astro-legacy-archive/iris-infrared-maps

IRIS Infrared Maps

The Improved Reprocessing of the IRAS Survey (IRIS) provides co-added infrared sky-brightness maps at 12, 25, 60 and 100 microns. These four configurations are the authors' native-resolution NSIDE-2048 nohole HEALPix products, frozen by NASA LAMBDA. HCON1, HCON2 and HCON3 were co-added, and DIRBE data fill the roughly two per cent of the sky not observed by IRAS.

Bands	12, 25, 60 and 100 microns
Pixelisation	HEALPix NSIDE 2048; 50,331,648 pixels
Ordering	RING
Coordinate frame	Galactic (COORDSYS = 'G')
Quantity	Sky-brightness intensity
Unit	MJy/sr, from the product documentation
Source column	INTENSITY

How to use

python -m venv .venv && .venv/bin/pip install datasets huggingface_hub pyarrow

from datasets import load_dataset
dataset = load_dataset("astro-legacy-archive/iris-infrared-maps", "IRIS_nohole_4_2048_v2", split="train[:1]", columns=["INTENSITY"])
row = dataset[0]
print(row["INTENSITY"][:3])

[1.4355186223983765, 1.4360263347625732, 1.540387511253357]

import json
from huggingface_hub import hf_hub_download
import pyarrow.parquet as pq
path = hf_hub_download("astro-legacy-archive/iris-infrared-maps", "IRIS_nohole_4_2048_v2.parquet", repo_type="dataset")
print(json.loads(pq.read_schema(path).metadata[b"healpix_map"])["schema"])

astro-legacy-archive/healpix-map

The embedded document records the ordering, pixel_rule, unit, and coverage codes. The documented undefined value and the measured NaN population are distinct records; validity and statistics follow those records.

Configurations and structure

Each configuration contains 49,152 source rows. Each INTENSITY value is a fixed-size list of 1,024 float32 pixels, preserving the source bintable's vector shape and row order.

config	wavelength	angular resolution	range excluding -32768 and NaNs (MJy/sr)	download
IRIS_nohole_1_2048_v2	12 micron	3.8 ± 0.2 arcmin	-1.6049669981 to 9879.0751953125	about 172 MB
IRIS_nohole_2_2048_v2	25 micron	3.8 ± 0.2 arcmin	-1.1838477850 to 19779.328125	about 175 MB
IRIS_nohole_3_2048_v2	60 micron	4.0 ± 0.2 arcmin	-3.4821598530 to 21098.47265625	about 176 MB
IRIS_nohole_4_2048_v2	100 micron	4.3 ± 0.2 arcmin	0.4944684505 to 24603.37890625	about 177 MB

All four source files omit TUNIT1. The documented MJy/sr unit and its source pin are recorded in columns[0].unit, and MJy/sr is present in the Arrow field metadata. PyArrow schema inspection exposes the source field metadata; Hugging Face Dataset.features exposes its type but not Arrow field metadata.

Every band contains 6,095 float32 values equal to -32768. LAMBDA calls that value “undefined,” and it appears in coverage.codes with the source documentation as its basis. Each band also contains four quiet NaNs with bit pattern 0x7fc00000 in INTENSITY; their semantics are not documented in the checked sources. They appear as a measured code with no meaning. coverage.blank_pixels is 6,099, the combined count of both code records. The example derives its validity mask from their source_bits values.

The nohole designation refers to the use of DIRBE information where IRAS had no observation. It does not imply that every stored value is a finite measurement. The companion product family includes ordinary coadds, individual HCON maps, lower-resolution products, projected images and separately produced uncertainty maps; none is included in these four intensity configurations.

LAMBDA notes two limitations. IRIS photometry is limited mainly by the zodiacal-light subtraction in the DIRBE data, based on the Kelsall et al. (1998) model. The ISSA plates were not constructed completely self-consistently: their overlaps usually differ negligibly, but in some cases, especially in bright regions, the difference can reach several percent.

Considerations for use

• These are co-added intensity products. No uncertainty field or uncertainty product is included.
• Negative finite intensities occur in the 12, 25 and 60 micron bands.
• NSIDE 2048 defines the delivered pixel grid, not the independent angular resolution of every feature in the underlying surveys.
• Configuration boundaries follow the four separately delivered FITS files. Wavelength is recorded per file in embedded metadata rather than inferred from the ordinal in consumer code.
• The _v2 designation belongs to LAMBDA's filenames. LAMBDA describes these copies as minor FITS-compliance header revisions; the card does not compose a new scientific release name from that suffix.

Provenance

IRSA is the canonical archive for the original 430 ISSA-format plates per band and assigns dataset DOI 10.26131/IRSA94. The IRIS overview and LAMBDA product page describe the products. The four converted source files are:

source file	SHA-256
IRIS_nohole_1_2048_v2.fits	97e496c585c085e3f262079650e0efebb2903123f5a3168e39b8836422881a89
IRIS_nohole_2_2048_v2.fits	485e18a9afeac2b1171509475069c529405cad1abf0f2187cf5abbace4a0ab73
IRIS_nohole_3_2048_v2.fits	a9fdbef7d54f2e185f7d9570640f11a7df42b7aabe9399c220fbbda9969fe7ab
IRIS_nohole_4_2048_v2.fits	402adba2396eae9b5f518898a2e39939f38d394fa3bc85e680e3b4955d5898ab

The complete source headers, URLs, hashes, retrieval date and regeneration command are embedded in every Parquet file. Rebuild with python tools/make_iris.py data/iris.

Licensing, credit and citation

HEASARC states verbatim:

HEASARC materials are all available freely for your use.

This is a free-use statement rather than a named formal licence.

No named formal product licence is published on the checked IRSA, authors', CADE or LAMBDA pages. The HEASARC free-use statement quoted above is represented as other. LAMBDA requests:

We acknowledge the use of the Legacy Archive for Microwave Background Data Analysis (LAMBDA), part of the High Energy Astrophysics Science Archive Center (HEASARC). HEASARC/LAMBDA is a service of the Astrophysics Science Division at the NASA Goddard Space Flight Center.

IRSA asks IRIS users to cite dataset DOI 10.26131/IRSA94. Please also cite Miville-Deschênes & Lagache (2005), ApJS 157, 302–323, DOI 10.1086/427938.

Errata and requests

To report an erratum, propose a correction, or ask a question about this dataset, open a Discussion or Pull Request in its Community tab; posting requires a free Hugging Face account. Dataset requests and archive-wide questions may be posted in the README space's Community tab or the Community tab of any dataset. Confirmed corrections are made in the source conversion tools and published by rebuilding the dataset.