeling. Second, ground-based 10-m class telescopes can observe the restframe far-UV spectra of star-forming galaxies at $z>3$ (Pettini et al. 2001). These spectra reach down to the Lyman limit and achieve a signal-to-noise rivaling that of UV spectra of their local counterparts.  As part of a \\fuse\\ Guaranteed Time Observer (GTO) program, spectra of more than 200 hot, luminous stars in the Galaxy and in the Magellanic Clouds have been collected.  These data, while astrophysically useful in their own right, are well suited to extend currently existing spectral libraries longward of 1200~\\AA\\ down to the Lyman limit. The individual spectra are discussed by Pellerin et al. (2002) and Walborn et al. (2002a). In the present paper we describe the creation of the far-UV spectral libraries, their implementation into existing evolutionary synthesis codes, and a parameter study using a standard grid of synthetic spectra produced with the new libraries. ",
        "conclusions": "We have created stellar spectral libraries of hot stars in the wavelength range 1003~--1183~\\AA\\ from \\fuse\\ archival data. These new libraries complement our earlier work in the satellite-UV longward of L$\\alpha$, providing almost continuous wavelength coverage from 1000 to 1800~\\AA. In comparison with the longer wavelength {\\it HST} data, the \\fuse\\ spectra have 10 times higher %Claus I don't see the point to compare this with IUE resolution and at the same time about the same S/N. The library stars are located in the Milky Way, the LMC, and the SMC and extend over a metallicity range of almost a factor of 10.  The spectral region below 1200~\\AA\\ shows strong line-blanketing due to stellar-wind, stellar photospheric, and interstellar lines. The stellar features generally originate from higher ionization stages than the features found above 1200~\\AA. The most prominent transition is the \\ovi\\ resonance doublet at 1032, 1038~\\AA\\ which displays a spectacular P~Cygni profile over a broad range of spectral types. At the resolution afforded by \\fuse, the blueshifted absorption component of the P~Cygni profile is resolved from nearby L$\\beta$ and can be distinguished from the narrow interstellar \\cii\\ at 1036~\\AA. The (redshifted) emission component of its P~Cygni profile is relatively unaffected by interstellar lines and provides additional diagnostic power. The \\ciii\\ $\\lambda$1176 line is at the long-wavelength end of the covered spectral range and can also be observed with spectrographs optimized for wavelengths longward of L$\\alpha$. Surprisingly, the line has received relatively little attention in the earlier literature. We find it a very good diagnostic of the properties of hot stars. \\ciii\\ is not a resonance transition, and consequently does not suffer from contamination by an interstellar component. \\ciii, like most other stellar lines, has a pronounced metallicity dependence, either directly via opacity variations, or indirectly via metallicity dependent stellar-wind properties.  The \\fuse\\ wavelength range is particularly rich in interstellar lines from molecular, atomic, and ionic transitions. Even at the spectral resolution of \\fuse, the blending of stellar and interstellar features can be complex, and care is required when interpreting the spectra. The numerous transitions of molecular hydrogen are dominant in the Galactic stars, but less so in the LMC and SMC stars. The lower column densities associated with the Clouds have allowed us to model and remove the H$_2$ lines in a subset of the LMC/SMC stars and to generate library stars virtually uncontaminated by H$_2$.  The libraries for different metallicities were integrated into the LavalSB and Starburst99 synthesis codes. A suite of standard synthetic spectra was generated to study the basic properties of stellar population spectrum as a function of the most relevant parameters. These model sets serve as a baseline for comparison with young galaxy spectra, both observed locally in the far-UV or in the distant universe when redshifted into the visual wavelength range. Such a comparison will provide insight into the properties of the stellar content and of the opacity of the intervening intergalactic medium. Additionally, fully theoretical far-UV line spectra for stars in any position on the Hertzsprung-Russell diagram will soon become widely available. Our empirical set of spectra can provide tests and guidelines for such theoretical approaches.  The readers are encouraged to explore a broader parameter range than discussed in this paper by visiting www.stsci.edu/science/starburst99/ and running a set of tailored models."
    },
    "0209/hep-ph0209063_arXiv.txt": {
        "abstract": "name}{\\large\\bfseries Abstract}   \\newcommand{\\FNAL}{Fermi National Accelerator Laboratory, P.O. Box 500, Batavia, IL 60510, USA} \\newcommand{\\OX}{Department of Physics, Theoretical Physics, University of Oxford, Oxford OX1\\hspace{0.2em}3NP, UK} \\newcommand{\\SNS}{Scuola Normale Superiore and INFN, Sezione di Pisa, \\\\ I--56126 Pisa, Italy} \\newcommand{\\PISA}{Dipartimento di Fisica, Universit\\`a di Pisa and INFN, Sezione di Pisa, I--56126 Pisa, Italy} \\newcommand{\\LBNL}{Department of Physics and Lawrence Berkeley National Laboratory \\\\ University of California, Berkeley, California 94720, USA} \\newcommand{\\DPNC}{Departement de Physique Nucleaire et Corpuscolaire, Universit\\'e de Gen\\`eve CH--1211 Gen\\`eve 4, Switzerland}   \\newcommand{\\preprintdate}{Month yyyy} \\newcommand{\\preprintnumber}{% SNS--PH/yy--nn} \\newcommand{\\hepnumber}{hep-ph/0209063} \\newcommand{\\titletext}{Bulk neutrinos and core collapse supernovae} \\newcommand{\\authortext}{\\large G. Cacciapaglia, M. Cirelli, Y. Lin, A. Romanino \\medskip\\\\\\em\\normalsize \\SNS} \\newcommand{\\abstracttext}{We discuss the phenomenology of neutrino mixing with bulk fermions in the context of supernova physics. The constraints on the parameter space following from the usual energy loss argument can be relaxed by four orders of magnitude due to a feedback mechanism that takes place in a broad region of the parameter space. Such a mechanism also affects the protoneutron star evolution through a non trivial interplay with neutrino diffusion.  The consistency with the \\SN\\ signal is discussed, as well as the implications for deleptonization, cooling, composition of the neutrino flux and the delayed explosion scenario.}   \\title{ \\normalsize \\LARGE\\bfseries\\titletext\\bigskip} \\author{\\begin{minipage}[t]{0.8\\textwidth} \\normalsize\\centering\\authortext \\end{minipage}} \\date{}  \\begin{document}  \\bigskip \\normalsize\\noindent \\abstracttext ",
        "introduction": "Sterile neutrinos from extra dimensions are not likely to play a primary role in atmospheric and solar neutrinos~\\cite{sno}. On the contrary, astrophysics, cosmology and rare decays still represent natural stages for their exotic performances. In this paper, we focus on the implications of a possible mixing of the electron neutrino with a Kaluza-Klein (KK) tower of sterile neutrinos in supernova (SN) physics.  The potential relevance of bulk fermions for neutrino physics has first been pointed out and analyzed in~\\cite{DDG,DS}. Fermions propagating in large extra dimensions appear as light, dense KK towers of sterile neutrinos in 4 dimensions, possibly mixing with the Standard Model ones. In turn, the existence of extra dimensions much larger than the inverse electroweak scale is natural in string inspired models with a fundamental scale in the TeV range~\\cite{TeV,TeV3}, if the standard relation between that scale and the Planck scale is implemented.  The basic ideas about supernova explosion and cooling, nicely confirmed by the neutrino signal from \\SN~\\cite{signal}, translate into significant constraints on the energy loss through invisible channels~\\cite{ann}. In fact, in order for the observed neutrino signal to be accounted for, the invisible energy loss should not cool the protoneutron star on a time scale shorter than about 10 seconds. The bulk of a large extra dimension provides at least two interesting examples of invisible channel: KK graviton emission~\\cite{graviton} and mixing of Standard Model and bulk neutrinos, the possibility we are interested in. With respect to the case of a single sterile neutrino~\\cite{sterile}, conversion into KK towers of sterile neutrinos is enhanced by the higher number of available states and especially by matter effects~\\cite{BCS}. The constraints that follow on the parameter space of bulk neutrino models look at first sight quite severe~\\cite{BCS}. However, those constraints are relaxed by an interesting feedback mechanism\\footnote{A different type of feedback was considered in~\\cite{Valle}.} that prevents unacceptable energy loss~\\cite{LRRR2} and affects the protoneutron star deleptonization and cooling through a non trivial interplay with diffusion. By suppressing the neutrino MSW potential and then making it negative everywhere, the mechanism leads to a self-limiting of the potentially dangerous, MSW-enhanced, energy loss into the bulk and requires a revisitation of the bounds in the literature.  The consequences for SN physics, in particular explosion, deleptonization, cooling and spectra, are not less interesting. From this point of view what we discuss here is an example of how the standard ideas about the protoneutron star evolution can be affected in presence of new physics in an unexpected and intriguing way.  The paper is structured as follows. In Section~\\ref{sec:framework} we shortly review the salient points about (electron) neutrino oscillations into bulk neutrinos. Section~\\ref{sec:main} is the central part of the paper. Since a reliable study requires taking into account diffusion, we first set up in Section~\\ref{sec:assumptions} a toy model of the protoneutron star core which, although minimal, contains all the neutrino transport physics relevant to our discussion. In Section~\\ref{sec:loss} we incorporate in the model the effect of electron neutrino and antineutrino conversion into bulk neutrinos. The effect on the evolution of the protoneutron star core is described in Section~\\ref{sec:evolution}. The consistency with \\SN\\ is shown in Section~\\ref{sec:implications}, where the implications for the neutrino luminosity, the flux spectrum and the shock reheating are also discussed.  Section~\\ref{sec:subleading} summarizes the constraints on subleading non self-limiting conversion probabilities, that provide one of the relevant bounds on our parameter space. We conclude in Section~\\ref{sec:summary}. ",
        "conclusions": "\\label{sec:summary}  The supernova plays an important role in the phenomenology of a possible mixing between SM neutrinos and fermions propagating in the bulk of large extra dimensions. We studied both the constraints on such a mixing following from the necessity of avoiding unacceptable energy loss and the implications for supernova physics. We considered the simple case of mixing between the electron neutrino and a single KK tower of 5D bulk neutrinos. The parameter space where interesting effects arise is quite broad, with the size of the extra dimension in the range $10^{-2}\\eV\\lesssim 1/R\\lesssim 1\\keV$ and the mixing mass parameter $m$ such that $10^{-12}\\eV\\lesssim \\mmR\\lesssim 10^{-8}\\eV$ and $mR\\lesssim 10^{-5}$. The interplay of neutrino diffusion and conversion into bulk neutrinos has an important role in the phenomenology we study and requires taking into account the relevant aspects of neutrino transport, which we did in the context of a simplified model of the protoneutron star core.  Due to the large number of available states and especially to matter effects, the rate at which neutrino disappearance in the bulk cools the protoneutron star is initially dangerously high. If taken at face value, such a rate would set quite a stringent bound on $\\mmR$, $\\mmR\\lesssim 10^{-12}\\eV$ in the case of neutrino conversion. However, the disappearance rate quickly reduces itself to acceptable values. This happens for different reasons. In the region where the MSW potential $V$ is positive, neutrinos escape in the bulk and the lepton fraction quickly drops to the value $\\Ya\\simeq 0.3$ at which $V = 0$, thus stopping the potentially most dangerous MSW enhanced conversion before a significant amount of energy is lost.  Then, on a much slower time scale, neutrino diffusion spoils the $V=0$ condition and unlocks the energy frozen in what was initially the $V>0$ region. The potential becomes slightly negative and antineutrinos start escaping in the bulk to restore the $V=0$ condition. This feedback keeps $\\Y$ close to $\\Ya$ until the temperature becomes too low to sustain the necessary antineutrino escape rate. In the region where the potential was initially negative, neutrino conversion never takes place. The initial loss rate is smaller because of the much lower antineutrino density. Still, it would be too high for large values of $\\mmR$. However, in the inner part of the $V<0$ region the antineutrino loss stops itself by drawing $V$ up to zero before all the local energy is lost, analogously to the $V>0$ case. In the outer part of the $V<0$ region, there is not enough energy to reach $V=0$. However, the conversion again controls itself. This happens this time because the energy loss, proportional to $T^{7/2}$, becomes less and less effective as the region cools.  The described mechanism allows to gain four orders of magnitude in the allowed range for $\\mmR$ ($10^{-12}\\eV\\rightarrow 10^{-8}\\eV$). In the subsequent four orders of magnitudes, the MSW enhanced conversion would still be under control but what were previously neglected as subleading contributions to the oscillation probability become too large. Such contributions, as well as the limits on other ``conventional'' invisible cooling channels, were discussed in Section~\\ref{sec:subleading}. In the gained portion of the parameter space, the implications for supernova physics are particularly interesting. Although compatible with the \\SN\\ signal, the evolution of the protoneutron star is significantly affected, especially for large values of $\\mmR$. Deleptonization and cooling take place on the same time scale as in absence of new physics.  However, the reluctance of the lepton fraction to get away from $\\Ya\\sim 0.3$ slows the deleptonization, while the energy loss accelerates the cooling. Up to a fourth of the energy lost by the protoneutron star goes to the bulk but the total luminosity in all neutrino species turns out to be approximately the same as in the standard case, at least in the first 20 seconds. Antineutrinos escape in the bulk in the effort of keeping $\\Y$ close to $\\Ya$, which boosts the lepton number radiated from the neutrinosphere. A quantitative study of the consequences of such an enhancement would require solving the evolution of the mantle. However, we expect it to result in an enhancement of the $\\nu_e$ component of the neutrino flux and therefore of the $\\nu_e$ luminosity, an effect observable in the existing neutrino detectors. Since we expect the temperature of the neutrinosphere not to be significantly affected, the energy deposition in a possibly stalling shock would also be enhanced at the expenses of a smaller $\\bar\\nu_e$ luminosity.  The exotic phenomenology discussed in this paper represents an example of how the standard ideas about protoneutron star evolution could be affected in an unexpected way by the presence of new physics."
    },
    "0209/astro-ph0209410_arXiv.txt": {
        "abstract": "{We investigate the properties of a galaxy similar to the Milky Way within the context of standard disk formation theory in a \\LCDM\\ universe. Using the standard assumption that baryons conserve specific angular momentum when they collapse, we conclude that the \\emph{mean} properties of the model galaxies are in good agreement with the Milky Way and other similar spiral galaxies, but the predicted scatter in disk scale lengths may be too large. A model in which half of the initial specific angular momentum is transfered to the dark matter may produce a smaller scatter, if very compact disks are unstable and evolve into spheroids or early type galaxies.}  \\resumen{} ",
        "introduction": "Cold Dark Matter (CDM) seems to provide a very successful paradigm for explaining many different kinds of observations on large scales, but suffers from several problems on small scales. Perhaps the most worrisome of these is the ``cusp'' problem: it seems that the radial profiles of dark matter halos produced in cosmological simulations based on CDM are inconsistent with the observed rotation curves of at least some dwarf and low surface brightness galaxies (e.g. van den Bosch \\& Swaters 2001 and references therein). It is important to establish whether or not this problem is peculiar to this particular class of galaxies. It is more difficult to assess whether the rotation curves of luminous, high surface brightness galaxies are consistent with CDM dark matter halos, because in these galaxies, baryons contribute significantly to the gravitational force in the central part of the galaxy, where rotation curves are observed. It is therefore expected that the inner dark matter profile is significantly modified by the collapse of the baryons. It is possible to calculate the effect of this baryon-induced ``contraction'' on the dark matter halo using a well-established analytic formalism. However, the large degeneracies due to the many unknown parameters make it difficult to obtain strong constraints from the observed rotation curves of most luminous galaxies.  The Milky Way galaxy offers a special opportunity to investigate this question. We know about the dynamical properties of our Galaxy in much greater detail and over a larger range of scales than any other galaxy. For example, the mass profile of our Galaxy as a function of radius is constrained by velocity measurements from scales of a few pc (from stellar velocities) to 100 kpc (from satellite galaxies). Also, observations of microlensing events towards the Galactic bulge place strong lower limits on the mass of \\emph{baryonic} material within about 3 kpc. In Klypin, Zhao \\& Somerville (2002; KZS02; see also the contribution by A. Klypin in this volume), we showed that these combined data place very strong constraints on the parameters of the Milky Way Galaxy and its dark matter halo. KZS02 concluded that, within the framework of the popular \\LCDM\\ cosmological model: 1) the Milky Way must occupy a halo with a total mass in the range 1--$2 \\times 10^{12}$ \\msun\\ 2) half of the baryons within the virial radius of this halo must have been ejected 3) standard disk formation models, in which the gas conserves its specific angular momentum during collapse, have difficulty obeying the combined microlensing and dynamical constraints. However, if angular momentum is \\emph{transfered} from the baryons to the dark matter, the dark matter gains angular momentum and so moves outward. The inner dark matter ``cusp'' is flattened out, leaving more room for baryons in the inner part of the Galaxy. KZS02 concluded that a model in which about half of the initial specific angular momentum was lost by the baryons could accommodate all of the data.  This brings up several further questions. How typical is our Galaxy? Does it lie near the mean of the distribution of objects predicted by the theory, or is it an outlier? Is it reasonable that such a large fraction of the baryons could be ejected from a relatively massive halo?  Are the photometric properties of a ``Milky Way'' produced in this framework consistent with observations? Is the distribution of stellar ages and metallicities in such a model consistent with observations in the Galaxy? These questions will be addressed in a companion paper to KZS02 (Somerville, Klypin \\& Zhao, in prep). Here, we will briefly address a few of these questions.  \\begin{figure} \\begin{center} \\end{center} \\includegraphics[width=\\columnwidth]{fig1.ps} \\caption{\\footnotesize Fundamental plane relations predicted by semi-analytic models, for an ensemble of ``Milky Way'' mass galaxies (total mass of baryons plus dark matter $10^{12} {\\rm M}_{\\odot}$), assuming that the specific angular momentum of the baryons is conserved. Large star symbols show the observed relations for the Ursa Major sample of normal spirals from Verheijen (2001). Shaded areas indicate the acceptable range of values for the Milky Way.  Dots show the model galaxies, with filled symbols indicating galaxies that should be globally stable, and open symbols showing galaxies that may be unstable to bar/bulge formation. \\label{fig:notrans}} \\end{figure}  \\begin{figure} \\begin{center} \\end{center} \\includegraphics[width=\\columnwidth]{fig2.ps} \\caption{\\footnotesize The same as Fig.~\\protect\\ref{fig:notrans}, except that angular momentum transfer is included using the formalism developed by KZS02. Note that while the agreement with the observational locus is better, a large number of unstable disks are predicted (shown by open symbols). \\label{fig:trans}} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209283_arXiv.txt": {
        "abstract": "We observed the entire course of the 1998 outburst of V592 Her, which was originally reported as a nova in 1968.  We have been able to construct a full light curve of the outburst, which is characterized by a rapid initial decline (0.98 mag d$^{-1}$), which smoothly developed into a plateau phase with a slower linear decline.  We detected superhumps characteristic to SU UMa-type dwarf novae $\\sim$7 d after the optical maximum.  The overall behavior of the light curve and the development of superhumps were characteristic to a WZ Sge-type dwarf nova.  Combined with the past literature, we have been able to uniquely determine the superhump period to be 0.05648(2) d.  From this period, together with a modern interpretation of the absolute magnitude of the outburst light curve, we conclude that the overall picture of V592 Her is not inconsistent with a lower main-sequence secondary star in contrast to a previous claim that V592 Her contains a brown dwarf. ",
        "introduction": "WZ Sge-type dwarf novae are a still enigmatic class of SU UMa-type dwarf novae [for recent summaries of dwarf novae and SU UMa-type dwarf novae, see \\citet{osa96review} and \\citet{war95suuma}, respectively], which is characterized by a long ($\\sim$ 10 yr) outburst recurrence time and a large ($\\sim$ 8 mag) outburst amplitude (cf. \\cite{bai79wzsge}; \\cite{dow81wzsge}; \\cite{pat81wzsge}; \\cite{odo91wzsge}; \\cite{kat01hvvir}).  In recent years, the secondary stars (mass-donor stars) of WZ Sge-type dwarf novae, or dwarf novae with extremely large outbursts amplitudes (TOADs, \\cite{how95TOAD}), have been regarded as promising candidates for brown dwarfs (\\cite{how97periodminimum}; \\cite{pol98TOAD}; \\cite{cia98CVIR}; \\cite{pat01SH}; \\cite{how01llandeferi}). The existence of a brown-dwarf secondary star has been also considered to play an important role in realizing an extremely low quiescent viscosity of WZ Sge-type stars required (\\cite{sma93wzsge}; \\cite{osa95wzsge}) from the disk-instability theory (\\cite{mey98wzsge}; \\cite{min98wzsge})\\footnote{ Arguments, however, exist against the extremely low quiescent viscosity.  \\citet{las95wzsge}, \\citet{war96wzsge} assuming evaporation/truncation of the inner disk are the best examples. \\citet{ham97wzsgemodel} and \\citet{bua02suumamodel} presented slight modifications of these ideas. The discovery of a WZ Sge-type phenomenon in a long-period system \\citep{ish01rzleo} suggests that the existence of a brown-dwarf secondary is not a necessary condition for manifestation of the WZ Sge-type phenomenon. } Observational confirmation of cataclysmic variables (CVs) with brown dwarf secondaries is also important in that it can provide an independent estimate of the upper limit of the age of the Universe (\\cite{pol98TOAD}; \\cite{szk02egcnchvvirHST}). In particular, \\cite{how01llandeferi} claimed the direct spectroscopic detection of a brown dwarf in LL And, but inconsistency in this interpretation has been later found \\citep{how02llandefpegHST}. \\citet{men01j1050} reported a discovery of a CV with a brown dwarf based on radial velocity studies. V592 Her is another object in which the existence of a brown dwarf has been claimed \\citep{vantee99v592her}.  V592 Her was discovered as a possible fast nova in 1968 on Sonneberg plates \\citep{ric68v592her}.  The observed maximum was $m_{\\rm pg}$ = 12.3. Although there was a gap in the 1968 observation, the outburst lasted at least for 30 d.  \\citet{ric68v592her} reported that the object was exceptionally blue based on a comparison between quasi-simultaneously taken blue- and yellow-sensitive plates.  Based on this conspicuously blue color at maximum, \\citet{due87novaatlas} suspected that the object may be either a dwarf nova or an X-ray nova resembling V616 Mon.  \\citet{ric91v592her} further studied Sonneberg plates, and discovered a second outburst in 1986.  The recorded maximum of the 1986 outburst was $m_{\\rm pg}$ = 13.6 (1986 May 12).  The limited coverage of this 1986 outburst made it difficult to draw a conclusion on the nature of this outburst.  The star has been intensively monitored by visual observers, members of the Variable Star Observers League in Japan (VSOLJ) since 1986 February.  The absence of visual detection of the 1986 May outburst suggests that the outburst was fainter than the 1968 one, or the brightness peak lasted for a very short time.  In spite of the intensive world-wide efforts, no further outburst had been detected until 1998.  On 1998 August 26.835 UT, Timo Kinnunen detected the object in outburst at $m_{\\rm v}$ = 12.0 (vsnet-alert 2067).\\footnote{ $\\langle$http://www.kusastro.kyoto-u.ac.jp/vsnet/Mail/\\\\alert2000/msg00067.html$\\rangle$.} He also noted a 0.5 mag variation within 0.08 d. The last negative observation (fainter than 13.2) was made by Patrick Schmeer on August 25.899 UT.  The object was reported to fade by 0.5--1.0 mag within 1 d of this detection.  Judging from the rapid fading and the presence of an immediately preceding negative observation, this outburst must have been caught around the peak brightness.  The dwarf nova-type nature was subsequently confirmed with spectroscopy (Mennickent et al., vsnet-alert 2087.\\footnote{ $\\langle$http://www.kusastro.kyoto-u.ac.jp/vsnet/Mail/\\\\alert2000/msg00087.html$\\rangle$.}; \\cite{men02v592her}). The large outburst amplitude ($>$ 9 mag, \\cite{due87novaatlas}; \\cite{ric91v592her}) and long recurrence times (10--20 yr) clearly qualifies V592 Her as a best candidate for a WZ Sge-type dwarf nova. \\citet{due98v592her} reported detection of superhumps, confirming that V592 Her belongs to SU UMa-type dwarf novae.  \\citet{vantee99v592her} argued, based on their quiescent photometry, that the secondary star of V592 Her can be a brown dwarf. We present a summary of our observations of the entire aspect of the 1998 outburst conducted as a part of VSNET Collaboration.\\footnote{ $\\langle$http://www.kusastro.kyoto-u.ac.jp/vsnet/$\\rangle$.} ",
        "conclusions": "\\subsection{Overall Outburst Light Curve}\\label{sec:lc}  Figure \\ref{fig:lc} shows the light curve of the 1988 superoutburst of V592 Her drawn from visual observations reported to VSNET.  Large and small dots represent positive and negative (upper limit) observations, respectively.  Open circles with error bars represent nightly averaged Kyoto CCD observations (table \\ref{tab:log}).  The overall light curve is characterized by the presence of a sharp maximum ($t$ = 0, JD 2451052 $\\pm$1 d) followed by a rapid decay.  The decay became slower as the object faded, and smoothly evolved into a gradually fading stage (plateau phase).  Between $t$ = 21 and $t$ = 25, the object experienced a sudden drop by 3.4$\\pm$0.5 mag.  The early development of the light curve more resembles those of very fast novae rather than those of usual SU UMa-type dwarf novae, which are characterized by the presence of a linear (exponential in flux scale) fading at a rate of 0.03--0.16 mag d$^{-1}$ (cf. \\cite{war85suuma}; see also \\cite{kat02v359cen} for a summary of recent well-documented examples). The sudden fading between $t$ = 21 and $t$ = 25 is characteristic of the termination of a superoutburst in an SU UMa-type star.  The later part of the outburst ($7 \\leq t \\leq $25) is more characteristic of a usual SU UMa-type superoutburst while the initial part is more unusual.  Similar departures of early light curves from the ``canonical\" light curve of SU UMa-type superoutbursts is rather commonly seen in WZ Sge-type outbursts.  In WZ Sge itself (e.g. \\cite{ort80wzsge}; \\cite{pat81wzsge}), there seems to have been such a sharp initial peak.\\footnote{ There exists an argument against the sharp, initial peak recorded in the past observations, because these feature were recorded on blue-sensitive photographs, which had a different sensitivity from visual observations.  See also \\citet{kat01hvvir}. }  Among well-observed WZ Sge-type outbursts, the present V592 Her most clearly showed this feature.  \\begin{figure*} \\begin{center} \\FigureFile(140mm,100mm){fig1.eps} \\end{center} \\caption{Light curve of the 1988 superoutburst of V592 Her drawn from visual observations reported to VSNET. Large and small dots represent positive and negative (upper limit) observations, respectively.  Open circles with error bars represent nightly averaged CCD observations listed in table \\ref{tab:log}. The epoch of maximum ($t$ = 0, see text) corresponds to JD 2451052. } \\label{fig:lc} \\end{figure*}  Such a deviation from a linear (exponential in flux scale) decay in a WZ Sge-type outburst is shown to be naturally understood as a consequence of a rapid viscous depletion of the large amount of stored gas during the initial stage of a WZ Sge-type outburst (\\cite{osa95wzsge}; see also \\cite{can93DI} for a basic model). \\citet{can01wzsge} recently successfully modeled the light curve of the 2001 superoutburst of WZ Sge with this mechanism.  \\citet{can02ugem1985} showed that this mechanism also worked in a system with a long orbital period. The case of V592 Her is more striking than in the 2001 superoutburst of WZ Sge.  A linear fit to the first 1 d of the light curve has yielded a mean decay rate of 0.98 mag d$^{-1}$.  The decay rate decreased to 0.05 mag d$^{-1}$ during the late half of the plateau phase.  The initial decay rate was 4 times larger than that of the 2001 superoutburst of WZ Sge \\citep{can01wzsge}.\\footnote{ The first 15-d average of the decline rate in V592 Her is comparable to that in WZ Sge \\citep{can01wzsge}.  Since the effect of a viscous decay is stronger near the outburst peak, we use the initial decline rate described in the text. }  By applying the relation between the viscous decay time-scale ($\\tau_\\nu$) and the surface density in the disk ($\\Sigma$), $\\tau_\\nu \\propto \\Sigma^{-3/7}$ \\citep{can01wzsge} to the initial decay rate, the initial surface density is estimated to be $\\sim$25 times larger than that in the initial part (derived from an average of the first 15 d) of the 2001 superoutburst of WZ Sge.  WZ Sge-type dwarf novae are known to frequently (but not always) show post-outburst rebrightenings [for a review, see \\citet{kat98super}. See also \\citet{ric92wzsgedip}; \\citet{how95TOAD}; \\citet{kuu96TOAD}; \\citet{kuu00wzsgeSXT}; \\citet{kat97egcnc}; \\citet{pat98egcnc}].\\footnote{ These phenomena are sometimes referred to as {\\it echo outbursts}, but we avoid this terminology because this idea was first proposed to describe the ``glitches\" or ``reflares\" in soft X-ray transients (SXTs) \\citep{aug93SXTecho}.  In SXTs, hard-soft transition is considered to be more responsible for the initially claimed phenomenon \\citep{min96SXTtransition}, which is clearly physically different from dwarf nova-type rebrightenings. }  Due to the faintness of the object, the existence of such a post-superoutburst rebrightening was not unambiguously confirmed during the present outburst of V592 Her, although there may have been a hint of such phenomenon on October 3 (JD 2451090).  However, a long-lasting rebeightening as observed in AL Com (\\cite{kat96alcom}; \\cite{nog97alcom}; \\cite{pat96alcom}), WZ Sge in 2001 (\\cite{ish02wzsgeletter}; \\cite{pat02wzsge}), and V2176 Cyg \\citep{nov01v2176cyg} was not recorded. There was no hint of a bright rebrightening as expected by \\citet{bua02suumamodel}.  \\subsection{Superhump Period}  \\citet{due98v592her} reported the detection of superhumps (period either 0.06007 d or 0.06391 d) based on their three-night observation. A closer look at the data by \\citet{due98v592her} left some uncertainty regarding this period determination, mainly because only one superhump per night was observed, which makes unique alias selection virtually impossible.  In order to solve this problem, we have digitized the figure in \\citet{due98v592her} and measured their observations to an accuracy of 0.001 mag and 0.0001$-$0.0002 d.  Although a period analysis of these data has confirmed the claimed periods by \\citet{due98v592her}, there remains substantial possibility around $P$ = 0.0567 d (see also the upper panel of figure \\ref{fig:pdm}).  \\begin{figure} \\begin{center} \\FigureFile(88mm,120mm){fig2.eps} \\end{center} \\caption{Representative nightly superhump light curves during the superoutburst plateau.  Superhumps with amplitudes of 0.1--0.2 mag are present.  These observations covered an earlier epoch than in \\citet{due98v592her}. } \\label{fig:nightly} \\end{figure}  We have further extracted the times of superhump maxima from our observations between 1998 September 2 and September 7 (early part of the superoutburst plateau, see figure \\ref{fig:nightly}). These observations covered an earlier epoch than in \\citet{due98v592her}.  The times of maxima were determined by fitting the average superhump light curve (given in figure \\ref{fig:shph}) to the observed data.  The maximum times and 1-$\\sigma$ errors of timing estimates were determined with Marquardt-Levenberg method \\citep{Marquardt}. The validity of the fits has also been confirmed with a comparison of independent eye extraction of maximum times. Table \\ref{tab:sh} lists the measured timings of the superhump maxima. The values are given to 0.0001 d in order to avoid the loss of significant digits in a later analysis.  These maxima are not well expressed by either of the two candidate periods listed in \\citet{due98v592her}. In particular, the interval of 0.396 d between the BJD 2451062.911 and 2451063.307 is only well expressed by a period near 0.057 d within their respective errors (this interval corresponds to 6.59 and 6.20 cycles of the two candidate periods \\citep{due98v592her} of 0.06007 and 0.06391 d, respectively).  We thus conclude that the short alias ($P \\sim$ 0.0567 d) is the true superhump period.  The cycle counts ($E$) in table \\ref{tab:sh} are calculated with this period.  A linear regression to the observed superhump times gives the following ephemeris (the errors correspond to 1-$\\sigma$ errors at the epoch of $E$ = 67) :  \\begin{equation} {\\rm BJD (max)} = 2451058.9005(10) + 0.056498(13) E. \\label{equ:reg1} \\end{equation}  \\begin{table} \\caption{Timings of superhumps.}\\label{tab:sh} \\begin{center} \\begin{tabular}{lrrrc} \\hline\\hline BJD$^*$ & Error$^\\dagger$ & $E$$^\\ddagger$ & $O-C_1$$^\\dagger$$^\\S$ & Ref.$^\\|$ \\\\ \\hline 58.9003 & 10 &   0 &  $-$2 & 1 \\\\ 58.9576 & 15 &   1 &     6 & 1 \\\\ 59.8634 & 23 &  17 &    24 & 1 \\\\ 59.9180 & 14 &  18 &     5 & 1 \\\\ 62.9107 & 23 &  71 & $-$12 & 1 \\\\ 63.3070 & 26 &  78 &  $-$3 & 1 \\\\ 63.8718 & 25 &  88 &  $-$5 & 1 \\\\ 63.9257 & 20 &  89 & $-$31 & 1 \\\\ 64.4910 &  7 &  99 & $-$28 & 2 \\\\ 65.5123 &  6 & 117 &    16 & 2 \\\\ 67.4913 &  9 & 152 &    31 & 2 \\\\ \\hline \\multicolumn{5}{l}{$^*$ BJD$-$2451000.} \\\\ \\multicolumn{5}{l}{$^\\dagger$ Unit 0.0001 d.} \\\\ \\multicolumn{5}{l}{$^\\ddagger$ Cycle count.} \\\\ \\multicolumn{5}{l}{$^\\S$ Against equation (\\ref{equ:reg1}).} \\\\ \\multicolumn{5}{l}{$^\\|$ 1: this work, 2: measured from} \\\\ \\multicolumn{5}{l}{\\phantom{$^\\|$} \\citet{due98v592her}} \\\\ \\end{tabular} \\end{center} \\end{table}  Figure \\ref{fig:pdm} shows the result of period analysis of superhumps with the Phase Dispersion Minimization (PDM, \\cite{PDM}).  The upper panel shows an analysis of the data in \\citet{due98v592her}, which shows the possibility of many one-day aliases.  The lower panel shows an analysis of the combined data (this work and \\cite{due98v592her}), which covered the superoutburst plateau between JD 2451057 (September 1) and 2451067 (September 11).  A strong preference of the frequency of 17.716(8) $d^{-1}$, which corresponds to a period of $P$ = 0.05645(2) d, is clearly seen. An exclusion of the data of \\citet{due98v592her} did not significantly change this trend.  The selection of the true alias is confirmed by these analyses. The significance level of this period is above 95 \\%.  Figure \\ref{fig:clean} shows period analysis of superhumps in V592 Her with the Clean method \\citep{CLEAN}, with a gain parameter of 0.01.  The data and the frequency range are the same as in the lower panel of figure \\ref{fig:pdm}. The Cleaned spectrum clearly shows that the frequency of 17.72 d$^{-1}$ is the only acceptable period.  We finally adopted $P_{\\rm SH}$ = 0.05648(2) from an average of superhump timing analysis and PDM analysis.  \\begin{figure} \\begin{center} \\FigureFile(88mm,120mm){fig3.eps} \\end{center} \\caption{Period analysis of superhumps in V592 Her with the Phase Dispersion Minimization (PDM).  (Upper) Analysis of the data in \\citet{due98v592her}.  (Lower) Analysis of the data between JD 2451057 (September 1) and 2451067 (September 11), which covered the superoutburst plateau.} \\label{fig:pdm} \\end{figure}  \\begin{figure} \\begin{center} \\end{center} \\FigureFile(88mm,120mm){fig4.eps} \\caption{Period analysis of superhumps in V592 Her with the Clean method \\citep{CLEAN}.  The data and the frequency range are the same as in the lower panel of figure \\ref{fig:pdm}. (Upper) Cleaned spectrum (power in arbitrary unit). The frequency of 17.72 d$^{-1}$ is the only acceptable period. (Lower) Window function. } \\label{fig:clean} \\end{figure}  Figure \\ref{fig:shph} shows a mean superhump profile phase-folded with a period of $P_{\\rm SH}$ = 0.05648 d.  The rapid rising and slowly declining profile is characteristic to SU UMa-type superhumps (\\cite{vog80suumastars}, \\cite{war85suuma}).  The mean amplitude (0.15 mag) of superhumps is smaller than those of usual SU UMa-type dwarf novae, but is close to that of a WZ Sge-type star, HV Vir \\citep{kat01hvvir}.  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig5.eps} \\end{center} \\caption{Mean superhump profile of V592 Her phase-folded with a period of $P_{\\rm SH}$ = 0.05648 d.} \\label{fig:shph} \\end{figure}  The newly established superhump period ($P_{\\rm SH}$ = 0.05648(2) d) is extremely close to those of WZ Sge ($P_{\\rm SH}$ = 0.05726(1) d: \\cite{ish02wzsgeletter}, \\cite{pat02wzsge}), AL Com ($P_{\\rm SH}$ = 0.05722(1) d: \\cite{kat96alcom}, \\cite{pat96alcom}), the two best-studied WZ Sge-type dwarf novae.  All known WZ Sge-type dwarf novae have $P_{\\rm SH}$ shorter than 0.060 d except RZ Leo and EG Cnc (see e.g. \\cite{kat01hvvir}).  Among them, the long period of RZ Leo is compatible with the evidence of a relatively massive secondary \\citep{ish01rzleo}.  Since the secondary of V592 Her is apparently less luminous \\citep{vantee99v592her} than in RZ Leo, the new period better fits the general WZ Sge-type characteristics without necessarily introducing, as we will see, a possibility of a brown dwarf secondary.  \\subsection{Superhump Period Change}  In contrast to the ``textbook\" decrease of the superhump periods in usual SU UMa-type dwarf novae (e.g. \\cite{war85suuma}; \\cite{pat93vyaqr}), WZ Sge-type dwarf novae are recently known to show virtually zero or even increase of the superhump periods (for a summary, see \\cite{kat01hvvir}). The quadratic term determined from the superhump maximum timings corresponds to $\\dot{P}$ = +1.2 $\\pm$ 0.4 $\\times$ 10$^{-6}$ cycle$^{-1}$ or $\\dot{P}/P$ = +2.1(0.8) $\\times$ 10$^{-5}$.  This value indicates a small, but significant, period increase in V592 Her (figure \\ref{fig:oc}). This rate is comparable to the period changes observed in WZ Sge (\\cite{ish02wzsgeletter}; \\cite{pat02wzsge}).  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig6.eps} \\end{center} \\caption{$O-C$ diagram of superhump maxima. The parabolic fit corresponds to $\\dot{P}$ = 1.2 $\\pm \\times$ 10$^{-6}$ cycle$^{-1}$.} \\label{fig:oc} \\end{figure}  \\subsection{Early Superhumps and Orbital Period}\\label{sec:ESH}  All well-observed WZ Sge-type dwarf novae are known to show double-humped modulations having a period very close to the system orbital period during the earliest stage of superoutbursts (\\cite{kat96alcom}; \\cite{mat98egcnc}; \\cite{pat96alcom}; \\cite{nog97alcom}; \\cite{ish01rzleo}; \\cite{kat01hvvir}; \\cite{ish02wzsgeletter}; \\cite{pat02wzsge}). These modulations are called early superhumps.\\footnote{ This feature is also referred to as {\\it orbital superhumps} \\citep{kat96alcom}, {\\it outburst orbital hump} \\citep{pat98egcnc} or {\\it early humps} \\citep{osa02wzsgehump}. }  The presence of early superhumps is the unique characteristic of WZ Sge-type dwarf novae \\citep{kat01wxcet}. Although the origin of early superhumps is controversial, several interpretations have been historically proposed: (1) enhanced hot spot caused by a sudden increase of the mass-transfer (\\cite{pat81wzsge}; \\cite{pat02wzsge}), (2) immature form of superhumps \\citep{kat96alcom}, (3) geometrical effect of a jet or a thickened edge of the accretion disk \\citep{nog97alcom}.  Most recently, \\citet{osa02wzsgehump} proposed that these humps are a manifestation of a tidal 2:1 resonance in the accretion disks of binary systems with extremely low mass ratios.  During the 2001 superoutburst of WZ Sge \\citep{ish02wzsgeletter}, a two-armed spiral velocity pattern in the Doppler tomograms of the He\\textsc{II} line was found (\\cite{ste01wzsgeiauc7675}; \\cite{bab02wzsgeletter}) at the same time of the appearance of early superhumps.  \\citet{kat02wzsgeESH} suggested that both early superhumps and the two-armed spiral velocity pattern can be naturally considered by taking into the effect of a velocity field of a tidally distorted disk (\\cite{sma01tidal}; \\cite{ogi02tidal}).  In the present case of V592 Her, the apparent presence of early epoch short-term variation (up to 0.5 mag) as inferred from visual observations seems to suggest the presence of early superhumps as in the 2001 outburst of WZ Sge.  However, the lack of time-resolved photometry during the earliest stage of the outburst makes it difficult to draw a firm conclusion.  By using the best-established fractional superhump excesses ($\\epsilon=P_{\\rm SH}/P_{\\rm orb}-1$) of 1.0$\\pm$0.1 \\% in WZ Sge (\\cite{ish02wzsgeletter}; \\cite{pat02wzsge}) and AL Com (\\cite{kat96alcom}; \\cite{nog97alcom}; \\cite{pat96alcom}), the orbital period ($\\sim$ period of early superhumps) is expected to be 0.05592(3) d.  Most recently, \\citet{men02v592her} reported candidate orbital periods of 91.2 $\\pm$ 0.6 min ($P_1$: 0.0633(4) d), 85.5 $\\pm$ 0.4 min ($P_2$: 0.0594(3) d) or 80.8 $\\pm$ 0.6 min ($P_3$: 0.0561(4) d) from an analysis of their spectroscopy taken a fews days after the maximum of the 1998 outburst. Based on our identification of the true $P_{\\rm SH}$, $P_3$ is now confirmed to be the true $P_{\\rm orb}$.  This difference of preferable period selection between \\citet{men02v592her} and this work can be reasonably attributed to a severe aliasing clearly seen in the Fig. 3 of \\citet{men02v592her}.  Our period and $P_3$ in \\citet{men02v592her} are consistent within their respective errors.  Figure \\ref{fig:esh} shows the light curves on 1998 August 30 and 31 ($t$ = 4 d and 5 d, respectively) phase-averaged with a period of 0.05592 d, assuming that these variations reflect early superhumps.\\footnote{ Due to the shortness of each runs, any trial period between 0.05592 d (adopted $P_{\\rm orb}$) and 0.05648 d ($P_{\\rm SH}$) gives the virtually same waveform.  Strictly speaking, we cannot distinguish early superhumps from (the growing stage of) superhumps from these observation only.  However, we consider it likely that these modulations reflect early superhumps because the transition from early superhumps to superhumps has been to confirmed to occur less than 1 d in WZ Sge (\\cite{ish02wzsgeletter}; \\cite{pat02wzsge}).  A chance to observe the growing stage of superhumps on two nights is expected to be very small. } On August 30, there is a hint of low-amplitude double-wave modulation (with a rather strong signature of minimum), resembling early superhumps in HV Vir \\citep{kat01hvvir}.  On August 31, only small-amplitude variation seems to have been marginally detected.  The amplitude was less than 0.08 mag on August 31.  The weakness of the signal on August 31 has made it impossible to make a period determination from these observations.  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig7.eps} \\end{center} \\caption{Light curves on 1998 August 30 and 31 phase-averaged with the expected ($P$ = 0.05592 d) period of early superhumps.  The phase zero was arbitrarily taken as BJD 2451000.  On August 30, there is a hint of low-amplitude double-wave modulation, resembling early superhumps in HV Vir \\citep{kat01hvvir}.  On August 31, only small-amplitude variation seems to have been marginally detected.} \\label{fig:esh} \\end{figure}  Since both \\citet{osa02wzsgehump} and \\citet{kat02wzsgeESH} imply that the amplitude of early superhumps is a strong function of the binary inclination, the low amplitude of the possible early superhumps suggests a low binary inclination.  This suggestion is consistent with the lack of a strong He\\textsc{II} emission line \\citep{men02v592her}, which was strongly seen in emission in the high-inclination system WZ Sge (\\cite{ste01wzsgeiauc7675}; \\citet{bab02wzsgeletter}), and the lack of large-amplitude variability in quiescence \\citep{vantee99v592her}.  \\subsection{Brown Dwarf Secondary?}  V592 Her is proposed to have a brown dwarf secondary because a main sequence secondary would imply a large distance, which is inconsistent with absolute magnitudes of ordinary dwarf novae (\\cite{vantee99v592her}).  Here we reexamine the distance and the nature of the secondary based on the more accurate orbital period of V592 Her.  We can estimate a distance of a dwarf nova from the observed peak magnitude and the empirically expected absolute magnitude.  The correlation between the peak magnitude ($M_V({\\rm max})$) and the orbital period ($P_{\\rm orb}$) provides $M_V({\\rm max})\\sim$ 3.8--5.3 for a superoutburst (\\cite{war95book}; \\cite{vantee99v592her}).  When we apply this method to WZ Sge stars, however, the observed peak magnitude is not suitable to estimate a distance because they experience a rapid fading phase just after the peak, during which the viscosity decays with time (\\cite{can01wzsge}; \\cite{osa95wzsge}).  After this viscosity decay phase, the surface density of the accretion disk is expected to follow the same time-evolution as in superoutbursts of ordinary SU UMa-type dwarf novae \\citep{osa95wzsge}.  The peak magnitudes of these ordinary outbursts are limited by the critical surface density ($\\Sigma_{\\rm max}$) required by the disk instability theory \\citep{can98DNabsmag}.  The calculated peak magnitudes are known to well reproduce the Warner's relation. \\citet{can98DNabsmag} originally restricted the discussion to SS Cyg-type dwarf novae, but the same discussion can be naturally extended to the upper limits of $M_V({\\rm max})$ of normal outbursts of SU UMa-type dwarf novae. Superoutbursts are generally $\\sim$0.5 mag brighter than upper-limit magnitudes of normal outbursts \\citep{war95suuma}, caused by an extra heating by tidal dissipation, a safe upper limit of $M_V({\\rm max})$ for ordinary SU UMa-type superoutbursts is estimated to be 0.5 mag brighter than the extrapolation of \\citet{can98DNabsmag}. In the case of V592 Her, this value corresponds to $M_V({\\rm max}) \\sim +4.8$, which is consistent with the reported $M_V({\\rm max})\\sim$ 3.8--5.3 for observed superoutbursts (\\cite{war95book}).  We should hence compare the expected $M_V({\\rm max})$ not with the observed peak magnitude, but with the magnitude at which the viscous decay finishes, in other words, an ordinary plateau phase begins.  As can be seen in figure \\ref{fig:lc} and a comparison with a simulation in \\citep{osa95wzsge}, V592 Her experienced this phase change at $V=14.0$ (considering that CCD observations tended to give slightly fainter magnitudes than visual observations, and considering the difficulty in accurately estimating such a faint magnitude visually, this magnitude would better be regarded as an upper limit of the plateau phase).  While \\citet{vantee99v592her} estimate a distance $d\\sim$ 220--440 pc using a peak magnitude of $V=12$, our estimation hence provides larger distances of $d\\geq$ 550--1100 pc.  Another caveat in \\citet{vantee99v592her} is that they used a wrong (longer) superhump period based on \\citet{due98v592her}.  By adopting the correct $P_{\\rm SH}$ = 0.05648(2) and estimated $P_{\\rm orb}$ = 0.05592(3) d, the expected absolute magnitude of a main-sequence secondary filling the Roche-lobe of this $P_{\\rm orb}$ is at least $\\sim$1.0 mag fainter \\citep{bar98lowmassstars}.  Based on the same method of estimate in \\citet{vantee99v592her}, the lower limit of the distance from a comparison of apparent magnitude and the absolute magnitude of a main-sequence secondary is now lowered to 900 pc or even lower.  This lower limit of the distance is now not at all inconsistent with an estimate from the outburst photometry.  The present new determinations of the true $P_{\\rm SH}$, $P_{\\rm orb}$ and the new distance estimate are thus consistent with a lower main-sequence secondary.  By considering a main-sequence secondary with $M_I$ = 12.4 (which corresponds to an upper limit of the luminosity of a main sequence filling the Roche-lobe of V592 Her), the observed color $R-I \\sim$ 0.2 can be naturally explained by a contribution of this secondary star.  In this case, we don't need to assume an extremely cold ($\\sim$10000 K) white dwarf as deduced in \\citet{vantee99v592her}. Although accurate determination of the white dwarf temperature should await optical-UV spectroscopy, this finding seems to be consistent with recent determinations of white dwarf temperatures in WZ Sge-type dwarf novae (EG Cnc: 11700--13000 K, HV Vir: 12500--14000 K \\cite{szk02egcnchvvirHST}; GW Lib: 14700 K in average \\cite{szk02gwlibHST}; LL And: 15000 K \\cite{how02llandefpegHST}).  The present conclusion is also comparable to recent result in WZ Sge itself \\citep{ste01wzsgesecondary}, who concluded that a lower main-sequence secondary is still viable in spite of all the past negative efforts in directly detecting a signature of emission from the secondary of WZ Sge. In conjunction with the present conclusion, the presence of a brown dwarf in WZ Sge-type dwarf novae is still an open question even in most promising cases.\\footnote{ \\citet{men02v592her} concluded, from their identification of $P_{\\rm orb}$, that $\\epsilon$ supports the brown-dwarf like nature of the secondary star.  However, as discussed in subsection \\ref{sec:ESH}, the correct $P_{\\rm orb}$ is their $P_3$ = 0.0561(4) d.  This value gives $\\epsilon$ = 0.7 $\\pm$ 0.7\\%, which essentially gives no constraints on the existence of a brown dwarf. Furthermore, velocity fields of emission lines in WZ Sge-type superoutbursts are known to be very complex \\citep{bab02wzsgeletter}, or even systematically vary \\citep{kat02wzsgeESH}.  Radial velocity variation of emission lines in WZ Sge-type superoutbursts thus may not reasonably trace the binary motion as {\\it a priori} assumed in \\citet{men02v592her}. }  \\subsection{Related Objects}  As stressed in \\citet{kat01hvvir}, light curves of some WZ Sge-type dwarf novae often display similar characteristics to those of very fast novae.  The present, first-ever fully obtained, light curve of V592 Her (figure \\ref{fig:lc}) marks an even stronger similarity.  In WZ Sge itself, either a lower surface density at the beginning of an outburst, or a self-shielding effect arising from a nearly edge-on view, may have reduced this effect.  In this context, the present light curve of V592 Her even ``better\" reproduce the expected light curve of a WZ Sge-type dwarf nova \\citep{osa95wzsge}.  As seen in subsection \\ref{sec:ESH}, this light curve may be a result of a low binary inclination in V592 Her. Among the stars listed in \\citet{kat01hvvir}, V358 Lyr \\citep{ric86v358lyr}, LS And \\citep{sha78lsand} and V4338 Sgr have very similar light curves to that of V592 Her.  These objects may comprise a group of WZ Sge-type dwarf novae which is either characterized by a stronger effect of initial viscous decay, or a low binary inclination.  None of these systems, including V592 Her \\citep{vantee99v592her}, have been detected in ROSAT surveys (\\cite{ver97ROSAT}; \\cite{ROSATRXP}).  Apparently low X-ray luminosities of these systems makes a striking difference from the relatively strong quiescent X-ray detection in WZ Sge \\citep{ver97ROSAT}.  This difference from WZ Sge may be a result of an even smaller quiescent viscosity, which could explain a stronger effect of initial viscous decay."
    },
    "0209/astro-ph0209230_arXiv.txt": {
        "abstract": "The distant globular cluster Palomar 13 has been found to have a very high mass-to-light ratio and its profile can be well fitted either by a King model with a tail, or with a NFW model [1]. This cluster may be the first case of the many clumps predicted by CDM simulations that would not be disrupted by the galactic halo potential. We make the hypothesis that Pal 13 is made of neutralinos and run the DarkSuspect code to estimate the high-energy photon flux due to the annihilation of neutralinos through various channels in some benchmark scenarios. These low fluxes may be used as targets to be reached in proposals for future ground-based high altitude Cerenkov telescopes. ",
        "introduction": "\\vspace*{-0.1cm} The distant globular cluster Palomar 13 has been found to have a very high M/L ratio of $40~M_\\odot/L_\\odot$ and its profile can be well fitted either by a King profile with a power-law tail or a NFW model [1] with scale radius $2.4 \\pm 0.2$pc and central density $80~ M_\\odot pc^{-3}$. A possible explanation is that this distant cluster $(D = 24.3~kpc)$ is one of the numerous dark clumps predicted by CDM scenarios, which was not destroyed by the galactic tidal field. It may be a disrupted cluster as well, out of dynamical equilibrium. Here we assume that the NFW profile is the signature of a halo made of cold particles. Physics beyond the standard model could be supersymmetry. The lowest massive supersymmetric particle, i.e. the neutralino, is a natural candidate for CDM. If R-parity is conserved the neutralino is stable, is its own antiparticle and has a very small cross-section for annihilation. We assume that the halo of Palomar 13 is made of neutralinos and calculate the flux in high energy $\\gamma$-rays due to their annihilation. \\vspace*{-0.1cm} ",
        "conclusions": ""
    },
    "0209/astro-ph0209006_arXiv.txt": {
        "abstract": "This report analyzes the $I$-band CCD photometry of Nova V1974 Cygni from the 1997 observational season. The analysis shows that both short-term modulations with periods 0.0813 and 0.085 days are still present in the light curve of the star. We confirmed the stability of the shorter period which is interpreted as the orbital period of the binary system. Its value, determined using the $O-C$ residuals, is $P_{orb}=0.08125873(23)$ days = $117.0126(3)$ min. The longer period, which appeared in the light curve in 1994, was decreasing until the beginning of 1995 but then started to increase quite rapidly. In October 1996 the value of the period was $122.67\\pm0.02$ min. Until the next observing run the period significantly decreased. Its value, determined from our observations performed in July 1997, was $121.87\\pm0.12$ min. This means that the rate of change of the period in 1996-1997 was as high as $\\dot P \\approx 10^{-6}$. Such a rapid change of the period requires a large amount of rotational kinetic energy, if we assume that a 122-min periodicity is the rotation period of a white dwarf. Thus the more probable explanation is the hypothesis is that the longer period including a superhump period is caused by the precession of an accretion disc surrounding a white dwarf primary.  \\noindent {\\bf Key words:} Stars: individual: V1974 Cyg -- binaries: close -- novae, cataclysmic  variables ",
        "introduction": "V1974 Cygni (Nova Cygni 1992) was one of the brightest galactic novae of the 20th century. At the end of February 1992 it reached its maximum brightness equal to $V=4.3$ mag.  During subsequent years Nova Cygni 1992 was intensively observed using several techniques. The X-ray flux of the object indicated high luminosity at the level of $10^{37}$ erg~s$^{-1}$ caused by hot material burning in thermonuclear reactions at the surface of a white dwarf (Krautter et al. 1996).  The earliest photometric observations obtained by DeYoung and Schmidt (1993, 1994) revealed the presence of a light curve modulation with a period of 0.0813 days (117 min) and an amplitude equal to 0.16 mag.  In 1994, Semeniuk et al. (1994) discovered another periodicity in the light curve of V1974 Cygni with a period equal to 0.085 days (122 min). Subsequent observations showed that this period is not stable and changes with $\\dot P$ at a level of ~$-10^{-6}$ (Semeniuk et al. 1995, Olech et al. 1996, Skillman et al. 1997).  The 117 min stable period was interpreted as the orbital period of the system (DeYoung and Schmidt 1994, Semeniuk et al. 1995, Olech et al. 1996). The nature of the 122 min period still remains unexplained. Until now, two major explanations have been considered: 1) it might be either a superhump period of a SU UMa permanent superhumper star (Leibowitz et al. 1995, Skillman et al. 1997, Retter et al. 1997); or 2) it might be either the rotation period of the magnetized white dwarf in the binary system (Semeniuk et al. 1995, Olech et al. 1996).  In this paper we present the results of CCD $I$-band photometry of V1974 Cygni performed during the end of the observational season 1996 and for the whole season of 1997. ",
        "conclusions": ""
    },
    "0209/gr-qc0209088_arXiv.txt": {
        "abstract": "The Einstein-Hilbert action (and thus  the dynamics of gravity) can be obtained by: (i)  combining the principle of equivalence, special relativity  and quantum theory in the Rindler frame and (ii) postulating that the horizon area must be proportional to the entropy. This approach  uses the local Rindler frame  as a natural extension of the local inertial frame, and leads to the interpretation that the gravitational action represents the free energy of the spacetime geometry. As an aside, one  obtains  an insight into the peculiar structure of Einstein-Hilbert action and a natural explanation  to the questions: (i) Why does the covariant action for gravity contain second derivatives of the metric tensor? (ii)  Why is the   gravitational coupling constant is positive ? Some geometrical features of gravitational action are clarified. ",
        "introduction": "The (i) existence of the principle of equivalence and (ii) the connection between gravity and thermodynamics are the two most surprising features of gravity. Among these two, the principle of equivalence  finds its natural expression when gravity is described as a manifestation of curved spacetime. This --- in turn --- makes gravity  the only interaction which is capable of wrapping up regions of spacetime so that information from one region is not accessible to observers at another region. Given the fact that entropy of a system is closely related to accessibility of information, it is inevitable that there will be some connection between gravity and thermodynamics (for a review, see references \\cite{birreltp} \\cite{tprealms}). But, in contrast to the principle of equivalence,  years of research in this field (see, for a sample of references \\cite{bhentropy}), has not led to  something more profound or fundamental arising out of this feature.  This suggests that we should  learn a lesson from the way Einstein handled the principle of equivalence and  apply it in the context of the connection between thermodynamics and gravity. Einstein did not attempt to ``derive\" principle of equivalence in the conventional sense of the word. Rather, he accepted it as a key feature which must find expression in the way gravity is described --- thereby obtaining  a geometrical description of gravity. Once the geometrical interpretation of gravity is accepted, it follows that there {\\it will} arise surfaces which act as one-way-membranes for information and  will thus lead to some connection with thermodynamics. It is, therefore, more in tune with the spirit of Einstein's analysis to {\\it accept} an inevitable connection between gravity and thermodynamics and ask what such a connection would imply. I will now elaborate this idea further in order to show how powerful it is \\cite{grf}.  The first step in the logic,  the principle of equivalence, allows one to define a coordinate system around any event ${\\cal P}$ in a region of size $L$ (with $L^2(\\partial^2 g/g)   \\ll 1$ but $L(\\partial g/g) $ being arbitrary) in  which the spacetime is locally inertial. As the second step, we want to give expression to the fact that there is a deep connection between one-way-membranes arising in a spacetime and thermodynamical entropy. This, of course, is not possible in the local inertial frame since the quantum field theory in that frame, say, does not recognize any non trivial geometry of spacetime. But it is possible to achieve our aim by using a uniformly accelerated frame around ${\\cal P}$. In fact, around  any event ${\\cal P}$ we have fiducial observers anchored firmly   in space with ${\\bf x} = $ constant and  the four-velocity $u^i = g_{00}^{-1/2}(1,0,0,0)$ and  acceleration $a^i = u^j \\nabla_j u^i$. This allows us to define a second natural coordinate system around any event by using the Fermi-Walker transported coordinates corresponding to these accelerated observers. I shall call this the local Rindler frame. [Operationally, this coordinate system is most easily constructed by first transforming to the locally inertial frame and then using the standard transformations between the inertial coordinates and the Rindler coordinates.] This local Rindler frame will lead to a natural notion of horizon and associated temperature. The key new idea  will be to postulate that the horizon in the local Rindler frame also has an entropy per unit transverse area and  demand that any description of gravity must have this feature incorporated in it.  What will such a postulate lead to? Incredibly enough, it leads to the {\\it correct Einstein-Hilbert action principle}   for gravity. Note that the original approach of Einstein making use of the principle of equivalence lead only up to the {\\it kinematics} of gravity --- viz., that gravity is described by a curved spacetime with a non trivial metric $g_{ab}$ --- and cannot tell us how the {\\it dynamics} of the spacetime is determined. Taking the next step, using the local Rindler frame and demanding that gravity must incorporate the thermodynamical aspects lead to the action functional itself.   This  approach also throws light on (what has been usually considered) a completely different issue: Why does the Einstein-Hilbert action contain second derivatives of the metric tensor? The new approach ``builds up'' the Einstein-Hilbert action from its surface behaviour and, in this sense, shows that gravity is intrinsically holographic \\cite{holo}. I use this term with the specific meaning that given the form of the action on a two dimensional surface, there is a way of obtaining the full bulk action. In the $(3+1)$ formalism, this leads to the interpretation of the gravitational action as the free energy of spacetime.  Einstein's equations are equivalent to the principle of minimization of free energy in thermodynamics. ",
        "conclusions": "The approach adopted here is a natural extension of the original philosophy of Einstein; viz., to use non inertial frames judiciously to understand the behaviour of gravity. In the original approach, Einstein used the principle of equivalence which leads naturally to the description of gravity  in terms of the metric tensor. Unfortunately, {\\it classical} principle of equivalence cannot take us any further since it does not encode information about the curvature of spacetime. However, the true world is quantum mechanical and one would like to pursue the analogy between non inertial frames and gravitational field into the quantum domain. Here the local Rindler frame arises as the natural extension of the local inertial frame and the study of the thermodynamics of the horizon shows a way of combining special relativity, quantum theory and physics in the non inertial frame. I have shown that these components are adequate to determine the action functional for gravity and, in fact, leads  to the Einstein-Hilbert action. This is remarkable because we did not introduce the curvature of spacetime explicitly into the discussion and --- in fact --- the analysis was done in a  Rindler frame which is just flat spacetime. The idea works because the action for gravity splits up into two natural parts {\\it neither} of which is generally covariant but are related to each other by the remarkable identity (\\ref{lageh}) which --- as far as I know --- was not noticed before. The sum of the two parts  is generally covariant but the expression for individual parts can be ascertained in the local Rindler frame specifically because these parts are {\\it not} generally covariant.   The fundamental postulate we use is in equation (\\ref{postulate}) and it does {\\it not} refer to any horizon. To see how this comes about, consider any spatial plane, say the $y-z$ plane, in flat spacetime. It is always possible to find a Rindler frame in the flat spacetime such that the chosen surface acts as the horizon for some Rindler observer. In this sense, any plane in flat spacetime must have an entropy per unit area. Microscopically, I would expect this to arise because of the entanglement over length scales of the order of $\\sqrt{{\\cal A}_P}$. We have defined in (\\ref{postulate}) the  entropy per unit area rather than the total entropy in order to avoid having to deal with global nature of the surfaces (whether the surface is compact, non compact etc.). This approach also provides a natural explanation as to why the gravitational coupling constant is positive. It is positive because entropy and area are positive quantities.  The result emphasizes the role of two dimensional surfaces in fundamental physics. A two dimensional surface is the basic minimum one needs to produce region of inaccessibility and thus entropy from lack of information. When one connects up gravity with spacetime entropy it is is inevitable that the coupling constant for gravity has the dimensions of area in natural units. The next step in such an approach will be to find the fundamental units by which spacetime areas are made of and provide a theoretical, quantum mechanical description for the same. This will lead to the proper quantum description of spacetime with Einstein action playing the role of the free energy in the thermodynamic limit of the spacetime."
    },
    "0209/astro-ph0209189_arXiv.txt": {
        "abstract": "{ The spiral edge-on galaxies from the `Revised Flat Galaxy Catalog' (RFGC) are identified with the Extended Source Catalog of the Two Micron All Sky Survey (2MASS). A relative number of the 2MASS detected galaxies is 2996/4236 = 0.71. We consider the statistical properties of the Tully-Fisher relations for the edge-on galaxies in the $B, I, J, H,$ and $K_s$ bands. The slope of derived TF relations increases steadily from 4.9 in the $B$ band to 9.3 in the $K$ one. The effect is mainly due to the internal extinction which is different in dwarf and giant spiral galaxies seen edge-on that leads to the tight correlation between galaxy color and luminosity. The moderate scatter of the RFGC galaxies in the ``color-luminosity'' diagram, 0$\\fm$86, provides us with a ``cheap'' method of mass measurements of distances to galaxies on the basis of modern photometric sky surveys. ",
        "introduction": "The Tully-Fisher (TF) relation (Tully \\& Fisher,1977) between luminosity and rotation velocity of spiral galaxies is a basic tool for studying large-scale motions of galaxies because it provides us with distances independent of galaxy redshifts. Using this relation, Tully \\& Fisher and their numerous followers excluded usually very tilted spiral galaxies burdened with strong internal extinction. However, as it was shown by Karachentsev (1989), internal extinction is not a dominant reason for scatter of spiral galaxies on the TF diagram, and edge-on galaxies may be successfully used to map cosmic streamings. For such a purpose the `Flat Galaxy Catalog', FGC, (Karachentsev et al, 1993) and its renewed version, RFGC (Karachentsev et al, 1999) were prepeared. Selection of galaxies into the RFGC was carried out based on two simple geometric criteria, when the major angular diameter of galaxies is greater than 0.6 arcmin and the apparent axial ratio, $a/b$, is greater than 7. Here the major and minor diameters correspond to the standard isophote of $25^m/\\sq\\arcsec$ in the $B$ band. The RFGC catalog covers the entire northern and southern sky, and contains 4236 galaxies mostly of Sc--Sd morphological types. The all-sky distribution of RFGC galaxies looks quite smooth because late-type spiral galaxies are mostly found in the general field, avoiding rich cluster cores with their large virial motions. Only a small number of the RFGC galaxies have apparent integral magnitudes, $B_t$, from RC3 (de Vaucouleurs et al. 1991). But for all its galaxies the RFGC presents marginal B- magnitudes derived from angular diameters of the galaxies as well as from their surface brightness and other parameters (Kudrya et al. 1997). Being reduced to the RC3 photometric system, these magnitudes are characterized with an error of $\\sigma(B) = 0\\fm3$. The TF relation `absolute magnitude vs. HI line width', plotted for $\\sim$800 RFGC galaxies, has a slope of $-$5.3 and a scatter of $\\sigma(M_B) = 0\\fm56$. ",
        "conclusions": ""
    },
    "0209/astro-ph0209140_arXiv.txt": {
        "abstract": "Evolution of primordial fluctuations in a Brans-Dicke type scalar-tensor gravity theory is comprehensively investigated. The harmonic attractor model, in which the scalar field has its harmonic effective potential in the Einstein conformal frame and the theory relaxes toward Einstein gravity with time, is considered. The evolution of adiabatic initial perturbations in flat SCDM models is examined from the radiation-dominated epoch up to the present. We discuss how the scalar-tensor gravity affects the evolution of metric and matter perturbations, mainly focusing on the observational consequences, i.e., the matter power spectrum and the power spectrum of cosmic microwave background temperature. We find that the early time deviation is characterized only by the large static gravitational constant while the late time behavior is qualitatively different from that in Einstein gravity because the time variation of the gravitational constant and its fluctuation have non-negligible effects. The attracting scalar-tensor gravity affects only small scale modes due to its attracting nature, the degree of which is far beyond the post-Newtonian deviation at the present epoch. ",
        "introduction": "The existence of massless scalar partners associated with the tensor field of Einstein gravity is generically predicted by the recent attempts toward unifying all elementary forces in nature based on supergravity, superstrings \\cite{GSW}, and other higher dimensional gravity theories. In these theories, time variations of fundamental constants such as the gravitational constant and the fine structure constant are naturally introduced. A recent claim of the time varying fine structure constant from observations of QSO absorption lines\\cite{Webb} may be a piece of evidence.  Scalar-tensor gravity theories, whose original version was proposed by Jordan \\cite{J} and Brans and Dicke \\cite{BD} and were extended in a more general framework later \\cite{BW}, naturally provide coupling between the massless scalar fields and the tensor field of Einstein gravity. Scalar-tensor gravity theories are almost only possible alternatives to Einstein gravity. Moreover, scalar-tensor theories may supply a new approach to implementing the inflationary scenario (called extended inflation) \\cite{LS}-\\cite{L}. The time-varying gravitational ``constant'' which is their distinctive feature slows the inflationary expansion from exponential to power-law in time, and then the inflationary epoch has finite period, thereby solving the so-called ``graceful exit'' problem. Furthermore, scalar-tensor theories provide a natural framework of realizing the time-variation of fundamental constants (gravitational constant) via the dynamics of the Brans-Dicke dilaton (for review see \\cite{CU}).  In the Jordan-Brans-Dicke theory \\cite{J},\\cite{BD} (hereafter, we refer to it as the Brans-Dicke theory for simplicity) which is the simplest example of scalar-tensor theories, a constant parameter $\\omega$ is introduced. In the limit $\\omega \\rightarrow \\infty$, the gravitational constant can not change and Einstein gravity is recovered. Although scalar-tensor theories including the Brans-Dicke theory are compatible with Einstein gravity in several aspects, they have many deviations from it. Weak-field experimental tests in solar-system have constrained the post-Newtonian deviation from Einstein gravity, $\\omega > 500$ \\cite{REA},\\cite{CMW}. Measurement of the signal time delay of millisecond pulsars or the light deflection of quasars may raise this limit \\cite{CMWGQ}.  In more general scalar-tensor theories, $\\omega$ can vary depending on the scalar field. In cosmological models based on such theories, it has been pointed out that there is generally an attractor mechanism that drives $\\omega$ to $\\infty$ in the late cosmological epochs \\cite{DN}. The nature of gravity can be significantly different in the early universe. Hence, information on the different cosmological epochs may constrain such theories. A simple and natural extension of the Brans-Dicke theory to the attractor model is the harmonic attractor model in which the scalar field has a quadratic effective potential of a positive curvature in the Einstein conformal frame. The analysis of big-bang nucleosynthesis (BBN) in this model \\cite{DP} restricts two parameters characterizing the potential (its curvature $\\beta$ and today's gradient $\\alpha_0$). It is concluded that the BBN limit on the possible deviation from Einstein gravity ($2\\omega_0+3={\\alpha_0}^{-2}$) is much stronger than the present observational limits in large $\\beta(>0.3)$ models. Aside from BBN, we have another source of the information about the early universe that is the cosmic microwave background (CMB). The trace of primordial fluctuation can be seen clearly in the CMB anisotropy spectrum where the information on the early universe up to the last scattering time ($z \\sim 1000$) is integrated on the acoustic peaks. The precise CMB data which will be provided in near future \\cite{MAP},\\cite{PLANCK} should further constrain the allowed parameter region. Toward probing gravity theories by CMB data, we examine the cosmological perturbation evolution in the context of scalar-tensor gravity theories.  Our primary goal is to clarify the growth process of perturbations in the scalar-tensor cosmology and understand the influence on the resultant observable power spectra. We also revisit the perturbations of the Brans-Dicke theory which has been investigated by several authors \\cite{NARI,LMB,CK}. Although the contemporary plausible model is a vacuum dominated model, we adopt flat standard CDM (SCDM) models for three reasons. The first reason is that the universe necessarily experiences matter domination once at least. Therefore, we should begin by revealing the effect of the scalar-tensor gravity up to the matter-dominated epoch without confusing curvature or vacuum energy effect with it. The second reason is that the dependence of the scalar evolution on the matter density ($\\Omega_0 h^2$) is quite simple as shown in Section IV. The final reason is that, if we take account of the current constraint for $\\omega_0$, the attracting scalar-tensor gravity virtually becomes Einstein gravity before vacuum domination and hence any extra process originating from the scalar-tensor gravity would not occur after vacuum domination.  This paper is organized as follows: In Section II, the field equations and the background cosmological evolution equations in the scalar-tensor theory are described. In Section III, we present the cosmological perturbation theory based on the scalar-tensor gravity. We demonstrate the numerical calculations in Section IV. The resultant matter power spectra and the CMB temperature anisotropy spectra are illustrated in Section V. Finally, some conclusions are in Section VI. In Appendix, the analytic approximate solutions of the background and the perturbation equations in the Brans-Dicke theory are discussed. ",
        "conclusions": "We have comprehensively studied the perturbation evolution in the scalar-tensor cosmological model. It is shown that the scalar-tensor gravity requires some modification to the standard evolution that is mainly caused by the variable gravitational constant and also brings additional process derived from its fluctuation. We provide the insight into their influence on matter evolution and thus it becomes possible to interpret the variation in the observable quantities.  In the radiation-dominated epoch, the scalar-tensor gravity behaves like Einstein gravity. The influence on the perturbation spectrum is the larger horizon length alone. Its observational consequence is the smaller turnover scale in the matter power spectrum. On the other hand, in the matter-dominated epoch, the nature of gravity is qualitatively different from Einstein gravity. Consequently, aside from the shift of the acoustic peak locations and the diffusion cut off scale, the enhancement and the attenuation which originate from the scalar field perturbation and others appear in the CMB anisotropy spectrum. The attracting scalar-tensor gravity especially affects only the small scale power because of its attracting nature and the degree is far beyond the post-Newtonian deviation at the present epoch.  In this paper, we have concentrated on studying the influence of the scalar-tensor gravity to cosmological perturbations and their resultant observable power spectra. In the future work, we will put constraint on the model parameters $\\alpha_0$ and $\\beta$ by using the precise CMB and large scale structure data provided in near future."
    },
    "0209/astro-ph0209154_arXiv.txt": {
        "abstract": "Using NASA's Submillimeter Wave Astronomy Satellite (SWAS) we have examined the production of water in quiescent and shocked molecular gas through a survey of the 556.936 GHz \\swash2o\\ transition of ortho-\\HtwoO\\ in the NGC 1333 molecular core. These observations reveal broad emission lines associated with the IRAS~2, IRAS~4, IRAS~7, and HH7-11 outflows.  Towards 3 positions we detect narrow ($\\Delta v \\sim$ 2--3 km s$^{-1}$) emission lines clearly associated with the ambient gas.   The SWAS observations, with a resolution of $\\sim 4'$, are supplemented with observations from the Infrared Space Observatory (ISO) and by an unbiased survey of a $\\sim$17$'$ $\\times$ 15$'$  area, with $\\sim 50''$ resolution, in the low-J transitions of CO, \\thCO , \\CeiO , \\NtwoHp , \\CHthreeOH , and SiO.  Using these combined data sets, with consistent assumptions, we find beam-averaged ortho-\\HtwoO\\ abundances of $> 10^{-6}$ relative to \\Htwo\\ for all four outflows.   A comparison of SWAS and ISO water data is consistent with non-dissociative shock models, provided the majority of the ortho-\\HtwoO\\ \\swash2o\\ emission arises from cool post-shock material with enhanced abundances. In the ambient gas the ortho-\\HtwoO\\ abundance is found to lie between 0.1 -- 1 $\\times 10^{-7}$ relative to \\Htwo\\ and is enhanced when compared to cold pre-stellar molecular cores. A comparison of the water emission with tracers of dense condensations and shock chemistry finds no clear correlation.  However, the water emission appears to be associated with the presence of luminous external heating sources which power the reflection nebula and the photodissociation (PDR) region. Simple PDR models are capable of reproducing the water and high-J \\thCO\\ emission, suggesting that a PDR may account for the excitation of water in low density undepleted gas as suggested by Spaans \\& van Dishoeck (2001). ",
        "introduction": "Water in cold quiescent molecular gas is theorized to form through a series of ion-molecule chemical reactions that occur early in the reaction sequence that links atomic oxygen to H$_2$, the most abundant molecule. Recently NASA's Submillimeter-Wave Astronomy Satellite detected water vapor emission arising from ortho-\\HtwoO\\ (\\oHtwoO ) ground state transition (\\swash2o ) in numerous dense cores within extended molecular clouds \\citep{snell_h2o}. Interestingly, there appears to be a dichotomy in the presence of water detected by SWAS in the sense that emission from water vapor has yet to be detected in cold (T $\\le 15-20$ K) star-less objects such as TMC-1, B68, or $\\rho$ Oph D \\citep{snell_h2o, bergin_b68h2o}. However, in warmer (T $>$ 20 K) giant molecular cloud cores, which are associated with sites of multiple star formation, water is readily detected.  In each case the derived water abundance, or abundance limit, is well below theoretical predictions, which is believed to be the result of the freeze-out, or depletion, of oxygen onto grain surfaces \\citep{bergin_impl, viti_h2o, charnley_swasiso}. It is not clear if the non-detections in colder cores are the result of any intrinsic difference in the chemistry due to lower temperature, because the excitation conditions for the \\swash2o\\ transition favors emission in warm gas. An alternative has been proposed by \\citet{spaans_h2o} who suggest that the enhanced penetration of ultraviolet (UV) photons in clumpy giant molecular clouds lowers the water vapor column density through photodissociation but enhances the emissivity due to the heating of the gas by the UV radiation. This would allow ion-molecule chemistry to account for water detections in warm giant molecular clouds but colder clouds, such as the star-less objects, require freeze-out of oxygen in the form of \\HtwoO\\ \\citep{spaans_h2o}.  Conditions in the interstellar medium also favor water production through a variety of other mechanisms besides ion-molecule chemistry.  Water is believed to be produced efficiently through reactions on the surfaces of cold dust grains \\citep{tielens_hagen}. However, the sublimation temperature of water ice is $\\sim 110$ K \\citep{fraser_h2olab}, much greater than typical dust grain temperatures, so this frozen reservoir is not returned to the gas phase. An additional formation channel is linked to a series of neutral-neutral reactions with activation barriers that can be overcome in gas that is heated to temperatures greater than $\\sim$300 K \\citep{wagner_oxy, graff_oxy}. Such high temperatures can be found in shocked gas, where material entrained in the bipolar outflows associated with the birth of a star impact the surrounding gas, or in the immediate vicinity of embedded sources. Indeed ESA's {\\em Infrared Space Observatory} (ISO) and SWAS have convincingly detected water vapor emission in hot gas towards several star-forming regions \\citep{liseau_hh54, cecc_h2o, harwit_h2o, melnick_bnkl, neufeld_h2o, molinari_hh711, wright_h2o, giannini_h2o, maret_h2o, benedettini_h2o}. The water abundance determined for these regions is also significantly enhanced, by several orders of magnitude, over that estimated in the ambient gas within other nearby clouds by \\citet{snell_h2o}. These abundance enhancements are expected to persist for $\\sim 10^{5}$ yr, whereupon the molecules deplete onto grain mantles \\citep{bmn98}. This led to the suggestion by \\citet{bmn98} that molecular gas could be chemically enriched in \\HtwoO\\ due to repeated exposure to numerous shocks. Furthermore, once the dynamical effects of a shock have dissipated, enhanced abundances of water vapor, above that produced by ion-molecule reactions, might be found as a shock ``legacy'' in the low velocity quiescent gas.  To examine the production of water in the interstellar medium we present the results of a biased survey of the NGC 1333 molecular core for emission in the 557 GHz ground state of ortho-\\HtwoO . This source is relatively close, at a distance of 220 pc \\citep{cernis90}, and is an excellent laboratory to investigate the various potential mechanisms for water formation and emission. First, NGC 1333 is a well known reflection nebula illuminated by two nearby B stars, BD $+$30.549 (B8V) and SVS~3 (B6V).    The radiation from these external sources allows for a search for water emission associated with photodissociation regions (PDR) within the clumpy NGC 1333 cloud \\citep{sk01}, as suggested by \\citet{spaans_h2o}. Second, near-infrared photometry has revealed a dense cluster of young stars embedded within the molecular gas \\citep{asr94, lal96, as97}. These clustered young stars have powered a burst of energetic outflow activity manifested in the detection of clumps of vibrationally excited H$_2$ emission and a  large number of Herbig-Haro objects \\citep{asr94, bdr96, hl95}. These clumps of high excitation are associated with a complex cluster of overlapping molecular outflows that dominate the structure in the dense core of the NGC 1333 molecular cloud and apparently have carved out large cavities within the dense gas \\citep{lcl96, lefloch_dust, ks00}. Thus, it is possible that the outflows have enriched the quiescent gas with water vapor, which is also a focus of this paper.  We report detections of spectrally resolved water vapor emission in several distinct outflows, and also from several positions tracing the quiescent gas. To aid in the analysis of the SWAS data, we present results from a supplementary survey of NGC 1333, using the Five College Radio Astronomy Observatory (FCRAO), in transitions of CO, \\thCO , \\NtwoHp , \\CHthreeOH , and SiO. Our analysis of these combined data sets suggests that the quiescent water abundance is enhanced relative to cold pre-stellar molecular cores. Despite the pervasive presence of outflowing gas we find no supporting evidence that the quiescent water vapor is due to past shock episodes.  Rather the correspondence between the water emission and high-J \\thCO\\ emission suggests that a PDR could provide additional heating to excite water in relatively low density quiescent gas. The observations and data reduction are presented in \\S 2, and in \\S 3 we present the results and a comparison of the integrated emissions maps with the water vapor distribution. In \\S 4 we derive an estimate of the \\oHtwoO\\ abundance in the high and low velocity gas. In \\S 5 we discuss the origin of water vapor emission in NGC 1333. Our conclusions are given in \\S 6. ",
        "conclusions": "We have presented the results of a biased survey for \\swash2o\\ emission from \\oHtwoO\\ in the core of the NGC 1333 molecular cloud using the Submillimeter Wave Astronomy Satellite.  This survey is biased in the sense that known centers of activity were centered within the 3\\farcm3 $\\times$ 4\\farcm5 SWAS beam. These data are supplemented by an unbiased FCRAO survey in the J=1--0 transitions of $^{12}$CO, $^{13}$CO, \\CeiO , \\NtwoHp , \\HCOp , and the J=2--1 transition of SiO.  Each of these transitions were mapped, with beam sampling, over an area that encompasses all SWAS observations. The primary results are listed below.  (1) We report the detection of broad \\oHtwoO\\ emission in the IRAS~2, IRAS~7, and HH7-11 outflows.  The detection of \\swash2o\\ emission in the IRAS~4 flow has been reported previously \\citep{neufeld_h2o}.   In each case the emission is accompanied by narrow absorption features at the quiescent cloud velocity. Towards 3 positions we detect narrow ($\\Delta v < 3$ \\kms ) emission clearly associated with the quiescent gas.    (2) Using these data sets we derive beam-averaged abundances for outflow emission in NGC 1333. In the outflows the water abundance relative to \\Htwo\\ is $>$ 10$^{-6}$. These abundances are averaged within the large SWAS beam and, hence, we have no information on structures smaller than $4'$.   However, with this caveat in mind, our results confirm the enhancement of \\oHtwoO\\ in the NGC~1333 molecular outflows. A combination of published observations, and those reported here, find broad emission in SiO and \\CHthreeOH\\ in the HH7-11, IRAS~2, and IRAS~4 outflows. These same sources are the strongest centers of \\oHtwoO\\ outflow activity detected by SWAS.  Although the sample is quite limited, this is suggestive of a potential correlation between regions with SiO and \\CHthreeOH\\ abundance enhancements and detectable water emission. Models of non-dissociative (C-type) shocks are used to constrain the origin of the 557 GHz outflow emission seen by SWAS and 179 $\\mu$m emission detected by ISO.   These models suggest that the water abundance is enhanced within the shocks, but fail to reproduce the water emission unless the majority of the \\swash2o\\ emission arises from cold post-shock gas within a portion of the extended outflows.  (3) In the quiescent gas beam-averaged abundances lie between $0.1 - 1 \\times 10^{-7}$. The quiescent water emission is correlated with unusually strong \\thCO\\ J = 5--4 emission and is found near the NGC 1333 reflection nebula and PDR. Through a PDR model we find that the local enhancements of the FUV field increase the gas temperature in relatively low density gas (\\nhtwo\\ $\\sim 10^4$ \\cc ) allowing for simple gas phase chemistry (with slightly lower gas phase abundances due to photodissociation) to  account  observed fluxes in both \\thCO\\ J = 5--4 and \\oHtwoO\\ \\swash2o. These results are consistent the results of \\citet{spaans_h2o} who examined water emission in S140. Thus the excitation of water in low density gas, that is undepleted due to longer depletion timescales and exposed to local enhancements of the FUV field, may account for the extended 557 GHz water emission found towards a variety of molecular sources. However, depletion of oxygen in the form of water is still required for regions with significant amounts of dense gas along the line of sight."
    },
    "0209/astro-ph0209362_arXiv.txt": {
        "abstract": "The large majority of extragalactic star cluster studies done to date have essentially used two or three-passband aperture photometry, combined with theoretical stellar population synthesis models, to obtain age estimates.  The accuracy to which this can be done depends on the number of observations through different (broad-band) filters available as well as, crucially, on the actual wavelength range covered.  I show, based on the examples of the nearby starburst galaxies NGC 3310 and M82, that the addition of, in particular, near-infrared passbands to a set of optical filters greatly enhances our ability to disentangle age, metallicity and extinction parameters for star clusters with ages younger than a few Gyr.  In addition, for the intermediate-age star cluster system in the fossil starburst region of M82 we find (i) a well-defined burst of cluster formation at slightly older ages than derived from previous estimates based on optical fluxes alone, and (ii) that by considering {\\it only} the clusters originating in this burst of cluster formation, we uncover the first conclusive observational evidence for an unambiguous turn-over in the luminosity function of a coeval star cluster system with an age as young as $\\sim 1$ Gyr. ",
        "introduction": "The distribution of cluster brightnesses, known as the Cluster Luminosity Function (CLF), is one of the most important diagnostics for the study of extragalactic compact star cluster populations.  For the old globular cluster (GC) systems the CLF shape is well-established: it is roughly Gaussian, with the peak or turn-over magnitude at $M_V^0 \\simeq -7.4$ and a FWHM of $\\sim 3$ mag (Harris 1991, Whitmore et al. 1995, Harris et al.  1998).  The well-studied young star cluster (YSC) population in the Large Magellanic Cloud, on the other hand, displays a power-law CLF (Elson \\& Fall 1985, Elmegreen \\& Efremov 1997).  {\\sl Hubble Space Telescope (HST)} observations have provided CLFs for young compact cluster systems in more distant galaxies and are continuing to do so.  Although incompleteness effects often preclude detection of a turn-over in the CLF (e.g., Whitmore \\& Schweizer 1995, Schweizer et al.  1996, Miller et al.  1997), in galaxies with YSC systems for which deep observations are available there is little evidence for an intrinsic turn-over (but see de Grijs, Bastian \\& Lamers 2002b).  In most of these cases, the CLF shapes are consistent with power laws down to the completeness threshold (but see Miller et al. 1997, de Grijs et al.  2001, 2002b).  GC formation models suggest that the distribution of the initial cluster masses (and, therefore, of the initial cluster luminosities) is closely approximated by a power law (e.g., Harris \\& Pudritz 1994, McLaughlin \\& Pudritz 1996, Elmegreen \\& Efremov 1997).  The processes responsible for the depletion of, preferentially, low-luminosity, low-mass star clusters over time-scales of a Hubble time, leading to the Gaussian CLFs observed, include tidal interactions with the gravitational field of the parent galaxy and evaporation of stars through two-body relaxation within clusters.  From the models of Gnedin \\& Ostriker (1997) and Elmegreen \\& Efremov (1997) it follows that {\\it any} initial mass (or luminosity) distribution will shortly be transformed into peaked distributions.  However, Vesperini (2000, 2001) has demonstrated that, due to dynamical friction effects affecting the high-mass clusters, considerable fine tuning of the model parameters is required to produce from an initial power-law distribution a Gaussian-type mass function with parameters similar to those observed for well-studied GC systems.  However, all of these models are valid {\\it only} for time-independent Milky Way-type gravitational potentials; galaxy-galaxy interactions will obviously have a major effect on the resulting (time varying) gravitational potential, in which the dynamical star cluster evolution is likely significantly different. ",
        "conclusions": ""
    },
    "0209/astro-ph0209538_arXiv.txt": {
        "abstract": "{ We present the last pointed observation of AM\\,Her carried out during the life of the BeppoSAX satellite. It was bright at the beginning of the observation, but dropped to the lowest X-ray level ever observed so far. The X-ray emission during the bright period is consistent with accretion occurring onto the main pole of the magnetized white dwarf. The rapid change from the {\\em active state} to the low deep state indicates a drop by a factor of 17 in the accretion rate and hence that accretion switched-off. The short timescale (less than one hour) of this variation still remains a puzzle. Optical photometry acquired simultaneousy during the low state shows that the white dwarf remains heated, although a weak emission from the accretion stream could  be still present. Cyclotron radiation, usually dominating the V and R bands, is negligible thus corroborating the possibility that AM\\,Her was in an off-accretion state. The X-ray emission during the inactive state is consistent with coronal emission from the secondary late type star. ",
        "introduction": "AM Her is the prototype of Polars, strongly magnetic Cataclysmic Variables (mCVs) (10-230\\,MG), and consists of a magnetized ($\\sim$ 14\\,MG) white dwarf accreting from a late type (M4V) Roche lobe-filling secondary star. Polars are characterized by long-term (months to years) high and low accretion states which, due to the absence of an accretion disc in these systems, reflect changes in the mass loss rate of the donor star. AM Her is the brightest and best  monitored Polar in the optical range and hence represents a test object to study the evolution of the instantanous mass accretion rate with time and then to understand the causes of the mass transfer variations from the secondary star.  Bright and faint luminosity states ($\\Delta V \\sim$ 2-3\\,mag) occur on irregular timescales from less than a day to months.  Different models have been discussed to account for the long-term  mass transfer variations in CVs (Livio \\& Pringle 1994; King \\& Cannizzo 1998) and the occurrence of starspots at the inner Lagrangian point appears to be the most likely explanation. Along this line, Hessman et al. (2000) derived the mass transfer rate history of AM Her, using the long term optical variability and convert it into starspot filling factors. They conclude that the density of the starspots  near the L1 point is unusually high (about 50$\\%$).  On the other hand, monitoring of the X-ray activity  has been relatively sparce, but has already brought further insights into the variability of the mass accretion rate. In particular, since its launch, we undertook a programme with the BeppoSAX satellite to monitor the  X-ray behaviour of AM Her, to infer  the evolution of the X-ray luminosity and its X-ray spectral variability. In this work we present new and the most recent observations of AM Her carried out during the life of BeppoSAX together with simultaneous optical photometry acquired at the Loiano Bologna Observatory, which reveal further new results on the accretion variability and on the possible identification of the coronal X-ray emission from the magnetically active secondary star. ",
        "conclusions": "The last pointed BeppoSAX observation of AM\\,Her caught the source during its deepest low state ever observed. The X-ray flux shows a rapid variability from an {\\em active} to a {\\em quiescent state} similar to that observed during another prolonged low state in 1997 (de Martino et al. 1998). Unfortunately, also for this second time, AM\\,Her was bright at the beginning of the observation and, although the bright state is consistent with X-ray accretion-induced emission, it is not possible to assess whether this was a temporary accretion event or if the source was previously in a  constant accretion epoch. However, the drop in X-ray flux indicates that the accretion rate decreased by a factor of $\\sim$ 17 in less than one hour. A very rapid variation (18\\,min) in the X-ray flux also has been recorded in XMM-Newton data of UZ\\,For during a deep low X-ray state and interpreted as an accretion event (Still \\& Mukai 2001).  For the {\\em active state}, we derive an estimate of the accretion luminosity, assuming that about half of the thermal bremsstrahlung and cyclotron radiation emitted from the post-shock region is intercepted by the white dwarf and re-emitted in the UV ($\\rm L_{UV}=L_{tb} + L_{cyc}$) and neglecting the contribution of a re-processed component in the EUV range (cfr. G\\\"ansicke et al. 1995 and similar reasoning in de Martino et al. 1998). We then estimate an accretion rate of $\\rm 4.9\\times 10^{-11}\\,M_{\\odot}\\,yr^{-1}$.  The drop in mass accretion rate therefore indicates that the X-ray emission during the {\\em quiescent state} is not due to accretion. The much lower temperature (1.5\\,keV) is compatible with coronal temperatures of M type dwarfs (Schmitt et al. 1990) as well as the low X-ray luminosity in the 0.05-3\\,keV of $\\rm 8.6\\times 10^{29}\\,erg\\,s^{-1}$ is  compatible with those observed in active late type stars (Pallavicini et al. 1990). Furthermore the emission measure derived during the {\\em quiescent state} EM= $\\rm 4.6\\times 10^{52}\\,cm^{-3}$ is also consistent with coronal values. We have also compared the X-ray luminosity during the {\\em quiescent state} with that expected from coronal emission of rapidly rotating late type stars. For a secondary star filling its Roche lobe in a 3.09\\,hr orbital period binary, the period-radius relation (Patterson 1984) gives for the secondary star $\\rm R_{sec}=0.32-0.36\\,R_{\\odot}$. The predicted saturation value for the X-ray luminosity due to  rotation is $ \\rm \\propto R_{sec}^{2}$ (Fleming et al. 1989), corresponding to $\\rm 4.9-6.3\\times 10^{29}\\,erg\\,s^{-1}$ (0.3-3\\,keV band), in remarkable agreement with the luminosity derived in the same range for the BeppoSAX {\\em quiescent state} ($\\rm 6.4\\times 10^{29}\\,erg\\,s^{-1}$).  \\noindent Moreover, the lack of V and R band modulations indicate that, even if present, cyclotron emission is very weak, thus implying that AM\\,Her really switched-off accretion during the BeppoSAX observation. All this indicates that the X-ray emission can be indeed and more safely ascribed to the secondary star than done previously for the {\\em quiescent state} in September 1997.  \\noindent As for the 1997 BeppoSAX data set, still remains unclear  the cause of the rapid drop of accretion flux, with timescales remarkably close to the dynamical timescale of the secondary star.  The observations of a  rapidly evolving burst in UZ\\,For observed with XMM-Newton further confirms that such rapid changes occur in AM\\,Her stars and a coronal mass ejection event at the L1 point may be the cause of the observed active state  (cfr. de Martino et al. 1998 for discussion).  It is clear that secondaries in these close  binary systems are still far from being understood and further X-ray observations during low states would help in assessing the nature of these active secondary stars."
    },
    "0209/astro-ph0209012_arXiv.txt": {
        "abstract": "I present results on the correlation between galaxy mass, luminosity, and metallicity for a sample of spiral and irregular galaxies having well-measured abundance profiles, distances, and rotation speeds. Additional data for low surface brightness galaxies from the literature are also included for comparison. These data are combined to study the metallicity-luminosity and metallicity-rotation speed correlations for spiral and irregular galaxies.  The metallicity luminosity correlation shows its familiar form for these galaxies, a roughly uniform change in the average present-day O/H abundance of about a factor 100 over 11 magnitudes in B luminosity. However, the O/H - $V_{rot}$ relation shows a change in slope at a rotation speed of about 125 km s$^{-1}$. At faster $V_{rot}$, there appears to be no relation between average metallicity and rotation speed. At lower $V_{rot}$, the metallicity correlates with rotation speed. This change in behavior could be the result of increasing loss of metals from the smaller galaxies in supernova-driven winds. This idea is tested by looking at the variation in effective yield, derived from observed abundances and gas fractions assuming closed box chemical evolution. The effective yields derived for spiral and irregular galaxies increase by a factor of 10-20 from $V_{rot}$ $\\approx$ 5 km s$^{-1}$ to $V_{rot}$ $\\approx$ 300 km s$^{-1}$, asympotically increasing to approximately constant $y_{eff}$ for $V_{rot}$ $\\ga$ 150 km s$^{-1}$. The trend suggests that galaxies with $V_{rot}$ $\\la$ 100-150 km s$^{-1}$ may lose a large fraction of their SN ejecta, while galaxies above this value tend to retain metals. ",
        "introduction": "The strong correlation between galaxy metallicity $Z$ and galaxy luminosity $L$ is one of the more significant phenomenological results in galaxy chemical evolution studies. The basic correlation of O/H vs. $L$ was demonstrated for irregular galaxies by \\cite{lrspt79} and confirmed by \\cite{skh89}. \\cite{gs87} extended this correlation to include spiral galaxies, demonstrating that a correlation between metallicity and luminosity extends over a factor of 100 in metallicity and 11 magnitudes in blue luminosity $M_B$. Parallel studies of elliptical galaxies \\citep{faber73,bh91} showed a similar metallicity-luminosity correlation. \\cite{zkh94} noted that both ellipticals and star-forming galaxies exhibited similar metallicity luminosity correlations, despite the different measurement techniques. This suggests that similar phenomena govern the metallicity-luminosity relationship in spiral/irregular and elliptical galaxies.  It is not clear that the primary correlation in this case is between metallicity and luminosity. For disk galaxies, luminosity, rotation speed (or total mass), surface brightness, and Hubble type are all correlated to some degree (Roberts 1994; de Jong \\& Lacey 2000), so it is not certain which is the main driver of the correlation. Rotation speed ($V_{rot}$), taken as the maximum measured speed or the speed on the flat part of the rotation curve, or $V^2_{rot}R$, where $R$ is the radius, may be preferable measures of galaxy mass, but \\cite{zkh94} and \\cite{gar97} found that the metallicity-$V_{rot}$ correlation was not noticeably better than the metallicity-luminosity correlation, at least for spiral galaxies.  The origin of this correlation is open to debate. A metallicity-luminosity correlation can arise if smaller galaxies have larger gas fractions than larger galaxies, which is seen statistically in the local universe \\citep{rh94,mdb97,bdj00}. This situation can arise if small galaxies either evolve more slowly (lower star formation rate per unit mass) or are younger on average than more massive systems, and have thus simply processed a smaller fraction of their gas into stars. The relatively blue colors of low-mass irregular galaxies indicates that this must be true at some level.  On the other hand, it has been popular recently to ascribe the correlation to the effects of selective loss of heavy elements from galaxies in supernova-driven outflows. Low-mass galaxies are expected to lose a larger fraction of SN ejecta than more massive systems because of their smaller gravitational potentials. However, the loss of gas from galaxies depends not just on gravitation potential, but also on the vertical structure of the ISM and details of radiative losses, among other things. Numerical modeling of outflows by \\cite{mlf99} showed that it is extremely difficult to `blow away' the ISM of a gas-rich dwarf galaxy via starburst superwinds, although selective loss of metals may be relatively easy. However, a generalized description of the conditions for loss of metals from galaxies is not yet available.  The question of loss of metals from galaxies is of high interest because of the apparently ubiquitous presence of metals in the intracluster medium (ICM) and the intergalactic medium (IGM; Ellison et al. 2000). The source of these metals, whether from galactic outflows, tidal/ram pressure stripping in dense environments, or pre-galactic stars, is debated. It is therefore worthwhile to look for evidence that galaxies are actually losing metals to the IGM. I begin here by examining the $L$-$Z$ correlation in a different way. Figure 1(a) shows the familiar O/H - $M_B$ correlation for spiral and irregular galaxies, taking the value of O/H at the disk half-light radius $R_{eff}$ as a surrogate for the average ISM abundance in spirals; the data used is discussed below.  B-band magnitudes are often considered to be a less-than-ideal surrogate for galaxy mass, however. The blue light from galaxies can be strongly affected by interstellar extinction, and variations in recent star formation history can make the relation between stellar mass and blue luminosity uncertain, especially for bursting dwarf galaxies. Therefore, in Figure 1(b) I plot O/H versus rotation speed $V_{rot}$ (obtained from rotation curves) for the same set of galaxies; data for this plot are listed in Tables 1-3. This panel shows that, interestingly, the correlation between O/H and $V_{rot}$ does not increase steadily, but rather turns over for rotation speeds greater than 125 km s$^{-1}$. Above this value, the data indicate that the mean metallicity for the most massive spirals is essentially constant. (To a large degree, this behavior becomes apparent because we have changed from a logarithmic scale [B magnitude] to a linear scale [$V_{rot}$]). This result suggests that there is a velocity/mass threshold below which the metallicity evolution of a galaxy varies according to mass. ",
        "conclusions": "As noted in Section 2, either outflows or metal-poor inflows can reduce the effective yield. Then how do we interpret the results shown in Figure 4? The trend of decreasing $y_{eff}$ with decreasing $V_{rot}$ and galaxy luminosity suggests increasing importance of SN-driven outflows in the smallest galaxies, as suggested by \\cite{ds86}, and we tend to favor this model as the simplest interpretation. The smooth increase of $y_{eff}$ with increasing $V_{rot}$ would then imply that the fraction of material lost is a simple function of the galaxy potential. Flattening of the trend for $V_{rot}$ $>$ 100-150 km s$^{-1}$ would indicate that such massive galaxies essentially retain all of the metals produced by stars in those galaxies; gas may be driven into the halo by SN-driven outflows, but the ejected gas eventually rains back down onto the galaxy to enrich the disk. \\cite{martin99}, using a different argument, suggested that escape of hot gas in galactic winds occurs for galaxies with $V_{rot}$ $<$ 130 km s$^{-1}$, similar to what I find here. Alternatively, it would be necessary to posit a conspiracy between inflow and outflow that keeps the massive spirals at roughly the same $y_{eff}$ regardless of mass.  In a strict sense, it is not possible to rule out pure inflow of metal-poor gas as a cause of the trend in Figure 4 from this data alone. The interpretation of Figure 4 in this case would be that dwarf galaxies have accreted a larger fraction of outside gas at relatively late times compared to spirals. Evidence for inflow of gas onto dwarf galaxies is much more scarce than evidence for outflows, although \\cite{ks95} and \\cite{tbh97} argue that NGC 5253 may be accreting gas that is fueling the current starburst, and \\cite{wm98} show that IC 10 has extended plumes of \\hi\\ that they interpret as gas infalling onto the galaxy. However, to explain the low effective yields for the dwarf galaxies would require as much as 80-90\\% of their gas to have been accreted at late times without much star formation and subsequent metal enrichment. This was demonstrated by \\cite{ke99} for galaxies for a wide range in the ratio of accretion rate to star formation rate. Only models in which gas was accreted much faster than the star formation rate acheived low effective yields; galaxies with slow accretion (and no outflows) tend to have effective yields that approached the true yield with time. Local Group dwarf irregular galaxies show a large range of star formation histories, but they are consistent with a roughly constant, if low, star formation rate over the past 10 Gyr \\citep{mateo98}, so unless they have all accreted large amounts of gas in the recent past, it would be difficult to explain their low effective yields with pure infall.  Stripping of gas can also reduce $y_{eff}$ by decreasing the gas fraction at a fixed metallicity. This could be particularly important for low-mass satellites of large galaxies. Stripping could indeed lead to a trend like that seen in Figure 4, since dwarf galaxies are more likely to be stripped of gas than the more massive spirals. This could certainly be relevant to galaxies like the Magellanic Clouds, where evidence for tidal stripping is strong. Moreover, the detection of hot gas in even small groups of galaxies \\citep{mdmb93} makes it more likely that ram pressure can play a role even in relatively low-density galaxy environments. \\cite{bc02} cite Holmberg II as an example of possible gas stripping from a dwarf galaxy despite its location on the edge of the M81 group, and IC 10 could also fall into this picture. On the other hand, Sextans A, which is on the edge of the Local Group, shows no evidence for a disturbed gas disk \\citep{wh02}. Sorting out the origin of extended gas structures in dwarf irregulars will require deep \\hi\\ mapping of many more galaxies. In the meantime, the dwarf irregulars included in this sample are relatively isolated, so we might expect that gas stripping is less likely to be important in determining the effective yields. Furthermore, stripping is much less efficient in reducing the effective yield than direct loss of metals: very small galaxies like Leo A would have to have some 80-90\\% of their gas stripped to account for their low effective yields.  If we accept the premise that the variation of $y_{eff}$ is the result of increasing importance of outflows of metals in low-mass galaxies, then the rotation speed tells us which galaxies are likely to be enriching the IGM. We can then discuss the fates of outflows observed in various starburst galaxies. For example, we can infer that low-mass starburst galaxies such as I Zw 18 ($V_{rot}$ $\\approx$ 30 km s$^{-1}$) and NGC 1569 ($V_{rot}$ $\\approx$ 40 km s$^{-1}$) probably lose a large fraction of the metals produced by their stars into the IGM. On the other hand, in more massive spirals with outflows, such as NGC 253 ($V_{rot}$ $\\approx$ 210 km s$^{-1}$) and NGC 4631 ($V_{rot}$ $\\approx$ 150 km s$^{-1}$), metals ejected into their halos probably remain bound, to fall back onto the disk in a fountain. M82 presents a more ambiguous case because of its peculiar rotation curve: the galaxy has a maximum rotation speed of about 200 km s$^{-1}$, but this declines with distance from the nucleus \\citep{sofue97}. Another complication is its location in the dense M81 group environment, since gas ejected into the halo can suffer tidal stripping.  If the fraction of metals that is lost by a galaxy is known, it is possible to estimate the contribution to IGM enrichment by galaxies of a given $L$ or $V_{rot}$. This is important to understanding the element abundance pattern in the IGM, as dwarf galaxies have a different element abundance pattern than spirals. Given the present uncertainties in data for effective yields, it is not yet possible to say with great precision what range of galaxies contribute most to IGM enrichment, but we can make a crude estimate of the relative contribution to IGM enrichment as a function of $M_B$.  I approach this by assuming a Schechter luminosity function of the form  \\begin{equation} \\phi(L) = (\\phi^*/L^*) (L/L^*)^{\\alpha} exp(-L/L^*), \\end{equation} where $L$ is the B-band luminosity, and I adopt $\\alpha$ = $-$1.2, log $L^*$ = 10.16 ($M_B$ = $-$20.2) and $\\phi^*$ = 1.0$\\times$10$^{-2}$ based on the field galaxy studies of \\cite{ellis96}, \\cite{blanton01}, and \\cite{folkes99}, scaled to a Hubble constant of 75 km s$^{-1}$ Mpc$^{-1}$. The actual normalization of $\\phi^*$ is not important since I will only examine the relative contributions of metals to the IGM. I further assume:  \\noindent \\begin{enumerate} \\item The trend of $y_{eff}$ vs. $V_{rot}$ is well described by the function \\begin{equation} log(y_{eff}) = -1.95 - {(320 - V_{rot})^4\\over 9.1\\times 10^9}. \\end{equation} I assume that spirals with $V_{rot}$ $>$ 150 km s$^{-1}$ retain all their newly-produced metals with log $y_{eff}$ = --1.95. The difference between this value and $y_{eff}$($V_{rot}$) given by the above expression represents the fraction of metals lost by a galaxy. \\noindent \\item The relation between $V_{rot}$ and $M_B$ for my galaxy sample can be represented by the expression \\begin{equation} M_B = -6.8~log(V_{rot}) - 4.56. \\end{equation} This is not meant to be a new derivation of the Tully-Fisher relation, only a parameterization of the data for this galaxy sample. \\noindent \\item The O/H-luminosity relation for the galaxy sample is represented by \\begin{equation} log(O/H) = -0.16 M_B - 6.4. \\end{equation} \\noindent \\item $M/L$ is taken to be equal to one for all galaxies. A factor two error in $M/L$ will turn out to have only a small effect on the relative contributions of ejected metals. \\end{enumerate}  As a crude approximation I take the total mass of oxygen in a galaxy to be the measured oxygen mass fraction times the mass in stars. Then the mass of oxygen lost by the galaxy is \\begin{equation} M_{lost}(O) = 12~(O/H)~L_B~{M\\over L}~{0.0112 \\over y_{eff}(V_{rot})} - 1, \\end{equation} where 0.0112 is the average $y_{eff}$ for the massive spirals. The factor twelve is the conversion from the number ratio O/H to oxygen mass fraction. Convolving this expression with the galaxy luminosity function gives the relative mass of oxygen ejected by galaxies of a given luminosity into the IGM in a given volume element. The results for this set of assumptions is illustrated in Figure 7, where I show the relative mass of oxygen ejected by of all galaxies of a given $M_B$ or $V_{rot}$.  Figure 7 indicates that dwarf galaxies with $M_B$ $>$ --16, $V_{rot}$ $<$ 50 km s$^{-1}$ completely dominate the enrichment of the IGM in the present day universe. Although the approximations made above are crude in some cases, this conclusion is fairly robust to those approximations. A systematic increase in M/L for more massive galaxies would shift the distribution to higher $M_B$ and $V_{rot}$, but is not likely to make the dwarfs less dominant. A steeper luminosity function would only increase the relative contribution of the dwarf galaxies. \\cite{trent94} has similarly noted that dwarf galaxies may contribute the bulk of the intracluster gas in galaxy clusters, although \\cite{gm97} argue that the dwarf galaxies are unlikely to account for all of the ICM in clusters. Note that the analysis presented here does not consider the total amount of gas that may be ejected by galaxies, and that there are other ways to account for the ICM, such as ram pressure stripping and tidal stripping of gas ejected during disk galaxy mergers in dense cluster environments \\citep{mihos01}.  If dwarf galaxies dominate in contributing metals to the IGM, the abundance pattern of the intergalactic medium should reflect that of the dwarf galaxies. Irregular galaxies tend to be deficient in nitrogen and carbon relative to oxygen. Fe/O is not well known in ionized nebulae because of depletion of iron onto grains, but could be subsolar or solar, depending on the galaxy's star formation history. If the outflows consist preferentially of supernova ejecta from massive stars, then we would expect C, N and Fe to be deficient relative to O and other alpha-capture elements. Note that this analysis may not apply to Ly$\\alpha$ systems at high redshift. These systems have young ages, and comparison with the Galactic halo abundance pattern may be more relevant, although in many respects the abundances in metal-poor dwarf galaxies are similar to those in the halo. Nevertheless, we might expect a similar relationship between metal loss and rotation speed to hold in the early universe as well, determining which systems contribute material to the IGM.  Ejection of metals by galaxies is closely connected to the question of feedback of stellar energy into galaxies, a subject of intense interest currently. Hierarchical clustering models for galaxy formation have some difficulty in reproducing the relation between luminosity and rotation speed in disk galaxies; too much angular momentum is transferred from the disk to the halo during the collapse, leading to disks that rotate too fast for their size. Feedback of energy from stars and supernovae into the ambient ISM is expected to relieve this problem, by reheating the gas and preventing it from collapsing too quickly. How feedback should be parameterized in galaxy formation models is poorly understood, however, because star formation itself is poorly understood. The fraction of metals ejected by a galaxy must be a function of the feedback process. This connection is complex, however, and is beyond the scope of this paper. Still, the relation between effective yield and rotation speed in galaxies, if it results from the loss of metals in galactic winds, must be related to the feedback of energy from stars into the ISM, and thus should be reproduced by successful models of feedback in galaxy evolution."
    },
    "0209/astro-ph0209224_arXiv.txt": {
        "abstract": "We present new near- and mid-IR observations of 19 Class I/flat-spectrum young stellar objects in the nearby $\\rho$ Ophiuchi ($d$ = 125 pc) and Serpens ($d$ = 310 pc) dark clouds. These observations are part of a larger systematic infrared multiplicity survey of Class I/flat-spectrum objects in the nearest dark clouds. We find 7/19 (37\\% $\\pm$ 14\\%) of the sources surveyed to be multiple systems over a separation range of $\\sim$ 150 -- 1800 AU. This is consistent with the fraction of multiple systems found among older pre-main-sequence stars in each of the Taurus, $\\rho$ Ophiuchi, Chamaeleon, Lupus, and Corona Australis star-forming regions over a similar separation range. However, solar-type main-sequence stars in the solar neighborhood have a fraction approximately one-third that of our Class I/flat-spectrum sample (11\\% $\\pm$ 3\\%). This may be attributed to evolutionary effects or environmental differences. An examination of the spectral energy distributions (SEDs) of the SVS 20 and WL 1 binaries reveals that the individual components of each source exhibit the same SED classifications, similar to what one typically finds for binary TTS systems, where the companion of a classical TTS also tends to be of the same SED type. ",
        "introduction": "The past two decades have witnessed substantial progress in the study of star formation and its aftermath, the birth of planetary systems. Observations have led the way, with much of the credit due to both spaceborne missions, such as the Infrared Astronomical Satellite (IRAS), the Infrared Space Observatory (ISO), and the Hubble Space Telescope (HST), and to ground-based infrared (IR) imaging surveys from NOAO/SQIID, the NASA Infrared Telescope Facility (IRTF) and other facilities. Advances in detection and instrumentation have been matched by equally exciting theoretical developments. Researchers have delineated key features of a young star's structure and the manner in which it affects its surroundings, both thermally and mechanically.  As we broaden our focus from individual stars, we find that there is a universal tendency for these objects to form in groups, rather than as isolated entities. On the largest scales, IR surveys have revealed stellar aggregates still embedded in their parent molecular clouds (e.g., \\cite{zmw93}). These must eventually become optically visible associations or bound clusters. However, the smallest entities created within such groups are generally $not$ single stars. We have known for many years that the majority of solar-to-late-type field stars are binaries or multiples (\\cite{al76}; \\cite{dm91}; \\cite{fm92}), but only in the past decade have significant numbers of pre-main-sequence (PMS) T Tauri stars (TTSs) been surveyed for multiplicity. These recent IR surveys have shown that young low-mass stars have binary fractions that are greater than or equal to that of the field (e.g., \\cite{gnm93}; \\cite{math94}; \\cite{simon95}). These pairs appear to be coeval, that is, at the same evolutionary age (\\cite{hss94}; \\cite{bz97}). Recent observations at millimeter continuum wavelengths also find multiple sources at the origins of extended molecular outflows and optical jets (\\cite{lmw97}). The current findings set this fact in its proper evolutionary context. Most stars $form$ as part of a binary system, rather than find their partners later in life.  The large binary frequency in visible, PMS stars prompts us to investigate even younger systems. Recent sub-arcsecond surveys of young stars have almost exclusively focused on PMS TTSs and have virtually ignored younger systems with Class I/flat-spectrum energy distributions. Thus we know very little about the multiplicity of self-embedded young stars. Is multiplicity the rule even among more deeply embedded objects, those still forming out of their parent clouds? What is their binary frequency, and how does this compare to that of PMS stars? What are the energy distributions of their components? New multiplicity surveys of known embedded low-mass stars must be conducted to address these questions.  We have initiated the first systematic near-to-mid-IR multiplicity survey of $\\sim$ 100 Class I/flat-spectrum young stellar objects in six nearby (d $\\la$ 300 pc) dark clouds. Our survey has several purposes: 1) to study the binary fractions and separations of protostars; 2) to diagnose the luminosities and evolutionary states of their components; and 3) use our near-IR photometry (in conjunction with existing IRTF and Keck spectra) to place protostars in Hertzsprung-Russell (H-R) diagrams to diagnose their stellar properties.  All of the sources in our survey were selected such that they have either Class I (protostellar) or flat-spectrum (slightly more evolved than Class I) spectral energy distributions (SEDs; see \\cite{als87}; hereafter ALS87, 1988; \\cite{lada87}). These objects are the direct evolutionary progenitors of the well-studied TTSs, and many are reasonably bright in the near-IR (m$_{K}$ $\\sim$ 10). Although these objects have peak fluxes at far-IR wavelengths, there are no far-IR high-resolution observatories (0.85 -- 3.0 m for SIRTF -- SOFIA) currently operating. Also, millimeter wavelength observations of the very few known Class I binaries show that their envelopes can be so large that they overlap, making it difficult to discern the individual components at these long wavelengths (\\cite{lmw00}).  In this contribution, we present the initial results of our survey. Specifically, we discuss our near- and mid-IR observations of 19 Class I/flat-spectrum objects in the $\\rho$ Ophiuchi ($d$ = 125 pc; \\cite{kh98}) and Serpens ($d$ = 310 pc; \\cite{del91}) dark clouds. We discuss the observations and data reduction procedures in $\\S$2. In $\\S$3, we present the results of our survey, and discuss the results in $\\S$4. We summarize our primary results in $\\S$5. ",
        "conclusions": "We have obtained new near- and mid-IR observations of 19 Class I/flat-spectrum objects in the $\\rho$ Ophiuchi and Serpens dark clouds. These observations are part of a larger systematic infrared multiplicity survey of Class I/flat-spectrum objects in the nearest dark clouds. The primary results and conclusions derived from the present study can be summarized as follows:  1. The CSF for the Class I/flat-spectrum sources surveyed is 7/19 (37\\% $\\pm$ 14\\%). These results are consistent with the CSFs derived for PMS stars in each of the Taurus, $\\rho$ Ophiuchus, Chamaeleon, Lupus, and Corona Australis star-forming regions over a similar separation range. However, the CSF for solar-type main-sequence stars in the solar neighborhood in this separation range (11\\% $\\pm$ 3\\%) is approximately one-third that of our sample, and may be attributed to evolutionary effects or environmental differences.  2. Among the sources which were found to be binary, and for which SEDs of both sources could be constructed, we find that the individual components of SVS 20 and WL 1 exhibit the same SED classifications. This behaviour is similar to what one typically finds for TTSs, where the companion of a classical TTS also tends to be a classical TTS (\\cite{ps97}).  3. The strength of the mid-infrared and sub-millimeter emission observed in IRS 43/VLA 1 suggests that, if an embedded, late-stage protostar were present in this source, we would have detected it in our near-infrared images as well; we did not. In addition, the near- and mid-IR ``secondary'' sources observed in IRS 51 are at different position angles and separations from each other. We therefore suggest that the companion sources observed in IRS 43 and IRS 51 may not be stellar objects, but rather are associated with molecular outflows from the primary YSO, although IRS 43 does form a wide binary with GY 263."
    },
    "0209/astro-ph0209542_arXiv.txt": {
        "abstract": "We present new constraints on the system parameters of the SW~Sextantis star DW~Ursae~Majoris, based on ultraviolet ($UV$) eclipse observations with the {\\em Hubble Space Telescope}. Our data were obtained during a low state of the system, in which the $UV$ light was dominated by the hot white dwarf (WD) primary. The duration of the WD eclipse allows us to set a firm lower limit on the mass ratio, $q = M_2/M_1 > 0.24$; if $q < 1.5$ (as expected on theoretical grounds) the inclination must satisfy $i > 71^{\\circ}$. We have also been able to determine the duration of WD ingress and egress from our data. This allows us to constrain the masses and radii of the system components and the distance between them to be $0.67\\leq M_1/M_{\\odot} \\leq 1.06$, $0.008\\leq R_1/R_{\\odot} \\leq 0.014$, $M_2/M_{\\odot} > 0.16$, $R_2/R_{\\odot} > 0.28$ and $a/R_{\\odot} > 1.05$. If the secondary follows Smith \\& Dhillon's mass-period relation for CV secondaries, our estimates for the system parameters become $M_1/M_{\\odot} = 0.77 \\pm 0.07$, $R_1/R_{\\odot} = 0.012 \\pm 0.001$, $M_2/M_{\\odot} = 0.30 \\pm 0.10$, $R_2/R_{\\odot} = 0.34 \\pm 0.04$, $q = 0.39 \\pm 0.12$, $i = 82^{\\circ} \\pm 4^{\\circ}$ and $a/R_{\\odot} = 1.14 \\pm 0.06$.  We have also obtained time-resolved $I$- and $K$-band photometry of DW~UMa during the same low state. Using Bessell's spectral-type vs $(I-K)$ color calibration, we estimate the spectral type of the donor star to be $M3.5 \\pm 1.0$. This latter result helps us to estimate the distance towards the system via Bailey's method as $d = 930 \\pm 160$~pc.  Finally, we have repeated Knigge et al.'s WD model atmosphere fit to the low-state $UV$ spectrum of DW~UMa in order to account for the higher surface gravity indicated by our eclipse analysis. The best-fit model with surface gravity fixed at $\\log{g}=8$ has an effective temperature of $T_{\\rm eff} = 50,000 \\pm 1000$~K. The normalization of the fit also yields a second distance estimate, $d = 590 \\pm 100$~pc. If we adopt this distance and assume that the mid-eclipse $K$-band flux is entirely due to the donor star, we obtain a second estimate for the spectral type of the secondary in DW~UMa, $M7 \\pm 2.0$. After discussing potential sources of systematic errors in both methods, we conclude that the true value for the distance and spectral type will probably be in between the values obtained by the two methods. ",
        "introduction": "Nova-likes (NLs) are a subgroup of cataclysmic variables (CVs) in which a late-type main sequence secondary loses mass onto a white dwarf (WD) primary via Roche lobe overflow. If the WD does not have a strong magnetic field, the transferred matter forms an accretion disk with a bright spot where the stream of matter hits the disk. Unlike the more commonly known dwarf nova type CVs, NLs do not undergo quasi-periodic outbursts. They are instead characterized by a high, steady accretion rate which prohibits the disk instability mechanism responsible for the dwarf nova outbursts. However, some NLs with periods between 3-4~hrs also exhibit low states during which mass transfer from the secondary and/or accretion onto the WD decreases or shuts off completely. The reason for the low states is uncertain; one possibility is that they are related to magnetic activity of the secondary star (see Hessman, G${\\rm \\ddot{a}}$nsicke \\& Mattei 2000). Supporting this theory is the fact that the orbital period of these systems is very close to the upper edge of the CV period gap between 2-3~hrs. The absence of CVs in this region of the orbital period distribution, is believed to occur as a consequence of a change in the internal structure of the secondary. More specifically, it is thought that at $P_{\\rm orb} \\approx 3$~hrs, the secondary becomes fully convective and magnetic braking ceases to be the main mechanism by which angular momentum is removed from the system. As a result, the secondary loses contact with its Roche lobe and mass transfer ceases. Gravitational radiation takes over as the dominant, but less efficient, angular momentum loss mechanism. At $P_{\\rm orb}\\approx 2$~hrs the orbit has shrunk enough for the secondary to re-establishes contact, and mass transfer resumes. For a thorough review of CVs refer to Warner (1995).  The subject of our paper, DW~Ursae~Majoris, is an eclipsing CV with a period of $P_{\\rm orb} =$ 3.28~hrs and belongs to a subclass of NLs called SW Sextantis stars (Thorstensen et al.~1991). SW~Sex systems are grouped together because they exhibit several unusual properties: (a) they are often eclipsing systems with orbital periods of 3-4~hrs; (b) their continuum eclipses are more V-shaped (as opposed to U-shaped) than those of other NLs; (c) their optical emission lines are single-peaked, instead of double-peaked (as expected for high-inclination, disk-formed lines); (d) their Balmer and HeI lines remain largely unobscured during primary eclipse, but display absorption events at the opposite orbital phase; (e) the radial velocity curves derived from their optical emission lines lag substantially behind the phase one expects from the WD on the basis of eclipses. Several models have  been put forward to account for the SW~Sex stars, and although each is capable of explaining a subset of the SW~Sex behavior, none has so far been able to explain all the features listed above (e.g. Knigge et al. 2000 and references therein).  One reason for our poor understanding of the SW~Sex phenomenon is the scarcity of reliable system parameters for members of this class. This scarcity is a direct consequence of the defining SW~Sex characteristics. More specifically, radial velocity studies of SW~Sex stars are of limited value, given the ubiquitous and significant phase lags seen in the radial velocity curves of these systems. Eclipse studies have been similarly unsuccessful, since WD contact points are not evident in the V-shaped eclipse light curves of SW~Sex stars (perhaps because the disks are self-occulting; Knigge et al. 2000). Also, the high accretion rates exhibited by these systems during normal states cause the late-type main sequence secondary to be invisible against the glare of the WD and/or accretion disk.  The goal of this paper is to improve on this situation by deriving a set of reliable system parameters for DW~UMa. This is possible because a recent {\\em Hubble Space Telescope} (HST) observation of the system found DW~UMa in a deep low state, during which the WD dominated the ultraviolet ($UV$) light. This is the second, in a series of papers, describing these observations. The first, Knigge et al. (2000), dealt with the low-state $UV$ spectrum. Here we analyze the time-resolved behavior around the eclipse and discuss $I$- and $K$-band photometry obtained during the same low state. The remainder of this article is organized as follows: In section 2, we describe the HST and ground-based observations and their reduction. Next, in section 3, we discuss our determination of the contact phases describing the eclipse of the WD by the secondary. In section 4, we use the contact phases an other information to determine the parameters of the binary system and its constituents. Then, in section 5, we calculate estimates for the distance to DW~UMa and for the spectral type of the donor star in two different ways: using our $I$- and $K$-band photometry and re-fitting the Knigge et al.'s (2000) WD model atmosphere to the low-state $UV$ spectrum of DW~UMa. Finally, in section 6, we discuss our results and compare the system parameters we have derived to those previously reported for DW~UMa. ",
        "conclusions": "SW~Sex stars display a range of peculiarities that do not seem to fit the standard steady accretion disk model for NLs. There is currently no single agreed-upon explanation for the SW~Sex phenomenon. Indeed, no real consensus has been reached about whether SW~Sex stars deserve a NL-sub-class label in the first place. Rather, the SW~Sex {\\em syndrome} is so widespread (also seen in X-ray binaries; Hynes et al. 2001) that we must consider the possibility that some important element is missing in our standard picture of the accretion flows. Whatever the nature of the SW~Sex stars, they may play an important role in CV evolution: with very few exceptions (e.g., BT Mon with $P_{\\rm orb}=8$~hrs; Smith, Dhillon \\& Marsh 1998), SW~Sex stars have orbital periods falling in the 3 -- 4 hrs range, just above the period gap.   Reliable system parameters are desperately needed in order to understand the origin of the SW~Sex phenomenon. Unfortunately, the very characteristics that distinguish SW~Sex stars from other NLs have also prevented us from accurately determining their physical and geometrical parameters. In particular, the phase-shifted radial velocity curves seen in these systems make it very difficult to accurately determine the $K$-velocities of the component stars (from which other system parameters could then follow). In addition, the high accretion rates exhibited by the SW~Sex stars hamper the detection of the individual components of the system since the accretion disk dominates the emission and may even be self-occulting (Knigge et al. 2000). These latter effects are greatly reduced during the sporadic low states displayed by at least some of the SW~Sex stars.  In the case of DW~UMa, radial velocity studies have been carried out based on both high-state (Shafter, Hessman \\& Zhang 1988) and low-state (Dhillon at al. 1994; also see Rutten \\& Dhillon 1994) observations. However, as emphasized by both sets of authors, neither study has provided reliable results. In the former study, significant ($55^{\\circ}-75^{\\circ}$) phase lags were seen in all the radial velocity curves from which $K_1$ was estimated; in the latter study, the radial velocity curve was based on emission lines that clearly arise on the irradiated front face of the secondary (Rutten \\& Dhillon 1994), and may therefore provide only a lower limit for $K_2$.  Our HST low state $UV$ observations of DW~UMa have provided a rare opportunity to accurately determine the system parameters of an SW~Sex star from the eclipses of its WD. This has been possible because the accretion disk contribution is dramatically reduced, revealing the sharp ingress and egress features that mark the WD eclipse. Therefore, we have been able to avoid many of the difficulties associated with radial velocity studies. However, potential sources of systematic errors remain and include (1) our assumption that the WD is entirely unobscured, (2) the application of the mass-period relation for CV secondaries, and (3) the application of the theoretical mass-radius relation for isolated WDs.  As it turns out, the system parameters listed in Table~1 agree reasonably well with the constraints inferred by Shafter et al. (1988) and Dhillon et al. (1994) from radial velocity analyses. However, as argued above, we consider our new estimates to be considerably more reliable, particularly the ones labelled ``fundamental'' in Table~1. Our estimate of the WD mass ($M_1/M_{\\odot} \\simeq 0.77$) is considerably higher than that obtained by Knigge et al. (2000) from their model atmosphere fit to the low state HST $UV$ spectrum ($M_1/M_{\\odot} \\simeq 0.5$). Their estimate was based on the surface gravity inferred from the spectral fit and essentially the same mass-radius relation used here but for a WD with $T_{\\rm eff}= 46,000$~K (Panei et al. 2000). Given the systematic uncertainties inherent in spectroscopic $\\log{g}$ estimates, we consider our new set of constraints on the WD parameters more reliable.  Both of our distance estimates, $d = 930 \\pm 160$~pc (from Bailey's method) and $d = 590 \\pm 100$~pc (from the WD model fit), indicate that DW~UMa is quite far away. For comparison, Marsh \\& Dhillon (1997) estimated $d \\gtappeq 850$~[$450$]~pc if the secondary is an $M4$~$[M5]$ star, based on the absence of clear secondary signatures in their low-state $I$-band spectroscopy. We note, however, that even though our estimate of $I \\simeq 18.7$ for the secondary star is fainter than the (out-of-eclipse) $I$-band magnitude they actually observed ($I \\simeq 18$), it is brighter than the limit of $I > 19.5$ ($<25\\%$) they suggest for the secondary's contribution to their spectrum. If the mid-eclipse $I$-band flux in our date is due to the secondary, the latter should have contributed roughly half of the $I$-band flux in their spectrum.  We conclude this paper by reiterating the need for accurate system parameters for the SW~Sex stars. Given that most SW~Sex stars are high-inclination systems, it is natural to try and exploit this in the way we have done here. However, eclipse analyses based on WD ingress and egress features are usually impossible, since the light from these systems is usually dominated by their (possibly self-occulting) accretion disks. In the case of DW~UMa, we have succeeded only because the system was caught in a deep low state. This raises the obvious question whether other SW~Sex systems may also exhibit such low states. This would obviously be interesting for its own sake, but would also open a new avenue of attack for determining their system parameters.  A partial answer to this question is already available: of the 14 objects listed as SW~Sex stars in the G\\\"{o}ttingen Online CV Catalog\\footnote{http://www.cvcat.org}, 5 (including DW~UMa) are already classified as VY~Scl stars (i.e., nova-like variables that exhibit occasional low states). We therefore advocate a long-term photometric monitoring program of SW~Sex stars. This would tell us whether all SW~Sex stars are also VY~Scl stars and permit us to exploit the low states when they occur."
    },
    "0209/astro-ph0209068_arXiv.txt": {
        "abstract": "{ Employing the most recent parametrization of the baryon-baryon interaction of the Nijmegen group, we investigate, in the framework of the Brueckner--Bethe--Goldstone many-body theory at zero temperature, the influence of neutrino trapping on the composition, equation of state, and structure of neutron stars, relevant to describe the physical conditions of a neutron star immediately after birth (protoneutron star). We find that the presence of neutrinos changes significantly the composition of matter delaying the appearance of hyperons and making the equation of state stiffer. We explore the consequences of neutrino trapping on the early evolution of a neutron star and on the nature of the final compact remnant left by the supernova explosion. ",
        "introduction": "Neutrinos play a crucial role in the physics of supernova explosions (Janka and M\\\"uller 1996) and in the early evolution of their compact stellar remnants (Burrows and Lattimer 1986, Janka and M\\\"uller 1995). During the collapse of the pre-supernova core, a large number of neutrinos is produced by electron capture process. Immediately following the core bounce the radius of the newly formed neutron star shrinks from about 100~km to about 10~km. During this same period (up to about 1~second after core bounce) substantial matter accretion occurs on the compact star (this accretion may eventualy led to the formation of a black hole). As the newly formed neutron star contracts the neutrino mean free path $\\lambda_\\nu$ decreases, and above a critical value of the density ({\\it neutrino trapping density}) $\\lambda_\\nu$ becomes smaller than the stellar radius. Under these physical conditions  neutrinos are {\\it trapped} in the star, {\\it i.e.,} the neutrino diffusion time is of the order of a few tens of seconds. Neutrino trapping has a strong influence on the overal {\\it stiffness} of the equation of state (EoS) of dense stellar matter. Thus, the physical conditions of the hot and lepton-rich newborn neutron star (the so-called protoneutron star) differ substantially from those of the cold and deleptonized neutron star. Nevertheless, this stage nearly fulfills the conditions of hydrostatical equilibrium (Burrows \\& Lattimer 1986).  The composition and the structure of protoneutron stars have been systematically investigated by Prakash {\\it et al.} (1997) and by Strobel {\\it et al.} (1999) using a large sample of modern equations of state of dense stellar matter. The implications of the early evolution of a protoneutron star on the concept of neutron star maximum mass have been studied by the authors of Refs. (Bombaci 1996, Prakash {\\it et al.} 1997, Strobel \\& Weigel 2001).  Due to the rapid increase of the nucleon chemical potentials with density, hyperons ($\\Lambda$, $\\Sigma^{-}$, $\\Sigma^{0}$, $\\Sigma^{+}$, $\\Xi^{-}$ and $\\Xi^{0}$ particles) are expected to appear in the core of neutron stars, as suggested in the pioneer work by Ambartsumyan and Saakyan (1960). Since then the structrural properties  of these {\\it hyperon stars} have been studied by many researchers using a variety of aproaches (see {\\it e.g.,} Pandharipande (1971), Glendenning (1985), Keil and Janka 1994, Shaffner and Mishustin (1996), Prakash {\\it et al.} (1997), Balberg and Gal (1997), Baldo {\\it et al.} (2000), Vida\\~na {\\it et al.} (2000a)).  All the previous studies of hyperonic matter with trapped neutrinos have been done in the framework of a relativistic theoretical field model of nucleons and hyperons interacting via meson exchange in a mean field approximation (Keil and Janka 1994, Prakash {\\it et al.} 1997). In the present work, we use a microscopic approach instead, which is based on the Brueckner--Bethe--Goldstone (BBG) many body theory. In our calculations the basic input is the baryon-baryon interaction for the complete baryon octet  ($n$, $p$, $\\Lambda$, $\\Sigma^{-}$, $\\Sigma^{0}$, $\\Sigma^{+}$, $\\Xi^{-}$ and $\\Xi^{0}$) developed recently by Stoks and Rijken (1999). Within this approach we compute the EoS of hyperonic matter with trapped neutrinos and the corresponding properties of newborn hyperon stars. A similar microscopic approach has been recently employed by Baldo {\\it et al.} (2000) and Vida\\~na {\\it et al.} (2000a) to study cold and deleptonized hyperon stars. The primery purpose of the present work is to investigate the effects of neutrino trapping on the structure and evolution of newly formed hyperon stars.  The paper is organized in the following way. A brief review of the Brueckner--Hartree--Fock (BHF) approximation of the BBG many-body theory at zero temperature extended to the hyperonic sector is given in Sec. \\ref{sec:sec2.1}. Equilibrium conditions and Eos of $\\beta$-stable matter are discussed in Sec. \\ref{sec:sec2.2}. Section \\ref{sec:sec3} is devoted to the presentation and discussion of the results. Finally, a short summary and the main conclusions of this work are drawn in Sec. \\ref{sec:sec4}. ",
        "conclusions": "\\label{sec:sec4}  In this paper we have investigated within the framework of the Brueckner--Hartree-Fock approximation the effects  of neutrino trapping on the properties of $\\beta$-stable neutron star matter including nucleonic and hyperonic degrees of freedom.  We have found that the presence of neutrinos changes significantly the compositon of matter with respect to the neutrino-free case: matter becomes more proton rich, muons are not present, and the appearance of hyperons is moved to higher densities. In additon, the number of strange particles is on average smaller and the EoS stiffer in comparison with the neutrino-free case.   We have found that the value of the maximun mass of hyperon stars decreases as soon as neutrinos diffuse out of the star, contrary to what happens when the only baryonic degrees of freedom considered are nucleons.   Using the microscopic EoS developed in the present work we have found that stars having at birth a gravitational mass between  1.28 -- 1.59 $M_\\odot$ are metastable, in other words these stellar configurations remain only stable for several seconds (the neutrino trapping time), collapsing afterwards into low-mass black holes."
    },
    "0209/astro-ph0209297_arXiv.txt": {
        "abstract": "{Following Sereno (\\cite{io02}), we discuss the bending of light rays by spherically symmetric lenses with angular momentum. For several astrophysical systems, such as white dwarfs and galaxies, gravitomagnetism induces a correction on the deflection angle as large as $0.1\\%$. ",
        "introduction": "Gravitational lensing is one of the best investigated phenomena of gravitation. In the framework of general relativity, its lowest-order predictions have been confirmed by observative astrophysics on very different scales. On the other hand, the impressive development of technical capabilities demands for a full treatment of lensing theory to any order of approximation. The study of higher order perturbative terms is the link between weak and strong regimes of the theory.  Mass-energy currents relative to other masses generate space-time curvature. This phenomenon, known as intrinsic gravitomagnetism, is a new feature of general relativity and other conceivable alternative metric theories of gravity and cannot be deduced by a motion on a static background (for a detailed discussion on gravitomagnetism we refer to Ciufolini \\& Wheeler (\\cite{ci+wh95})). In particular, the effect of the angular momentum of the deflector has been studied by several authors (Epstein \\& Shapiro \\cite{ep+sh80}; Ib\\'{a}\\~{n}ez \\& Mart\\'{\\i}n \\cite{ib+ma82}; Ib\\'{a}\\~{n}ez \\cite{iba83}; Dymnikova \\cite{dym86}; Glicestein \\cite{gli99}; Sereno \\cite{io02}).  One of us (Sereno \\cite{io02}) showed as the gravitomagnetic correction to the lensing quantities can be evaluated in the usual framework of lensing theory (see also Capozziello et al. \\cite{cap+al99}), i.e. {\\it i)} weak field and slow motion approximation for the lens; {\\it ii)} thin lens hypothesis (Schneider et al. \\cite{sef}; Petters et al. \\cite{pet+al01}).  In this letter, we consider the gravitomagnetic contribution to the deflection angle for extended gravitational lenses of astrophysical interest. We discuss spherically symmetric deflectors. ",
        "conclusions": "We have investigated the effect of dragging of inertial frames in gravitational lensing for spherically symmetric lenses and a general expression for the deflection angle, to the order $c^{-3}$, has been derived. We have explicitly considered isothermal spheres, power law models and the homogeneous sphere. Both for galaxies and white dwarfs, the gravitomagnetic correction can be as large as $0.1\\%$.  The satellite Hipparcos, launched in 1989 by ESA, can measure the position of stars with accuracy of nearly a milliarcsec. New generation space interferometric mission, such as SIM by NASA (scheduled for launch in 2009), should greatly improve this accuracy. Measurements of deflection of electromagnetic waves could give one of the first experimental evidences of gravitomagnetism."
    },
    "0209/astro-ph0209404_arXiv.txt": {
        "abstract": "{ We present an H-band image of the companion of $\\chi^1$~Orionis taken with the Keck adaptive optic system and NIRC~2 camera equipped with a 300\\,mas-diameter coronographic mask. The direct detection of this companion star enables us to calculate dynamical masses using only Kepler's laws (M$_{\\rm A} =1.01\\pm0.13\\,{\\rm M}_{\\odot}$, M$_{\\rm B} =0.15\\pm0.02\\,{\\rm M}_{\\odot}$), and to study stellar evolutionary models at a wide spread of masses. The application of \\cite{Baraffe1998} pre-main-sequence models implies an age of 70-130\\,Myrs. This is in conflict to the age of the primary, a confirmed member of the Ursa Major Cluster with a canonical age of 300\\,Myrs. As a consequence, either the models at low masses underestimate the age or the Ursa Major Cluster is considerably younger than assumed. } ",
        "introduction": "$\\chi^1$~Ori is a G0V-star and is known to be a single-lined spectroscopic and astrometric binary. The orbital parameters were first derived by \\cite{Lipp1978}. Since then \\cite{Irvin1992} published precise radial-velocity measurements of the orbit. \\cite{Gatewood1994} published an astrometric parallax of the orbit of $\\chi^1$~Ori. Recently, \\cite{Han2002} using their new astrometric data and the radial velocity data from \\cite{Marcy1998} presented a period of ${\\rm P} = 5156.7 \\pm 2.5$~days and a mass ratio $q = {\\rm M}_{\\rm B}/{\\rm M}_{\\rm A} = 0.15 \\pm 0.005$.\\\\ \\begin{figure}[h] {\\includegraphics[angle=0, width=8.5cm]{Eh282_fig1.eps}} \\caption{The H-band image of $\\chi^1$~Ori behind the coronograph in the center and the companion to the left. Note the diffraction ring around the companion.} \\label{fig:image} \\vspace{-0.5cm} \\end{figure} \\cite{McCarthy1986} claimed to have detected the companion directly by speckle imaging techniques, but this has not been confirmed yet. They derive M$_{\\rm V}=6.1$\\,mag, which would place the companion star to $\\chi^1$~Ori about 4\\,mag above the main sequence (\\cite{Henry1999}). \\cite{Han2002} claim that \\cite{McCarthy1986} and subsequent attempts by speckle observations have not been able to detect the companion directly due to instrument limitations.\\\\ The G-type star $\\chi^1$~Ori and its companion form a binary with a very small mass ratio. A direct detection of the secondary would be significant as it would allow the masses to be determined without astrophysical assumption. The derived mass and observed luminosity allow the age to be inferred from comparison to pre-main-sequence evolutionary tracks, which in turn enables a calibration of other alternate estimators. ",
        "conclusions": "\\begin{figure}[h] {\\includegraphics[angle=270, width=8.5cm]{Eh282_fig4.eps}} \\caption{Baraffe et al. (1998) isochrones for solar metallicity in a mass-luminosity plot compared to the position of $\\chi^1$~Ori~B. The error-bars for the mass are derived by the spectroscopy (solid) and for the dynamical mass (dots). The age for $\\chi^1$~Ori~B ranges from 70-130\\,Myrs using the dynamical mass.} \\label{fig:m_mh} \\end{figure} \\begin{figure}[ht] \\vspace{-0.5cm} {\\includegraphics[angle=90, width=9.5cm]{Eh282_fig5.eps}} \\vspace{-0.6cm} {\\includegraphics[angle=90, width=9.5cm]{Eh282_fig6.eps}} \\vspace{-0.6cm} {\\includegraphics[angle=90, width=9.5cm]{Eh282_fig7.eps}} \\caption{Baraffe et al. (1998) tracks for solar metallicity. The horizontal line in the first plot gives M$_{\\rm H}$ for the companion star with the top shaded area indication the $1 \\sigma$~error for M$_{\\rm H}$ and the temperature range. In panel (a), the bottom shaded area is the age range determined for the Ursa Major cluster using different methods. With a mass of $0.15\\,{\\rm M}_{\\odot}$ the companion appears younger compared to the age range of the Ursa Major cluster. In the other two panels the same tracks plotted are for the primary, indicating the position of the primary by the shaded area. In panel (a) and (b) the model parameters are [M/H]=0, Y=0.275 and $L_{\\rm mix} = H_{\\rm P}$. For (c) the parameters have been adjusted to fit the sun to [M/H]=0, Y=0.282 and $L_{\\rm mix} = 1.9 H_{\\rm P}$.} \\label{fig:teff_mh} \\end{figure} The mass of the companion to $\\chi^1$~Ori has been determined precisely to $(0.15\\pm0.005)\\,{\\rm M}_{\\chi^1~{\\rm Ori}}$ (Han \\& Gatewood 2002). The main uncertainty is ${\\rm M}_{\\chi^1~{\\rm Ori}}$. This leads to a spectroscopic ($0.15 \\pm 0.01\\,{\\rm M}_{\\odot}$) and dynamic mass ($0.15 \\pm 0.02\\,{\\rm M}_{\\odot}$), which are both in good agreement.\\\\ The position of the  $\\chi^1$~Ori~B in the mass-luminosity plot (Fig.~\\ref{fig:m_mh} and \\ref{fig:teff_mh}a) compared to the isochrones provided by \\cite{Baraffe1998} indicates that the star lies about $0.50\\pm0.10$\\,mag above the main sequence.\\\\ Figure~\\ref{fig:teff_mh}b and c show H-R diagrams for the primary star including the tracks of \\cite{Baraffe1998}. Figure~\\ref{fig:teff_mh}b shows models for $[M/H] = 0$, $Y=0.275$, and the mixing length of $L_{\\rm mix} = H_{\\rm P}$. \\cite{Baraffe1998} acknowledge that these models do not reproduce the sun at present age. Those tracks and isochrones also do not reproduce $\\chi^1$~Ori~A.\\\\ Figure~\\ref{fig:teff_mh}c shows the same as Fig.~\\ref{fig:teff_mh}b except that the parameters $[M/H] = 0$, $Y=0.282$, and the mixing length of $L_{\\rm mix} = 1.9\\,H_{\\rm P}$ were adjusted to fit the sun. With these parameters the present sun could be reproduced and for $\\chi^1$~Ori~A they also seem to work. The M$_{\\rm H}$~predicted by \\cite{Baraffe1998} is a bit lower than the measured M$_{\\rm H}$~value for $\\chi^1$~Ori~A. This could be because $\\chi^1$~Ori~A is slightly iron underabundant (${\\rm [Fe/H]} = -0.07\\pm0.07$) and the tracks were calculated for solar abundance. No tracks for masses of 0.15 - 0.175\\,M$_{\\odot}$ are available for the model with the parameter set to fit the sun.\\\\ The age prediction by the pre-main-sequence models can be directly compared to other age determinations for the Ursa Major Cluster. While the canonical value for the age of the Ursa Major Cluster is 300\\,Myrs (cf. e.g. \\cite{Soderblom1993}, and references therein) derived by comparing the members of the Ursa Major Cluster nucleus stars in a color-magnitude diagram to theoretical isochrones computed by \\cite{VandenBerg1985}, more recent observations of Sirius' white dwarf companion led \\cite{Holberg1998} to suggest an age of 160\\,Myrs with reference to the cooling tracks of \\cite{Wood1992}. Since Sirius~B is also well-known as a fairly massive degenerate white dwarf with a mass of ${\\rm M}=1.034\\pm0.026~{\\rm M}_{\\odot}$ (Holberg et al. 1998), the initial-final mass relation suggests a progenitor of about 6-7~${\\rm M}_{\\odot}$ which means that we can expect another $\\sim$60-70~Myr for the pre-white-dwarf evolution. Hence, an age only somewhat above 200\\,Myrs may be more in line with this nearby open cluster. More recent white dwarf cooling models of \\cite{Salaris2000} (models with a pure hydrogen atmosphere) suggest the age of the white dwarf of 111\\,Myrs derived from the V-magnitude and the temperature published by \\cite{Holberg1998}. Assuming the lifetime of the progenitor of the white dwarf of 46\\,Myrs this leads to an age of the UMa cluster of 157\\,Myrs. \\\\ The comparison of the age using \\cite{Baraffe1998} (70-130\\,Myrs) to the ages of the Ursa Major Cluster (200-300\\,Myrs) indicate that either: (i) the  Ursa Major Cluster has a larger than expected age spread, (ii) there are problems with the models at a solar and/or at $\\sim 0.15\\,{\\rm M}_{\\odot}$~mass, (iii) the canonical age for the Ursa Major Cluster is too high (300\\,Myrs), or (iv) $\\chi^1$~Ori is not a member of the Cluster. Considering possibility (i), we note that the age spread of 70-300\\,Myrs seems too large for a Cluster.  As for the option (iv),  $\\chi^1$~Ori is a classical member of the Ursa Major Cluster, located near the cluster center. The spectrum of $\\chi^1$~Ori~A would support an age of 200\\,Myrs regarding the activity indicators, as would the cooling tracks for the Sirius B white dwarf."
    },
    "0209/astro-ph0209318_arXiv.txt": {
        "abstract": "From adaptive optics observations with the Palomar 5-meter telescope we place upper limits on the masses of any planetary companions located between $\\sim$30--230~AU away from Vega, where our data are sensitive to depths ranging from $H=12.5$~mag to $H=19.0$~mag fainter than Vega itself.  Our observations cover a plus-shaped area with two $25\\arcsec \\times 57\\arcsec$ elements, excluding $7\\arcsec \\times 7\\arcsec$ centered on the star.   We have identified 2 double and 4 single point sources.  These projected companions are 14.9--18.9~mag fainter than Vega, and if physically associated would have masses ranging from 4 to 35~M$_{\\rm Jup}$ and orbital radii 170--260~AU.  Recent simulations of dusty rings around Vega  predict the presence of a perturbing body with mass $<$2--3~M$_{\\rm Jup}$ and orbital radius $\\sim$40--100~AU, though more massive ($\\lesssim$10~M$_{\\rm Jup}$) planets cannot be excluded.  None of the detected objects are this predicted planet.  Based on a color-magnitude, spectroscopic, and proper motion analysis, all objects are consistent with being background sources. Given the glare of Vega, a 2~M$_{\\rm Jup}$ object near the expected orbital radii would not have been visible at the 5$\\sigma$ level in our data, though any $>$10~M$_{\\rm Jup}$ brown dwarf could have been seen at separation $>$80~AU. ",
        "introduction": "}  The A0V star Vega is famously known since the early days of data return from IRAS as a young main sequence star surrounded by dust \\citep{aum84}. Its age \\citep[270--380~Myr;][]{son01} combined with the large fractional excess luminosity at infrared wavelengths \\citep[$L_{\\rm excess}/L_\\ast \\approx 10^{-5}$ or $M_{\\rm dust} \\approx 1/2 M_{\\rm moon}$;][]{bac93} imply that dust is being generated at the current epoch by either grinding collisions between larger rocky bodies, a.k.a.\\ planetesimals \\citep*{harp84,wei84,zuc93}, or in cometary ejecta \\citep[and references therein]{beu89,beu90}. If the dust is not continuously regenerated it will be depleted by a combination of Poynting-Robertson drag and radiation pressure on a time-scale much shorter than the age of Vega. Discovery of the infrared excess around Vega and other main sequence stars too old to possess the so-called primordial dust and gas disks that are commonly found around 1--10~Myr old stars, led to coining of the term ``debris disk.'' Searches for other examples of ``the Vega phenomenon'' have led to cataloging of a mere tens of objects \\citep[see, e.g.][]{man98,sil00}, mostly early-type stars whose dust was detectable with {\\sl IRAS} or {\\sl ISO}, or observable from the ground with mid-infrared instrumentation on large telescopes.  The mid- and far-infrared (25--850~$\\micron$) emission from Vega is extended over tens of arcseconds \\citep*{aum84,harv84,zuc93,hei98,hol98}. Aperture synthesis imaging at 1.3~mm \\citep*{koe01,wil02} resolved several dust clumps located $\\sim$8--14$\\arcsec$ from the central source (60--110~AU, assuming the Hipparcos parallax of 128.9~milli-arcsec).  One interpretation is that these clumps trace the densest portions of the already inferred face-on circumstellar ring \\citep{den00}.  Additional support for a ring interpretation comes from Vega's spectral energy distribution, which is close to photospheric at shorter wavelengths \\citep[$\\lesssim$20~$\\micron$;][]{hei98}, and suggests an inner gap in the density distribution which may or may not be entirely devoid of hot dust.  At 11.6~$\\micron$ extensions larger than 1/4$\\arcsec$ are ruled out by the imaging of \\citet*{kuc98}. Interferometric work by \\citet{cia01}, however, did suggest extended emission at 2.2~$\\micron$.  Observations of structure in the circumstellar dust around Vega have spawned detailed models for a planetary perturber \\citep{gor01,wil02}. Resonance trapping and gravitational scattering induced by a body of mass 2--3~$M_{\\rm Jup}$ are consistent with the \\citet{hol98} map, and with the interferometric observations of \\citet{koe01} and \\citet{wil02}.  Due to degeneracies in dynamical models \\citep[e.g.,][]{wil02}, more massive planets ($\\sim$10~M$_{\\rm Jup}$) also cannot be ruled out.  Modeling to date assumes a face-on orientation of the presumed dust disk or ring.  Evidence for this geometry comes both from a ring-shaped \\citep[e.g.,][]{hei98} albeit clumpy \\citep{koe01,wil02} dust distribution, and from detailed analysis of stellar line profiles \\citep*[assuming parallel disk and stellar rotation axes;][]{gul94}.  Our experiment was designed to search for low-mass companions within 4--30$\\arcsec$ of Vega, in part to test the aforementioned planetary perturber predictions.  Imaging observations close to this bright source are usually ``burned out'' in survey data such as POSS or 2MASS. Ground-based coronagraphic observations \\citep*{smi92,kal96} have also lacked sufficient sensitivity. Except for NICMOS images \\citep*{sil02} with sensitivity comparable to ours, high dynamic-range observations have not been previously reported. ",
        "conclusions": "We find 8 faint objects within $35\\arcsec$ of Vega that are 15--19~mag fainter than the star at $H$-band. If associated, at the 330~Myr age for Vega, current brown-dwarf cooling models \\citep{bur01,cha00} set their masses at 5--35~$M_{\\rm Jup}$.  The number of detected objects is however consistent with estimates of field star density, and their colors and proper motion indicate that they are not associated with Vega.  We thus exclude the possibility of a distant (80--220~AU; $\\sim$83\\% of this area is imaged), massive ($>$10~$M_{\\rm Jup}$; $>$6~$M_{\\rm Jup}$ for 120--220~AU) planetary/brown-dwarf companion causing the observed dust distribution around Vega. We also detect nothing at the positions of the predicted planetary perturbers, with upper mass limits 7--15~$M_{\\rm Jup}$ ($H$$<$$17$--13), well above the 2--3~$M_{\\rm Jup}$ predictions.  We detect nothing at the position of the mid-infrared dust clumps, placing limits on the possibility of their extragalactic interpretation."
    },
    "0209/astro-ph0209527_arXiv.txt": {
        "abstract": "The neutron capture cross section of the unstable nucleus \\WV\\ has been derived from experimental photoactivation data of the inverse reaction \\WVI \\rgn \\WV . The new result of $\\sigma =  (687 \\pm 110)$\\,mbarn confirms the theoretically predicted neutron capture cross section of \\WV\\ of $\\sigma \\approx 700$\\,mbarn at $kT = 30$\\,keV. A neutron density in the classical $s$-process of $n_{\\rm n} = (3.8 ^{+0.9} _{-0.8}) \\times 10^8$ cm$^{-3}$ is derived from the new data for the \\WV\\ branching. In a stellar $s$-process model one finds a significant overproduction of the residual $s$-only nucleus $^{186}$Os. ",
        "introduction": "\\label{sec:intro} The unstable nucleus \\WV\\ is a so-called branching point in the slow neutron capture process ($s$-process). The nucleus \\WV\\ is produced by neutron capture in the $s$-process from the stable $^{184}$W. At small neutron densities \\WV\\ $\\beta$-decays to $^{185}$Re with a half-life of $T_{1/2} = 75.1$\\,d, and it has been pointed out that the $\\beta$-decay half-life does practically not depend on the temperature at typical $s$-process conditions \\citep{Taka87}. At higher neutron densities \\WV\\ may capture one more neutron leading to the stable \\WVI . It is obvious that the branching between $\\beta$-decay and neutron capture depends on the $\\beta$-decay half-life, the neutron capture cross section, and the neutron density. The half-life and the neutron capture cross section can be measured in the laboratory, and therefore one can determine the neutron density from the observed abundances of the various tungsten isotopes \\citep{Kaepp91}. Additionally, this branching has minor influence on the $^{187}$Os/$^{187}$Re cosmochronometer \\citep{Bosch96}.  Up to now, only theoretical estimates are available for the neutron capture cross section of \\WV\\ because direct neutron capture experiments with radioactive targets are very difficult. Theoretical predictions for the Maxwellian averaged capture cross section at a typical temperature of 30\\,keV vary significantly from 532\\,mbarn \\citep{Kaepp91} and 560\\,mbarn \\citep{Rau00} to 794\\,mbarn \\citep{Hol76}. In a recent compilation a value of $(703 \\pm 113)$\\,mbarn has been adopted \\citep{Bao00}. All calculations used the statistical model. The differences in the results come from the parameterizations of the level density, the gamma-ray strength function, and the neutron-nucleus optical potential.  In order to reduce the uncertainties, a new experiment was performed on the inverse reaction \\WVI \\rgn \\WV . The idea is to find a parameter set for the calculations which reproduces the cross section of \\WVI \\rgn \\WV , and to apply these parameters for the prediction of the \\WV \\rng \\WVI\\ cross section. Such a prediction should be more reliable for one special reaction than previous calculations which used global or local systematics to derive the relevant parameters from neighboring nuclei. The relevant energy region is located close above the threshold of the \\rgn\\ reaction at $S_n = 7194$\\,keV \\citep{Mohr01}. At higher energies experimental data on the \\WVI \\rgn \\WV\\ reaction are available in literature \\citep{Ber69,Gor78,Gur81}, and the results can be found in the compilations of \\citet{Die88} and in CDFE \\citep{CDFE}. We have performed an additional measurement at energies close above the threshold.  In \\S\\,\\ref{sec:exp} we present our experimental set-up. In \\S\\,\\ref{sec:calc} we calculate the cross sections of the \\rng\\ and \\rgn\\ reactions, in \\S\\,\\ref{sec:branch} we derive the $s$-process neutron density from our experimental data, and we apply a stellar $s$-process model to the \\WV\\ branching. \\S\\,5 gives a summary and conclusions. ",
        "conclusions": "\\label{sec:conc} We have measured the photodisintegration cross section of the \\WVI \\rgn\\ \\WV\\ reaction at energies near the reaction threshold. The experimental data have been used to restrict model predictions for the $A = 186$ system and to derive the neutron capture cross section of the inverse \\WV \\rng \\WVI\\ reaction. The result of $\\sigma =  (687 \\pm 110)$\\,mbarn is close to the calculated cross section which was recommended by \\citet{Bao00}, but exhibits significantly improved reliability. %  The $s$-process flow at the branch point isotope \\WV\\ has been analyzed within the classical $s$-process model and within a realistic stellar model for AGB stars. With the classical model one obtains a neutron density of $3.8 \\times 10^8$/cm$^3$ compatible with the analyses of the branchings at $A=147/148$, but incompatible with the branchings at $A=169/170$ and 191/192. This inconsistency indicates that the assumptions of the classical model are too schematic to account for the stellar situation, where the $s$ process takes place. The corresponding analysis based on a more realistic stellar model overestimates the $^{186}$Os abundance by 20\\,\\%. Presently, we are facing the question whether this mismatch is related with remaining uncertainties in other nuclear physics data or whether it originates from the $s$-process model itself. If the nuclear physics uncertainties can be further reduced, the $s$-process branching at \\WV\\ can be interpreted as a sensitive test of models for the important AGB phase of stellar evolution."
    },
    "0209/astro-ph0209182_arXiv.txt": {
        "abstract": "In this paper we present new observations of the gravitational lens system JVAS B0218+357 made with a global VLBI network at a frequency of 8.4~GHz. Our maps have an rms noise of 30~$\\mu$Jy~beam$^{-1}$ and with these we have been able to image much of the extended structure of the radio jet in both the A and B images at high resolution ($\\sim$1~mas). The main use of these maps will be to enable us to further constrain the lens model for the purposes of $H_0$ determination. We are able to identify several sub-components common to both images with the expected parity reversal, including one which we identify as a counter-jet. We have not been successful in detecting either the core of the lensing galaxy or a third image. Using a model of the lensing galaxy we have back-projected both of the images to the source plane and find that they agree well. However, there are small, but significant, differences which we suggest may arise from multi-path scattering in the ISM of the lensing galaxy. We also find an exponent of the radial mass distribution of $\\beta\\approx1.04$, in agreement with lens modelling of published 15-GHz VLBI data. Polarisation maps of each image are presented which show that the distributions of polarisation across images A and B are different. We suggest that this results from Faraday rotation and associated depolarisation in the lensing galaxy. ",
        "introduction": "Few gravitational lens systems are as well studied as JVAS B0218+357 \\citep{patnaik93}, perhaps the best example of a lens system for which the method of \\citet{refsdal64} can be used to determine the Hubble parameter, $H_0$. Unlike many other methods, this is done in a single step and involves well understood, and relatively simple, astrophysics. We point out that there remain significant uncertainties in $H_0$ as determined by traditional methods \\citep[e.g.][]{shanks01}. For example, the use of Cepheid-calibrated distances to determine the distance to galaxies hosting Type Ia supernovae has resulted in final values of $H_0$ that do not agree at the $1\\sigma$ ($\\sim$10~per~cent) level \\citep{parodi00,freedman01}. This is despite a large overlap between the samples of SNIa.  Whilst the gravitational lens route to $H_0$ can be a particularly ``clean'' one, and despite many lens systems now having measured time delays, almost all suffer in one way or another from effects that significantly increase the uncertainty in the final determination of $H_0$. In most cases the main source of uncertainty is in the modelling of the gravitational potential responsible for the image splitting and distortion. A lack of observational data to constrain the lens model and multiple-galaxy lenses are both significant factors in this regard \\citep[see e.g.][]{schechter00}. A list of the time delays measured to date can be found in \\citet*{courbin02}.  With JVAS B0218+357 on the other hand, the potential exists to reduce the uncertainty in the lens model to a level comparable to that in the time delay; this currently stands at 3~per~cent ($\\tau = 10.5\\pm0.4$~d; Biggs et al., 1999). Other authors find a value for the time delay consistent with this \\citep{cohen00}. This reduction in model uncertainty is possible due to the large number of observational model constraints available from multi-frequency, multi-resolution imaging. Both images (A and B) of the $z=0.96$ background quasar/BL~Lac are easily resolved with VLBI into two subcomponents \\citep*[e.g.][]{patnaik95,kemball01} which constrain the radial mass profile to be close to isothermal. The lens systems with the largest numbers of model constraints are those that contain Einstein rings, such as B0218+357. The resolved arcsec-scale structure of the ring probes the lens potential along multiple lines of sight and along all azimuthal position angles relative to the lens centre. The ring (see Fig.~\\ref{vlamap}) has been mapped at high resolution using combined MERLIN and VLA data at 5~GHz \\citep{biggs01} and the model constrained using a version of the LensClean algorithm \\citep{kochanek92}; preliminary results can be found in \\citet{wucknitz01}. Finally, the deflecting mass is concentrated in a single, isolated galaxy and so the effect of tidal shear due to nearby structure is very small ($\\sim$1~per~cent, Leh\\'{a}r et al. 2000). This results in a relatively uncomplicated mass model compared to other lens systems. A VLA map of B0218+357 can be seen in Fig.~\\ref{vlamap}.  The main stumbling block to date in measuring $H_0$ with JVAS B0218+357 has been that, due predominantly to the small size of the system (A-B separation of 334~mas), it has been difficult to measure the position of the $z=0.6847$ lensing galaxy relative to the lensed images accurately using optical ($HST$) data. For example, the positions derived from two NICMOS observations differ by 46~mas \\citep{lehar00} and this uncertainty translates into a large uncertainty in $H_0$. With LensClean we have estimated the galaxy position to a theoretical accuracy of a few mas, but this relies to some extent on the lens model and so could be biased.  \\begin{figure*} \\begin{center} \\includegraphics[scale=0.35]{fig1b.ps} \\includegraphics[scale=0.35]{fig1a.ps} \\caption{CJ1 VLBI maps of JVAS B0218+357 at 5~GHz. Left: B, right: A. The restoring beam is shown in the bottom-left corner of each map and has a FWHM of $5 \\times 4.2$~mas at a position angle of $-1$\\fdg9. Contours are plotted at multiples ($-1$, 1, 2, 4, 8, 16, etc) of 3$\\sigma$ where $\\sigma$ is the off-source rms noise in the map (330~$\\mu$Jy~beam$^{-1}$). Both maps are plotted on the same angular scale.} \\label{cj} \\end{center} \\end{figure*}  In this paper we present new 8.4-GHz observations of JVAS B0218+357 taken with a global VLBI array that were designed to increase the available model constraints by other means. With the great sensitivity of a global array we hoped we might detect: \\begin{enumerate} \\item extended structure associated with the radio jet in both A and B. This is detected in 5-GHz data from the First Caltech-Jodrell Bank (CJ1) VLBI survey \\citep{xu95} that we have re-analysed to produce maps (Fig.~\\ref{cj}) that are superior to those published in \\citeauthor{xu95}. With a resolution of $\\sim$5~mas these new maps reveal a jet that extends out to 30~mas from the core in each image. \\item a third image, as detected in, e.g. APM~08279+5255 \\citep{ibata99,lewis02} and MG~1131+0456 \\citep*{chen93,chen95}, and \\item the galaxy core. Emission from the lensing galaxy has been detected in several systems e.g. B0957+561 \\citep{harvanek97} and CLASS B2045+265 \\citep{fassnacht99}. A third radio component is detected in PMN~J1632-0033, but it is not known if this is a lensed image or the lens galaxy \\citep{winn02}. \\end{enumerate}  The observations have also been used to investigate the mas-scale polarisation structure of this system. Whilst the cores of most quasars/BL~Lacs are polarised at a level of about 2--3~per~cent \\citep{saikia88}, images A and B of B0218+357 are both much more highly polarised, reaching $\\sim$10~per~cent at frequencies 8.4~GHz and above \\citep{patnaik93,biggs99}. Also, the polarisation position angles of the images are not the same, in conflict with the expected behaviour of a lens system which preserves the polarisation position angle on the sky \\citep{dyer92}. It is thought that Faraday rotation in the magnetoionic medium of the lensing galaxy itself rotates the position angles of the images by different amounts \\citep{patnaik93}. Thus the extended nature of the emission in images A and B allows the interstellar medium (ISM) of a high-redshift galaxy to be studied on parsec scales. Both images also depolarise at low frequencies. ",
        "conclusions": "Our global VLBI 8.4-GHz maps have revealed a great deal of substructure in images A and B of JVAS B0218+357. Neither the lens galaxy core nor the third image were detected, both of which would be very useful constraints on the lens model, especially in this system where the centre of the lensing galaxy is currently uncertain.\\footnote{The lack of detection of a third image can be used though to place an upper limit on a finite core radius of the lens galaxy. This has been done for B0218+357 by \\citet{norbury01}.} In this section we discuss three main topics: the constraints that the new maps can already put on lens mass models, the evidence that the image suffers from multi-path scattering and the different polarisation distributions in each image.  \\subsection{Mass model constraints}  \\begin{figure*} \\begin{center} \\includegraphics[scale=0.45]{fig5b.ps} \\includegraphics[scale=0.45]{fig5a.ps} \\caption{Maps of images A (right) and B (left) after being back-projected into the source plane. The restoring beam is shown in the bottom-left corner and has a FWHM of $2.61 \\times 0.49$~mas at a position angle of $-13$\\fdg7.} \\label{backproj} \\end{center} \\end{figure*}  The extensive substructure that we have detected will be useful for lens modelling and ultimately it is our intention to use the LensClean algorithm on these data for this purpose, the results of which will be presented in a future paper. In the interim we have back-projected each image into the source plane in order to qualitatively demonstrate the accuracy of the model. This has been done using the {\\sc clean} components and an isothermal model of the lens galaxy optimized using the positions of the 15GHz VLBI sub-components \\citep{patnaik95}, a flux density ratio of 3.75 and the galaxy position found from our LensClean of VLA and MERLIN data \\citep{wucknitz01}. With $x$ and $y$ pointing along the major and minor axis of the ellipsoid, the potential is written as \\begin{equation} \\label{thefirstequation} \\phi= \\alpha_0 \\sqrt{\\frac{x^2}{(1+\\epsilon)^2} + \\frac{y^2}{(1-\\epsilon)^2}} \\quad . \\end{equation} The critical radius is $\\alpha_0=160.6\\,\\rmn{mas}$, the ellipticity $\\epsilon=0.0583$ and the position angle of the major axis $\\theta=-41.9\\,\\rmn{deg}$, measured from north through east.  The centre of the lens is at $\\bmath{z}_0 = (260,117.5)\\,\\rmn{mas}$ relative to A.  In order to compensate for the very different effective resolution of A compared to B, the back-projected CLEAN components of both images have been restored with the same beam. This beam is just large enough that both the A and B back-projected beams are encompassed by it, which in practice means that it is approximately equal to the restoring beam in the image plane back-projected from image B to the source plane. This makes the nominal resolution in each the same. The back-projected maps are shown in Fig.~\\ref{backproj}.  Compared to the lens-plane images of Fig.~\\ref{global1}, the source-plane images look quite similar. Worthy of note is the excellent correspondence between the counter-components to the west of the core region which are probably images of a weak counterjet. This is intriguing though as the lensed radio source in B0218+357 has been classified as a BL~Lac \\citep{odea92,browne93,stickel93}, a class of objects that are not expected to display counterjets according to the unification schemes of extragalactic radio sources. This theoretical prediction is supported by the non-detection of counterjets in BL~Lac sources, a notable exception being PKS~1413+135 \\citep{perlman94,perlman96}.  A closer comparison of the A and B source-plane maps shows that the B jet seems to be stretched by about 10~per~cent relative to the A jet i.e. the jet in B is longer. This stretching can only be explained with a different radial mass profile, compared to that used in the model, since the jet is directed more or less radially relative to the galaxy's centre. Writing the radial surface mass density as a function of radius as a power law, $\\Sigma(r) \\propto r^{-\\beta}$, we find $\\beta\\approx1.04$. A very similar value is found ($\\beta=1.06\\pm0.03$) if we use the 15-GHz sub-component positions of \\citet{patnaik95} as constraints \\citep{wucknitz01}.  \\subsection{Multi-path scattering} \\label{scattering} In Fig.~\\ref{backproj} image A looks smoother than image B. We suggest that this is due to scatter-broadening, which may also explain the frequency-dependent sizes of the images at low ($\\le$2.3~GHz) frequencies (Biggs et al., in preparation). Other examples of where scattering is believed to modify the surface brightness of gravitationally lensed images are PKS~1830-211 \\citep{jones96,guirado99} and CLASS B1933+503 \\citep{marlow99}. In B0218+357, as well as in the other two systems, the scattering would most likely originate in the lensing galaxy as we know from the observed Faraday rotation and differential depolarisation that this is both highly ionised and non-homogeneous. Also, at the lens redshift of $z=0.6847$ it is perfectly reasonable to expect that the ISM in front of each image, with a separation of 334~mas (2.4~kpc)\\footnote{Throughout this paper we assume a flat universe with $H_0=70$\\,km\\,s$^{-1}\\,$Mpc$^{-1}$, $\\Omega_0=0.3$ and $\\lambda_0 = 0.7$.} between them, could be significantly different.  In order to quantify the effect, we have fitted Gaussians to sub-components 1 and 2 of the back-projected images of A and B using the task {\\sc jmfit}. The deconvolved major ($a$) and minor ($b$) axes, as well as the position angle of the major axis ($\\phi$), are shown in Table~\\ref{jmfit} where it can be seen that for each sub-component the geometric mean of the axes (equivalent circular size) is bigger in A than in B; the area of the deconvolved A1 is approximately eight times that of B1. Also, whilst the position angles of B are as expected (they point more or less down the jet as in the image plane) those of A remain approximately perpendicular to this.  The advantage of model-fitting to the core sub-components in the source plane is that the back-projection removes the differences between the two images, reducing the need for removal of confusing pixels corresponding to the jet emission and rendering any bias that might exist identical in each. The disadvantage of fitting in the source plane is the potential for errors to accumulate during the many convolutions and deconvolutions that have been applied to the data at various stages of the data reduction. However, we note that the measured sizes in the image plane also show that image A is larger than image B, by a factor greater than the relative magnification. From the deconvolved sizes of Table~\\ref{modelfit} the image-plane area ratios A1/B1 and A2/B2 are approximately 14 and 9. For the remainder of this analysis we only consider the source-plane size measurements.  \\begin{table} \\begin{center} \\caption{Deconvolved major ($a$) and minor ($b$) axes and orientation of ellipticity ($\\phi$) of Gaussians fitted to the back-projected maps as well as their geometric means and the size of the derived scattering discs ($\\theta_{\\mathrm{sc}}$). All quantities are measured in mas apart from position angles which are measured in degrees. We estimate errors on the major and minor axes of about 5~per~cent.} \\begin{tabular}{lllrll} \\hline Image & $a$ & $b$ & $\\phi$ & $\\sqrt{a\\times b}$ & \\multirow{1}*{$\\theta_{\\mathrm{sc}}$} \\\\ \\hline A1 & 1.07 & 0.67 & 151 & 0.85 & \\multirow{2}*{0.79} \\\\ B1 & 0.34 & 0.28 & 57  & 0.31 & \\\\ \\hline A2 & 1.24 & 0.96 & 149 & 1.09 & \\multirow{2}*{0.83} \\\\ B2 & 0.78 & 0.64 & 76  & 0.71 & \\\\ \\end{tabular} \\label{jmfit} \\end{center} \\end{table}  We assume that the observed size of an A sub-component is the sum, in quadrature, of the intrinsic size and a scattering scale size ($\\theta_{\\mathrm{sc}}$). We further assume that, for each sub-component, the true intrinsic size is actually that of its image B counterpart i.e. that this image is unaffected by scattering at this frequency. This assumption is supported by the similarity of the deconvolved size of B1 as measured from these data and the 15-GHz data of \\citet{patnaik95}. The derived size of $\\theta_{\\mathrm{sc}}$ is also shown in Table~\\ref{jmfit} and is the same for both sub-components, 0.8~mas. The true size of the scattering disc (we have measured sub-component sizes in the source plane) is \\begin{equation} \\label{thesecondequation} \\hat{\\theta}_{\\mathrm{sc}}=\\theta_{\\mathrm{sc}} \\frac{D_{\\mathrm{s}}}{D_{\\mathrm{ds}}}, \\end{equation} where $D_{\\mathrm{s}}$ is the angular diameter distance between the Earth and the lensed source and $D_{\\mathrm{ds}}$ that between the lensing galaxy and the lensed source. The relation in equation~(\\ref{thesecondequation}) is analogous to that between the true deflection angle ($\\hat\\alpha$) and the {\\em apparent} deflection angle ($\\alpha$) \\citep[e.g.][]{refsdal94}. No correction for the lens magnification is required as the displacement in the image plane due to scattering is the scattering angle magnified by the lens, and this was removed when back-projecting each image. We calculate $\\hat{\\theta}_{\\mathrm{sc}} = 3.4$~mas.  The size of the scattering disk is equal to the refractive length scale ($r_{\\mathrm{ref}}$) subtended at the distance of the lensing galaxy \\citep{narayan92}. Using the formulation of \\citet{walker98} we have estimated the scattering strength (SM) of the ionised medium. With the observed frequency of 8.4~GHz corrected to the redshift of the lensing galaxy we measure $\\mathrm{SM \\approx 150~kpc~m^{-20/3}}$. This is a very large value compared to those seen along typical lines of sight through the Galaxy, although comparable (and in some cases much higher) values have been measured, particularly towards the Galactic centre \\citep{cordes02}. Therefore, although the scattering in front of B0218+357 image A is extreme, it is not ruled out by observations in the Galaxy.  \\subsection{Depolarisation}  The polarisation properties of this system continue to intrigue. From observations of this system with MERLIN and the VLA, a general picture has emerged of image A being less polarised than image B and of both depolarising with increasing wavelength, albeit much more steeply in A. The reason for this is probably inhomogeneities in the magneto-ionic medium that is responsible for the Faraday rotation of the polarisation position angles, inhomogeneities that are greater in the region of A than B. Complications that hinder the interpretation of the polarisation results include ``beating'' of the polarisation position angles of sub-components 1 and 2 (for observations which do not resolve the sub-components), changing source structure with frequency and different magnification gradients across each image.  The resolution afforded by VLBI observations allows the polarisation structure of the source to be resolved and constraints to be put on the angular scale of the inhomogeneities in the Faraday screen. In Fig.~\\ref{global2} we see that image B, the least resolved of the images, is seen to have a higher peak polarisation (20~per~cent) than that of A (12~per~cent). This shows that we are not seeing the effects of {\\em intrinsic} changes in the polarisation structures across the radio source as the lower effective resolution of B would cause these to be averaged incoherently more than in A i.e. we would expect image B to be less polarised than image A. Therefore, our global VLBI maps support the theory that the depolarisation is due to non-uniformities in the Faraday screen which are greater in front of image A. They also allow us to constrain the size of these irregularities which must be smaller than the synthesised beam i.e. $\\la$1~mas. This corresponds to a linear distance of $\\la$7~pc at the redshift of the lensing galaxy.  We also note that the scatter-broadening identified in Section~\\ref{scattering} could cause the depolarisation. This is because scatter-broadening effectively blurs a source, causing regions with different polarisations to overlap and be averaged together. As the scattering is different in front of each image the polarisations of each will be different and, as scattering increases at longer wavelengths, the observed frequency dependence of the depolarisation could also result.  Another interpretation is that the differences between the images could result from the combination of intrinsic polarisation variability and time delay. VLA monitoring \\citep{corbett96,biggs99} has shown that this source is variable in polarisation as well as in total flux density. Two-epoch VLBI observations separated by the time delay, $10.5\\pm0.4$~d, could help to clarify the polarisation structure in this source. At the same time the picture of A being more depolarised than B is too simplistic. From earlier VLA monitoring observations \\citep{corbett96} it was found that, with the time delay removed, the polarisation of image B was {\\em lower} than that of image A at 15~GHz, the opposite to what is usually observed. This could signify time variability within the magneto-ionic medium or movement of the source relative to this."
    },
    "0209/astro-ph0209307_arXiv.txt": {
        "abstract": "Sampling fluctuations in stellar populations give rise to dispersions in observables when a small number of sources contribute effectively to the observables. This is the case for a variety of linear functions of the spectral energy distribution (SED) in small stellar systems, such as galactic and extragalactic H{\\sc ii} regions, dwarf galaxies or stellar clusters. In this paper we show that sampling fluctuations also introduce systematic biases and multi-modality in non-linear functions of the SED, such as luminosity ratios, magnitudes and colours.  In some cases, the distribution functions of rational and logarithmic quantities are bimodal, hence complicating considerably the interpretation of these quantities in terms of age or evolutionary stages.  These biases can be only assessed by Monte Carlo simulations. We find that biases are usually negligible when the effective number of stars, $\\Neff$, which contribute to a given observable is larger than 10.  Bimodal distributions may appear when $\\Neff$ is between 10 and 0.1. Predictions from any model of stellar population synthesis become extremely unreliable for small $\\Neff$ values, providing an operational limit to the applicability of such models for the interpretation of integrated properties of stellar systems.  In terms of stellar masses, assuming a Salpeter Initial Mass Function in the range 0.08 -- 120 M$_\\odot$, $\\Neff$=10 corresponds to about 10$^5$ M$_\\odot$ (although the exact value depends on the age and the observable).  This bias may account, at least in part, for claimed variations in the properties of the stellar initial mass function in small systems, and arises from the discrete nature of small stellar populations. ",
        "introduction": "The comparison of observations to theoretical models is one of the basic steps which allows an understanding of natural phenomena. The comparison is not always satisfactory and leads to either new insights on the nature of the phenomena observed or to revisions of the theoretical framework within which the models are made, or both.  In general models have an {\\it intrinsic} uncertainty, in addition to possible systematic effects, and one seeks to minimise both in order to infer a robust interpretation of the observations. In the case of stellar population synthesis, the former uncertainties have often been neglected in comparison with the latter, where external errors (i.e. uncertainties on the input assumptions) have been much discussed, e.g. \\citet{Lei96} or \\citet{Bru01}.  However, the {\\it predicted} integrated properties of small stellar systems suffer from large intrinsic dispersions arising from the very nature of the systems: sampling fluctuations from, say, the initial mass function (IMF) will give rise to fluctuations in the properties and hence to an intrinsic dispersion in the predictions \\citep[e.g.][]{Chi88,SF97,LM99,CLC00,Cer01,Bru02Tuc,Cer02}. This effect has been overlooked by most synthesis codes, which usually take an analytical IMF and produce predictions which are only valid in the limit where the number of stars is formally infinite.  \\begin{figure} \\centerline{\\includegraphics[width=6cm,angle=270]{MC801f1.eps}} \\caption{Integrated $B-V$ colour vs. $V-K$ colour of 10 Myr old stellar clusters with 1000, 100, 10 and 1 star, respectively, with masses between 2 and 120 M$_\\odot$ distributed following a Salpeter IMF (triangles). Each triangle represents a Monte Carlo simulation with these parameters. The location of a stellar population with an infinite number of stars is indicated with the $\\odot$ symbol. The colours of the sparsely-populated clusters show a multimodal distribution in this diagram, and their average values are biased with respect to the colours of an infinite stellar population.} \\label{fig:GTC} \\end{figure}  In addition to this dispersion --which can be quantified and properly taken into account with the tools developed by \\citet{Cer02}-- there is also a more subtle effect arising from this incomplete sampling.  Integrated quantities which scale linearly with the number of stars (such as the luminosity in a given photometric band, or the overall spectral energy distribution, SED) can be predicted for any stellar system from the scaled-down outputs of synthesis codes with effectively predict properties for very large stellar systems, that is, systems which fully sample the distribution function of masses. However, for quantities which do {\\it not} scale {\\it linearly} with the number of stars, such as luminosity ratios, magnitudes, colours, equivalent widths, etc, such a scaling cannot be performed properly, and the only solution is to make extensive Monte Carlo simulations. We can already foresee that biases will arise from the incomplete sampling of the underlying distribution function. For instance, infrared colours may be dominated, at some ages, by a handful of high luminosity, IR-bright stars.  Monte Carlo simulations calculated by \\citet{SF97} ~show that the average colour is a function of the number of stars in the simulation.  Similarly, the average optical colours of stellar clusters, at a given age, are functions of the total mass in the cluster, as shown for instance by \\citet{GB93} and more recently by \\citet{G00} and \\citet{Bru02Tuc}.  This is best illustrated with an example in a simple colour-colour diagram. Figure \\ref{fig:GTC} shows the point (indicated by the $\\odot$ symbol) where a 10 Myr-old stellar population with an infinite number of stars would be. Each triangle in each panel represents one Monte Carlo simulation of a 10 Myr-old stellar population with a total number of stars (with masses between 2 and 120 M$_\\odot$ distributed with a Salpeter IMF) of 1000, 100, 10 and 1 star, respectively\\footnote{For reference, $\\Ntot$=1000 stars within 2 and 120 M$_\\odot$ have a weight of $\\Mtot \\approx 6 \\times 10^3$ M$_\\odot$, and would the IMF be extended down to 0.08 M$_\\odot$, the mass of the cluster would be $M_{\\rm tot}$=$2.25 \\times 10^4$ M$_\\odot$.}. Figure \\ref{fig:GTC} shows three important effects:  {\\it (a)} The distribution of colours in this diagram is clearly bimodal or even multimodal. The points do not cluster around a well-defined center, but are very scattered over the entire diagram, at odds with the (standard) predicted value for an infinitely-large stellar population.  {\\it (b)} If the average colour is computed for clusters with the same number of stars, its value is very different from the one given by the standard prediction : the average colour is biased, and the bias depends on the total number of stars present in the population.  {\\it (c)} The area covered by the points (each representing one Monte Carlo simulation) is not a monotonic function of the number of stars in the simulation.  Note that these effects are not limited of course to clusters of this age, but do also appear in more evolved populations, since these effects are a generic feature of under-sampled clusters\\footnote{See for example the analysis by \\cite{Bru02Tuc} where the positions of LMC clusters are interpreted in terms of Monte Carlo simulations of clusters of $10^5$, $10^4$ and $10^3$ M$_\\odot$ at different ages.}.  In this paper we provide a first order analytical estimation on {\\it when and why synthesis models which use an analytical formulation of the IMF are not be able to reproduce the observed properties of stellar systems} and therefore when Monte Carlo simulations are absolutely required in order to make predictions.  We show that bias and multimodality features are a {\\it generic} property of non-linear functions of the SED, and will always be present when trying to apply population synthesis models to small size stellar populations. Their importance cannot be stressed enough: many of the seemingly 'peculiar' properties of some systems can, at least in part, be accounted for by this bias and multimodality effects.  This is an important issue since odd values of these properties often lead to non standard interpretations. For instance, when colours or equivalent widths (EWs) of small stellar systems appear to be peculiar (that is, peculiar with respect to the values predicted by synthesis models of a large number of stars), they are often interpreted in terms of IMF variability : either the mass cutoffs have to be truncated, and/or the slope of the IMF has to be changed in order to produce the 'correct' colours or EWs.  Since the lower mass cutoff determines the overall mass-to-light ratio, it is usually the upper mass limit which is varied to account for the data. Examples of these interpretations abound in the recent literature, starting perhaps with \\citet{ST76} who, on the basis of the variation of the equivalent width of H$\\beta$ in H{\\sc ii} regions across M 101, inferred a dependence of the upper mass cutoff with metallicity.  Line ratios included in classical diagnostic diagrams have also been used (e.g.  \\citealt{O97}, and \\citealt{BKG99} for the softness parameter) and seem to show that the IMF changes in extragalactic H{\\sc ii} regions \\cite[but see however][ for an alternative explanation based on sampling effects]{BK02}.  Colours, and colour-colour diagrams, sometimes including the H$\\alpha$ line, of clusters and starbursts also seem to point to variations in the parameters of the IMF, from the LMC \\citep{DH98}, to clusters in the starburst galaxy IC10 \\citep{H01}.  The common features of all these interpretations which point to a variation of the IMF are (i) the systematic use of non-linear quantities (EWs, line ratios, colours) and (ii) observations of small stellar systems (H{\\sc ii} regions, stellar clusters, starbursts, etc).  Given the scaling arguments explained above, it is interesting to see whether sampling effects could account for this type of observations.  In this paper we show that whilst the {\\it average} values of the properties (and their {\\it dispersions}) that scale with the number of stars are reasonably well reproduced by the models, non-linear functions such as luminosity ratios or colours may be heavily biased. It is beyond the scope of this paper to analyse in detail each and every observation, but given our results, it appears very likely that the 'peculiar' properties observed in some systems can be simply understood in terms of the bias and multimodality introduced by sampling fluctuations of the IMF in these stellar systems.  This has profound implications for the interpretation of colours, average magnitudes and line ratios in such systems, as well as for the universality (or otherwise) of the IMF.  These effects seem to be large enough that most observations may point to the universality of the IMF, despite a variety of physical conditions (e.g. \\citealt{K02} for a general review and \\citealt{MHK99} for the case of starburst galaxies), and therefore provide a unique insight on star formation processes in a variety of environments.  The structure of the paper is the following: in Sec. \\S2 we recall the scaling properties of synthesis models.  In Sec. \\S3 we analyse the statical properties of functions of the SED, showing the biases and multimodalities that appear in their distribution functions.  We discuss these results and their implications in Sec. \\S4 and \\S5. ",
        "conclusions": "In this work we have found important effects which have been overlooked in the application of population synthesis codes to small stellar systems such as clusters, starbursts, dwarf galaxies or pixels.  Most current synthesis codes predict the average values of observables for an infinitely large population of stars, which samples perfectly a given IMF. Extrapolating these predictions to small stellar systems may be misleading.  First, scaling observables with the mass in stars is not correct for small systems.  The proper scaling should be performed with the total number of stars present in the system.  Second, the distribution functions of observables which are non-linear functions of the monochromatic luminosity (or SED) may show multimodality. This implies that for a given population, of fixed size, and all other parameters kept the same, there will be a large dispersion in the observable, whose average value will not necessarily be the same as the one predicted when the IMF is fully sampled. Most synthesis codes will therefore lead to biased predictions when applied to small stellar systems.  Third, we have also shown that bimodality is a natural effect that appears when stellar systems are affected by sampling (i.e. the number of stars in the clusters is not enough to sample completely the IMF).  It is related to the evolutionary speed of different masses, or in more graphical terms, to the absence/presence of stars in some evolutionary phases.  This effect was already shown 14 years ago by \\cite{Chi88}.  We have now established the range of $\\cal{N}$ values (from $\\approx$ 5 to 1) where bimodality appears. An additional result is that bimodality will be more easily observed when infrared colours are used since such colours are more affected by the presence of short-lived post-MS phases.  Fourth, while we cannot confirm or reject the hypothesis of possible variations in the properties of the IMF in small stellar systems, these results suggest that an important fraction of the dispersion observed in non-linear observables is not necessarily evidence for IMF variations, but can rather be accounted for by sampling fluctuations.   Finally, we have also derived an operational limit to the validity of standard synthesis codes. There is a critical effective number of stars (unrelated to an actual number of stars), with a value of around 10, which defines a boundary where multimodality may appear, i.e.  cluster masses larger than 10$^5$ M$_\\odot$ for a Salpeter IMF in the range 0.08 -- 120 M$_\\odot$ (but the exact value for the cluster masses depends on the age and the observable).  Below this critical value, the results of the codes must be taken with caution, since not only they can be biased, but also they may underestimate the actual dispersion of the observable, and Monte Carlo simulations are required. The reason for the existence of this critical value is directly related to the presence (or otherwise) of stellar evolutionary stages which are not fully sampled in small systems.  We encourage the inclusion of the calculation of this effective number of sources for a given observable in all synthesis codes, so that observers (and theoreticians alike) can estimate the expected dispersion (and possible biases) in observables."
    },
    "0209/astro-ph0209594_arXiv.txt": {
        "abstract": "We present atmospheric parameters and Fe abundances derived for the majority of dwarf stars (north of -30 degrees declination) which are up to now known to host extrasolar planets. High-resolution spectra have been obtained with the Sandiford Echelle spectrograph on the 2.1m telescope at the University of Texas McDonald Observatory. We have used the same model atmospheres, atomic data and equivalent width modeling program for the analysis of all stars. Abundances have been derived differentially to the Sun, using a solar spectrum obtained with Callisto as the reflector with the same instrumentation. A similar analysis has been performed for a sample of stars for which radial velocity data exclude the presence of a close-in giant planetary companion. The results are compared to the recent studies found in the literature. ",
        "introduction": "To examine the relation between the metallicity of the stellar host and the existence of close-in massive planets, we have analyzed high-resolution, high signal-to-noise spectra of the dwarf stars with super-Jupiter planets listed in the ``Extrasolar Planets Encyclopaedia''\\footnote{http://www.obspm.fr/encycl/encycl.html} (44 stars at the time when this poster was prepared, hereafter CGP dwarfs). Note that this list does not include stars with companions with $M_{\\rm p} \\sin i > 12 M_{\\rm J}$. A list of comparison stars was compiled using the results of the Lick planet search (Cumming et al. 1999). We analyzed all stars for which radial velocity data exclude the presence of a close-in giant planetary companion (23 stars, hereafter no-CGP dwarfs). The selection process is illustrated in Figure~\\ref{limits}, which shows the upper limits for planetary masses (in Jupiter-masses) as a function of orbital radii (in AU) for this sample, as derived from radial-velocity data, and masses and semi-major axes for known extrasolar planets. Those of the latter which would fall below the lower limit for the comparison sample have been excluded.  \\begin{figure} \\plotfiddle{heiteru1_01.ps}{11cm}{270}{60}{60}{-220}{350} \\caption{Red dots: Masses and semi-major axes of currently known extrasolar planets. Blue lines: Upper (mass) and lower (orbital radius) limits for stars from the Lick survey (Cumming et al.1999).} \\label{limits} \\end{figure} ",
        "conclusions": "At this point of time we cannot draw any hard conclusions, because \\begin{itemize} \\item the sample sizes are limited; \\item the analysis is still incomplete; \\item the samples could be affected by selection bias. \\end{itemize}  \\noindent However, Figure~\\ref{results} indicates that \\begin{itemize} \\item in contrast to the no-CGP sample, there seems to be a lack of cool stars ($T_{\\rm eff} \\le$ 5400 K) in the CGP sample. Does this mean a bias in the sample selection? \\item there is a ``clump'' of metal-rich CGP stars at solar temperature. Is this a lack of metal poor giant planet hosts or a bias -- either from selection or from a lack of metal poor solar type stars (with or without planets)? \\item there seems to be {\\em no} correlation between metallicity and planet hosting at hotter temperatures ($T_{\\rm eff} \\ge$ 6000 K). \\end{itemize}  \\noindent Planned future work includes: \\begin{itemize} \\item Abundance analysis of a second comparison sample: Stars which have spectral types like those of the dwarfs with (CG) planets selected randomly from the Bright Star Catalog (and Supplement); \\item Abundance analysis of ``Very Strong Lined'' dwarfs (Eggen 1978); \\item Abundance determination for elements other than Fe, for all stars of the four samples. \\end{itemize}"
    },
    "0209/astro-ph0209077_arXiv.txt": {
        "abstract": "SBS 1150+599A is a blue stellar object at high galactic latitude discovered in the Second Byurakan Survey. New high-resolution images of SBS 1150+599A are presented, demonstrating that it is very likely to be an old planetary nebula in the galactic halo, as suggested by Tovmassian et al (2001). An H$\\alpha$ image taken with the WIYN 3.5-m telescope and its ``tip/tilt\" module reveals the diameter of the nebula to be $9\\farcs2$, comparable to that estimated from spectra by Tovmassian \\etal Lower limits to the central star temperature were derived using the Zanstra hydrogen and helium methods to determine that the star's effective temperature must be $>$68,000K and that the nebula is optically thin. New spectra from the MMT and FLWO telescopes are presented, revealing the presence of strong [\\ion{Ne}{5}] $\\lambda3425$, indicating that the central star temperature must be $>$100,000K. With the revised diameter, new central star temperature, and an improved central star luminosity, we can constrain photoionization models for the nebula significantly better than before. Because the emission-line data set is sparse, the models are still not conclusive. Nevertheless, we confirm that this nebula is an extremely metal-poor planetary nebula, having a value for O/H that is less than 1/100 solar, and possibly as low as 1/500 solar. ",
        "introduction": "Planetary Nebulae (PNe) in our galactic halo provide a unique insight into the mechanisms of stellar evolution at low metallicities and old ages. Since many chemical abundances of the progenitor star are not changed during stellar evolution, halo PNe may probe the metallicity of the Galactic halo at the formation epoch of their progenitors (Torres-Peimbert \\& Peimbert 1979).  Because of their rarity, however, only a few halo or Type IV PNe have been discovered (e.g., Howard, Henry, \\& McCartney 1997), where in contrast, there are over a thousand disk PNe known (e.g., Acker et al. 1992).  In this paper, we report our study of a likely new halo PN: SBS 1150+599A (PN G135.9+55.9).  Garnavich \\& Stanek (1999) obtained a spectrum of SBS 1150+599A (11:53:24.73 +59:39:57.0 J2000) along with 2 other cataclysmic variable (CV) candidates identified in the stellar object list of the Second Byurakan Sky Survey (SBS; Balayan 1997). Of these 3 stars, SBS 1150+599A was uncharacteristic, having narrow emission lines from the Balmer series as well as \\ion{He}{2} $\\lambda4686$ and [\\ion{Ne}{5}] $\\lambda3425$. Because the object was small and unresolved, Garnavich \\& Stanek suspected it to be a young PN or symbiotic star. Subsequently, Tovmassian et al. (2001; TSCZGP) obtained a series of spectra of this object to evaluate its nature. They confirmed some of the emission characteristics briefly noted by Garnavich \\& Stanek (1999) and they attempted to derive a number of astrophysical quantities by modeling the photoionization properties of the nebula. Remarkably, TSCZGP found that the nebula had to be extremely oxygen-poor, despite the relatively few constraints on their models. In fact, the nebula would be the most oxygen-poor PN known by about a factor of 10, implying that it derived from one of the most oxygen-poor stars.  Because this object is so extreme, we wished to verify that it was, in fact, a genuine PN. If it was another class of object (\\eg nova shell, galaxy, cataclysmic variable), then the TSCZGP abundance models would be suspect.  If the object was found to be a PN, then we hoped to improve the abundance models by adding additional observational constraints. In Section 2, we describe the high spatial resolution images we obtained with the VATT and WIYN telescopes that confirm the PN hypothesis, as well as additional spectra obtained at the MMT. In Section 3, we derive the properties of the PN. In Sections 4 and 5, we discuss our photoionization model and abundance analysis results. ",
        "conclusions": "We have developed a fairly consistent picture in which SBS 1150+599A is an old PN derived from a progenitor in the Galactic halo. The central star must be $\\sim100,000$K and the nebula must be very oxygen-poor.  The picture, though, is not completely satisfying because the observational data are so limited. Future studies should include UV observations; that region of the spectrum is likely to pay the most dividends in constraining the abundances for carbon and nitrogen. These elements control the cooling rate of the nebula, and thus, the abundance of oxygen cannot be measured accurately until that rate is determined. Alternatively, one can attempt to measure the electron temperature directly from the line ratio of [\\ion{O}{3}] $\\lambda4363$ to $\\lambda5007$, but doing so would be a difficult observation due to the low fluxes of these lines.  Nevertheless, the evidence is very strong that SBS 1150+599A is extremely underabundant in oxygen, and possibly is the PN with the lowest oxygen abundance. There is some chance that oxygen has somehow been transformed to neon within the progenitor star, thereby explaining the very high Ne/O and N/O ratios.  That scenario has been proposed by Howard et al (1997) for the anomalous halo PN BoBn~1. If so, then the composition of this PN no longer represents the original composition of its progenitor star, and the abundances measured in  SBS 1150+599A become more valuable for studying stellar physics than for studying Galactic chemical history.  More observational study of this unique object is clearly warranted."
    },
    "0209/astro-ph0209255_arXiv.txt": {
        "abstract": "Five years of \\it Rossi X-ray Timing Explorer (RXTE) \\rm observations of the Galactic black-hole candidates \\onee\\ and \\grs\\ show a periodic modulation with amplitude 3-4\\% in each source at $12.73 \\pm 0.05$~dy and $18.45 \\pm 0.10$~dy, respectively.  We interpret the modulations as orbital, suggesting that the objects have red-giant companions. Combining the \\it RXTE \\rm data with earlier data \\citep{Zh97} from the Burst and Transient Source Experiment on the \\it Compton Gamma-Ray Observatory\\rm, we find a long period or quasi-period of about 600~dy in \\onee, and a suggestion of a similar 600-dy period in \\grs.  These timescales are longer than any yet found for either precessing systems like \\her\\ and \\ssfour\\ or binaries like \\lmcx\\ and \\cyg\\ with more irregular long periods. ",
        "introduction": "The Galactic-bulge x-ray sources \\onee\\ and \\grs\\ are generally called black-hole candidates due to the similarity of their x-ray spectral and timing behavior to that of \\cyg\\ in its usual hard state.  Like \\cyg, both sources occasionally enter an intermediate or soft state, but the evolution of their spectral hardness and luminosity is very different \\citep{Sm02}.  Both have prominent, bright radio lobes about an arcminute in size \\citep{Mi92,Ro92}, while \\cyg\\ does not \\citep{Ma96}, showing only a milliarcsecond jet near its core \\citep{St01}.  The counterparts of \\onee\\ and \\grs\\ are unknown due to high extinction, therefore there are no orbital solutions or estimated masses.  For \\grs, \\citet{He02} recently used \\it Chandra \\rm data to confirm the association of the x-ray source with the radio core source (\"VLA-C\") and \\citet{Cu01} did the same for \\onee. Marti et al. (1998) identified two candidate counterparts to VLA-C/\\grs\\ in \\it I- \\rm and \\it K-\\rm band images. The brighter and closer candidate was found through multi-band photometry and near-infrared spectroscopy to be a likely K0 III giant.  Revised astrometry \\citep{Ro02} of infrared observations by \\citet{Ei01a} confirm that this star (``star A'') is consistent with VLA-C at the 3$\\sigma$ level. \\citet{Ma00} and \\citet{Ei01a} agree on several possible high-mass candidates for the companion of \\onee\\ in its more crowded and obscured field, but a low-mass companion would be unobservable at the current \\it K\\rm-band sensitivity. ",
        "conclusions": ""
    },
    "0209/astro-ph0209063_arXiv.txt": {
        "abstract": "Using the cross-spectral method, we confirm the existence of the X-ray hard lags discovered with cross-correlation function technique during a large flare of \\mkn\\ observed with \\sax . For the 0.1--2 versus 2--10~keV light curves, both methods suggest sub-hour hard lags. In the time domain, the degree of hard lag, i.e., the amplitude of the 3.2--10~keV photons lagging the lower energy ones, tends to increase with the decreasing energy. In the Fourier frequency domain, by investigating the cross-spectra of the 0.1--2/2--10~keV and the 2--3.2/3.2--10~keV pairs of light curves, the flare also shows hard lags at the lowest frequencies. However, with the present data, it is impossible to constrain the dependence of the lags on frequencies even though the detailed simulations demonstrate that the hard lags at the lowest frequencies probed by the flare are not an artifact of sparse sampling, Poisson and red noise. As a possible interpretation, the implication of the hard lags is discussed in the context of the interplay between the (diffusive) acceleration and \\sy\\ cooling of relativistic electrons responsible for the observed X-ray emission. The energy-dependent hard lags are in agreement with the expectation of an energy-dependent acceleration timescale. The inferred magnetic field ($B \\sim 0.11$~Gauss) is consistent with the value inferred from the Spectral Energy Distributions of the source. Future investigations with higher quality data that whether or not the time lags are energy-/frequency-dependent will provide a new constraint on the current models of the TeV blazars. ",
        "introduction": "\\label{sec:intro}  It has been well established that blazars are extra-galactic sources possessing relativistic jets aligned close to the line of sight, nevertheless, it is still poorly understood how the jets are powered, formed and collimated, and how particles are efficiently accelerated. One of the best observational approaches would be to use temporal and spectral analysis of the emission from the jets. Although blazars are not the unique hosts of jets, being dominated by Doppler effects (causing the observed emission to be enhanced and the \\tss\\ shortened), they are the ideal targets for studying jet physics.  The dominant radiation mechanisms in blazars are thought to be \\sy\\ and inverse Compton by relativistic electrons in a tangled magnetic field, which can reproduce the two peaks of the Spectral Energy Distributions (SEDs) in the $\\nu F_{\\nu} - \\nu$ diagram: synchrotron radiation is responsible for the low energy peak, while inverse-Compton upscattering by the same population of electrons produces the high energy one (e.g., Urry \\& Padovani 1995). In such a picture, the blazar family could be unified on the basis of the SEDs whose properties are determined by the bolometric luminosities (Ghisellini \\etal 1998).  According to this scenario, the variability of blazars is expected to be energy-dependent, with the highest energy part of each emission component showing the most rapid variations as produced by the highest energy part of the relativistic electron distribution which evolves most rapidly. For high-energy (usually UV/soft X-rays) \\sy\\ peaked blazars (HBLs), X-rays provide an ideal radiative window for studying the variations because (1) rapid variability indicates that the X-rays arise from the innermost region of the jets, and give direct clues on the central source; (2) \\sy\\ X-ray emission probes the electrons accelerated to the highest energies, which plausibly have the longest acceleration and the shortest cooling times.  The detected TeV blazars at TeV energies are typical HBLs, including three well-studied classical \\bls , Mrk~421, Mrk~501, and PKS~2155--304. The former two have been detected as bright and variable TeV emitters. In particular, \\mkn\\ ($z=0.031$) is the brightest blazar at UV, X-ray and TeV wavelengths. These sources have thus received particular attention as ideal targets for detailed temporal and spectral variability studies in the broadest spectral ranges. Extensive multi-wavelength monitoring campaigns and long looks with various satellites have been conducted.  Intensive monitoring has shown that the synchrotron peak energy up-shifts with flux in these sources (Pian \\etal 1998; Fossati \\etal 2000b; Tavecchio \\etal 2001; Zhang \\etal 2002). They have also exhibited quite different variability properties and spectral evolution from flare to flare, indicating that the high energy photons can lead or lag the low energy ones (e.g., Zhang \\etal 1999, 2002; Fossati \\etal 2000a; Tanihata \\etal 2001). The so-called soft lag (lower energy X-ray photons lagging higher energy ones) is consistent with the picture of the energy-dependent cooling time of relativistic electrons---higher energy electrons cool faster. The opposite behavior, i.e., the so-called hard lag (higher energy X-ray photons lagging the lower energy ones) has been found (as not rare) with recent long looks of the three TeV blazars with \\sax\\ and ASCA. This ``unusual'' hard lag has been thought to give direct information on electrons acceleration: it takes longer time for higher energy electrons to accelerate to the radiative energy.  In this paper, through the studies of the time lags in the time and frequency domains, the cross-correlation function (CCF) and the cross-spectral methods are used to re-examine the discovery of the X-ray hard lags during a large flare of \\mkn\\ detected by \\sax\\ (Fossati \\etal 2000a). The latter method was historically used to analyze the X-ray variability of Galactic black hole candidates (GBHCs; e.g., Miyamoto \\etal 1988, 1991; Nowak \\etal 1999). Recently, Papadakis, Nandra \\& Kazanas (2001) adopted this technique to study the X-ray variability of a Seyfert galaxy. Since the cross-spectrum can give more information (i.e., the Fourier frequency-dependent time lags) than the CCF can do, it is able to impose stronger constraints on the emission models.  The paper is organized as follows. The \\lcs\\ of the flare are presented in \\S\\ref{sec:lc}. The characteristic feature of the hard lag is examined by showing the evolutionary behavior of the hardness ratio versus the count rate. In \\S\\ref{sec:lage} the dependence of hard lags on photon energies is studied with the CCF method incorporating with a model-independent Monte Carlo simulations. We investigate in \\S\\ref{sec:flag:results} the time lags in Fourier frequency domain using the cross-spectral technique; detailed simulations are performed in \\S\\ref{sec:flag:simulations} to investigate the effects of Poisson and red noise, sampling and signal-noise (S/N) ratio of the data sets. In \\S\\ref{sec:disc:summary} we discuss and compare the results derived in time and frequency domains; the physical implications of the results are preliminarily explored in \\S\\ref{sec:disc:implications}; we also briefly compare in \\S\\ref{sec:disc:comparison} the time lags in different black hole accreting systems. Finally, we present our conclusions in \\S\\ref{sec:conc}. ",
        "conclusions": "\\label{sec:conc}  X-ray observations have been a powerful diagnostic for the physical processes taking place in the vicinity of the central engines of blazars. \\sax\\ observed a well-defined flare of \\mkn\\ on 21 April 1998, from which an X-ray hard lag was discovered in a blazar. We investigated with different methods the hard lag character of the flare, and discussed the relevant physical implications.  We examined the conditions that may produce the hard lag variability pattern. For the first time, the energy-dependent acceleration time is assumed to be account for the energy-dependent hard lags. $\\xi \\sim 3.06\\times 10^4$ is suggested if the maximum characteristic energy of synchrotron emission were $\\sim$ 10~keV, where \\tacc\\ $\\sim$ \\tcool . By fitting the energy-dependent hard lags using the energy-dependent acceleration timescale of relativistic electrons, we found $B \\sim 0.11$ Gauss, which is self-consistent with the values inferred from the SEDs. We then deduced that the hard lags are caused by \\tacc\\ $\\sim$ \\tcool , and \\tacc\\ is energy-dependent.  Finally, we emphasize that our interpretation is based on the simplest scenario. We consider only \\tacc\\ and \\tcool\\ without any other complexities involved, in particular the simplified interpretation did not take into account the Fourier frequency-dependence of hard lags (if observably confirmed). However, to understand the relationship between electron acceleration/cooling and time variability, such a study should be qualitatively meaningful, but more accurate analysis needs detailed numerical simulations involving energy-dependent electron acceleration timescale and information of the frequency-dependence of time lags. More importantly, even though there is no doubt about the sign of hard lag, the results presented in this work are still suggestive, and need further confirmation with higher quality data. Further investigations that whether or not the time lags are energy- and Fourier frequency-dependent will put important implications for the blazar models."
    },
    "0209/astro-ph0209580_arXiv.txt": {
        "abstract": "{We present detailed photometric investigations of the recurrent nova CI Aql. New data obtained after the 2000 outburst are used to derive a 3D geometrical model of the system. The resulting light curves clearly indicate the existence of an asymmetric spray around the accretion disk, as claimed in the past e.g. for the super-soft X-ray source CAL87 in the LMC. The simulated light curves give us the mass transfer rates varying from $\\dot M \\approx 2.5 \\times 10^{-8}\\,\\,{\\rm M}_\\odot\\,{\\rm yr}^{-1}$ in 1991-1996 to \\mbox{$5.5 \\times 10^{-8} < \\dot M < 1.5 \\times 10^{-7}\\,\\,{\\rm M}_\\odot\\,{\\rm yr}^{-1}$} in 2001/2002. The distance and the interstellar foreground extinction resulting from the model are 1.55~kpc and \\mbox{E$_{\\rm B-V}$ = 0\\fm98} respectively. During fast photometry sequences in 2002 short timescale variations ($t_{\\rm F} \\approx 13$ minutes) of the mass loss are found. Moreover a change in the orbital period of the system is detectable and results in a mass loss of \\mbox{$2.2 \\times 10^{-6} < \\Delta M < 5.7 \\times 10^{-6}\\,\\,{\\rm M}_\\odot$} during the nova explosion. ",
        "introduction": "CI Aql is one of the 9 to 10 members of the class of recurrent novae: U~Sco, V394~CrA, RS~Oph, T~CrB, V745~Sco, V3890~Sgr, T~Pyx, CI~Aql and IM~Nor (Webbink et al. \\cite{We87}; Sekiguchi \\cite{Sec}; Schmeja et al. \\cite{schmeja}; Liller \\cite{iauc7789}). Webbink et al. (\\cite{We87}) also mention V1017~Sgr as possible class members although its status is still not clear. The first known outburst of CI Aql was discovered on Heidelberg plates recorded in June 1917 (Reinmuth \\cite{ci_1}). Williams (\\cite{ci_2}) completed the light curve by using records on Harvard College Observatory patrol plates. Schaefer (\\cite{ci_3}) found another outburst in 1941 again on Harvard plates. Schaefer argues, it might be a recurrent nova with a timescale of 20 years and the 1960 and 1980 outburst were missed. As the timescales of other recurrent novae often change and as there are no observations available, we assume for our calculations a quiescence phase of 60 years before the 2000 event. Anyway this does not affect the results of the model presented here but our resulting pre--outburst accretion rate indicates a long recurrence timescale. CI Aql was found to be an eclipsing binary system with a period of 0\\fd618355(9) by Mennickent \\& Honeycutt (\\cite{M_H}). It is, to our knowledge, the only eclipsing system investigated photometrically in such detail before and after an outburst. Following the classification of Sekiguchi (\\cite{Sec}), CI\\,\\,Aql is of U\\,\\,Sco subclass (slightly evolved main sequence star and accreting white dwarf). For a detailed discussion of the outburst data we refer to Matsumoto et al. (\\cite{Mat01}) and Kiss et al. (\\cite{kiss}).  \\noindent In this paper we deduce a detailed model of the system basing on our optical photometry of 2001 and 2002. Further we follow the final decline to quiescence and find quasi periodic short timescale variations in this system. Together with pre--outburst data of Honeycutt (\\cite{honey01}) we finally determine a period change. ",
        "conclusions": "The models presented allow us to determine the physical parameters of the system within a very small range. This is mainly achieved by the combination of data from different bands and different epochs what is the main advantage to the work of Hachisu \\& Kato (\\cite{HaKa}, \\cite{HaKaII}). The model of the recurrent nova two years after the outburst, when it reached a mean quiescent photometric state still differs significantly from the one of the pre--outburst phase. Even speculative, one may assume that the energy transferred during the outburst, when the secondary was completely enclosed by the WD shell, caused a significant extension over the Roche lobe. This increased the mass transfer to \\mbox{$5.5 \\times 10^{-8} < \\dot M < 1.5 \\times 10^{-7}\\,\\,{\\rm M}_\\odot\\,{\\rm yr}^{-1}$} in 2001/2002. Thus if the mass transfer rate derived from the pre--outburst phase is used as average, it does not reach up to the accreted mass during the 60 years of quiescence. The mass transfer rate of $\\dot M \\approx 2.5 \\times 10^{-8}\\,\\,{\\rm M}_\\odot\\,{\\rm yr}^{-1}$ derived for 1991-1996 differs from that given by Hachisu \\& Kato by a factor of four due to the different model of the accreting region. The period shift determined in this paper gives us a good estimate of the expelled mass $2.2 \\times 10^{-6} < \\Delta M < 5.7 \\times 10^{-6}\\,\\,{\\rm M}_\\odot$. Within the errors (the uncertainty due to the angular momentum carried by the ejected material and the uncertainty of the average mass transfer throughout the 60 years of quiescence) the determined expelled mass corresponds to the accretion. Thus the evolution towards the critical mass of the WD runs rather slow at this evolutionary stage. Assuming a net mass increase of $< 10^{-6}$\\,M$_\\odot$ per outburst leads to a few 10$^7$ years to reach the critical mass. On the other hand the evolutionary tracks of the secondary show it at a position with extremely rapid increase of its diameter (about \\mbox{5\\,\\,10$^{-10}$\\,R$_\\odot$~yr$^{-1}$}). Moreover the decrease of the orbit is about 10$^{-7}$\\,R$_\\odot$ per outburst. Thus the mass transfer should increase and may even evolve towards steady hydrogen burning (van den Heuvel et al. \\cite{vH}). The derived distance and interstellar extinction gives us a somewhat higher luminosity than in the outburst models of Hachisu \\& Kato (\\cite{HaKa}). This affects also the considerations in Oka et al. (\\cite{oka}). Because of our  smaller inclination we also see some parts of the disk during the primary minimum and thus it is not necessary to increase the temperature of the SE in 2001 to obtain a higher flux."
    },
    "0209/astro-ph0209313_arXiv.txt": {
        "abstract": "We present the detection of a $2.0(+1.4,-0.8)\\times10^4~\\Msun$ black hole (BH) in the stellar cluster G1 (Mayall II), based on data taken with the Space Telescope Imaging Spectrograph onboard the {\\it Hubble Space Telescope}. G1 is one of the most massive stellar clusters in M31. The central velocity dispersion (25~\\kms) and the measured BH mass of G1 places it on a linear extrapolation of the correlation between BH mass and bulge velocity dispersion established for nearby galaxies. The detection of a BH in this low-mass stellar system suggests that (1) the most likely candidates for seed massive BHs come from stellar clusters, (2) there is a direct link between massive stellar clusters and normal galaxies, and (3) the formation process of both bulges and massive clusters is similar due to their concordance in the $M_{\\bullet}-\\sigma$ relation. Globular clusters in our Galaxy should be searched for central BHs. ",
        "introduction": "The questions of how the nuclei of galaxies form and why they contain massive black holes (BHs) remain unsolved. However, the recent discovery of a tight correlation between central black hole (BH) mass and bulge velocity dispersion (hereafter the $M_{\\bullet}-\\sigma$ relation; Gebhardt et al. 2000b; Ferrarese \\& Merritt 2000) does shed some light on the evolutionary history of massive BHs and their host galaxies.  Many theories (e.g., Silk \\& Rees 1998; Haehnelt \\& Kauffmann 2000; Ostriker 2000; Adams, Graff, \\& Richstone 2001) predict such a correlation, and the exact details (i.e., slope and normalization) can discriminate among the various models. Presently, however, the data are inadequate to do this.  One difficulty is that the galaxies studied so far have limited coverage in parameter space. There are not enough observations at the low-dispersion end, and yet this region provides the tightest constraints on determining both the slope and offset. The main reason for this lack of low-dispersion systems is that there are few that the {\\it Hubble Space Telescope (HST)}\\ can reasonably observe.  Due to their high central densities and proximity, globular clusters provide an alternative to studying galaxies to explore the low mass end. Furthermore, there is evidence that at least some globular clusters may be nuclei of accreted galaxies (Freeman 1993, Ferguson~\\etal\\ 2002), and thus may contain central black holes if all galaxies contain them (Magorrian~\\etal\\ 1998). The clusters G1 (Mayall II) in M31 and M15 are excellent objects for these studies. They both have high central densities and short central relaxation times ($10^7$ years for M15 and $10^8$ for G1).  Gebhardt~\\etal\\ (2000c), van~der~Marel~\\etal\\ (2002) and Gerssen~\\etal\\ (2002) present results for M15 showing that it likely contains a central BH of a few thousand solar masses. If large (nonstellar) BHs exist in these stellar clusters, which have low escape velocities and apparently lack dark halos, they will pose a severe challenge to nearly every theory for the formation of massive BHs.  The cluster G1 lies 40 kpc from the nucleus of M31, projected approximately on its major axis.  It is the most luminous stellar cluster in the Local Group and has a higher central surface brightness than any Galactic globular cluster. Djorgovski et al. (1997) report a velocity dispersion of 25 km s$^{-1}$ from ground-based spectroscopy, and Meylan~\\etal\\ (2001) derive a total mass of $(7-17) \\times 10^6M_\\odot$, with the uncertainty due to their lacking a velocity dispersion profile. To place G1 among Galactic globular clusters, we note that the compilation of Trager, Djorgovski, \\& King (1995) gives no cluster with central surface brightness $<14.5$ $V$ mag arcsec$^{-2}$, fully one magnitude fainter than G1 (Rich et al. 1996). NGC 5139 ($\\omega$~Cen) is about as massive and luminous as G1, but its central surface brightness is 16.8 $\\rm mag\\ arcsec^{-2}$.  The highest measured velocity dispersion for any Galactic globular cluster is 18 km s$^{-1}$, for NGC 6441 and NGC 6388 (Pryor \\& Meylan 1993). It is interesting to note that both G1 (Meylan~\\etal\\ 2001; Ferguson~\\etal\\ 2002) and $\\omega$~Cen (Freeman 1993) are possibly nuclei of accreted galaxies.  The high luminosity and remarkable central surface brightness of G1 led us to propose to obtain STIS spectroscopy of its nucleus (GO-9099; PI: Rich). This {\\it Letter} reports the discovery of a $2.0\\times10^4\\Msun$ BH in G1. Analysis of central population gradients and detailed ground-based spectra will be reported in a future paper (Rich et al. 2003). ",
        "conclusions": "G1 has traditionally been called a globular cluster. However, Meylan~\\etal\\ (2001) offer the hypothesis that it is a nucleus of an accreted small galaxy, similar to NGC~205. Furthermore, Ferguson~\\etal\\ (2002) report the discovery of a disrupted galaxy near G1, speculating that the two may have been part of the same system. A similar origin for $\\omega$~Cen has been proposed by Freeman (1993). Evidence in favor of this interpretation for both clusters include their large spread in metallicity and their exceptionally large velocity dispersions relative to most globular clusters. If G1 is the nucleus of an accreted galaxy, and if all bulge systems have BHs in their centers (Magorrian~\\etal\\ 1998), then, to the extent that a massive, bound cluster can be viewed as a ``mini-bulge,'' it is no surprise that G1 has a BH as well. Combined with the results for M15, it may be that {\\it every}\\ dense stellar system hosts a central BH.  There are significant consequences for these small systems having central BHs.  First, it provides a direct link between stellar clusters and galaxies. Galaxy correlation studies that include globular clusters (Burstein~\\etal\\ 1997, Geha~\\etal\\ 2002) show that they typically lie near to, but slightly offset from, the correlations for the nearby galaxies. However, the large scatter prevents a definitive comparison. The tightness of the $M_{\\bullet}-\\sigma$ relation allows us to explore their connection in better detail.  The current results show that there is little difference between the smallest and largest dense stellar systems.  Second, the existence of large (nonstellar) BHs in these small systems constrains BH formation models. The low escape velocity of M15 and G1 ($<$100~\\kms) makes it difficult to grow a BH slowly over time. Growing black holes from adiabatic accretion of gas is difficult since globular clusters have a hard time holding onto any gas (namely that from mass loss in evolved stars) due to their low escape velocities. If BHs are grown from accretion of stars and stellar remnants, the cluster must be able overcome the large recoil velocities due to two-body interaction near the center.  A possible solution to this dilemma is to have a large initial seed mass for the BH that cannot subsequently get ejected from two-body interactions. Miller \\& Hamilton (2002) and Portegies~Zwart \\& McMillan (2002) discuss a mechanism in which massive BHs can exist in globular clusters.  Third, one of the more important aspects of theories for the formation of supermassive BHs in galaxies is that each has to start with a seed BH. There are multiple explanations as to where these seeds come from; whether they are primordial, created during the formation of the galaxy, or formed in subsequent evolution is unknown. If small stellar clusters contain BHs, due to their ages, they are a natural candidate for the formation sites of seed BHs. A stellar cluster that formed before the host protogalaxy collapsed could have easily donated its BH to the galaxy center.  All of the theoretical models require only a modest-sized BH to act as a seed, around $1\\times10^4\\,\\Msun$ or less.  As an interesting counterexample, the galaxy M33 appears not to have a BH.  Gebhardt~\\etal\\ (2001) measure an upper limit of $1500\\Msun$. The main difference between M33 and other galaxies with detected BHs is that M33 does not have a a clear bulge component. However, M33 does have a compact nucleus, whose stellar density is as high as those in globular clusters. Thus, if any dense cluster has a BH then it is puzzling that M33's nucleus has none. There are no obvious reasons for this difference. However, M33's nucleus has a very different age than G1; the former contains a significant population of stars younger than a few Gyr (e.g., O'Connell 1983), whereas the latter is older than 10 Gyr (Meylan~\\etal\\ 2001). It is possible that either conditions to make a massive BH were better in the past (i.e., different initial mass function) or that M33's nucleus has not had enough time to create one. In any event, we need more data on a larger set of nuclei and clusters in order to explore this issue.  The model that we use for G1 assumes a constant stellar mass-to-light ratio. The relaxation time for G1 near the center is short and we expect heavy remnants there. By not including them, we overestimate the BH mass. We estimate the effect by extrapolation of the Fokker-Plank simulations of Dull~\\etal\\ (1997) for M15. They find that $\\sim 1000\\,\\Msun$ of remnants is in the central regions of that cluster (see discussion in Gerssen~\\etal\\ 2002). G1 is $\\sim$5 times more massive than M15, implying that it has 5 times more remnants given the same initial mass function. However, the relaxation times are significantly longer in G1 (about a factor of ten), suggesting that the presence of heavy remnants is even less of a problem. Furthermore, Gerssen~\\etal\\ include models which incorporate the appropriate mass-to-light variation and find that the required black hole mass {\\it increases} slightly. The reason is that even though the remnants increase the mass-to-light ratio at the smallest radius, the giant star cause it to drop towards the center since they are centrally-concentrated. This drop in mass-to-light balances the increase from the heavy remnants, thereby causing litle effect on the BH mass.  A possible concern comes from comparing the {\\it HST}/STIS dispersions with that measured from the ground using different setups and analysis. Therefore, we also ran models in which we use only the {\\it HST}\\ data. We find essentially the same BH mass, but the confidence band decreases slightly. The difference in $\\chi^2$ between the no-BH mass model and the best fit changes from 3.0 to 2.5. We find no reason to suspect that either dispersion measurement is biased and use both in the dynamical models.  The dynamical constraints for G1 can be improved with more extensive ground-based kinematic observations (i.e., multiple position angles). In addition to G1 and M15, there are a significant number of globular clusters that can be exploited for these studies. Moreover, ground-based observations should have sufficient spatial resolution to measure a central BH. The most important challenge, however, lies in understanding the contribution of heavy remnants. As explained above, the impact of heavy remnants on G1 should not be severe because of its long relaxation time, but more quantitative estimates of this effect using evolutionary models would be highly desirable. Based on experiments done so far in Gerssen~\\etal\\ (2002), it appears that the BH mass estimates in M15 and G1 are unbiased."
    },
    "0209/astro-ph0209125_arXiv.txt": {
        "abstract": "Recent studies of the 55 Cancri system suggest the existence of three planets with periods of $\\sim$15 days, $\\sim$45 days, and $\\sim$5500 days (Marcy et al. 2002).  The inner two planets are near the 3:1 mean motion commensurability and it is likely that these two planets became trapped in the resonance while farther from the star and migrated together.  As the innermost planet begins to dissipate energy through tides the planets break out of the resonance.  The final state of the system gives important information about its past history, such as the migration timescale that led to capture. ",
        "introduction": "Tidal evolution leading to capture into resonance has been invoked to explain the many commensurabilites in our solar system (e.g. Goldreich 1965), but the problem of breaking resonance due to tidal effects is less well-studied.  The presence of even a small amount of dissipation can alter the dynamical evolution of the system.  In this case, the dissipation causes the resonance to accelerate the circularization of the inner planet compared to what would happen in isolation.  The resonant interaction between the planets in the 55 Cancri system imply that fully self-consistent fits are necessary to obtain the orbital elements of the system.  The 3:1 resonance covers a narrow range of semi-major axes, so determining the location of the system within the resonance will require accurate fits, which requires additional data.  Direct integrations of the orbital elements given by one fully self-consistent fit (Laughlin 2002) indicate that the resonant angles are circulating and the system in {\\em not} in resonance.  To study the long term dynamics of the system, we use the classical disturbing function together with Lagrange's planetary equations to lowest order in eccentricity (e.g. Murray \\& Dermott 1999).  We numerically integrate these equations using a variable time step Bulirsch-Stoer integrator.  We consider tidal evolution in the regime where energy is dissipated and angular momentum is conserved.  The eccentricity damping rate is given by $\\dot{e}=-e/\\tau_e$ and $\\tau_e=GMme^2Q/anE_0$ where $G$ is the gravitational constant, $M$ is the mass of the star, $m$ is the mass of the inner planet, $Q$ parameterized tidal energy dissipation, $a$ is the planet's semi-major axis, $n$ is the planet's mean motion, and $E_0$ is the maximum amount of energy stored in the tidal deformation of the planet.  The planet's eccentricity e-folding time in isolation is given by $\\tau_e$. ",
        "conclusions": "We used the classical disturbing function together with Lagrange's planetary equations accurate to lowest order in eccentricity to study resonance breaking through tidal dissipation.  It is possible to derive analytic expressions for the inner planet's eccentricity when the resonance is broken and the time required to break the resonance (equations \\ref{ebreakeqn}, \\ref{tbreakeqn}).  These expressions are in good agreement with the secular theory (figure \\ref{tbreak})  This is applicable to the 55 Cancri system because the inner planet is near the regime where tidal effects are starting to play a role and current fits to the radial velocity curve give orbital elements where the planets are {\\em not} in the resonance.  We argue that the two planets became trapped in resonance while farther from the star and then broke the resonance when tidal dissipation began to play a role in the inner planet's evolution."
    },
    "0209/physics0209024_arXiv.txt": {
        "abstract": "We present a formalism for Newtonian multi-fluid hydrodynamics derived from an \\emph{unconstrained} variational principle. This approach provides a natural way of obtaining the general equations of  motion for a wide range of hydrodynamic systems containing an arbitrary number of interacting fluids and superfluids. In addition to spatial variations we use ``time shifts'' in the variational principle, which allows us to describe dissipative processes with entropy creation, such as chemical reactions, friction or the effects of external non-conservative forces. The resulting framework incorporates the generalization of the \\emph{entrainment} effect originally discussed in the case of the mixture of two superfluids by Andreev and Bashkin. In addition to the conservation of energy and momentum, we derive the generalized conservation laws of vorticity and helicity, and the special case of Ertel's theorem for the single perfect fluid.  We explicitly discuss the application of this framework to thermally conducting fluids, superfluids, and superfluid neutron star matter. The equations governing thermally conducting fluids are found to be more general than the standard description, as the effect of entrainment usually seems to be overlooked in this context. In the case of superfluid $^4$He we recover the Landau--Khalatnikov equations of the two-fluid model via a translation to the ``orthodox'' framework of superfluidity, which is based on a rather awkward choice of variables. Our two-fluid model for superfluid neutron star matter allows for dissipation via mutual friction and also ``transfusion'' via $\\beta$-reactions between the neutron fluid and the proton-electron fluid. ",
        "introduction": "The main purpose of this work is to develop a formalism that allows one to derive the equations of motion for a general class of multi-constituent systems of interacting charged and uncharged fluids, such as conducting and non-conducting fluids, multi-fluid plasmas, superfluids and superconductors. For the sake of clarity of presentation we restrict ourselves here to uncharged fluids, while the case of charged fluids and their coupling to the electromagnetic field will be treated in a subsequent paper \\cite{prix03:_variat_II}.  Long after the completion of classical Hamiltonian particle mechanics, the quest of finding a variational (or ``Hamlitonian'') description of hydrodynamics has surprisingly been a long-standing problem, which started only a few decades ago to be fully understood. The reason for this can be traced to the nature of the hydrodynamic equations, which are most commonly expressed in their Eulerian form in terms of the \\emph{density} $\\rho$ and \\emph{velocity} $\\vv$, where the information about the underlying flowlines has been hidden. Fluid particle trajectories, i.e. flowlines, can still be recovered by integrating the velocity field, but they are not independent quantities of the Eulerian description. However, it turns out that the ``true'' fundamental field variables of Hamiltonian hydrodynamics are the flowlines, which determine $\\rho$ and $\\vv$ as derived quantities.  Consider as an example the Lagrangian density $\\Lagr$ describing a barotropic perfect fluid, which in analogy to classical mechanics one would postulate to be $$ \\Lagr(\\rho,\\vv) = {1\\over 2}\\rho \\vv^2 - \\E(\\rho)\\,, $$ where $\\E(\\rho)$ represents the internal energy density of the fluid. We note that the internal energy defines the chemical potential $\\mut$ and the pressure $P$ as $$ d \\E = \\mut \\, d \\rho\\,,\\qaq P + \\E = \\rho \\, \\mut\\,. $$ The corresponding action is defined in the usual way as \\mb{$\\Act\\equiv\\int \\Lagr\\,d V\\,d t$}, and the variation $\\d\\Lagr$ of the Lagrangian density is $$\\d\\Lagr = \\rho\\vv\\cdot \\d\\vv + (\\vv^2/2 -\\mut)\\, \\d\\rho\\,.$$ Requiring the action $\\Act$ to be stationary with respect to \\emph{free variations} $\\d\\rho$ and $\\d\\vv$ is immediately seen to be useless, as this leads to the over-constrained equations of motion \\mb{$\\rho\\vv=0$} and \\mb{$\\mut=\\vv^2/2$}. In fact, it has been shown \\cite{schutz77:_variat_aspec} that an unconstrained variational principle with $\\rho$ and $\\vv$ as the fundamental variables cannot produce the Eulerian hydrodynamic equations. The reason for this is rather intuitive, as it is evident that free variations of density and velocity  probe configurations with different masses (i.e. different numbers of particles), which is not an actual degree of freedom of the dynamics of the system. Therefore the variational principle has to be constrained or reformulated in some way in order to restrict the variations to the physically meaningful degrees of freedom.  The historic approach to this problem in Newtonian physics has been to supplement the Lagrangian with appropriate constraints using Lagrange multipliers. This method was pioneered by Zilsel \\cite{zilsel50:_liquid_helium_ii} in the context of the two-fluid model for superfluid $^4$He, who used the constraints of conserved particles (i.e. mass) and entropy.  However, as pointed out by Lin \\cite{lin63:_hydrod_helium_ii}, this is generally insufficient, as it results in equations of motion restricted to \\emph{irrotational flow} in the case of uniform entropy. Lin showed that one has to add a further constraint, namely the ``conservation of identity'' of fluid particles in order to obtain the most general hydrodynamic equations. We can label particles by their initial positions $\\va$, and so we can write their flowlines as \\mb{$\\vx=\\vx(\\va,t)$}. The famous ``Lin constraint'' is \\mb{$\\partial_t \\va + \\vv\\cdot \\vnabla \\va = 0$}, i.e. the identity or label of a particle is conserved under its transport.  For reviews of this approach and its relation to the ``Clebsch representation'' we refer the reader to \\cite{seliger68:_variat,salmon88:_hamil_fluid_mechan,zakharov97:_hamil}, and references therein.  Although this method produces the correct equations of motion, it does not seem very natural due to the rather ad~hoc introduction of constraints, and the need for unphysical auxiliary fields (the Lagrange multipliers). It was pointed out by Herivel~\\cite{herivel55:_deriv_equat} that the \\emph{Lagrangian} as opposed to Eulerian formulation of hydrodynamics results in a much more natural variational description, and this approach was further developed and clarified by Seliger and Whitham \\cite{seliger68:_variat}. Instead of using $\\rho$ and $\\vv$ as fundamental variables, hydrodynamics can also be understood as a field theory in terms of the \\emph{flowlines} $\\vx(\\va,t)$, or equivalently \\mb{$\\va=\\va(\\vx,t)$}. It turns out that this formulation allows for a perfectly natural \\emph{unconstrained} variational principle. This seems rather intuitive considering that hydrodynamics is a smooth-averaged description of a many-particle system, which is described by a variational principle based on the particle trajectories, i.e. $\\vx_N$ and $\\dot{\\vx}_N$.  We can express the velocity and density in terms of the flowlines as \\mb{$\\vv =\\partial_t \\vx(\\va,t)$} and \\mb{$\\rho(\\vx,t) = \\rho_0(\\va) / \\det({\\Jac^i}_j)$}, where \\mb{${\\Jac^i}_j = \\partial x^i / \\partial a^j$} is the Jacobian matrix corresponding to the map \\mb{$\\va\\mapsto\\vx(\\va,t)$} between the physical space $\\vx$ and the ``material space'' $\\va$. Any further comoving quantities like the entropy $s$ are determined in terms of their initial value $s_0(\\va)$. Substituting these expressions into the Lagrangian $\\Lagr$, one obtains an unconstrained variational principle for the field $\\vx(\\va,t)$, which results in the correct equations of motion. It is interesting to note that this approach implicitly satisfies Lin's constraint, as we are varying the particle trajectories $\\vx(\\va,t)$, along which $\\va$ is a constant by construction. Also, we do not need to impose an a~priori constraint on the conservation of mass, as it is automatically satisfied by these ``convective'' variations: shifting around flowlines obviously conserves the number of flowlines, and therefore the number of particles. One can actually \\emph{derive} the Lin constraint by transforming this Lagrangian framework back into a purely Eulerian variational principle \\cite{seliger68:_variat,salmon88:_hamil_fluid_mechan}, which shows that these two approaches are formally equivalent.  As pointed out by Bretherton \\cite{bretherton70:_hamilton_fluid}, one can even more conveniently use a ``hybrid'' approach, in which the Lagrangian is expressed in terms of the Eulerian hydrodynamic quantities $\\vv$, $\\rho$, $s$ etc, but one consider them as functions of the underlying flowlines. Their variations are therefore naturally \\emph{induced} by variations $\\vxi$ of the flowlines $\\vx(\\va,t)$. In general relativity the same idea was pioneered by Taub \\cite{taub54:_gr_variat_princ}, and has subsequently been largely developed and extended by Carter \\cite{carter73:_elast_pertur,carter83:_in_random_walk,carter89:_covar_theor_conduc}, who also coined the term ``convective variational principle'' for this approach. Carter and Khalatnikov \\cite{carter92:_equiv_convective_potential} have further demonstrated the formal equivalence of the convective approach and the more common Clebsh formulation that results from an Eulerian variational approach. A ``translation'' of the covariant convective formalism into a Newtonian framework  (albeit using a spacetime-covariant language close to general relativity) is also available \\cite{carter94:_canon_formul_newton_superfl,carter04:_covar_newtonI}. The convective approach in relativity has independently been developed by Kijowski \\cite{kijowski79:_sympl_framew}, and Hamiltonian formulations have been constructed by Comer and Langlois \\cite{comer93:_hamil_multi_constituent} and Brown \\cite{brown93:_action_fluids}. Here we are using the convective (or ``hybrid'') variational principle in order to derive the Newtonian multi-fluid equations, and our notation and formalism follows most closely the framework developed by Carter.  We conclude our example of the simple barotropic fluid by using the convective variational principle to derive the Euler equation. The expressions for (Eulerian) variations of density and velocity \\emph{induced} by infinitesimal spatial displacements $\\vxi$ of the flowlines are well known\\footnote{A generalization of these expressions to include time-shifts is derived in Appendix~\\ref{sec:Variations}} (e.g. see \\cite{friedman78:_lagran}), namely $$ \\d\\rho = - \\vnabla\\left( \\rho \\vxi \\right) \\,,\\qaq \\d \\vv = \\partial_t \\,\\vxi + (\\vv\\cdot\\vnabla) \\vxi - (\\vxi\\cdot\\vnabla) \\vv\\,. $$ Inserting these expressions into the variation of the action \\mb{$\\d\\Act=\\int\\d\\Lagr\\,d V\\,d t$} with $\\d\\Lagr$ given above, and after some integrations by parts and dropping total divergences and time derivatives (which vanish due to the boundary conditions), we find \\begin{eqnarray*} \\d\\Act =& & -\\int \\vxi\\cdot\\biggl[\\rho(\\partial_t + \\vv\\cdot\\vnabla)\\vv + \\rho\\vnabla\\mut \\\\ && + \\vv \\left\\{\\partial_t \\rho + \\vnabla\\cdot(\\rho\\vv) \\right\\}\\biggl]\\,d V\\,d t\\,. \\end{eqnarray*} If we assume conservation of mass\\footnote{This will be seen to be a consequence of the variational principle rather than an a-priori assumption when time-shift variations are included.}, i.e. \\mb{$\\partial_t \\rho + \\vnabla\\cdot(\\rho\\vv)=0$}, then stationarity of the action (i.e. \\mb{$\\d\\Act=0$}) under free variations $\\vxi$ directly leads to Euler's equation, namely $$ (\\partial_t + \\vv\\cdot\\vnabla)\\,\\vv + {1\\over\\rho}\\vnabla P = 0\\,, $$ where we have used the thermodynamic identity \\mb{$\\rho\\vnabla\\mut = \\vnabla P$}. This shows that an unconstrained convective variational principle produces to the correct hydrodynamic equations of motion in a surprisingly simple and straightforward way.   The spatial variations $\\vxi$ have three degrees of freedom, resulting in one vector equation, which represents the conservation of momentum. In order to complete the description we will need a fourth variational degree of freedom to produce the missing energy equation. This can be achieved by considering time-shifts, which are a natural part of the covariant relativistic approach, but which we have to be considered explicitly in the conventional ``3+1'' language of Newtonian space-time. These time-shifts variations allow us to take this formalism to its full  generality, as we can now describe even dissipative processes with entropy creation, particle transformations (i.e. chemical reactions), resistive frictional forces etc. These dissipative systems are of course still \\emph{conservative} as long as one includes entropy, which is why they can be described by an action principle. The second law of thermodynamics, however, is obviously not contained in the action principle and has to be imposed as an additional equation on the level of the equations of motion.  We note that the equations we derive here do not explicitly include shear- and bulk-viscosity effects. However, the current \\emph{form} of the equations is in principle general enough to allow for both of these effects: bulk viscosity is caused by heat flow or chemical reactions due to thermal or chemical disequilibrium, both of which can already be described in the current formulation. Shear viscosity on the other hand has to be introduced as an ``external'' force, the problem therefore consists in prescribing a  physically reasonable model for a multi-fluid generalization of the shear stresses. Including viscosity should therefore not be a matter of actually \\emph{extending} the current framework but rather of appropriately applying it in order to describe such processes. An explicit discussion of this is postponed to future work. Further work is also necessary in order to extend this formalism to include elasticity (as pioneered in the relativistic framework \\cite{carter72:_found_elasticity}), and especially to allow for an elastic medium interpenetrated by fluids as encountered in the inner neutron star crust, or any type of conducting solid. As shown in \\cite{carter95:_kalb_ramond}, a Kalb-Ramond type extension is required for the macroscopic treatment of quantized vortices in superfluids. With the present formalism we can describe superfluids either on the local irrotational level, or on the smooth-averaged macroscopic level by neglecting the (generally small) anisotropy induced by the quantized vortices.  The plan of this paper is as follows: in Sect.~\\ref{sec:GeneralMultiConst} we derive the general form of the equations of motion for multi-constituent systems using the convective variational principle. In Sect.~\\ref{sec:TotalConservationLaws} we show the conservation of energy and momentum implied by these equations. In Sect.~\\ref{sec:FlowlineConservations} we derive conserved quantities under transport by the flow, namely the vorticity and helicity. We then give the explicit functional form of the Lagrangian density for hydrodynamic systems in Sect.~\\ref{sec:Hydrodynamics}, and in Sect.~\\ref{sec:Applications} we discuss several applications of the foregoing formalism to particular physical systems. ",
        "conclusions": ""
    },
    "0209/astro-ph0209369_arXiv.txt": {
        "abstract": "{\\small Multifrequency VLA and OVRO observations of the radio outburst of Cygnus X-3 in September 2001 are presented, illustrating the evolution of the spectrum of the source over a period of six days. An estimate of the magnetic field in the emitting region is made from the spectral turnover and possible explanations for the spectral evolution are suggested.} ",
        "introduction": "Cygnus X-3 is an X-ray binary system located in the Galactic plane at a distance of $\\approx$10\\,kpc (\\eg \\cite{Pre00}, \\cite{Dic83}) in or behind one of the spiral arms.  There is ongoing debate as to the nature of the compact object (\\eg \\cite{Mit98},\\cite{Sch96}).  The large interstellar extinction to the source precludes any optical spectroscopy, making it difficult to obtain a reliable mass function and hindering identification of the companion star. Analysis of infrared spectra \\cite{vanKer96} suggests that the companion is a WN7 Wolf-Rayet star, although more recent observations \\cite{Fuc02} indicate a WN8 subclass.  The orbital period, as confirmed in X-ray and infrared flux modulations, is 4.8 hours (\\eg \\cite{Par72}).  The source occasionally undergoes huge radio outbursts where the flux density increases to a level of up to 20\\,Jy.  Two-sided jets have been seen on arcsecond scales in a N-S orientation \\cite{Mar01}, whereas a highly-relativistic ($\\beta \\geq 0.81$) one-sided jet with the same orientation has been reported on milliarcsecond scales with the VLBA \\cite{Mio01}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209475_arXiv.txt": {
        "abstract": "{ The state of the matter that is obscuring a small circumnuclear region in active galactic nuclei can be probed by observations of its broad emission lines. Infrared lines are particularly useful since they penetrate significant columns of obscuring matter, the properties of which can be constrained by comparing infrared and X-ray obscuration. We report on new 4$\\mu$m spectroscopy with ISAAC at the ESO VLT of a sample of 12 Seyfert 2 galaxies, probing for broad components to the Brackett~$\\alpha$ 4.05$\\mu$m hydrogen recombination line. Broad components are observed in 3 to 4 objects. All objects with a broad component exhibit relatively low X-ray obscuring columns, and our results are consistent with a Galactic ratio of 4$\\mu$m obscuration to the BLR and X-ray column. In combination with observations of a {\\em non}-Galactic ratio of {\\em visual} obscuration of BLRs and X-ray obscuring column in Seyferts, and interpreted in a unified AGN scheme, this result can be reconciled with two interpretations. Either the properties of dust near the AGN are modified towards larger grains, for example through coagulation, in a way that significantly flattens the optical/IR extinction curve, or the ratio of dust obscuration to X-ray column varies for different viewing angles with respect to the axis of symmetry of the putative torus. Our spectra also provide a survey of emission in the [Si\\,IX] 3.94$\\mu$m coronal line, finding variation by an order of magnitude in its ratio to Br$\\alpha$. The first extragalactic detection of the [Ca\\,VII] 4.09$\\mu$m  and [Ca\\,V] 4.16$\\mu$m coronal lines is reported in the spectrum of the Circinus galaxy. ",
        "introduction": "\\label{sect:intro} Unified scenarios have been highly successful in explaining several aspects of the AGN phenomenon, by assuming that different manifestations of the AGN phenomenon correspond to similar objects viewed from different directions. The detection in polarized light of broad emission lines in Seyfert 2 galaxies (Antonucci \\& Miller \\cite{antonucci85}, Antonucci \\cite{antonucci93}) has been central to the development of these scenarios. Subsequently, spectropolarimetry has become the prime tool for detecting hidden broad lines in larger samples (e.g., Miller \\& Goodrich \\cite{miller90}; Tran et al. \\cite{tran92}; Young et al. \\cite{young96}; Heisler et al. \\cite{heisler97}; Moran et al. \\cite{moran00}; Lumsden et al. \\cite{lumsden01}; Tran \\cite{tran01}). X-ray spectroscopy has been the second key observation, finding Seyfert 2s on average much more highly obscured than Seyfert 1s and quantitatively establishing the absorbing column densities in neutral and `warm', i.e. highly ionized material (e.g., Turner et al. \\cite{turner97}; Bassani et al. \\cite{bassani99}). Still, relatively little is known about the actual distribution and physical state of the material obscuring the central engine of Seyfert 2 galaxies from our view. Is it in the form of a compact parsec scale torus (Krolik \\& Begelman \\cite{krolik86})? Or does obscuration on scales of tens or hundreds of parsec play a significant role (e.g. Maiolino \\& Rieke \\cite{maiolino95}, Malkan et al. 1998)? Mass arguments suggest that at least the Compton-thick absorbers are on scales of tens of parsecs or less (Risaliti et al. \\cite{risaliti99}), but lower column components are more difficult to constrain. Do outflows contribute to forming the obscuration (e.g. K\\\"onigl \\& Kartje \\cite{koenigl94}; Elvis \\cite{elvis00})? How does the state of the obscuring matter differ from the interstellar medium in a normal galaxy, given the extreme conditions close to a powerful AGN? High obscuring columns can be explained by different scenarios, and the warm dust emission seen in the mid-infrared cannot uniquely distinguish between compact and more extended configurations either (e.g., Pier \\& Krolik \\cite{pier92}; Efstathiou et al. \\cite{efstathiou95}; Granato et al. \\cite{granato97}).  One way to address some of these issues is to study the obscuration of the central engine in wavelength ranges that are {\\em partially} transparent at the column densities of interest, and compare the results. Of particular value are X-rays and the infrared range, but care has to be taken to compare obscuration towards similar regions. Narrow Line Region (NLR) emission and mid-infrared dust emission probe regions that are much larger than the central X-ray source. Therefore, their obscuration may involve different foreground material, and cannot be compared directly to X-ray results. In contrast, reverberation mapping results show the Broad Line Region (BLR) to be well below a parsec in size (Netzer \\cite{netzer90}), allowing a meaningful comparison of BLR and X-ray obscuration, by material that might be found in both parsec-scale and larger regions. Our goal is to constrain infrared obscuration, and thus the state of the obscuring matter by the detection or nondetection of infrared BLRs in Seyfert 2s of various X-ray obscuring columns. Recently, Maiolino et al. (\\cite{maiolino01a}) have compared visual and X-ray obscuration in Seyferts, finding large deviations from standard Galactic values. In many objects, the ratio of reddening $E_{\\rm B-V}$ and X-ray column $N_{\\rm H}$ is low by about an order of magnitude compared to the Galactic value. These results provide additional motivation to explore the relation between infrared and X-ray obscuration, and compare it with that in the visual.  Practical considerations drive the choice of the optimal transition for Broad Line Region searches in the infrared. Standard interstellar extinction laws have a minimum in the 3-8$\\mu$m range, then rise through the silicate features and drop again steeply beyond 30$\\mu$m. This tends to argue in favour of the  longest wavelength infrared observations. However, the flux of the strongest ($\\alpha$) recombination lines drops approximately with the second power of wavelength, while dust continuum increases making their detection increasingly difficult. On the basis of such reasoning and of ISO spectroscopy of Brackett~$\\beta$, Brackett~$\\alpha$, and Pfund~$\\alpha$ in NGC\\,1068, Lutz et al. (\\cite{lutz00a}) concluded that Brackett~$\\alpha$ with its fairly high line to continuum ratio and low obscuration is the most promising line for infrared BLR searches. Using sensitive instruments such as ISAAC on 8m telescopes like the VLT, it is now possible to perform such observations and take a next step beyond earlier infrared searches for BLRs, which either focussed on lines that have a good line to continuum but still relatively high obscuration (Pa$\\beta$ 1.28$\\mu$m) or are suffering less extinction but are difficult to measure because of a low line-to-continuum ratio (Br$\\gamma$ 2.17$\\mu$m). Searches in these lines detected several broad components, but cannot probe beyond equivalent visual obscurations of 10--20\\,mag (e.g. Rix et al. \\cite{rix90}; Blanco et al. \\cite{blanco90}; Goodrich et al. \\cite{goodrich94}; Ruiz et al. \\cite{ruiz94}; Veilleux et al. \\cite{veilleux97}; Gilli et al. \\cite{gilli00}).  This paper is organized as follows. Sect.~\\ref{sect:obs} discusses the observations and data analysis, Sect.~\\ref{sect:coronal} the result of a coronal line survey obtained from our data, Sect.~\\ref{sect:bra} presents the results of the Brackett~$\\alpha$ spectroscopy including line decompositions. We discuss the results and compare to X-ray and optical work in Sect.~\\ref{sect:discussion} and conclude in Sect.~\\ref{sect:conclusions}. ",
        "conclusions": "\\label{sect:conclusions}  We have presented new 4$\\mu$m spectroscopy of a sample of 12 Seyfert 2 galaxies that are well-studied in the X-ray, and combine these data with previous spectroscopy of NGC\\,1068. The observations are designed to probe for the presence of optically obscured Broad Line Regions. The main results are\\\\ (i) Broad components  to Brackett~$\\alpha$ are detected in 3 to 4 of these 13 galaxies.\\\\ (ii) The detections and limits are consistent with a Galactic ratio of infrared and X-ray obscuring columns. This result can be reconciled with the low ratios of optical to X-ray obscuring columns observed for several AGN if either the obscuring dust consists of large grains leading to a modified extinction curve, or if variation in dustiness or dust properties exists  between directions probing right through the putative torus and directions closer to its opening.\\\\ (iii) A survey of the coronal [Si\\,IX] 3.94$\\mu$m line shows considerable variation in its ratio to Brackett~$\\alpha$.\\\\ (iv) Two coronal lines of [Ca\\,VII] and [Ca\\,V] are detected for the first time in an extragalactic object, the Circinus galaxy."
    },
    "0209/astro-ph0209196_arXiv.txt": {
        "abstract": "We present an abundance analysis of six main sequence turnoff, subgiant, and giant branch stars toward the Galactic bulge that were observed with Keck/HIRES during microlensing events.   This is an early look at the first detailed chemical analysis of main sequence stars in the Galactic bulge. Lensing events allow the effective aperture of Keck to be increased beyond its current dimensions; although, some events still stretched its spectroscopic capabilities.  Future large telescopes with high resolution and high throughput spectrometers will allow the study of abundances in distant stellar populations and in less evolved stars with greater ease. ",
        "introduction": "\\label{sect:intro}  %  The distance to the Galactic bulge has until recently kept the prospects of observing directly the chemical properties of its constituent stars just past the edge of possibility. Even now, with large-aperture telescopes such as the Kecks, the VLT, the HET, and the Gemini twins among others, only the brightest giants in the Bulge have been observable~\\cite{RM2000}. A novel approach employed by the MACHO microlensing team has taken advantage of the temporary brightening of the source star during the lensing event to effectively decrease that once prohibitive distance to the Galactic center and allow for the observation of less evolved stars. For example, Minniti et al.~\\cite{Minniti} first published results in 1997 on the presence of a Li I feature observed in a hot ($T_{\\rm eff}~=~6000$~K) main sequence turn-off star. In this paper we discuss the full analysis of the that spectrum along with the spectra of five other stars observed during lensing events between 1997 and 1999. The results to date are still preliminary; however, they offer a glimpse into the diverse chemical evolution scenarios that can be found in and around the center of the Milky Way. Future telescopes with effective apertures in the 30-m or more range will make such observations commonplace rather than rare. ",
        "conclusions": "Microlensing has allowed us to derive a reasonably large and detailed set of abundances for a group of stars that have heretofore been beyond our capabilities. While some work on the details remains to be done before conclusions can be firmly drawn, we have been able to study abundance trends in stars as diverse as solar analogs and metal-poor red giants. Our early results show a trend of overabundance in nearly all the elements regardless of metallicity or spectral class. The two giants in the study also have the lowest metallicities and the highest radial velocities, and they are easily ruled out as members of the Sagittarius dwarf spheroidal. We have also re-analyzed the Li abundance in the star 97-BLG-45 and have found that we cannot conclusively rule for or against the presence of Li. All the spectra show some hint of Li ${\\lambda}$6708 and we will examine these in kind to determine abundances or upper limits.  Our results are likely to change for some of the elements where hfs or isotopic corrections are still necessary. When possible we corrected for hfs effects by correcting one line and applying a shift to the results from all the lines. Sometimes the shift was large ($>~0.5$ dex) and sometimes it was nonexistent.  Thanks to a helping hand from Nature, we were able to get a glimpse at the functioning of Keck as a much larger telescope than its current 10-m diameter. Future 30-m and larger class telescopes will make such observations commonplace and allow us to learn truly the diverse histories of the Galactic bulge stellar population."
    },
    "0209/astro-ph0209533_arXiv.txt": {
        "abstract": "{We present the results of bispectrum speckle interferometry of the B[e] star MWC\\,349A obtained with the SAO 6\\,m telescope. Our diffraction-limited $J$-, $H$-, and $K$-band images (resolutions 43--74\\,mas) suggest the star is surrounded by a circumstellar disk seen almost edge-on. The observed visibility shape is consistent with a two-component elliptical disk model, probably corresponding to the gaseous and dusty components of the disk. We show that the classification of the object as a pre-main-sequence star or a young planetary nebula is problematic. An analysis of the uncertainties in the basic parameter determination lead us to the conclusion that MWC\\,349A is probably either a B[e] supergiant or a binary system, in which the B[e]-companion dominates the observed properties. ",
        "introduction": "\\object{MWC\\,349} consists of two apparently close (the angular distance is 2.4\\,arcsec) objects: \\object{MWC\\,349A}, which exhibits a strong emission-line spectrum and IR-excess, and \\object{MWC\\,349B}, a weak emission-line source which is $\\sim$40\\% fainter than MWC\\,349A in the optical region (Cohen et al. \\cite{cbdw85}, hereafter C85). C85 argued that MWC\\,349A and MWC\\,349B are a physical pair, as they are connected by an arc of emitting matter, and estimated the distance (1.2\\,kpc) toward it based on the spectral type and luminosity of MWC\\,349B (B0\\,{\\sc iii} and M$_{V}=-$5.0\\,mag, respectively). However, Meyer et al. (\\cite{mnh02}) found that the A and B components have a different level of interstellar polarization, suggesting that they are not connected to each other and that MWC\\,349A could be a member of Cyg\\,OB2. In this paper we concentrate on MWC\\,349A and use MWC\\,349B only as a reference star for our observations.  MWC\\,349A is a peculiar stellar object with one of the strongest emission-line spectra ever observed. It is located in the southern part of the \\object{Cyg OB2} association, a heavily reddened stellar cluster at a distance of 1.7\\,kpc (e.g., Kn\\\"odlseder \\cite{knodl00}). The object does not show any photospheric lines in its spectrum (Andrillat, Jaschek, \\& Jaschek \\cite{ajj96}), although the presence of strong He\\,{\\sc i} emission lines indicates a high temperature of the underlying source. In most of the studies devoted to the nature of MWC\\,349A its spectral type is estimated as late O (e.g., Hartmann, Jaffe, \\& Huchra \\cite{hjh80}). The co-existence of forbidden emission lines in the optical spectrum and a near-IR excess brought MWC\\,349 into the group of B[e] stars (Allen \\& Swings \\cite{as76}). On the other hand, MWC\\,349A is considered to be a pre-main-sequence object because of the surrounding nebula, which is formed by an optically-thick bipolar outflow seen at radio wavelengths (Olnon \\cite{o75}). The bipolar structure of the nebula suggests that the stellar core is surrounded by a disk, whose presence is inferred from a high level of polarization (Elvius \\cite{elv74}, Zickgraf \\& Schulte-Ladbeck \\cite{zsl89}, Yudin \\cite{y96}; Meyer, Nordsieck, \\& Hoffman \\cite{mnh02}) and double-peaked emission-line profiles (Hartmann et al. \\cite{hjh80}). MWC\\,349A also shows maser and laser line emission in the IR and radio spectral regions, which makes it rather unique among similar sources (see Gordon et al. \\cite{g01} for a recent update).  Since MWC\\,349A is a distant object, observations with high spatial resolution are crucial for revealing its nature. So far results of only a few high-resolution near-IR observations have been published. Leinert (\\cite{l86}) marginally resolved the IR source and found it elongated with a gaussian FWHM of the brightness distribution of $85\\pm19$\\,mas in the east-west direction at L$^{\\prime}$ and $38\\pm18$\\,mas in the north-south direction at K. The highest resolution measurements were presented by Danchi, Tuthill, \\& Monnier (\\cite{dtm01}). These authors showed that the IR source can be fitted by uniform ellipses with major axes of $36\\pm2$\\,mas at 1.65\\,$\\,\\mu$m, 47$\\pm$2\\,mas at 2.25\\,$\\,\\mu$m and $62\\pm1$\\,mas at 3.08\\,$\\,\\mu$m, axial ratios of $\\sim$0.5, and a position angle of $100^{\\degr} \\pm 3^{\\degr}$. Under a flat disk approximation, Danchi et al. estimated the inclination angle of the disk plane to the line of sight to be $\\sim21^{\\degr}$. They also found the observed disk sizes at the above wavelengths consistent with the theoretical predictions for the inner and outer dimensions of photoevaporating accretion disks around young stars reported by Hollenbach et al. (\\cite{hjls94}).  To verify the results by Danchi et al. (\\cite{dtm01}) and to extend the spatial information toward shorter wavelengths, we obtained new speckle interferometric observations of MWC\\,349A in $J$, $H$, and $K$ broadband filters and a narrowband filter at 2.09 $\\mu$m. Our results are presented in Sect.~\\ref{obsres}. Additionally, we review the existing information about the object, as its nature and evolutionary state remain uncertain. Most of the investigators focus on details of the circumstellar (CS) structures (mainly on the disk) in efforts to interpret the observed features, while properties of the illuminating source are only vaguely known. In Sect.~\\ref{nature} we re-estimate physical parameters of the object and examine existing hypotheses on its nature. In Sect.~\\ref{discuss} we consider the reliability of the adopted parameters and summarize our findings in Sect.~\\ref{conclus}.  \\begin{figure*} \\begin{center} \\epsfxsize=50mm \\mbox{\\epsffile{2826_f01.eps}}\\hspace{4mm} \\epsfxsize=56mm \\mbox{\\epsffile{2826_f05.eps}} \\epsfxsize=50mm \\mbox{\\epsffile{2826_f09.ps}}\\\\[4mm] \\vspace{0.5cm}  \\epsfxsize=50mm \\mbox{\\epsffile{2826_f02.eps}}\\hspace{4mm} \\epsfxsize=56mm \\mbox{\\epsffile{2826_f06.eps}} \\epsfxsize=50mm \\mbox{\\epsffile{2826_f10.ps}}\\\\[4mm] \\vspace{0.5cm}  \\epsfxsize=50mm \\mbox{\\epsffile{2826_f03.eps}}\\hspace{4mm} \\epsfxsize=56mm \\mbox{\\epsffile{2826_f07.eps}} \\epsfxsize=50mm \\mbox{\\epsffile{2826_f11.ps}}\\\\[4mm] \\vspace{0.5cm}  \\epsfxsize=50mm \\mbox{\\epsffile{2826_f04.eps}}\\hspace{4mm} \\epsfxsize=56mm \\mbox{\\epsffile{2826_f08.eps}} \\epsfxsize=50mm \\mbox{\\epsffile{2826_f12.ps}}\\\\[4mm] \\end{center} \\caption{Left column: two-dimensional visibilities. Middle column: cuts through the long and short axis of the visibilities and the visibilities of the best-fitting two-dimensional elliptical uniform-disk models fitted up to the telescope cut-off frequency. Right column: images of MWC\\,349A reconstructed by bispectrum speckle interferometry. The contours are plotted in steps of 0.3\\,mag (from 0.3\\,mag to 4.2\\,mag of peak intensity).} \\label{f1} \\end{figure*}   \\begin{figure*} \\begin{center} \\epsfxsize=66mm \\mbox{\\epsffile{2826_f13.eps}}\\hspace{4mm} \\epsfxsize=66mm \\mbox{\\epsffile{2826_f14.eps}}\\\\[8mm] \\epsfxsize=66mm \\mbox{\\epsffile{2826_f15.eps}}\\hspace{4mm} \\epsfxsize=66mm \\mbox{\\epsffile{2826_f16.eps}}\\\\[8mm] \\end{center} \\caption{The visibilities of the best-fitting two-component visibility models consisting of two Gaussian functions fitted to the cuts through the short and long axes of the measured visibilities (corresponding to the major and minor axes of the Gaussian disk in image space). The fit range chosen is up to the telescope cut-off frequency. The numbers in parenthesis are the FWHM diameters  of the large/small component of the two-component model.} \\label{f2} \\end{figure*} ",
        "conclusions": "We have obtained and analyzed diffraction-limited (resolutions 43--74\\,mas) speckle interferometric $J-$, $H-$, and $K$-band observations of MWC\\,349A. The images are elongated at all wavelengths with an axial ratio of $\\sim$0.7. Their position angles coincide with those obtained by Danchi et al. \\cite{dtm01}. The best-fit model to the observed visibilities consists of 2 elliptical components (e.g. $J$-band size: $\\sim 22\\times12$\\,mas and $\\sim 119\\times74$\\,mas; Table\\,\\ref{t3}). The inner component probably corresponds to the CS gas emission, while the outer one can be produced by a dusty disk, whose symmetry plane is seen at a low inclination angle.  We also summarized existing information about observed parameters of MWC\\,349A, and hypotheses previously suggested to explain its nature. This analysis resulted in a new luminosity estimate, which is based on the assumption that the object belongs to Cyg\\,OB2 (see Meyer et~al. \\cite{mnh02} for the justification) and indicates that it is more luminous than was suggested before. We concluded that MWC\\,349A is unlikely to be a pre-main-sequence star or a young planetary nebula. Its fundamental parameters and observed features (the strong emission-line spectrum and dust emission) are more consistent with those of B[e] supergiants.  A binary nature of MWC\\,349A cannot be excluded, but this hypothesis needs further observational testing. In particular, optical photometric observations to search for possible brightness variations, and high-resolution spectroscopy to search for radial velocity variations both associated with the orbital motion are very important for further progress in the understanding of this puzzling object. Quantitative analysis of our speckle images within this model will be presented elsewhere."
    },
    "0209/hep-ph0209094_arXiv.txt": {
        "abstract": "We re-examine the sensitivity of solar neutrino oscillations to noise in the solar interior using the best current estimates of neutrino properties. Our results show that the measurement of neutrino properties at KamLAND provides new information about fluctuations in the solar environment on scales to which standard helioseismic constraints are largely insensitive. We also show how the determination of neutrino oscillation parameters from a combined fit of KamLAND and solar data depends strongly on the magnitude of solar density fluctuations.  We argue that a resonance between helioseismic and Alfv\\'en waves might provide a physical mechanism for generating these fluctuations and, if so, neutrino-oscillation measurements could be used to constrain the size of magnetic fields deep within the solar radiative zone. ",
        "introduction": "Current solar \\citep{ahm02a,ahm02b,Fukuda:2002pe,chlorine,sage,gallex} and atmospheric \\citep{Fukuda:1998mi} neutrino data give compelling evidence that neutrino conversions take place. For the simplest case of oscillations, the relevant parameters are two mass-squared differences $\\Delta m^2_{\\mathrm{sol}}$ and $\\Delta m^2_{\\mathrm{atm}}$, three angles $\\theta_{12}$, $\\theta_{23}$, $\\theta_{13}$ plus a number of CP violating phases~\\citep{Schechter:1980gr}. One knows fairly well now that $\\theta_{23}$ is nearly maximal (from atmospheric data) and that the preferred solar solution for $\\theta_{12}$ is the so-called large mixing angle (LMA) solution~\\citep{Gonzalez-Garcia:2000aj}, while the third angle $\\theta_{13}$ is strongly constrained by the result of reactor experiments~\\citep{Apollonio:1999ae}. The CP phases are completely unknown at present. A recent analysis of solar and atmospheric data in terms of neutrino oscillations is given in \\citet{Maltoni:2002ni}, and finds the currently-favored LMA solution of the solar neutrino problem has \\begin{equation} \\label{eq:lmabfp} \\tan^2\\theta = 0.46\\,, \\quad \\Delta m^2 = 6.6\\times 10^{-5}~\\eVq \\end{equation} and corresponds to oscillations into active neutrinos. Other recent analyses of solar data can be found in: \\citet{Fogli:2002pt,bah02,ban02a,ban02b,barger02,dho02,cre01}.  The recent results from the KamLAND reactor experiment~\\citep{kamland} have brought neutrino physics to a new stage. For the first time the solar neutrino anomaly has been probed using terrestrial neutrino sources.  This is fundamental for two reasons. First, among the various proposed solutions of the solar neutrino problem, such as the possibility of neutrino spin-flavor-precession~\\citep{Schechter:1981hw,akh88,lim88,mir01a,mir01b,bar02}, or non-standard neutrino matter interactions~\\citep{Guzzo:2001mi}, which may arise in models of neutrino mass~\\citep{moh86,hal86}, it singles out a unique ``oscillation-type'' solution: the LMA MSW solution.  Second, it brings to fruition one of the initial motivations for studying solar neutrinos in the first place~\\citep{Bahcall}: the use of solar neutrinos to infer the equilibrium properties of the solar core.  In this article we make the following points: \\begin{itemize} \\item We show how the determination of neutrino oscillation parameters from a combined fit of KamLAND and solar data shows a strong dependence on the magnitude of solar density fluctuations.  \\item We show that the fact that the KamLAND results largely support LMA neutrino oscillations, can be used to provide new information about fluctuations in the solar core on much shorter scales than those which existing constraints (like helioseismology) can presently probe.  \\item We propose a physical process which could arise in the core, that may produce fluctuations on the scales to which solar neutrinos are sensitive. \\end{itemize} ",
        "conclusions": "\\label{sec:summary-conclusion}  We have re-examined the sensitivity of solar neutrino oscillations to fluctuations in the solar density profile, using the best current estimates of neutrino properties, especially the new reactor data from KamLAND. Our results show that the measurement of neutrino properties in the latter experiment provides new information about fluctuations in the solar environment on scales to which standard helioseismic constraints are largely insensitive. Conversely we have seen how the determination of solar neutrino parameters from a fit of the data in the case of a noisy Sun differ from the quiet Sun case.  We have also argued that a resonance between helioseismic and Alfv\\'en waves can provide a physical origin for such fluctuations and, if so, neutrino-oscillation measurements could be used to constrain the size of magnetic fields deep within the solar radiative zone."
    },
    "0209/astro-ph0209568_arXiv.txt": {
        "abstract": "We report the discovery of two spectroscopic binaries in the field of the old open cluster M\\,67 -- S\\,1063 and S\\,1113 -- whose positions in the color-magnitude diagram place them $\\approx$ 1 mag below the subgiant branch. A ROSAT study of M\\,67 independently discovered these stars to be X-ray sources. Both have proper-motion membership probabilities greater than 97\\%; precise center-of-mass velocities are consistent with the cluster mean radial velocity. S\\,1063 is also projected within one core radius of the cluster center.   S\\,1063 is a single-lined binary with a period of 18.396 days and an orbital eccentricity of 0.206. S\\,1113 is a double-lined system with a circular orbit having a period of 2.823094 days. The primary stars of both binaries are subgiants. The secondary of S1113 is likely a 0.9  $M_{\\odot}$ main-sequence star, which implies a  1.3  $M_{\\odot}$ primary star. We have been unable to explain securely the low apparent luminosities of the primary stars. The colors of S1063 suggest 0.15 mag higher reddening than found for either M67 or through the entire Galaxy in the direction of M67. S1063 could be explained as an extincted M67 subgiant, although the origin of such enhanced extinction is unknown. The photometric properties of S1113 are well modeled by a cluster binary with a 0.9  $M_{\\odot}$ main-sequence secondary star. However, the low composite luminosity requires a small (2.0 $R_{\\odot}$) primary star that would be supersynchronously rotating, in contrast to the short synchronization timescales, the circular orbit, and the periodic photometric variability with the orbital period. Geometric arguments based on a tidally relaxed system suggest a larger (4.0 $R_{\\odot}$) primary star in a background binary, but such a large star violates the observed flux ratio. Thus we have not been able to find a compelling solution for the S1113 system. We speculate that S1063 and S1113 may be the products of close stellar encounters involving binaries in the cluster environment, and may define alternative stellar evolutionary tracks associated with mass-transfer episodes, mergers, and/or dynamical stellar exchanges. ",
        "introduction": "The open cluster M\\,67 is one of the most comprehensively studied of all star clusters, and has long been the prototype for old (4 Gyr) open clusters in the Galaxy. Indeed, one of the first photoelectric color-magnitude diagrams was derived for M\\,67 by Johnson \\&\\ Sandage (1955).  Since that time the precision of stellar photometry has steadily improved, and with it has the definition of the M\\,67 giant branch (Janes \\&\\ Smith 1984, Montgomery et al. 1993). Indeed the remarkable narrowness of the M\\,67 giant branch has served as a precise touchstone against which innumerable single-star evolution models have been tested (e.g., Dinescu et al. 1995).  However, even after application of strict proper-motion membership criteria, the color-magnitude diagram of M\\,67 remains littered with stars that do not fall on a single-star isochrone (Fig.~\\ref{s1063:cmd}). Some of these seeming anomalies can be accounted for. For example, one luminous star to the blue of the giant branch is a spectroscopic binary whose composite light can be explained by a giant--main-sequence pairing; another is a giant--white dwarf pair with a complicated history (Mathieu et al.  1990, Verbunt \\&\\ Phinney 1995, Landsman et al. 1997). Many of the stars immediately above the turnoff and subgiant branch are spectroscopic binaries and consequently overluminous. And the parallel sequence of stars to the red of the main sequence is assuredly comprised of binaries as well (Montgomery et al. 1993).  Still, there remain stars that are not so easily explained. Most famous are the blue stragglers, first noted in M\\,67 by Johnson \\&\\ Sandage. The origin of blue stragglers in an open cluster environment is still not securely understood today (Bailyn 1995, Hurley et al. 2001). Similarly, the yellow giant S\\,1072, lying 1.2 magnitudes in V above the main-sequence turnoff, has defied explanation. A kinematic member in all three dimensions and lying in projection in the cluster core, its explanation as a non-member remains possible but not satisfying (Mathieu \\&\\ Latham 1986, Nissen et al. 1987). In this paper we consider two more stars -- S\\,1063 and S\\,1113 -- which by every indication are cluster members, yet whose location in the M\\,67 color-magnitude diagram is dramatically inconsistent with single-star evolutionary theory. Specifically, as shown in Fig.~\\ref{s1063:cmd}, these stars lie below the subgiant branch.  Our attention has converged on these stars from two directions. We have underway a multi-decade survey of the spectroscopic binary population in M\\,67 (Mathieu et al. 1990, Latham et al. 1997). Both S\\,1063 and S\\,1113 were found to be spectroscopic binaries, with S\\,1113 in particular being notable for the rapid rotation of its primary star. Belloni et al. (1998) have also undertaken a comprehensive study of stellar X-ray sources in M\\,67 based on ROSAT observations. S\\,1063 and S\\,1113 were independently discovered to be X-ray sources, of which there are only 25 known among cluster members. Both stars are also photometric variables (van den Berg et al. 2002), which are infrequent in a cluster of this age (Stassun et al. 2002).  In this paper we present a comprehensive discussion of the orbital, spectroscopic, and photometric properties of S\\,1063 and S\\,1113. Regrettably, like the blue stragglers and S\\,1072, their interpretation remains a puzzle. ",
        "conclusions": "\\label{discussion} {\\bf S\\,1063} The binary S1063 is remarkable in three ways: low apparent brightness compared to M67 subgiants of similar color, enhanced reddening compared to Galactic extinction along the line of sight, and weak H$\\alpha$ emission observed at a velocity different from the primary star velocity. In addition, the long timescale light variations are not easily explained if they are aperiodic. If they are periodic on timescales longer than 18 days, then they likely are due to star spots on the primary star and indicate that the primary is not rotating pseudosynchronously.  Given both its close kinematic association with M67 and projection upon the core of the cluster, we have explored explanations for S1063 in the context of cluster membership. In this context, we have previously inferred a radius of 2.4 $R_{\\odot}$ for the primary star. As such, the primary star does not approach its critical surface at periastron passage. Thus the radius of the star is not confined by the presence of the secondary, nor is there motivation for present mass transfer. Similarly, the eccentricity of the orbit does not suggest either a large evolved star or mass transfer in the past, both of which would likely have circularized the orbit.  We have explored the possibility that rather than being an underluminous subgiant the primary star is instead an overluminous main-sequence star. For example, a recent merger of two main-sequence stars would deposit kinetic energy into the merging stars. Here we consider two types of mergers: coalescence and collision. In the coalescence scenario, the S\\,1063 system was originally a triple system in which the present secondary was the tertiary star. If we assume that the orbital period that we observe now was equal to the period of the original tertiary star, we can use stability arguments to place an upper limit on the period of the original inner binary star. Using the coplanar formulation of Mardling \\&\\ Aarseth (1998) for the stability of a triple system, and adopting a mass ratio of $q = 0.5$ (outer star to inner binary), we find that an original inner binary would have had an orbital period of less than 2.22 days. Such a period is physically permitted, and so that S\\,1063 was formerly a triple system is possible.  Coalescence via multiple mechanisms is plausible on time scales of a few Gyr (St{\\c e}pie\\'n 1995). Four contact binaries are known in M67 (see light curves in van den Berg et al. (2002) and Stassun et al. (2002)), and the old (6 Gyr) cluster NGC 188 has many contact binaries (Baliunas \\&\\ Guinan 1985, Rucinski 1998).  A collision scenario has been suggested by Leonard \\&\\ Linnell (1992) as a possible mechanism for the formation of some blue stragglers in M\\,67. In the low-density environment of an open cluster, collisions of single stars are improbable. However, binary-binary encounters have substantially higher probabilities, and in the course of such encounters the probability of a stellar collision is not negligible. In this scenario, the present S\\,1063 would consist of three of the stars involved in the binary-binary encounter, two having merged into the primary and a third being the present secondary. An eccentric orbit is a natural result of such a resonant encounter. A prediction of either of these merger scenarios is that the resulting star will rotate rapidly. In the coalescence scenario rapid rotation is expected as a result of synchronous rotation prior to the merger. With respect to the collision scenario, the simulations of Sills et al. (2001) show that a collision product is born with a high rotation rate. We argued in Sect. 5 that for aligned rotation and orbital axes, the true equatorial rotation velocity could be as large as 13 km s$^{-1}$, and for a primary radius of 2.4 $R_{\\odot}$ the rotational period could be as short as 10 days. However such a rotation period is still long compared to those expected from mergers or collisions. In the collision scenario, of course, the primary rotation axis and the orbital axis need not be aligned, particularly since tidal circularization has not come to completion. As such, the surface rotation velocity of the primary could be significantly higher. A high inclination angle could also possibly explain a lack of periodic photometric modulation.  Most problematic, these impulsive origins must face the challenge that -- even if the merger product would briefly have the properties of the S\\,1063 primary star -- the thermal time scale of the primary is very short compared to the cluster lifetime. If, for the sake of example, we adopt a primary mass of 0.7 $M_{\\odot}$, the present thermal time scale $E_{\\rm pot}$/$L$ is only 3.4 Myr. A merger product would be expected to readjust to its new mass on such time scales (and perhaps become a present or future blue straggler depending on the combined mass). Given such short time scales, the probability of observing the primary prior to its new equilibrium state is very low. On the other hand, should a star be found shortly after a merger, the presence of residual circumstellar material resulting from the merger might also be expected, and represents a possible explanation for the enhanced reddening of this system. Lastly, if formed recently such a system may not yet be pseudo-synchronized.  We stress that the nature of the secondary is unknown; we know only that it is substantially fainter than the primary at $V$. Thus the secondary is permitted to be a white dwarf. An isolated white dwarf of the cluster age would have a mass of order 0.6--0.7 $M_{\\odot}$ (Wood 1992, Richer et al. 1998); a white dwarf deriving from a prior mass transfer scenario could be much different. A hot source of radiation might help explain the weak H$\\alpha$ emission observed at a velocity different from the primary star velocity. How the secondary could become a white dwarf without circularizing the orbit during its prior post--main-sequence evolution would need to be explained; again, a dynamical exchange might play a role.   {\\bf S\\,1113} In Section 5 we derived  the properties of the stars in S1113 from two independent lines of reasoning, geometric and photometric. The two arguments produced quite different conclusions regarding the luminosity ratio of the stars, or similarly regarding the nature of the primary star. To bring these two lines of reasoning into agreement, we must give up a basic premise of at least one of the arguments. Here we explore two possibilities, and then discuss the measurement uncertainties.  a) Both the geometric and photometric arguments converge on a 0.9 $M_{\\odot}$ main-sequence secondary. This suggests that the resolution of the contradiction in luminosity ratios might be found in the primary star. Under the assumption of membership in M67 the geometric argument must be found wrong, for its implied 4800 K primary star of radius 4.0 $R_{\\odot}$ would be 1.2 mag more luminous in V than observed. The problem cannot be solved with extinction, since the reddening vector for the S1113 primary does not pass through an M67 4.0 $R_{\\odot}$ giant.  Examining the premises of the geometric argument, we first note that relaxing the assumption of alignment of orbital and rotational axes  cannot in itself resolve the problem. Even if the inclination angle of the primary rotation axis is taken to be 90$^{\\circ}$, its radius is reduced only to 3.0 $R_{\\odot}$. (Recall that the primary radius derived from the photometric argument is 2.0 $R_{\\odot}$.)  Supersynchronous rotation of the primary leads to more success. Specifically, a rotation period of 1.4 days - or supersynchronous rotation by a factor of 2 - would bring the primary radii derived from both the geometric and photometric arguments into agreement on a primary radius of 2.0  $R_{\\odot}$.  Such supersynchronism is not expected theoretically given the close proximity of the stars, the primary's large radius, and the surface convection zones on both stars. Unless continuously driven, the duration of a supersynchronous state would be short. Furthermore, supersynchronous rotation of the primary would also require that the observed variability of the composite light with the orbital period must derive from elsewhere in the system. A cool spot on the secondary seems an unlikely origin. Given a secondary/primary V luminosity ratio of 0.35, a 38\\% light modulation from the secondary would be required to reproduce the observed 10\\% variation in the composite light at V.  We note that making the secondary subsynchronous would imply a larger, more luminous secondary which would also resolve the flux ratio discrepancy. However the consequent change in mean density moves the secondary off the main sequence into a domain not occupied by standard stellar evolution (Figure 7). While perhaps this is the case, this path permits essentially unconstrained modeling of the binary.  Finally, the unusual CMD position of S1113 is not easily solved by enhanced extinction of a cluster member. Large extinctions (in excess of 1 mag) along a standard reddening vector would imply a primary star near the top of the M67 main sequence, with much higher effective temperatures than found spectroscopically.   b) Alternatively, we consider arbitrarily discounting the spectroscopically determined flux ratio at V of 0.32 (Sec. 3.1) and adopt the geometric model for the system. This model indicates a secondary/primary flux ratio at V of 0.11. Corresponding decomposition of the composite V magnitude leads to a primary V = 13.88 and a secondary V = 16.31. The geometric model indicates a 0.9 $M_{\\odot}$ main-sequence secondary, and so without considering additional reddening the secondary V magnitude places the binary 1 mag beyond M67. A similar conclusion is reached by comparing the luminosity of a 4800 K, 4.0 $R_{\\odot}$ primary star to the composite light. At this distance the 1.3 $M_{\\odot}$ primary becomes very similar to a star at the base of the M67 giant branch, and as such can be explained by standard stellar evolution theory (albeit not as a cluster member).  The larger primary radius of 4.0  $R_{\\odot}$ implied by the geometric model represents a large fraction of the primary Roche radius. The Roche radii are 4.5 $R_{\\odot}$ and 3.8 $R_{\\odot}$ for the primary and secondary, respectively. An equipotential surface  about the primary whose volume equals that of a 4.0 $R_{\\odot}$ sphere extends 74\\% of the way to the $L_{1}$ point. Thus, within the geometric model, the role of mass transfer in the evolution of S1113 merits consideration. Evidently the 0.9 $R_{\\odot}$ main-sequence secondary lies well within its Roche radius.  Interestingly, both the phasing of the periodic photometric variation and the ultraviolet excess are suggestive of a hot spot near the secondary star powered by an accretion stream. Such spots are formed on the following side of the secondary (e.g., Flannery 1975), in qualitative agreement with the phasing of the light and velocity curves shown in Figure 4. The 10\\% system brightness variation at V requires a mass accretion rate of 2.5 x 10$^{-8}$  $M_{\\odot}$/yr, presuming that the energy is released by a 10,000 K spot on the surface of the secondary star.  We note that at present no spectral signatures of such a hot spot have been seen. Van den Berg et al. (1999) did find broad H$\\alpha$ from S1113, but this emission was kinematically associated with the primary star. They argued that this emission could be consistent with the higher chromospheric activity driven by the rapid rotation of the star, in analogy to V711 Tau. In this context, we note that the photometric variation can also be well modeled by a large cool spot on the primary star. For example, we have reproduced the light variations with the addition of a cool spot in the upper hemisphere of the primary and located 90$^{\\circ}$ from the major axis of the orbit. Specifically the spot properties are: latitude 40$^{\\circ}$, longitude 270$^{\\circ}$, radius 27$^{\\circ}$, temperature 0.82 of the stellar surface temperature. The location of the spot at 270$^{\\circ}$ longitude is motivated by the observations rather than an independent physical argument.  The large extent to which the primary star fills its Roche lobe in this model implies that its shape is significantly asymmetric. Using the Wilson-Devinney formalism we have investigated the expected magnitude of photometric variations due to the anticipated ellipsoidal shape of the primary. Adopting the spot parameters above, a temperature of 4800 K for the primary star, a limb darkening coefficient (linear law) of 0.6 for each star, a gravity-brightening coefficient of 0.32, and a reflection coefficient of 0.5, we find the peak-to-peak ellipsoidal variations to be 0.06 mag in $V$. Most importantly, these ellipsoidal variations produce a double-peaked light curve over the orbital period, in marked contrast to the observed single-peaked light curve. It is possible that these small ellipsoidal variations have gone undetected in contrast to the larger photometric variations of the system. For example, in Fig.~\\ref{s1063:ellips} we show the light curve combining both the cool spot described above and the ellipsoidal variations. Evidently the ellipsoidal variations are lost to inspection in this synthetic light curve, which in fact looks very similar to the observed light curve.  To summarize, the geometric line of reasoning suggests that S1113 is a field binary located behind M67. In this scenario it may be a detached RS CVn whose rapid rotation is producing the several emission diagnostics of enhanced chromospheric activity as well as a large spot. Alternatively, mass transfer may be underway, producing a hot spot in the vicinity of the secondary star. However, to adopt this interpretation of the system, both the kinematic association with M67 and the agreement of the primary mass with the M67 turnoff mass must be taken as merely coincidental. In addition, the measured V flux ratio must be ignored. As we discuss below, we find this last requirement to be a serious counterargument.  c) Measurement uncertainties We have explored whether the discrepancy between the conclusions of the geometric and photometric analyses are the result of our measurement uncertainties. On the photometric side, broadband photometric measurements of S\\,1113 are several and corroborative. The effective temperature for the primary derived from photometric colors is similar to the effective temperature derived from our high-resolution spectra, indicating that both determinations of the effective temperature are reasonable. Finally, we find it unlikely that the bolometric correction for the primary star overestimates the luminosity by a factor 3, particularly given the observed excess flux in the U-band. However, given that only UBV photometry is available, photometry over a wider range of wavelength is much needed.  The geometric luminosity ratio can be brought into accord with the observed V ratio if the effective temperatures and projected rotation velocities of both stars are all adjusted by 1.5 $\\sigma$ in the appropriate senses. While {\\it ad hoc} and improbable, the required adjustments are small enough relative to their uncertainties that further precise measurements of these quantities are in order.  The measured flux ratio at V is pivotal in this discussion. External checks of this quantity on other eclipsing binary systems, using  similar spectroscopic material and with the same techniques used here, have not shown significant systematic errors.  For example, a comparison with independent determinations available from the light-curve solutions of double-lined eclipsing binaries typically agree within 10\\% or better with our spectroscopic determinations (see, e.g., Lacy et al.\\ 1997; 2000).  Thus we have no reason to believe that our flux-ratio measurement for S1113 would be in error by as much as a factor 3. Nonetheless, given the importance of this measurement we encourage additional observational study.  In closing, we note that Hurley et al. (2001) have suggested that S1113 is subluminous due to an evolutionary response to mass transfer. Albrow et al. (2001) identify stars below the subgiant branch of the globular cluster 47 Tuc which they note are similar to S1063 and S1113. They suggest that such systems may result from mass transfer, with the subsequent contraction of the primary Roche radius deflating the primary star and making it less luminous. Our analyses indicate that, considered as a cluster member (i.e., the photometric model), both stars in S1113 presently fall well within their Roche radii. Put another way, if the primary of S1113 fills its Roche lobe (e.g., the geometric model) and has an effective temperature of 4800K, the observed brightness places it behind M67 (and as noted above this conclusion of non-membership cannot be removed via extinction arguments). Thus at least presently S1113 would not seem to be a candidate for a Roche-filling cluster binary."
    },
    "0209/astro-ph0209042_arXiv.txt": {
        "abstract": "Two pseudo-Newtonian potentials, which approximate the angular and epicyclic frequencies of the relativistic accretion disk around rotating (and counter rotating) compact objects, are presented. One of them, the Logarithmically Modified Potential, is a better approximation for the frequencies while the other, the Second-order Expanded potential, also reproduces the specific energy for circular orbits in close agreement with the General Relativistic values. These potentials may be included in time dependent hydrodynamical simulations to study the temporal behavior of such accretion disks. ",
        "introduction": "\\label{sec: I} X-ray binaries are known to be powered by accretion disks around Neutron Stars and Black Holes. The rapid variability of these sources indicate that the X-ray emission arises from the inner accretion disk where the effects of strong gravity are important. High frequency ($\\approx$ kHz) Quasi Periodic Oscillations (QPO) have been observed in neutron star systems (see van~der~Klis~2000 for a review) while  slightly lower frequencies ($\\approx 450$ Hz) QPOs have been detected in black hole systems (Strohmayer 2001). For neutron star systems the kHz QPO tends to be observed in pairs. Associating these frequencies with a Keplerian frequency in the disk, leads to the conclusion that the phenomena originate at radii less than 20 gravitational radius ($r_g \\equiv GM/c^2$).  A number of theoretical ideas have been proposed to explain the phenomenology of kilohertz QPO.  In all these models, one of the two frequencies observed in neutron star systems, is identified as the Keplerian frequency of the innermost orbit of an accretion disk. The sonic point model (Miller et al.~1998) identifies the second frequency as the beat of the primary QPO with the spin of the neutron star, while according to the two oscillators model (Osherovich \\& Titarchuk~1999), the secondary frequency is due to the transformation of the primary (Keplerian) frequency in the rotating frame of the neutron star magnetosphere. On the other hand, Stella \\& Vietri (1999) have proposed a General Relativistic (GR) precession/apsidal motion model, wherein the primary frequency is the Keplerian frequency of a slightly eccentric orbit and the secondary is due to the relativistic apsidal motion of this orbit i.e. the secondary frequency is the Keplerian frequency minus the epicyclic one. These models in general are based on identifying the characteristic frequencies of the system with observed ones and often do not address the issue of how such oscillations occur in the accreting flow.  A complete understanding of the QPO phenomena would require a self consistent hydrodynamical simulation of the accreting flow in general relativity.  While such an ambitious endeavor has been impeded for several reasons, the main difficulties can be identified to be (a) the development of a self-consistent turbulent viscosity and (b) the inclusion of GR effects. In hydrodynamical simulations, turbulent viscosity has typically been introduced in a parametric form like the $\\alpha$-parameterization (e.g. Taam \\& Lin 1984). Since the temporal behavior of accretion disks is expected to depend on the form of the viscosity law, the results of such simulations were not conclusive. A promising mechanism for driving the turbulence responsible for angular momentum and energy transport is the action of the magneto-rotational instability (MRI) that is expected to take place in such disks (Balbus \\& Hawley 1991). Recent 3D magneto hydrodynamical (MHD) simulations have shown that indeed the MRI can give rise to a turbulent viscosity which leads to the accretion flow in a Keplerian disk (Hawley, Balbus \\& Stone 2001). While presently such simulations do not include radiation (and hence do not describe optically thick accretion flow), it is expected that self-consistent simulations will be possible in the near future and the temporal behavior of accretion disks can be studied with confidence.  Despite these recent advances, it is still extremely difficult to simulate realistic accretion flows in a complete GR framework. However, relativistic effects may be approximately simulated by using modified Newtonian (or Pseudo-Newtonian) potentials in the non-relativistic radial-momentum equation. Paczy\\'nski \\& Wiita (1980) proposed such a pseudo-Newtonian potential which has been frequently used in simulations (e.g. Milsom \\& Taam 1997; Hawley \\& Balbus 2002). Here the Newtonian potential has been replaced by $\\phi = GM/(r-2r_g)$. The attractive feature of the potential is that it reproduces the last stable orbit exactly and the specific energies of circular orbits within $10\\%$ of the GR values (i.e. for Schwarzschild geometry). Several other pseudo-Newtonian potentials have been proposed and used in the literature (e.g. Chakrabarti \\& Khanna 1992). Artemova et al. (1996) have considered several such potentials and concluded that the Paczy\\'nski-Wiita potential is better than the rest based on the above criteria for non-rotating compact objects. Recently, Mukhopadhyay (2002) has proposed a pseudo- potential which is valid for rotating compact objects. This potential reproduces the GR values of last stable orbit exactly and is a good approximation ($< 10\\%$ error) for the specific energy at last stable circular orbit in case of Kerr geometry. It also reduces to the Paczy\\'nski-Wiita potential when the spin of the black hole is set to zero.  However, these potentials are not a good approximation (with error $ > 50\\%$) for the angular and epicyclic frequencies for radii $ < 20 r_g$. Thus, while they are adequate to approximate the relativistic effects for a steady state accretion disk, they can not quantitatively reproduce the temporal behavior of a disk since that is expected to depend on the disk's characteristic (i.e. the angular and epicyclic) frequencies. Nowak \\& Wagoner (1991) have proposed a potential for a non rotating black hole which reproduces the Keplerian frequencies (with deviations $<15\\%$) and the epicyclic frequencies (with deviations less than $45\\%$) and hence is better than the Paczy\\'nski-Wiita potential for such applications.  In this paper, we present two pseudo-Newtonian potentials which may be used to simulate the relativistic time varying effects in accretion disks around a compact object that may be co-rotating or counter-rotating with respect to the disk with the  spin parameter $a < 0.99$. For faster spin rates the predictions of these potentials deviate pronouncedly (with errors $> 200\\%$) and hence are no longer a good approximation. The first has been named the {\\it Second-order Expansion Potential (SEP)} since it contains terms up to $(r_{ms}/r)^2$, where $r_{ms}$ is the marginally stable orbit. This potential reproduces the specific energy  and the angular frequency with deviations $<10\\%$ and $< 25\\%$, respectively from GR values (i.e. for Kerr geometry). The deviations in epicyclic frequency range from $25-170\\%$ (for $a\\leq 0.9$) depending on the spin rate of the compact object. When the object is not rotating, the potential reduces to the one proposed by Nowak \\& Wagoner (1991). The second has been named {\\it Logarithmically Modified Potential (LMP)} since it contains a logarithmic term. This potential reproduces well the angular (with deviations $< 20\\%$ for co-rotating and $<40\\%$ for counter-rotating flows) and epicyclic frequencies (with deviations $< 60\\%$) but predicts specific energies which are around $30\\%$ different from the GR values. ",
        "conclusions": "\\label{sec: III}  In this work, we have presented two pseudo-Newtonian potentials which approximate the general relativistic effects on an accretion disk around rotating compact objects. These two potentials are designed particularly to approximate the angular and epicyclic frequencies of the accretion disk as seen by an observer at infinity. Table 1 summarizes the results by comparing the maximum percentage deviations from relativistic values (in Kerr geometry) for the two potentials and comparing them with those of another standard pseudo-potential.  The SEP not only approximates the frequencies well, but also the specific energies for circular orbits turn out to be remarkably close to the relativistic values. Thus based on such criteria, this potential is better than other pseudo-Newtonian potentials given in the literature and can be used to simulate both the steady state and time varying accretion disks. The LMP while being a better approximation to the frequencies than SEP, gives rather large ($\\approx 30\\%$) deviation from the GR results for the specific energies. Hence its utility is perhaps limited to the time-dependent studies of accretion disks.  Which one of these two potentials should be used in a hydrodynamical simulation depends on problem being addressed. Acoustic waves (which depend on the epicyclic frequencies) would perhaps be better simulated by the LMP while the SEP may be more suited for the long term temporal behavior (which may depend also on the energy dissipation). Moreover, a temporal behavior detected in a simulation could be an artifact of the pseudo-Newtonian potential rather than true GR effects. Hence, it will be prudent to confirm the behavior using both the potentials. Since the mathematical forms of the two potentials are quite different any temporal behavior detected for both the potentials would imply that the behavior is indeed due to relativistic effects. Use of these potentials in hydrodynamical simulations of accretion disk will help in the understanding of relativistic effects and may serve as a guideline for advanced simulations in general relativity."
    },
    "0209/astro-ph0209274_arXiv.txt": {
        "abstract": "A variety of observations indicate that the universe is dominated by dark energy with negative pressure, one possibility for which is a cosmological constant.  If the dark energy is a cosmological constant, a fundamental question is: Why has it become relevant at so late an epoch, making today the only time in the history of the universe at which the cosmological constant is of order the ambient density.  We explore an answer to this question drawing on ideas from unimodular gravity, which predicts fluctuations in the cosmological constant, and causal set theory, which predicts the magnitude of these fluctuations.  The resulting ansatz yields a fluctuating cosmological ``constant'' which is always of order the ambient density. ",
        "introduction": "The most startling discovery to emerge from the recent plethora of cosmological data is that the universe appears to be dominated by dark energy~\\cite{Concordance,SN,CMB}.  We know that this dark energy accounts for roughly seventy percent of the energy density in the universe, does not cluster like ordinary matter, and has negative pressure. Otherwise, we are in the dark about the nature of this extraordinary phenomenon.  Perhaps the most popular explanation is that the dark energy is due to a cosmological constant, for such a parameter was introduced into general relativity at its birth~\\cite{Einstein} and has remained an important tool for cosmologists seeking to model the observed universe~\\cite{age}. The strongest argument against the cosmological constant is that naively we expect it to contribute an energy density, $\\rho_\\Lambda$, of order $m_p^4 = (8\\pi G)^{-2}$, where $G$ is Newton's constant and $m_p$ is the reduced Planck Mass\\footnote% {In this paper, we will use units in which $\\hbar=c=m_p=1$.}. This estimate is some one hundred and twenty orders of magnitude larger than the observed value.  An equivalent way of stating the problem is to note that only today is the cosmological constant of order the ambient density in matter or radiation.   At all past epochs, $\\rho_\\Lambda$ was sub-dominant and immensely so. Many people have felt that no theory could naturally predict such a tiny value for $\\Lambda$ (or equivalently such a late epoch for it to become relevant) without predicting $\\Lambda$ to vanish entirely, and for this reason they have sought other explanations of the observations.   Many alternatives to the cosmological constant have been proposed.  Most significant among these are quintessence models in which the dark energy is due to a homogeneous scalar field shifted away from the true minimum of its potential \\cite{quintessence}. Like a simple cosmological constant, many of these suffer from the ``Why Now?'' problem: Why does the quintessence field come to dominate only recently? They also typically need to explain the small mass scale necessary for the field to be important today ($m \\la 10^{-33}$ eV).   Even more disturbing, none of them are connected to realistic particle physics models.  Perhaps then, instead of altering the energy content of the universe, we need to look in another direction and modify gravity in order to explain the dark energy today.   The simulations reported here flesh out an old heuristic prediction~\\cite{LamPred} of a fluctuating cosmological term arising from the basic tenets of causal set theory.\\footnote% {For an introduction to the causal set hypothesis see~\\cite{prlcs}.} In normal usage, the words ``cosmological term'' refer to a contribution to the effective stress-energy-momentum tensor of the form $T_{\\mu \\nu} = g_{\\mu \\nu} \\Lambda(x)$. However, in classical General Relativity (GR) such a $\\Lambda(x)$ must be constant if the total energy momentum in other matter components is separately conserved.  Here we consider a specific modification of GR motivated by the search for a theory of quantum gravity based on causal sets.   Although the ultimate status and precise interpretation of the prediction of a fluctuating $\\Lambda$ must await the development of a quantum dynamics for causal sets~\\cite{dprrds}, the basic lines of the argument are simple and general enough that they have a certain independence of their own. In this paper we review the motivation for a fluctuating $\\Lambda$ from causal set theory, propose an ansatz for the form of these fluctuations, apply the latter to the Friedmann equation with time-dependent cosmological term, and find that we can have a viable cosmology for some fraction of the solutions.  Finally, we address issues related to our choice of evolution equations. ",
        "conclusions": "It is still too early to understand the full implications of recent cosmic discoveries that point to dark energy in the universe. A number of possibilities have previously been explored in detail, including a non-zero cosmological constant $\\Lambda$ and zero $\\Lambda$ with dark energy hidden in a scalar field.   It is also possible, though, that the measurements are telling us that we need to modify our understanding of space and time. In particular, the notion that space-time is continuous may be simply an approximation that breaks down on scales as small as the Planck scale. If so, drawing on ideas from causal set theory -- which postulates a discrete space-time -- and unimodular gravity, we have shown that the cosmological ``constant'' need not be a fixed parameter.  Rather, it arguably fluctuates about zero with a magnitude $1/\\sqrt{\\spv}$, $\\spv$ being some measure of the past four volume.  The amplitude of these fluctuations is then of the right order of magnitude to explain the dark energy in the universe. This argument is so general that it would apply at all times, and, indeed, we expect the energy density in the cosmological ``constant'' to always be of order the ambient density in the universe.   In \\S IV, we presented a number of issues which inevitably will confront anyone wishing to implement this idea. Until these issues are resolved, it will be difficult to make unambiguous, robust predictions. Nevertheless, one can already see that this theory of a fluctuating $\\Lambda$ differs significantly from most other solutions to the dark energy problem. Most important for its testability is the notion that it may have affected the evolution of the universe at early times. Thus, the primordial generation of perturbations during a possible inflationary phase;  production of light elements in Big Bang Nucleosynthesis; acoustic oscillations in the background radiation; and the evolution of structure at more recent times all may yield clues and tests of the idea of an everpresent $\\Lambda$.   This work was supported by the DOE at Fermilab, by NASA grant NAG5-10842 and by NSF Grant PHY-0079251.  It was also supported at Syracuse University by NSF Grant PHY-0098488 and by an EPSRC Senior Fellowship at Queen Mary College, University of London.  SD and RDS would like to acknowledge the hospitality of the Aspen Center for Physics where their collaboration began, and RDS would like to acknowledge the hospitality of Goodenough College, London, where part of this work was done.  PBG would like to acknowledge useful conversations with J. Moffatt at the University of Toronto and the hospitality of the Canadian Institute for Theoretical Astrophysics while there.   \\newcommand\\spr[3]{{\\it Physics Reports} {\\bf #1}, #2 (#3)} \\newcommand\\sapj[3]{ {\\it Astrophys. J.} {\\bf #1}, #2 (#3) } \\newcommand\\sprd[3]{ {\\it Phys. Rev. D} {\\bf #1}, #2 (#3) } \\newcommand\\sprl[3]{ {\\it Phys. Rev. Letters} {\\bf #1}, #2 (#3) } \\newcommand\\np[3]{ {\\it Nucl.~Phys. B} {\\bf #1}, #2 (#3) } \\newcommand\\smnras[3]{{\\it Monthly Notices of Royal Astronomical Society} {\\bf #1}, #2 (#3)} \\newcommand\\splb[3]{{\\it Physics Letters} {\\bf B#1}, #2 (#3)}"
    },
    "0209/astro-ph0209104_arXiv.txt": {
        "abstract": "We report results of a sensitive search for cold dust and molecular gas in the disks around 8 T~Tauri stars in the high-latitude cloud MBM~12.  Interferometric observations of 3~mm continuum emission in 5 fields containing 6 of the objects, and literature values for the remaining two, limit the disk masses to $M_{\\rm disk}<0.04$--0.09~M$_\\odot$ (gas+dust), for a gas:dust mass ratio of 100 and a distance of 275~pc. By coadding the 3~mm data of our five fields, we set an upper limit to the average disk mass of $\\bar M_{\\rm disk}(N=5)<0.03$~M$_\\odot$.  Simultaneous observation of the CS $J$=2--1 and the N$_2$H$^+$ 1--0 lines show no emission. Single-dish observations of the $^{13}$CO 2--1 line limit the disk mass to (5--10)$\\times 10^{-4}$ M$_\\odot$ for a standard CO abundance of $2\\times 10^{-4}$. Depletion of CO by up to two orders of magnitude, through freezing out or photodissociation, can reconcile these limits. These mass limits lie within the range found in the Taurus-Auriga and $\\rho$~Oph star-forming regions (0.001--0.3~M$_\\odot$), and preclude conclusions about possible decrease in disk mass over the 1--2 Myr age range spanned by the latter two regions and MBM~12. Our observations can exclude the presence in MBM~12 of T~Tauri stars with relatively bright and massive disks such as T~Tau, DG~Tau, and GG~Tau. ",
        "introduction": "}  Characterizing the disks of young stars is an outstanding challenge in star- and planet-formation research. Millimeter emission from cold dust and molecular gas has revealed rotating disks a few hundred AU in radius (\\citealt{mannings:haebe,mannings:haebe2}; \\citealt{dutrey:tts13co,dutrey:gmaur}; \\citealt*{dutrey:dmtau+ggtau}; \\citealt*{simon:ttsmass}) around T~Tauri and Herbig Ae/Be stars, while millimeter and infrared observations have shown the mass in small dust particles to decrease gradually from the classical-T~Tauri to late pre--main-sequence stage (\\citealt*{robberto:sf99}; \\citealt{rayjay:mirtwa, carpenter:ic348}). Evidence suggests that the age range of 1--3~Myr is pivotal in the disk's evolution, when small dust in the inner disk regions starts to clear out \\citep{haisch:freq_clusters, luhman:mbm12, hartmann:agespread}. Some of the dust may have coagulated into larger, centimeter-sized objects, which are invisible to (sub-) millimeter or infrared observations. The fate of the gas that makes up 99\\% of the disk's mass is only now being investigated \\citep{thi:h2_ggtau,duvert:exdisk,thi:h2_debris,thi:h2_co_disks,herczeg:h2twa,richter:h2disk}. This paper investigates the dust- and gas-content of the disks around several young stars in the MBM~12 cloud.  The cloud MBM~12 (distance 275~pc; but see below) is only one of two high-latitude clouds known to harbor young stellar objects (YSOs); the other being MBM~20 \\citep*{sandell:l1642}. \\citet{hearty:mbm12+20} identified 8 late-type young stars in MBM~12, and two main-sequence stars with unclear relation to the cloud. \\citet{luhman:mbm12} found four additional pre--main-sequence members from 2MASS data, and estimated the cluster's age at $2^{+3}_{-1}$~Myr. \\citet{rayjay:mbm12_mir} detected mid-infrared excess toward six of the original eight objects. Based on their H$\\alpha$ equivalent line widths, these objects are classified as six classical T~Tauri stars and two `weak-line' T~Tauri stars (Table \\ref{t:coords} and Fig. \\ref{f:map}).  Their frequency of K- (20\\%) and L-band infrared excesses (70\\%) suggests clearing of their disks \\citep{luhman:mbm12}. A crucial but unresolved question is how much material, if any, still resides in more extended, outer disks. Thus far, upper limits to the millimeter continuum of only two of the T~Tauri stars (E02553+2018 and LkH$\\alpha$~264) have been reported \\citep{pound:mbm12}.  The group of young stars in MBM~12 is among several nearby `isolated' groups that have recently been identified and investigated for the nature of their disks \\citep{rayjay:newblock,rayjay:youngstars}. Some of these groups (the TW~Hydrae Association (TWA) at $\\sim 55$~pc, \\citealt{kastner:twa}; and $\\eta$ Chamaeleontis cluster at $\\sim 97$~pc, \\citealt*{mamajek:etacha}) are far from obvious parent molecular clouds. This suggests ages for these stars of 5--10~Myr, and little circumstellar disk material is expected. The association of the MBM~12 stars with cloud material supports their younger derived age, and they may be represent an earlier epoch of groups like TWA and $\\eta$~Cha.  Until recently, MBM~12 was thought to be the nearest star-forming cloud to the Sun, at $\\sim 65$~pc \\citep{hearty:mbm12_rosat}. New evidence suggests it may be significantly further \\citep{luhman:mbm12}, at $\\sim 275$~pc. \\citet*{idzi:mbm12_aas} found extinction at 65, 140 and 275~pc along the MBM~12 line of sight, so the distance of the stars is unclear. We follow \\citet{luhman:mbm12} and adopt 275~pc, which gives the most plausible location of the stars on the the Herzsprung-Russel Diagram. In any case, the young age indicated by MBM~12's association with cloud material and its relative proximity make it an excellent target to study circumstellar disks in detail.  This paper presents the results of sensitive millimeter-wave searches for cold dust and molecular gas associated with the six classical T~Tauri stars in MBM~12 known to harbor disks for which no millimeter data exist, and the two weak-line T~Tauri stars (Table \\ref{t:coords}). We did not include the newly identified members of MBM~12 \\citep{luhman:mbm12}, but fortuitously include the edge-on disk source MBM~12~3C \\citep{rayjay:mbm12_ao} which happened to fall in our field containing LkH$\\alpha$~262 and LkH$\\alpha$~263. Since the completion of these observations, \\citet{luhman:mbm12} has concluded that one object, RXJ0306.5+1921, is likely an older interloper. For completeness, we still report the data on this object.  We used a millimeter interferometer to measure the continuum emission, because the high spatial resolution avoids confusion with the surrounding cloud. We use single-dish submillimeter observations of $^{13}$CO $J$=2--1 to trace any cold ($\\sim 16$ K) and relatively low density ($n_{\\rm H_2}\\approx 10^4$ cm$^{-3}$) gas that may reside around the objects. Section \\ref{s:obs} describes the observations and section \\ref{s:results} the resulting mass limits. Section \\ref{s:discussion} discusses these limits in the context of other nearby star-forming regions and young associations. Section \\ref{s:summary} concludes the paper with a short summary. ",
        "conclusions": "}  Section \\ref{s:results} places upper limits on the mass of cold dust and gas associated with eight T~Tauri stars in MBM~12. The 3~mm continuum limits the mass to $<0.04$--0.09~M$_\\odot$. The $^{13}$CO lines provide limits that are much lower, (5--10)$\\times 10^{-4}$~M$_\\odot$. However, the abundance of molecules such as CO may be decreased as commonly observed in T~Tauri disks (e.g., \\citealt*{dutrey:ggtau}; \\citealt{dutrey:tts13co}), possibly by freezing out onto dust grains and by photodissociation by (inter-) stellar ultraviolet photons as suggested by chemical models (e.g., \\citealt{aikawa:chem2d}; \\citealt{willacy:photodisk}). These processes can easily reduce the CO abundance by factors of tens or a hundred, in which case the limits of Table \\ref{t:lines} are no longer lower than those of Table \\ref{t:cont}. We stress that if CO is significantly frozen out, the gas mass of the disk does not change appreciably, since it is dominated by undepleted H$_2$. If CO is significantly photodissociated, the total molecular gas mass may be reduced if a sizable fraction of H$_2$ is also photodissociated.  \\citet{rayjay:mbm12_mir} report N-band excess in the six classical T~Tauri stars, indicating the presence of material close to the stars. This places a lower limit to the disk mass of $\\sim 10^{-5}$~M$_\\odot$, well below our upper limits.  The edge-on disk source MBM~12A~3C near LkH$\\alpha$~263 \\citep{rayjay:mbm12_ao} requires a mass of $\\sim 2\\times 10^{-3}$~M$_\\odot$ to explain its scattered light image. This mass is comfortably bracketed by our lower and upper limits. \\citeauthor{rayjay:mbm12_ao} model the scattered light with a distribution of dust sizes. At the observing wavelength of 3~mm, our observations are sensitive to particles much larger than those doing the infrared scattering, but both populations are connected through the adopted dust-size distribution and mass emissivity coefficients, and therefore refer to a similar mass reservoir.  With the obtained limits, the disk masses in MBM~12 are indistinguishable from those found in Taurus-Auriga and $\\rho$~Oph of 0.001--0.3~M$_\\odot$ \\citep{beckwith:mmdisks, osterloh:mm,andre:rhoophmm}. We therefore cannot draw any conclusions about the evolution of disks in the 1--2 Myr age range spanned by these regions and MBM~12. While the work by \\citet{luhman:mbm12} suggests that the dust in the disks in MBM~12 is starting to clear out, more sensitive measurements of the dust continuum with, e.g., SCUBA or SIRTF, are needed to investigate the fate of the colder dust at larger radii. \\citet{carpenter:ic348} limits the disk masses of the members of the 2~Myr cluster IC~348 to $<0.025$~M$_\\odot$ (or 0.002~M$_\\odot$ averaged over his 95 sources), somewhat below our limits on MBM~12.  Our mass limits do exclude the presence of objects such as T~Tau, GG~Tau, and DG~Tau with large millimeter fluxes \\citep{beckwith:mmdisks}. We would easily have detected these fluxes, at 20--40~mJy when scaled to 275~pc and extrapolated to 3~mm (adopting spectral indices between 2.5 and 4). But even in Taurus-Auriga these objects are rare, and given the small number of T~Tauri stars in MBM~12 ($\\sim 10$; \\citealt{luhman:mbm12}) no such bright objects would necessarily be expected.   }  We obtained upper limits on the 3~mm continuum flux and $^{13}$CO 2--1 line intensity of eight T~Tauri stars in the MBM~12 region. These limits constrain the disk masses to $<0.04$--0.09 M$_\\odot$ (gas+dust), not inconsistent with the distribution of masses in slightly younger regions like Taurus-Auriga and $\\rho$~Oph ($\\sim 1$~Myr vs.\\ $\\sim 2$~Myr), and consistent with the mass of $2\\times 10^{-3}$ M$_\\odot$ derived for the edge-on disk of MBM~12A~3C.  We exclude the presence of objects such as T~Tau, GG~Tau, and DG~Tau with bright millimeter emission. More sensitive searches with, e.g., SCUBA and SIRTF, will probe the evolution of the cold disk material at larger radii at the moment when the inner disks start to clear."
    },
    "0209/astro-ph0209618_arXiv.txt": {
        "abstract": "We present a numerical method for solving the Poisson equation on a nested grid.  The nested grid consists of uniform grids having different grid spacing and is designed to cover the space closer to the center with a finer grid. Thus our numerical method is suitable for computing the gravity of a centrally condensed object. It consists of two parts: the difference scheme for the Poisson equation on the nested grid and the multi-grid iteration algorithm. It has three advantages: accuracy, fast convergence, and scalability.  First it computes the gravitational potential of a close binary accurately up to the quadraple moment, even when the binary is resolved only in the fine grids. Second residual decreases by a factor of 300 or more by each iteration. We confirmed experimentally that the iteration converges always to the exact solution of the difference equation.  Third the computation load of the iteration is proportional to the total number of the cells in the nested grid.  Thus our method gives a good solution at the minimum expense when the nested grid is large. The difference scheme is applicable also to the adaptive mesh refinement in which cells of different sizes are used to cover a domain of computation. ",
        "introduction": "Astronomical objects such as stars, clouds, and galaxies have enormous dynamic range both in density and in size.  To illustrate this enormous dynamic range, we consider star formation as an example.  Stars form in molecular clouds of which the mean density is 10$^3$ atoms~cm$^{-3}$.  The molecular clouds contain condensations named molecular cloud cores, from which stars form owing to the self-gravity.  The molecular cloud cores have typical size of 10$^{17}$ cm and typical density of 10$^5$ atoms~cm$^{-3}$. On the other hand, the central density of a star is 10$^{11}$ atoms cm$^{-3}$ at the very beginning of its protostellar stage and the present Sun has the central density of 10$^{26}$ atoms~cm$^{-3}$.  The radius of a protostar is 10$^{14}$ cm when the central density is 10$^{12}$ atoms~cm$^{-3}$.  As the central density increases, the radius decreases down to 10$^{10}$ -- 10$^{12}$ cm until the star reaches its main sequence (hydrogen burning) stage.  This enormous dynamic range restrict us to achieve high spatial resolution only in the small regions of interest.  To generate finer grids in the region of interest, people have developed various mesh generation methods. Adaptive mesh refinement (AMR) and nested grids (NG) are typical of the mesh generation methods developed in the past decade. AMR and NG generate finer grids hierarchically in the region of interest.  AMR was invented by \\citet{berger84} and has been advanced by many researchers. NG is a variant of AMR and generates only one sub-grid at each hierarchical level, while AMR has no restriction on the number of sub-grids. AMR and NG have succeeded in simulations of star formation and galaxy formation in which compact objects form by condensation of diffuse clouds.  Some recent numerical simulations on star formation and cosmology apply either AMR or NG to achieve high spatial resolution.  \\citet{truelove97} studied gravitational collapse of a molecular cloud core with AMR to resolve fragmentation of the highly condensed cloud core. Since then AMR is frequently used in numerical simulations of fragmentation during gravitational collapse \\citep{truelove98,boss00}. Using NG \\citet{burkert93,burkert96} studied fragmentation of a centrally condensed protostar.  They succeeded in resolving spiral arms formed by the self-gravitational instability in the protoplanetary disk. Using NG \\citet{tomisaka98} computed gravitational collapse of rotating magnetized gas cloud and found magnetohydrodynamical driven outflow emanating from a very compact central disk. \\citet{norman99} has reviewed application of AMR to cosmological simulations.  Though simulations based on AMR and NG are successful, some technical problems still remain for AMR and NG. One of them is a numerical algorithm for solving the Poisson equation on a grid in which a cell faces several smaller cells. In other words, difficulty arises at boundaries between the regions covered with different size cells. For later convenience we name these boundaries the grid level boundaries. In most AMR and NG of three dimensions, a parent cell faces four child cells at each grid level boundary in case of three dimensions.  When cells are uniform on a grid, a simple central difference scheme gives us second order accuracy and the difference equation can be solved fast with the multi-grid iteration or some other methods.  Since the simple central difference breaks down at the grid level boundary, we need some modifications at the boundaries.  An ideal scheme should provide an accurate solution and require minimum computation load. As shown later, some schemes published in recent literature lack in accuracy.  One might think that we could obtain a good solution by relatively small computation cost if we would solve the Poisson equation successively from coarse grids to fine grids. One can solve the Poisson equation on the coarsest grid if an appropriate boundary is given at the boundary of the grid.  Then the solution gives boundary conditions for the sub-grid if we interpolate it appropriately. Thus we can solve the Poisson equation on the sub-grid and repeat the same procedure down to the finest grid. This algorithm requires rather small computation load but the solution is not accurate enough.  To illustrate the problem of the algorithm mentioned above, we consider binary or multiple stars of which separation is too short to resolve in a coarse grid.  For simplicity we assume that the binary or multiple stars dominate the gravity.  At a large distance from them, the gravity is approximately the sum of the point mass gravity and the quadraple moment of gravity.  If we compute the gravity with a coarse grid without any knowledge on the fine structure, we will miss the quadraple moment of gravity.   Although it is smaller than the point mass gravity, the quadraple moment of gravity is the leading term in the gravitational torque.  We need to take account of mass distribution in a fine grid to evaluate the gravitational torque.  In this paper we present a difference equation for the Poisson equation on a nested grid.   The solution of this difference equation is accurate enough in the sense that it reproduces the quadraple moment of gravity quantitatively even in a coarse grid.  Moreover the difference equation can be solved fast with the multi-grid iteration.   The computation load is scalable, i.e., proportional to the total number of cells involved in the nested grid.  The above difference equation is also applicable to AMR. The difference equation is likely to be solved fast also with a multi-grid iteration.  In \\S 2 we describe our difference equation.  In \\S 3 we denote our algorithm for solving the difference equation. In \\S 4 we show the performance of our scheme with emphasis on the accuracy of the solution and speed of computation. In \\S 5 we discuss reason for success of our difference equation in reproducing the quadraple moment of gravity.  We also discuss extension of our method to AMR. ",
        "conclusions": "As shown in the previous section, our numerical method provides an accurate solution with a reasonably small computation cost.  The computation load is scalable in the sense that it is proportional to the number of the cells contained in the nested grid. Our discrete Poisson equation is robust in the sense that it can be applied also to AMR as far as a parent cell is subdivided into two in one dimension.  We discuss the strength of our scheme while comparing with other schemes given in literatures.  \\citet{suisalu95} took another approach when solving the Poisson equation in their AMR computation. They adopted an cubic interpolation formula at the grid level boundary to ensure the continuity of the solution.  They have noticed that a linear interpolation ensures only the continuity in the potential but not that in the gravitational force.  Discontinuity in the force may cause a spurious feature across the grid level boundary. Although not mentioned explicitly in \\citet{suisalu95}, the discontinuity causes a more  serious problem; it introduces spurious mass on the grid level boundary. If the gravity $ \\mbox{\\boldmath$g$} $ has different values at a grid level boundary, the solution does not satisfies the Gauss's theorem. The spurious mass on the grid boundary gives a serious error on the solution in a long range.  The error decreases only inversely proportional to the distance from the spurious mass. This error dominates over the gravitational torque which decreases more steeply.  Compared with adoption of a higher order interpolation formula at the grid level boundary, our scheme has several advantages. First our approach ensures absence of a spurious mass on the grid level boundary while it is not guaranteed in a higher oder interpolation formula.  Second higher order interpolation formula increases computation load.  Third higher order interpolation may slow down efficiency of convergence. As far as in our experience, a simple multi-grid iteration does not work when Equations (\\ref{interpolation1}) and (\\ref{interpolation2}) are replaced with a higher order interpolation formula.  We discuss the third point further in the following.  Our discrete Poisson equation has the following favorable character.  When it is rewritten in the form, \\begin{equation} \\rho _i \\; = \\; \\sum _j a _{i,j} \\phi _j \\; , \\end{equation} the coefficient $ a _{i,j} $ is always positive except for $ i $ = $ j $.  Here symbols $ i $ and $ j $ denote the cell number after sorted in one dimension.  This property ensures that the Gauss-Seidel iteration converges always since it always underestimates the correction.  In other words, the convergence of the Gauss-Seidel iteration is ensured since the matrix $ a _{i,j} $ is diagonally dominant, \\begin{equation} \\sum _{i\\ne j} \\vert a _{i,j} \\vert \\; \\le \\; \\vert a _{i,i} \\vert \\; . \\label{Ddominance} \\end{equation} When we use a higher order interpolation formula, this the dominance of the diagonal element [Eq. (\\ref{Ddominance})] is lost. The coefficient $ a _{i,j} $ is no longer always positive.  Then the Gauss-Seidel iteration may overestimate the correction and may not converge.  If we apply successive under-relaxation, the convergence slows down.  \\citet{ricker00} have proposed another idea for solving the Poisson equation in a nonuniform grid. They have used a nonuniform grid in which the spacing in a certain direction varies but only in the direction. For example, the spacing in the $ x $-direction varies with $ x $ but not in the $ y $- and $ z $-directions. Consequently each cell is rectangular and may have different side lengths.  Each cell has a neighboring cell in each direction and faces 6 neighboring cells in total except for a cell located on the grid.  They evaluated the gravity on the cell surface using the second order interpolation.  Their Poisson equation can be rewritten as \\begin{equation} \\frac{g _{x,i+1/2,j,k} \\, - \\, g _{x,i-1/2,j,k}} {\\Delta x _i} \\, + \\, \\frac{g _{y,i,j+1/2,k} \\, - \\, g _{y,i,j-1/2,k}} {\\Delta y _j} \\, + \\, \\frac{g _{z,i,j,k+1/2} \\, - \\, g _{z,i-1/2,j,k-1/2}} {\\Delta z _k} \\, = \\; 4 \\pi G \\rho _{i,j,k} \\; , \\label{ricker} \\end{equation} where \\begin{equation} g _{x,i+1/2,j,k} \\; = \\; F _{2,i} \\phi _{i+2,j,k} \\, + \\, F _{1,i} \\phi _{i+1,j,k} \\, + \\, F _{0,i} \\phi _{i,j,k} \\; \\end{equation} \\begin{eqnarray} F _{2,i} & = & \\frac{2 \\, (\\Delta x _i \\, - \\, \\Delta x _{i+1})} {(\\Delta x _i \\, + \\, \\Delta x _{i+1} \\, + \\, \\Delta x _{i+2}) \\, (\\Delta x _{i+1} \\, + \\, \\Delta x _{i+2})} \\, \\\\ F _{1,i} & = & - \\, \\frac{2 \\, [\\Delta x _i ^2 \\, - \\, 3 \\Delta x _{i+1} \\, ( \\Delta x _{i+1} \\, + \\, \\Delta x _{i+2}) \\, - \\, \\Delta x _{i+2} ^2]} {(\\Delta x _i \\, + \\, \\Delta x _{i+1} \\, + \\, \\Delta x _{i+2}) \\, (\\Delta x _{i+1} \\, + \\, \\Delta x _{i+2}) \\, (\\Delta x _i \\, + \\, \\Delta x _{i+1})} \\; , \\\\ F _{0,i} & = & \\frac{2 \\, (2 \\Delta x _{i+1} \\, + \\, \\Delta x _{i+2})} {(\\Delta x _i \\, + \\, \\Delta x _{i+1} \\, + \\, \\Delta x _{i+2}) \\, (\\Delta x _i \\, + \\, \\Delta x _{i+1})} \\; . \\end{eqnarray} The symbols, $ \\Delta x _i $, $ \\Delta y _j $, and $ \\Delta z _j $ denote the side length of each  cell. We omitted the interpolation formula for $ g _y $ and $ g _z $ to save space. Integrating Equation (\\ref{ricker}) over the whole volume, we obtain \\begin{equation} \\sum _{i,j,k} \\rho _{i,j,k} \\, \\Delta x _i \\, \\Delta y _j \\, \\Delta z _k \\; = \\; \\sum _{i,j,k} H _{i,j,k} \\phi _{i,j,k} \\; , \\end{equation} where $ H _{i,j,k} $ does not vanish in a nonuniform grid. This means that $ \\phi _{i,j,k} $ is not necessary to vanish even when $ \\rho _{i,j,k} $ = 0.  In other words Equation (\\ref{ricker}) has multiple solutions for a given density distribution and boundary conditions.  This is a serious problem.  As shown above, the approach based on a higher order interpolation formula has some fundamental problems.   Although our scheme is only the first order accurate at the grid boundaries, it give a quantitatively good solution.  This is likely to be due to the fact that our difference scheme satisfies global conditions for the proper Poisson equation to satisfy.  First our scheme satisfy the Gauss's theorem as mentioned repeatedly. Second our scheme ensures the Stokes' theorem, \\begin{equation} \\oint \\mbox{\\boldmath$g$} \\cdot d\\mbox{\\boldmath$s$} \\; = \\; 0 \\; . \\label{stokes} \\end{equation} Equation (\\ref{interpolation1}) is equivalent to \\begin{equation} \\phi (\\mbox{\\boldmath$r$} _1) \\; = \\; \\phi (\\mbox{\\boldmath$r$} _2) \\; + \\; \\int _1 ^2 \\, \\mbox{\\boldmath$g$} \\cdot d\\mbox{\\boldmath$r$} \\; . \\end{equation}  We adopted a principle that the difference equations should be consistent with the global properties, i.e., the Gauss's theorem and Stoke's theorem. We were satisfied with the first order accuracy at the grid level boundaries.  In other words we gave priority to the global properties over the higher order accuracy at a given point. As a result we succeeded in computing the gravity of a close binary and in reproducing the quadraple moment. This success is analogous to that of a Total Variation Diminishing (TVD) scheme for wave equations and that of symplectic integrator for a Hamiltonian system \\citep{yoshida90}. TVD scheme gives priority to monotonicity of the solution over the local accuracy \\citep[see, e.g.][]{hirsch90}. The solution is free from numerical oscillations.  A symplectic integrator gives a solution which satisfies the conservation of the volume in phase space.   As a result there is no secular change in the total energy, even though it is not conserved at each time step due to the limited accuracy. These examples confirm that the global properties are more important than the higher order accuracy.  An alternative  method was proposed by \\citet{truelove98}. Their numerical scheme is based on that of \\citet{almgren98} who solved the dynamics of an incompressible fluid with AMR. They evaluated the gravitational potential, $ \\phi $, not on the cell centers but on the vertexes.  While they applied the cell centered difference to the hydrodynamical equations, they applied the vertex centered difference to the Poisson equation.  This method has an advantage that the different level cells share the same gravitational potential automatically.  In other words, this method ensures the continuity of the gravitational potential between the grids of different levels.  This method, however, does not satisfies the Gauss's theorem and accordingly does not ensure the continuity of the gravity, $ \\mbox{\\boldmath$g$} $.  The method of \\citet{truelove98} has another disadvantage that it underestimates gravity.  This disadvantage is due to averaging used to evaluate the vertex centered density.   They evaluated the density at a vertex since the Poisson equation is evaluated on the vertex center. Averaging lowers the peak density and broadens the density distribution.  Thus the gravitational potential evaluated at the vertex is shallower than that evaluated at the cell center.  This difference is seriously large when the gas is concentrated in a few cells.  We have applied our Poisson equation solver to numerical simulations of binary star formation.   The results will be published in near future."
    },
    "0209/astro-ph0209332_arXiv.txt": {
        "abstract": "Based on {\\it XMM-Newton} observations of a sample of galaxy clusters, we have measured the elemental abundances (mainly O, Si, S, and Fe) and their spatial distributions in the intracluster medium (ICM). In the outer region of the ICM, observations of the O:Si:S:Fe ratio are consistent with the solar value, suggesting that the metals in the ICM were produced by a mix of supernovae (SNe) Ia and II. On the other hand, around the cD galaxy, the O/Fe ratios are about half of the solar value because of a central excess of the Fe abundance. An increase of the relative contribution from SNe Ia in the cD galaxy to the metal production towards the center is the most likely explanation. ",
        "introduction": "Clusters of galaxies are filled with an X-ray emitting ICM. The ICM is not only a dominant baryon component in the nearby Universe but also contains comparable or much amount of heavy elements than those in galaxies. Therefore, the distribution and composition of metals in the ICM is essential for understanding the history of metals in galaxies and clusters.  Following early measurements, {\\it ASCA} and {\\it BeppoSAX} observations have revealed several important properties of the ICM metallicity. These include Fe abundance increases around cD galaxies (e.g. Fukazawa 1998; De Grandi and Molendi 2001) and variations in Si/Fe ratio within a cluster (e.g. Finoguenov et al. 2000) and among clusters (e.g. \\cite{ttamura-B3:fu98}). {\\it XMM-Newton} observations with higher capability should improve the accuracy of the measurements. In particular, thanks to the high spectral resolution (RGS) and better sensitivity (EPIC) in the soft X-ray band, significant improvement of the O abundance determination is expected. The O abundance is crucial to understand the origin of the ICM metal since O is expected to be produced mostly by SNe II.  Early results from {\\it XMM-Newton} of S\\'ersic~159-03, A~1835, A~1795, and A~496 were reported in Kaastra et al. (2001), Peterson et al. (2001), Tamura et al. (2001a) and \\cite*{ttamura-B3:ta01b}, respectively. Here we present new results of abundance measurements. Table~\\ref{ttamura-B3_tbl:sample} summaries our sample.  Note that most of our sample have a giant elliptical galaxy (cD galaxy) at the X-ray center. Around the galaxy, we often find metal-rich and cooler X-ray emission compared to the outer region of the cluster.  \\begin{table} \\caption[]{ The sample. (2) redshift. (3) The ICM temperature in keV. (4) {\\it XMM-Newton} publications and notes.} \\label{ttamura-B3_tbl:sample} \\begin{center} \\leavevmode \\footnotesize \\begin{tabular}{lccl} \\hline (1) target\t& (2) z\t& (3) T\t& (4) \\\\ \\hline M87/Virgo\t&  0.0044\t& 2.5\t& BBK01, BSB01\\\\ NGC~533\t\t&  0.017\t& 1.2\t& a galaxy group.\\\\ A~262\t\t&  0.016\t& 2.2\t& \\\\ S\\'ersic~159-03\t&  0.056\t& 2.5\t& K01\\\\ A~496\t\t&  0.033\t& 4.5\t& T01b\\\\ Hyd-A\t\t&  0.054\t& 4.0\t& \\\\ A~1795\t\t&  0.063\t& 6\t& T01a\\\\ \\hline \\end{tabular} \\end{center} \\end{table} ",
        "conclusions": "Our measurements of the elemental abundances along with the {\\it XMM-Newton} results of M87 (B\\\"ohringer et al. 2001) are summaried in Fig.~\\ref{ttamura-B3_fig:abundance}. Main results are followings. \\begin{enumerate} \\item The average (and standard variance among clusters) of the O/Fe ratio around the cD galaxy are 0.5 (and 0.16) times the solar value, respectively. \\item Those of the O/Fe, Ne/Fe, Si/Fe, S/Fe, and (Ar=Ca)/Fe in the outer region of clusters are 0.80 (0.17), 0.7 (0.6), 1.11 (0.29), 1.06 (0.24), and 1.2 (1.0) solar, respectively. These variances among clusters are comparable to the statistical errors and insignificant. \\item There is significant change in the O/Fe ratio between the center and outer regions of clusters. \\end{enumerate}  \\subsection{Comparisons with {\\it ASCA} results} Fukazawa et al.(1998) measured Si and Fe abundances in $\\sim$ 40~clusters excluding the central cool region. They reported that the Si/Fe ratio varies depending on the ICM temperature; groups and poor clusters ($1\\sim2$~keV) have Si/Fe of $\\sim$ 1 solar, while rich clusters exhibit a higher value of 2--3 solar. Since our sample consists of groups and poor clusters (1.2--4~keV), our results of Si/Fe $\\sim$ 1 solar is roughly consistent with the ASCA measurements.  \\subsection{Origin of the metals in the ICM} We interpret our new abundance measurements along with other {\\it XMM-Newton} results. Here we assume that the distribution of the abundance ratios in the ICM have not change since the ejection of those metals from galaxies. However, we should note that the change in Si/Fe ratio among clusters from the ASCA results suggests that a large amount of $\\alpha$-elements such as O, Si, and S have selectively escaped from the gravitational potential in poorer systems (Fukazawa et al. 1998).  The result (2) indicates that in member galaxies in the clusters metal has been enriched in a similar way to that in solar neighborhood, and then ejected into the intergalactic space. Then, we compare the observed abundance ratios among O, Si, S, and Fe with a mixture of SN Ia and SN II prediction in Fig.~\\ref{ttamura-B3_fig:sniaii}. This comparison indicates that the observed O:Si:S:Fe ratio is consistent with a SN Ia originating Fe mass fraction of 0.6--0.8 (or equivalently a SN Ia/SN II frequency ratio of 0.2--0.7).   How about the cluster center ? The result (3) implies that there are at least two different origins for the metals in the ICM, irrespective of any theoretical model for the metal production. An increase of the relative contribution from SNe Ia in the cD galaxy to the metal production towards the center is one possibility. This is because a SN Ia is supposed to produce O/Fe ratio smaller than the solar value. In fact, Fukazawa et al. (1998) found that the excess Fe mass around the cD galaxy in general can be produced by the standard SN Ia rate over a Hubble time. The abundances of other elements such as Si and S at the cluster center, which is not addressed here, are important to examine this idea more quantitatively (Fukazawa 1998; Finoguenov et al. 2000).  \\begin{figure} \\resizebox{0.8\\hsize}{!}{\\includegraphics[angle=-90]{ttamura-B3_fig3.ps}} \\caption{The observed abundance relative to Fe normalized to the solar value. The values of the central cool component (from RGS) and the ICM (from EPIC) are indicated by a open circle and a filled-circle, respectively. } \\label{ttamura-B3_fig:abundance} \\end{figure}  \\begin{figure*} \\begin{center} \\resizebox{0.30\\hsize}{!}{\\includegraphics[]{ttamura-B3_fig4a.ps}} \\resizebox{0.30\\hsize}{!}{\\includegraphics[]{ttamura-B3_fig4b.ps}} \\resizebox{0.30\\hsize}{!}{\\includegraphics[]{ttamura-B3_fig4c.ps}} \\caption{The O/Fe, Si/Fe, and S/Fe ratios: Observation vs. SN Ia$+$SN II model predictions. Our results of the abundance range are shown in horizontal lines. As a function of the ICM SN Ia originated fraction of Fe, predictions based on a SN Ia yield (Thielemann et al. 1993) and SN II yields from three different models are plotted . Three SN II models are T95=Tsujimoto et al. (1995), W95;A;Z and W95;A;$10^{-4}$ = model A in Woosley \\& Weaver (1995) with Z=Z$_{\\odot}$ and Z=$10^{-4}$Z$_{\\odot}$, where Z is metallicity of the stars, respectively. We used yields calculations by Gibson et al. (1997) who averaged elemental yields over the progenitor mass range 10--50~\\hbox{M$_{\\odot}$} for a Salpeter IMF. } \\label{ttamura-B3_fig:sniaii} \\end{center} \\end{figure*}"
    },
    "0209/astro-ph0209381_arXiv.txt": {
        "abstract": "We have developed a time-dependent model of ionization, excitation  and energy balance of SN~1987A atmosphere at the photospheric epoch to study early behavior of hydrogen and helium lines. The ionization freeze-out effects play a key role in producing both the strong H$\\alpha$ during nearly the first month and the He~I 5876\\,\\AA\\ scattering line on day 1.76. Using an extended reaction network between hydrogen molecules and their ions demonstrates that ion-molecular processes are likely responsible for the blue peak in H${\\alpha}$ profile at the Bochum event epoch. ",
        "introduction": "We understand a spectrum formation of SNe~II not better than we do the hydrogen and helium spectra of SN~1987A at the early photospheric epoch. Despite many attempts to explain H$\\alpha$ and He I 5876\\,\\AA\\ lines we are still far from satisfactory results~\\cite{viu01,viu02,viu03,viu04}. The stumbling block is that any model produces too week H$\\alpha$ and He I 5876\\,\\AA\\ at the early photospheric epoch.  Yet, it has already become clear that a residual ionization (freeze-out effects) may result in the enhanced excitation of hydrogen compared to steady-state regime~\\cite{viu05,viu06}. However, in our previous study~\\cite{viu06} we assumed simple laws for the  behavior of electron temperature in the atmosphere. Here we will confirm a vital role of freeze-out effects in the hydrogen excitation at the photospheric epoch using the upgraded model in which time-dependent chemical kinetics is solved along with time-dependent energy balance. Moreover, here we consider neutral helium, not only hydrogen, and will show a key role of the time-dependent kinetics for He I 5876\\,\\AA\\ line too.  We also recapitulate here our previous results on modeling the blue peak of H$\\alpha$ at the Bochum event phase upon the basis of the time-dependent kinetics with hydrogen-composed species~\\cite{viu06}. This will demonstrate that not only first-order effects, like freeze-out, but subtle molecular processes may be also important for hydrogen spectrum formation in SN~II.   \\begin{figure}[t] \\begin{center} \\includegraphics[width=0.9\\textwidth]{utrobinF1.ps} \\end{center} \\caption[]{ Time evolution of H$\\alpha$ from day 1.76 to 29.68 (\\textbf{a-d}) and He I 5876\\,\\AA\\ on day 1.76 (\\textbf{a}) in SN~1987A. The observed spectra from day 1.76 to 8.69~\\cite{viu12} and on day 29.68~\\cite{viu13} ({\\it thick solid line\\,}) are compared to those computed with the full time-dependent energy balance ({\\it thin solid line\\,}), with adiabatic evolution of the kinetic temperature ({\\it short dashed line\\,}), with the electron temperature equal to the local radiative temperature ({\\it long dashed line\\,}), and for the steady state ({\\it dotted line\\,}) } \\label{utrobinF1} \\end{figure}   \\begin{figure}[t] \\begin{center} \\includegraphics[width=0.9\\textwidth]{utrobinF2.ps} \\end{center} \\caption[]{ {\\it Left panel}: The role of molecular reactions for the Bochum event. (\\textbf{a}) H$\\alpha$ profiles on day 29.68 calculated in time-dependent approach with ({\\it thin solid line\\,}) and without ({\\it dotted line\\,}) molecular reactions. (\\textbf{b}) The behavior of the Sobolev optical depth of H$\\alpha$ in both models. {\\it Right panel}: Physical conditions in the supernova atmosphere ($v>10700$\\,km/s) on day 4.64. (\\textbf{c}) The density ({\\it thick solid line\\,}) and electron temperature (in units of 1000\\,K). The temperature is given for the model with the full energy balance ({\\it thin solid line\\,}) and for that with the adiabatic approximation ({\\it dotted line\\,}). (\\textbf{d}) The behavior of the Sobolev optical depth for H$\\alpha$ and fractional abundance of H$^+$ for both models } \\label{utrobinF2} \\end{figure} ",
        "conclusions": "Our present study of the time-dependent effects and the influence of molecules and radioactivity in the photospheric epoch of SN~1987A may be summarized as follows: \\begin{itemize} \\item the time-dependent effects are a key prerequisite for the strong H$\\alpha$ line during nearly the first month and the He I 5876\\,\\AA\\ line on day 1.76; \\item the additional excitation from radioactive $^{56}$Ni decays is not required to account for the H$\\alpha$ line during about the first month; \\item the molecular processes among hydrogen-composed species turn out important factor of hydrogen neutralization which is manifested by the emergence of the blue peak of H$\\alpha$ at the Bochum event phase. \\end{itemize} We believe that these effects are important for normal SNe II-P at the plateau epoch as well."
    },
    "0209/astro-ph0209348_arXiv.txt": {
        "abstract": "We present a spatially resolved spectroscopic analysis of the young Galactic supernova remnant \\kes\\ (SNR G29.7-0.3) using the \\chandra\\ X-ray Observatory.  \\kes\\ is one of an increasing number of examples of a shell-type remnant with a central pulsar powering an extended radio/X-ray core.  We are able to pinpoint the location of the recently discovered pulsar, \\psr, and confirm that X-rays from the remnant's core component are consistent with non-thermal power-law emission from both the pulsar and its surrounding wind nebula. We find that the spectrum of the pulsar is significantly harder than that of the wind nebula. Fainter, diffuse emission is detected from throughout the volume delineated by the radio shell with a surface brightness distribution strikingly similar to the radio emission. The presence of strong lines attributable to ionized Mg, Si, and S indicate that at least some of this emission is thermal in nature. However, when we characterize the emission using a model of an underionized plasma with non-solar elemental abundances, we find we require an additional diffuse high-energy component. We show that a significant fraction of this emission is an X-ray scattering halo from the pulsar and its wind nebula, although a nonthermal contribution from electrons accelerated in the shock cannot be excluded. ",
        "introduction": "Our observation of \\kes\\ was obtained on $10-11$ Oct 2000 with the \\chandra\\ X-ray Observatory (Weisskopf, O'Dell, \\& van Speybroeck 1996). Photons were collected using the Advanced CCD Imaging Spectrometer (ACIS), a mosaic of ten X-ray CCD chips, with the target placed on the ACIS--S3 chip, offset $2^{\\prime}$ from the nominal aim-point to avoid losing portions of the remnant to the CCD gaps. The back-side illuminated ACIS--S3 CCD is sensitive to photons in the $\\sim 0.2-10$ keV energy range with a spectral resolution of $E/\\Delta E \\sim 10$ at 1 keV. The on-axis angular point-spread function (PSF) of the telescope is $\\sim 0\\farcs5$ at 1 keV and is undersampled by the CCD pixels ($\\sim 0\\farcs49 \\times 0\\farcs49$). As a consequence of the 3~s CCD readout time, no timing information for the pulsar ($P = 324$~ms) is available with the ACIS instrument.  Data reduction and analysis were performed using the CIAO, FTOOLS, and XSPEC X-ray analysis software packages.  To correct for the detrimental effects of charge transfer inefficiency (CTI) on the gain and resolution of the CCD, we reprocessed the Level 1 event data using the custom software of Townsley et al. (2000). The resulting CTI-corrected Level 2 event file was then time-filtered to exclude intervals of unacceptably high background activity caused by flares in the rate of solar cosmic ray particles. These intervals were identified using a $3\\sigma$ iterative clipping algorithm applied to a lightcurve created using all data on the ACIS-S3 chip but with the supernova remnant emission excluded (defined as emission from a radius of $r>100^{\\prime\\prime}$ centered on the SNR radio shell). The subsequent time-filtering resulted in a useful exposure of 33.7 ks. The instrument response files used in the spectral fits were calculated using the RMF and quantum efficiency uniformity (QEU) calibration files that accompany the CTI-removing software. ",
        "conclusions": "\\subsection{The distance to \\kes}  Becker and Helfand (1984) presented a 21 cm hydrogen absorption spectrum for \\kes\\ which shows clear evidence of absorption at negative velocities, indicating a location beyond the solar circle; we adopt their distance of 19 kpc. Assuming a hydrogen spin temperature of 100 K, these measurements indicate a neutral atomic hydrogen column density to the source of $\\sim2 \\times 10^{22}$ cm$^{-2}$. Since roughly half the hydrogen along any line of sight through the Galaxy is either in ionized or molecular form, we should expect an X-ray absorption column density of $\\sim 4 \\times 10^{22}$ cm$^{-2}$, in excellent agreement with that derived from our fit to the pulsar wind nebula spectrum ($3.96 \\times 10^{22}$ cm$^{-2}$). The lower $N_H$ values found in the various thermal fits reported in Table 2 and 3 are implausible; the discrepancy is probably due to inadequate plasma models. We thus use the wind nebula value in assessing the intrinsic luminosities of all components of the remnant.  \\subsection{The X-ray luminosity of the pulsar and its PWN}  The nonthermal luminosity from the pulsar is $4.1 \\times 10^{35}$ erg s$^{-1}$ in the $0.5-10$ keV band, and the luminosity of its wind nebula\\footnote{We note here that there is an error in the X-ray luminosities for this source quoted in Blanton and Helfand (1996). The observed, rather than intrinsic, X-ray fluxes were used. The correct values from the {\\it ASCA} analysis ($L_x(0.5-8.0~\\rm{keV})_{core} = 2.5 \\times 10^{36}$ erg s$^{-1}$ and $L_x(0.5-8.0~\\rm{keV})_{shell} = 1.1 \\times 10^{37}$ erg s$^{-1}$) agree well with the values derived here.} is $1.7 \\times 10^{36}$ erg s$^{-1}$.  Both values are second only to the Crab Nebula and its pulsar among known Galactic objects. However, the efficiencies with which this pulsar converts its rotational kinetic energy to X-rays in its magnetosphere and in its surrounding nebula are both  greater than the Crab. The pulsar value of $L_{x-pulsar}/\\dot E\\sim 1.6\\%$ is more than six times this value for the Crab, although it is very similar to the value for PSR~J0540-693, the 50 ms pulsar in the Large Magellanic Cloud. The value of $L_{x-nebula}/\\dot E \\sim 6.5\\%$ is the highest known -- in this case, comparable to that for the Crab, but several times higher than that for PSR~J0540-693.  The similarity of the pulsars and their wind nebulae in \\kes\\ and PSR~0540-693 is noteworthy. Morphologically, they are two of the best representatives of the composite remnant class, with prominent radio and X-ray shells and bright pulsar-driven synchrotron cores. Their relatively high values of $L_{x-nebula}/\\dot E$ could result from the confinement of the outflowing pulsar wind by the evident shells. The characteristic age of PSR~0540-693 is roughly twice that of \\psr\\ (1400 vs 700 yrs), although the detailed optical study of Kirshner et al. (1989) found an expansion velocity for the shell surrounding PSR~J0540-693 of $\\sim 3000$ km s$^{-1}$ and a dynamical age of $\\sim 760$ yrs. There are several striking differences between these two systems however: their spin periods, period derivatives (and, thus, inferred magnetic field strengths), and their shell diameters. It is not implausible that very different pulsar birth parameters have led to the difference in the remnant sizes.  \\subsection{The large shell diameter of \\kes}  The radius of \\kes, 9.7 pc, is enormous for the young characteristic age of its pulsar, implying a {\\it mean} expansion velocity for the remnant of over 13,000 km s$^{-1}$. For even a modest $5 M_{\\odot}$ of material ejected into a vacuum, the kinetic energy required is nearly 10$^{52}$ erg. For 10 $M_{\\odot}$ of ejecta going off into a medium with a mean density of 0.5 cm$^{-3}$, the required kinetic energy reaches nearly 10$^{53}$ erg, comparable to the entire gravitational binding energy of the neutron star whose formation produced the explosion. The shell X-ray luminosity derived in \\S5 suggests a density closer to 1 cm$^{-3}$. Such kinetic energy values are both unprecedented and highly implausible.  There are several possible ways to reduce the required explosion energy. A smaller distance for the remnant would reduce the inferred velocity, albeit only as $d^{0.5}$. The HI absorption spectrum of Becker and Helfand (1984) leaves little possibility of reducing the distance by more than $\\sim 30\\%$, in that all velocities are seen in absorption to the solar circle and beyond.  Alternatively, the remnant could be considerably older than the pulsar's characteristic age $\\tau_c = P/2\\dot P$. This would run counter to the situation for other remnants. For the Crab, $\\tau_c = 1240$ yrs, while its true age $t = 945$ yrs, for PSR~0540-693 $\\tau_c = 1672$ yrs (Zhang et al. 2001) while the dynamical age estimate is 760 yrs, and for PSR~J1811.5-1926 in G11.2-0.3, $\\tau_c = 24,000$ yrs, while its association with SN 386AD means $t=1616$ yrs (Kaspi et al. 2001); a plausible explanation in each of these cases is that the pulsar's initial period was not negligible compared to its present spin rate (an assumption of the approximation for $\\tau_c$ defined as above).  Indeed, for a pulsar braking index $n= - \\nu \\ddot \\nu / \\dot \\nu^2 = 3$, expected for braking by pure magnetic dipole radiation, the characteristic age is {\\it always} an upper limit to the true age. It is possible for the true age to exceed $\\tau_c$, however, if $n<3$. But for the range of most measured indices ($\\sim 1.4<n<2.9$), $t\\le 5\\tau_c$, and is equal only for a very high initial spin rate, $P_o$. The recent results of Mereghetti et al. (2002) suggest a value for n in the range $1.86<n<2.48$ depending on the glitch history of the star. For high initial spin rates, this yields ages in the range $980<t<1770$ years. If we push all the numbers in one direction -- reduce the distance to 15 kpc, posit only 5 $M_{\\odot}$ of ejecta, require an ambient density of $n = 0.1$ cm$^{-3}$, and double the age, thus requiring a small $P_o$ -- the energy needed {\\it just} to explain the observed kinetic energy is $2 \\times 10^{51}$ ergs, at the upper end of the inferred distribution of supernova explosion energies.  A consequence of requiring a fast initial spin rate for \\psr\\ is that, given its enormous magnetic field of $5 \\times 10^{13}$ G, the amount of rotational kinetic energy dumped into the remnant shortly after the explosion will exceed the fraction of the explosion energy itself that couples to the surrounding matter (typically $\\sim 1\\%$). For $P_0 = 0.01P \\sim 3$~ms, $E_{rot} = 2.2 \\times 10^{51}$ ergs. As Blanton and Helfand (1996) showed, the current total energy in radio-emitting relativistic particles and magnetic field in the synchrotron nebula is only $\\sim 10^{48}$ ergs. If \\psr\\ had been born with a spin period only a factor of two or so below its current value, it could have just supplied this inferred energy with none left over to help power the expansion of the remnant. However, if it were born spinning a hundred times faster, the current synchrotron nebula represents only $\\sim 10^{-4}$ of its total energy output, with the remainder transferred to the kinetic energy of the shell.  \\subsection{The hard component to the shell emission}  As we showed in \\S4, the X-ray emission from outside the clearly demarcated pulsar wind nebula includes a hard spectral component that is inconsistent with an extension of the thermal emission that accounts for the bulk of the shell's X-ray luminosity. Both a very high-temperature plasma ($T\\sim 20$ keV) and a power-law can satisfactorily fit the high-energy spectrum. We explore here the possible origins for this emission.  There are two possible sources of relativistic electrons which could contribute to the power-law emission: leakage of pulsar wind nebula particles into the shell cavity, and particles accelerated at the blastwave shock. The first explanation can be shown to be implausible.  There is no strong gradient in the amount of high energy emission with radius; in particular, 5 keV photons in excess of those expected from the low-temperature thermal plasma are found to the edge of the remnant shell.  The particle energy required for 5 keV radiation $E \\sim 2.7 \\times 10^4$ GeV. The synchrotron lifetime of such a particle must, at minimum, exceed the light travel time from the pulsar to the rim of the remnant: $\\tau_{synch}>c/r\\approx10^9$~s or 32 years. This sets a constraint on the strength of the magnetic field in the remnant $B<422~{\\rm s}~\\tau_{synch}^{-1} E_{GeV}^{-1}\\approx 4\\mu$G -- comparable to the interstellar field. By contrast, the equipartition field in the pulsar wind nebula is $300\\mu$G; furthermore, propagation of the particles from the wind nebula boundary to the remnant's rim at $c$ is highly unrealistic, lowering further the allowed field strength within the shell. Thus, it seems highly unlikely that the bulk of the high energy diffuse emission in the remnant is from pulsar-injected particles.  In recent years, synchrotron X-rays from particles accelerated at the outward moving supernova shock have been discovered in several remnants (Allen, Gotthelf \\& Petre 2000 and references therein; Slane et al. 1999; Slane et al. 2001). The observed power law spectral indices are quite steep, ranging from $\\Gamma = 2.4$ in G347.3-0.5 (Slane et al. 1999) to $\\Gamma = 3.3$ in RCW 86 (although it should be noted that the XTE results of Allen et al. (2000) on young remnants uses a harder energy band -- and yields steeper slopes -- than are obtained for the two ASCA synchrotron shells which are fitted in a softer band). Contributions from nonthermal emission to the total X-ray flux in the $1-10$ keV band range from a few percent to $>90\\%$.  The fitted power law component in the shell of \\kes\\ has a photon index on the flat end of this distribution (although it is not very well-constrained -- see Table 4) and an implied luminosity of $\\sim 3.2 \\times 10^{35}$ erg s$^{-1}$. Given the high shell expansion velocity, it is not implausible that some of this emission is indeed direct synchrotron radiation from high energy electrons. However, when the effect of dust scattering from the PWN is included (see below), the inferred spectral index is flatter than that seen in any other remnant.  One component that must be present at some level arises from the dust-scattering halo of the central nebula plus pulsar (e.g., Mauche and Gorenstein 1986).  A substantial literature exists on this topic, although it has been problematic to derive quantitative constraints on the relevant parameters -- grain size distribution, composition, internal structure, and distribution along the line of sight -- from data of limited spatial resolution. The recent \\chandra\\ observation of the Galactic X-ray binary GX 13+1 by Smith, Edgar, and Shafer (2002) shows simultaneously the lack of agreement with previous models as well as the sensitivity of such observations to various instrumental effects. Nonetheless, as it provides a direct empirical measurement of dust-scattering effects at the highest available resolution, we use these results to estimate the contamination of the \\kes\\ shell by the dust scattering halo of the PWN.  Smith et al. provide radial profiles at three energies for the halo of GX 13+1.  We have measured the mean surface brightness in each $10\\asec$ annulus from $50\\asec$ to $100\\asec$ (the outer limit of the remnant shell) and calculated the fraction of scattered flux detected. We have then extrapolated the observed surface brightness to $25\\asec$ (the inner boundary of the diffuse shell emission in question) and obtained a crude estimate of the fraction of the intensity scattered into the SNR shell region from the pulsar plus its surrounding nebula (ignoring the effect of the nebula's extent). We obtain values of $\\sim 15\\%$, $\\sim 10\\%$, and $\\sim 8\\%$ for the $\\Delta E = 100$ eV energy bands centered at 2.15, 2.95, and 3.75 keV, respectively. As noted by Smith et al., the energy dependence $\\sim E^{-1}$ is shallower than the theoretically expected value of $E^{-2}$. We then correct for the difference in column density between GX 13+1 and \\kes\\ ($\\sim 2.8 \\times 10^{22}$ cm$^{-2}$ vs.  $\\sim 4 \\times 10^{22}$ cm$^{-2}$) by finding $\\tau_{scat}$ from Figure 4 in Predehl (1997) and using the simple relation for the fractional scattered intensity $I_{frac} = (1 - e^{-\\tau_{scat}})$; this increases the estimates above by a factor of 1.2.  In the band $2-4$ keV, then, it is possible to explain most of the putative power-law emission as a simple dust-scattering halo of the PWN.  In the $1-2$ keV band, however, we would expect a contribution from the halo that exceeds the total power law component. Two effects are relevant here. Mathis and Lee (1991) show that multiple scattering is expected to set in for $\\tau_{scat}>1.3$ and broaden the halo; for our column density, this limit is exceeded at 1.7 keV, although the effect is not large for the $25\\asec$ to $100\\asec$ region of interest here. Secondly, the thermal component of the fit absorbs a large (and uncertain) fraction of the power in the $1-2$ keV band where the Mg and Si lines are found. If some of the photons in this band are in fact attributable to dust scattering, it might well raise the equivalent widths of these lines and boost their sub-solar abundances to (the expected) higher values. It could also help explain the disturbing trend toward lower temperatures at larger shell radii noted above.  In the high energy ($4-7$ keV) band, there appears to be some power law contribution beyond the expected halo emission; whether this is evidence of an unusually flat-spectrum synchrotron contribution from the shell, a small thermal contribution from the fast blastwave shock, or further unexpected scattered flux remains unclear.  A quantitative analysis of the spectrum including all of the effects discussed above is beyond the scope (not to mention the calibration uncertainties and photon counting statistics) of this paper. Less distant remnant shells are better targets for attempts to untangle the complicated physics of thermal and non-thermal X-ray emission. However, the apparent importance of considering dust-scattering in this case should serve as a cautionary tale for observers using the new high-energy imaging capabilities to explore absorbed remnants in the Galactic plane.   \\subsection{Conclusions}  Our high-resolution imaging observations of \\kes\\ have revealed all the components expected surrounding the site of recent stellar demise. The young pulsar and its wind nebula are shown to be among the most efficient known at turning rotational kinetic energy into X-ray emission; the extraordinary magnetic field strength of the pulsar may be responsible for this notoriety.  Thermal X-ray emission is seen from throughout the remnant shell, although deriving constraints on the remnant's evolutionary state and elemental abundances is compromised by the presence of a significant diffuse nonthermal component which we attribute largely to the dust-scattering halo of the pulsar and its nebula. The presence of such halos needs to be considered when deriving inferences concerning the presence of nonthermal components in the spectra of distant shell-type and composite remnants."
    },
    "0209/astro-ph0209512_arXiv.txt": {
        "abstract": "Planetary companions to the source stars of a caustic-crossing binary microlensing events can be detected via the deviation from the parent light curves created when the caustic magnifies the star light reflecting off the atmosphere or surface of the planets.  The magnitude of the deviation is $\\deltap \\sim \\ep \\rhop^{-1/2}$, where $\\ep$ is the fraction of starlight reflected by the planet and $\\rhop$ is the angular radius of the planet in units of angular Einstein ring radius. Due to the extraordinarily high resolution achieved during the caustic crossing, the detailed shapes of these perturbations are sensitive to fine structures on and around the planets.  We consider the signatures of rings, satellites, and atmospheric features on caustic-crossing microlensing light curves.  We find that, for reasonable assumptions, rings produce deviations of order $10\\% \\deltap$, whereas satellites, spots, and zonal bands produce deviations of order $1\\%\\deltap$.  We consider the detectability of these features using current and future telescopes, and find that, with very large apertures ($>$30m), ring systems may be detectable, whereas spots, satellites, and zonal bands will generally be difficult to detect.  We also present a short discussion of the stability of rings around close-in planets, noting that rings are likely to be lost to Poynting-Robertson drag on a timescale of order $10^5$ years, unless they are composed of large ($\\gg$1~cm) particles, or are stabilized by satellites. ",
        "introduction": "Precise radial velocity surveys have detected over 100 planetary companions to FGKM dwarf stars in the solar neighborhood (see http://cfa-www.harvard.edu/planets/catalog for a list of planets and discovery references).  Among the interesting trends that have been uncovered in this sample of planets are a positive correlation between the frequency of planets and metallicity of the host stars \\citep{gonzalez1997, gonzalez1998, laughlin2000, santos2001, reid2002}, a paucity of massive, close-in planets \\citep{zucker2002,patzold2002}, and a `piling-up' of less-massive, close-in planets near periods of $P\\simeq 3\\days$.  This latter trend is important because the number of planets which transit their parent stars is roughly proportional to $1/a$, where $a$ is the semi-major axis. The discovery and interpretations of these global trends provide clues to the physical mechanisms that affect planetary formation, migration, and survival.  A somewhat different way of obtaining clues about the physical processes at work in planetary systems is to acquire detailed information about individual planets.  With radial velocity measurements alone, such information is limited only to the minimum mass $\\mp\\sin{i}$ of the planet, and the semi-major axis $a$ and eccentricity of its orbit.  However, if the planet also transits its parent star, then it is possible to infer considerably more information.  A basic transit measurement allows one to infer the radius, mass, and density of the planet, as has been done with the only known transiting extrasolar planet, HD209458b \\citep{charbonneau00, henry2000}.  This in turn allows one to place constraints on the planet's orbital migration history \\citep{burrows2000}.  More detailed photometric and spectroscopic data during (and outside of) the transit can be used to study the composition of, and physical processes in, the planetary atmosphere \\citep{seager2000, seager2000b, charbonneau2002, brown2002}, measure the oblateness, and thus constrain the rotation rate, of the planet \\citep{hui2002, seager2002}, and to search for rings and satellites associated with the planet \\citep{sartoretti99, schneider99, brown01}.   The `classical' method of searching for planets via microlensing was first proposed by \\citet{mao91}, and subsequently further developed by \\citet{gould92}.  In this method, a planetary companion to the primary lens star produces a small perturbation atop the smooth, symmetric lensing light curve created by the primary.  The microlensing method has several important advantages over other methods, as well as several disadvantages (see \\citealt{gaudi2003} for a review).  The most important advantage is that the strength of the planet's signal depends weakly on the planet/primary mass ratio and thus it is the only currently feasible method to detect Earth-mass planets \\citep{bennett96}.  The other advantage is that it enables one to detect planets located at large distances of up to several tens of kiloparsecs.  However, it also has disadvantages, the most important of which is that the only useful information one can obtain is the mass ratio between the planet and the primary.  Thus classical microlensing searches only allow one to identify the existence of the planet, and build statistics about the types of planetary systems, but cannot be used to obtain detailed information about the discovered planets.  This is especially problematic in light of the fact that follow-up of the discovered systems will generally be difficult or impossible.   Recently, \\citet{graff00} and \\citet{lewis00} proposed a novel method of detecting planets via microlensing.  They suggested that one could detect close-in giant planets orbiting the {\\it source} stars of caustic-crossing binary-lens events via accurate and detailed photometry of the binary-lens light curve.  In this method, the planet can be detected because the light from the planet is sufficiently magnified during the caustic crossing to produce a noticeable deviation to the lensing light curve of the primary.  The magnitude of the deviation is $\\deltap \\sim \\ep \\rhop^{-1/2}$, where $\\ep$ is the ratio of the (unlensed) flux from the planet to the (unlensed) flux from the star, and $\\rhop$ is the angular radius of the planet in units of the angular Einstein ring radius $\\thetae$ of the lens system.  The Einstein ring radius is related to the physical parameters of the lens system by \\begin{equation} \\thetae=\\sqrt{{{2 \\rsch}\\over \\drel} }, \\label{eqn:thetae} \\end{equation} where $\\rsch=2GM/c^2$ is the Schwarzschild radius of the lens, $M$ is the total mass of the lens, $\\drel\\equiv \\dos\\dol/\\dls$, and $\\dos$, $\\dol$, and $\\dls$ are the distances between the observer-source, observer-lens, and lens-source, respectively.  For searches in the optical, the light from the planet will be dominated by the reflected light from the star, and $\\ep \\sim 10^{-4}$ for close-in planets.\\footnote{Because the fraction of reflected light decreases as $\\ep \\propto a^{-2}$, optical searches will generally only be sensitive to close-in planets.  However, planets may have significant intrinsic flux in the infrared, enabling the detection of more distant companions at longer wavelengths.}   Adopting typical parameters, $\\rho_{\\rm p}\\sim 10^{-4}$, and thus $\\deltap \\sim 1\\%$.  \\citet{graff00} demonstrated that this level of photometric precision is currently within reach of the largest aperture telescopes.  The exquisite resolution afforded by caustics may allow one to study features on and around the source in detail, and with larger aperture telescopes, one may able to study spots and bands on the surfaces of detected planets by looking for small deviations to the nominal light curve \\citep{graff00}. Here we study the signatures of these and other structures on lensing light curves, quantify the magnitude of the deviations, and assess their detectability using current and future instrumentation. Specifically, we consider the signatures of rings and satellites, as well as atmospheric features such as spots, zonal bands, and scattering. \\citet{lewis00} considered using variations in the polarization during the planetary caustic crossing to probe the composition of the planetary atmosphere. The effects of the phase of the planet on the light curve were considered previously by \\citet{ashton01}.   The layout of the paper is as follows.  In \\S\\ref{sec:binlens}, we discuss binary lenses and their associated caustic structures, and describe the magnification patterns near caustics.  We discuss expectations for the existence, stability, and properties of rings, satellites, and atmospheric features of close-in extrasolar planets in \\S\\ref{sec:gen}.  In \\S\\ref{sec:quantitative}, we layout the formalism for calculating microlensing light curves, and apply this formalism to make a quantitative predictions for the deviations caused by planets (\\S\\ref{sec:planet}), satellites (\\S\\ref{sec:satellites}), rings (\\S\\ref{sec:rings}), and atmospheric features (\\S\\ref{sec:atmosphere}). We address the detectability of these deviations in \\S\\ref{sec:detect}, and summarize and conclude in \\S\\ref{sec:summary}. ",
        "conclusions": "}  Planetary companions to the source stars of caustic crossing microlensing events can be detected via the brief deviation created when the caustic transits the planet, magnifying the reflected light from the star.  The magnitude of the planetary deviation is $\\delta_{\\rm p} \\sim \\epsilon_{\\rm p} \\rho_{\\rm p}^{-1/2}$, where $\\epsilon_{\\rm p}$ is the fraction of the flux of the star that is reflected by the planet, and $\\rho_{\\rm p}$ is the angular size of the planet in units of the angular Einstein ring radius of the lens. For giant, close-in planets (similar to HD20958b), $\\epsilon_{\\rm p} \\sim 10^{-4}$, and for typical events toward the Galactic bulge, $\\rho_{\\rm p} \\sim 10^{-4}$.  Thus $\\delta_{\\rm p} \\sim 1\\%$, which is accessible to 10m-class ground-based telescopes.  Due to the extraordinarily high angular resolution afforded by caustic crossings, fine structures in and around the planet are, in principle, also detectable.  We first presented a brief discussion on the existence and stability of satellites, rings and atmospheric features of close-in planets, concluding that although rings and satellites may be short-lived due to dynamical forces, the ultimate fate of such structures is not clear.  There are good reasons to believe that atmospheric features may be important in close-in planets.  We therefore considered the signatures of satellites, rings, spots, zonal bands, and non-uniform surface brightness profiles on the light curves of planetary caustic-crossings.  Where possible, we used semi-analytic approximations to derive useful expressions for the magnitude of the deviations expected for these features, as a function of the relevant parameters, such as the albedo or size of the feature.  We express these deviations in terms of $\\delta_{\\rm p}$, the magnitude of the planetary deviation.  We find that rings produce deviations of amplitude $\\sim 10\\% \\delta_{\\rm p}$, whereas spots, zonal bands, and satellites all produce deviations of order $\\sim 1\\%\\delta_{\\rm p}$.  These semi-analytic estimates are supported by more detailed numerical simulations. We also find that the light curve of a planet with the surface brightness profile expected from Lambert scattering deviates from that of a uniform source by $\\sim 10\\%\\delta_{\\rm p}$.  This affords the possibility of probing the physical processes of the atmospheres of distant extrasolar planets by constraining their surface brightness profiles, and therefore the scattering properties of their constituent particles.  We assessed the detectability of spots, rings, satellites, and bands with current and future telescopes.  We found that, for reasonable assumptions and 10m-class telescopes, the planetary deviation will have a signal-to-noise of $\\sim 15$, a ring system will only be marginally detectable with a signal-to-noise of $\\sim 6$, and all other features will be completely undetectable.   For 30m-class or larger telescopes, rings should be easily detectable, The detection of the non-uniform nature of the planetary surface brightness profile arising from Lambert scattering requires 100m-class telescopes for bare detection.  Spots, satellites and zonal bands are essentially undetectable for even the largest telescopes apertures.  \\smallskip"
    },
    "0209/astro-ph0209038_arXiv.txt": {
        "abstract": "{We present a spectroscopic study of the long-recurrence-time dwarf nova V\\,592 Herculis based on observations obtained during its August 1998 superoutburst.  From the analysis of  the radial velocities of the H$\\alpha$ emission line we find a most likely orbital period of 85.5 $\\pm$ 0.4 minutes, but the 91.2 $\\pm$ 0.6 minutes alias cannot be completely discarded. Both periods imply a very small period excess and supports the brown-dwarf like nature of the secondary star. ",
        "introduction": "Cataclysmic variable stars (CVs) are interacting binaries consisting of a white dwarf accreting matter from a red dwarf donor. In non-magnetic CVs, the transferred gas spirals onto the white dwarf, forming an accretion disk. Due to the partial hydrogen ionization, the disk  is  thermally unstable, jumping in temperature when a certain critical density is reached. This hot state is accompanied by increased viscosity and a release of  luminous energy when the material rapidly drops onto the white dwarf. This event is called an outburst. In some CVs, quasi-periodic humps are seen in the light curve during extended outbursts; they are called superhumps and the outbursts are called superoutbursts. According to current theories, these peculiar CVs, the so-called SU UMa stars, should contain low mass secondaries; some of them  could have been eroded after a long time of mass transfer. It is even possible that many of them harbour secondary stars with masses less than the minimum mass needed to sustain hydrogen fusion in their cores (e.g. Howell et al.\\,2001). Such objects should be found among ultra-short orbital period systems ($P_{o}$ $\\sim$ 80 minutes) with very low mass transfer rates ($\\dot{M} \\sim 10^{15}$ g/s), and should be characterized by rare and large-amplitude outbursts. This subgroup of the SU UMa stars is usually called the WZ Sge stars. Until now, a relatively small number of these objects have been studied in detail (Kato et al.\\, 2001), in part due to the fact that their low luminosities make them hard to study in quiescence, even with large aperture telescopes. In this paper we make a contribution to the understanding of these rare objects presenting the first spectroscopic study of V\\,592 Herculis during superoutburst. ",
        "conclusions": "\\begin{itemize}  \\item From spectroscopy obtained during superoutburst, we have found the most likely orbital period of the dwarf nova V\\,592 Her, viz\\, $85.5 \\pm 0.4$ minutes. The 91.2 $\\pm$ 0.5 minutes alias cannot be completely ruled out.  \\item We show that both periods, when combined with the reported superhump period and theoretical relationships, give support for the view that V\\,592 Her harbours a brown dwarf like secondary star.  \\item The detection of He\\,II 4686 emission, makes V\\,592 Her  a good candidate to look for spiral shocks during superoutburst.  \\end{itemize}"
    },
    "0209/astro-ph0209506_arXiv.txt": {
        "abstract": "The problem of the origin of cosmic rays in the shocks produced by supernova explosions at energies below the so-called 'knee' (~at $\\sim 3 \\cdot 10^6$ GeV~) in the energy spectrum is addressed, with special attention to the propagation of the particles through the inhomogeneous interstellar medium and the need to explain recent anisotropy results, \\cite{Tsune}. It is shown that the fractal character of the matter density and magnetic field distribution leads to the likelihood of a substantial increase of spatial fluctuations in the cosmic ray energy spectra. While the spatial distribution of cosmic rays in the vicinity of their sources (~{\\em eg.} inside the Galactic disk~) does not depend much on the character of propagation and is largely determined by the distribution of their sources, the distribution at large distances from the Galactic disk depends strongly on the character of the propagation. In particular, the fractal character of the ISM leads to what is known as 'anomalous diffusion' and such diffusion helps us to understand the formation of the Cosmic Ray Halo. Anomalous diffusion allows an explanation of the recent important result from the Chacaltaya extensive air shower experiment \\cite{Tsune}, viz. a Galactic Plane Enhancement of cosmic ray intensity in the Outer Galaxy, which is otherwise absent for the case of the so-called 'normal' diffusion.  All these effects are for just one reason: anomalous diffusion emphasizes the role of local phenomena in the formation of cosmic ray characteristics in our Galaxy and elsewhere. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209440_arXiv.txt": {
        "abstract": "We present the $K$-band local luminosity function derived froma sample of 1056 bright ($K<15$) $K$-selected galaxies from the Hawaii-AAO $K$-band redshift survey. The Hawaii-AAO $K$-band redshift survey covers 4 equatorial fields with a total area of 8.22 $\\deg^2$. We derive both the non-parametric and Schechter luminosity function from our data, and determine $M^*(K)=-23.70\\pm0.08+5\\log_{10}(h)$, $\\alpha=-1.37\\pm0.10$ and $\\phi^*=0.013\\pm0.003~h^3~\\Mpc^{-3}$ for a universe with $\\Omega_m=0.3$ and $\\Omega_{\\Lambda}=0.7$. We also measure the $K$-band luminosity function for the early- and later-type galaxies from our morphologically classified subsample. It appears that later-type galaxies have a fainter $M^*$ and a steep slope, while early-type galaxies have a much brighter $M^*$ and a quite flat slope in their $K$-band luminosity functions. This is consistent with what have been found in optical type dependent luminosity function. The $K$-band luminosity density derived using our luminosity function is now measured at a similar redshift depth to optical luminosity densities in the SDSS redshift survey. It is 2 times higher than the previous measurement from the shallower 2MASS sample and resolves the previously reported discrepancies between optical and near-IR luminosity densities. ",
        "introduction": "The galaxy luminosity function is an important quantity in the study of galaxy evolution and formation. Traditionally, galaxy luminosity functions have been derived in optical bands. It is now clear that extragalactic studies in optical bands suffer several systematic uncertainties and complexities compared to those using near infrared bands. In particular, dust extinction has little effect on $K$-band magnitudes; the K-correction in the $K$-band is a much smaller and better understood  quantity than in the optical bands, and it is independent of galaxy spectral types for $z<1$. Because of this it is easier to detect a high redshift elliptical galaxy in an infrared band than in optical bands. A galaxy's near infrared luminosity is also a good tracer of its stellar mass independent of spectral type \\citep{col01,gla02}. Recent theoretical studies show that the $K$-band galaxy luminosity function is a powerful constraint on galaxy formation theory \\citep{bau98, kau98}.  Most near infrared surveys, however, have been modest in size due to the small size of available infrared detectors. The advent of large format infrared array detectors has made possible a variety of wide-field near-infrared surveys, ranging from several at the 10 deg$^2$-level \\citep{gar96, hua97} up to the largest of them all, the 2 Micron All Sky Survey (2MASS) \\citep{skr97}. Obtaining optical redshifts for a $K$-selected sample, however, is difficult because the wide range in optical-infrared colours results in the requirement for a wide range of exposure times to acquire optical spectroscopy. In particular, very long exposures are required to secure redshifts for the reddest objects. Because of this there are only a few $K$-band luminosity functions available. The early $K$-band luminosity functions were derived from small size samples with number of galaxies ranging from 100 to 500 \\citep{mor93, gla95, gar97, szo98, lov00}. After the second incremental release of 2MASS data, two teams \\citep{col01,koc01} used the overlap between the 2MASS Extended Source Catalog (http://pegasus.phast.umass.edu) and two existing optical redshift databases, CfA2 \\citep{del88} and 2dFGRS (http://www.mso.anu.edu.au/2dFGRS), to obtain very large ($>4000$), if very shallow ($K<13$), $K$-selected redshift samples to derive the local $K$-band luminosity function. However, the mean redshift for these samples is shallow, 0.025 for the CfA sample\\citep{koc01} and 0.05 for the 2df sample\\citep{col01}.    The observational goal of the Hawaii-AAO $K$-band redshift survey was to obtain a large medium-deep $K$-selected galaxy sample with redshifts and optical-infrared colors. Both optical ($B$ and $I$) and near-infrared ($K$) images were taken at Mauna Kea Observatory with total infrared coverage of totaling 8.22 $\\deg^2$ \\citep{hua97}. The spectroscopic observations were carried out on the Anglo-Australian Telescope (AAT) with the Two Degree Field facility (2dF). In this paper we report $K$-band luminosity functions derived from a sub-sample of 1056 bright ($K<15$) galaxies. The median redshift of this sample is 0.138 with the redshift distribution tail extending to z=0.5. In this paper, we adopt the $\\Omega_m=0.3$ and $\\Omega_{\\Lambda}=0.7$ and $H=100~\\rm km~s^{-1}~Mpc^{-3}$ cosmology model. A comparison of the $K$-band luminosity functions derived using different cosmology models is also presented. We briefly summarize the observation and data reduction in $\\S$2; in $\\S$3, we present the $K$-band luminosity functions; in $\\S$4, we compare our luminosity function with other's and discuss the difference; we summarize our results in $\\S$5. ",
        "conclusions": "We present a large and highly complete large $K$-selected redshift sample down to $K\\le 15$, and use it to derive the $K$-band luminosity function using both non-parametric and STY methods. By comparing with previous $K$-band luminosity functions, our $K$-band luminosity function has a significantly brighter characteristic luminosity and steeper slope. The slope is the same as in the optical determinations. The $K$-band luminosity density measured from our sample is 2 times higher than those measured from the 2MASS redshift sample, the largest local $K$-selected redshift sample. We argue that this deeper survey, of comparable depth to the optical and SDSS surveys, is more strictly comparable in luminosity density and that we have in fact resolved the discrepancy of the 2MASS survey. We are also able to re-produce the the 2MASS $K$-band luminosity function using similar sunsample with $z<0.1$.   We also derive the $K$-band luminosity functions for both early and later type galaxies. The early type galaxies have a bright $M^*$ and a flat slop $\\alpha\\sim -1$, while the later type galaxies have  a faint $M^*$ and a steep slope $\\alpha=-1.4$. A steep slope for later type galaxies is also found in the current large optical redshift surveys."
    },
    "0209/astro-ph0209489_arXiv.txt": {
        "abstract": "Future experiments will produce high-resolution temperature maps of the cosmic microwave background (CMB) and are expected to reveal the signature of gravitational lensing by intervening large-scale structures.  We construct all-sky maximum-likelihood estimators that use the lensing effect to estimate the projected density (convergence) of these structures, its power spectrum, and cross-correlation with other observables.  This contrasts with earlier quadratic-estimator approaches that Taylor-expanded the observed CMB temperature to linear order in the lensing deflection angle; these approaches gave estimators for the temperature-convergence correlation in terms of the CMB three-point correlation function and for the convergence power spectrum in terms of the CMB four-point correlation function, which can be biased and non-optimal due to terms beyond the linear order. We show that for sufficiently weak lensing, the maximum-likelihood estimator reduces to the computationally less demanding quadratic estimator.  The maximum likelihood and quadratic approaches are compared by evaluating the root-mean-square (RMS) error and bias in the reconstructed convergence map in a numerical simulation; it is found that both the RMS errors and bias are of order 1 percent for the case of Planck and of order 10--20 percent for a 1 arcminute beam experiment.  We conclude that for recovering lensing information from temperature data acquired by these experiments, the quadratic estimator is close to optimal, but further work will be required to determine whether this is also the case for lensing of the CMB polarization field. ",
        "introduction": "Introduction}  Gravitational weak lensing of the cosmic microwave background (CMB) has been recognized as a potential indicator of large-scale structure in the universe.  Compared to galaxy surveys, weak lensing has the advantage of directly tracing the matter density, thus avoiding the uncertainties associated with the relationship between the distributions of galaxies and of mass \\cite{1999PhRvL..82.2636S}.  Because the CMB is the most distant background object that can be used for weak lensing studies, it probes the matter distribution at higher redshifts than can be reached by galaxy weak lensing and is sensitive to the largest observable scales in the universe \\cite{1999PhRvL..82.2636S,1999PhRvD..59l3507Z,2001PhRvD..63d3501B, 2001ApJ...557L..79H,2000PhRvD..62d3007H}.  In addition to providing data on the power spectrum of density fluctuations on these large scales, CMB weak lensing may yield constraints on the expansion history of the universe by making possible a measurement of the integrated Sachs-Wolfe (ISW) effect.  The ISW effect (the change in temperature of the CMB radiation as it passed through a changing gravitational potential) is smaller than the primary CMB fluctuations produced in the early universe and consequently can be detected only through the cross-correlation of CMB observations with some tracer of the gravitational potential.  Because it is sensitive directly to the potential, weak lensing is an ideal candidate for this cross-correlation \\cite{1999PhRvD..60d3504S,1999PhRvD..59j3002G}.  Because detection of CMB weak lensing may be possible with near-future satellite experiments, such as Planck and possibly even MAP, several algorithms have been proposed for estimating matter distributions, power spectra, and ISW cross-correlations from CMB temperature maps.  Some of these methods are based on local statistics, such as the products of gradients of the temperature field \\cite{1999PhRvL..82.2636S}.  Recently Hu \\cite{2000PhRvD..62d3007H,2001ApJ...557L..79H}, working to linear order in the deflection angle, determined the optimal quadratic estimator (i.e. quadratic in the CMB temperature map) for the deflection field.  Within this linear approximation, the corresponding power spectrum estimator makes full use of the information in the CMB four-point correlation function \\cite{2000PhRvD..62f3510Z,2001PhRvD..64h3005H}.  However, the limits to the validity of the linear order approximation have not been well-determined, and the possibility of obtaining more information on lensing from higher-order correlation functions has not been studied in detail. Neglect of nonlinear terms may also create a bias in the quadratic estimators of the power spectrum. The nonlinear terms may be important whenever the deflection angle is comparable to the scale of CMB fluctuation used in the reconstruction of lensing potential. The deflection angle is of the order of several arcminutes and for high resolution experiments significant amount of lensing information comes from CMB modes on the same scale, indicating that the nonlinear terms may be important. In order to address these issues, we use the likelihood function to construct estimators rather than assuming an estimator with a particular form (local, quadratic, etc.) and avoid linearizing in the deflection field except to compare our results to previous work and where necessary for computational tractability.  We work principally in position space rather than harmonic space.  This is done partly because real data are obtained in position space, and partly to show how the harmonic-space estimators \\cite{2001ApJ...557L..79H} can be derived from position-space arguments; also, the generalization of the position-space analysis to anisotropic instrument noise is more transparent.  We also do not consider the reconstruction of matter distributions from CMB polarization; although polarization can theoretically yield much better information about lensing than CMB temperature fluctuations \\cite{2002ApJ...574..566H}, it is also computationally more demanding, so we defer a more careful analysis to a future work.  We will proceed as follows: Section \\ref{sec:s2} introduces our formalism and notation, and defines the basic mathematical operations that will be used in the rest of the paper.  Section \\ref{sec:s3} considers the likelihood function for the CMB and its dependence on the lensing potential (the potential that generates the deflection field).  In Section \\ref{sec:s4} we consider the maximum likelihood estimators for the power spectrum of the lensing potential and its cross-correlation with the CMB.  In Section \\ref{sec:s5}, we describe our numerical implementation of the estimators from Sections \\ref{sec:s3} and \\ref{sec:s4}; the performance of the estimators, as determined numerically, is described in Section \\ref{sec:s6}.  We conclude in Section \\ref{sec:s7}. ",
        "conclusions": ""
    },
    "0209/astro-ph0209326_arXiv.txt": {
        "abstract": "We use three dimensional hydrodynamic simulations to investigate the effects of a soft X-ray background, that could have been produced by an early generation of mini-quasars, on the subsequent cooling and collapse of high redshift pregalactic clouds.  The simulations use an Eulerian adaptive mesh refinement technique with initial conditions drawn from a flat $\\Lambda$-dominated cold dark matter model cosmology to follow the nonequilibrium chemistry of nine chemical species in the presence of both a soft ultraviolet Lyman-Werner \\HH photodissociating flux of strength $F_{LW}=10^{-21}$~\\fluxunit and soft X-ray background extending to $7.2$~keV including the ionization and heating effects due to secondary electrons. Although we vary the normalization of the X-ray background by two orders of magnitude, the positive feedback effect of the X-rays on cooling and collapse of the pregalactic cloud expected due to the increased electron fraction is quite mild, only weakly affecting the mass threshold for collapse and the fraction of gas within the cloud that is able to cool, condense and become available for star formation. Inside most of the cloud we find that \\HH is in photodissociation equilibrium with the soft UV flux. The net buildup of the electron density needed to enhance \\HH formation occurs too slowly compared to the \\HH photodissociation and dynamical timescales within the cloud to overcome the negative impact of the soft UV photodissociating flux on cloud collapse. However, we find that even in the most extreme cases the first objects to form do rely on molecular hydrogen as coolant and stress that our results do not justify the neglect of these objects in models of galaxy formation. Outside the cloud we find the dominant effect of a sufficiently strong X-ray background is to heat and partially ionize the inter-galactic medium, in qualitative agreement with previous studies. ",
        "introduction": "\\label{sec:introduction}  One of the most important questions in current cosmology is to understand how the cosmological dark ages ended by identifying the nature of the first luminous sources and determining their impact on subsequent structure formation and the reionization of the universe. Recent observations are beginning to constrain the epoch of reionization and give modest information about possible first sources. Observations of metals throughout even low column density \\Lya lines (Ellison \\etal 1999, 2000; Schaye \\etal 2000) suggest the need for an early population of stars to pre-enrich the inter-galactic medium (IGM). Observed spectra of high redshift quasars such as SDSS 1030+0524 (Fan \\etal 2001) at $z=6.28$ and \\Lya emitters at these redshifts (e.g. Hu \\etal 2002) start to constrain the overlap stage of hydrogen reionization. Observations of patchiness in the HeII optical depth in the Lyman alpha forest at $z \\sim 3$ (Reimers \\etal 1997) might signal a recent period of helium reionization. Multi-wavelength observations planned in the near future hope to study the reionization epoch in detail. Direct imaging of quasars and star clusters at $z > 10$ may be possible using the Next Generation Space Telescope (Haiman \\& Loeb 1999, Barkana \\& Loeb 2000). Emission measurements in the 21 cm line using LOFAR and the Square Kilometer Array could identify the first epoch of massive star formation (Tozzi \\etal 2000). Searches for polarization effects and secondary anisotropies in the Cosmic Microwave Background induced by scattering off the increased electron fraction produced during reionization are planned by Planck and the next generation of millimeter telescopes such as ALMA.  In order to correctly interpret the findings of these observations, we need to understand the predictions of the current cosmological structure formation paradigm. The initial stages of collapse of structure at high redshift in cold dark matter cosmologies have been well studied analytically and numerically (see, for example, the excellent review by Barkana \\& Loeb 2001 and references therein). Structure forms from small density perturbations via gravitational instability where smaller clumps merge to form larger clumps within and at the intersections of filaments. By redshifts $30 \\la z \\la 20$ pregalactic clouds with masses (virial temperatures) $M  \\sim 10^5$ - $10^6\\,\\Ms$ ( $T_{vir} \\sim 1000$~\\kel) have formed sufficient molecular hydrogen to begin to cool reducing pressure support in their central regions and allowing their cores to collapse to high density (Haiman, Thoul \\& Loeb 1996; Tegmark \\etal 1997; Omukai \\& Nishi 1998; Abel \\etal 1998; Fuller \\& Couchman 2000). However, it is only recently that 3-D numerical simulations have achieved sufficient resolution to follow this collapse reliably from cosmological initial conditions to the extremely high densities and stellar spatial scales expected for the first luminous sources (Abel, Bryan, \\& Norman 2000, 2001). These  and related studies (e.g. Bromm, Coppi \\& Larson 2001; Nakamura \\& Umemura 2001) show that the final stages of collapse are slow (quasi-static) and that fragmentation in the primordial, metal-free gas is difficult. Thus the first sources were most probably massive stars. This has led to renewed interest in the evolutionary properties of such metal-free, massive stars (Fryer, Woosley \\& Heger 2001; Schneider \\etal 2001; Oh 2001; Oh \\etal 2001). However, the micro-galaxies ($\\ga 10^6\\Ms \\approx 10^{-6}$ times the mass of the Milky Way) which  host the first stars collapse over a large redshift interval. At the same time in different regions of space also rarer but much larger objects are forming whose integrated emitted light cannot be reliably predicted nor constrained strongly from existing observations. Consequently, the radiation spectrum from the first luminous sources is uncertain.  Clearly, a background of soft ultraviolet (UV) radiation is expected from the first stars. The neutral primordial gas remains optically thin to soft UV photons below the hydrogen ionization edge until the gas collapses to high density. In particular photons in the Lyman-Werner bands ($11.2\\,{\\rm eV}\\, < E_\\gamma < 13.6$~eV) can travel large distances and readily photodissociate the fragile \\HH coolant in their own and neighboring clouds via the two-step Solomon process. Such a soft UV background alone is expected to suppress the subsequent collapse of low mass clouds ($T_{vir} < 10^4$~\\kel) that require \\HH to cool (Dekel \\& Rees 1987; Haiman, Rees \\& Loeb 1997; Omukai \\& Nishi 1999; Ciardi \\etal 2000; Haiman, Abel \\& Rees 2000; Glover \\& Brand 2001; Machacek, Bryan \\& Abel 2001; Oh \\& Haiman 2001). In Machacek, Bryan \\& Abel (2001, hereafter referred to as MBAI) we used fully 3-dimensional AMR simulations in the optically thin approximation starting from cosmological initial conditions in a \\LCDM cosmology to follow the evolution and collapse of pregalactic clouds in the presence of varying levels of soft UV flux, $0 \\leq F_{LW} \\leq 10^{-21}$~\\fluxunit. We confirmed that the presence of a  Lyman-Werner flux $F_{LW}$ delays the onset of collapse  until the pregalactic clouds evolve to larger masses and found a fitting formula for the mass threshold for collapse given the mean flux in the Lyman-Werner bands. In MBAI we also investigated what fraction of gas could cool and condense and thus become available for star formation, an important input parameter for semi-analytical models of galaxy formation, stellar feedback and its impact on the process of reionization (Ciardi \\etal 2000, Ciardi, Ferrara, \\& Abel 2000, Madau, Ferrara, \\& Rees 2001). We found that the fraction of gas that could cool and become dense in metal-free pregalactic clouds in our simulations depended primarily on two numbers, the flux of soft UV radiation and the mass of the cloud, and that once above the collapse mass threshold determined by the level of photodissociating flux, the fraction of cold, dense gas available for star formation in these early structures increased logarithmically with the cloud's mass.  If early luminous sources include a population of mini-quasars or other X-ray emitting sources such as X-ray binaries they will produce a background radiation field of soft X-rays with energies above the Lyman limit ($\\ga 1$~keV).  It has been suggested (Haiman, Rees \\& Loeb 1996; Haiman, Abel \\& Rees 2000; Oh 2001; Ricotti, Gnedin \\& Shull 2001) that the increased electron fraction produced by the ionizing photons would promote the formation of \\HH thereby undoing the negative feedback effect of the soft, Lyman-Werner UV flux on the collapse of low mass pregalactic clouds.  Haiman, Rees \\& Loeb (1996) found that the formation rate of \\HH could be enhanced in a dense ($n_H \\ga 1$~cm$^{-3}$), stationary, homogeneous gas cloud of primordial composition irradiated by an external, uniform power-law background flux with photon energies $\\la 40$~keV. Their calculations assumed chemical equilibrium for the species. They used Lepp \\& Shull (1983) cooling functions for molecular hydrogen and mimicked radiative transfer effects by assuming a mean absorbing column density of $10^{22}$~\\cms. Haiman, Abel \\& Rees (2000) again considered \\HH formation in a static, isolated primordial cloud in the presence of a power-law radiation flux extending to $10$~keV. However, they adopted the more realistic profile of a truncated isothermal sphere at its virial temperature and followed the time evolution of nine chemical species.  They concluded that in the cores of these objects the negative effect of the photodissociating flux on collapse is erased if as little as $10\\%$ of the radiation field is from mini-quasars with energies extending into the soft X-ray band. In two recent papers, Ricotti \\etal (2002a, 2002b) have also examined the effect of radiative feedback on cooling in low-mass halos and find that high-energy photons can have a positive net effect on the star formation rate.  In this paper we extend the results of our fully 3-D Eulerian AMR simulations of the formation and collapse of primordial pregalactic structure in the presence of a soft Lyman-Werner UV background (MBAI) to include the contribution of X-rays with energies extending to $7.2$~keV.  This work, as in MBAI, improves upon earlier studies by following the time evolution of a collection of collapsing protogalaxies evolving together in a $1$~Mpc$^3$ (comoving) simulation volume from cosmological initial conditions drawn from a flat \\LCDM model. Thus it treats consistently the density evolution of the cloud and includes the effects of gravitational tidal forces and merging that also impact cooling and collapse. We develop statistics on the amount of gas that can cool due to molecular hydrogen and the fraction of gas that is cold and dense enough to be available for star formation in these objects when exposed to both a soft \\HH photodissociating flux and various levels of X-rays.  We also use the more recent Galli \\& Palla (1998) \\HH cooling functions in this work and fitting functions for the energy deposition from high energy electrons from Shull \\& Van Steenberg (1985) that take into account the primordial composition of the gas.  This paper is organized in the following way: In  \\S\\ref{sec:sims} we review the set-up of our simulations with particular emphasis on our treatment of photoionization and the effects of secondary electrons induced by the X-ray background. In \\S\\ref{sec:data} we discuss our peak identification method and the general characteristics of our simulated data set of pregalactic clouds. In \\S\\ref{sec:fractions} we investigate how varying the intensity of the X-ray background affects the amount of gas that can cool and condense, thus becoming available for star formation, in the presence of both an \\HH photodissociating flux $F_{LW}$ and ionizing X-ray background. In \\S\\ref{sec:profiles} we use radial profiles of cloud properties to elucidate the effects of the competing physical processes important to cooling and collapse. We summarize our results in  \\S\\ref{sec:conclude}. ",
        "conclusions": "\\label{sec:conclude}  In this paper we used high resolution numerical simulations to investigate the effect of radiative feedback on the formation of $10^5$ --$10^7 \\Ms$ pregalactic clouds when the radiation spectrum extends to energies above the Lyman limit ($ \\ga 1$~keV). Such an ionizing X-ray component is expected if the initial mass function of the first luminous sources contains an early generation of miniquasars or very massive stars. The range of pregalactic objects we consider is important because they are large enough to form molecular hydrogen, but too small to cool by hydrogen line cooling. Thus any process that affects the amount of \\HH coolant within the cloud affects its ability to cool, lose pressure support and collapse to high density. The soft, UV flux in the $11$--$13$~eV Lyman-Werner band produced by the first stars can destroy the fragile \\HH in these objects delaying subsequent collapse and star formation until later redshifts when the objects have evolved to larger masses.  We test whether the presence of ionizing X-rays can mitigate or even reverse this effect by increasing the electron fraction in the gas and thus enhancing the formation of molecular hydrogen coolant. Since the relative amplitude of the X-ray to soft UV components in the background spectrum of the first luminous sources is unknown, we study four cases with relative X-ray normalizations ranging from zero to ten for mean soft UV flux at $12.86$~eV of $10^{-21}$~\\fluxunit. We compare these results to the case with no background radiation fields.  We draw our initial conditions from a \\LCDM cosmological model. The simulations evolve the nonequilibrium rate equations for $9$ species of hydrogen and helium including the effects of secondary electrons. A summary of our main findings are as follows: \\begin{itemize} \\item Ionizing X-rays do have a positive effect on subsequent structure formation, but the effect is very mild. Even in the presence of X-rays photodissociation is rapid delaying the collapse of the cloud until later redshifts when larger objects have collapsed. \\item The mass thresholds for gas to cool and for gas to cool and become dense decrease only weakly with increasing X-ray flux up to relative X-ray normalization $\\epsilon =1$ when compared to the case with only a soft UV radiation field, but remain $ \\ga$ a factor three more massive than the mass collapse threshold found when no radiation fields were present. Equivalently, the redshift for collapse decreases weakly with increasing X-ray flux up to $\\epsilon_x = 1$ from that for the case with only a soft UV radiation field, but collapse occurs significantly later than in the case with no background radiation field. \\item The fraction of gas that can cool or cool and become dense (and thus become available for star formation) within a cloud increases with increasing X-ray flux.  We fit these fractions with a simple fitting formula (Equation \\ref{eq:regression}) that increases logarithmically with cloud mass and find that the slope of this fitting formula increases by as much as a factor $\\sim 2$ with increasing X-ray flux for X-ray normalizations $\\epsilon_x \\le 1$. \\item The weak positive effect of the ionizing X-rays appears maximal for relative normalization $\\epsilon_x=1$. For significantly higher X-ray fluxes the positive trends described in the previous two items is reversed. Heating becomes important both within the cloud and in the surrounding intergalactic medium thus weakening the characteristic accretion shock near the virial radius. The mean temperature of the cloud is raised well above its virial temperature causing a significant fraction of the gas to be evaporated into the surrounding intergalactic medium. \\end{itemize}  We conclude that although an early X-ray background from quasars or mini-quasars does enhance cooling in pregalactic objects, the effect is weaker than found in previous studies that did not follow the evolution of the collapsing cloud. The net impact on subsequent structure formation is still negative due to photodissociation of the \\HH coolant by the soft UV radiation spectrum of the first stellar sources, and the pattern of subsequent structure formation is only weakly changed by including an ionizing X-ray component. We should point out that that we have not included star formation and its subsequent effect on the forming halos in these simulations (except insofar as we have modelled the radiative background), so obviously more work on this subject is required.  In this context, it is interesting to note that Ricotti \\etal (2002a, 2002b) have simulated star formation and radiative transfer in somewhat more massive halos (albeit with a mass resolution 100 times lower than used here) and find a self-regulated feedback loop that includes positive feedback.   Together with the findings of MBAI our results suggest that radiative feedback from cosmological radiation backrounds have subtle effects on the formation of luminous objects within the micro--galaxies. The negative feedback of a soft UV background may change the minimum mass of a dark halo within which gas may cool by a factor of a few. However, even in the most extreme cases considered the first objects to form rely on molecular hydrogen as coolant. Hence, our results do {\\sl not} justify the neglect of halos cooling by molecular hydrogen in all current studies of galaxy formation.  We found that heating from an early X-ray background only slightly modifies the temperature and density profiles of halos at the time when a cool core is first formed in their centers. One may speculate that such temperature variations may lead to varying accretion rates onto the proto--star which will form within them (Abel, Bryan \\& Norman 2002). If so early radiation backgrounds may influence the spectrum of initial masses of Population III stars. To answer such detailed questions will rely on carrying out yet higher resolution simulations than the ones presented here.    This work is supported in part by National Science Foundation grant ACI-9619019.  The computations used the SGI Origin2000 at the National Center for Supercomputing Applications. M.E.M. gratefully acknowledges the hospitality and support of the MIT Center for Space Research where most of this work was done."
    },
    "0209/astro-ph0209110_arXiv.txt": {
        "abstract": "The superior resolution of HST and the light gathering power of large 8-10m class telescopes are now providing information on distant globular clusters (GCs) that is comparable to that obtained in early 1990s for Local Group systems. Here I summarise what has been learnt from the imaging and limited spectroscopy of GCs in other galaxies. The GC systems of spirals and ellipticals reveal remarkable similarities. The vast bulk of GCs appear to have formed at early epochs, with mergers making a limited contribution to the overall GC system at later epochs. These observational findings are placed in the context of galaxy formation. ",
        "introduction": "Much has been learnt about the formation and evolution of our Milky Way Galaxy from the system of globular clusters (GCs). It is natural to extend such studies to the GC systems of other galaxies. There are several advantages in doing this:\\\\ $\\bullet$ Greater numbers, eg the MW has only 150 known GCs, M31 has a total population of $\\sim$ 400 and M87 $\\sim$ 10,000.\\\\ $\\bullet$ Probing to higher metallicity GCs.\\\\ $\\bullet$ The Local Group has no giant ellipticals.\\\\ $\\bullet$ No foreground stars in the GC spectra.\\\\ $\\bullet$ Discover something new !\\\\  For Local Group studies (D $\\sim$ 1 Mpc), 3-6m telescopes have collected optical and IR colours, and measured sizes for large numbers of GCs. Spectra also exist in large numbers for V $\\sim$ 16 GCs, giving metallicities, abundances, relative ages and system kinematics. Classic papers for Local Group GCs are those of Brodie \\& Huchra in 1990s (eg Brodie \\& Huchra 1990, 1991) More recently the 0.1$^{''}$ resolution of HST has provided CMD morphologies for several Local Group GCs (see various posters at this conference).  Further afield (eg Virgo/Fornax distance of 15 Mpc) 8-10m telescopes are starting to obtain similar information for more distant GC systems. Optical colours have now been measured for $\\sim$50 systems. IR colours are lagging behind but quickly catching up (see posters by Puzia and Hempel). At 15 Mpc, GCs are partially resolved by HST allowing sizes to be measured (eg Larsen et al. 2001). Typical mags of V $\\sim$ 22 have made obtaining high quality spectra (for ages, abundances etc) and large numbers for system kinematics slow going.  Only a handful of systems have good quality spectra for individual GCs (the vast majority coming from the SAGES project.) A database of imaging and spectral information for extragalactic GCs can be found at: astronomy.swin.edu.au/dforbes ",
        "conclusions": "\\noindent The observations of extragalactic GC systems indicate that:\\\\ $\\bullet$ The inner metal-rich GCs in the Milky Way and other spirals has a bulge (not disk) origin.\\\\ $\\bullet$ The GC systems of spirals and ellipticals show remarkable similarities.\\\\ $\\bullet$ Blue and red GCs have similar ages $\\sim$ 12 Gyrs (but red GCs could be younger by 2--4 Gyrs).\\\\ $\\bullet$ Some young ($\\sim$ 2-8 Gyrs) GCs have been found in `old' ellipticals.\\\\ $\\bullet$ Blue and red GCs appear to have supersolar alpha ratios.\\\\ $\\bullet$ Red GCs trace elliptical galaxy star formation.\\\\  \\noindent The modelling of GC formation in a $\\Lambda$CDM universe indicates that:\\\\ $\\bullet$ Blue GCs formed $\\sim$ 12 Gyrs ago in all ellipticals.\\\\ $\\bullet$ Red GCs have a mean age of 8--10 Gyrs in field ellipticals.\\\\ $\\bullet$ S$_N$ is driven by the number of blue GCs in a galaxy and is largely determined at early epochs; late stage mergers have little effect on S$_N$.\\\\ $\\bullet$ In general, spirals and ellipticals have similar GC systems.\\\\  \\noindent Areas of research that we will no doubt hear more about in the near future are:\\\\ $\\bullet$ 8m wide field K band imaging studies (eg Puzia et al. 2002).\\\\ $\\bullet$ Age structure within the red GCs (Beasley et al. 2002).\\\\ $\\bullet$ Nature of the large (R$_{eff}$ $\\sim$ 10 pc) red low luminosity (M$_V$ $\\sim$ --6) clusters (Brodie this meeting).\\\\ $\\bullet$ Importance of shredded dwarf galaxies (eg Bekki et al. 2001).\\\\ $\\bullet$ HST+ACS CMDs for Local Group GCs (eg Rich this meeting).\\\\ $\\bullet$ Halo mass estimates, GC kinematics vs Xrays (eg Kissler-Patig et al. 1998).\\\\ $\\bullet$ Continued improvements in SSP grids (eg Maraston this meeting)\\\\"
    },
    "0209/astro-ph0209260_arXiv.txt": {
        "abstract": "Due to their star formation history (SFH), the stellar population in Dwarf Galaxies (DG) is likely to have a metallicity spread which is best traced by the morphology of the Red Giant Branch (RGB). We probe here a purely empirical approach aimed at estimating average metallicity ($Z$) and $Z$ spread by comparing the Color-Magnitude Diagrams (CMD) of galactic Globular Clusters (GCs) with those of two DGs:  Leo A (HST data) and Phoe (VLT Fors2 data). ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209056_arXiv.txt": {
        "abstract": "\\noindent We compute the reionization histories of hydrogen and helium due to the ionizing radiation fields produced by stars and quasars.  For the quasars we use a model based on halo-merger rates that reproduces all known properties of the quasar luminosity function at high redshifts.  The less constrained properties of the ionizing radiation produced by stars are modeled with two free parameters: (i) a transition redshift, $z_{\\rm tran}$, above which the stellar population is dominated by massive, zero-metallicity stars and below which it is dominated by a Scalo mass function; (ii) the product of the escape fraction of stellar ionizing photons from their host galaxies and the star-formation efficiency, $f_{\\rm esc}f_{\\star}$.  We constrain the allowed range of these free parameters at high redshifts based on the lack of the HI Gunn-Peterson trough at $z\\la 6$ and the upper limit on the total intergalactic optical depth for electron scattering, $\\tau_{\\rm es}<0.18$, from recent cosmic microwave background (CMB) experiments.  We find that quasars ionize helium by a redshift $z\\sim 4$, but cannot reionize hydrogen by themselves before $z\\sim 6$. A major fraction of the allowed combinations of $f_{\\rm esc}f_{\\star}$ and $z_{\\rm tran}$ lead to an early peak in the ionized fraction due to metal-free stars at high redshifts.  This sometimes results in two reionization epochs, namely an early HII or HeIII overlap phase followed by recombination and a second overlap phase. Even if early overlap is not achieved, the peak in the visibility function for scattering of the CMB often coincides with the early ionization phase rather than with the actual reionization epoch. Consequently, $\\tau_{\\rm es}$ does not correspond directly to the reionization redshift. We generically find values of $\\tau_{\\rm es}\\ga 7\\%$, that should be detectable by the MAP satellite. ",
        "introduction": "Following cosmological recombination at a redshift $z\\sim 10^3$, the baryonic gas filling up the universe became predominantly neutral.  Given that this gas is known to be mostly ionized today, one arrives at two of the major questions in current extragalactic astronomy: {\\it (i) when were the cosmic hydrogen and helium re-ionized?} and {\\it (ii) which sources dominated this reionization process?} The answers to both questions are likely to be different for hydrogen and helium (see review by Barkana \\& Loeb 2001). Recent observations provide preliminary answers to the first question.  The absorption spectra of SDSS quasars at $z\\sim 6$ indicate that the neutral fraction of hydrogen increases significantly at $z\\ga 6$ (Becker et al. 2001; Fan et al. 2002), and the UV spectrum of quasars implies that helium is fully ionized only at $z\\la 3$ (Jacobsen et al 1994; Tytler 1995; Davidsen et al. 1996; Hogan et al. 1997; Reimers et al. 1997; Heap et al. 2000; Kriss et al. 2001; Smette et al. 2002).  The latter observations also indicate, through a cross-correlation between the hydrogen and helium forest of absorption lines (Kriss et al. 2001; Smette et al. 2002), that the ionization of helium at $z\\sim 3$ had significant contributions from both quasars and stars. No analogous evidence exists for hydrogen at $z\\ga 6$.  In this paper we make theoretical predictions for the reionization histories of hydrogen and helium. We calculate the contribution from quasars using a model that matches all the existing observational data on the quasar luminosity function at high redshifts (see Wyithe \\& Loeb 2002 for details).  We model the less constrained stellar contribution using two free parameters, which are in turn constrained by existing observational data. Since the recombination times of both hydrogen and helium are shorter than the age of the universe in overdense regions of the intergalactic medium (IGM), it is possible that these species experienced more than one epoch of reionization.  One of the goals of our detailed study is to identify the parameter values for which multiple reionization epochs are possible.  The reionization history has important implications for the temperature and polarization anisotropies of the cosmic microwave background (CMB). Anisotropies on scales smaller than the angular size of the horizon at hydrogen reionization are suppressed by a factor $\\sim e^{-\\tau_{\\rm es}}$ where $\\tau_{\\rm es}$ is the line-of-sight optical depth for electron scattering (Hu \\& White 1997; Haiman \\& Loeb 1997), and secondary anisotropies are added (Hu 2000 and references therein).  Reionization is the primary source of polarization anisotropies on large angular scales (Zaldarriaga \\& Seljak 1997; Kamionkowski et al. 1997; Hu 2000). Polarization anisotropies might be detected in the near future by MAP on large angular scales (Kaplinghat 2002) or by ground-based experiments on small angular scales. So far, the available data on the temperature anisotropies of the CMB provides an upper limit of $\\tau_{\\rm es}\\la 0.18$ (Wang, Tegmark \\& Zaldarriaga 2002; Bond et al. 2002).  We will use this upper limit to constrain the free parameters of our model.  Throughout the paper we assume density parameters values of $\\Omega_{m}=0.35$ in matter, $\\Omega_{b}=0.052$ in baryons, $\\Omega_\\Lambda=0.65$ in a cosmological constant, and a Hubble constant of $H_0=65~{\\rm km\\,s^{-1}\\,Mpc^{-1}}$ (or equivalently $h=0.65$). For calculations of the Press-Schechter~(1974) mass function (with the modification of Sheth \\& Tormen~1999) we assume a primordial power-spectrum with a power-law index $n=1$ and the fitting formula to the exact transfer function of Cold Dark Matter, given by Bardeen et al.~(1986).  Unless otherwise noted we adopt an rms amplitude of $\\sigma_8=0.87$ for mass density fluctuations in a sphere of radius $8h^{-1}$Mpc. ",
        "conclusions": "We have explored the reionization histories of hydrogen and helium due to stars and quasars. The results were analyzed as a function of the two free parameters in our model, namely: (i) the transition redshift, $z_{\\rm tran}$, above which the stellar population is dominated by massive, zero metallicity stars; and (ii) the product of the escape fraction of ionizing photons and the star-formation efficiency, $f_{\\rm esc} f_\\star$.  The quasar model was not varied since it provides an excellent fit to all existing data on the luminosity function of quasars up to redshift $z\\sim 6$ (Wyithe \\& Loeb 2002).  Figures~\\ref{fig2} and \\ref{fig3} show sample reionization histories for different choices of $z_{\\rm tran}$ and $f_{\\rm esc}f_\\star$. We find that a wide range of $z_{\\rm tran}\\ga 6$ is allowed for $f_{\\rm esc}f_\\star\\sim 2\\times10^{-3}$ if the star-formation efficiency does not depend on galaxy mass (figure~\\ref{fig4}) or $f_{\\rm esc}f_\\star\\sim 10^{-2}$ if the efficiency is suppressed in low mass galaxies (figure~\\ref{fig5}). This wide range satisfies the constraints that overlap of the HII regions must be achieved by $z\\sim 6$ (Fan et al 2002) and that the optical depth for electron scattering must be limited to $\\tau_{\\rm es}<0.18$.  The allowed range leads generically to $\\tau_{\\rm es}\\ga 7\\%$ (see figures~\\ref{fig4} and \\ref{fig5}). The MAP satellite is expected to have sufficient sensitivity to detect values of $\\tau_{\\rm es}$ as small as $\\sim 5\\%$ (Kaplinghat et al. 2002), well below this range of expected values.  A major fraction of the allowed range of $f_{\\rm esc}f_{\\star}$ and $z_{\\rm tran}$ leads to an early peak in the ionized fraction due to metal-free stars at high redshifts.  Often this peak results in a small but significant filling factor that is subsequently reduced (temporarily) due to recombination. In a restricted range of the allowed parameter values, we find that either hydrogen or helium experience two overlap epochs, separated by recombination (see shaded regions in figures~\\ref{fig6} and \\ref{fig8}). That helium might have been reionized twice due to the presence of early very massive stars was previously pointed out by Oh et al.~(2001). The first overlap phase is caused by the population of zero-metallicity, massive stars and the second is dominated by the quasars for helium or by stars and quasars for hydrogen.  Even if early overlap is not achieved, the peak in the visibility function for scattering of the CMB often coincides with the early partial ionization peak rather than with the actual reionization epoch (see figures~\\ref{fig7} and \\ref{fig9}). The resulting value of $\\tau_{\\rm es}$ is therefore larger than expected based the reionization redshift alone.  Thus, the CMB visibility function may be probing the nature of the early generation of stars rather than the reionization epoch itself.  Future CMB experiments, such as MAP and Planck, will provide tighter constraints on $\\tau_{\\rm es}$ and will reduce the range of allowed $f_{\\rm esc}f_\\star$ in our model (see figures~\\ref{fig4} and \\ref{fig5}). Any additional observational data on the composition of the stellar population or the abundance of HeI or HeII at $z\\ga 4$ will provide stronger lower limits on $z_{\\rm tran}$."
    },
    "0209/astro-ph0209370_arXiv.txt": {
        "abstract": "We present {\\em Chandra} X-ray observations of the supernova remnant (SNR) N63A in the Large Magellanic Cloud (LMC). N63A, one of the brightest LMC remnants, is embedded in an H {\\sc ii} region and probably associated with an OB association. The optical remnant consists of three lobes of emission contained within the approximately three times larger X-ray remnant. Our \\chandra\\ data reveal a number of new physical structures in N63A.  The most striking of these are the several ``crescent''-shaped structures located beyond the main shell that resemble similar features seen in the Vela SNR. In Vela, these have been interpreted as arising from high speed clumps of supernova ejecta interacting with the ambient medium. Another distinct feature of the remnant is a roughly triangular ``hole'' in the X-ray emission near the location of the optical lobes and the brightest radio emission.  X-ray spectral analysis shows that this deficit of emission is a result of absorption by an intervening dense cloud with a mass of $\\sim$450 $M_\\odot$ that is currently being engulfed by the remnant's blast wave. We also find that the rim of the remnant, as well as the crescent-shaped features, have considerably softer X-ray spectra than the interior.  Limits on hard X-ray emission rule out a young, energetic pulsar in N63A, but the presence of an older or less active one, powering a wind nebula with a luminosity less than $\\sim$$4\\times 10^{34}$ erg s$^{-1}$, is allowed. ",
        "introduction": "Supernova remnants (SNRs) are valuable tools for understanding the composition of the interstellar medium (ISM), recent star formation, and supernovae (SNe) themselves.  The structure of SNRs and their interactions with the surrounding medium give us insight into their origin and effects on their environment.  The second brightest X-ray SNR in the Large Magellanic Cloud (LMC), N63A, provides us with an excellent laboratory for studying such structures and interactions. This remnant is embedded in a larger H {\\sc ii} region, N63, and appears to be located within the OB association NGC 2030 \\citep{chu2}. N63A is believed to be the product of the explosion of a massive star in a dense and complex environment \\citep{shull,hughes} and is the first confirmed SNR in an H {\\sc ii} region \\citep{shull}.  The X-ray size of the SNR ($34^{\\prime\\prime}$ or 8.2 pc radius for an LMC distance of 50 kpc) is about three times the size of the optical remnant, which appears as a three-lobed structure \\citep{mathewson}. Two lobes are shock-heated, while the third (southwestern) is a photoionized H {\\sc ii} region \\citep{levenson}.  The shock-heating is a result of interactions with the ISM, rather than with SN ejecta, based on the derived abundances \\citep{rusdop90}.  Similarly, the X-ray emission from N63A is also consistent with swept-up ISM (Hughes et al.~1998). In the radio, the remnant appears as a thick shell, with the brightest emission corresponding to the two shock-heated eastern lobes of optical emission \\citep{dickel}.  N63A was shown by \\citet{graham} to be a significant IRAS source, with $L_{\\rm IR}=1.5\\times10^5 L_\\odot$.  However, the Columbia CO survey of the LMC revealed no significant molecular gas associated with the SNR \\citep{cohen} and no CO emission was detected in a higher angular resolution pointed observation by SEST \\citep{israel}.  Hughes et al.~(1998) estimate the age of N63A to be in the range 2000-5000 yr.  We targeted N63A with \\chandra\\ to study the interaction of the SNR with the ISM, to find evidence for SN ejecta, and to search for a compact remnant and its associated pulsar wind nebula (PWN).  Here we report on the most prominent new features of the remnant derived from the \\chandra\\ imaging and spectral data. ",
        "conclusions": "The crescent-shaped plumes in N63A resemble features seen in the {\\em ROSAT} image of the Vela supernova remnant \\citep{asch}.  These are the only SNRs where such features have been detected.  The plumes in N63A extend from the main shock boundary, distorting the otherwise very symmetrical shape of the remnant.  The largest one in the southwest reaches approximately 4.2 pc (in projection) beyond the main shock wave, the one just north of this protrudes about 3.5 pc, and the smaller northern region about 1.6 pc.  The lateral size of the largest crescent feature is about 6 pc measured halfway out from the shell. These sizes should be compared to the 8.2 pc radius of the nearly circular main blast wave of N63A.  Three other crescent features appear closer in projection to the main shell in the northeast, east, and south.  \\citet{wang} use two-dimensional hydrodynamic simulations to model the formation of such structures.  They find that dense, high velocity clumps of ejecta may protrude beyond the forward shock as the remnant interacts with the surrounding medium.  Shocks moving through the clump at first crush it and then cause it to expand laterally, eventually taking on the shape of a crescent, such as is seen in N63A.  The origin of crescent-shaped features as ejecta is supported by \\citet{miyata}, who found an overabundance of Si in their study of Vela shrapnel A.  {\\em ASCA} and \\chandra\\ observations of Vela bullet D show strong O, Ne, and Mg emission \\citep{slane,plucinsky}, suggesting an origin as a dense knot of supernova ejecta.  However, \\citet{plucinsky} argue that a nearly-solar abundance plasma far from ionization equilibrium is a more likely explanation for their observations. N63A is about 200 times more distant than Vela, limiting our ability to do a similar detailed spectral study. We find that none of the crescent regions in N63A shows strongly enhanced abundances, although there is weak evidence for enhanced Ne, Mg, and Si line emission at the apex of the northern crescent (compared to the sides). We can confidently conclude that the plumes are not {\\em dominated} by ejecta.  Thus, for the high-speed ejecta clump scenario to be correct, the clumps in N63A must be significantly mixed with ambient material, perhaps signalling the onset of their fragmentation and destruction.  On the other hand, our image analysis has revealed that both the crescent-shaped regions and the entire rim of the SNR show considerably softer spectra than average for N63A.  This argues for a common origin in terms of shocked ISM.  However, confirmation of this hypothesis will require more careful study of spatially resolved X-ray spectra from these regions. Based on models of SNR evolution as well as other \\chandra\\ observations (e.g. Rakowski et al.~2002), we expect the most rapid changes in the thermodynamic state of the post-shock gas to occur immediately behind the blast wave, and it is at the projected rim where these variations can be most cleanly resolved.  It is somewhat puzzling, therefore, that our initial fits for the integrated soft rim spectrum appear to support a relatively lower column density, which, if taken literally, would imply a screen of absorbing matter covering most of the SNR that falls off near the projected rim.  We find this to be a rather contrived geometry and believe the low fitted value of $N_{\\rm H}$, as well as the higher fitted value of $kT$, to be spurious and a result of our oversimplified spectral modelling, i.e., the use of a single component NEI planar shock model for a region with spatially varying spectral characteristics.  In fact when the rim region is examined in finer detail, it is apparent that there is spectral evolution of the ionization state with position behind the shock front on scales of an arcsec or so, qualitatively showing the advance of post-shock ionization state. Further work is in progress to study this effect and use it to quantify the dynamical state of the remnant.  We interpret the hard region to be absorption by an intervening cloud in the LMC with a mass of $\\sim$450 $M_\\odot$, determined by integrating the $N_{\\rm H}$ map over the extraction region for the spectrum in Figure 3c.  If we assume that the cloud's depth along the line-of-sight is roughly half of its length (3.8 pc), which is reasonable given our discussion below, then the mean density of the cloud is on the order of 250 cm$^{-3}$. This value is consistent with estimates of $\\sim$50--300 cm$^{-3}$ (see Shull 1983 and references cited therein) for the preshock density of the optical remnant.  Two other smaller absorbing clouds, north and northeast of center, are roughly 20-30 $M_\\odot$ in size.  The large region of excess absorption is situated in nearly the same area as the bright radio and optical emission from N63A (see Figure 4).  To explain the X-ray absorption, the cloud must be on the near side of N63A with little to no residual X-ray emission coming from in front of it.  The radio and optical emission signal the presence of slow shocks (several 100 km s$^{-1}$ or less) driven into the dense gas of the cloud.  However, the modest extinction to the optical knots (consistent with the mean $N_{\\rm H}$ to N63A rather than the higher value associated with the cloud) means that we are predominantly seeing the shocks being driven into the {\\em near} side of the cloud. This is further supported by the observation of only blue-shifted emission from the shock-heated optical lobes \\citep{shull}.  In essence, then, we are viewing the dense cloud as it is being engulfed by the remnant's blast wave.  The clumpy, knot-like features apparent in the absorption map and {\\em HST} image suggest that the dense cloud's interaction with the blast wave is causing it to break-up and fragment. Indeed the highest measured radial velocities in the optical band (up to 250 km s$^{-1}$) appear mainly around the edges of the lobes \\citep{shull}, while the interior portions appear to be moving toward us at less than half this speed. Numerical simulations of an ISM cloud overrun by a SN blast-wave (e.g., Klein, McKee, \\& Colella 1994) show, rather generically, that the edges of the cloud are accelerated more rapidly (hence attaining higher speeds) than the cloud core. When viewed perpendicular to the direction of motion of the blast wave, this results in a characteristic shape for the shocked cloud: the edges are swept forward along with the blast wave, while the core trails behind. An example of this type of interaction can be seen in an isolated cloud on the eastern limb of the Cygnus Loop \\citep{fesen}. If the shocked cloud is viewed {\\em along} the direction of the blast wave's motion, then the edges of the cloud should show higher radial velocities than the core.  This situation appears, at least qualitatively, to describe the optical characteristics of N63A. Further studies of this remnant across the wavebands should help to confirm this scenario and provide deeper insights into the interaction of SNRs with their environments."
    },
    "0209/astro-ph0209146_arXiv.txt": {
        "abstract": "A model is presented that explains the `torsional oscillation' pattern of deviations in the solar rotation rate as a geostrophic flow. The flow is driven by temperature variations near the surface due to the enhanced emission of radiation by the small scale magnetic field. The model explains the sign of the flow, its amplitude and the fact that the maxima occur near the boundaries of the main activity belts. The amplitude of the flow decreases with depth from its maximum at the surface but penetrates over much of the depth of the convection zone, in agreement with the data from helioseismology. It predicts that the flow is axisymmetric only on average, and in reality consists of a superposition of circulations around areas of enhanced magnetic activity. It must be accompanied by a meridional flow component, which declines more rapidly with depth. ",
        "introduction": "The so-called torsional oscillation (Howard and LaBonte 1980) was discovered as a small time- and latitude-dependent modulation of the rotational velocity of the Sun as measured from Doppler shifts. In averages of the azimuthal component of the solar surface velocity field (synoptic maps) it appears as a band of  increased velocity that drifts towards the equator during the sunspot cycle, together with the magnetic activity. Its amplitude is of the order $\\sim 5$ m/s, and is most prominent on the equatorward side of the main activity belt.  The obvious connection of the flow with the solar cycle, and its relatively small amplitude, suggest that it may be a secondary effect, somehow caused by the magnetic fields that are the main manifestation of the cycle\\footnote{If this is the case, the term `oscillation' would be somewhat misleading, as it suggests a cyclic variation due to some restoring force intrinsic to the flow itself.}. One possibility would be that it is driven by uncompensated Lorentz forces; this hypothesis has been put forward immediately after the discovery of the oscillation (Sch\\\"ussler, 1981), and has been the basis of several subsequent interpretations. An exception is Ulrich's (2001), which proposes that a hydrodynamic mode of the solar envelope must play an essential role, since the flow  is already present at activity minimum, when no active regions are present. The energy in the oscillation is small, however, compared with the inferred magnetic energy of the cycle, hence the suggestion in Ulrich et al. (2002) that the oscillation is in step with the cycle because it triggers the eruption of active regions.  Instead of being a direct consequence of Lorentz forces, the oscillation could be the secondary result of a thermal effect of the cycle's magnetic fields. The field wound up by differential rotation in the interior of the convection zone could cast a `thermal shadow' on the layers above it by interfering with the efficiency of convection (Spruit, 1977; Parker, 1987). Gilman (1992) proposes that Coriolis forces acting on the downflow created by such a shadow would set up a pattern of differential rotation. The advantage of this idea is that it produces a concentration of the flow at the boundaries of the active latitude belt, as observed. It is not clear if thermal shadows would be of sufficient amplitude for this to work, especially if the shadowing takes place near the base of the convection zone. In a turbulent diffusion model the temperature changes due to shadowing are quite small, because of  the very high effective thermal conductivity of the convective envelope (Spruit, 1977). They are likely to be even smaller (Spruit, 1997) when the extreme asymmetry between upward and downward flows due to the density stratification of the envelope is taken into account, since this causes the upflows to be much more identical in entropy than in mixing length models (for results of recent numerical simulations see Stein and Nordlund (1998), Asplund et al. (2000).  The bulk of the flux and energy of the solar magnetic field is believed to be located near the base of the convection zone (D'Silva and Choudhuri, 1993; Caligari et al., 1994; Fan et al., 1994). These authors have shown that the phenomenology of magnetic fields observed at the surface can be quantitatively understood as due to the eruption of the loops from this magnetic layer at the base, as in Leighton's (1969) classical interpretation of the solar cycle. Models in which the torsional oscillation is a consequence of magnetic fields therefore typically also place the source of the oscillation in the deeper layers of the convection zone.  With only surface observations available, this prediction could not be tested but with the detailed measurements of the Sun's internal rotation made possible by helioseismology this has now changed. The torsional oscillation has been clearly detected in the variation of the rotation profile during the cycle (Woodard and Libbrecht, 1993b; Basu and Antia, 1998, Schou et al., 1998). Perhaps surprisingly, the oscillation appears to have its largest amplitude at the solar surface. In the most detailed results so far (Howe et al., 2000, Vorontsov et al., 2002) the pattern can be followed to a depth of about 100\\,000 km. The systematic decline of the amplitude with depth, however, is hard to reconcile with an origin in the deeper layers of the convection zone.  The new interpretation proposed in this paper takes the apparent surface origin seriously. It proposes that the oscillation actually is i) a secondary consequence of the cycle's magnetic fields, and ii) is caused by the {\\it thermal effect} which small scale surface magnetic fields of the cycle have on their surroundings, not the Lorentz force. As shall be demonstrated below, this provides a simple and easily quantifiable explanation for the oscillation. It agrees with most of its observed properties and makes detailed predictions.  The key element of the theory is the observation that horizontal flows on length scales like those of the oscillation are dominated by the Coriolis force, hence approximately {\\it geostrophic}, like large scale flows in the Earth's atmosphere and oceans.  \\subsection{Relation of the oscillation to the magnetic field}  The oscillation pattern travels with the main activity belt in a butterfly diagram, but its maximum does not coincide with that of the magnetic activity. To illustrate the well known relation between the field and the flow, I reproduce  in Figure \\ref{relat} a synoptic map based on the observations made at Mt. Wilson from 1986 till 1999 (Ulrich 2001).  \\begin{figure} \\mbox{}\\hfill\\epsfxsize 0.5\\hsize\\epsfbox{bav.eps} \\hfill\\epsfxsize 0.5\\hsize \\epsfbox{vav.eps} \\hfill\\mbox{}\\break \\mbox{}\\hfill\\epsfxsize 0.8\\hsize\\epsfbox{uplo.eps}\\hfill\\mbox{} \\caption{Top panels: synoptic maps of the absolute value of the magnetic field (left) and azimuthal velocity residual (right) measured at Mt. Wilson Observatory (Ulrich, 2001). Bottom panel: composite map in which the absolute value of the magnetic field is shown in intensity and deviations in rotation velocity in color, with blue (red) for faster (slower) rotation than the mean. Color saturation codes for the velocity amplitude. Data have been averaged over 3 rotations.\\label{relat}} \\end{figure}  Deviations in rotation velocity are clearly associated with the magnetic activity, but avoid the center of the main belt of activity. As the lower panel shows, the velocity deviations correlate with the latitudinal {\\it gradient of the magnetic activity}. On the leading (equatorial) side of the belts the gradient is strongest, and the flows most pronounced. On the trailing side, the flows are of opposite sign. The boundary of the activity belt is more diffuse on this side, and the flows are of lower amplitude.  The representation in Fig. 1 is chosen to emphasize the relation which I want to draw attention to. It  somewhat deemphasizes a known problem with the oscillation, namely the fact that it starts before the magnetic field of the new cycle becomes noticeable on synoptic magnetograms. This problem is intrinsic to explanations that assume the cycle's magnetic field to be the cause of the zonal flows: the flows are quite strong already before the new cycle, as measured by the main magnetic activity indicators, sets in. I return to this problem in the discussion section.  The correspondence between velocity and gradient of the (absolute value of the) field strength suggests that the flows are geostrophic, like the large scale flows in the earth atmosphere and oceans. In these flows, there is an approximate balance between the horizontal pressure gradient and the Coriolis force, while the contribution of small scale (`turbulent') momentum transport processes is small on these length scales. The flow speed is therefore proportional to the horizontal pressure gradient. The direction of the flow is perpendicular to the horizontal pressure gradient, leading to the characteristic cyclonic and anticyclonic flows circling around low and high pressure systems, respectively (for a systematic treatment see Pedlosky, 1982).  On the Sun, the pressure gradients driving geostrophic flows could be thermal, as in the geophysical case, or could have a magnetic origin. The magnetic field could cause flows directly, through the Lorentz force. A second possibility is an indirect, thermal effect they have on their surroundings; an example is the thermal shadow effect mentioned. Another is the well known surface effect of magnetic fields: by inhibiting convection (in spots) or facilitating radiative loss (in the small scale field), magnetic fields reduce or enhance the radiative cooling of the solar surface. This effect is entirely passive: the energy content of the magnetic structures is not affected, they only act as `valves' changing the radiative energy flux.  These thermal surface effects, though indirect, are not small. The excess emission due to small scale magnetic fields persists for as long as the magnetic structures are present on the surface. The total amount of excess energy emitted at the surface by a small flux tube during it life can therefore be much larger than its total magnetic energy content (e.g. Spruit et al. 1991). I show in this paper that this thermal effect produces flows of the observed nature and magnitude. ",
        "conclusions": "I  have presented a geostrophic flow model for the torsional oscillation, which identifies it as a consequence of enhanced surface cooling by the small scale magnetic field in active regions. Its achievements are: \\begin{itemize} \\item{} The model is quantitative and makes detailed predictions. \\item{} It explains the sign of the oscillation and its concentration along the boundaries of the active latitude belt. \\item{} Using observed values for the enhanced emissivity of active regions, and a standard value for the turbulent viscosity due to granulation, it reproduces the correct amplitude of the flow. \\item{} It explains why the flow has its maximum amplitude at the surface, and reproduces the approximate depth dependence as measured with helioseismology. \\end{itemize} The accuracy of the model can still be improved with better input data. The most critical data it needs are the bolometric brightness contrast of active regions, more generally: all small scale magnetic fields including ephemeral regions and the network. The surface observations required for this currently have limited accuracy, or they rely on proxy indicators like Calcium emission. This will probably improve in the future, since these measurements are the same as are needed for an improved identification of the cause of  solar irradiance variations.  The main difficulty the model faces is that the correct correlation of the oscillation with magnetic activity (as seen in magnetograms) breaks down around minimum activity, when the flow is already strong but little evidence for magnetic fields of the new cycle is seen. This problem is addressed below in \\ref{early}.  \\subsection{Active regions as cooling agents}  The model  is based on the fact that active regions must be areas of lower temperature below the surface. This is perhaps somewhat counter to the prevailing intuition. One might feel that, since active regions are brighter, they should have higher temperatures. This intuition is also somewhat rooted in the idea that the energy flux of the convection zone comes from below, so that it is natural to associate higher surface temperatures with higher heating from below.  This intuition is misleading, however. In a steady state, the heating from below is equal to the cooling at the surface by radiation into space, and it is hard to say which of the two is more fundamental in determining the structure of a convective envelope. In a time dependent situation, the two are generally not equal. Which of the two agents, heating from below and cooling at the top, provides the most appropriate physical view when considering {\\em changes} in the convection zone depends on the {\\em time scale} and location of the phenomenon studied. On the time scale of stellar evolution, it is the change in conditions at the center that determine the changing luminosity of the star. The thermal time scale of the star, $\\approx 10^7$ yr, is very long compared to the phenomena of solar activity, however. Even the thermal time scale of the convection zone alone, $\\sim 10^5$ yr, is very long in this respect.  Because of this, changes in surface brightness due to variations in conditions in the deeper layers of the convection zone, though not impossible, are much harder to bring about than changes in brightness caused by surface effects. The darkness of sunspots due to the suppression of convection by their magnetic field is a well-known example of such a surface effect. The opposite of this effect happens in small scale magnetic fields, which act as leaks in the surface through which an increased heat flux escapes from the convection zone (Spruit 1977). These effects can be viewed simply as modulations of the effective emissivity of the solar surface, i.e. changes in the surface heat flux at {\\it fixed} thermal conditions in the {\\it bulk} of the convection zone.  The large differences in time scale and energy required for thermal changes in the bulk compared with changes at the surface is due to the enormous stratification of the density through the convection zone, by a factor of $10^6$. This has no equivalent in the laboratory- and kitchen analogs of convecting fluids on which conventional intuition is based.  \\subsection{Flows: a proxy for surface temperature variations} As shown above (\\ref{sol}) the velocity amplitude produced by a 0.1\\% change in surface  emissivity is of the order 5 m/s, for a nominal horizontal length scale of $3\\,10^{10}$ cm. Velocity amplitudes of this magnitude are easily detected with current Doppler and helioseismic measurements. The brightness changes themselves are much more difficult to observe directly. So far, the sensitivity of  the order 0.03\\% needed for such a detection has been achieved only with disk-integrated measurements like the ACRIM irradiance data. Surface flows like the torsional oscillation may therefore be a more sensitive way of detecting surface temperature changes associated with the solar cycle.  \\subsection{The early start of the flow pattern} \\label{early} A difficulty with the model presented here is the strength of the observed oscillation at high latitudes, before the beginning of the spot cycle. In synoptic magnetograms, there is very little evidence of magnetic fields at these latitudes and phase in the cycle (e.g. Ulrich, 2001, Ulrich et al. 2002). The model presented here predicts that substantial magnetic flux is in fact present there, but in a form which somehow escapes detection in synoptic magnetograms.  This is not a new suggestion. Significant coronal activity at high latitudes before the beginning of the spot cycle has been noticed already in the 60's, in particular in the Pic du Midi observations of Trellis et al. (cf.\\ Leroy and Trellis, 1974). It has also been consistently seen in the coronographic observations at Sacramento Peak (Altrock, 1986; Guhathakurta et al., 1993). Their possible connection with the torsional oscillation was first proposed by Wilson et al. (1987).  The p-mode frequency shifts during the activity cycle show a similar effect: the latitude dependence of the sources derived from these shifts agrees well with the surface distribution of magnetic fields (Antia et al. 2001), except at high latitude, where the effect is stronger than expected from magnetograms (Moreno-Insertis and Solanki, 2000). The inferred high-latitude magnetic fields peak around activity minimum (Dziembowski and Goode, 2002; Goode et al., 2002).  The form in which these high latitude fields can exist is significantly constrained by their absence in synoptic magnetograms. They would have to exist in the form of small scale flux with polarities nearly balancing within the pixel scale of the synoptic maps. High-resolution observations at the relevant latitudes around solar minimum could test this possibility. Ephemeral active regions, known to have a wider latitude distribution than regular active regions and to be most prominent at activity minimum (Harvey et al. 1986, Harvey 1989), may be a manifestation of these fields (Wilson et al. 1987). It is also possible that magnetograms are less sensitive to magnetic flux in very small scale form, due to radiative transfer effects (e.g.\\ S{\\' a}nchez Almeida 2000)."
    },
    "0209/gr-qc0209101_arXiv.txt": {
        "abstract": "We extend and improve earlier estimates of the ability of the proposed LISA (Laser Interferometer Space Antenna) gravitational wave detector to place upper bounds on the graviton mass, $m_g$, by comparing the arrival times of gravitational and electromagnetic signals from binary star systems. We show that the best possible limit on $m_g$ obtainable this way is $\\sim 50$ times better than the current limit set by Solar System measurements. Among currently known, well-understood binaries, 4U1820-30 is the best for this purpose; LISA observations of 4U1820-30 should yield a limit $\\approx 3-4$ times better than the present Solar System bound. AM CVn-type binaries offer the prospect of improving the limit by a factor of $10$, {\\it if} such systems can be better understood by the time of the LISA mission. We briefly discuss the likelihood that radio and optical searches during the next decade will yield binaries that more closely approach the best possible case. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209236_arXiv.txt": {
        "abstract": "We investigate the consequences of magnetic flares on the surface of asymptotic giant branch (AGB) and similar stars. In contrast to the solar wind, in the winds of AGB stars the gas cooling time is much shorter than the outflow time.  As a result, we predict that energetic flaring will not inhibit, and may even enhance, dust formation around AGB stars.  {{{If magnetic flares do occur around such stars, we expect some AGB stars to exhibit X-ray emission; indeed certain systems including AGB stars, such as Mira, have been detected in X-rays. However, in these cases, it is difficult to distinguish between potential AGB star X-ray emission and, e.g., X-ray emission from the vicinity of a binary companion.  Analysis of an archival ROSAT X-ray spectrum of the Mira system suggests an intrinsic X-ray luminosity $\\sim2\\times10^{29}$ erg s$^{-1}$ and temperature $\\sim10^7$ K. These modeling results suggest that magnetic activity, either on the AGB star (Mira A) or on its nearby companion (Mira B), is the source of the X-rays, but do not rule out the possibility that the X-rays are generated by an accretion disk around Mira B.}}} ",
        "introduction": "Asymptotic giant branch (AGB) stars represent the final stages of evolution for intermediate mass ($\\sim1-8 M_\\odot$) stars, just before these stars (potentially) generate planetary nebulae. Such stars have exhausted their core hydrogen, have completed core helium and H-shell burning phases, and are now burning H and He in concentric shells around the exhausted stellar core (this core, once exposed, will become a degenerate white dwarf). AGB stars are ascending fully convective Hayashi tracks in the H-R diagram, and the very large luminosities of these stars ($\\sim10^4 L_\\odot$), in combination with copious dust grain formation in their cool, extended atmospheres, lead to dusty, radiatively-driven winds. Hence, AGB stars lose mass rapidly, with rates ranging from $\\sim 10^{-8}$ to $\\sim 10^{-4}$ $M_\\odot$ yr$^{-1}$ (as ascertained from far-IR continuum and millimeter-wave molecular line emission; e.g. Loup et al.\\ 1993).  There are several independent pieces of evidence for the presence of magnetic fields in and around cool giants and, in particular, in and around AGB stars.  These include: \\begin{enumerate} \\item Maser polarization around AGB stars and other cool stars. SiO masers are found close to (typically within less than a few stellar radii of) the stellar surface (e.g. Kemball \\& Diamond 1997), OH masers are found at at $\\sim 10^{15}-10^{16} \\cm$ from the star (e.g., Szymczak, Cohen, \\& Richards 1999; Palen \\& Fix 2000; Bains et al. 2003), and H$_2$O masers are found in between (Vlemmings, Diamond, \\& van Langevelde 2002). \\item Polarization of OH maser spots in planetary nebulae (PNs) and proto-PNs, which are the immediate descendants of AGB stars (e.g., Zijlstra et al.\\ 1989).  Miranda et al.\\ (2001) find polarization in the 1,665-MHz OH maser line, which indicates the presence of $\\sim 10^{-3} \\G$ magnetic fields at $\\sim 10^{16} \\cm$ from the central star of the young PN K3-35. \\item X-ray emission from cool giant stars (H\\\"unsch et al. 1998; Schr\\\"oder, H\\\"unsch, \\& Schmitt 1998; H\\\"unsch 2001). H\\\"unsch (2001) used {\\it Chandra} to detect X-ray emission from two M giant stars, which display hard X-ray luminosities of $L_x \\simeq 10^{30} \\erg \\s^{-1}$. \\end{enumerate}  Although it is widely agreed that radiation pressure on dust is the main acceleration process of winds from AGB stars (Elitzur, Ivezic, \\& Vinkovic 2002, and references therein), there is still disagreement over the role the magnetic field plays in shaping the circumstellar matter. First, it is not clear whether magnetic fields play any role in shaping the winds; many other models exist, e.g., acceleration via radiation pressure on dust in fast rotating AGB stars spun up by companions (e.g., Dorfi \\& H\\\"ofner 1996; Reimers, Dorfi \\& H\\\"ofner 2000), and the influence of a companion outside the envelope (e.g., Mastrodemos \\& Morris 1999; Soker 2001). Second, there is a disagreement on whether the magnetic fields play a dynamical role, i.e., whether the magnetic pressure and/or tension become comparable to the other relevant forces in the flow (e.g., Pascoli 1997; Chevalier \\& Luo 1994; Garc\\'{\\i}a-Segura 1997; Garc\\'{\\i}a-Segura et al.\\ 1999; Garc\\'{\\i}a-Segura, \\& L\\'opez 2000; Garc\\'{\\i}a-Segura, L\\'opez, \\& Franco 2001; Matt et al.\\ 2000; Blackman et al.\\ 2001; Gardiner \\& Frank 2001; Falceta-Goncalves \\& Jatenco-Pereira 2002; see review by Garc\\'{\\i}a-Segura 2002), or whether the magnetic field is relatively weak on a global scale, and plays only secondary effects (Soker 2002; Soker \\& Zoabi 2002 and references therein).  In earlier papers Soker and collaborators (e.g., Soker 1998; Soker \\& Clayton 1999) propose that cool magnetic spots facilitate the formation of dust, hence locally increasing the mass loss rate. In this process the magnetic field has a secondary role and becomes dynamically important only in relatively small regions of the stellar surface and circumstellar environment. In this model the large scale magnetic field is too weak to play a dynamic role and directly influence the wind from the AGB star. However, the magnetic field is strong enough above and near cool spots that reconnection may occur {{{(if magnetic pressure exceeds thermal pressure);}}} such reconnection events likely would lead to magnetic flares analogous to those on the surface of the sun. The reconnection process is basically cancellation of magnetic field lines having opposite directions. The energy released by the reconnecting magnetic fields heats the gas and accelerates it to high speeds. The gas is heated by MHD shocks (e.g., Chen et al.\\ 1999). In the present paper we discuss some implications of these events. In $\\S 2$ the potential of impact of flaring on circumstellar dust formation is considered. In $\\S 3$ we discuss the expected observational signatures of AGB star magnetic activity in general and of flares in particular. Our summary is in $\\S 4$. ",
        "conclusions": "Recently, evidence has accumulated for the presence of relatively strong magnetic fields around AGB and similar stars. We have investigated certain theoretical and observational consequences of magnetic activity in such cool giant stars. The main results of this investigation are as follows. \\begin{enumerate} \\item In contrast to the solar wind, in the expanding circumstellar envelopes of AGB stars the post-flare gas cooling time is much shorter than the typical flow timescale. As a result, we predict that energetic flaring will not inhibit dust formation around AGB stars, {{{ and may even enhance dust formation in some cases. }}} \\item Magnetic reconnection events near the stellar surface should lead to localized, long duration (timescales $\\sim$ few days to a year) flares. {{{ \\item X-ray observations should provide indications as to whether AGB stars display magnetic flare activity. Analysis of an archival ROSAT X-ray spectrum of the Mira system yields results for intrinsic X-ray luminosity ($\\sim2\\times10^{29}$ erg s$^{-1}$) and temperature ($\\sim10^7$ K) that are consistent with such activity, either on Mira A or its companion, Mira B, although we do not rule out the possibility that the X-ray emission arises from an accretion disk around Mira B. }}}{{{{ X-ray spectroscopic observations with XMM-Newton, whose sensitivity and spectral resolution far exceed those of ROSAT/PSPC, should better constrain the X-ray emission mechanism of the Mira system. Meanwhile, the exceptional spatial resolution of the Chandra X-ray Observatory likely provides the only means to determine unambiguously the source of the X-ray emission. The separation between Mira A and B, $\\sim0.6''$ (Karovska et al.\\ 1997), is just at the limit of Chandra's spatial resolving power and absolute astrometry.  }}}} \\item Observations of polarized maser emission from the inner circumstellar envelopes of AGB stars may indicate the presence of localized, highly magnetized wind clumps --- analogous to magnetic clouds in the solar wind --- rather than large-magnitude global magnetic fields. \\end{enumerate}"
    },
    "0209/astro-ph0209285_arXiv.txt": {
        "abstract": " ",
        "introduction": "We present a new evolutionary synthesis (ES) tool specifically developed for the analysis of multi-color data of young star clusters (YSCs) in interacting galaxies. Our ESO/ST-ECF ASTROVIRTEL\\footnote{The support given by ASTROVIRTEL, a Project funded by the European Commission under FP5 Contract No. HPRI-CT-1999-00081, is acknowledged.} project provides an unprecedented database of UV--optical--NIR observations from the {\\sl HST}/WFPC2, {\\sl HST}/NICMOS and {\\sl VLT} data archives. Comparison of these data with ES models for clusters of various metallicities which include gaseous emission as well as various degrees of dust extinction, allows one to independently determine metallicities, ages and extinction for individual clusters. These, in turn, are the basis to derive the mass functions of the YSC systems and to predict the future evolution of their luminosity functions and color distributions. Comparing YSC systems of various ages will, we anticipate, allow us to ``see cluster disruption processes at work''. ",
        "conclusions": ""
    },
    "0209/astro-ph0209550_arXiv.txt": {
        "abstract": "The SuperNova / Acceleration Probe (SNAP) is a space-based experiment to measure the expansion history of the Universe and study both its dark energy and the dark matter.  The experiment is motivated by the startling discovery that the expansion of the Universe is accelerating.  A 0.7~square-degree imager comprised of 36 large format fully-depleted $n$-type CCD's sharing a focal plane with 36 HgCdTe detectors forms the heart of SNAP, allowing discovery and lightcurve measurements simultaneously for many supernovae. The imager and a high-efficiency low-resolution integral field spectrograph are coupled to a 2-m three mirror anastigmat wide-field telescope, which will be placed in a high-earth orbit.  The SNAP mission can obtain high-signal-to-noise calibrated light-curves and spectra for over 2000 Type Ia supernovae at redshifts between $z=0.1$ and 1.7. The resulting data set can not only determine the amount of dark energy with high precision, but test the nature of the dark energy by examining its equation of state. In particular, dark energy due to a cosmological constant can be differentiated from alternatives such as ``quintessence'', by measuring the dark energy's equation of state to an accuracy of $\\pm 0.05$, and by studying its time dependence. ",
        "introduction": "\\label{sect:intro}  %  In the past decade the study of cosmology has taken its first major steps as a precise empirical science, combining concepts and tools from astrophysics and particle physics.  The most recent of these results have already brought surprises.   The Universe's expansion is apparently accelerating rather than decelerating as expected solely due to gravity.  This implies that the simplest model for the Universe --- flat and dominated by matter --- appears not to be true, and that our current fundamental physics understanding of particles, forces, and fields is likely to be incomplete.  The clearest evidence for this surprising conclusion comes from the recent supernova measurements of changes in the Universe's expansion rate that directly show the acceleration.  Figure~\\ref{confcmbclust} shows the results of Ref.~\\citenum{p99} (see also Ref.~\\citenum{riess98}) which compare the standardized brightnesses of 42 high-redshift Type~Ia supernovae (SNe~Ia) ($0.18<z<0.83$) with 18 low-redshift SNe~Ia to find that for a flat universe $\\Omega_\\Lambda=0.72\\pm0.08$ ($\\Omega_M=1-\\Omega_\\Lambda$), or a deceleration parameter $q_0=-0.58$, and constrain the combination $0.8\\,\\Omega_M-0.6\\,\\Omega_\\Lambda$ to $-0.2\\pm0.1$.  This evidence for a negative-pressure vacuum energy density is in remarkable concordance with combined galaxy cluster measurements \\cite{bahcall00}, which are sensitive to $\\Omega_M$, and CMB results \\cite{balbi2000,lange2001}, which are sensitive to the curvature $\\Omega_k$ (see Fig.~\\ref{confcmbclust}).  Two of these three independent measurements and standard inflation would have to be in error to make the cosmological constant (or other negative pressure dark energy) unnecessary in the cosmological models.  These measurements indicate the presence of a new, unknown energy component that can cause acceleration, hence having equation of state $w\\equiv p/\\rho<-1/3$.  This might be the cosmological constant. Alternatively, it could be that this dark energy is due to some other primordial field for which $\\rho \\ne -p$, leading to different dynamical properties than a cosmological constant.  The fundamental importance of a universal vacuum energy has sparked a flurry of activity in theoretical physics with several classes of models being proposed (e.g.  quintessence \\cite{caldwell98,zlatev99}, Pseudo-Nambu-Goldstone Boson (PNGB) models \\cite{frieman95,coble97}, cosmic defects \\cite{vilenkin84,vilenkin94}).  Placing some constraints on possible dark energy models, Refs.~\\citenum{p99,garnavich98,ptw99} find that for a flat Universe, the data are consistent with a cosmological-constant equation of state with $0.2\\ls\\Omega_M\\ls 0.4$ (Fig.~\\ref{wconf}), or generally $w<-0.6$ at 95\\% confidence level. The cosmic string defect theory ($w=-1/3$) is already strongly disfavored.  \\begin{figure}[h] \\begin{center} \\includegraphics[height=4.5truein]{confcmbclust.eps} \\caption[$\\Omega_M$---$\\Omega_\\Lambda$ confidence regions with current SN, galaxy cluster, and CMB results.] {There is strong evidence for the existence of a cosmological vacuum energy density.  Plotted are $\\Omega_M$---$\\Omega_\\Lambda$ confidence regions for current SN \\cite{p99}, galaxy cluster, and CMB results.  These results rule out a simple flat, [$\\Omega_M=1$, $\\Omega_\\Lambda=0$] cosmology.  Their consistent overlap is a strong indicator for dark energy.  Also shown is the expected confidence region from the SNAP satellite for an $\\Omega_M=0.28$ flat Universe. } \\label{confcmbclust} \\end{center} \\end{figure}  \\begin{figure}[ht] \\begin{center} \\includegraphics[height=4.0truein]{snapsatwconf.eps} \\caption[Confidence regions in the $\\Omega_{\\rm M}$--$w$ plane.] {Best-fit 68\\%, 90\\%, 95\\%, and 99\\% confidence regions in the $\\Omega_{\\rm M}$--$w$ plane for an additional energy density component, $\\Omega_w$, characterized by an equation-of-state $w = p/\\rho$.  (For Einstein's cosmological constant, $\\Lambda$, $w = -1$.) The fit is constrained to a flat cosmology ($\\Omega_{\\rm M} + \\Omega_w =1$).  Also shown is the expected confidence region allowed by SNAP assuming $w=-1$ and $\\Omega_M=0.28$. } \\label{wconf} \\end{center} \\end{figure}   In this paper, we attempt to formulate a definitive supernova study that will determine the values of the cosmological parameters and measure the properties and test possible models for the dark energy. In \\S\\ref{exec_uncertain} we identify and discuss how to minimize systematic errors that fundamentally limit the precision with which this probe can measure cosmological parameters.  A SN dataset that maximizes the resolving power of the redshift-luminosity relation under the constraint of these systematic errors is constructed in \\S\\ref{asprobe}.  We present in \\S\\ref{proposed_experiment} the SuperNova / Acceleration Probe (SNAP) whose observing strategy and instrumentation suite is tailored to provide the data that satisfy both our statistical and systematic requirements. ",
        "conclusions": "The surprising discoveries of recent years make this a fascinating new era of empirical cosmology, addressing fundamental questions.  SNAP presents a unique opportunity to extend this exciting work and advance our understanding of the Universe."
    },
    "0209/astro-ph0209599_arXiv.txt": {
        "abstract": "We report on the detection of variability from the proposed IR counterpart to the Anomalous X--ray Pulsar (AXP) \\src\\ based on Chandra and ESO optical/IR deep observations carried out in 2001--2002.  Within the narrow \\AXAF\\ uncertainty region for \\src\\ we found only one relatively faint ($J$=22.1$\\pm$0.3, $J$--$Ks$=2.4) source, while the recently proposed IR counterpart was not detected down to a limiting $Ks$ magnitude of $\\sim$20.7 (3$\\sigma$ confidence level). This implies a remarkable IR brightening of this object, $\\Delta\\,Ks$$>$1.3, on a timescale of about 50 days. Although our knowledge of the IR properties of AXPs is rather limited (there is only another source, \\ee, for which IR variability has been detected), the observed IR variability of the proposed counterpart strengthens its association with \\src. Our results make the IR (and presumably optical) variability a likely common characteristic of AXPs, and provide new constraints on this class of objects. ",
        "introduction": "After more than 20 years since the discovery of X--ray pulsations from the prototype source \\ee, the {\\rc current observational properties of AXPs are not yet conclusive to unambiguously assess their nature}. There are currently five confirmed members of the AXP class plus two candidates.  Although we can be reasonably confident that AXPs are magnetic rotating neutron stars (NSs), their energy production mechanism is still uncertain. It is also unclear whether they are solitary objects (either low magnetised or magnetars) or are in binary systems with very low mass companions (for a review see Israel et al. 2002; Mereghetti et al 2002 and references therein).  Different production mechanisms for the observed X--ray emission have been proposed, involving either accretion or the dissipation of magnetic energy. The recent detection of X--ray bursts from \\ee\\ and {\\rc \\src} has strengthened the possible connection of AXPs with Soft Gamma--ray Repeaters (SGRs; Kaspi \\& Gavriil 2002; {\\rc Gavriil et al. 2002}). Moreover the detection of relatively large pulsed fraction pulsations in the optical flux of \\uu\\ (at the same period of X--ray pulses) seems to rule out the X--ray reprocessing and favors models based on isolated NSs (Kern \\& Martin 2002). {\\rc The magnetar model, originally proposed by Thompson \\& Duncan (1993) to explain the properties of SGRs, appears currently to be fairly succesful at interpreting these properties of AXPs.}  As done in the past in the case of High and Low Mass X--ray binary system (HMXBs and LMXBs), finding the optical/IR counterparts to AXPs might give the key to definitively assess the nature and the emission mechanisms of AXPs. The lack of any known distinctive property of AXPs in the optical/IR bands, makes the X--ray positional accuracy at the (sub)arcsec level a fundamental preliminary step to carry out very deep searches for potential optical/IR counterparts. Moreover a high X--ray spatial resolution can be exploited to search for small scale diffuse emission or the presence of structures in the vicinity of the sources.  \\src\\ has been extensively studied in the X--ray band since its discovery in 1979 (Seward et al. 1986). Previous searches for an optical counterpart gave negative results (Seward et al. 1986; Mereghetti et al. 1992; Wang \\& Chakrabarty 2002a). Recently, a relatively faint IR counterpart ($J$=21.7; $J$--$Ks$=2.3) to \\src, with unusual IR colors, has been proposed (Wang \\& Chakrabarty 2002b; hereafter WC02).  In this paper we report on the early results obtained for \\src\\ as a part of a joint ESO/Chandra large project aimed at the identification and study of the optical/IR counterparts of AXPs in the southern sky, and the study of the spatial distribution of X--ray emission from AXPs (results concerning \\AXAF\\ observations of \\rxj\\ and \\axj\\ will be reported elsewhere). ",
        "conclusions": "The NTT IR data show that no object is consistent with the \\AXAF\\ position of \\src\\ in the optical band down to a $V$ limiting magnitude of about 25.5 (see Figure\\,2). On the other hand, a faint object (marked {\\tt X3} in Figure\\,2) is present in the IR $Ks$ frames ($J$=22.1$\\pm$0.3, $Ks$=19.7$\\pm$0.3). {\\rc This object is also detected in the WC02 frames, although its magnitudes are not listed. A visual inspection of the WC02 image suggests that the source {\\tt X3} has remained nearly unchanged in flux with respect to our frames} (superimposing to our IR images the \\AXAF\\ position and uncertainty radius reported by WC02 for \\src, the object {\\tt X3} falls within the circle). {\\rc Note that during the NTT IR observations carried out on 2001, the object {\\tt X3} is not detected due to the poor seeing conditions and contamination with the nearby bright object.} More surprisingly, we note that no object is detected at the position of the source {\\tt X1} reported by WC02 (the position of {\\tt X1} is marked with a square in Figure\\,2). For comparison, the object marked {\\tt X2} in Figure\\,2 has a magnitude of $J$=20.63$\\pm$0.06, $H$=19.73$\\pm$0.15 and $Ks$=19.0$\\pm$0.2 (consistent, within 1--1.5$\\sigma$, with the values reported by WC02). {\\rc As the limiting magnitudes of our IR frames are well above the values measured by WC02 for {\\tt X1}, we conclude that the source was considerably fainter during our observations}. This IR brightening ($\\Delta Ks>$1.3; {\\rc about a factor of 3 in flux}), displayed by the source on a timescale of about 50 days (the time interval between our last IR images and those of WC02) further strengthens the link between the IR source and \\src. Moreover, the chance probability of finding a highly variable IR source within a circle of 0\\farcs8 is negligible.  An IR counterpart was recently found for the prototype of the AXP class, \\ee, which also showed IR variability at a level of $\\Delta Ks\\sim$1.3 (Kaspi et al. 2002). This IR variability {\\rc might be associated to the short--term X--ray bursts (similar to those displayed by SGRs) observed from the source three days before the IR observations} (Kaspi \\& Gavriil 2002). Recently, short X--ray bursts were detected in the X--ray flux of \\src\\ (Gavriil et al. 2002). All these findings together, suggest {\\rc a close similarity of \\src\\ and \\ee, such that the IR counterpart of the former might be expected to be variable. For this reason {\\tt X1} is the most likely IR counterpart to \\src, even though source {\\tt X3} is included in the \\AXAF\\ uncertainty circle and its color ($J$--$Ks$=2.4) is within the value interval predicted for AXPs (see also Figure \\,2 of WC02)}.  {\\rc The fact that \\src\\ showed two X--ray bursts close to each other, implies that many other bursts might have been missed due to the sparse X--ray monitoring with the \\RXTE. Moreover since our non--detection of the IR candidate predates the WC02 detection, it would appear more natural to associate (if at all) the IR variability to some X--ray activity that took place between the two IR campaigns (as opposed to the few second--duration X--ray bursts that was detected 4 months earlier). Finally we note that the X--ray/IR observations of \\ee\\ are such that we do not know whether the detected IR brightening follows or precedes that in the X--ray band. Future multi--wavelength monitoring programs will probably allow us to better understand these points.}   The non--detection of {\\tt X1} in our IR frames implies that the ``quiescent'' (or low state) IR emission level of \\src\\ is still undetected. {\\rc Correspondingly, it cannot be ruled out yet that the spectral flattening proposed by WC02 might result from wavelength--dependent variability}. Similarly, the measured IR fluxes for the counterpart cannot be used to study the overall (from X--ray to IR) energy spectrum of \\src, since they reflect a different energy distribution than the quiescent (or low state) one.  Deeper IR observations are needed to determine the quiescent emission state of this source.  IR variability might represents a new, fairly common, characteristic of AXPs (and possibly SGRs), and potentially a diagniostic of their nature. We note that, so far, very little is known on the expected optical/IR emission from a magnetar. Both models based on magnetar and accretion from ``fossil'' disk have difficulties in accounting for transient {\\rc and/or variable} emission {\\rc at wavelengths shorter than those in the X--ray band (Thompson \\& Duncan 1996 and references therein; Chatterjee \\& Hernquist 2000 and references therein)}. {\\rc On the other hand, the analogy with the X--ray bursts of LMXBs, suggests that the effects of the X--ray burst of AXPs might propagate towards longer wavelengths (e.g., via reprocessing)}. It is somewhat intriguing that at least the values of $\\Delta Ks$ and colors of the IR counterpart of \\ee\\ and \\src\\ are similar to those displayed by some classes of accreting cataclysmic variables and low mass X--ray binaries hosting a neutron star.  The superb \\AXAF\\ HRC--I spatial resolution allowed us to look for extended X--ray emission around the pulsar; {\\rc this might result from an X--ray nebula powered by the AXPs as well as a scattering halo.} A first detailed study was carried out for \\ee, where emission beyond 4\\arcsec\\ and extending up to 100\\arcsec\\ was found (Patel et al 2001). However in this case the source is embedded in the diffuse emission from the supernova remnant G109.1--1.0, and it is very difficult to disentangle this component from that (possibly present) around the AXP. The present \\AXAF\\ HRC--I data are by far the best spatial resolution observation carried out for an AXP and no significant diffuse X--ray emission was found beyond $\\sim$2\\arcsec."
    },
    "0209/astro-ph0209620_arXiv.txt": {
        "abstract": "In this review I focus on a few selected topics, where recent theoretical and/or observational progress has been made and important developments are expected in the future. They include: 1) Evolution of isotopic ratios, 2) Mixing processes and dispersion in abundance ratios,  3) Abundance gradients in the Galactic disk (and abundance patterns in the inner Galaxy), 4) The question of primary Nitrogen and 5) Abundance patterns in extragalactic damped Lyman-$\\alpha$ systems (DLAs). ",
        "introduction": "Despite the lack of a reliable model of Galactic Chemical Evolution (GCE) considerable progress has been made in the past few years, due to:   - an impressive amount of observational data concerning abundance ratios either in the Milky Way (Carretta et al. 2002, Truran et al. 2002, Andriefsky et al. 2002, and references therein) or in extragalactic systems (e.g. Prochaska and Wolfe 2002);  - the publication of detailed nucleosynthesis yields (e.g. van den Hoek and Groenewegen 1997 for intermediate mass stars; Woosley and Weaver 1995, Nomoto et al. 1997, Limongi et al. 2000, for massive stars; Iwamoto et al. 1999 for SNIa).   Interpreted in the framework of simple ``toy'' models of GCE these data offer mainly information about: i) evolutionary timescales of corresponding production sites, ii) physics of those sites, iii) nature of relevant nucleosynthetic processes.  In this, highly biased, review I focus on a few selected topics, where recent theoretical and/or obervational progress has been made and important developments are expected in the future.       \\begin{figure*}[htb] \\begin{center} \\psfig{figure=prantzos_f1.eps,angle=-90,width=0.7\\textwidth} \\caption{{\\it Top left:} Overproduction factors of the Mg isotopes as a function of stellar mass and for various initial metallicities (from WW95). {\\it Bottom left:} Abundance ratios of the Mg isotopes (integrated over the Kroupa et al. 1993 IMF) as a function of initial stellar metallicity. {\\it Right:} Evolution of Mg isotopic ratios in Milky Ways' halo and local disk stars (from Goswami and Prantzos 2000) with WW95 yields and comparison to observations. } \\end{center} \\end{figure*} ",
        "conclusions": ""
    },
    "0209/astro-ph0209416_arXiv.txt": {
        "abstract": "A planetary system may undergo significant radial rearrangement during the early part of its lifetime.  Planet migration can come about through interaction with the surrounding planetesimal disk and the gas disk---while the latter is still present---as well as through planet-planet interactions.  We review the major proposed migration mechanisms in the context of the planet formation process, in our Solar System as well as in others. ",
        "introduction": "The word planet is derived from the Greek word ``planetes'', meaning wandering star.  Geocentric views of the Universe held sway until the Middle Ages, when Copernicus and Kepler developed a better phenomenological explanation of planetary wanderings, which with small modifications has withstood the test of time.  Kepler's first law of planetary motion states that planets travel along elliptical paths with one focus at the Sun.  Thus, although planets wander about the sky, in this model their orbits remain fixed and they do not migrate. In his physical model of the Solar System, Newton theorized that planets gradually altered one another's orbits, and he felt compelled to hypothesize occasional divine intervention to keep planetary trajectories well-behaved over long periods of time. In the early 1800s, Poisson pointed out that planetary-type perturbations cannot produce secular changes in orbital elements to second order in the mass ratio of the planets to the Sun, but Poincare's work towards the end of the 19th century suggests that the Solar System may be chaotic. The stability of mature planetary systems is a fascinating topic, but we shall be concerned herein with the potentially much more rapid migration of planets during and immediately following the epoch of their formation.  Modern research into this topic began when Goldreich and Tremaine (1980) showed that density wave torques could have led to significant orbital evolution of Jupiter within the protoplanetary disk on a timescale of a few thousand years, and research accelerated when giant planets were found much closer to their stars (Mayor and Queloz 1995) than predicted by models of their formation (Lin et al. 1996, Bodenheimer et al. 2000).  In Section 2, we discuss models for the migration of giant planets within our own Solar System which may have occurred as the results of interactions of the planets with one another and with small solid bodies.  Section 3 summarizes models of the potentially substantial planetary migration that results from (primarily gravitational) interactions between a planet and a gaseous protoplanetary disk.  The predictions of this model are compared to observations of Saturn's rings and moons in Section 4.  In Section 5, the models are applied to extrasolar planetary systems. ",
        "conclusions": "The longstanding view of planet formation as an orderly process, involving little radial migration of material, is being made to look increasingly inaccurate by both observational and theoretical findings.  In our own Solar System, the high eccentricities of objects in exterior mean-motion resonances with Neptune imply an outward migration of several AU by the outer ice giant; modeling suggests that Uranus, and to a lesser degree Jupiter and Saturn, likewise underwent migration as they cleared the surrounding planetesimal disk.  An earlier and more violent period of migration could have occured if Uranus and Neptune originally formed among proto-Jupiter and -Saturn; such a model would alleviate the longstanding formation timescale problem of Uranus and Neptune, and could simultaneously help to account for the gravitationally stirred-up state of the Kuiper belt.  In the first ten million years or so of a planetary system's life, the nebular gas is still present and provides a much larger sink/source of angular momentum than the planetesimal disk.  Growing protoplanets exchange angular momentum by launching density waves at resonance sites in the disk.  From theoretical consideration, this ought to bring about a net loss of angular momentum and rapid orbital decay---Type I migration---for bodies of order a few Earth masses. For a sufficiently massive body, the torques between it and the disk will be strong enough to open a gap, thus locking the body into the subsequent viscous evolution of the disk in what is called the Type II mode of migration.  The core accretion model of giant planet formation seems to require that the Type I to II transition occur at small masses, otherwise growing cores will spiral into the star before they can acquire a massive envelope.  Alternatively, it is possible that planet formation simply is an enormously wasteful process, which dumps a steady stream of growing protoplanets onto the primary, and the end result is whatever happens to be left over when the gas fades away . This is sometimes called the ``last of the Mohicans'' scenario. Significant post-formation migration is quite likely responsible for the large number of planets detected on close-in orbits (``giant Vulcans'', also referred to as ``hot Jupiters'').  In multiple-planet systems, convergent and divergent migration of planets can be invoked to explain, respectively, resonant capture and eccentricity excitation.  As the nebular gas dissipates, it is likely that the tables are eventually turned; the planets, heretofore at the mercy of the gas, assert themselves and serve as anchors to slow down the viscous evolution of the last remains of the disk, so that migration ends in a Type III phase.  Clearly, the present observational and theoretical ``state of the art'' still requires us to use a liberal amount of conjecture in attempting to sketch a coherent picture of planet migration.  However, it seems equally clear that migration is intimately linked with the formation of planetary system, and a complete picture of the latter will require a full understanding of the former."
    },
    "0209/astro-ph0209544_arXiv.txt": {
        "abstract": "Narrow-band searches for \\lya\\ emission are an efficient way of identifying star-forming galaxies at high redshifts.  We present Keck telescope spectra confirming redshifts $z\\approx 5.7$ for three objects discovered in the Large Area Lyman Alpha (LALA) survey at Kitt Peak National Observatory. All three spectra show strong, narrow emission lines with the asymmetric profile that is characteristically produced in high redshift \\lya\\ emitters by preferential HI absorption in the blue wing of the line.  These objects are undetected in deep $B_W$, $V$, $R$, and $\\lambda \\approx 6600$\\AA\\ narrow-band images from the NOAO Deep Wide-Field Survey and from LALA, as expected from Lyman break and \\lya\\ forest absorption at redshift $z\\approx 5.7$.  All three objects show large equivalent widths ($\\ga 150$\\AA\\ in the rest-frame), suggesting at least one of the following: a top-heavy initial mass function, very low stellar metallicity, or the presence of an active nucleus. We consider the case for an active nucleus to be weak in all three objects due to the limited width of the \\lya\\ emission line ($< 500 \\kms$) and the absence of any other indicator of quasar activity. The three confirmed high redshift objects were among four spectroscopically observed targets drawn from the sample of 18 candidates presented by Rhoads \\& Malhotra (2001).  Thus, these spectra support the \\lya\\ emitter population statistics from our earlier photometric study, which imply little evolution in number density from $z=5.7$ to $z=4.5$ and provide strong evidence that the reionization redshift $\\zre > 5.7$. ",
        "introduction": "Narrow-band searches were originally proposed as a method of finding young, high redshift galaxies in their first throes of star-formation (Partridge \\& Peebles 1967). This method has now proven its worth in discovering star-forming galaxies at ever increasing redshifts. A large fraction (9 out of 11) of known $z > 5$ galaxies have strong \\lya\\ emission (Dey et al. 1998, Dawson et al. 2002, Weyman et al. 1998, Hu et al. 1999, Hu et al. 2002, this paper), the only exception being a pair of Lyman break galaxies at z=5.34 (Spinrad et al 1998).  This is partly due to the ease of confirming galaxies with strong emission lines and partly because one can avoid night sky lines by using narrow-band selection methods.  Ultimately, the \\lya\\ emitters at high redshifts will provide useful constraints on models of galaxy formation and evolution. A more immediate application is to constrain the epoch of reionization. In a neutral universe, \\lya\\ photons from galaxies are resonantly scattered into large, diffuse halos, whose extremely low surface brightness will make their detection very challenging.  Thus, the transition to a neutral universe will be marked by a sharp drop in \\lya\\ galaxy counts.  In the Large Area Lyman Alpha (LALA) survey (Rhoads et al. 2000, Paper I) we use narrow-band imaging with the wide-field Mosaic cameras at Kitt Peak National Observatory (KPNO) and Cerro Tololo Inter-American Observatory (CTIO) 4-meter telescopes to image a relatively large area ($36'\\times 36'$) of the sky at once. This allows us to obtain good statistics on the candidates selected by imaging, and use them to study galaxy formation and evolution at high redshifts. Confirming candidates from narrow-band imaging by spectroscopy requires large telescopes and is a slower process.  But as we show in this and companion papers, the success rate for spectroscopic confirmations is high, so conclusions drawn from mere imaging candidates are probably accurate (Rhoads \\& Malhotra 2001, hereafter RM01; Malhotra \\& Rhoads 2002; Ouchi et al. 2002). ",
        "conclusions": "\\label{disc} We now consider the physical implications of our observations, beginning with the properties of the individual sources and then moving on to properties of the galaxy population and of the universe at $z\\approx 5.7$.  The three spectroscopically confirmed \\lya\\ sources presented here all show very large equivalent widths, with $2\\sigma$ lower bounds of $150$\\AA\\ (rest frame) and up.  Such large equivalent widths are very hard to achieve with only the ionizing flux from a normal stellar population, which can provide an equivalent width beyond 200\\AA\\ for only a few $\\times 10^6$ years (Charlot \\& Fall 1993).  To produce equivalent widths beyond $\\sim 200$\\AA, the ultraviolet radiation in the galaxies must be harder than usual.  This can be achieved in galaxies with active nuclei, with a stellar metallicity near zero ($Z \\ll 0.05 Z_\\odot$, Tumlinson \\& Shull 2000; Bromm, Kudritzki \\& Loeb 2001), or with a stellar initial mass function dominated by high mass stars (Malhotra \\& Rhoads 2002; Kudritzki et al. 2000).  However, conventional (broad-lined) active galactic nuclei are excluded by the observed \\lya\\ line widths in these sources, which have upper bounds from $220$ to $460 \\kms$ (Table~\\ref{spectab}).  Thus, if the sources are AGN, they must be type~II AGNs, i.e., systems whose broad-line regions are hidden from direct view.  Even type~II AGN generally have larger \\lya\\ line widths than we observe.  For radio galaxies, a composite spectrum (Stern et al 1999) shows FWHM $= 1130 \\kms$, while an individual case (4c41.17; Dey et al 1997) shows a narrow component of $613 \\kms$ but also a broad component of similar line flux and FWHM $= 1373 \\kms$.  For X-ray selected type~II AGN, we have one measurement at $1520 \\kms$ (Stern et al 2002) and one upper limit at $< 1130 \\kms$ (Norman et al 2002).  Thus, line widths seem to favor star formation activity over active nuclei as the driver for the \\lya\\ line in these LALA sources.  The remaining explanations for the observed equivalent widths both suggest star formation in primitive conditions.  Low metallicity stars obviously require low metallicity gas, while top-heavy initial mass functions may be a natural consequence of the way low-metallicity gas ($Z/Z_\\odot \\la 5\\times 10^{-4}$) cools and fragments (Bromm et al 2001).  A final possibility is that anisotropy in the \\lya\\ radiation biases the equivalent width measurements substantially (see Charlot and Fall 1993; Neufeld 1991), but this explanation would require a very special source geometry.  Further tests of the narrow-lined AGN hypothesis are possible, including searches for X-ray emission and for other optical emission lines characteristic of AGN (especially CIV and HeII).  Such tests are underway for the LALA survey's larger $z\\approx 4.5$ sample.  To date we have seen no obvious evidence for AGN activity in any of our confirmed high-redshift LALA spectra.  The confirmation of 3 $z\\approx 5.7$ \\lya\\ emitters among 4 candidates observed allows us to refine the statistics of this population that were presented in RM01.  There we presented a sample of 18 candidates, based on detection significance, emission line strength, and an absence of blue flux.  A complication in candidate selection is that the measured equivalent widths of high redshift objects may be changed by intergalactic hydrogen absorption of either continuum or line flux. A ``worst case'' (and somewhat unrealistic) scenario is where the broad-band flux is substantially reduced while the narrow-band flux is unaffected.  If we assume this case and correct the $I$ band fluxes of our 18 candidates for intergalactic hydrogen absorption using a factor of $2.1$ (based on the prescription of Madau 1995), we find that 13 candidates still pass the equivalent width cutoff for selection.  The remaining 5 candidates could in principle be continuum sources at redshift just below $5.7$, where an apparent narrowband excess of a factor $\\approx 2$ could be produced by IGM hydrogen absorption in the blue side of the broad band filter.  We will call the 18 candidates the ``full sample'' and the 13 remaining after our worst case IGM absorption correction the ``reduced sample'' in the following discussion. All four candidates with Keck spectra belong to the reduced sample.  Applying a 3/4 correction factor to the observed source counts for the spectroscopic success rate presented here, we find a number density of 540 (390) \\lya\\ emitters per square degree per unit redshift for the full (reduced) sample. These counts are above a detection threshold of $(1.47 \\pm 0.08) \\times 10^{-17} \\ergcm2s$, where the error bar indicates the range in detection threshold between the two narrow-band filters.      The implied \\lya\\ luminosity density at $z\\approx 5.7$ then becomes $1.26 \\times 10^5$ ($0.80 \\times 10^5$) $\\lsun/\\Mpc^3$ for sources with $L_\\lya > 1.36 \\times 10^9 \\lsun$, based on the full (reduced) sample.  (Here and throughout this section we assume a cosmology with $H_0 = 70 \\kmsMpc$, $\\Omega_m = 0.3$, and $\\Omega_\\Lambda = 0.7$.) This number is a lower bound, and the total luminosity density could be substantially increased if the luminosity function rises steeply below our detection threshold.  These \\lya\\ luminosities can be translated into star formation rates under an assumed model for the stellar initial mass function (IMF) and for the escape fraction of \\lya\\ photons from their parent galaxies. To enable comparison with other work, we follow Hu, McMahon, \\& Cowie (1999) and apply a conversion factor of $1 \\Msun/\\year = 10^{42} \\ergsec$.  This factor is obtained by taking the \\lya\\ to H$\\alpha$ ratio for Case B recombination and no dust, together with the Kennicutt (1983) conversion between H$\\alpha$ luminosity and star formation rate.  This corresponds to an IMF slope $2.5$ (close to the Salpeter value of $2.35$) over the range $1 < M/M_\\odot < 100$, continued to a lower mass cutoff of $0.1 \\Msun$ with slope $1.4$, with solar metallicity for all stars.  We hasten to add that these assumptions appear inconsistent with the observed equivalent width distribution of the LALA \\lya\\ emitters (Malhotra \\& Rhoads 2002), but as we cannot uniquely determine the correct IMF and metallicity from the available data, it is not clear how we should replace these assumptions.  The resulting star formation rate density at $z\\approx 5.7$ is $5 (3) \\times 10^{-4} \\Msun/\\year/\\Mpc^3$ in objects with $L_\\lya > 7\\times 10^{42} \\ergsec$, for the full (reduced) samples after correction for spectroscopic completeness. The star formation rates per object range from $7$ to $14 \\Msun/\\year$.  The \\lya\\ luminosities allow us to constrain the metal production rate density in these objects much more reliably than the total star formation rate density.  The ratio of ionizing photon production to metal production depends only weakly on the details of a stellar population, because the most massive stars dominate both production rates (e.g., Madau \\& Shull 1996).  Every observed \\lya\\ photon implies the production of $\\ge 1.5$ ionizing photons (where equality is expected in the limit of zero dust attenuation). The metal production rate is approximately $d M_Z / d t = 500 L_i / c^2$, where $L_i$ is the luminosity in ionizing photons and $M_Z$ is the mass in elements with atomic number $Z \\ge 6$  (Madau \\& Shull 1996). Thus, our \\lya\\ luminosity density implies a metal production rate density $\\ge 8.6 (5.5) \\Msun/\\Myr/\\Mpc^3$ for the full (reduced) sample. Malhotra \\& Rhoads (2002) suggest that the \\lya\\ emitters in the LALA sample suffer little dust attenuation, based on their large observed equivalent widths.  Our lower bound is then likely to be near the total metal production in \\lya\\ emitting galaxies. However, Malhotra \\& Rhoads (2002) also note that additional objects (accounting for $\\sim 90\\%$ of all galaxies) could be removed from the sample by dust obscuration.  This would result in a total metal production rate $\\sim 10$ times larger.  For comparison, we estimate the metal production rate density in Lyman break galaxies at $z \\approx 4$ to be $\\sim 14$ times larger than that in our observed \\lya\\ sample, based on the UV luminosity densities in Steidel et al. (1999). Earlier work by Madau et al (1996), adjusted to our assumed cosmology, gives $85 \\Msun/\\Myr/\\Mpc^3$ at $z \\approx 4$, for continuum selected 450~nm dropouts in the Hubble Deep Field.  This is again a lower bound that can be increased by corrections for low luminosity sources and for dust absorption.  Our spectroscopic results also reinforce the conclusion of RM01 that the space number density of \\lya\\ emitters with line {\\it luminosity\\/} $L_\\lya \\ga 7 \\times 10^{42} \\ergsec$ shows little evolution between $z\\approx 4.5$ and $z\\approx 5.7$.  In contrast, the surface density per unit redshift of \\lya\\ emitters above observed line {\\it flux\\/} $\\approx 1.5 \\times 10^{-17} \\ergcm2s$ decreases by a factor of $\\ga 4$ over the same range.  This is in contrast to the baseline model of Haiman \\& Spaans (1999), which would predict a decrease by at most a factor of $2$ over this redshift range.  Finally, the spectroscopic confirmation of these low-luminosity \\lya\\ sources at $z\\approx 5.7$ secures the conclusion (RM01) that the reionization redshift $\\zre > 5.7$.  The basic physics underlying this conclusion is simple.  \\lya\\ photons propagating through intergalactic gas with a substantial neutral fraction are subject to resonant scattering by atomic hydrogen.  Thus, even flux on the red side of line center can be hidden by the red damping wing of a substantially neutral IGM (Miralda-Escude 1998; Miralda-Escude \\& Rees 1998).  This would obscure the \\lya\\ line from view, so that detection of the \\lya\\ line at some redshift $z$ implies a predominantly ionized medium around the line source, and suggests $\\zre > z$ (Haiman \\& Spaans 1999).  While the resonantly scattered photons will eventually redshift in the Hubble flow and escape to infinity, their effective photosphere (or last scattering surface) is expected to subtend an angle $\\sim 15''$ (Loeb \\& Rybicki 1999), reducing their surface brightness by a factor $\\ga 100$ and thereby rendering them undetectable to surveys like LALA.  This test was first applied by RM01 to show that $\\zre > 5.7$, and subsequently by Hu et al. (2002) to argue for $\\zre > 6.56$.  To make these reionization limits firm, several nuances in the above physical argument deserve close consideration.  In particular, these objects have to produce ionizing photons in order to generate the \\lya\\ line by recombination.  If the escape fraction $\\fesc$ for the ionizing flux is large, the source will be surrounded by an ionized bubble in the IGM.  The size of this bubble is limited by the (time-integrated) ionizing photon production of the source.  RM01 show that the ionized bubble (in the dust-free case) will be large enough for \\lya\\ photons to escape freely (i.e., $1.2 \\Mpc$) if $L_{43} t_8 \\fesc/(1-\\fesc) \\ga 5$, where $L_{Ly-\\alpha} = 10^{43} L_{43} \\ergsec$ and the source is $10^8 t_8$ years old, but that $L_{43} t_8 \\fesc/(1-\\fesc) \\la 2$ for the observed $z\\approx 5.7$ LALA sources.  The critical constraint on $t_8$ and $\\fesc$ is derived from the equivalent widths of the \\lya\\ sources, which can only achieve their large observed values in young galaxies with modest escape fractions.  The Hu et al. (2002) source, HCM~6A, has a more modest equivalent width ($25$\\AA\\ rest frame) that could be produced in systems with $t_8 \\sim 1$ and $\\fesc \\sim 0.75$, but also shows lower line luminosity $L_{43} \\approx 0.3$.  It therefore fulfills $L_{43} t_8 \\fesc/(1-\\fesc) \\la t_8 \\ll 8$, where the last inequality comes from the age of the universe at this redshift.  This should be compared with a bound $L_{43} t_8 \\fesc/(1-\\fesc) \\ga 7$ to produce a Stromgren sphere of radius $1.2\\Mpc$ in the denser IGM at $z=6.56$.  Thus, it again appears unlikely that HCM~6A could ionize a sufficiently large bubble in a fully neutral IGM to allow its \\lya\\ photons to escape freely to infinity.  It is possible, of course, for any particular \\lya\\ source near $\\zre$ to be more luminous than we think and have its \\lya\\ flux substantially attenuated an effective \\lya\\ optical depth of a few.  Thus, HCM~6A al object may be consistent with $\\zre \\sim 6$ (e.g. Haiman 2002). However, the \\lya\\ luminosity function is quite steep at these flux levels, so the number of detectable \\lya\\ sources is still expected to drop sharply at $\\zre$. Haiman (2002) estimates that the suppression of observed \\lya\\ flux would be a factor $\\sim 10$ if HCM~6a is in a fully neutral IGM.  This would imply an intrinsic line luminosity $\\sim 3\\times 10^{43} \\ergsec$, making HCM~6a more luminous than any object found by the LALA survey in a volume $\\sim 30\\times$ greater than surveyed by Hu et al.  We therefore regard this possibility as unlikely.  Other possible loopholes involve the detailed kinematics of the \\lya\\ emitters and the neutral IGM, including the possible effects of gas infall, galactic peculiar motions, and galactic winds.  Detailed consideration of these effects is beyond the scope of the current paper.  However, we can make quick estimates of these effects.  First, infall of the IGM to a newly forming galaxy can blueshift the gas nearest the galaxy outside the \\lya\\ line core, but there must be some point along the line of sight where infalling gas gives way to the Hubble flow and the IGM has zero velocity relative to the emitted line photons.  Hence, infall will not greatly reduce the resonant scattering opacity.  Next, note that the required velocity scale to allow free escape of \\lya\\ is $\\sim 1000 \\kms$.  Peculiar velocities of this magnitude are restricted to the richest clusters of galaxies today, and should be vanishingly rare in the $z\\sim 6$ universe.  Finally, a spherically symmetric galactic wind with (peculiar) velocity $\\sim 1000 \\kms$ at the source, smoothly decreasing to match the Hubble flow with zero peculiar velocity at $\\sim 1.2 \\Mpc$, would require a total energy $\\sim 10^{61.5} \\erg$.  Spread over a Hubble time at $z\\sim 6$, this corresponds to a mechanical luminosity of $\\sim 10^{11.5} \\lsun$, substantially exceeding the total UV/optical light output of these objects ($\\la 10^{11} \\lsun$). The required energy would of course be reduced in an IGM with substantial clumping or ionization.  However, mechanical energy is typically $\\sim 1\\%$ and at most $\\approx 10\\%$ of the starlight in a starburst (Leitherer \\& Heckman 1995), a factor of $\\ga 100$ below the na\\\"{\\i}ve calculation above.  Thus, while galactic winds are common features of starbursts and probably play a role in \\lya\\ escape from local galaxies (Kunth et al 1998), they may not be able to ``save'' \\lya\\ emission in a substantially neutral IGM.  A final concern is the possibility that a low-luminosity \\lya\\ source may be visible at $z>\\zre$ if it is located within the ionized bubble of a more luminous neighbor.  In this case, the neighbor might be visible to the \\lya\\ survey itself or to a Lyman break survey of the region (such as the NDWFS).  The required luminosity corresponds to $I_{AB} \\la 24-7.5*lg(R/2 \\Mpc)$, for a Stromgren sphere radius $R$ and a normal starburst ionizing spectrum.  This number is only a lower bound because it assumes $\\fesc = 1$, which is almost certainly false for the ionizing photons.  A search for NDWFS sources with $I_{AB} < 24$ and nondetections at the $2\\sigma$ level in both $R$ and $B_W$ filters finds no viable candidates within $2 \\Mpc$ (projected physical separation) of any of the three spectroscopically confirmed sources.  In any case, Stromgren spheres of bright neighbors may explain an isolated \\lya\\ emitter prior to reionization (such as the Hu et al. source at $z=6.56$), but these Stromgren spheres include a minority of the volume before the overlap phase of reionization.  Thus, a statistical sample of \\lya\\ emitters such as the LALA survey provides should still see a drop in counts by a factor of $\\ga 2$ at the overlap phase.  The fact that we see no such drop in our $z=5.7$ sample (RM01) reinforces the argument that $\\zre > 5.7$ on the line of sight to the LALA Bootes field.  The neutral fraction required to effectively scatter most of the \\lya\\ flux from low-luminosity \\lya\\ sources can be estimated by setting $\\tau \\approx 1$ at a velocity $100 \\Delta v_{100} \\kms$ on the red side of line center.  RM01 find that this requires a homogeneously mixed neutral fraction $\\approx 0.1 \\Delta v_{100} [(1+z)/6.7]^{-3/2}$.  Figure~\\ref{rei_sum} summarizes our present observational understanding of reionization by combining this estimate for the neutral fraction limits placed by \\lya\\ sources with the mass-averaged neutral fraction evolution that Fan et al (2002) infer from Gunn-Peterson trough searches in high redshift quasars (Becker et al 2001; Pentericci et al 2002).  More detailed modeling that combines the \\lya\\ luminosity function with a more realistic, inhomogeneous IGM is planned and will reduce the uncertainties in figure~\\ref{rei_sum}.  Still, because the Gunn-Peterson trough cannot effectively probe the $x_{HI} \\gg 0.01$ regime, it is already clear that the \\lya\\ emitting galaxies have a role to play in extending our knowledge of reionization to $z \\ga 6.5$.   \\begin{figure}[ht] \\plotone{rei_sum.eps} \\caption{ Observational constraints on reionization.  The solid circle shows the upper limit on the (spatially homogeneous) neutral fraction inferred from \\lya\\ galaxy statistics at $z\\approx 5.7$ (this paper and RM01). The open squares with error bars show the evolution of the mass-averaged neutral fraction $x_{HI}$ that Fan et al (2002) derive from the \\lya\\ forest and Gunn-Peterson trough in a sample of quasars up to redshift $z=6.28$, using an inhomogeneous IGM model.  Finally, the open circle shows the upper limit to the neutral fraction suggested by the discovery of a \\lya-emitting galaxy at $z=6.56$ (HCM~6a; Hu et al 2002).  Several loopholes might allow this source to be observed in a predominantly neutral IGM (e.g., Haiman 2002), but we regard these as unlikely (see section~\\ref{disc}). \\label{rei_sum}} \\end{figure}    To summarize, we have presented spectra confirming redshifts $z\\approx 5.7$ for three \\lya\\ emitting galaxies selected from the Large Area Lyman Alpha (LALA) survey.  These objects were among four candidates observed from a total sample of 18 candidates.  This implies a total sample of $\\sim 13$ real $z\\approx 5.7$ galaxies, easily the largest uniformly selected galaxy sample at $z > 5.5$.  These \\lya\\ emitters all have narrow lines (inconsistent with broad-lined quasars) and very weak to absent continuum emission.  They belong to the class of high equivalent width \\lya\\ sources found in the LALA survey at $z\\approx 4.5$.  The population properties of these high equivalent width objects are reasonably explained by young galaxies with comparatively low metal and dust content (Malhotra \\& Rhoads 2002).  The $z\\approx 5.7$ population is similar to the $z\\approx 4.5$ one in its number counts.  This suggests that the sources reside in a predominantly ionized universe and hence that the reionization redshift $\\zre > 5.7$."
    },
    "0209/astro-ph0209291_arXiv.txt": {
        "abstract": "The location of the images in a multiple-image gravitational lens system are strongly dependent on the orientation angle of the mass distribution. As such, we can use the location of the images and the photometric properties of the visible matter to constrain the properties of the dark halo. We apply this to the optical Einstein Ring system 0047-2808 and find that the dark halo is almost spherical and is aligned in the same direction as the stars to within a few degrees. ",
        "introduction": "Numerical simulations of Cold Dark Matter (CDM) have been very successful in reproducing the observed large scale structure of the universe. The CDM model predicts that the dark matter (DM) haloes of today's galaxies are assembled through successive mergers of smaller haloes. Simulations using only dark matter predict that the haloes should be quite prolate, however it is not clear how gas and/or stars interacting with the dark matter will change the shape of the halo. Studies have suggested that the DM halo can become more or less cuspy ({El-Zant}, {Shlosman}, \\&  {Hoffman}, 2001; {Tissera} \\& {Dominguez-Tenreiro}, 1998) and rounder ({Evrard}, {Summers}, \\& {Davis}, 1994; {Dubinski}, 1994) after the interaction with stars and gas. An important test of galaxy formation and evolution models will be to compare the shape and profile of galaxy haloes with observed haloes. Thus, simple questions such as: ``Do we expect the visible and dark matter to be aligned in elliptical galaxies?'' and ``Is the dark matter density in the central regions changed by the gravitational dominance of the stars?'' must be answered with observations. For instance: the Milky Way, despite being a spiral galaxy, appears to have an almost spherical halo ({Ibata} {et~al.}, 2001).  Gravitational lensing offers a method to tightly constrain the shape of DM haloes in the population of medium redshift ($0.1 < z < 1.0$) lens galaxies. The image positions in a lens system are highly sensitive to the orientation of the overall mass profile. {Keeton}, {Kochanek}, \\&  {Falco} (1998) showed that the \\emph{overall} mass distribution is typically aligned with the visible matter using a sample of lens galaxies and a simple SIE mass model. However, depending on the lens galaxy, the stellar mass can contribute a substantial fraction of the total mass inside the image. The extreme case is the lensed QSO 2237+0305 where the dark matter constitutes only 4\\% of the projected mass inside the images ({Trott} \\& {Webster}, 2002). In this case we expect the visible matter orientation and the total matter orientation derived from a lensing analysis to be very similar. The logical next step is to use a more complicated (stars + halo) model for the lens galaxy to determine the properties of the DM halo alone.  In this paper we use an implementation of the LensMEM algorithm ({Wallington}, {Kochanek}, \\&  {Narayan}, 1996) and a stars+halo lens model to study the optical Einstein Ring 0047-2808 ({Warren} {et~al.}, 1999, 1996) using data from the HST. This system is well suited for the study because it is an isolated lens galaxy so we expect any external shear contributions to be small. The system is a $z=0.485$ elliptical which is lensing a background starbursting galaxy at $z=3.6$.  The algorithm we employ performs a non-parametric source reconstruction to match the observed data for a given lens model. The goodness-of-fit of the model is calculated using a $\\chi^2$ taking into account the degrees of freedom used in the source. In this paper we assume $H_0=70$ kms$^{-1}$Mpc$^{-1}$ and $(\\Omega_m,\\Omega_{\\lambda})=(0.3,0.7)$. ",
        "conclusions": "By using a lens model which separates the stars from the halo, we have been able to determine some of the basic properties of the dark matter halo in the lens system 0047-2808. We find that the projected ellipticity of the halo is not circular, but is substantially rounder than the observed stellar ellipticity. A small range of halo core radii values ($1.5\\arcsec < r_c < 2.5\\arcsec$) allow the projected halo mass to be circular.  The halo's core, modelled as a constant density region, must be $< 7\\arcsec$ to fit the observation. The core size could be further constrained by applying realistic limits to the stellar M/L which we intend to do in further work.  Finally, we find that although the halo is less elliptical than the stars, the orientation angle of the star's and halo's major axis are the same within errors."
    },
    "0209/astro-ph0209258_arXiv.txt": {
        "abstract": "{Deep \\ha$+$\\nitrogen~CCD images have been obtained in the area of the Pegasus Constellation. The resulting mosaic covers an extent of $\\sim$ 7\\degr.5$\\times$~8\\degr.5 and filamentary and diffuse emission is discovered. Several long filaments (up to $\\sim$1\\degr) are found within the field, while diffuse emission is present mainly in the central and northern areas. The filaments show variations in the intensity along their extent suggesting inhomogeneous interstellar clouds. Faint soft X--ray emission is also detected in the ROSAT All--Sky Survey. It is mainly concentrated in the central areas of our field and overlaps the optical emission. The low ionization images of \\sulfur\\ of selected areas mainly show faint diffuse emission, while in the medium ionization images of \\oxygen\\ diffuse and faint filamentary structures are present. Spectrophotometric observations have been performed on the brightest filaments and indicate emission from photoionized or shock--heated gas. The sulfur line ratios indicate electron densities below $\\sim$600 cm$^{-3}$, while the absolute \\ha\\ emission lies in the range of 1.1 -- 8.8 $\\times$ \\flux. The detected optical line emission could be part of a single or multiple supernova explosions. ",
        "introduction": "Several surveys have been made in the last decade concerning galactic supernova remnants (SNRs -- Arendt 1989; Seward 1990; Koo \\& Heiles 1991; Saken et al. 1992;). Green (2001) published a revised catalogue containing 231 Galactic SNRs and details for a number of possible and probable SNRs. In an effort to deepen our knowledge on the properties of the optically detected remnants, many imaging and spectral observations have been performed (Fesen \\& Hurford 1995, Fesen et al. 1995, 1997; Boumis et al. 2001; Mavromatakis et al. 2001, 2002a, 2002c) while, new optical SNRs have also been discovered (Boumis et al. 2002; Mavromatakis \\& Strom 2002; Mavromatakis et al. 2002b).  The Pegasus constellation is a region without any historical record of SN events. The Palomar Observatory Sky Survey (POSS) plates do not provide clear evidence of optical emission however, careful examination of the POSS plates reveals very weak extended optical emission. In particular, two of the filaments were detected on the POSS plates and we considered that such structures in high galactic latitudes would be interesting to study. The imaged area was expanded as more emission line structures were discovered. However, the final field size was restricted by the available telescope time. The published radio maps do not provide evidence for excess non--thermal emission which could be attributed to a supernova remnant.  In this paper, we report the discovery of faint optical filamentary and diffuse X--ray emission from the Pegasus constellation. We present an \\hnii\\ mosaic which covers an area of $\\sim$7\\degr.5$\\times$~8\\degr.5, and \\sii~and \\oiii~images of selected regions showing filamentary and diffuse structures. Spectrophotometric observations of the brightest filaments were also performed and the emission lines were measured. In Sect. 2, we present information concerning the observations and data reduction, while the results of the imaging and spectral observations are given in Sect. 3 and 4, respectively. Information about the X--ray data are given in Sect. 5, while in the last section (Sect. 6) we discuss the physical properties of the newly detected structures. ",
        "conclusions": "Unknown filamentary and diffuse structures have been discovered in Pegasus constellation through deep imaging and spectral observations. These structures appear more filamentary in \\hnii\\ emission lines images than the \\oiii\\ and \\sii\\ images. Diffuse X--ray emission is detected in the area and may be associated with the optical line emission. The flux calibrated long--slit spectra indicate that the emission arises from both photoionized and shock heated gas. The \\siirat\\ ratio indicates low electron densities while shock velocities around 100 \\vel\\ were found. The equivalent hydrogen column density was estimated from the extinction derived from \\ha/\\hbeta\\ ratio to be between 1.0 to 3.8 $\\times 10^{21}$~cm$^{-2}$. It may be possible that most of the optical emission is part of one or more supernova remnants."
    },
    "0209/hep-ex0209028_arXiv.txt": {
        "abstract": "A search for the relic neutrinos from all past core-collapse supernovae was conducted using 1496 days of data from the \\SK\\, detector.  This analysis looked for electron-type anti-neutrinos that had produced a positron with an energy greater than 18~MeV.  In the absence of a signal, 90\\% C.L. upper limits on the total flux were set for several theoretical models; these limits ranged from 20 to 130~$\\bar{\\nu}_e$~cm$^{-2}$~s$^{-1}$.  Additionally, an upper bound of $1.2\\; \\bar{\\nu}_e$~cm$^{-2}$~s$^{-1}$ was set for the supernova relic neutrino flux in the energy region $E_\\nu > 19.3$~MeV. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/hep-ph0209301_arXiv.txt": {
        "abstract": "\\noindent Properties of neutrinos, the lightest of all elementary particles, may be the origin of the entire matter-antimatter asymmetry of the universe. This requires that neutrinos are Majorana particles, which are equal to their antiparticles, and that their masses are sufficiently small. Leptogenesis, the theory explaining the cosmic matter-antimatter asymmetry, predicts that all neutrino masses are smaller  than $0.2$~eV, which will be tested by forthcoming laboratory experiments and by cosmology. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209014_arXiv.txt": {
        "abstract": "We have detected modulation of the Ca II H\\&K reversal structure in four out of five 51~Peg-type stars whose planets have orbital periods between 3 and 4 days. We observe two effects in the K-core: (1) a broad 3-\\AA~ variation at $\\approx$1\\% level and (2) changes on a scale of 0.5 \\AA~ ($\\sim$1-3\\%) in each of the three reversal features. The nightly variations are coherent in both H and K.  From differential radial velocities measured to better than 20 m/s, up-to-date phases were extracted.  The enhancements in the reversals tend to be greatest at the sub-planetary points which may imply that there is a magnetic interaction between the star's outer layers and the magnetosphere of the planet.  These high-S/N (500 per pixel in the continuum) and high-resolution (R = 110,000) data are too few to confirm orbital synchronization. ",
        "introduction": "Current planet detection methods give basic information on the planet: a minimum mass if the orbital inclination is not known, an estimate of surface temperature, and, in the case of the one transiting system, a possible sodium detection in the planet's atmosphere.  There remains a lack of constraints on the planet's structure leaving astronomers to explore new observational probes.  Cuntz, Saar, \\& Musielak (2000) suggested that there may be an observable interaction between a parent star and a close-in giant planet.  The effect could be tidal, magnetic, or a combination of the two.  Such an interaction could manifest itself in the form of chromospheric and coronal heating which are predicted to produce a variation of a few percent (Saar \\& Cuntz 2001). If the surface of the star is indeed heated by the planet, then a hot-spot would follow the planet around its orbit, appearing hottest at the sub-planetary point if the interaction is with the planet's magnetosphere, or twice per orbit if the interaction is tidal. There are several indicators of increased chromospheric activity, one of which is the Ca II H \\& K line reversal.  Due to the accessibility of these lines with ground-based telescopes, we chose to monitor the spectral region over several nights for five planetary systems whose orbital periods are between 3.1 and 4.6 days. ",
        "conclusions": ""
    },
    "0209/astro-ph0209222_arXiv.txt": {
        "abstract": "Here we report some results from an ESO-VLT programme to observe individual stars in nearby dwarf galaxies at high resolution with the UVES spectrograph (Tolstoy, Venn, Shetrone, Primas, Hill, Kaufer \\& Szeifert 2002, submitted to AJ).  We mainly concentrate on illustrating the issues and uncertainties surrounding our efforts to determine the ages of stars for which we have accurately measured [Fe/H] and [$\\alpha$/Fe]. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209478_arXiv.txt": {
        "abstract": "We report the detection of polarized anisotropy in the Cosmic Microwave Background radiation with the Degree Angular Scale Interferometer (DASI), located at the Amundsen-Scott South Pole research station.  Observations in all four Stokes parameters were obtained within two $3\\fdg4$ FWHM fields separated by one hour in Right Ascension.  The fields were selected from the subset of fields observed with DASI in 2000 in which no point sources were detected and are located in regions of low Galactic synchrotron and dust emission. The temperature angular power spectrum is consistent with previous measurements and its measured frequency spectral index is $-0.01$ ($-0.16$ -- 0.14 at 68\\% confidence), where 0 corresponds to a 2.73~K Planck spectrum.  The power spectrum of the detected polarization is consistent with theoretical predictions based on the interpretation of CMB anisotropy as arising from primordial scalar adiabatic fluctuations.  Specifically, $E$-mode polarization is detected at high confidence ($4.9\\sigma$). Assuming a shape for the power spectrum consistent with previous temperature measurements, the level found for the $E$-mode polarization is 0.80 (0.56 -- 1.10), where the predicted level given previous temperature data is 0.9 -- 1.1. At 95\\% confidence, an upper limit of 0.59 is set to the level of $B$-mode polarization with the same shape and normalization as the $E$-mode spectrum.  The $TE$ correlation of the temperature and $E$-mode polarization is detected at 95\\% confidence, and also found to be consistent with predictions. These results provide strong validation of the underlying theoretical framework for the origin of CMB anisotropy and lend confidence to the values of the cosmological parameters that have been derived from CMB measurements. ",
        "introduction": "Measurements of the Cosmic Microwave Background (CMB) radiation reveal the conditions of the universe when it was $\\sim400,000$ years old with remarkable precision. The three most fundamental properties of the CMB are its frequency spectrum and the angular power spectra of the temperature and polarization fluctuations.  The frequency spectrum was well determined by the COBE FIRAS instrument \\markcite{mather94,fixsen96}({Mather} {et~al.} 1994; {Fixsen} {et~al.} 1996). The initial detection of temperature anisotropy was made on large angular scales by the COBE DMR instrument \\markcite{smoot92}(Smoot {et~al.} 1992) and recently there has been considerable progress in measuring the anisotropy on finer angular scales \\markcite{miller99,halverson02,netterfield02,lee02}({Miller} {et~al.} 1999; {Halverson} {et~al.} 2002; {Netterfield} {et~al.} 2002; {Lee} {et~al.} 2001). There have been many efforts to measure the polarization (see below) but so far, detection of this property of the CMB has remained beyond the reach of the most sensitive observations.  In the past several years, a standard cosmological model has emerged \\markcite{hudodelson02}(see, e.g., {Hu} \\& {Dodelson} 2002). In this model, the structure of the CMB angular power spectrum at degree angular scales is assumed to arise from acoustic oscillations of the photon-baryon fluid sourced by primordial scalar adiabatic fluctuations. At decoupling, the modes at maximal amplitude lead to excess power in the observed CMB angular power spectrum resulting in a harmonic series of peaks and troughs. Within this theoretical framework, and given knowledge of the temperature angular power spectrum, a prediction can be made for the level of the CMB polarization with essentially no free parameters \\markcite{kaiser83,bond84,polnarev85,kamionkowski97,zaldarriaga97,hu_w97,kosowsky99}({Kaiser} 1983; Bond \\& Efstathiou 1984; {Polnarev} 1985; Kamionkowski, Kosowsky, \\&  Stebbins 1997; {Zaldarriaga} \\& {Seljak} 1997; Hu \\& White 1997; {Kosowsky} 1999). A determination of the CMB polarization would therefore provide a critical test of the underlying theoretical framework \\markcite{huspergelwhite97,kinney01,bucher01}({Hu}, {Spergel}, \\& {White} 1997; {Kinney} 2001; {Bucher}, {Moodley}, \\& {Turok} 2001) and therefore of the validity of cosmological parameters derived from CMB measurements. Polarization measurements also offer the potential to triple the number of observed CMB quantities and to enhance our ability to constrain cosmological parameters.  CMB polarization arises from Thompson scattering by electrons of a radiation field with a local quadrupole moment \\markcite{rees68}({Rees} 1968). In the primordial plasma, the local quadrupole moment is suppressed until the photon mean free path grows during decoupling. At this time, the largest contribution to the local quadrupole is due to Doppler shifts induced by the velocity field of the plasma \\markcite{zaldarriaga_h95}(Zaldarriaga \\& Harari 1995). In this way, CMB polarization directly probes the dynamics at the epoch of decoupling. For a Fourier mode of the acoustic oscillations, the electron velocities are perpendicular to the wavefronts, leading to either a parallel or perpendicular alignment of the resulting polarization. These polarization modes are referred to as the scalar $E$-modes in analogy with electric fields; they have no curl component. Since the level of the polarization depends on velocity, one expects that the peaks in the scalar $E$-mode power spectrum correspond to density modes that are at their highest velocities at decoupling and are therefore at minimum amplitude. The location of the harmonic peaks in the scalar $E$-mode power spectrum are therefore expected to be out of phase with the peaks in the temperature spectrum \\markcite{kamionkowski97,zaldarriaga97,hu_w97}(Kamionkowski {et~al.} 1997; {Zaldarriaga} \\& {Seljak} 1997; Hu \\& White 1997).  Primordial gravity waves will lead to polarization in the CMB \\markcite{polnarev85,crittenden93}({Polnarev} 1985; {Crittenden}, {Davis}, \\&  {Steinhardt} 1993) with an $E$-mode pattern as for the scalar density perturbations, but will also lead to a curl component, referred to as $B$-mode polarization \\markcite{kamionkowski97,seljak97}(Kamionkowski {et~al.} 1997; {Seljak} \\& {Zaldarriaga} 1997).  The $B$-mode component is due to the intrinsic polarization of the gravity waves. In inflationary models, the level of the gravity wave induced $B$-mode polarization power is set by the energy scale of inflation to the fourth power. While the detection of $B$-mode polarization would provide a critical test of inflation, the signal is likely to be very weak and may have an amplitude that is effectively unobservable \\markcite{lyth97}({Lyth} 1997). Furthermore, distortions to the scalar $E$-mode polarization by the gravitational lensing of the intervening large scale structure in the universe will lead to a contaminating B-mode polarization signal which will severely complicate the extraction of the gravity-wave induced signal \\markcite{zaldarriaga98,hu02,knox02}({Zaldarriaga} \\& {Seljak} 1998; {Hu} \\& {Okamoto} 2002; {Knox} \\& {Song} 2002). The possibility, however, of directly probing the universe at energy scales of $\\sim10^{16}$ GeV by measuring the gravity-wave induced polarization \\markcite{kamionkowski99}(see, e.g., Kamionkowski \\& Kosowsky 1999) is a compelling goal for CMB polarization observations.  Prior to the results presented in this paper, only upper limits have been placed on the level of CMB polarization. This is due to the low level of the expected signal, demanding sensitive instruments and careful attention to sources of systematic uncertainty \\markcite{staggs99}(see {Staggs}, {Gunderson}, \\& {Church} 1999, for a review of CMB polarization measurements).  The first limit to the degree of polarization of the CMB was set by \\markcite{penzias65}{Penzias} \\& {Wilson} (1965) who stated that the new radiation that they had discovered was isotropic and unpolarized within the limits of their observations. Over the following 20 years, dedicated polarimeters have been used to set much more stringent upper limits on angular scales of order several degrees and larger \\markcite{caderni78,nanos79,lubin79,lubin81,lubin83,sironi97}({Caderni} {et~al.} 1978; {Nanos} 1979; {Lubin} \\& {Smoot} 1979, 1981; {Lubin}, {Melese}, \\& {Smoot} 1983; Sironi {et~al.} 1997). The current best upper limits for the $E$-mode and  $B$-mode polarizations on large angular scales are 10$~\\mu$K at 95\\% confidence for the multipole range $2 \\le l \\le 20$, set by the POLAR experiment \\markcite{keating01}({Keating} {et~al.} 2001).  On angular scales of order one degree, \\markcite{wollack93}{Wollack} {et~al.} (1993) used the Saskatoon experiment to set the first upper limit to the CMB polarization ($25~\\mu$K at 95\\% confidence for $l \\sim 75$); this limit is also noteworthy in that it was the first limit that was lower than the level of the CMB temperature anisotropy. The current best limit on similar angular scales was set by the PIQUE experiment \\markcite{hedman02}({Hedman} {et~al.} 2002), who reported a 95\\% confidence upper limit of $8.4~\\mu$K to the $E$-mode signal, assuming no B-mode polarization. \\markcite{cartwright02}{Cartwright} {et~al.} (2002) presented a preliminary analysis of CBI data that indicated an upper limit similar to the PIQUE result, but on somewhat smaller scales.  On much finer angular scales of order an arcminute, polarization measurements have also been pursued and upper limits set \\markcite{partridge97,subrahmanyan00}(e.g., {Partridge} {et~al.} 1997; {Subrahmanyan} {et~al.} 2000).  However, at these angular scales, corresponding to multipoles $\\sim 5000$, the level of the primary CMB anisotropy is strongly damped and secondary effects due to the interactions with large scale structure in the universe are expected to dominate \\markcite{hudodelson02}({Hu} \\& {Dodelson} 2002).  In this paper, we present the results of CMB polarization measurements made with the Degree Angular Scale Interferometer (DASI) located at the NSF Amundsen-Scott South Pole research station during the 2001 and 2002 austral winter seasons. DASI was successfully used to measure the temperature anisotropy from $140 < l < 900$ during the 2000 season. Details of the instrument, the measured power spectrum and the resulting cosmological constraints were presented in a series of three papers ~\\markcite{leitch02a,halverson02,pryke02}(Leitch {et~al.} 2002b; {Halverson} {et~al.} 2002; {Pryke} {et~al.} 2002, hereafter Papers, I, II and III, respectively). Prior to the start of the 2001 season, DASI was modified to allow polarization measurements in all four Stokes parameters over the same $l$ range as the previous measurements. The modifications to the instrument, observational strategy, data calibration and data reduction are discussed in detail in \\markcite{leitch02b}Leitch {et~al.} (2002a) (hereafter Paper IV).  This paper is organized as follows. In  \\S\\ref{sec:instrument} we briefly summarize the modifications to the instrument, the observing strategy and the data calibration from Paper IV and in \\S\\ref{sec:observations} the CMB observations and data reduction are discussed. The noise model and detection of signal in our data are discussed in \\S\\ref{sec:noiseandsignal}. The analysis method is presented in  \\S\\ref{sec:lhformalism} and the results from the likelihood analysis which models and parameterizes the signal in terms of CMB polarization and temperature angular power spectra are given in \\S\\ref{sec:lhresults}. In \\S\\ref{sec:systematics} we discuss systematic uncertainties including instrumental effects and the possibility of foreground contamination. Conclusions are summarized in \\S\\ref{sec:conclusion}. ",
        "conclusions": "\\label{sec:conclusion}  In this paper, we present the first detection of polarization of the CMB.  These results are the product of two years of observations with the DASI telescope within two $3\\fdg4$ FWHM fields.  For the observations described here, DASI was reconfigured with achromatic polarizers to provide sensitivity in all four Stokes parameters.  As described in Paper IV, observations of both polarized and unpolarized astronomical sources give us confidence that the gain and instrumental polarization of the telescope have been precisely characterized.  We have performed extensive consistency tests on various splits and subsets of the visibility data.  For those modes expected to have high s/n, a simple comparison of the polarization data with the measured instrumental noise results in a robust detection of a polarized signal with a significance of approximately 5$\\sigma$. These tests show no indication of systematic contamination and strongly support a celestial origin of the polarized signal. We employ a full likelihood analysis to determine confidence intervals for temperature and polarization models parameterized by shaped and flat band powers. Unlike the DASI temperature angular power spectrum reported in Paper II, the temperature power spectrum presented in this paper is strongly dominated by sample variance. However, the high s/n achieved in the deep polarization presented here permits a precise determination of the spectral index of the CMB temperature anisotropy, $\\beta_T = -0.01$ ($-0.16$ -- 0.14 at 68\\% confidence).   A likelihood ratio test is used to demonstrate the agreement of the observed CMB temperature and polarization anisotropy signals with a concordance $\\Lambda$CDM model,  and strongly rejects models without CMB polarization. From this analysis we determine that we have detected $E$-mode CMB polarization with a significance of $4.9\\sigma$. Specifically, assuming a shape for the power spectrum consistent with previous temperature measurements, the level found for the $E$-mode polarization is 0.80 (0.56 -- 1.10), where the predicted level given previous temperature data is 0.9 -- 1.1.   The spectral index determined for the observed $E$-mode polarization signal, $\\beta_E = 0.17$ ($-1.63$ -- 1.92), is consistent with CMB. At 95\\% confidence, an upper limit of 0.59 is set to the level of $B$-mode polarization with the same shape and normalization as the $E$-mode spectrum.  The $TE$ correlation of the temperature and $E$-mode polarization is detected at 95\\% confidence, and also found to be consistent with predictions.       We have considered the possibility that our results are contaminated by foreground emission in the form of a distribution of polarized radio point sources and high Galactic latitude synchrotron emission. Simulated distributions of radio sources are shown to contribute insignificant polarization compared to the observed signal. The strongest constraints against diffuse synchrotron emission come from the DASI dataset itself. The observed $TE$ correlation, combined with the precisely thermal spectrum of the temperature anisotropy creates a compelling argument that the $E$-mode polarization we observe was created at the surface of last scattering. Although the constraint on the $E$-mode polarization spectral index is not nearly as strong as those for the temperature anisotropy, this result is incompatible with Galactic synchrotron as the source of the observed polarization at nearly $2\\sigma$.  In general, foregrounds are expected to produce comparable amplitude in both $E$- and $B$-mode spectra. Our data therefore provide additional evidence against a strong contribution from foreground emission to the degree that our results limit the ratio of $B$- to $E$-mode polarization.   The likelihood results and tests to which we have subjected the data provide self-consistent and strong support for the detection of the polarization induced on the CMB at the surface of last scattering. These results provide strong validation of the underlying theoretical framework for the origin of CMB anisotropy and lend confidence to the values of the cosmological parameters that have been derived from CMB measurements."
    },
    "0209/astro-ph0209152_arXiv.txt": {
        "abstract": "We use the block model to generate merger trees for the first star clusters in a $\\Lambda$CDM cosmology.  Using a simple collapse model and cooling criterion, we determine which halos are able to form stars before being disrupted by mergers.  We contrast the mass functions of all the resulting star clusters and those of primordial composition, i.e. star clusters that have not been contaminated by subclusters inside them.  In confirmation of previous work, two generations of primordial star clusters are identified: low-temperature clusters that cool via molecular hydrogen, and high-temperature clusters that cool via electronic transitions.  The former dominate by number, but the two populations contain a similar mass with the precise balance depending upon the details of the model.  We speculate on the current-day distribution of Population-{\\sc iii} stars. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209364_arXiv.txt": {
        "abstract": "We present three-dimensional numerical simulations of the interaction of a circular-orbit planet with a protoplanetary disk.  We calculate the flow pattern, the accretion rate, and torques on a planet. We consider planet masses ranging from 1 Earth mass to 1 Jupiter mass. ",
        "introduction": "Planet formation is believed to involve accretion from the surrounding disk of material. We consider this gas accretion onto a planet. A non-Keplerian flow pattern develops near the planet. Furthermore, the planet is subject to gravitational torques due to its interaction with the gas disk which result in planetary migration. We have extended the 2D simulations by Lubow, Seibert, and Artymowicz (1999) to 3D, in order to analyze these effects more realistically. We expect 3D effects to be important for planet masses below $1$M$_{\\rm J}$, since the Hill sphere radius becomes smaller than the disk thickness. ",
        "conclusions": "\\begin{itemize} \\item The migration rate scales linearly with $M_{\\rm p}$ for planets that do not open a gap.  The results are in excellent agreement with the linear analysis of Tanaka et al.~(2002).  Planets with masses $>0.1$ M$_{\\rm J}$ migrate on the disc's viscous timescale. \\item The mass accretion rate increases linearly with planet mass $M_{\\rm p}$ for low-mass planets and peaks at $\\approx 0.1$ M$_{\\rm J}$. At higher masses, the mass accretion rate declines as a gap is opened in this disk. \\end{itemize}"
    },
    "0209/nucl-ex0209017_arXiv.txt": {
        "abstract": "The \\pa\\ reaction is recognized as one of the most important reaction for nova gamma--ray astronomy as it governs the early $\\leq$ 511~keV emission. However, its rate remains largely uncertain at nova temperatures due to unknown low--energy resonance strengths. In order to better constrain this reaction rate, we have studied the one--nucleon transfer reaction, \\dpa, at the CRC-RIB facility at Louvain~La~Neuve. ",
        "introduction": "Gamma--ray emission from classical novae is dominated, during the first hours, by positron annihilation following the beta decay of radioactive nuclei. The main contribution comes from the decay of \\fo\\ (half--life of 110~mn) and hence is directly related to \\fo\\ nucleosynthesis during the outburst\\cite{Gom98,Her99,F00}. The \\pa\\ reaction is the main mode of \\fo\\ destruction and has been the object of many recent experiments\\cite{Gra00,Bar01,Bar02} but its rate remains poorly known at nova temperatures. The uncertainties are directly related to the unknown proton widths of the first three \\nen\\ levels above proton emission threshold ($E_x$, $J^\\pi$ = 6.419~MeV, 3/2$^+$; 6.437~MeV, 1/2$^-$ and 6.449~MeV, 3/2$^+$). The tails of the corresponding resonances (at respectively $E_R$ = 8~keV, 26~keV and 38~keV) can dominate the astrophysical factor in the relevant energy range\\cite{F00}. As a consequence of these nuclear uncertainties, the \\fo\\ production in nova and the early gamma--ray emission is uncertain by a factor of $\\approx$300\\cite{F00}. This supports the need of new experimental studies to improve the reliability of the predicted annihilation gamma--ray fluxes from novae. ",
        "conclusions": ""
    },
    "0209/astro-ph0209109_arXiv.txt": {
        "abstract": "We have developed methods for tracing rays and performing radiative transfer through a magnetoactive plasma in a general relativistic environment.  The two electromagnetic plasma modes propagate differently due to a combination of dispersive and gravitational effects.  We have found that, when given an appropriate environment surrounding the central black hole, it is indeed possible to generate a significant degree of circular polarisation without an appreciable amount of linear polarisation due to these effects alone. ",
        "introduction": "Polarisation measurements now exist for many accreting compact objects (ostensibly black holes) at a number of frequencies. Typically, emission mechanisms are called upon to explain polarisation observations (see \\eg \\opencite{Brom-Meli-Liu:01};~\\opencite{West:59}). However, recent observations of Sgr A$^*$ and M81 (see \\eg \\opencite{Brun-Bowe-Falc-Mell:01} ;~\\opencite{Bowe-Falc-Back:99};~ \\opencite{Saul-Macq:99}), as well as a number of blazars (see \\eg \\opencite{Kome-Robe-Miln-Rayn-Cook:84}), have exhibited a significant amount of circular polarisation (CP) apparently unassociated with any linear polarisation (LP).  This has proved difficult to explain with the standard set of polarised emission mechanisms alone, often requiring specialised magnetic field or disk structures.  In addition to its anomalous size, the CP typically does not change in sign despite having a variability larger in frequency and magnitude than that of the LP (if present) and total intensity.  Both of these suggest that the region  responsible for the polarisation is compact, and perhaps the central  compact object is playing a significant role, if only in moderating the local plasma and/or magnetic field structure.  As a result, a substantial effort has been made to investigate the effects of the accretion environment upon polarisation.  These efforts have been primarily concentrated in two directions: ({\\em i}) propagation effects due to a magnetised plasma (see \\eg \\opencite{Rusz-Bege:02} ;~\\opencite{Macq:02};~\\opencite{Jone-ODel:77a} \\& \\citeyear{Jone-Odel:77b}), and ({\\em ii}) vacuum propagation effects due to general relativity, in particular near a rotating black hole (see \\eg \\opencite{Falc-Meli-Agol:00};~\\opencite{Agol:97};~ \\opencite{Laor-Netz-Pira:90};~\\opencite{Conn-Star-Pira:80}).  Most of these require an initial source of polarisation, presumably provided by the emission mechanism.  A notable exception is the scintillation mechanism proposed by \\citeauthor{Macq-Melr:00}.  However, for realistic conditions this has been unable to produce a polarisation of constant sign.  The studies regarding ({\\em i}) have thus far ignored general relativistic effects (and hence are inapplicable near the compact object), focusing upon non-dispersive plasma effects, \\eg Faraday rotation and conversion. The work considering ({\\em ii}) has found general relativity to have a depolarising influence on LP due to frame dragging for photons passing near the black hole. However, the studies of general relativistic effects have ignored plasma effects completely, and hence are not always applicable in the case of a thick disk or when a dense and/or magnetised corona is present.  In contrast, magnetoionic effects, including dispersion, have been studied in detail in the context of radio waves in the upper atmosphere.  This has, of course, been done in the absence of general relativity, where it has been found that for a specific range in frequency the dispersive effects can have a significant impact upon the propagation and polarisation of the radio waves (see \\eg \\opencite{Budd:64}).  Here we present a fully general relativistic magnetoionic theory which takes into account general relativity as well as dispersive and non-dispersive plasma effects.  This is a natural, albeit currently less well developed, extension of the previous investigations into the polarisation effects of accretion flows onto compact objects.  The development of the theory can be succinctly separated into the problems of tracing rays and performing the radiative transfer.  As such, these proceedings will be presented in five sections with \\S2 discussing ray tracing, \\S3 explaining the radiative transfer, \\S4 presenting results for Bondi flows, and \\S5 containing conclusions. ",
        "conclusions": "Dispersive effects coupled with general relativistic effects will produce considerable amounts of CP when the plasma and/or cyclotron frequencies are commensurate with those being observed. This method of producing CP is unique in that it does not require a polarised emission mechanism --- even unpolarised emission will become polarised after passing near a black hole.  Unlike the non-dispersive processing mechanisms, \\eg Faraday conversion, this does not require uniform large scale magnetic fields over the entire disk.  Rather, only uniformity near the black hole horizon is necessary, where the black hole's influence can in principle moderate the magnetic field geometry.  This is neither dependent upon the details of the emission mechanism being employed nor contaminated by large degrees of LP.  The requirements of the dispersion mechanism place some constraints upon the emission mechanisms.  The first is that the mechanism must be able to operate near $\\omega_P^{},\\omega_B^{} \\sim \\omega$. This can be relaxed somewhat by having the black hole being backlit, eliminating the necessity for an emission mechanism that is capable of operating near the hole. A second constraint upon the emission mechanism is that it needs to have a large brightness temperature.  This is equivalent to the fact that the fraction of the total intensity propagating through the inner $\\sim 5-10 M$ must be larger than the CP fraction. For blazars, this all but rules out this mechanism (however jets and/or plasma distributions which use dispersion to magnify the emitting regions may yet make a difference). For Sgr A$^*$ and M81, brightness temperatures on the order of $10^{12}$ K are necessary, pushing the upper bounds given by the paucity of the X-ray fluxes.  Still, this remains a tenable source for the CP in low luminosity AGN and may be at work in Sgr A$^*$ and/or M81. Recent LP results (which confirm the earlier observations of \\opencite{Aitk_etal:00}) suggest that the gas density close to the black hole in Sgr A$^*$ is far lower than expected for a conservative accretion flow \\cite{Bowe-Wrig-Falc-Back:02}.  Therefore, the Sgr A$^*$ environment is conducive to seeing relativistic magnetoionic effects at high frequencies close to the black hole.  In addition to applications to low luminosity AGN, this mechanism can have implications in stellar mass black hole systems as well.  The degree of CP depends upon the relative sizes of the black hole and the accretion flow.  Hence, high mass X-ray binaries can be expected to exhibit a significant amount of CP if ({\\em i}) the accretion flow is optically thin at radio frequencies (\\eg if the black hole is being seen through the corona) and ({\\em ii}) magnetic fields are present, presumably generated via the magnetorotational instability and ordered by the black hole."
    },
    "0209/astro-ph0209279_arXiv.txt": {
        "abstract": "Not the way one might have thought.  In hydrodynamic simulations of galaxy formation, some gas follows the traditionally envisioned route, shock heating to the halo virial temperature before cooling to the much lower temperature of the neutral ISM.  But most gas enters galaxies without ever heating close to the virial temperature, gaining thermal energy from weak shocks and adiabatic compression, and radiating it just as quickly.  This ``cold mode'' accretion is channeled along filaments, while the conventional, ``hot mode'' accretion is quasi-spherical.  Cold mode accretion dominates high redshift growth by a substantial factor, while at $z<1$ the overall accretion rate declines and hot mode accretion has greater relative importance.  The decline of the cosmic star formation rate at low $z$ is driven largely by geometry, as the typical cross section of filaments begins to exceed that of the galaxies at their intersections. ",
        "introduction": " ",
        "conclusions": ""
    },
    "0209/astro-ph0209086_arXiv.txt": {
        "abstract": "Ultraviolet spectra from the GHRS instrument on board the {\\em Hubble Space Telescope} reveal the presence of a mysterious absorption feature in the Mg~II h \\& k lines of the nearby ($d=20.0$~pc) K5~III star $\\alpha$~Tau.  The narrow absorption looks like an interstellar absorption feature but it is in the wrong location based on our knowledge of the local ISM flow vector.  Since the absorption is close to the rest frame of the star, it has been interpreted as being from the interaction region between $\\alpha$~Tau's massive, cool wind and the interstellar medium, i.e., $\\alpha$~Tau's ``astrosphere''.  We compute hydrodynamic models of the $\\alpha$~Tau astrosphere in order to see if the models can reproduce the Mg~II absorption feature.  These models do predict that stellar wind material heated, decelerated, and compressed after passing through a termination shock a few thousand AU from the star should produce a Mg~II absorption feature with about the right width at roughly the right velocity.  However, our first models underestimate the Mg~II column density by an order of magnitude.  A much larger parameter search is necessary to see whether the observed Mg~II absorption can be reproduced by acceptable changes to the adopted stellar wind and ISM properties. ",
        "introduction": "Figure~1 shows the Mg~II k line from $\\alpha$~Tau (K5~III) observed by the Goddard High Resolution Spectrograph (GHRS) on board the {\\em Hubble Space Telescope} (HST) on 1994 April 8, first studied by Robinson et al.\\ (1998).  The center of the stellar line profile is dominated by very broad absorption from the massive, cool wind of the red giant star.  However, near the rest frame of the star there is a much narrower absorption feature, which is also seen in the Mg~II h line. The narrow absorption looks very much like interstellar absorption, and the strength of the absorption in the k and h lines has the expected 2 to 1 ratio of the oscillator absorption strengths, but Robinson et al.\\ (1998) noted that the location of the absorption makes an interstellar origin very unlikely. \\begin{figure}[t] \\plotfiddle{fig1.eps}{3.6in}{90}{60}{60}{233}{-10} \\caption{HST/GHRS Mg~II k line spectrum of $\\alpha$~Tau, plotted on a heliocentric velocity scale, with the rest frame of the star indicated by the vertical dashed line.  Broad absorption from $\\alpha$~Tau's wind dominates the central part of the profile.  The narrow absorption feature shaded in green is the absorption that is presumed to be from the wind/ISM interaction region.} \\end{figure}  The flow vector of the Local Interstellar Cloud (LIC) from Lallement et al.\\ (1995) suggests that LIC absorption should be at $+25.5$ km~s$^{-1}$, which is within the saturated core of the wind absorption in Figure~1, and is therefore undetectable. Interstellar absorption components other than that of the LIC are occasionally seen for stars as nearby as $\\alpha$~Tau ($d=20.0$~pc), but they are generally close to the expected LIC velocity.  Redfield \\& Linsky's (2002) survey of ISM absorption features within 100~pc demonstrates just how discrepant the location of the $\\alpha$~Tau absorption feature is, if it is interstellar.  Noting the absorption feature's location near the stellar rest frame, Robinson et al.\\ (1998) proposed that the absorption is instead from the interaction region between $\\alpha$~Tau's wind and the ISM.  The absorption would therefore be somewhat analogous to the H~I Ly$\\alpha$ absorption detected from the interaction regions between the winds of solar-like stars and the local ISM (Linsky \\& Wood 1996; Wood et al.\\ 2000, 2001, 2002; M\\\"{u}ller et al.\\ 2001).  This ``astrospheric'' (analogous with ``heliospheric'') absorption has proved to be a very useful diagnostic for the wind properties of solar-like stars (Wood et al.\\ 2002). ",
        "conclusions": ""
    },
    "0209/astro-ph0209445.txt": {
        "abstract": "We have constructed a very deep $K^\\prime$-selected multicolor $BVRIz^\\prime JK^\\prime$ sample of $439$ field galaxies. Based on this multicolor data, a photometric redshift for each sample galaxy was estimated. The overall redshift distribution N(z) for the $K^\\prime \\leq 21.0$ sample is consistent with previous observations, and for the first time we derive N(z) down to $K^\\prime=24.0$. After taking account of the dust extinction and selection effects of the sample, the observed N(z) distribution is well described with the PLE model, while the hierarchical galaxy formation model shows an apparent deficiency of galaxies especially at $z\\gtrsim2$. The photometric redshift and the best-fit SED model evaluations allow the derivation of the rest-frame $K^\\prime$, $B$, and UV($2000$\\AA)-band luminosity functions and their evolutions. The rest $K^\\prime$-band luminosity function shows almost no evolution up to $z=3$, while the rest $B$ luminosity function shows mild luminosity evolution, and the rest UV luminosity function shows strong luminosity evolution. These findings seem to be qualitatively in favor of the PLE model. No evolution in the rest $K^\\prime$-band luminosity function can also be consistent with the hierarchical galaxy formation model if $M/L_K$ decreases with redshift, that is, if the number density of $K^\\prime$-band luminosity- selected galaxies does not significantly change with redshift while the number density of stellar mass-selected galaxies decreases with look-back time. %At $z>3$, all the rest-frame luminosity functions decline their amplitudes. This trend corresponds to the evolution of the rest UV$(2000$\\AA$)-K^\\prime$ color, which gets bluer with increasing redshift. We also found that more massive galaxies are redder in this rest-frame color in every epoch. The rest-UV luminosity function of our $K^\\prime$-selected galaxies shows a much shallower faint end slope at $z=3$ than that of previous estimations for rest-UV selected Lyman break galaxies. As a consequence, the contribution to the global star formation rate of our $K^\\prime$-selected galaxies is about $42\\%$ of that derived from the integration of luminosity function of Lyman break galaxies at $z=3$. This result suggests that a large fraction of the star formation rate density at $z>1.5$ comes from the contribution from the faint ($M_{2000}>-20$) blue galaxy population at high redshift universe that have not yet obviously been identified. ",
        "introduction": "The large galaxy sample with redshift determined to faint magnitudes allows us to understand the evolution of galaxies. Previous deep spectroscopic surveys have shown several evolutionary trends for galaxies at $0\\lesssim z\\lesssim 1$, especially for the evolution of luminosity functions (LFs), which is one of the fundamental observational tools to trace the galaxy evolution. \\citet{lil95} have shown that the LF for red galaxies does not evolve significantly over $0\\lesssim z\\lesssim 1$, which infers that red massive galaxies must have already been assembled by $z=1$. On the other hand, their blue subsample is evolving, which indicates that they are formed later than the red subsample. \\citet{cow96} have shown the evidence of the ^^ ^^ downsizing\" trend in which more massive galaxies appear to be forming at higher redshifts.  Although it is important to see these evolutionary trends in the earlier epoch beyond $z=1$, it is difficult to construct a large galaxy sample in the familiar optical wavelength except for the distinct high z star-forming populations of Lyman break galaxies and/or Ly$\\alpha$ emitters at $z\\gtrsim 3$ \\citep[Paper II]{ste99,ouc03}. There are some possible interpretations of such luminous high z galaxies such as a core of today's massive $L>L^*$ galaxies \\citep{ste96} or as a low-mass starbursting building blocks of the more massive galaxies seen today \\citep{low97}. However, it remains uncertain how these $z>3$ galaxies are related to present-day galaxies. %Although dust extinction is skewing the estimate of their intrinsic luminosities, these observations indicate a scenario in which massive galaxies formed at $z\\gtrsim 3$ and were then followed by a sequence of less and less massive galaxies forming at lower and lower redshifts, leading down to the formation of dwarfs at recent ($z\\lesssim 0.5$) epochs.  One of the explicit difficulties in tracing systematically the evolution of galaxies is that a single observed frame does not correspond to a single rest frame over wide redshift range, say $0\\lesssim z\\lesssim 3$. The rest-optical frame in which we have seen the evolution trend at $0\\lesssim z\\lesssim 1$ shifts to the NIR region at $z\\gtrsim 1.5$. An observation in wide wavelength range is undoubtedly required to know the SEDs of galaxies and follow their evolution based on the same rest frame in which we can draw almost similar stellar contents at any redshifts. %It is undoubtedly required to observe in wide wavelength range and to know the SEDs of galaxies in order to follow their evolution based on the rest-frame in which we can draw almost similar stellar contents at any redshifts.  %2 rest-frmae\u0082\u00cc\u0092\u00c7\u0082\u00a2\u0095\u00fb\u0082\u00c9\u0082\u00cd\u0082\u00b1\u0082\u00ea\u0082\u00b1\u0082\u00ea\u0082\u00aa\u0082\u00a0\u0082\u00e9\u0081B  There are two approaches to probe the rest-frame luminosity of galaxies. One is to change the observed band accordingly and follow the redshifting rest frame. This straightforward approach was taken by many studies. \\citet{cow99} traced the rest-$B$ band up to $z=1$ by $UBVI$ multiband observation. \\citet[Paper IV]{fur03} used multicolor $BVRi^\\prime z^\\prime$ bands and traced the rest $B$ band LF up to $z=1.25$. NIR photometry for Lyman break galaxies enables them to derive their rest $V$-band luminosity at $z=3$ \\citep{sha01}. This approach gives a reliable way to trace the rest frame directly, though the study is restricted to a single rest band.  The other approach is to utilize the best-fit SED templates determined on photometric redshift method \\citep{saw97,rud01}. As have been already mentioned by many authors, the photometric redshift technique has less precision for redshift estimates than the spectroscopic redshift. However, spectroscopic redshifts are practical only for bright objects. Photometric redshifts can be extensively applied to fainter objects with a fair degree of confidence. They are also of great use in constructing a statisticaly large sample in the redshift range between $z=1.5$ and $z=2.5$, where redshifts can be identified only by NIR spectroscopy. It works predominantly as a statistical technique rather than as a constraint on parameters that determine the SED of an individual object. Studies based on the photometric redshifts and those based on the spectroscopic redshifts complement each other. % \u0095\u00aa\u008c\u00f5\u0083T\u0083\u0093\u0083v\u0083\u008b\u0082\u00c5\u008dL\u0082\u00a2\u0094g\u0092\u00b7\u0094\u00cd\u0088\u00cd\u0082\u00cc\u0083X\u0083y\u0083N\u0083g\u0083\u008b\u0082\u00f0\u0091\u00e5\u0097\u00ca\u0082\u00cc\u008b\u00e2\u0089\u00cd\u0082\u00c9\u0082\u00c2\u0082\u00a2\u0082\u00c4\u008b\u0081\u0082\u00df\u0082\u00e9\u0082\u00cc\u0082\u00cd\u0093\u00ef\u0082\u00b5\u0082\u00a2\u0082\u00aaphotometric redshift method\u0082\u00f0\u0097p\u0082\u00a2\u0082\u00ea\u0082\u00ce\u0082\u00bb\u0082\u00ea\u0082\u00aa\u0089\u00c2\u0094\\\u0082\u00c9\u0082\u00c8\u0082\u00e9\u0081B If the SED of each sample galaxy in wide wavelength is given, the rest-frame luminosity of the galaxy can be estimated. This method has the advantage to deriving the rest-frame luminosity at any redshift and at any wavelength, though the estimated luminosity inherits an uncertainty in determining the photometric redshift.  %3 \u0096{\u0098_\u0095\u00aa\u0082\u00c5\u0082\u00cd\u008c\u00e3\u008e\u00d2\u0082\u00cc\u0095\u00fb\u0096@\u0082\u00f0\u008e\u00e6\u0082\u00e8\u0081Arest-frame\u0082\u00c5\u0082\u00cc\u008b\u00e2\u0089\u00cd\u0082\u00ccLF\u0081Acolor\u0082\u00cc\u0090i\u0089\u00bb\u0082\u00f0\u0092\u00c7\u0082\u00a4\u0081B  In this study, we have constructed a very deep $K^\\prime$ band-selected sample of galaxies. We then adopt the second approach described above with photometric redshifts derived from multicolor $BVRIz^\\prime JK^\\prime$ deep images and investigate the evolution of the LFs in the three ($K^\\prime$, $B$, and UV) rest-frame bands out to $z=3.5$ where the $K^\\prime$-band still traces the rest wavelength longer than an age-sensitive spectral break at $4000$\\AA. %6 K\u0081|Band\u0082\u00cc\u0097L\u0097\u0098\u0090\u00ab\u0082\u00cd\u0082\u00b1\u0082\u00ea\u0082\u00b1\u0082\u00ea\u0082\u00c5\u0082\u00a0\u0082\u00e9\u0081B NIR band-selected samples may provide significant advantages over optical band-selected samples in studying galaxy evolution due to smaller extinction by dust and less type-dependent $k-$corrections \\citep{man01} at these wavelengths. Moreover, NIR selection provides samples that are not biased towards star-forming galaxies and allows estimation of the mass of galaxies over a wide range of redshift \\citep{sar01}. NIR photometry in general improves the estimate of photometric redshift (e.g., \\citealp{bol00}). % or the near-infrared light in galaxies is produced by giants drawn from the population of old evolved stars which dominate the stellar mass. %7 SDF\u0082\u00cc\u0082\u00b1\u0082\u00ea\u0082\u00dc\u0082\u00c5\u0082\u00cc\u008a\u00cf\u0091\u00aa\u0082\u00cd\u0082\u00b1\u0082\u00ea\u0082\u00b1\u0082\u00ea\u0082\u00c5\u0082\u00a0\u0082\u00e9\u0081B  The Subaru Deep Field (SDF) was imaged quite deeply at near-infrared with wide field near infrared camera CISCO and yielded some of the faintest galaxies ever observed, down to magnitudes of $J=25.5$ and $K^\\prime=24.5$ \\citep[Paper I]{mai01}, which is much deeper than most of previous $K^\\prime$-band surveys (e.g., \\citealp{cow96,bar99,fon00,rud01}). % and as deep as but much wider than \\citet{hog97}. The field was well selected, taking care to avoid large Galactic extinction and nearby clusters of galaxies and to get smaller airmass than the HDF at Mauna Kea (Paper I). The following optical imaging enabled us to estimate the photometric redshifts of these faint objects and offered the discovery of faint high z object candidates. In the SDF, we have already evaluated the number counts in $J$ and $K^\\prime$ bands, the near-infrared color distribution, and contribution to extragalactic background light (Paper I, see also \\citealp{tot01a}). \\citet{tot01b} have investigated a detailed comparison of the number counts, colors, and size distribution for the SDF galaxies with theoretical galaxy formation models. \\citet{nag03} have also analyzed the SDF galaxies by using a semianalytic model of galaxy formation based on the CDM model. In this paper, we will compare the redshift distribution with their models. %8 \u0082\u00b1\u0082\u00cc\u0098_\u0095\u00b6\u0082\u00cc\u0091g\u0082\u00dd\u0097\u00a7\u0082\u00c4\u0082\u00cd\u0082\u00b1\u0082\u00cc\u0092\u00ca\u0082\u00e8\u0081B   This paper will be organized as follows. We present the process to construct the $K^\\prime$-limited sample of SDF galaxies and estimating photometric redshifts as well as their spectroscopic calibration in \\S 2. In \\S 3, we show the redshift distribution of our sample and compare it with galaxy formation models. We discuss the evolution of LFs on rest $K^\\prime$, $B$, and UV(2000\\AA) bands in \\S 4 and the evolution of rest UV$-K^\\prime$ color in \\S 5. We will also describe in \\S 6 the evaluation of star formation rate density (SFRD) derived from our LF estimate. We give a summary in \\S 7. Throughout this paper we assume a flat, matter dominated universe ($\\Omega_m=1$) with $H_0=75$kms$^{-1}$Mpc$^{-1}$ to compare with other major previous observations. ",
        "conclusions": "We have constructed a very deep $K^\\prime$-selected multicolor $BVRIz^\\prime JK^\\prime$ sample of field galaxies on the Subaru Deep Field. Based on this multicolor data, a photometric redshift for each sample galaxy was estimated. Our principal conclusions are the following:  \\begin{enumerate}  \\item The overall N(z) distribution for the $K^\\prime \\leq 21$ sample is consistent with previous observations, and we for the first time derived the N(z) down to $K^\\prime=24$. Taking account of dust extinction and selection effects of the sample, the observed N(z) distribution at the most secure magnitude bin, $22<K^\\prime \\leq 23$, is well described with the PLE model, while a recent version of the hierarchical galaxy formation model shows an apparent deficiency of galaxies, especially at $z>2$. The PLE model with realitic dust evaluation can account for the abundance of the high z ($z>2$) galaxies. In the faintest bin of N(z), $23<K^\\prime \\leq 24$, both of the model predictions are found to be inconsistent with the observations. The discrepancy in the faintest bin is presumably caused by the photometric errors that give an incorrect solution of photometric redshifts.  \\item The rest $K^\\prime$, $B$, and UV($2000$\\AA) band luminosity functions and their evolution have been investigated. The rest $K^\\prime$-band luminosity function shows almost no evolution up to $z=3$. The rest $B$ luminosity function shows mild brightning with epoch up to $z=2.5$, though its brightening degree is within the $M^*$ error caused by photometric and redshift errors. The rest $2000$\\AA~LFs also show a luminosity evolution that is more significant than in the rest $B$ LF and continues up to $z=3.5$. The $M_{2000}^*$ value is getting luminous from $z=0.6$ to $z=3.5$ over $1$ mag, which is more significant than $M_{2000}^*$ errors.  \\item The evolutionary trend of rest luminosity functions appears to correspond to the bluing trend of the rest $UV-K^\\prime$ color with increasing redshift. We also found that more massive galaxies are redder at every epoch.  \\item The findings about the evolution of rest luminosity functions are qualitatively in favor of the PLE model. The star formation activity in a galaxy changes the rest blue luminosity, while the amount of old stellar population does not change. The resulting rest $K^\\prime$-band luminosity function showing almost no evolution up to $z=2.5$ can also be consistent with the hierarchical galaxy formation model if $M/L_K$ decreases with redshift; that is, the number density of $K^\\prime$-band luminosity-selected galaxies does not significantly change with redshift, while the number density of stellar mass-selected galaxies decreases with look-back time.  %However, the result can also be consistent with the hierarchical galaxy formation model if $M/L_K$ decreases with redshift, which means that the number density of $K^\\prime$-band luminosity selected galaxies does not change significantly with redshift while the number density of stellar-mass selected galaxies decreases with look-back time.  \\item We have evaluated the contribution to the cosmic star formation rate density of our $K^\\prime$-selected galaxies. The star formation rate density derived from integration of our rest UV luminosity function is about $42\\%$ of that derived from Lyman break galaxies at $z=3$. The result suggests that a large fraction of global star formation rate density at $z>2$ could represent the contribution of faint blue galaxy population, which has not yet explicitly been identified in a high z universe.  \\end{enumerate}  Though the present results rely heavily on an availability of photometric redshifts, we are confident that we have successfully derived a coherent picture of evolution of $K^\\prime$, $B$, and UV rest-frame luminosity functions over the range $0.6<z \\leq 3.5$ to the faint end below the limit of the today's spectroscopic studies. Our result raises the question of how such massive galaxies have formed at high redshift with efficient star formation, which is apparently consistent with the PLE picture in the $K$ band, within the framework of the structure formation in the CDM universe. Of course, deeper NIR imaging surveys in future will allow us to derive more secure results for N(z) at the faintest magnitude and on the faint end of the rest LFs, in which we have obtained robust results in this study. Deep observation at longer wavelength with $SIRTF$ and $ASTRO$-$F$ will be also required to constraint the SEDs of galaxies at $z>3$. The technique that we have adopted in this study will be applied to future wide-field $K$-band surveys to determine the rest luminosity at any wavelength at any redshift down to the faintest magnitude. On the other hand, ongoing large multiobject spectroscopic surveys are of course extremely important to reveal the galaxy evolution more reliably."
    },
    "0209/astro-ph0209203_arXiv.txt": {
        "abstract": "Abstract:  We discuss the evolution of binaries with moderately high masses ($\\ts 10 - 30\\Msun$), and with periods of $\\ts 3 - 300\\thin$d, corresponding mostly to early Case B. These are usually thought to evolve either by reasonably conservative Roche-lobe overflow, if the initial mass ratio is fairly mild, or else by highly non-conservative common-envelope evolution, with spiral-in to short periods ($\\ts$ hours, typically), if the initial mass ratio is rather extreme. We discuss here a handful of binaries from part of this period range $(\\ts 50 - 250\\thin$d), which appear to have followed a different path: we argue that they must have lost a large proportion of initial mass ($\\ts 70 - 80\\%$), but without shortening their periods at all. We suggest that their behaviour may be due to the fact that stars of such masses, when evolved also to rather large radii, are not far from the Humphreys-Davidson limit where single stars lose their envelopes spontaneously in P Cygni winds, and so have envelopes which are only lightly bound to the core. These envelopes therefore may be relatively easily dissipated by the perturbing effect of a companion. In addition, some or all of the stars considered here may have been close to the Cepheid instability strip when they filled their Roche lobes. One or other, or both, of high luminosity and Cepheid instability, in combination with an appropriately close binary companion, may be implicated. ",
        "introduction": "\\par It is well known that stars of high mass, $\\tgs 30\\Msun$, appear to be strongly affected by mass loss at some stage in their evolution across the Hertzsprung-Russell diagram (HRD). Humphreys \\& Davidson (1979) found that stars are absent above a line (the Humphreys-Davidson limit) in the HRD that slopes down gently from left to right. Theoretical evolutionary tracks for stars $\\tgs 30\\Msun$ should cross this line, but apparently real stars do not. Furthermore stars near the Humphreys-Davidson limit are often highly variable, and have indications (P Cygni line profiles, variable light-curves) of fast, copious and erratic stellar winds. Thus it is likely that as a massive star approaches the Humphreys-Davidson limit it loses considerable mass and instead of evolving further to the red is stripped almost to its helium-burning core. It then evolves to a small hot remnant, a Wolf-Rayet (WR) star. \\par Evolutionary tracks of stars of lower mass do not intersect the Humphreys-Davidson limit. This does not mean that they suffer no mass loss at all, but it may mean that they do not suffer much mass loss until after they have crossed the HRD and become red supergiants. One can hope for guidance here from observed masses of late supergiants in binaries; but there are not many of these, and the difficulty of measuring the small radial-velocity amplitudes of the hot, and therefore typically broad-lined, component makes some determinations quite uncertain. We discuss some possibly relevant systems below. \\par Conservative early Case B evolution, starting from a mass ratio which is not strongly different from unity, is expected to pass through (i) Roche-lobe overflow (RLOF), then (ii) a detached phase, with the loser having become small, and (iii) a late stage of {\\it reversed} Roche-lobe overflow (RLOF), as the gainer's evolution proceeds. We distinguish two main subtypes of Case B, depending on whether the loser is able to complete its evolution (presumably to a supernova explosion, followed by a neutron star) before the gainer evolves to RLOF, or on the other hand is still in a helium-burning stage when this happens. Following the discussion of Nelson \\& Eggleton (2001), hereinafter Paper I, on Case A evolution, we call these subtypes Case BN (`no overtaking') and Case BL (`late overtaking'). In Section 2 we discuss three observed systems near the Case B/A borderline, that seem to be reasonable examples of the conservative model during the first RLOF phase, and three more systems that appear to be reasonable examples of the detached phase that follows. In Section 2 we also emphasise that there at least three other variants of Case B, which we call Cases BB, BR and BD, by analogy with three subtypes of Case A in Paper I. In Section 3 we discuss four observed binary systems that we believe are highly evolved and which do not fit well the usual assumptions of Case B. Four is not a large number, but all four appear to us to show strong evidence that a considerable amount of mass has been lost from each system, and yet the orbital periods ($\\ts 50 - 225\\thin$d) are by no means as small as one would expect if the mass-loss process was driven primarily by the release of orbital energy during a common-envelope phase of evolution. We call this unexpected variant Case BU. In Section 4 we discuss a possible physical mechanism, and in Section 5 we discuss some implications for the evolution of binary stars. ",
        "conclusions": "\\par We have seen that some massive and fairly close binaries, the first group of Table 1, can be accounted for reasonably well by conservative RLOF in Case A or very early Case B, and that some other, somewhat wider, binaries -- the second group -- can be seen as later stages of similar systems. But four systems at least (the third group in Table 1), which were probably substantially wider than the first group but comparable to the second group when RLOF began, are hard to account for with the same model. Note that we are not distinguishing here between the first episode of RLOF, in the forward direction, and the second, in the reverse direction: in our preferred model of V379 Cep the first occurred at a short period, and was conservative, but the second and non-conservative one was at a much longer period. In our preferred models of the other three difficult systems the {\\it first} RLOF occurred at a fairly long period. A highly non-conservative model is required for these, but one which, unlike common-envelope evolution, does {\\it not} shrink the orbital period down to a day or less. Although our difficult group only come from the mass range $\\ts 10 - 20\\Msun$, we expect that it will extend to $\\ts 30\\Msun$. \\par What we believe distinguishes the difficult group from the first group and also the last group in Table 1 is their periods, with the implication that if a massive star which is about to fill its Roche lobe has evolved far into the Hertzsprung gap, but not so far as to reach the red (super)giant branch, then it is prone to lose almost its whole envelope to infinity rather than to its companion. This might be because the Humphreys-Davidson limit for single stars is in effect lower for binary stars with close companions -- close, that is, once the star has expanded to supergiant radius -- or it might be that pulsational instability of the Cepheid variety is much more vigorous in a star with a close companion (in the same sense). It might even be a combination of the two. \\par Possibly the mass ratio immediately before the ejection episode is also important. However, we seem to have a substantial spread. For 0045-7319 the mass ratio of 1.2 that we hypothesise is rather mild, while for V2174 Cyg, $\\upsilon$ Sgr and V379 Cep the values are 2, 3.3 and 11. Four cases hardly make a good statistical base, but there is no evidence here that mass ratio is an important discriminant between our envelope ejection process on the one hand and conservative RLOF or common-envelope evolution on the other. \\par Case B is traditionally divided into two major subtypes, early or late, where star 1 has a radiative or a convective envelope respectively at the onset of RLOF. We conclude that several more subtypes are necessary. There appears to be a significant difference between `very early Case~B' and `moderately early Case~B', with a division that may correspond to whether a star is on the left and below, or on the right and above, some boundary that goes through the middle of the Hertzsprung gap. The first group may have fairly conservative early Case B, and the second a version that is highly non-conservative of mass but relatively conservative of angular momentum. We cannot be very precise about this boundary, but stars which have spectral types $\\ts$AI/FI/GI, and have masses of $\\ts 10 - 30\\Msun$, as they approach their Roche lobes appear to be in the second category. Within the first category, we should remember that even conservative RLOF has several subtypes, which we have identified as Cases BD, BR, BL, BB and BN; the second category we call BU (`B -- unusual'), for the present. Case BD, which has a large mass ratio, probably leads to common-envelope evolution, and quite possibly a complete merger rather than a short-period remnant. The fate of BR systems is very unclear, but we suspect they also end up as merged single stars, by a slower and somewhat more conservative process. Cases BL -- BN lead to wider detached binaries, and then to {\\it reversed} RLOF, which we conclude will usually be Case BU (rev). \\par Case C is presumably similar. In the upper part of our range of masses stars may ignite helium well before they have crossed the Hertzsprung gap, so that there will be early Case C as well as late Case C systems. There will however probably not be `very early Case C' systems, since helium does not usually ignite before the star is about half-way across the Hertzsprung gap. We believe that what we have concluded about `moderately early Case B' will apply to `moderately early Case C', and in effect to most early Case C systems. \\par We may have to go to {\\it late} Case B (or late Case C) for the kind of systems which undergo common-envelope evolution with spiral-in to short but non-zero periods, producing low-mass X-ray binaries at high (star 1) mass and cataclysmic variables at low mass. But such evolution is not guaranteed; we suspect that it depends strongly on mass ratio, unlike case BU. We hope to show in a future paper that short-period highly-evolved systems require not just a convective envelope in the loser, but a mass ratio $q\\tgs 4$ as well. Systems with lower mass ratios seem, like Case BU, to suffer much mass loss with little orbital shrinkage. \\par If we are right about the existence of Case BU, then there appear to be implications for various evolutionary scenarios. For example low-mass X-ray binaries are often conjectured to arise from common-envelope evolution with spiral-in. We start with a binary containing an OB star and a GK dwarf, and require that when the OB star has developed  a helium-burning core and a large envelope the GK dwarf spirals into the envelope and ejects it, reducing the period from hundreds of days to less than a day. The helium-burning core in this close binary then evolves to a supernova explosion. But if the initial period is such that star 1 reaches its Roche lobe somewhat before it reaches the red (super)giant branch, we would claim on the evidence presented above that the period does {\\it not} shrink drastically. We may need substantially wider initial binaries (late Case B/C), where star 1 has room to develop a deep convective envelope, to do this. \\par The DJEHUTY project at Lawrence Livermore National Laboratory is an attempt to model stars, including both hydrodynamic and radiation-transport processes, with a fully 3-D grid. We believe this will be the best way to investigate the interaction of a lobe-filling A/F/G supergiant with its companion. Such modelling would allow the possible effect of Cepheid instability to be included. \\par This work was undertaken as part of the DJEHUTY project at LLNL.  Work performed at LLNL is supported by the DOE under contract W7405-ENG-48."
    },
    "0209/astro-ph0209035_arXiv.txt": {
        "abstract": "Considerable progress has been made over the last decade in the study of the evolutionary trends of the population of galaxy clusters in the Universe. In this review we focus on observations in the X-ray band. X-ray surveys with the {\\it ROSAT} satellite, supplemented by follow-up studies with {\\it ASCA} and {\\it Beppo--SAX}, have allowed an assessment of the evolution of the space density of clusters out to $z\\approx 1$, and the evolution of the physical properties of the intra-cluster medium out to $z\\approx 0.5$. With the advent of {\\it Chandra} and {\\it Newton-XMM}, and their unprecedented sensitivity and angular resolution, these studies have been extended beyond redshift unity and have revealed the complexity of the thermodynamical structure of clusters.  The properties of the intra-cluster gas are significantly affected by non-gravitational processes including star formation and Active Galactic Nucleus (AGN) activity.  Convincing evidence has emerged for modest evolution of both the bulk of the X-ray cluster population and their thermodynamical properties since redshift unity. Such an observational scenario is consistent with hierarchical models of structure formation in a flat low density universe with $\\Omega_m\\simeq 0.3$ and $\\sigma_8\\simeq 0.7-0.8$ for the normalization of the power spectrum.  Basic methodologies for construction of X-ray--selected cluster samples are reviewed and implications of cluster evolution for cosmological models are discussed. ",
        "introduction": "Galaxy clusters arise from the gravitational collapse of rare high peaks of primordial density perturbations in the hierarchical clustering scenario for the formation of cosmic structures (e.g. Peebles 1993, Coles \\& Lucchin 1995, Peacock 1999).  They probe the high--density tail of the cosmic density field and their number density is highly sensitive to specific cosmological scenarios (e.g. Press \\& Schechter 1974, Kofman et al. 1993, Bahcall \\& Cen 1993, White et al. 1993a). The space density of clusters in the local universe has been used to measure the amplitude of density perturbations on $\\sim\\!  10$ Mpc scales. Its evolution, which is driven by the growth rate of density fluctuations, essentially depends on the value of the matter density parameter $\\Omega_m$\\footnote{ The matter-density parameter is defined as $\\Omega_m=\\bar\\rho/\\rho_c$, where $\\bar \\rho$ is the cosmic mean matter density; $\\rho_c=1.88\\,10^{-29}h^2$ g cm$^{-3}$ is the critical density; $h$ and $h_{50}$ denote the Hubble constant $H_0$ respectively in units of 100 and 50 km s$^{-1}$ Mpc$^{-1}$. $\\Omega_\\Lambda$ is referred to as the contribution to the total mass-energy density of the Universe associated with the cosmological constant $\\Lambda$.}  (e.g. Oukbir \\& Blanchard 1992, Eke et al. 1998, Bahcall et al. 1999). Figure~\\ref{fi:hubblevol} shows how structure formation proceeds and the cluster population evolves in two cosmological models, characterized by different values of $\\Omega_m$. High and low density universes show largely different evolutionary patterns, which demonstrate how the space density of distant clusters can be used as a powerful cosmological diagnostic. What cosmological models actually predict is the number density of clusters of a given mass at varying redshifts. The cluster mass, however, is never a directly observable quantity, although several methods exist to estimate it from observations.  Determining the evolution of the space density of clusters requires counting the number of clusters of a given mass per unit volume at different redshifts. Therefore, three essential tools are required for its application as a cosmological test: {\\it i)} an efficient method to find clusters over a wide redshift range; {\\it ii)} an observable estimator of the cluster mass and {\\it iii)} a method to compute the selection function or equivalently the survey volume within which clusters are found.  Clusters form {\\it via} the collapse of cosmic matter over a region of several megaparsecs. Cosmic baryons, which represent approximately 10--15\\% of the mass content of the Universe, follow the dynamically dominant dark matter during the collapse. As a result of adiabatic compression and of shocks generated by supersonic motions during shell crossing and virialization, a thin hot gas permeating the cluster gravitational potential well is formed. For a typical cluster mass of $10^{14}$--$10^{15}M_\\odot$ this gas reaches temperatures of several $10^7$ K, becomes fully ionized and, therefore, emits via thermal bremsstrahlung in the X-ray band.  Observations of clusters in the X-ray band provide an efficient and physically motivated method of identification, which fulfills the three requirements above. The X-ray luminosity, which uniquely specifies the cluster selection, is also a good probe of the depth of the cluster gravitational potential. For these reasons most of the cosmological studies based on clusters have used X-ray selected samples. X-ray studies of galaxy clusters provide: (1) an efficient way of mapping the overall structure and evolution of the Universe and (2) an invaluable means of understanding their internal structure and the overall history of cosmic baryons.  X-ray cluster studies made substantial progress at the beginning of the 90s with the advent of new X-ray missions.  Firstly, the all--sky survey and the deep pointed observations conducted by the \\rosat satellite have been a goldmine for the discovery of hundreds of new clusters in the nearby and distant Universe. Follow-up studies with {\\it ASCA} and {\\it Beppo--SAX} satellites revealed hints of the complex physics governing the intra--cluster gas. In addition to gas heating associated with gravitational processes, star formation processes and energy feedback from supernovae and galactic nuclear activity are now understood to play an important role in determining the thermal history of the intra--cluster medium (ICM), its X-ray properties and its chemical composition. Studies utilizing the current new generation of X-ray satellites, {\\it Chandra} and {\\it Newton-XMM}, are radically changing our X-ray view of clusters. The large collecting area of {\\it Newton--XMM}, combined with the superb angular resolution of {\\it Chandra}, have started to unveil the interplay between the complex physics of the hot ICM and detailed processes of star formation associated with cool baryons.  The aim of this article is to provide an up-to-date review on the methodology used to construct X-ray selected cluster samples and to investigate their evolutionary properties. We emphasize the evolution of the space density of clusters and their physical parameters. Additional reviews on galaxy clusters include: Forman \\& Jones (1982) and Sarazin (1988) for historical reviews on $X$-ray properties of galaxy clusters; Bahcall (1988) and Borgani \\& Guzzo (2001) for large--scale structure studies of galaxy clusters; Fabian (1994) for the physics of cooling flows in clusters; Mulchaey (2000) for the $X$-ray properties of galaxy groups; Birkinshaw (1999) and Carlstrom et al. (2001) for cluster studies with the Sunyaev--Zeldovich effect; Mellier (1999) for studies of the mass distribution of clusters via gravitational lensing and van Dokkum \\& Franx (2001) for the study of galaxy populations in clusters.   \\begin{figure} \\centerline{ \\psfig{figure=Figs_astroph/fig_nat.ps,width=5in} } \\caption{The evolution of the cluster population from N--body simulations in two different cosmologies (from Borgani \\& Guzzo 2001). Left panels describe a flat, low--density model with $\\Omega_m=0.3$ and $\\Omega_\\Lambda=0.7$ (L03); right panels are for an Einstein--de-Sitter model (EdS) with $\\Omega_m=1$. Superimposed on the dark matter distribution, the yellow circles mark the positions of galaxy clusters with virial temperature $T>3$ keV, the size of the circles is proportional to temperature. Model parameters have been chosen to yield a comparable space density of nearby clusters. Each snapshot is $250h^{-1}$ Mpc across and $75h^{-1}$ Mpc thick (comoving with the cosmic expansion). } \\label{fi:hubblevol} \\end{figure} ",
        "conclusions": ""
    },
    "0209/astro-ph0209345_arXiv.txt": {
        "abstract": "We present photometric and spectroscopic observations of the optical companion to the millisecond radio pulsar PSR J1740-5340 in the globular cluster NGC 6397. An analysis of the photometric variability in the $B$-, $V$-, and $I$-bands indicates an inclination of the system of 43.9 $\\pm$ 2.1 degrees if the optical companion fills its Roche lobe (a semi-detached configuration). The spectroscopic data show a radial velocity variation with a semi-amplitude of $K = 137.2 \\pm 2.4$ km/sec, and a system velocity $\\gamma$ = 17.6 $\\pm$ 1.5 km/sec, consistent with cluster membership.  We use these results to derive a mass of the optical companion of $M_{1}=0.296 \\pm  0.034$~$M_{\\sun}$ and $M_{2}=1.53 \\pm  0.19$~$M_{\\sun}$ for the pulsar. There is evidence for secular change of the amplitude of the optical light curve of the variable measured over seven years. The change does not have interpretation and its presence complicates reliable determination of the  absolute parameters of the binary. ",
        "introduction": "The millisecond radio pulsar  PSR J1740-5340 was discovered in the field of the globular cluster NGC~6397  by D'Amico et al. (2001a) in the course of a survey conducted with the Parkes radio telescope.  Follow up pulse timing observations (D'Amico et al. 2001b) led to a determination of several parameters of the system, including orbital period, projected semi-major axis and mass function.  Ferraro et al. (2001) identified the optical companion of the pulsar with a variable object detected earlier by Taylor et al. (2001).  Photometry presented by both groups shows that the optical companion to the MSP is a relatively bright star, $V_{max}\\approx 16.7$, located slightly to red of the turn-off region on the cluster color-magnitude diagram.  Remarkably, despite being located only 26$\\arcsec$ from the cluster center, the variable is a relatively isolated star (see Fig. 2 in Ferraro et al. 2001), permitting optical observations with ground-based telescopes. In this paper we present the results of photometric and spectroscopic observations obtained to determine the  masses for both components of the binary.  The evolutionary status of PSR J1740-5340 has been discussed by Burderi et al. (2002) and Ergma \\& Sarna (2002). Both groups present some detailed scenarios which attempt to  explain the current status of the binary and the observed position of its optical component on the cluster color-magnitude diagram. The system has been detected in the X-ray domain with the $Chandra$ observatory by Grindlay et al. (2001, 2002). ",
        "conclusions": "An additional complication to the  interpretation of the optical observations of the optical companion is the fact that its light curve clearly evolves on a time scale of a few years.  The range of variability in the $V$ band has changed from $\\Delta V\\approx 0.23$ mag in 1995 to $\\Delta V\\approx 0.15$ in 2002.  We discuss briefly 3 possible interpretations of these changes.  1. {\\it Secular change of the inclination of the orbit of binary}: Assuming a  semi-detached configuration the observed change of $\\Delta V$  requires a change of orbital inclination of about  8 degrees (from $i=52$~deg in 1995 to $i=44$~deg in 2002).  There are a few eclipsing binaries with observed variation in the inclination of their orbital plane due to an interaction with a third body  (Drechsel et al. 1994; Milone et al. 2000).  However the rate of variation is at a level of a few tenths of a degree per year at best.  Pulsar timing observations covering a 6 month interval (D'Amico et al. 2001b) show no evidence for a dynamical interaction of PSR J1740-5340 with a hypothetical \"third body\".  Further radio timing  observations of the pulsar should provide very strong limits on any changes of orientation of the orbital plane of the binary.  2. {\\it Variable \"third light\" contributions to the light curve}: Luminous streams of gas around pulsar or variable heating of the optical component of the system could lead to changes in the light curve. However in this  case  we should observe not only changes in the  amplitude of the light curves but also changes in the maximum observed light. One may estimate how much of 3rd light is needed to diminish $\\Delta V$ from $\\approx 0.23$ (1995 season) to $\\approx 0.15$ (2002 season), where $\\Delta V$ is magnitude difference for phases 0.25 and 0.50. Denoting the flux level at phase 0.25 in 1995 season by $l_{1V}$ we obtain $l_{3V}=0.48 \\, l_{1V}$. That in turn implies that at quadrature the system would be brighter by 0.42 mag in 2002 season as compared with 1995 season.  Our data show no indications for any change of light level at quadratures between the 1995 and 2002 seasons, which would exceed 0.02--0.03 mag.}  3. {\\it Intrinsic variability of the optical companion}: It is possible that the observed variability is intrinsic to the optical companion, perhaps in the form of star-spots. The companion has a high rotational velocity if it is tidally locked ($v_{rot} \\simeq 50$ km/sec) and such a high rotational velocity in a fully convective atmosphere normally leads to the formation of star-spots. We have used WD code to perform some light curve simulations for a model including one dark spot. The starting point was the light curve solution obtained for the semi-detached configuration and the 2002 data. It turns out that one may indeed increase the depth of the minimum observed at phase 0.5 to 0.23~mag, as seen in 1995 season, by putting a dark spot in a region around inner Lagrangian point $L1$. Specific parameters of such a  spot are $\\Delta T=1000$K and radius equal to 20~degrees (as seen from the center of the star). The corresponding change of $V$ magnitude at quadratures would then amount to only 0.02~mag. It is worth to note in that context that light curves presented by Ferraro et al. (2001; Fig. 4) show a clear asymmetry at minimum light which occurs at phase 0.5. Spot hypothesis offers a possible way to explain such a distortion.  We do not have a definite  explanation for the observed variations in the amplitude of the light curves. The hypothesis invoking stellar spots seems to be a viable option for a moment. Clearly further monitoring of the system would be desirable."
    },
    "0209/astro-ph0209490_arXiv.txt": {
        "abstract": "Faint submillimeter sources detected with the Submillimeter Common-User Bolometer Array on the James Clerk Maxwell Telescope have faced an identification problem due to the telescope's broad beam profile. Here we propose a new method to identify such submillimeter sources with a mid-infrared image having a finer point spread function. The Infrared Space Observatory has provided a very deep 6.7\\,$\\mu$m image of the Hawaii Deep Field SSA13. All three faint 850\\,$\\mu$m sources in this field have their 6.7\\,$\\mu$m counterparts. They have been identified with interacting galaxy pairs in optical images. These pairs are also detected in the radio. Two of them are optically faint and very red ($I>24$, $I-K>4$), one of which has a hard X-ray detection with the $Chandra$ satellite. As these observing properties are similar to those of local ultraluminous infrared galaxies, their photometric redshifts are derived based on submillimeter to mid-infrared flux ratios assuming a spectral energy distribution (SED) of Arp220. Other photometric redshifts are obtained via $\\chi^2$ minimization between the available photometry data and template SEDs. Both estimates are in the range $z=1$--2, in good agreement with a spectroscopic redshift and a millimetric one. The reconstructed Arp220 SEDs with these redshift estimates are consistent with all the photometry data except $Chandra$'s hard X-ray detection. The sources would be a few times more luminous than Arp220. With an assumption that AGN contributions are negligible, it appears that extremely high star formation rates are occurring in galaxies at high redshifts with massive stellar contents already in place. ",
        "introduction": "The installation of the Submillimeter Common-User Bolometer Array (SCUBA) on the 15\\,m James Clerk Maxwell Telescope has brought us the discovery of a new population of faint submillimeter sources \\citep{SIB97,BCS+98,HSD+98,ELG+99}. Meanwhile, the far-infrared background (FIRB) was detected with the Far Infrared Absolute Spectrometer (FIRAS) and the Diffuse Infrared Background Experiment (DIRBE) on the COBE satellite (\\citealp{PAB+96}; \\citealp*{SFD98}; \\citealp{HAK+98,FDM+98}). The FIRB is comparable to or larger than that in the optical wavelengths, and much of it could be explained by integrating individual submillimeter sources down to the faintest limit \\citep{BKI+99}. Thus, it is indispensable to investigate the faint submillimeter population in order to understand overall history of energy production in the Universe.  Although SCUBA has unveiled the faint submillimeter population, its beam is too broad (15{\\arcsec} FWHM at 850\\,$\\mu$m) to pinpoint their optical counterparts \\citep*[e.g.][]{BCR00}. The high density of the submillimeter sources also raises a problem of source confusion \\citep{ELW+00,H01}. Up until now, methods to bypass the broad submillimeter beam have included centimetric radio interferometry with VLA \\citep{ISB+98,R99,SIK+99,BCR00} and millimetric interferometry with IRAM or OVRO \\citep{DNG+99,GLS+00,BCM+00,FSIS00}. Most recently, an attempt to observe at a longer wavelength has started at the IRAM 30\\,m telescope \\citep{BCM+00}. The larger dish provides a slightly smaller beam (10{\\arcsec} at 1.3\\,mm).  Here we investigate the identification of SCUBA sources with a mid-infrared deep image having a relatively small beam (7\\farcs2 at 6.7\\,$\\mu$m). Faint submillimeter sources, presumably dusty systems at high redshifts, could be bright in the mid-infrared regardless of their energy sources. If the submillimeter emission was produced by dust heated by star forming activity, there should be stellar systems following the stellar initial mass function. The stellar systems could be luminous at the rest-frame near-infrared ($>1\\,\\mu$m) even in the presence of dust. In the case that the source was powered by an active galactic nucleus (AGN), very hot dust surrounding the AGN could emit its reprocessed energy at $>2\\,\\mu$m. For both cases, the emitted light could be received at the mid-infrared, even though the sources were at high redshifts. ",
        "conclusions": "All three of the faint 850\\,$\\mu$m sources (2--4\\,mJy) in the SCUBA SSA13 deep field have been found to have ISOCAM 6.7\\,$\\mu$m counterparts (10--30\\,$\\mu$Jy). Utilizing the smaller beam size at 6.7\\,$\\mu$m, we find that all the three sources are coincident with interacting galaxy pairs in the $HST$ $I$ band images. They all have VLA 1.4\\,GHz counterparts, and only one of them is detected with $Chandra$ at the 2--10\\,keV band. Based on the properties similar to local ULIGs, we used submillimeter to mid-infrared flux ratios to estimate photometric redshifts. We derived other photometric redshifts utilizing the X-ray to radio photometry data. Both estimates are consistent with a spectroscopic redshift and a millimetric one. An Arp220 SED at $z=1$--2 provides a good fit to the available data. With an assumption that AGN contributions are negligible, the implied SFRs are in the range (1--2) $\\times$ $10^{3}\\,M_\\sun\\,\\mathrm{yr}^{-1}$ and the mid-infrared emission requires stellar masses in the range (3--7) $\\times$ $10^{11}\\,M_\\sun$."
    },
    "0209/astro-ph0209173_arXiv.txt": {
        "abstract": "High redshift radio galaxies are great cosmological tools for pinpointing the most massive objects in the early Universe: massive forming galaxies, active super--massive black holes and proto--clusters. We report on deep narrow--band imaging and spectroscopic observations of several $z > 2$ radio galaxy fields to investigate the nature of giant Ly--$\\alpha$ nebulae centered on the galaxies and to search for over--dense regions around them. We discuss the possible implications for our understanding of the formation and evolution of massive galaxies and galaxy clusters. ",
        "introduction": "\\label{sect:why}  %  High redshift radio galaxies (\\hzrgs; $z > 2$) are great beacons for pinpointing the most massive objects in the early universe, whether these are galaxies, black holes or even clusters of galaxies.  At {\\it low} redshifts powerful, non--thermal radio sources are uniqely associated with massive ellipticals. Their twin--jet, double--lobe morphologies and large luminosities suggested already early on that such galaxies must also have spinning, super--massive black holes (SMBH's) in their centers\\cite{Rees78,Blandford82}.  We now know that the masses of the stellar bulges of galaxies and their central black holes are correlated\\cite{Magorrian98,Gebhardt00,Ferrarese00}, suggesting a causal connection. If radio sources are powered by SMBH's then it is no longer a surprise that their parent galaxies occupy the upper end of the galaxy mass function.  There is excellent evidence that radio galaxies are also the most massive systems at {\\it high} redshifts, even though their parent galaxies are very young and may still be forming.  The combined near--infrared `Hubble' $K-z$ relation for radio and field galaxies\\cite{DeBreuck02a} shows that \\hzrgs\\ are among the most luminous systems at any given epoch up to $z\\sim 5$. Between $0 < z < 2.5$ this $K - z$ diagram can be modeled using passive evolution of 5$L_\\star$ ellipticals\\cite{Jarvis01} which formed at $5 < z < 10$.  Other evidence includes the direct detection of absorption lines from massive young stars\\cite{Dey97}, large ($\\sim 50 - 70$ kpc), multi--component rest--frame UV and optical morphologies\\cite{Pentericci99}, the 'hyper' luminous rest--frame far--infrared luminosities ($L_{FIR} \\simgt 10^{13} \\Lsun$) and huge implied star formation rates\\cite{Archibald01,Reuland02} ($\\simgt 2000 \\Msunpyr$), and last but not least, the very extended ($\\sim 30 - 50$ kpc) molecular gas and dust clouds that have recently been discovered\\cite{Papadopoulos00,DeBreuck02b} around several \\hzrgs\\ showing that the star formation occurs on galaxy wide scales.  \\hzrgs\\ are also excellent tools for finding over--dense regions (`proto--clusters') at high redshift. This is because, in standard Cold Dark Matter (CDM) scenarios, galaxy formation is a highly `biased' process: the most massive galaxies, and the largest clusters of galaxies, are expected to emerge from regions with the largest over--densities\\cite{Kaiser84}. Simply put: the most massive systems (galaxies, SMBH's and galaxy clusters) hang out together, and radio sources are a great way to find them and to investigate their interrelations and evolution.  In this paper we report on deep narrow--band imaging and spectroscopic observations of several $z > 2$ radio galaxy fields using the Keck and ESO/VLT telescopes to investigate the nature of giant Ly--$\\alpha$ nebulae centered on the galaxies and to search for over--dense regions around them. We discuss the possible implications for our understanding of the formation and evolution of massive galaxies and galaxy clusters. We will adopt the cosmological parameters $\\OmM = 0.3$, $\\OmL = 0.7$, $H_{0} = 65 \\kmps\\,\\rm Mpc^{-1}$, for which the age of the Universe at $z \\sim$ 2, 3 and 4 is 3.5, 2.3 and 1.6 Gyr respectively, and the angular--to--linear transformations are 9.0, 8.3 and 7.5 \\kpc\\ arcsec$^{-1}$. ",
        "conclusions": ""
    },
    "0209/astro-ph0209459_arXiv.txt": {
        "abstract": "{In this paper, we present a  set of photometric observations in 15 colors of stars in the Landolt SA95 field with the BATC system. The wavelengths covered by the system range from 300 nm to 1000 nm. Visual magnitudes of the stars being studied in the field are from 10th to 20th mag.  The observational methodology and the data reduction procedures are described.  The relationships between the BATC intermediate-band system and the Landolt $UBVRI$ broad band system are obtained. A catalogue of the photometry has been produced which contains the SEDs of 3613 stars.  The electronic form of this catalogue can be accessed at the CDS via anonymous ftp to cdsarc.u-strasbg.fr. ",
        "introduction": "Landolt (1983) gave his measurements of $UBVRI$ magnitudes of 233 standard stars in a strip about two degrees wide centered on the celestial equator for the purpose of homogeneous flux calibration from both hemispheres. Further more, Landolt (1992) presented the $UBVRI$ photometry of another 298 standard stars around the celestial equator, within a visual magnitude range of 11.5 -- 16.0 and a $B-V$ colour range from $-0.3$ to $+2.3$. The Landolt catalogues have been widely used by observers using intermediate to large size telescopes. By using the Landolt standards catalogues, Galad\\'{\\i}-Enr\\'{\\i}quez et al. (2000) obtained $UBVRI$ photometry of 11 different fields around Landolt standards and gave 681 secondary standards with a visual magnitude range from 9.7 to 19.4 and a $B-V$ range from 1.15 to 1.97.  With the capability of imaging large fields, Beijing-Arizona-Taipei-Connecticut (BATC) survey program has observed a number of selected fields which include Landolt SA95 field centred at $\\alpha = \\rm{3^{h}54^{m}17^{s}}$, $\\delta= +00^\\circ19^{\\prime}08^{\\prime\\prime}{\\mbox{}\\hspace{-0.1cm}}$ (2000.0).  So far Landolt SA95 has been one of the fields with higher observing quality.  Thus we select it to demonstrate the photometric quality of the BATC system.  Standard BATC data reduction processes have been applied to this field. In this field, 7 stars have been measured in Landolt (1983), and 45 are listed in the catalogue of Landolt (1992). In addition, 7 stars in this field were included in the secondary standard catalogue of Galad\\'{\\i}-Enr\\'{\\i}quez et al. (2000).  The main purpose of this paper is to present the spectral energy distributions (SEDs) of the field stars in the format of 15 colours, and to show the relationships between the BATC  and the $UBVRI$ photometric systems derived by using the stars in  the catalogues of Landolt (1983), Landolt (1992), and Galad\\'{\\i}-Enr\\'{\\i}quez et al. (2000). In Sect. 2, we present the observing processes of the Landolt SA95 field.  In Sect. 3, we describe  the method of data reduction. The magnitude error of observation and data reduction are discussed in Sect. 4. The discussion of the system transformation is given in Sect. 5.  Sect. 6 gives the conclusions of this study. ",
        "conclusions": "In a total of 41 nights in the period from December 13, 1994 to December 16, 1999, we made observations of the Landolt SA95 field. A total of 189 images with acceptable quality were selected for photometric measurements. A complete SED catalogue in 15 colours in the BATC system for the stars in the field of Landolt SA95 is presented.  The wavelengths coverage with the 15 intermediate filter bands are from 300 nm to 1000 nm. Visual magnitude ranges from 10th to 20th mag. We describe the methods of observation and data reduction, and analysed the possible error of our measurement. By comparison with Landolts $UBVRI$ broad band photometric magnitudes of 48 stars, the relationships between the BATC intermediate-band system and Landolt $UBVRI$ broad band photometic system are obtained.  A catalogue is created in which a total of 3613 stars is included.  The catalogue is also published in electronic form and is available at CDS ftp site to cdsarc.u-strasbg.fr (130.79.128.5)."
    },
    "0209/astro-ph0209167_arXiv.txt": {
        "abstract": "With the advent of high-quality surveys in cosmology the full three-point correlation function will be a valuable statistic for describing structure formation models. It contains information on cosmological parameters and detailed halo properties that cannot be extracted from the two-point correlation function. We use the halo clustering model to analytically calculate the three-point correlation function (3PCF) for general cosmological fields. We present detailed results for the configuration dependence of the 3-dimensional mass and galaxy distributions and the 2-dimensional cosmic shear field.  We work in real space, where higher order correlation functions on small scales are easier to measure and interpret, but halo model calculations get rapidly intractable. Hence we develop techniques that allow for the accurate evaluation of all the contributing terms to real space correlations. We apply them to the 3PCF to show how its configuration and scale dependence changes as one transits from the nonlinear to the quasilinear regime, and how this depends on the relative contributions from the 1-, 2- and 3-halo terms.  The 3PCF violates the hierarchical ansatz in both its scale and configuration dependence. We study the behavior of the coefficient $Q$ in the expansion: $\\zeta(r_{12},r_{23},r_{31}) =Q[\\xi(r_{12}) \\xi(r_{23})+\\xi(r_{12})\\xi(r_{31})+\\xi(r_{23})\\xi(r_{31})]$ from large, quasilinear scales down to about 20 kpc. We find that the nonlinear 3PCF is sensitive to the halo profile of massive halos, especially its inner slope. We model the distribution of galaxies in halos and show that the 3PCF of red galaxies has a weaker configuration and scale dependence than the mass, while for blue galaxies it is very sensitive to the parameters of the galaxy formation model. The 3PCF from weak lensing on the other hand shows different scalings due to projection effects and a sensitivity to cosmological parameters. We discuss how our results can be applied to various analytical calculations: covariances of two-point correlation function, the pairwise peculiar velocity dispersion, higher-order shear correlations, and to extend the halo model by including the effects of halo triaxiality and substructure on statistical measures. ",
        "introduction": "Understanding the nature of the large-scale structure and the evolutionary history of the universe is the central aim of cosmology. Therefore, quantifying the observed distribution of matter rigorously is of fundamental importance, and $n$-point correlation functions have been the most widely used statistical tools for this purpose (e.g., see Peebles 1980). The inflationary scenario predicts Gaussian initial conditions. The statistical properties of the Gaussian field are fully characterized by the two-point correlation function (2PCF) or its Fourier transform, the power spectrum.  Based on this idea, the two-point statistical measures for various cosmological fields have been extensively used for testing the paradigm as well as constraining cosmological parameters; the galaxy distribution in the redshift catalog (e.g., Totsuji \\& Kihara 1969; Hamilton \\& Tegmark 2002), the cosmic microwave background anisotropy (e.g., de Bernardis 2000) and the cosmic shear field (e.g., Van Waerbeke et al. 2001). However, even for Gaussian initial conditions, nonlinear gravitational instability induces non-Gaussian signatures in the mass distribution, which contain information on the nature of gravity and the dark matter. The three-point correlation function (3PCF), or equivalently, its Fourier-transformed counterpart, the bispectrum, is the lowest order statistical tool to probe the non-Gaussianity. The 3PCF can place strong constraints on models of structure formation. As an example, the measurements of the bispectrum in the galaxy distribution at the large scales ($\\simgt 10~ h^{-1}{\\rm Mpc}$)  have been used to determine the bias parameter as well as to constrain the primordial non-Gaussianity (e.g., Feldmann et al. 2001; Verde et al. 2001).  It is also expected that the 3PCF of the cosmic shear field can be used to precisely determine the cosmological parameters complementary to the measurements of the 2PCF (e.g., Bernardeau, Mellier \\& Van Waerbeke 2002; Bernardeau, Van Waerbeke, \\& Mellier 2003).  In this paper, we focus on developing a theoretical model of the 3PCF in real space for 3D and 2D cosmological fields.  In practice, on small scales the 3PCF would be easier to measure from observational data over the bispectrum, since it does not require the Fourier transform for the survey data that usually have a complicated geometry of data fields.  Theoretical models of the weakly nonlinear 3PCF have been well studied in the literature based on perturbation theory \\cite{Fry84b,Jing97,Gaz98,Frie99,Barriga02}. Perturbation theory can describe properties of the dark matter and galaxy clustering on large scales $\\simgt 10~ h^{-1}{\\rm Mpc}$ and predict that the 3PCF depends on the shape of triangle configuration and, as a result, contains information of the primordial power spectrum and the galaxy biasing. Historically, the pioneering measurement of the galaxy 3PCF done by Peebles \\& Groth (1975) (also Groth \\& Peebles 1977) proposed the ``hierarchical form'', $\\zeta(r_{12},r_{23},r_{31})=Q[\\xi(r_{12})\\xi(r_{23}) +\\xi(r_{23})\\xi(r_{31})+\\xi(r_{31})\\xi(r_{12})], $ with the constant $Q\\simeq 1.3$. However, subsequent work has revealed that the measured 3PCF does not obey the hierarchical form rigorously and the large-scale amplitudes can be explained by the perturbation theory results of the cold dark matter (CDM) models, if the biasing relation is correctly taken into account for the analysis \\cite{Jing98,Frie99}.  On the other hand, a quantitative theoretical model of the 3PCF in the strongly nonlinear regime is still lacking except for studies relying on $N$-body simulations (Matsubara \\& Suto 1994; Suto \\& Matsubara 1994; Jing \\& B\\\"orner 1998) \\footnote{The nonlinear bispectrum has been well studied based on extended perturbation theory (Scoccimarro \\& Friemann 1999; also see Bernardeau et al. 2002a) or the recently developed dark matter halo approach (Ma \\& Fry 2000c; Scoccimarro et al. 2001; Cooray \\& Hu 2001a).}. Simulations provide only limited physical insight into the complex non-linear phenomena involved in gravitational clustering, and are intractable for performing multiple evaluations in parameter space.  Therefore, the main purpose of this paper is to develop an analytical model for predicting the 3PCF applicable to both the linear and nonlinear regimes. For this purpose, we need a model to correctly describe the redshift evolution and statistical properties of gravitational clustering up to the three-point level. We employ the so-called dark matter halo model, where gravitational clustering is described in terms of correlations between and within dark matter halos.  Originally, this model was developed to express nonlinear clustering as the real-space convolution of halo density profiles (Neymann \\& Scott 1952; Peebles 1974; McClelland \\& Silk 1977; and also see Scherrer \\& Bertschinger 1991; Sheth \\& Jain 1997; Yano \\& Gouda 1999; Ma \\& Fry 2000b,c). Most recent works have relied on the the Fourier-space formulation, since the forms of the power spectrum and the bispectrum become much simpler (Seljak 2000; Ma \\& Fry 2000c; Peacock \\& Smith 2000; Scoccimarro et al. 2001; Cooray \\& Hu 2001a,b; Berlind \\& Weinberg 2002; Hamana, Yoshida \\& Suto 2002; Takada \\& Jain 2002 hereafter TJ02; Scranton 2002; and also Cooray \\& Sheth 2002 for a recent review). The halo model appears remarkably successful in that, even though it relies on rather simplified assumptions, it has reproduced results from numerical simulations (Seljak 2000; Ma \\& Fry 2000c; Scoccimarro et al. 2001; TJ02) and also allowed for interpretations of observational results of galaxy clustering \\cite{Seljak00,Scocci01}.  We formulate the 3PCF model so that it can be applied to general 3D and 2D cosmological fields, such as the mass and galaxy distributions and the cosmic shear fields.  Our method is built on the real-space formulation for the correlations of three particles in one halo.  This is because the real-space approach enables us to compute the one-halo contribution to the 3PCF by a 4-dimensional integration, which is an advantage compared to the Fourier space approach. For the 2- and 3-halo terms, we rely on the Fourier-space approach and the approximations presented in Scoccimarro et al. (2001; also see TJ02). We study the transition from the quasi-linear to nonlinear regimes and the relative contribution of the different terms to the 3PCF. We show the halo model predictions for the 3PCF of the mass and galaxy distribution and the weak lensing convergence field for the currently favored CDM model. To do this, we will focus on the dependences of the 3PCF on the triangle configurations as well as on the properties of the halo profile, its inner slope and concentration.  This paper is organized as follows.  In \\S \\ref{halo} we present the formalism to compute the 3PCF of the 3D density field based on the real-space halo model and review the Fourier-space halo approach and halo model ingredients (halo density profile, mass function and bias model) we will use in this paper. In \\S \\ref{ang3pt}, we present the method to compute the angular 3PCF. In \\S \\ref{results} we show the results of the halo model predictions for the 3PCF of the mass and galaxy distributions and the weak lensing convergence field. Finally, \\S \\ref{discuss} is devoted to a summary and discussion. We give some useful approximations for calculating the 3PCF in Appendix \\ref{app3pt} and \\ref{app}. We use the currently favored CDM model ($\\Lambda$CDM) with $\\Omega_{m0}=0.3$, $\\Omega_{\\lambda0}=0.7$, $h=0.7$ and $\\sigma_8=0.9$.  Here $\\Omega_{\\rm m0}$ and $\\Omega_{\\lambda0}$ are the present-day density parameters of matter and cosmological constant, $h$ is the Hubble parameter, and $\\sigma_8$ is the rms mass fluctuation in a sphere of $8h^{-1}$Mpc radius.  The choice of $\\sigma_8$ for this model is motivated by the cluster abundance analysis \\cite{Eke96}. ",
        "conclusions": "\\label{discuss} In this paper, we have used the halo clustering model to compute the 3-point correlation function (3PCF) of cosmological fields. We have shown results for the three-dimensional mass and galaxy distributions and the two-dimensional weak lensing convergence field.  The halo model enables us to separately consider the contributions to the 3PCF arising from triplets in a single halo or two or three different halos.  Thus we can understand how gravitational clustering transitions from the quasi-linear regime to the strongly non-linear regime as one goes to smaller scales. We found that the single halo contribution, which is dominant on small scales, can be computed using the real-space formulation of the halo model far more easily than the Fourier-space approach used in the literature.  We also developed approximations for computing the 2- and 3-halo contributions to the 3PCF.  Since measuring the real space 3PCF on small scales is  likely to be easier than the bispectrum, our model predictions will allow for the extraction of cosmological information from forthcoming galaxy surveys and weak lensing surveys.  We obtained the following results by applying our halo model to the concordance CDM model ($\\Lambda$CDM). The 3PCF on small scales $r\\simlt 1~ h^{-1}{\\rm Mpc}$ is dominated by the 1-halo term. Hence it probes properties of massive halos and can be used to constrain halo profiles as discussed below. The quasi-linear 3PCF predicted by perturbation theory can be reproduced by the 3-halo contribution for $r\\simgt 3~ h^{-1}$Mpc. For plausible halo model parameters, the hierarchical ansatz for the reduced 3PCF parameter, $Q= {\\rm constant}$, does not hold over the range of scales we have considered. However the $Q$ parameter does have a weaker dependence on  triangle configurations in the non-linear regime than in the quasi-linear regime (see Figures \\ref{fig:qpara} and \\ref{fig:qjing}).  These results qualitatively verify the results in Ma \\& Fry (2000b,c). Ma \\& Fry also pointed out that the halo model for plausible model parameters violates the stable clustering hypothesis which has been widely used to develop analytical prediction of the non-linear gravitational clustering, as done in the popular PD fitting formula.  In fact, very recently Smith et al. (2002) showed a weak violation of the stable clustering hypothesis using high-resolution $N$-body simulations.  However, our halo model predictions for the 3PCF do not match the asymptotic shape proposed in Ma \\& Fry (2000b) because of the assumptions employed in the asymptotic formula.  Therefore, it it merits further study to carefully investigate how small-scale gravitational clustering can be described by the halo model ingredients. Such a study can be carried out with the halo model methods developed in this paper.  We also found that the non-linear 3PCF is most sensitive to the inner slope of the halo profile and to the halo concentration parameter for given cosmological parameters (see Figures \\ref{fig:qprofr001}, \\ref{fig:qprofr01} and \\ref{fig:qalpha}). Combinations of the inner slope and the concentration can be adjusted so that the halo model matches the 2PCF result (see Figure \\ref{fig:2ptdep}).  However only one combination can match both the 2- and 3PCFs. Hence in combination with the 2PCF, the non-linear 3PCF could be used to constrain the properties of dark matter halos. We suggest that the use of higher order correlations is a useful way of measuring the properties of massive halos. For example, while lensing surveys of clusters can be used to measure cluster halo profiles directly, this involves identifying cluster centers and assigning masses to clusters to measure averaged profiles. In contrast, by measuring the 2- and 3PCFs, parameters of the halo mass function and profile can be fitted for. While this approach is more challenging computationally, and requires some theoretical assumptions, it treats the data more objectively.  Our halo model predictions match the simulation results of Jing \\& B\\\"orner (1998) in the non-linear regime as well as the quasi-linear regime, as shown in Figure \\ref{fig:qjing}. However, the halo model seems to be less accurate on the intermediate transition scale $\\sim 1~h^{-1}{\\rm Mpc}$.  Figure \\ref{fig:qparauv} shows that the predicted $Q$ parameter for equilateral triangles has a bump  at this scale, which corresponds to the bump found for the reduced bispectrum at $k\\sim 1~ h\\ {\\rm Mpc}^{-1}$ by Scoccimarro et al. (2001). This is unlikely to reflect real properties of dark matter clustering. Rather, we argued that the bump feature is sensitive to the sharp cutoff of halos at the virial radius and the exclusion effect for the 2- and 3-halo terms. We explored alternative prescriptions as shown in Figure \\ref{fig:excl}. To resolve this, detailed calculations and comparisons of the halo model predictions with simulation results will be needed.  We extended the halo model to predict the 3PCF of the galaxy distribution.  Once we model how galaxies populate a halo of given mass, the halo occupation number, we can straightforwardly predict the galaxy 3PCF based on the halo model developed here. For the halo occupation number for red galaxies we have used, the galaxy 3PCF has a smaller amplitude and weaker dependence on triangle configuration compared to the mass 3PCF (Figures \\ref{fig:qgal} and \\ref{fig:qgaluv}).  This trend is indeed consistent with the actual measurements in Jing \\& B\\\"orner (1998).  On the other hand, the 3PCF of blue galaxies displays complex features reflecting the input model of the halo occupation number. Thus, the galaxy 3PCF can be used to constrain the galaxy formation scenario as a function of host halo properties and  galaxy type. Further work is needed to model the expected properties of galaxies by type for specific surveys and compute the resulting 3PCF.  As an example of the angular 3PCF, we have computed the 3PCF of the weak lensing convergence field. In particular, we employed the real-space halo model to compute the single halo contribution, as for the 3D case, which enabled us to compute the 1-halo term by a 4-dimensional integration. We verified that the real-space halo model is equivalent to the Fourier-space model well studied in the literature (see Figure \\ref{fig:ang2pt}).  We also developed approximations for calculating the 2- and 3-halo terms. Because of projection effects, the 2- and 3-halo terms contribute to the 3PCF over a wider range of scales for the angular 3PCF compared to the 3D 3PCF. The resulting 3PCF does not  obey the hierarchical ansatz over the angular scales we have considered.  The lensing 3PCF is sensitive to cosmological parameters, in particular $\\Omega_{\\rm m0}$. Comparing measurements with model predictions  of the 3PCF can be a useful tool to break degeneracies between the power spectrum and $\\Omega_{\\rm m0}$. We intend to compute 3-point functions of the shear field following Schneider \\& Lombardi (2002) and Zaldarriaga \\& Scoccimarro (2002) as these are easier to measure from the data. In Takada \\& Jain (2003), we presented a brief investigation of the 3PCFs of shear fields, where we found good agreement between the halo model prediction and the results measured from the ray-tracing simulations. It is also expected that the $n$-point correlations of weak lensing on sub-arcminute scales contain a wealth of information on properties of massive halos, beyond their dependence on cosmological parameters. The real-space halo model we have developed will be a useful analytical tool for such calculations.  The halo model presented in this paper allows for several interesting applications. First, the model can be extended to investigate the effect of triaxial halo shapes on the 3PCF, since it is the lowest-order statistical probe of non-sphericity.  So far halo model applications assume a single, spherically symmetric profile. Recently, based on high-resolution simulations, Jing \\& Suto (2002) showed that halos are more accurately described by  triaxial halo profiles than the spherically symmetric NFW profile. They claimed that the axis ratios typically have vales of  $\\sim 0.6$ for the smallest and largest axis. The non-sphericity of the halo profile could lead to a characteristic configuration dependence of the 3PCF.  This effect should also affect the interpretation of cosmic shear measurements (Jing 2002). Likewise, the formulation developed in Sheth \\& Jain (2002) can be used to discuss the effects of substructure within halos on the 2PCF and  3PCF.  The real-space halo model can be directly used to compute the covariance of the 2PCF. As discussed in Cooray \\& Hu (2001b), the mass distribution on small scales displays pronounced non-Gaussian features induced by non-linear gravitational clustering. Hence one needs to take into account the non-Gaussian errors arising from the 4-point correlation function (4PCF).  On the  scales of interest the single halo contribution should dominate the error. The real-space halo model allows us to analytically compute the error contribution arising from the 4PCF with no additional computational effort than for the 3PCF (see \\S \\ref{rhalo}).  In Appendix \\ref{app}, we constructed an approximation for calculating the 1-halo term of the 3PCF based on the Fourier-space halo model. We showed that the the approximation accurately describes the amplitude and configuration dependence of the 3PCF (see Figure \\ref{fig:appq}). We will employ this approximation to perform an analytical study of the pairwise peculiar velocity dispersion (PVD) within the BBGKY hierarchy formalism (Peebles 1980). The PVD can be measured through the redshift distortions inferred from galaxy surveys (e.g., Zehavi et al. 2002).  The BBGKY picture tells us that the PVD arises mainly from the 3PCF on small scales.  However, there has been no comprehensive analytical model to describe the non-linear PVD.  This is because of the complex form of the BBGKY hierarchy equations. Peebles (1980) (see Mo, Jing \\& B\\\"orner 1997; Jing et al. 1998 for a detailed study) assumed the hierarchical form for the non-linear 3PCF, although it turns out to be violated for CDM models. The isothermal assumption for the velocity distribution within a given halo was employed to analytically obtain the PVD based on the halo model (Sheth et al. 2001).  It is important to clarify whether the BBGKY hierarchy leads to a self-consistent PVD for the CDM model. Moreover, we can easily combine the halo model prediction with models of the halo occupation number of galaxies to predict the PVD for different galaxy types, following the approach  in \\S \\ref{gal}.  \\bigskip  We would like to thank R. Sheth, R. Scoccimarro, I. Szapudi, A. Taruya and L. Hui for several valuable discussions. We thank Y. P. Jing for kindly providing us with his simulation data results. Helpful comments from the referee, Chung-Pei Ma, led to improvements in the paper. This work is supported by NASA grants NAG5-10923, NAG5-10924 and a Keck foundation grant.  \\appendix"
    },
    "0209/astro-ph0209484_arXiv.txt": {
        "abstract": "We present observations of the intrinsic absorption in the Seyfert 1 galaxy NGC 3783 obtained with the Space Telescope Imaging Spectrograph (STIS) on the {\\it Hubble Space Telescope} ({\\it HST\\,}) and the {\\it Far Ultraviolet Spectroscopic Explorer} ({\\it FUSE\\,}).  We have combined 18 STIS and 5 {\\it FUSE} observations to obtain a high signal-to-noise averaged spectrum spanning 905--1730~\\AA .  The averaged spectrum reveals absorption in \\ion{O}{6}, \\ion{N}{5}, \\ion{C}{4}, \\ion{N}{3}, \\ion{C}{3} and the Lyman lines up to Ly$\\epsilon$ in the three blueshifted kinematic components previously detected in the STIS spectrum (at radial velocities of $-$1320, $-$724, and $-$548~km~s$^{-1}$).  The highest velocity component exhibits absorption in \\ion{Si}{4}. We also detect metastable \\ion{C}{3} in this component, indicating a high density in this absorber. No lower ionization lines, i.e., \\ion{C}{2} and \\ion{Si}{2}, are detected.  A weak, fourth absorption component is tentatively detected in the high ionization lines and Ly$\\alpha$ and Ly$\\beta$ at a radial velocity of $-$1027~km~s$^{-1}$.  The Lyman lines reveal a complex absorption geometry. The strength of the higher order lines indicates Ly$\\alpha$ and Ly$\\beta$ are saturated over much of the resolved profiles in the three strongest absorption components and, therefore, their observed profiles are determined by the covering factor.  We separate the individual covering factors of the continuum and emission-line sources as a function of velocity in each kinematic component using the Ly$\\alpha$ and Ly$\\beta$ lines.  The covering factor of the BLR is found to vary dramatically between the cores of the individual kinematic components, ranging from 0 to 0.84.  Additionally, we find that the continuum covering factor varies with velocity within the individual kinematic components, decreasing smoothly in the wings of the absorption by at least 60\\%. Comparison of the effective covering factors derived from the \\ion{H}{1} results with those determined directly from the doublets reveals the covering factor of \\ion{Si}{4} is less than half that of \\ion{H}{1} and \\ion{N}{5} in the high velocity component.  Additionally, the FWHM of \\ion{N}{3} and \\ion{Si}{4} are narrower than the higher ionization lines in this component.  These results indicate there is substructure within this absorber.  We also find evidence for structure in the column density profiles of the high ionization lines in this component.  We derive a lower limit on the total column ($N_H \\geq$~10$^{19}$~cm$^{-2}$) and ionization parameter ($U \\geq$~0.005) in the low ionization subcomponent of this absorber. The metastable-to-total \\ion{C}{3} column density ratio implies $n_e \\approx$~10$^9$~cm$^{-3}$ and an upper limit on the distance of the absorber from the ionizing continuum of $R \\leq$~8~$\\times$~10$^{17}$~cm. The decreasing covering factor found in the wings of the absorption and the extreme compactness of the C III$^*$ absorber are suggestive of a clumpy absorption gas with low volume filling factor. ",
        "introduction": "Observations with the {\\it Hubble Space Telescope} ({\\it HST\\,}) have revealed that intrinsic UV absorption is a common phenomenon in Seyfert 1 galaxies, appearing in over half of the objects with available spectra \\citep{cren99a}.  The UV absorption resonance lines are typically blueshifted in the rest frames of the host galaxies, indicating radial outflow.  The absorption is often highly variable and, in some cases, only partially covers the continuum and emission-line regions of the active nuclei, which implies the absorbers are intrinsic to the AGN environments.  The observed variations may be due to changes in ionization in the absorbers \\citep{krol97,shie97,cren00} or changes in the total absorbing column, i.e., as a result of motion into and out of our line-of-sight \\citep{cren99b}.  NGC~3783 is a bright Seyfert 1 galaxy that exhibits strong UV absorption features and X-ray ``warm absorption\".  Several observations with {\\it HST} over the past decade have revealed dramatic variability in the UV absorption. NGC~3783 also has a highly variable UV continuum source.  An {\\it International Ultraviolet Explorer} ({\\it IUE\\,}) monitoring campaign revealed a factor of $\\sim$2 flux variations over timescales of 20--40 days \\citep{reic94,onke02}.  Intrinsic absorption in Ly$\\alpha$ and \\ion{C}{4}~$\\lambda\\lambda$1548,1551 was first detected in NGC~3783 with Faint Object Spectrograph (FOS) observations by \\citet{reic94}.  Three subsequent spectra obtained over a period of about two years with the Goddard High Resolution Spectrograph (GHRS) revealed highly variable \\ion{C}{4} absorption \\citep{mara96,cren99a}. There was no detectable \\ion{C}{4} absorption in the 1993 February GHRS spectrum, however, by 1995 April, two kinematic components appeared, at radial velocities of $-$1365~km~s$^{-1}$ (referred to as component 1) and $-$548~km~s$^{-1}$ (component 2) relative to the systemic redshift \\citep[we adopt $z =$~0.009760~$\\pm$~0.000093 throughout this paper;][]{deva91}. Additionally, a GHRS spectrum of the \\ion{N}{5}~$\\lambda\\lambda$1239,1243 spectral region revealed absorption coincident in velocity with component 2 just 16 days after the 1993 February observation of \\ion{C}{4} that showed no absorption, suggesting rapid variability \\citep{lusa94}. A Space Telescope Imaging Spectrograph (STIS) medium resolution echelle spectrum of NGC 3783 was obtained on 2000 February 27 revealing a third kinematic component in \\ion{C}{4} ($v_r=-$724 km s$^{-1}$, component 3), in addition to the components seen in the final GHRS spectrum \\citep{krae01a}. Ly$\\alpha$ and \\ion{N}{5}~$\\lambda\\lambda$1239,1242 also appeared in these three kinematic components in the STIS spectrum.  \\ion{Si}{4}~$\\lambda\\lambda$1394,1403 absorption was only found in component 1 and no lines from lower ionization species, i.e., \\ion{C}{2}, \\ion{Si}{2}, Mg II, were detected.  Using the \\ion{N}{5} doublet lines, \\citet{krae01a} found that all three absorption systems have a non-unity effective covering factor.  No correlation was found between the strength of the UV absorption features and continuum flux in the GHRS and STIS spectra by \\citet{krae01a}. They concluded the observed absorption variations were due largely to a change in total column.  However, it was not possible to constrain tightly the variation timescales due to the sampling of these observations.  Additionally, these observations did not sample the rapid changes in the continuum flux that were observed in the study by \\citet{reic94}, thus, it remains unclear what affect variable ionization has on the absorption.  A large ($>$~10$^{22}$ cm$^{-2}$) and variable column of ionized gas was measured in the X-ray spectrum of NGC~3783 with the {\\it Advanced Satellite for Cosmology and Astrophysics} ({\\it ASCA\\,}) \\citep[e.g.,][]{geor98}.  Subsequent observations with the High Energy Transmission Grating Spectrometer (HETGS) aboard the {\\it Chandra X-ray Observatory} ({\\it CXO\\,}) showed numerous absorption lines with a mean radial velocity of $\\sim-$610~$\\pm$~130~km~s$^{-1}$, consistent with components 2 and 3 identified in the UV \\citep{kasp00,kasp01}.  The coincidence in velocities suggests a link between the UV and X-ray absorbers.  Photoionization modeling of the STIS spectrum revealed that, although the UV absorbers can produce some of the observed X-ray columns, the ionization is too low and the total column too small to account for all of the features in the X-ray \\citep{krae01a}.  Hence, the exact relationship of the UV and X-ray absorption remained uncertain.  Additionally, \\citet{krae01a} found that two zones are required to explain the UV absorption columns measured in component~1.  The strength of the \\ion{Si}{4} absorption is inconsistent with the large \\ion{N}{5}/\\ion{C}{4} column density ratio measured in the STIS spectrum.  They proposed a model where a relatively high density absorber is co-located with more tenuous gas that is more highly ionized.  Various studies of intrinsic absorption in AGNs have demonstrated the importance of determining the absorption covering factor in the line-of-sight to the nucleus. In addition to affecting column density measurements, the covering factor constrains the absorption and emission geometry, providing tests for physical models. Non-unity effective covering factors, $C_f$, can arise from scattering of light into our line-of-sight by an extended scatterer \\citep[e.g.,][]{cohe95,good95,krae01b} or partial occultation of the emission sources \\citep{wamp93,barl97a,hama97}.  In the most general case, the different covering factors of the continuum and emission-line region need to be determined \\citep{barl97a,gang99}, and $C_f$ can vary across the profile of an individual absorption component \\citep[e.g.,][]{barl97b,arav02}. Additionally, instrumental scattering can affect the derived $C_f$ and column densities \\citep{cren98}.  High-resolution spectra with high signal-to-noise (S/N) are required to separate these effects.  We have undertaken an intensive multiwavelength monitoring campaign to probe the intrinsic absorption in NGC 3783 using observations in the UV, far-UV, and X-ray.  In this paper, we present an analysis of the averaged spectrum in the UV and far-UV from STIS and the {\\it Far Ultraviolet Spectroscopic Explorer} ({\\it FUSE\\,}).  We have co-added all observations to produce a high S/N spectrum that samples numerous lines from a range of ionization states.  Analysis of the mean X-ray spectrum is presented in \\citet{kasp02}, hereafter Paper~I. Details of the variable nature of the absorption will be presented in future papers by Gabel et al.\\ (in preparation) and George et al.\\ (in preparation). This paper is organized as follows: in \\S 2, we present details of the STIS and {\\it FUSE} observations and data reduction; in \\S 3, measurements of the covering factors and column densities for the UV absorbers are given; in \\S 4 we derive constraints on the physical conditions and geometry of the absorbers; we summarize our results in \\S 5. ",
        "conclusions": "NGC~3783 was observed with {\\it HST}/STIS at 18 epochs and with {\\it FUSE} at 5 epochs as part of an intensive multiwavelength monitoring campaign designed to investigate the intrinsic absorption in this Seyfert~1 galaxy. We have combined the observations to produce a high S/N averaged spectrum in the UV and far-UV. The major findings of our study follow.  1.  \\ion{O}{6}, \\ion{N}{5}, \\ion{C}{4}, \\ion{N}{3}, and the Lyman lines up to Ly$\\epsilon$ appear in the three kinematic components identified by \\citet{krae01a} in the initial STIS spectrum (components 1, 2, and 3 at radial velocities $-$1320, $-$548, and $-$724~km~s$^{-1}$, respectively) .  Galactic contamination prevents the detection of higher order Lyman lines. \\ion{Si}{4} appears in component~1; \\ion{C}{3}~$\\lambda$977 is heavily contaminated with Galactic absorption and cannot be measured in this component, but is present in components 2 and 3. We detect absorption from the \\ion{C}{3}$^* \\lambda$1175-76 multiplet in component~1, indicating a high density in this absorber.  This is the first detection of metastable \\ion{C}{3} in a Seyfert galaxy other than NGC~4151 \\citep{brom85}. No lower ionization lines appear in absorption, and we place stringent upper limits on their column densities.  A fourth kinematic component of intrinsic absorption is tentatively identified at a radial velocity of $-$1027~km~s$^{-1}$.  This component appears strong only in Ly$\\beta$, which may be contaminated with Galactic absorption, and \\ion{O}{6}.  In all other lines, the detection of this component is very sensitive to the fit to the intrinsic emission.  2.  The Lyman lines reveal a complex absorption geometry in NGC~3783 and highlight the importance of determining the individual covering factors of the continuum and emission-line sources. The strength of the higher order Lyman lines indicates Ly$\\alpha$ and Ly$\\beta$ are saturated in components 1--3. The continuum and emission-line covering factors were separated using the Ly$\\alpha$ and Ly$\\beta$ absorption in each kinematic component, both in the cores of absorption and as a function of radial velocity.  The covering factor of the BLR varies significantly between the cores of the absorption components, with $C_f^l =$~0.33, 0.84, 0.55, and 0.03 derived in components 1, 2, 3, and 4, respectively. The large residual fluxes measured in Ly$\\alpha$ in components~1 and 3 imply that, at least in these components, the non-unity emission-line covering factors cannot be fully accounted for by light from an extended scatterer, and thus represent partial coverage of the BLR. We also find evidence for variation of the continuum covering factor with velocity in the individual kinematic components.  Specifically, $C_f^c$ decreases by $\\sim$60\\% over several resolution elements in the wings of the absorption.  This smoothly decreasing coverage is a signature of substructure within the absorbers.  3.  The individual continuum and line covering factors derived from \\ion{H}{1} were used to derive effective covering factors for all lines. Comparison to the covering factors derived directly by the doublet method reveals the \\ion{Si}{4} covering factor is smaller than that of \\ion{H}{1} and \\ion{N}{5} by more than a factor of two in component~1. This implies substructure in component~1 and is consistent with the prediction by \\citet{krae01a} that this component is comprised of at least two zones of UV absorption.  Furthermore, the FWHM of \\ion{Si}{4} and \\ion{N}{3} are narrower than the higher ionization lines in component~1, providing additional evidence for substructure. Employing the \\ion{H}{1} covering factor profiles for the higher ionization lines, we find the relatively weak apparent \\ion{C}{4} and \\ion{O}{6} absorption compared to \\ion{N}{5} in component~1 is due to low effective covering factors rather than small column densities. If the covering factors derived from \\ion{H}{1} are valid for \\ion{C}{4} and \\ion{N}{5}, then the column densities of these lines are not uniform across the profile of component~1; they are saturated in the blue wing, but smaller in the red wing. Additionally, \\ion{O}{6} is found to be saturated in components~2 and 3, as well as in the blue wing of component~1.  4.  We place lower limits on the total hydrogen column density and ionization parameter in the low ionization subcomponent in component~1, $N_H \\geq$~10$^{19}$~cm$^{-2}$ and $U \\geq$~0.005.  Combining model predictions of the total \\ion{C}{3} column with our measured \\ion{C}{3}$^*$ column gives $n_e \\approx$~10$^{9}$~cm$^{-3}$.  These results imply an upper limit on the distance of the absorber from the ionizing continuum source of $\\leq$~8~$\\times$~10$^{17}$~cm. Using upper limits on the size of the BLR from the variability study by \\citet{onke02}, the projected area of the absorbers at velocities exhibiting partial coverage of the emission-line source is found to be $\\sim$(10$^{16}$)$^2$~cm$^2$. The decreasing covering factor exhibited in the wings of the Lyman lines and extreme compactness of the \\ion{C}{3}$^*$ absorber are suggestive of a clumpy absorbing medium with a low filling factor.   J. R. G., D. M. C., and S. B. K. acknowledge support from NASA grant HST-GO-08606.13-A.  W.N.B acknowledges CXC grant GO1-2103 and NASA LTSA grant NAG5-8107.  F.H. acknowledges NSF grant AST 99-84040. We thank B-G Andersson and the {\\it FUSE} team for assistance with the {\\it FUSE} spectra and Derck Massa for providing his code for modeling the H$_2$ absorption.  We also thank the anonymous referee for very helpful comments.  \\clearpage"
    },
    "0209/astro-ph0209351_arXiv.txt": {
        "abstract": "We photometrically observed the VY Scl-type cataclysmic variable KR Aurigae after its final rise from the fading episode in 2000--2001. Time-resolved observation revealed that the light curve is dominated by persistent short-term variation with time-scales of minutes to tens of minutes.  On some nights, quasi-periodic variations with periods of 10--15 min were observed.  No coherent variation was detected.  The power spectral density of the variation has a power law component ($f^{-1.63}$). The temporal properties of the short-term variations in KR Aur present additional support for the possibility that flickering in CVs may be better understood as a result of self-organized critical state as in black-hole candidates.  The light curve lacks ``superhump\"-type signals, which are relatively frequently seen in VY Scl-type systems and which are suggested to arise from tidal instability of the accretion disk induced by changing mass-transfer rates.  The present observation suggests a borderline of superhump excitation in VY Scl-type stars between mass ratios $q$=0.43 (MV Lyr) and $q$=0.60 (KR Aur). ",
        "introduction": "Cataclysmic variables (CVs) are close binary systems consisting of a white dwarf and a red dwarf secondary transferring matter via the Roche lobe overflow.  The resultant accretion disk becomes thermally stable in systems with high mass-transfer rates ($\\dot{M}$).  Such systems are called novalike (NL) stars, because they lack outbursts characteristic to dwarf novae [see \\citet{osa96review} for a review].  Among NL stars, there exists a small group which shows a temporary reduction or cessation of $\\dot{M}$ from the secondary.  These systems are called VY Scl-type stars or anti-dwarf novae (\\cite{war95book}).  Although accretion disks in NL stars are thermally stable, the disk can be tidally unstable (\\cite{osa95eruma}; \\cite{osa96review}). Numerical simulations have shown that this instability (tidal instability: \\cite{whi88tidal}) only appears below a certain mass ratio ($q=M_2/M_1$): $q<0.25-0.33$ depending on calculations (\\cite{hir90SHexcess}; \\cite{woo00SH}; \\cite{whi91SH}; \\cite{mol92SHexcess}; \\cite{mur98SH}). In recent years, several systems above this stability limit are known to show superhumps \\citep{pat99SH}.  Since many of them are VY Scl-type stars, there has been a theoretical interpretation that accretion disks can be tidally unstable upon variation of $\\dot{M}$ \\citep{mur00SHvyscl} even in intermediate $q$ systems.  Temporary appearance of superhump signals in a recent low state of a VY Scl-type star, BZ Cam \\citep{kat01bzcam} may support this interpretation. However, observations have not yet fully illustrated the upper $q$ limit for superhumpers in VY Scl-type stars. KR Aur ($q$=0.60) is an ideal system to examine such a condition, since this system has the smallest $q$ among VY Scl-type stars which have not been reported to show (or studied for) superhumps.  \\begin{figure*} \\begin{center} \\FigureFile(160mm,80mm){fig1.eps} \\end{center} \\caption{Long-term light variation of KR Aur between 1994 and 2002 drawn from visual observations reported to VSNET. Large dots and small dots represent positive and negative (upper limit: object undetected) observations, respectively.  Typical errors of visual observations are 0.2--0.3 mag. The open circle represents our snapshot CCD observations. The mean epoch of our time-series CCD photometry is shown by an arrow. } \\label{fig:vis} \\end{figure*} ",
        "conclusions": "\\subsection{Long-Term Variation}  Figure \\ref{fig:vis} long-term shows light variation of KR Aur between 1994 and 2002 drawn from visual observations reported to VSNET Collaboration.\\footnote{ $\\langle$ http://www.kusastro.kyoto-u.ac.jp/vsnet/ $\\rangle$. }  Typical errors of visual observations are 0.2--0.3 mag, which will not affect the following analysis. The light curve shows predominant low states (durations: several months to two years) when the object is fainter than mag 14, and occasional high states when the object is typically between mag 13 and 14. Such a high occurrence of low states is relatively rare among VY Scl-type stars \\citep{gre98vyscl}.  The long-term behavior of the system most resembles ``superminimum\" of MV Lyr (\\cite{rob81mvlyr}; \\cite{wen83mvlyr}; \\cite{fuh85mvlyr}).  At a closer look, the system underwent a small dwarf nova-like outburst around JD 2450745--2450755.  The object then became inactive.  Before the fully recovery, the object underwent a rather complex brightening around JD 2451580--2451630.  This behavior is quite reminiscent of the fading and recovery processes observed in MV Lyr (\\cite{pav98mvlyr}; \\cite{shu98mvlyr}; \\cite{pav99mvlyr}). Although such behavior is unexpected for variable mass-transfer rates on a usual CV \\citep{hon94v794aql}, \\citet{lea99vyscl} showed that, in the presence of heating from a hot white dwarf, the irradiation on the accretion disk suppresses the thermal instability, which can reproduce the observed light curve of VY Scl-type systems.  From these observations, we propose that MV Lyr and KR Aur comprise the most ``active\" subgroup of VY Scl-type stars.  Figure \\ref{fig:overall} presents overall light variation of KR Aur in 2001 November--December drawn from observations in table \\ref{tab:log}. Although some irregular variation is superimposed, no major ``flare\"-like brightening (cf. \\cite{pop78kraur}) was observed.  \\subsection{Superhumps}\\label{sec:sh}  Figure \\ref{fig:perana} presents a period analysis of the November 23--December 17 data, after subtracting a linear fit to the overall light curve.  The upper and lower panels represent the results of period analysis with Phase Dispersion Minimization \\citep{PDM} and with the CLEAN algorithm \\citep{CLEAN}, respectively.  The orbital period is marked with a tick on the upper panel.  No significant coherent periodicity was detected at or near the orbital period, indicating that no detectable periodic or quasi-periodic variations related to orbital modulations or superhumps were present (see also subsection \\ref{sec:short}).  The absence of superhumps makes a clear contrast to other VY Scl-type stars with superhumps (TT Ari: \\citet{uda88ttari}; \\citet{ski98ttari}; \\citet{and99ttari}; \\citet{kra99ttari}; \\citet{sta01ttariSHspec}, MV Lyr: \\citet{ski95mvlyr}, V751 Cyg: \\citet{pat01v751cyg}). This finding suggests that the mechanism proposed by \\citet{mur00SHvyscl} is ineffective in the mass ratio of KR Aur. We propose that there is a borderline of superhump excitation between mass ratios $q$=0.43 (MV Lyr) and $q$=0.60 (KR Aur) in VY Scl-type stars.  Future determination of mass ratios in longer $P_{\\rm orb}$ systems (i.e. candidates for systems with higher $q$ than MV Lyr) BZ Cam, V751 Cyg and TT Ari is expected to provide a more stringent constraint to this limit (see table \\ref{tab:vyscl} for a summary of superhumps and binary parameters of VY Scl-type stars).  \\begin{table} \\caption{Parameters of VY Scl-Type Stars$^*$.}\\label{tab:vyscl} \\begin{center} \\begin{tabular}{llccc} \\hline\\hline Object & $P_{\\rm orb}$ (d) & $P_{\\rm SH}$ (d) & $M_1$ & $q$ \\\\ \\hline VY Scl      & 0.232    & $\\cdots$ & 1.22 & 0.32 \\\\ KR Aur      & 0.16280  & $\\cdots$ & 0.59 & 0.60 \\\\ LX Ser$^\\dagger$ & 0.158432 & $\\cdots$ & 0.41 & 0.88 \\\\ BZ Cam      & 0.153693 & 0.15634  & $\\cdots$ & $\\cdots$ \\\\ V794 Aql    & 0.1533   & $\\cdots$ & $\\cdots$ & $\\cdots$ \\\\ V425 Cas    & 0.1496   & $\\cdots$ & $\\cdots$ & $\\cdots$ \\\\ VZ Scl      & 0.144622 & $\\cdots$ & 1:   & 0.7: \\\\ V751 Cyg    & 0.144464 & 0.1394   & $\\cdots$ & $\\cdots$ \\\\ PG 1000+667 & 0.144384 & $\\cdots$ & $\\cdots$ & $\\cdots$ \\\\ TT Ari      & 0.137551 & 0.133160 & $\\cdots$ & $\\cdots$ \\\\ &          & 0.148815 &      &      \\\\ DW UMa$^\\dagger$ & 0.136607 & 0.1330   & 0.9  & 0.32 \\\\ MV Lyr      & 0.1329   & 0.138    & $\\cdots$ & 0.43 \\\\ V442 Oph$^\\dagger$ & 0.124330 & 0.12090  & $\\cdots$ & $\\cdots$ \\\\ \\hline \\end{tabular} \\end{center} {\\footnotesize $^*$ LQ Peg \\citep{kat99lqpeg} is also known as a VY Scl-type star. The orbital period has not been reported.  $^\\dagger$ SW Sex star (see \\cite{hel00swsexreview} for a recent review; see also \\cite{tho91pxand}).  A few other SW Sex stars [PX And (\\cite{sti95pxand}); BH Lyn (\\cite{hoa97bhlyn})] are claimed to show some degree of low/high state transitions (mainly from spectroscopic observations). Such a variation may more represent a change in the disk state (cf. \\cite{gro01swsex}) rather than VY Scl-type temporary reduction or cessation of $\\dot{M}$ from the secondary. Some DQ Her stars (intermediate polars) also show temporary fadings (\\cite{gar88vyscldqher}; \\cite{hes00CVlowstate}), but they are not usually considered as VY Scl-type stars.  {\\bf References:} VY Scl: \\citet{mar00vyscl}; KR Aur: \\citet{sha83kraur}; LX Ser: \\citet{you81lxser}; BZ Cam: \\citet{pat96bzcam}, \\citet{kat01bzcam}; V794 Aql: \\citet{hon98v794aql}; V425 Cas: see \\citet{kat01v425cas} and the references therein; VZ Scl: \\citet{war75vzscl}, \\citet{rob76CVmass}, \\citet{odo87vzscl}, \\citet{she84vzscl}; V751 Cyg: \\citet{pat01v751cyg}; PG 1000+667: \\citet{hil98pg1000}; TT Ari: \\citet{cow75ttari}, \\citet{tho85ttari}, \\citet{uda88ttari}, \\citet{ski98ttari}, \\citet{and99ttari}, \\citet{kra99ttari}, \\citet{sta01ttariSHspec}; DW UMa: \\citet{sha88dwuma}, \\citet{pat99SH}, \\citet{bir00dwuma}; MV Lyr: \\citet{ski95mvlyr}; V442 Oph: \\citet{pat99SH}, \\citet{hoa00v442oph}, \\citet{dia01v442oph}. } \\end{table}  \\subsection{Short-Term Variations}\\label{sec:short}  Figure \\ref{fig:nightly} shows typical examples of nightly light curves. Three longest runs (November 23 and 30, December 6) were selected as representatives of different epochs of the present observation. All light curves show distinct short-term (minutes to tens of minutes) quasi-periodic variations.  No systematic variation close to the orbital period (0.16280 d) was observed (cf. subsection \\ref{sec:sh}). Aside from the lack of superhumps, these variations look remarkably similar to those of quasi-periodic oscillations (QPOs) observed in another VY Scl-type star, TT Ari (e.g. \\cite{mar80ttari}; \\cite{jen83ttari}; \\cite{hol92ttariQPO}; \\cite{tre96ttari}; \\cite{and99ttari}).  Figure \\ref{fig:power} shows power spectra of nightly observations. On some nights (November 23, 30, December 11), an increased power around the frequencies 100--150 d$^{-1}$ (corresponding to the periods of 10--15 min) was observed.  There was no common period between the nightly observations. This finding confirms the quasi-periodic nature of the short-term variations. We have not been able to confirm the presence of 25-min periodicity claimed by \\citet{bir90kraur}.  The periods of the presently observed QPOs are close to the periodicities (480--780 s) recorded by \\citet{sin93kraurQPO}. We have also confirmed night-to-night variation of dominant periods as was originally claimed by \\citet{sin93kraurQPO}.  We also note that the overall profiles of short-term variations (figure \\ref{fig:nightly}) and the nightly variation of the power spectra are also similar to those observed in the peculiar symbiotic binary V694 Mon (\\cite{mic93v694mon}; \\cite{dob96v694mon}; \\cite{ish01v694mon}).  There are two major types of ``quasi-periodic\" oscillations in CVs: dwarf nova oscillations (DNOs), which are oscillations observed only in dwarf nova outbursts, having periods of 19--29~s, and have long (several tens to $\\sim$100 wave numbers) coherence times (\\cite{rob73DNO}; \\cite{szk76DNO}; \\cite{hil80DNO}), and (in a narrower sense) QPOs (for a review, see \\citet{war95book}).  The present QPOs in KR Aur correspond to the latter classification.  Several models have been proposed to account for QPOs, including vertical or radial oscillations of the accretion disk \\citep{kat78QPO}, reprocessing of the light by the orbiting blobs \\citep{pat79aeaqr}, non-radial pulsations of the accretion disk (\\cite{pap78CVQPO}; \\cite{vanhor80DNQPO}), radial oscillation of the accretion disk (\\cite{cox81QPO}; \\cite{blu84QPO}; \\cite{oku91QPO}; \\cite{oku92QPO}), excitation of trapped oscillations around the discontinuity of physical parameters \\citep{yam95DNoscillation}.  There has been also a suggestion that the magnetism of the white dwarf can be responsible for some kinds of QPOs \\citep{mik90chcygflickering}.  Although present observations are not able to constrain the origin of QPOs, a future search for coherent X-ray, ultraviolet or optical pulsations, which are a well-known signature of a magnetic white dwarf, would be helpful in discriminating the possibilities.  \\begin{figure} \\begin{center} \\FigureFile(88mm,60mm){fig6.eps} \\end{center} \\caption{Power spectral density (PSD) of the entire data set. High-frequency white noise has been subtracted from the PSD. Above the frequency $\\log f$(d$^{-1}$)=1.5, the PSD is proportional to $f^{-1.63}$. } \\label{fig:powave} \\end{figure}  \\subsection{Power Spectrum of Flickering}  Figure \\ref{fig:powave} shows power spectral density (PSD) of the entire data set.  High-frequency white noise has been subtracted from the PSD.  Above the frequency $\\log f$=1.5 d$^{-1}$ (corresponding to time-scales shorter than $\\sim$45 min), the PSD is proportional to $f^{-1.63}$.  These short-term variations (flickering) are one of the most characteristic features in CVs.  Although the existence of flickering in CVs has been well-documented since the 1940's (see \\cite{bru92CVflickering} for an extensive historical review), their physical origin has not been well understood.  Historically, \\citet{war71ugem} demonstrated that flickering almost disappeared during eclipses of the eclipsing dwarf nova U Gem.  This finding indicated that the origin of flickering is strongly associated with the hot spot (the stream impact point on the accretion disk).  \\citet{jam87spotflickering} noted the presence of power law-type frequency dependence, and proposed that a multiple scattering from the flickering source (hot spot) is responsible for this frequency dependence.  More recently, \\citet{bru91NLflickering} more extensively studied the properties of flickering, and summarized an observational review \\citep{bru92CVflickering}. From the analysis of frequency and color dependencies, \\citet{bru92CVflickering} and \\citet{bru93CVboundarylayer} suggested that the inner part of the accretion disk (rather than the hot spot) is more responsible for flickering.  More direct observational evidence for a major contribution from the inner accretion disk to flickering has been demonstrated through eclipse observations of CVs (HT Cas: \\cite{wel95htcas}; \\cite{bru00htcasv2051ophippeguxumaflickering}, Z Cha: \\cite{bru96zchaflickering}). It is now widely believed that the originally proposed idea of stream impact-type flickering \\citep{war71ugem} applies to a only limited sample of CVs, or contributes to a small extent to overall flickering activity, and that most of CVs have a strong concentration of flickering activity toward the inner accretion disk \\citep{bru96zchaflickering}.  In order to reproduce this power-law spectrum, \\citet{yon97CVflickering} proposed a superposition of ``shots\" in a self-organized critical state (SOC), which was originally introduced to explain time variations in black-hole candidates (BHCs) (e.g. \\cite{min94BHADSOC}; \\cite{min94BHfluctuation}; \\cite{tak95BHADfluctuation}; \\cite{kaw00BHADfluctuation}). Taking this analogy into account, the power law-type temporal properties of the short-term variations in KR Aur present additional support for the possibility that flickering in CVs may be better understood as a result of SOC as in BHCs.  Although the detailed mechanism of energy release was not identified at the time of \\citet{yon97CVflickering}, \\citet{wil02ADflare} recently tried to explain flickering by a superposition of flares, resulting from an energy release in the photosphere of the accretion disk of injected high-energy electrons originating from reconnections of magnetic field lines.  This type of energy release in visual wavelengths would be a promising candidate for explaining flickering in CVs.  Further quantitative comparisons with numerical simulations and observed properties in CVs will be a next step toward understanding flickering in CVs."
    },
    "0209/astro-ph0209398_arXiv.txt": {
        "abstract": "We combine deep, wide-field near-IR and optical imaging to demonstrate a reddening-independent quasar selection technique based on identifying outliers in the (g $-$ z) / (z $-$ H) colour diagram.  In three fields covering a total of $\\approx 0.7$ deg$^2$ to a depth of m$\\rm_H$$\\sim$18, we identified 68 quasar candidates.  Follow-up spectroscopy for 32 objects from this candidate list confirmed 22 quasars (0.86$<$z$<$2.66), five with significant IR excesses.  2 of 8 quasars from a subsample with U band observations do not exhibit UVX colours.  From these preliminary results, we suggest that this combined optical and near-IR selection technique has a high selection efficiency ($> 65$\\% success rate), a high surface density of candidates, and is relatively independent of reddening.  We discuss the implications for star/galaxy separation for IR base surveys for quasars.  We provide the coordinate list and follow-up spectroscopy for the sample of 22 confirmed quasars. ",
        "introduction": "Quasar spectral energy distributions are diverse and comparing surveys at different wavelengths, including the infrared, is important for characterizing the quasar population.  In addition, compared to optical surveys, infrared selection is less effected by dust extinction and reddening (due to both dust within host galaxies or along the line of sight).  This has implications not only for characterizing the influence of dust obscuration on the observed population, but also for understanding biases in lensing (Kochanek 1996) and Damped Lyman-$\\alpha$ (DLA) absorption system studies based on optical selected quasar samples (Pei and Fall 1995, Fall, Pei, McMahon 1989).  Infra-red observations were first used in quasars selection by Braccesi, Lynds and Sandage (1968) where they showed how Infrared excess could be used in conjunction with UV excess, derived using U$-$B colours, to distinguish quasars from galactic foreground stars. These early observations used red sensitive photographic emulsions which were sensitive to wavelengths beyond that of the human eye but which we would now call the the optical I band(7000-9000\\AA). The underlying physical principle behind the technique was the use of a wide wavelength range, 3500-8000\\AA, large enough to distinguish the black body dominated stellar spectra from the non-thermal powerlaw dominated spectral energy distribution of quasars.  The use of IR data alone to discover quasars was demonstrated by Beichman \\etal (1998) who discovered a z=0.147 quasar using JHK data from the 2 Micron All Sky Survey (2MASS).  However, due to current technological limitations, obtaining deep multi band IR observations over a sufficiently wide field of view to allow the construction of a sample of satisfactory size at cosmologically interesting redshifts (i.e. z=1-2) is prohibitively expensive in terms of telescope time. The availability of large infrared mosaic cameras makes possible wide-field IR surveys and the selection of large quasar samples at IR wavelengths.  The advantage of the IR approach compared to optical is the reduced influence of both dust extinction and reddening.  The removal of sources from a flux limited sample by dust extinction will be more severe at optical wavelengths than in the infrared.  In addition, as discussed below, the reddening vector does not push quasars into the stellar locus.  For example, Francis, Whiting, and Webster 2000 show that red quasars are indistinguishable from the stellar locus in optical multi-colour surveys, but often can be distinguished with the addition of near-IR data.  Similarly, Barkhouse and Hall 2001 recommend the combination of optical and near-IR as most effective (verse optical or near-IR separately).  A complementary approach is to use radio or hard X-rays as a method to identify AGN in a manner that is not strongly biased by dust. In a recent program, Ellison \\etal (2001) report the results of an survey for DLAs within a sample of radio selected quasars.  Ellison \\etal (2001) used a sample of 878 flat spectrum radio source with S$\\rm_{2.7GHz}$$>$250mJy.  However it is not obvious that these surveys can discover enough quasars with z$>$2, which are luminous enough in the optical that searches for damped Lyman-$\\alpha$ can be carried out, since typically $\\sim$1\\% of quasars have radio emission at the 250mJy flux level reached by Ellison \\etal (2001).  Hooper \\etal (1995) observed 3 out of 256 quasars (1.2\\%) with m$\\rm_B$$<$18.5 and S$\\rm_{8GHz}$$>$250mJy , rising to 14 (5.5\\%) with S$\\rm_{8GHz}$$>$ 25mJy). Using a deeper radio survey is possible but at the mJy level, radio surveys are dominated by faint galaxies and the follow-up of such surveys could be prohibitive.  Our approach is similar to the KX technique of Warren, Hewett, and Foltz (2000) which proposes selecting quasar candidates by their K-excess in a V$-$J/J$-$K colour-colour diagram (see Croom, Warren and Glazebrook (2001) for an application of the method).  As an alternative to VJK based candidate selection we demonstrate a technique based on gzH observations.  There are two reasons for our alternate band selection.  1) Only one near IR band is required.  The use of the z band, readily obtained with large format CCD based cameras, leads to an increase in the practicality of surveying a large area. The relatively small field of view of IR array cameras is imposed only on H band observations.  2) We demonstrate that after accounting for sky brightness, K-correction and potential differential dust extinction effects between the H and K bands, observations in either band are equivalent with respect to practical observational requirements. The current availability of wide area IR imaging observations in the H band with CIRSI (section \\ref{Near-IR and Optical Observations}) make sample definition possible over a wide field of view (in excess of 10deg$^2$).  Based on information obtained from the ESO ISSAC and SOFI exposure time calculator (http://www.eso.org/observing/etc/) we find the H and K sky brightnesses to be H=14.4 and K=13.0.  The quasar (H$-$K) K-correction is estimate to be $-$0.6 (assuming a simple power law quasar model $\\alpha$=$-$0.5).  Under these assumptions quasars are 0.8mag brighter relative to the sky in H than in K.  Strong reddening (A$\\rm_{V(rest)}$$>$2.0) in the quasar rest frame or along the line of site is required before this situation is reversed due to the shallow slope of the reddening function at longer wavelengths under most parameterization of the reddening law (figure \\ref{reddening}).  Although several papers have used the colours of previously known quasars to discuss quasar selection using combined optical and near-IR imaging, we demonstrate this technique by carrying out a survey for previously unknown quasars.  We present follow-up spectroscopic observations that confirm the technique, and suggest high selection efficiency, high surface density, and relative independence from reddening.  In the following section (\\ref{Near-IR and Optical Observations}) we discuss the infrared and optical imaging data, section (\\ref{colours}) describes the selection of quasar candidates using colour criteria, section (\\ref{host}) is concerned with the implications from detection of the host galaxy.  The spectroscopic observations are described in section (\\ref{spectra}), and the sample of confirmed quasars is presented in section (\\ref{quasars}).  Unless stated otherwise we use conventional Vega magnitudes and H$_{0}=$50 km s$^{-1}$ Mpc$^{-1}$, q$_{0}$=0.5 throughout this work.  \\begin{figure} \\psfig{file=reddening_1_paper.ps,width=8cm} \\psfig{file=reddening_2_paper.ps,width=8cm} \\caption[]{\\label{reddening} a) The (H$-$K) colour for a power law model quasar spectrum ($\\alpha$=$-$0.5) is compared to that of a power law with reddening at the level of A$\\rm_{V(rest)}$=1.0.  The (H$-$K) sky colour is indicated by the solid line.\\\\ b) Absorption as a function of redshift for A$\\rm_V$=1.0 in the rest frame is shown in the gzH and K bands.\\\\The galactic reddening law of Seaton 1979 is used.} \\end{figure} \\begin{figure} \\psfig{file=sdss_2mass_gzh.ps,width=8cm} \\caption[Model Colours]{\\label{model}An optical and near-IR colour diagram showing quasar model colours (described in the text).  The colours of Galactic stars are computed from the atlas of Bruzual, Persson, Gunn and Stryker included in the IRAF/STSDAS package SYNPHOT. The reddening vector does not drive the quasars into the stellar locus unlike purely optical colour-colour plots (see figure \\ref{ugr}).  The unfilled circles indicate the colours of known quasars in SDSS (Richards \\etal 2001), for which we obtained H magnitudes by looking for matches (within 2'') in the 2MASS second incremental release point source catalogue.} \\end{figure} \\begin{figure} \\epsfig{file=plotdh.eps,width=8cm} \\caption[Magnitude Calibration]{\\label{plotdh}The magnitude difference between CIRSI and 2MASS catalogues are shown for the 1204$-$0736 field. The rising tail at bright magnitudes is due to saturation in the CIRSI data.  The median magnitude offset is 0.07m.  After correcting for this median offset, the rms scatter between the two samples is 0.12m.} \\end{figure} \\begin{figure} \\psfig{file=cirsi_qso_gzh.ps,width=8cm} \\caption[Colour Diagram]{\\label{plotcd}The observed optical and near-IR colour diagram is shown for the 0218$-$0500 field (Table \\ref{fields table}). Identified quasars are indicated by filled circles.  The selection boundary (dot-dashed line), chosen based on model quasar colours, is $(z-H) > 0.59 \\times (g-z) - 0.06$.} \\end{figure} \\begin{figure} \\epsfig{file=plotspc.eps,width=8cm} \\caption[Quasar Spectra]{\\label{plotspc}The follow-up spectroscopy for the quasar sample is shown.  The first six spectra in the plot were taken with a long-slit spectrograph on the du Pont telescope.  The remaining spectra were taken with the Hydra multi-object spectrograph on WIYN. The flux calibrated spectrum of object 6 shows an IRX quasar while the fiber spectra of objects 14, 15, 16 and 19 are consistent with the IRX photometry.} \\end{figure} \\begin{figure} \\epsfig{file=plotz.eps,width=8cm} \\caption[Redshift distribution]{\\label{plotz}The redshift distribution for the sample of 22 confirmed quasars is shown.  The observed redshifts range from $z = 0.858$ to $z = 2.660$ with a median redshift $z = 1.211$.} \\end{figure} ",
        "conclusions": "We report the identification of a gzH selected sample of z$<$3 quasars.  The $g-z/z-H$ colour diagram is used with high selection efficiency ($\\sim$65\\%) and reduced sensitivity to reddening and extinction.  The initial sample has 22 confirmed quasars, but the full INT CIRSI survey will contain several hundred.  To date two objects within a subsample of the identified quasars, for which UV observations are available, do not possess a significant UV excess and would not have been recovered by a UVX based selection.  The full CIRSI-INT gzH quasar sample will cover an area of in excess of 10deg$\\rm^2$ \\ including observations across the full range of bands ugrizJ and H.  Contrasting the sample with those compiled through a range of selection techniques, such as UVX colour selection and radio identifications from overlapping surveys such as the VLA-FIRST survey, will provide an excellent tool for identifying biasing within quasar selection methodologies.  \\begin{table*} \\caption{\\label{fields table}Survey fields} \\begin{center} \\begin{tabular}{ccccccc} Name & \\multicolumn{2}{c}{R.A. (J2000) Dec} & Area & \\multicolumn{3}{c}{Depth (Vega)} \\\\ & & & deg$^2$ & $g$ & $z$ & $H$ \\\\ \\hline 1204$-$0736 & 12:04:50   &  $-$07:36:00 & 0.18  &  24  & 21  & 18 \\\\ 1636+4101 & 16:36:50     &  +41:01:50 & 0.35 & 24 & 21 & 18 \\\\ 0218$-$0500 & 02:18:00   &  $-$05:00:00 & 0.17 & 24 & 21 & 17.5 \\\\ \\hline \\end{tabular} \\end{center} \\end{table*} \\begin{table*} \\label{quasars table} \\caption{The quasar Sample.} \\begin{center} \\begin{tabular}{cccccccccc} Name & \\multicolumn{2}{c}{R.A. (J2000) Dec} & \\multicolumn{3}{c}{magnitude (Vega)} & $z$ & g-z & z-H & Comments\\\\ & & &$m_{g}$ & $m_{z}$ & $m_{H}$\\\\ \\hline CIRSI1 & 12:03:48.61 & $-$07:19:00.7 & 19.98 & 18.76 & 17.77 & 0.858               & 1.22 & 0.99 \\\\ CIRSI2 & 12:04:21.62 & $-$07:22:01.0 & 19.46 & 18.41 & 17.78 & 1.189               & 1.05 & 0.63 \\\\ CIRSI3 & 12:04:09.20 & $-$07:24:22.8 & 19.89 & 18.96 & 18.12 & 1.625               & 0.93 & 0.84 \\\\ CIRSI4 & 12:04:08.18 & $-$07:21:45.6 & 20.13 & 19.41 & 18.26 & 1.097               & 0.72 & 1.15 \\\\ CIRSI5 & 12:05:34.59 & $-$07:41:21.4 & 19.08 & 17.97 & 17.19 & 2.660               & 1.11 & 0.78 \\\\ CIRSI6 & 12:05:07.68 & $-$07:45:26.6 & 21.01 & 19.40 & 17.89 & 1.420               & 1.61 & 1.51 & IRX, No UV data\\\\ CCS88 163351.6+410628 & 16:35:31.05 & +41:00:27.2 & 19.14 & 18.25 & 17.21 & 1.151  & 0.89 & 1.04 & Crampton \\etal (1988)\\\\ CIRSI8 & 16:37:10.03 & +40:56:42.9 & 20.21 & 19.22 & 18.06 & 1.434                 & 0.99 & 1.16 & Candidate Crampton \\etal (1992)\\\\ CIRSI9 & 16:37:00.64 & +41:05:55.2 & 19.69 & 18.78 & 18.10 & 2.061                 & 0.91 & 0.68 & Candidate Crampton \\etal (1992)\\\\ CIRSI10 & 16:36:47.16 & +41:03:35.0 & 20.54 & 19.37 & 18.29 & 1.077                & 1.17 & 1.08 \\\\ CIRSI11 & 16:37:14.80 & +41:12:32.6 & 20.16 & 19.18 & 18.50 & 1.642                & 0.98 & 0.68 & Candidate Crampton \\etal (1992)\\\\ CIRSI12 & 16:35:30.51 & +41:10:41.6 & 20.86 & 19.36 & 18.04 & 1.211                & 1.50 & 1.32 \\\\ CCS88 163447.3+405448 & 16:36:27.11 & +40:48:48.9 & 19.72 & 18.89 & 17.90 & 0.904  & 0.83 & 0.99 & Crampton \\etal (1988)\\\\ CIRSI14 & 16:37:34.03 & +41:16:09.0 & 22.42 & 20.05 & 18.27 & 1.384                & 2.37 & 1.78 & IRX, No UV data\\\\ CIRSI15 & 02:17:56.51 & $-$05:06:51.7 & 20.72 & 19.79 & 18.01 & 1.087              & 0.93 & 1.78 & IRX, low UVX\\\\ CIRSI16 & 02:18:34.44 & $-$05:13:56.9 & 20.95 & 19.04 & 17.32 & 1.351              & 1.91 & 1.72 & IRX, low UVX\\\\ CIRSI17 & 02:17:11.99 & $-$04:46:19.8 & 18.92 & 18.00 & 16.73 & 1.102              & 0.92 & 1.27 & Non UVX\\\\ CIRSI18 & 02:18:08.56 & $-$05:12:23.9 & 20.22 & 19.43 & 18.18 & 1.038              & 0.79 & 1.25 \\\\ CIRSI19 & 02:18:17.42 & $-$04:51:12.5 & 19.41 & 18.19 & 16.68 & 1.085              & 1.22 & 1.51 & IRX \\\\ CIRSI20 & 02:18:13.94 & $-$04:52:41.1 & 19.80 & 18.67 & 17.43 & 1.443              & 1.13 & 1.24 \\\\ CIRSI21 & 02:18:30.57 & $-$04:56:22.7 & 17.60 & 16.61 & 15.61 & 1.401              & 0.99 & 1.00 \\\\ CIRSI22 & 02:17:21.66 & $-$05:06:28.9 & 19.97 & 18.68 & 17.31 & 0.983              & 1.29 & 1.37 & Non UVX\\\\ \\hline \\end{tabular} \\end{center} \\end{table*} \\begin{figure} \\centering \\psfig{file=qso_and_E_bw.ps,width=8cm} \\caption[Comparison of a quasar and elliptical galaxy]{\\label{qso and E}The redshift z=0 quasar and elliptical galaxy spectra used in the discussion are shown against the rest frame filter pass bands.  Both SEDS are normalized to 1 at $\\lambda$4400\\AA\\ , the effective wavelength of the B bands.  The optical ugriz and IR JHK filter sets are shown for comparison.  Quasar - $S_\\nu\\propto \\nu^{\\alpha}$ ($\\alpha$=$-$0.50) with an emission line spectrum based on the Francis \\etal (1991) composite spectrum , Elliptical galaxy - Coleman, Wu and Weedman (1980).} \\end{figure} \\begin{figure} \\psfig{file=cirsi_qso_gzh.ps,width=7cm} \\psfig{file=cirsi_qso_gzh_stargal.ps,width=7cm} \\caption[]{\\label{no stargal} Figure \\ref{plotcd} is reproduced to aid comparison with the identical colour selection diagram but including unresolved and extended objects. Objects with profiles indicative of cosmic rays or image defects are removed.  A limiting magnitude of m$\\rm_g$=21.0 and m$\\rm_H$=18.0 is applied, based on the current spectroscopic survey.  At fainter limiting magnitudes the growth in the number of extended sources and of the region of colour space they occupy would make sample selection without star/galaxy separation prohibitively expensive.} \\end{figure}"
    },
    "0209/gr-qc0209066_arXiv.txt": {
        "abstract": "We experiment with modifications of the BSSN form of the Einstein field equations (a reformulation of the ADM equations) and demonstrate how these modifications affect the stability of numerical black hole evolution calculations.  We use excision to evolve both non-rotating and rotating Kerr-Schild black holes in octant and equatorial symmetry, and without any symmetry assumptions, and obtain accurate and stable simulations for specific angular momenta $J/M$ of up to about $0.9 M$. ",
        "introduction": "\\label{intro}  Binary black holes are among the most promising sources for the gravitational wave laser interferometers currently under development, including LIGO, VIRGO, GEO, TAMA and LISA.  The identification and interpretation of possible signals requires theoretically predicted gravitational wave templates.  For the late epoch of the binary inspiral, numerical relativity is the most promising tool for the computation of such templates.  The numerical simulation even of single black holes has encountered numerous difficulties, which presumably arise from the complexity of Einstein's equations, the existence of a singularity inside the black hole, and the gauge (or coordinate) freedom inherent in general relativity.  Some recent developments, however, have led to significant and very promising advances.  Traditionally, most numerical relativity simulations were based on the $3+1$ decomposition of Arnowitt, Deser and Misner (ADM, \\cite{adm62}), which has been shown to develop instabilities often \\cite{bonc2,baut}. To avoid these instabilities, a number of hyperbolic formulations of Einstein's equations have been developed (see, e.g., \\cite{bonc2,ay99} as well as \\cite{r98} and references therein). Alternatively, Shibata and Nakamura \\cite{shim} and Baumgarte and Shapiro \\cite{baut} introduced a modification of the original ADM equations that involves a conformal-traceless decomposition and the introduction of a new auxiliary variable (see Section \\ref{Bf} below). This formulation, now commonly referred to as the BSSN formulation, has led to significant improvements over the original ADM equations, and has been widely adopted (e.g.~\\cite{baut2,shim2,shim3,shim4,alcm4,alcm5}).  Singularities inside black holes were traditionally avoided by using ``singularity avoiding'' coordinate conditions, including maximal or polar slicing \\cite{smal,eard,barj,plst8586}.  Typically, these conditions lead to grid pathologies that cause codes to crash after relatively short times.  An alternative strategy takes advantage of the fact that the black hole exterior is causally disconnected from the interior, so that a region of the black hole interior including the singularity can be excised from the computational grid.  These ``singularity excision'' techniques \\cite{unrw84,thoj2,seie,thoj3,annp,schm2} have led to large improvements in the simulation of single black holes, and are a promising tool for binary black hole evolutions (see \\cite{bras} for preliminary results).  The application of singularity excision requires a coordinate system that is regular across black hole horizons, allowing smooth horizon penetration.  For single black holes, one such coordinate system is the Kerr-Schild form, which can be used to represent both Schwarzschild and Kerr black holes (and which is also invariant under boosts).  Binary black hole initial data based on Kerr-Schild coordinates can be constructed by solving the constraint equations of general relativity for ``corrections'' arising from superpositions of two boosted Kerr-Schild black holes \\cite{mrhs99,bisn,marp,stp02,pct02}.  Alcubierre and Br\\\"ugmann \\cite{ambb01} recently combined the BSSN formalism with a particularly simple singularity excision method to evolve single black holes in Kerr-Schild coordinates.  Restricting the evolution to octant symmetry, all fields settle down to equilibrium (of the finite-difference equations), and no instabilities are encountered.  If, however, the symmetry assumption is relaxed, instabilities develop and the code crashes after a few hundred $M$. Similar findings were reported in \\cite{s00}, where a completely independent formulation and implementation was adopted.  Improvements over these results were discussed in \\cite{klst01,lpsd02,ls02}, but even these do not completely eliminate the instabilities for evolutions without symmetry assumptions.  Recent results suggest that adding constraints to the evolution equations affects the numerical stability of the system (e.g.~\\cite{klst01,d87,ys01,kllpsst01,ygsh02}; see \\cite{kwb02} for an illustration in electrodynamics).  In this paper we experiment with adding the new constraints that appear in the BSSN formulation to the evolution equation of the new auxiliary functions, and, following \\cite{ygsh02}, the Hamiltonian constraint to the evolution equation for the spatial metric.  We also experiment with schemes for imposing algebraic constraints on the conformally related metric and extrinsic curvature, as well as with different shapes for the excised region inside the black hole.  With these modifications we obtain evolutions of single black holes that last over several thousand $M$, independent of any symmetry assumptions, without encountering any evidence of a growing instability.  These findings hold both for static black holes and for rotating black holes with specific angular momentum of up to $0.9M$.  The paper is organized as follows: We summarize the BSSN formulation in Sec.~\\ref{Bf} and black hole spacetimes in Kerr-Schild coordinates in Section \\ref{idfiksbh}.  Our modifications of the BSSN scheme are described in Sec.~\\ref{mar}.  In Section \\ref{nr} we present results of our simulations for both static and rotating BHs.  We summarize and discuss the implications of our findings in Sec.~\\ref{conc}.  We also include an Appendix that explains our evaluation of the ADM mass and angular momentum.  Throughout the paper we adopt geometrized units with $G=c=1$. ",
        "conclusions": "\\label{conc}  We experiment with various modifications of the BSSN formulation and study their effect on the stability of numerical evolution calculations of static and rotating black holes.  We force the determinant of the conformally related metric to be unity and the trace of the traceless part of the extrinsic curvature to be zero.  We modify the evolution equation for the new auxiliary conformal connection functions by adding their constraint equation, and also experiment with adjustments of the other evolution equations suggested by \\cite{ygsh02}.  Most importantly, we find that an instability that arises when octant symmetry is relaxed \\cite{ambb01,s00,klst01,lpsd02,ls02} can be overcome when the above modifications are employed.  We demonstrate that both static and moderately rapidly rotating black holes can be evolved stably without encountering any growing modes.  Any changes in grid functions settle down to round-off error and remain there for several 1000 $M$.  We find that the dynamically enforced ``Gamma-driver'' spatial gauge condition for the shift leads to more stable evolution than using the analytical shift.  We also find that cubical excision surfaces, which are more straight-forward to implement in Cartesian coordinates, work better than spherical excision surfaces.  While our modifications to not solve all stability problems (e.g.~for the most extreme rapidly rotating black holes), we believe that they lead to significant improvements that may be a helpful step towards simulations of binary black holes and their coalescence."
    },
    "0209/astro-ph0209217_arXiv.txt": {
        "abstract": "{ We present morphological and spectroscopic studies  of the northwest rim of the supernova remnant RX~J1713.7$-$3946 based on observations  by  the \\emph{Chandra X-ray observatory}. We found a complex network of nonthermal (synchrotron) X-ray filaments, as well as a 'void' type structure -- a dim region of a circular shape -- in the northwest rim. It is remarkable that  despite distinct brightness variations, the X-ray spectra everywhere in this region can be well fitted with a power-law model with photon index ranging $\\Gamma =$ 2.1--2.5. We briefly discuss some  implications of these results and argue  that the resolved  X-ray features in  the northwest rim may challenge the perceptions  of the standard diffusive shock-acceleration models concerning the  production, propagation and radiation of relativistic  particles in supernova remnants. ",
        "introduction": "The supernova remnant (SNR) RX~J1713.7$-$3946 has proven to be a prominent source of nonthermal X-rays and presumably also $\\gamma$-rays, thus providing strong evidence that  shell-type SNRs are sites of acceleration of galactic cosmic rays. This source was  discovered during the \\emph{ROSAT} All-Sky Survey (Pfeffermann \\& Aschenbach \\cite{pfeffermann}). Observations with \\textit{ASCA} have revealed intense synchrotron X-ray emission from the entire remnant (Koyama et al.\\ \\cite{koyama97}; Slane et al.\\ \\cite{slane}). Unlike SN~1006, no evidence for thermal X-ray components has yet been found.   At the north perimeter of RX~J1713.7$-$3946, a molecular cloud (cloud~A) has been  found with a high CO\\,($J$=2--1)/CO\\,($J$=1--0) ratio that suggests possible interaction between  the cloud and the SNR shell (Butt et al.\\ \\cite{butt01}). The cloud has positional association with the unidentified $\\gamma$-ray source 3EG J1714$-$3857 (Butt et al.\\ \\cite{butt01}). Most recently, Uchiyama et al.\\ (\\cite{uchiyama02})  reported an unusually flat-spectrum X-ray source (AX J1714.1$-$3912) coincident with this cloud, and  argued that the flat spectrum can be best interpreted by bremsstrahlung from either sub-relativistic protons or mildly relativistic electrons. The high-energy particles responsible for the X-ray and $\\gamma$-ray emission from this cloud are likely associated (in one way or another) with the SNR-cloud  interaction. For the preferred distance to SNR RX~J1713.7$-$3946 of $d \\simeq 6$~kpc (Slane et al.\\ \\cite{slane}), based on the kinematic distance to cloud~A, the age of the remnant is estimated to be $\\gtrsim 10 \\,000$~yr. A younger age of $\\sim 2000$~yr cannot be, however, excluded if $d=\\mbox{1--2}$~kpc.  The CANGAROO collaboration (Muraishi et al.\\ 2000) reported the detection of TeV $\\gamma$-ray emission from the direction of the northwest (NW) rim,  the brightest region of synchrotron X-rays in RX~J1713.7$-$3946. If confirmed, the TeV $\\gamma$-radiation would provide direct and  unambiguous evidence for the presence of particles (electrons and/or protons)  accelerated to very high energies.  Recently Enomoto et al.\\ (\\cite{enomoto}) published the spectrum of TeV emission based on the new CANGAROO observations. The spectrum  is claimed to be quite  steep with  a  power-law photon index $\\Gamma = 2.8\\pm 0.2$ between 400 GeV and 8 TeV. They argued that the steep spectrum  is inconsistent with the inverse  Compton (IC) model,  but could be explained by $\\pi^0$-decay gamma-rays. If true,  for the canonical shock-acceleration spectrum of protons with power-law index $s \\sim 2$, this would imply  a cutoff energy in the proton spectrum around 100 TeV. Subsequently, Reimer \\& Pohl (\\cite{reimer}) and Butt et al.\\ (\\cite{butt02}) argued that this interpretation would violate the $\\gamma$-ray flux upper limits set by the EGRET instrument. However, this is not a sufficiently robust argument  to be used to dismiss the hadronic origin of the reported TeV emission. Adopting a slightly harder proton spectrum, e.g. with spectral index $s \\leq 1.9$, it is possible to avoid the conflict with the EGRET data. Even for proton spectra steeper than $s=2$, it is still possible to suppress the GeV $\\gamma$-ray flux, if one invokes the effects of energy-dependent propagation of protons while traveling from the  accelerator (SNR shocks) to the nearby clouds (see e.g. Aharonian \\cite{aharonian01}). Moreover, the lack of GeV $\\gamma$-rays can be naturally explained by confinement of low-energy (GeV)  protons in the supernova shell, in contrast to the effective  escape of high-energy (TeV) protons.  On the other hand, the arguments against the IC model of TeV emission should be backed by thorough theoretical studies based on higher quality data from the radio, X-ray and $\\gamma$-ray domains. X-ray observations are of particular interest   because the synchrotron X-ray spectra reliably ``control''  the predictions of IC emission at TeV energies. In this paper, we report on the X-ray study of the NW rim of SNR RX~J1713.7$-$3946 using archival data obtained with the \\emph{Chandra X-ray Observatory} (Weisskopf et al.\\  \\cite{weisskopf96}). We demonstrate that the X-ray emission has remarkable substructure with  bright filaments accompanied by curious dark voids. The observed features set new standards in X-ray studies of SNRs which should help to understand deeper  the nonthermal processes  of particle acceleration,  propagation and radiation in supernova shocks. ",
        "conclusions": "The \\emph{Chandra} image has revealed that the synchrotron X-ray emission from the northwestern rim of SNR RX~J1713.7$-$3946 has remarkable fine-structure: the complex network of synchrotron X-ray filaments surrounded by fainter diffuse plateau, and a dark region with a circular shape. By examining individual spectra, we found that despite significant brightness variations, the spectral shapes of the X-ray spectra everywhere in this region are more or less similar, being well fitted with a power-law model of photon index $\\Gamma \\simeq$ 2.3.  The observed hard power law requires rather high synchrotron cutoff frequency (energy), set by the condition ``acceleration rate = synchrotron loss rate'', which is most likely to be $\\gtrsim10$ keV taking account of the effects of spectral steepening due to synchrotron losses. We need unreasonably high shock speed exceeding 5000 km s$^{-1}$ to explain such a high cutoff energy within the standard formalism of the diffusive shock acceleration model. A possible solution to this difficulty could be obtained if one assumes low magnetic field in the acceleration sites, and high magnetic field in the emission regions. Otherwise we should invoke faster, as yet unknown, electron acceleration mechanism.  We testify the scenario that the filaments/hot-spots are the acceleration sites of bulk of multi-TeV electrons without specifying acceleration mechanism, by adopting a ``two-zone'' model. Then we found that our time-dependent treatment including the effects of fast energy losses and diffusive escape of electrons is capable of  accounting for the fluxes and spectral shapes of X-ray emissions from both the filament and plateau regions. We note that the TeV spectrum reported by the CANGAROO collaboration cannot be readily explained by the IC scattering of the CMBR photons off X-ray-emitting multi-TeV electrons. Higher quality data from the radio, X-ray and $\\gamma$-ray bands should help us to draw more definitive conclusions concerning the nature of acceleration mechanisms and the origin of the observed TeV $\\gamma$-rays."
    },
    "0209/astro-ph0209571_arXiv.txt": {
        "abstract": "We have analyzed {\\it JHKL\\,} observations of the stellar population of the $\\approx 9$ Myr-old $\\eta$ Chamaeleontis cluster. Using infrared (IR) colour-colour and colour-excess diagrams, we find the fraction of stellar systems with near-IR excess emission is $0.60 \\pm 0.13$ (2$\\sigma$).  This results implies considerably longer disc lifetimes than found in some recent studies of other young stellar clusters.   For the classical T Tauri (CTT) and weak-lined T Tauri (WTT) star population, we also find a strong correlation between the IR excess and H$\\alpha$ emission.  The IR excesses of these stars indicate a wide range of star-disc activity; from a CTT star showing high levels of accretion, to CTT -- WTT transition objects with evidence for some on-going accretion, and WTT stars with weak or absent IR excesses.  Of the 15 known cluster members, 4 stars with IR excesses $\\Delta(K-L) > 0.4$ mag are likely experiencing on-going accretion owing to strong or variable optical emission. The resulting accretion fraction ($0.27 \\pm 0.13$; 2$\\sigma$) shows that the accretion phase, in addition to the discs themselves, can endure for at least $\\sim 10$ Myr. ",
        "introduction": "Circumstellar discs are a natural by-product of the star formation process (e.g. Shu, Adams \\& Lizano 1987). Lada et al. (2000) found that 97 percent of the optical proto-planetary discs in the Trapezium cluster exhibit excess in the {\\it JHKL\\,} colour-colour diagram, indicating that the most likely origin of the observed IR excesses are the circumstellar discs.  The {\\it disc lifetime\\,} derived by examining the fraction of IR excess stars (the {\\it disc fraction\\,}) of young stellar populations as a function of age provides an empirical limit on the duration of the disc accretion phase.  This is critical for understanding the evolutionary paths followed by PMS stars in the HR diagram, their angular momentum histories, and the timescales available for planet building (e.g. Hartmann et al. 1998; Hillenbrand et al. 1998; Telesco et al. 2000).  Considerable debate has waged concerning the longevity of  discs around low mass PMS stars.  The effort is difficult because samples are often incomplete and biased:  CTT stars are typically located by H$\\alpha$ and near-IR excess surveys, while WTT stars are mostly found through X-ray surveys (Feigelson \\& Montmerle 1999).  An influential early study on the timescale for disc dissipation by Strom et al. (1989), based on several dozen PMS stars in the Taurus-Auriga complex, reported the CTT/WTT transition occurs around an age $t \\approx 3$ Myr. This result is supported by a recent {\\it JHKL\\,} survey of 7 clusters with mean ages from 0.5 to 5 Myr that shows half the stars lose their discs within 3 Myr and essentially all lose their discs in 6 Myr (Haisch, Lada \\& Lada 2001).  At a later age of $t \\simeq 13$ Myr, only 1/110 Sco-Cen late-type stars show spectroscopic CTT emission lines and $K$-band excesses (Mamajek, Meyer \\& Liebert 2002).  However, other studies suggest discs are more enduring.  No evolution in disc fraction is found in the stellar populations of the Orion Nebula Cluster from $t < 0.1-2$ Myr (Hillenbrand et al. 1998), and in NGC 2264 from $t < 0.1-5$ Myr (Rebull et al. 2002). The Chamaeleon I cloud population, where the sample is enhanced through X-ray and {\\it ISO\\,} surveys, shows no difference in the age distribution of CTT and WTT stars from $t < 1-10$ Myr (Lawson, Feigelson \\& Huenemoerder 1996).  Spectroscopic study of the $\\sim 10$ Myr-old TW Hydrae Association members found active accretion in 2 stars; TW Hya and Hen 3-600A (Muzerolle et al. 2000), albeit at mass accretion rates $1-2$ magnitudes lower than is derived for younger accreting systems.  The measurement of disc lifetimes is muddied by several other issues. First, despite the broad overlap in CTT and WTT age distributions, few intermediate cases are known suggesting that the transition between the two states is rapid (Wolk \\& Walter 1996).  Secondly, some studies suggest that two disc lifetimes must be considered; one associated with a rapid decline in accretion onto the star and another associated with a slower dissipation of the outer disc (Clarke, Gendrin \\& Sotomayor 2001).  Third, efforts to reconcile rotational distributions of both PMS and zero-age main sequence stellar clusters assuming star-disc rotational coupling during the accretion phase have deduced relatively long $10-15$ Myr disc lifetimes for slowly rotating stars (Bouvier, Forestini \\& Allain 1997; Tinker, Pinsonneault \\& Terndrup 2002), but $\\sim 3$ Myr disc lifetimes may suffice if the stellar interiors have sufficient radial differential rotation (Barnes, Sofia \\& Pinsonneault 2001).  Finally, the astrophysical processes leading to disc dissipation are highly uncertain.  Possible mechanisms include accretion onto the star, bipolar outflow, stellar winds, photoevaporation, close gravitational encounters and incorporation of disc material into planets (see the review by Hollenbach, Yorke \\& Johnstone 2000).  The recently discovered $\\eta$ Chamaeleontis cluster has the potential of clarifying some of the observational issues.  It is a nearby ($d \\simeq 97$ pc),  compact (extent $\\sim$1 pc) and coeval ($t = 9 \\pm 1$ Myr; Lawson \\& Feigelson 2001) system of  PMS stars with a small (15 known primaries) population of stars spanning a relatively large range in mass ($M = 0.2-3.4$ M$_{\\odot}$; Mamajek, Lawson \\& Feigelson 1999, 2000;  Lawson et al.\\ 2001, 2002).  The cluster includes $\\eta$ Cha (spectral type B8), HD 75505 (A5), the A7+A8 binary and $\\delta$ Scuti system RS Cha, 11 WTT stars (K5$-$M4) and 1 CTT star (M2). Importantly, this census is known to be virtually complete in the inner region from a combination of a deep $ROSAT$ exposure, optical photometry and proper motion study of the field. Unlike most other PMS populations, these membership criteria are independent of the presence or absence of  circumstellar discs. We can thus use the $\\eta$ Cha population as an unbiased laboratory to investigate the fraction of IR excess and accretion at a critical intermediate-age phase of disc evolution. ",
        "conclusions": "Considerable uncertainty reigns concerning the longevity, or more likely, the distribution of longevities, of circumstellar discs (Section 1).  Despite its modest population, the $\\eta$ Cha cluster provides a rare opportunity to examine -- at high sensitivity -- disc properties of PMS stars with intermediate ages whose selection is unbiased with respect to disc existence\\footnote{The importance of the $\\eta$ Cha cluster in this respect is recognized by its inclusion in the first year guaranteed time programme of the {\\it Space Infrared Telescope Facility}.}.  If discs decay rapidly as indicated by some past studies, then no discs at all are expected in this cluster.  We find, however, that 9/15 or 60 percent of $\\eta$ Cha primaries show IR excesses in the $(H-K)/(K-L)$ diagram (the late-type stars identified in Figure 2, plus $\\eta$ Cha and HD 75505).  The excess cannot be attributed to errors in photometry (Section 2), binarity (Section 3.4.1) or reddening (Section 3.4.2).  Long-lived circumstellar discs are the only plausible explanation.  One of these stars, ECHA J0843.3-7905, is a CTT star with active accretion (Lawson et al.\\ 2002), and the H$\\alpha - \\Delta(K-L)$ correlation seen in the late-type population suggests that up to 3 other stars may be accreting (see Figure 3).  High-resolution spectroscopic study now underway will address this issue.  Why do we find a high disc fraction at $t \\simeq 9$ Myr when some other studies find discs largely disappear by $t = 3-6$ Myr?  We first recognize that, except for ECHA J0843.3-7905, the $\\eta$ Cha discs would have been mostly missed from {\\it JHK\\,} colours alone (see Figure 1a).  Sensitive $L-$band surveys are essential for the detection of aging PMS discs.  The principal discrepancy among $L-$band studies lies between our high disc fraction (9/15 or 0.60) in $\\eta$ Cha and the low disc fraction (9/75 or 0.12) for the 5 Myr old cluster NGC 2362 found by Haisch et al. (2001).  We suggest several explanations for this difference.  First, the assigned age of NGC 2362 PMS stars relies solely on the turnoff age of the O9Ib supergiant $\\tau$ CMa (Balona \\& Laney 1996) and the assumption that all stars in the cluster are coeval.  Secondly, NGC 2362 is 1480 pc distant compared to 97 pc for $\\eta$ Cha.  The distance ratio alone degrades the $L$-band sensitivity by a factor of 200.  Because of this, faint discs in NGC 2362 may not have been detected.  Also, the limiting mass of the Haisch et al. (2001) study of NGC 2362 is $M \\approx 1$ M$_{\\odot}$ (spectral type mid-K) compared to $M \\approx 0.2$ M$_{\\odot}$ (spectral type M4) in our study of $\\eta$ Cha.  As we discuss in Section 3.1, it is easier to detect an IR excess in a late-M star, compared to a K-type star, using standard IR colour-colour plane analysis. Also, it is possible that disc lifetimes are shorter in higher mass stars. Third, there might be variance in the disc destruction rate amongst clusters due to different rates of close encounters or photoevaporation due to massive stars, e.g. the discs in NGC 2362 might have been stripped by the UV/wind of $\\tau$ CMa (see Hollenbach et al. 2000 for theory).  A combination of the above factors can explain why there is a $2-4$ dispersion in the observed disc fraction for PMS star clusters of a similar age (Hillenbrand \\& Meyer 1999).  While noting these differences, our results seen together with studies of other older nearby PMS stars (e.g. study of the TW Hydrae Association members by Muzerolle et al. 2000) indicate that IR-detected discs can be present in $\\sim 60$ percent, and accretion discs can be present in $\\sim 30$ percent, of $\\sim 10$ Myr-old PMS stars."
    },
    "0209/astro-ph0209092_arXiv.txt": {
        "abstract": "We present further observations of the extremely red object ERO~J094258+4659.2, identified by \\citet{iye00} as ERO R1 in their deep images of the cluster A851. We estimate its redshift independently by eight-band photometric redshift determination and cross-correlation of a new $H$-band spectrum with the optical spectra of local E/S0 galaxies, and conclude that it lies at $z \\sim 1.5$. Although its colors are consistent both with an elliptical galaxy and an S0 galaxy at that redshift, its elongated shape and exponential luminosity profile suggest the presence of an evolved stellar disk component. We rule out the possibility that these properties are strongly influenced by gravitational lensing by the foreground cluster, and therefore conclude that this object is more likely to be an S0-like galaxy, rather than a lensed elliptical.  The $H$-band spectrum does not show strong H$\\alpha$ emission and the star formation rate therefore appears to be very modest.  The presence of such a galaxy with an apparently relaxed disk of stars at this high redshift provides a new and strong constraint on theoretical models which aim to explain the formation and evolution of galaxies. ",
        "introduction": "The cluster of galaxies A851 (= Cl~0939+4713) at $z=0.4$ was observed during the first light period of Subaru Telescope using the optical imager, Suprime-Cam, and the infrared imager, CISCO, both mounted at the Cassegrain focus \\citep{iye00}. Five EROs with $R-K' > 5$ were found in the field of the cluster, and the object identified as ERO R1 was noted for its outstanding color ($R-K'=7.4$) and elongated shape ($b/a = 0.4$) with its major axis aligned with the equipotential surface of the cluster.  This object, formally named to as ERO~J094258+46592 (but referred to as ERO~R1 throughout this paper for brevity), was first noted in the literature by \\citet{sma99} as object \\#333 but did not receive special mention. In Figure~\\ref{fig:images} we present postage stamp images of ERO R1 in various bands.  Preliminary analysis by \\citet{iye00} from its \\textit{RJK$'$} magnitudes suggested that ERO~R1 is likely to be a passively-evolving early-type galaxy at $1.0<z<1.6$ with modest dust extinction ($E(B-V) < 0.5$).  The direction of the galaxy's major axis suggested that it might be the gravitationally lensed image of a rounder background galaxy, although the smooth and symmetrical shape argued in favor of intrinsic elongation. \\citet{iye00} therefore suggested that ERO~R1 might be an S0 or disk galaxy at high redshift.  The existence of early-type galaxies with dynamically-relaxed stellar disks, as ERO~R1 appears to be, is an important issue for galaxy formation models to deal with. Since few EROs have been studied in detail, we have made further observations of this object in an attempt to determine its properties unambiguously and determine its relevance and importance in our understanding of the ERO phenomenon and galaxy formation. We report on the results of these observations in this paper. Unless otherwise stated, we adopt $H_0 = 70\\rm\\,km\\,s^{-1}\\,Mpc^{-1}$, $\\Omega_{\\rm m}=0.3$, and $\\Omega_\\Lambda=0.7$. ",
        "conclusions": "\\subsection{Clustering at $z=1.5$?}  \\citet{dad00} found a strong clustering signal in their sample of 400 EROs and argued as a result that the ERO population is dominated by high-redshift elliptical galaxies, rather than dusty starbursts. Since ERO~R1 appears to be a member of the class of passively-evolving early-type galaxies, it may belong to some element of large-scale structure and therefore there may be an overabundance of galaxies at $z \\sim 1.5$. The Suprime-Cam imaging of Cl~0939+4713 by \\citet{kod01} was intended to study the large-scale structure in the distribution of galaxies, and a preliminary (unpublished) photometric redshift analysis based on their \\textit{BVRI\\/} data shows a slight excess of galaxies at $z=1.5$. However the excess is very modest and the uncertainty in photometric redshifts determined solely from optical data is rather large at $z>1$. Further studies, including wider-field infrared imaging, are clearly needed before a firm statement about the presence or absence of a distant cluster can be made.   \\subsection{Evaluation of possible lensing effects}  ERO~R1's major axis is almost perpendicular to the direction toward the surface mass density peak of the cluster as derived from the weak lensing analysis of \\citet{iye00}. This led \\citet{iye00} to suggest that the elongated nature of ERO~R1 could be due to image shear of an intrinsically rounder galaxy, caused by the strong gravitational field of the intervening cluster. \\citet{sei96} performed a weak-lensing analysis of Cl~0939+4713 and found that the cluster mass distribution closely traces the distribution of bright member galaxies. High-resolution X-ray observations by \\citet{sch98} confirmed the presence of concentrations of hot plasma consistent with the distribution of mass and galaxies, and indicate that this cluster is still actively evolving towards a virialized state.  The proximity of ERO~R1 to the cluster center, whether defined as the optical center ($\\alpha=09^{\\mathrm h}42^{\\mathrm m}56\\fs9$, $\\delta=+46^{\\circ}59\\arcmin23\\farcs1$, 21\\arcsec\\ from R1) or the X-ray peak M1 ($\\alpha=09^{\\mathrm h}42^{\\mathrm m}58\\fs2$, $\\delta=+46^{\\circ}58\\arcmin52\\arcsec$, 20\\farcs4 from R1) makes it conceivable that the object's morphology suffers considerable gravitational distortion. In order for the light from a galaxy at a projected distance $r$ from the cluster center to be strongly lensed, a lensing mass with radius $r$ of $M(<r) > \\pi \\sigma_{\\mathrm cr} r^2$ is required, where the critical surface mass density, $\\sigma_{\\rm cr}$, depends on the distances to the lens and the background source, as well as the cosmological model. Since we know the redshifts to the source ($z=1.5$) and lens ($z=0.4$), we can calculate the required lensing mass, which is $M(<r) = 1.4 \\times 10^{16} r_{\\mathrm Mpc}^2 M_\\odot$, or $3 \\times 10^{13} M_{\\odot}$ for $r = 44$\\,kpc, assuming the center of the cluster to be the location where the reconstructed surface mass density peaks (see Figure~13 of \\citealt{iye00}).  Figure~\\ref{fig:massprof} shows the integrated radial mass profile of the cluster derived from the weak lensing analysis using the method described in \\citet{kai95}. Adopting the surface mass density peak as the center of the cluster, we have performed the calculation for two values of the mean redshift of the background galaxies, $z_{\\mathrm s}=1.0$ and $z_{\\mathrm s}=3.0$. Although we cannot evaluate exactly the value of $z_{\\mathrm s}$, it is likely to be closer to 1.0 than 3.0. The integrated mass of the cluster CL~0939+4713 was estimated using X-ray observation by Schindler et al.(1998) and they deduced a mass of $\\sim 1.5 \\times 10^{13} M_\\odot$ at 100 kpc from the X-ray peak M1 whose location coincides with the mass peak deteced by lensing. The present mass estimate from lensing, as is shown in Figure 7, is consistent with their result.  From this figure, the mass wihin the projected distance to ERO~R1 is at most $(1.0 \\pm 0.4) \\times 10^{13} M_{\\odot}$, or a factor of 3 lower than the mass required for strong lensing. We made additional estimates of the mass after displacing the assumed cluster center in four directions, but found that the result was changed by less than 30\\%. Note in particular that if the cluster center is at the X-ray peak M1, the distance to R1 increases and the required mass becomes even larger. We therefore conclude that the cluster mass is insufficient to strongly lens the light from ERO~R1.  Since the distance of ERO R1 from the cluster center is relatively small, the image, if lensed, would show distinct bending, whose radius of curvature is approximately equal to the distance of ERO R1 from the lens center. The observed image of ERO R1 does not show any  expected  curvature and hence also is not consistent with  strong lensing interpretation.  Although ERO~R1 is not strongly lensed, the large mass of the intervening cluster may affect its observed shape, and we have attempted to reconstruct its unlensed image. We model the cluster as an isothermal sphere, using Schindler et al.'s (1998) estimate of the total mass, for five different assumed locations of the center of mass. Figure~\\ref{fig:unlensed} (Case~0) presents the unlensed image if the center of mass is located at the X-ray peak M1. In Figure~\\ref{fig:unlensed4}, we present four additional images where we have displaced the center of mass by 10\\arcsec\\ (equal to the smoothing applied to the X-ray image) from M1; we refer to these as Cases N, W, E, and S, according to whether the displacement was made to the north, west, east, or south, respectively. The calculation for Case~N requires data in an unobserved region, and we assigned the lowest observed surface brightness to this region. We constructed the luminosity profiles of the five unlensed images and confirmed that they were all well-fit with an exponential disk profile, as expected. ERO~R1 therefore possesses a significant disk component and its elongated shape is not due to gravitational lensing of a rounder elliptical galaxy.  \\subsection{Implications for the early formation of an S0 galaxy}  Our interpretation of an apparently relaxed, massive S0 galaxy at $z=1.5$ is made difficult since we do not know whether it lies in a cluster or in the field. The luminous early-type populations of rich clusters are known to be remarkably homogeneous and the bulk of their stellar populations appear to be old \\citep{bow92, ell97}. \\citet{jon00} found from their spectroscopic data that there is no statistically significant difference between the luminosity-weighted age of the E and S0 galaxies in clusters at $0.37 < z < 0.56$, supporting the conventional formation model of old, coeval E and S0 galaxies. In nearby clusters, the density--morphology relation and the formation of S0 galaxies are often related to the effects of ram pressure stripping \\citep{dre80}. This is unlikely to be the cause of ERO~R1's morphology, however, since it is not clear whether ram pressure stripping could occur effectively at such an early cosmic epoch, and our optical--infrared images provide no evidence for the existence of a rich cluster. We note that \\citet{soi99} identified another ERO in the field of Cl~0939+4713 (named Cl~0939+4713~B) as a passively-evolving galaxy at $z=1.58$ (the redshifts of ERO~R1 and Cl~0939+4713~B are marginally consistent given the uncertainties), although this is $\\gtrsim 700\\,$kpc from R1 and therefore cannot be said to belong to the same `cluster'.  It is far more likely, therefore, that ERO~R1 is simply a field galaxy. A morphological analysis of galaxies in the Hubble Deep Fields by \\citet{rod01} led these authors to conclude that massive E/S0 galaxies tend to disappear from flux-limited samples at $z>1.4$. They therefore suggested that the era $1<z<2$ corresponds to a very active phase for the assembly of massive E/S0 galaxies in the field.  \\citet{smi02} studied the gravitationally lensed ERO~J003707+0909.5 at $z=1.6$ in the field of the cluster A68. From their reconstructed image and photometric analysis, they concluded that this object is an $L^*$ early-type disk-galaxy, and therefore very similar to ERO~R1. In addition to the known classes of EROs, i.e., dusty starbursts such as HR~10 \\citep{hu94,gra96,dey99} and passively-evolving ellipticals, \\citeauthor{smi02} suggested that a significant fraction of EROs may evolve into luminous spiral galaxies by the present epoch.  If the absence of high over density of galaxies around R1 is confirmed, some constraints will be placed as to the physical mechanisms of forming S0 galaxies.  As already discussed, ram-pressure stripping will not be a dominant factor and instead a mechanism which can be effective in lower density regions will be required. This is consistent with the recent findings that the truncation of star formation in galaxies is not the phenomenon restricted to the dense cluster cores but is wide spread to relatively low density regions (e.g., Kodama et al.2001; Balogh et al.1999; 2002, Lewis et al.2002; Gomez et al.2002)."
    },
    "0209/astro-ph0209437_arXiv.txt": {
        "abstract": "I describe the principle of using the 1.6\\um\\ H$^-$ spectral feature as a photometric redshift indicator and demonstrate that the technique holds promise by successfully recovering the redshifts of a small sample of $z$=0--1 galaxies using only their infrared ($JHKL$) photometry.  I then consider the applicability of the technique to the 3.6--8\\um\\ SIRTF filter set and investigate the systematic errors that could arise in photometric redshifts from random photometric errors or from a mismatch between target galaxies and fitting templates in metallicity, star formation history, and amount of interstellar dust. It appears that SIRTF near-IR data alone should be sufficient to estimate redshift of most galaxies that are at $z$\\gap 1.5 and are dominated by stellar populations older than \\gap 20Myr.  Galaxies whose photometric fits indicate them to be at lower redshifts, \\zfit\\lap1.5, or dominated by very young stellar populations, \\agefit\\lap20Myr, suffer from severe degeneracies in photometric redshift, and a reliable photometric determination of their redshifts will have to include either IR observations at shorter wavelengths ($H$ and $K$) or optical data.  Overall, it appears that --- with care and caveats --- the 1.6\\um\\ bump can provide a powerful way of estimating redshifts of distant galaxies in deep infrared imaging surveys that will soon be provided by SIRTF, and, eventually, by NGST. ",
        "introduction": "The study of galaxy formation and evolution will gain an enormous boost from space-borne observations in the infrared (IR), first with the advent of the Space Infrared Telescope Facility (SIRTF) and then, at the end of this decade, with the appearance of the Next Generation Space Telescope (NGST).  SIRTF, scheduled for launch in early 2003, will allow us to efficiently observe distant galaxies at mid- and far-infrared wavelengths.  In particular, SIRTF's Infrared Array Camera (IRAC) will operate at 3.6\\um, 4.5\\um, 5.8\\um, and 8\\um, providing a window on the rest-frame near-IR properties of galaxies at $z$\\gap1.  A number of deep SIRTF imaging surveys that will detect many thousands of galaxies to high redshift are already scheduled (e.g., GOODS\\footnote{see http://www.stsci.edu/science/goods/}, SWIRE\\footnote{see http://www.ipac.caltech.edu/SWIRE/}), and it will be enormously important to efficiently estimate redshifts of galaxies in these samples.  However, large-scale spectroscopic follow-up of these IR-selected surveys will be very time-consuming and only small sub-samples of objects are likely to be targeted.  Instead, spectroscopic redshifts are likely to be complemented by the less precise but much more efficient photometric redshifts.  Photometric redshifts enjoy great popularity as a way to estimate the redshifts of faint galaxies (e.g., Connolly et al.\\ 1995; Gwyn \\& Hartwick 1996; Sawicki, Lin, \\& Yee 1997; Fern\\'andez-Soto, Lanzetta, \\& Yahil 1999; Rudnick et al.\\ 2001).  Traditionally, photometric redshifts rely on the photometric signatures of the Lyman and 4000\\AA\\ breaks to estimate redshifts from optical and near-IR data.  However, the availability of mid-IR data from SIRTF (and, later, NGST) surveys opens up the possibility of using a hitherto untried photometric redshift technique that could obviate the need for optical imaging of SIRTF and NGST survey fields, or provide independent redshift estimates where such imaging is already present.  Specifically, the spectral ``bump'' at 1.6\\um, caused by the minimum in the opacity of the H$^-$ ion present in the atmospheres of cool stars, is a nearly ubiquitous feature of almost all stellar populations and the photometric signature of this bump can be expected to provide a means of estimating redshifts to distant galaxies.  The importance of the imprint of the \\bump\\ on SIRTF IRAC colors has been appreciated for a long time (e.g., Wright, Eisenhardt, \\& Fazio 1994), and indeed the filter transmission curves of the IRAC filters have been designed partially with photometric redshifts in mind by modeling the expected colors of high-$z$ galaxies (Simpson \\& Eisenhardt 1999).  The purpose of the present paper is to demonstrate the potential of the 1.6\\um\\ bump as a photometric redshift indicator and to explore some of the possible limitations of the technique at high redshift.  Section~\\ref{lowz} illustrates, with real data at 0$<$$z$$<$1, that the 1.6\\um\\ bump {\\it is} an effective photometric redshift indicator and describes the source of the photometric redshift signal.  Section~\\ref{sirtf} discusses the applicability of the technique to SIRTF IRAC data and explores some of the limitations that could be present at high redshift due to effects such as metallicity, dust, and star formation history.  The main conclusions of the paper are summarized and discussed in Section~\\ref{summary}. ",
        "conclusions": "\\label{summary}  This paper considers the viability of the \\bump\\ as a photometric redshift indicator.  In \\S\\ref{HDF} it was shown that $JKHL$ photometry can be used to estimate the redshifts of $z$$<$1 galaxies, thus demonstrating with real data that the \\bump\\ is a viable photometric redshift indicator.  The analysis of the dependence of galaxy colors on redshift (\\S\\ref{source_of_signal}) leads to the conclusion that photometric redshift signal at a given redshift comes primarily from two filters: an ``anchor'' filter which at all redshifts of interest remains in the invariant part of the spectrum redward of the \\bump, and a second filter whose redshifted bandpass corresponds to 1.6\\um.  To cover an extensive redshift range and to reduce redshift degeneracies, one needs --- in addition to the long-wavelength anchor filter that is always redward of (1+$z$)1.6\\um\\ --- multiple filters that span the possible range of redshifted 1.6\\um.  Filters that are located blueward of the redshifted \\bump\\ do not constrain redshift well, but instead provide information about the age of the stellar population.  In this manner, the $HKL$ filter set was used to determine redshifts of galaxies at $z$$\\approx$0--1 (the $J$ filter does not provide redshift information but could be used as an age indicator), while the SIRTF IRAC \\iracfilters\\ filters can be expected to work at higher redshifts.  Simulations (\\S\\ref{photometricerrors}) suggest that SIRTF IRAC colors alone may be sufficient to measure photometric redshifts of high-$z$ galaxies, provided that care is taken to reject the degenerate objects whose fits claim them to be either at $z$\\lap 1.5 or younger than $\\sim$20 Myr (these redshift and age cut-offs can be adjusted depending on the levels of completeness and contamination that are acceptable).  At higher redshifts ($z$\\gap 2.5) low-$z$ contaminants can be separated from young high-$z$ galaxies by the addition of UV-optical data that are sensitive to the presence of the Lyman break; at $z$\\lap 1.5, the accuracy of photometric redshifts would be restored with the addition of $H$ and $K$ bands.  As was illustrated in \\S\\ref{systematics}, the ability to determine photometric redshifts from IRAC photometry should not be significantly compromised by the unknown metallicities, dust content, and star formation histories of high-$z$ galaxies.  Provided the caveats about rejecting galaxies with \\zfit \\lap 1.5 or \\agefit \\lap 20Myr are followed, SIRTF-based photometric redshifts should provide a powerful way to estimate the redshifts of distant galaxies.  With SIRTF IRAC data alone, photometric redshifts can be expected to be most accurate for galaxies at $z$$\\approx$2, with the accuracy decreasing to higher redshifts; they can be expected to be completely inaccurate for galaxies at $z$\\lap 1.5.  The epoch at $z$$\\approx$2 is the transitional era where the cosmic star formation appears to plateau (e.g., Sawicki, Lin, \\& Yee 1997; Steidel et al.\\ 1999) after its rapid rise with lookback time over 0$<$$z$\\lap 1 (Lilly et al.\\ 1996); the expected high accuracy of IRAC photometric redshifts at $z$$\\approx$2 means that SIRTF surveys are well-poised to study galaxies at this hitherto unexplored epoch of transition.  Although the \\bump\\ can be used to provide a good estimate of redshift from SIRTF data, particularly at $z$$\\approx$2, ages of stellar populations cannot be determined with great precision from SIRTF data alone at any redshift: large scatter in \\agefit\\ can be expected in the presence of even small photometric uncertainties (\\S\\ref{photometricerrors}), while significant systematic effects will be present due to the unknown metallicity, star formation history, and dust content of the target galaxy (\\S\\ref{systematics}).  The addition of optical and near-IR data could alleviate some of these problems: while SIRTF data are used to estimate redshifts, near-IR data that span the age-sensitive rest-frame 4000\\AA\\ break can be added to estimate ages, and dust effects can be constrained in the rest-UV with optical data (see, e.g., Sawicki \\& Yee, 1998).  The addition of spectroscopic redshifts might then free the SIRTF photometry to be used to better constrain star formation histories and/or metallicities of distant galaxies.  It is beyond the scope of this paper to explore what constraints can be placed on the evolutionary parameters of high-$z$ galaxies by multiwavelength photometry augmented by spectroscopic redshifts; however, such studies that combine spectroscopy with photometry that ranges from rest-frame UV to IR are sure to be an important path to understanding the nature of distant galaxies.  Although \\S\\ref{HDF} illustrates that reliable photometric redshifts are possible using the \\bump, the essential test of the \\bump\\ as a photometric redshift indicator for SIRTF will come only in the form of a comparison between spectroscopic redshifts and photometric ones based on SIRTF data.  The \\bump\\ technique is expected to be most useful at $z$$\\approx$2 --- precisely where our ability to determine spectroscopic redshifts is limited by the lack of readily accessible strong spectral features.  However, we should be able to reliably assemble large samples of $z$$\\approx$2 galaxies by first using photometric redshifts to select $z$$\\approx$2 candidates, then obtaining spectroscopic redshifts using deep spectroscopy that targets absorption lines that are weak but accessible in the optical (e.g., \\ion{C}{4} 1550, \\ion{Mg}{2} 1670), and finally using the spectroscopic redshifts to test and calibrate the photometric redshifts.  The \\bump\\ photometric redshift technique applied to SIRTF (and, later, NGST) data appears to be ideally suited to allow a detailed study of galaxies at the crucial epoch around $z$$\\approx$2."
    },
    "0209/astro-ph0209601_arXiv.txt": {
        "abstract": "{This letter presents a numerical study of the evolution of an emission line cloud of initial density 10~cm$^{-3}$, temperature $10^4$~K, and size 200~pc, being overtaken by a strong shock wave. Whereas previous simple models proposed that such a cloud would either be completely destroyed, or simply shrink in size, our results show a different and more complex behaviour: due to rapid cooling, the cloud breaks up into many small and dense fragments, which can survive for a long time.  We show that such rapid cooling behaviour is in fact expected for a wide range of cloud and shock properties.  This process applies to the evolution of emission line clouds being overtaken by the cocoon of a radio jet. The resulting small clouds would be Jeans unstable, and form stars. Our results thus give theoretical credibility to the process of jet induced star formation, one of the explanations for the alignment of the optical/UV and radio axis observed in high redshift radio galaxies. ",
        "introduction": "Regions of luminous optical line and continuum emission near high redshift radio galaxies ($z > 0.6$) are often found to be extended along the direction of the radio axis \\citep{Chambers87,McCarthy87}. One obvious explanation for these alignments, is that star formation takes place in regions where the shock bounding the radio jet, has passed.  Recent observations seem to support this idea. Deep spectra of the radio galaxy 4C41.17 at $z = 3.8$, show that the bright, spatially extended rest-frame UV continuum emission is unpolarized and contains P Cygni-like absorption features, indicating the presence of a large population of young, hot stars \\citep{Deyetal97}. \\citet{bicknelletal00} argue that this can best be understood if the shock associated with the radio jet has triggered star formation within the emission line clouds.  A nearby example where stars might be formed under the influence of a radio source is the case of Cen A. Here, young stars are found near filaments of ionized gas in a radio lobe \\citep{mouldetal2000}.  \\citet{Rees89} and \\citet{Begelmanetal89} analytically explored the evolution of intergalactic medium (IGM) clouds, overtaken by shocks from the cocoon of a radio jet. They argue that these clouds would be compressed and then gravitationally contract to form stars. However, \\citet{Icke99} claimed that the destructive aspects of the interaction between the expanding cocoon and the clouds would dominate the evolution of the clouds. In his scenario the clouds evaporate and their material mixes into the jet cocoon.  Given the complexity of the interaction between the clouds and the jet cocoon, numerical studies are a good tool to investigate this problem. Although the `shock-cloud interaction' problem was studied numerically before, none of these studies addresses the effects of radiative cooling, important for intergalactic clouds. Here we present new results of a numerical hydrodynamic study of the shock-cloud interaction problem, including the effects of radiative cooling.  In Sect.~2 we describe the general problem of shock-cloud interaction and the application to IGM clouds. Section~3 deals with the numerical method, and Sect.~4 contains the results, which we further discuss in the fifth section. We sum up the conclusions in Sect.~6. ",
        "conclusions": "We have for the first time simulated the cooling dominated evolution of an intergalactic cloud which is overrun by the cocoon of a passing radio jet. Previous analytical studies conjectured that the cloud would either be compressed, or be completely destroyed and evaporate into the cocoon. We instead find a new picture. Radiative cooling is so rapid, that nearly all of the cloud mass is compressed into many small and dense fragments with a long hydrodynamical survival time. These fragments are likely to collapse and form stars, in line with the scenario of jet induced star formation.  This type of fragmentation is expected whenever the cooling time is much shorter than the cloud crushing time. Evaluating this condition, shows this to be case for a wide range of parameters, stretching from intergalactic to interstellar conditions, see Eq.~1. The collapse-and-fragment sequence we find, may well be the way to create long lived mass loading flows inside post-shock regions \\citep{HartquistDyson}.  These simulations are only a first step, and definitely more work is needed. In future papers we plan to explore the effects three-dimensionality and self-gravity have on the fragmentation process."
    }
}