Chapter 19. Submarine Cables and Pipelines
Group of Experts: Alan Simcock (Lead member)
1. Submarine communications cables
1.1 Introduction to submarine communications cables
In the last 25 years, submarine cables have become a dominant element in th world’s economy. It is not too much to say that, without them, it is hard to see ho the present world economy could function. The Internet is essential to nearly al forms of international trade: 95 per cent of intercontinental, and a large proportio of other international, internet traffic travels by means of submarine cables. This i particularly significant in the financial sphere: for example, the SWIFT (Society fo Worldwide Interbank Financial Telecommunication) system was_ transmittin financial data between 208 countries via submarine cables in 2010. As long ago a 2004, up to 7.4 trillion United States dollars were transferred or traded on a dail basis by cables (Rauscher, 2010). The last segment of international internet traffi that depended mainly on satellite communications was along the East coast o Africa: that was transferred to submarine cable with the opening of three submarin cables along the East coast of Africa in 2009-2012 (Terabit, 2014). Submarine cable have advantages over satellite links in reliability, signal speed, capacity and cost: th average unit cost per Mb/s capacity based on 2008 prices was 740,000 dollars fo satellite transmission, but only 14,500 dollars for submarine cable transmissio (Detecon, 2013).
Submarine telegraph traffic by cable began between England and France in 1850 1851. The first long-term successful transatlantic cable was laid betwee Newfoundland, Canada, and Ireland in 1866. The early cables consisted of coppe wire insulated by gutta percha, and protected by an armoured outer casing. Th crucial development that enabled the modern systems was the development o fibre-optic cables: glass fibres conveying signals by light rather than electric current The first submarine fibre-optic cable was laid in 1986 between England and Belgium the first transatlantic fibre-optic cable was laid in 1988 between France, the Unite Kingdom and the United States. It was just at that time that the Internet wa beginning to take shape, and the development of the global fibre-optic network an the Internet proceeded hand in hand. The modern Internet would not have bee possible without the vastly greater communications possibilities offered by fibre optic cables (Carter et al., 2009). Over the 25 years from 1988 to 2013, an average o 2,250 million dollars a year was invested in laying 50,000 kilometres of cable a year However, this includes a great burst in the development of the global fibre-opti network that took place in 2000-2002, in conjunction with the massive interest i investment in companies based on the Internet: the so-called dot-com bubble. A the peak, in 2001, 12,000 million dollars were invested in submarine cables in on year. After the dot-com bubble burst in 2002, the cable-laying industry contracte severely, but by 2008 had recovered to what has since been a steady growth
© 2016 United Nations  

(Terabit, 2014). Figures 1 and 2 show diagrammatically the transatlantic an transpacific submarine communications cables that exist. More detaile diagrammatic maps showing submarine cables in the Caribbean, the Mediterranean North-West Europe, South and East Asia, and Sub-Saharan Africa can be found here http://submarine-cable-map-2014.telegeography.com/.
Two Arctic submarine communications cables are reported to be planned, linkin Tokyo and London: one will go around the north of the Eurasian continent, th other around the north of the American continent through the North-West passage both would service Arctic communities en route. In 2012, both were planned to b in service by 2016. The link by the American route is said to be under constructio but is not now expected to be complete until 2016. The link around the Eurasia route is reported to be stalled (Hecht, 2012; Arctic Fibre, 2014; Telegeography, 2013 APM, 2015).
Deployed international bandwidth (in other words, the total capacity of the world’ international cables) increased at a compound annual growth rate of 57 per cen between 2007 and 2011. It reached 67 Terabits per second (Tbps) in 2011, whic was six times the bandwidth in use in 2007 (11.1 Tbps). It has increased steadil since then and was estimated to be increasing to about 145 Tbps in 2014 (Detecon 2013). Submarine cable bandwidth is somewhat lower, as shown in Table 1. Th investment necessary to support this steady stream of investment is provide through consortia. The precise balance of the different interests varies from case t case, but the major players are nearly always national telecommunication Operators, internet service providers and private-sector equity investors Governments are rarely involved, except through government-owne national telecommunications operators (Terabit, 2014; Detecon, 2013).
© 2016 United Nations  

Table 1. Activated Capacity on Major Undersea Routes (Tbps), 2007-2013
South Asia  Middle East Inter continental
Australia & Ne Zealand Intercon tinental
Global Transoce anic Bandwidth
(Tbps)
Source: Terabit, 2014.
CAGR 2007-2013
1.2 Magnitude of the impact of submarine cables on the marine environment
In 2007, the total route length of submarine fibre-optic cables was about 1 millio route kilometres (Carter et al., 2009). This has now extended to about 1.3 millio route kilometres, given the extensions reported in the 2014 Submarine Cable Repor (Terabit, 2014). Although these are great lengths, the breadth of the impact on th marine environment is much, much less: the diameter of the fibre-optic cables o the abyssal plain is about 17-20 millimetres — that is, the width of a typical garde hose. On the continental shelf, the width of the cable has to be greater — about 28 50 millimetres — to allow for the extra armour to protect it from impacts an abrasion in these more dynamic waters and the greater threats from shipping and
bottom trawling (Carter et al., 2009).
© 2016 United Nation 

The cable is normally buried in the seabed if the water depth is less than 1,000-1,50 metres and the seabed is not rocky or composed of highly mobile sand. This is t protect the cable against other users of the sea, such as bottom trawling. Know areas where mineral extraction or other uses are likely to disturb the seabed ar avoided. In greater water depths, the cable is normally simply laid on the seabe (Carter et al., 2009). Where a cable is buried, this is normally done by a ploug towed by the cable ship that cuts a furrow into which the cable is fed. In a soft t firm sedimentary seabed, the furrow will usually be about 300 millimetres wide an completely covered over after the plough has passed. On other substrates, th furrow may not completely refill. The plough is supported on skids, and the tota width of the strip disturbed may be between two and eight metres, depending o the type of plough used. Various techniques have been used to minimis disturbance in specially sensitive areas: on the Frisian coast in Germany, a speciall designed vibrating plough was used to bury a cable through salt marshes (recover was monitored and the salt-marsh vegetation was re-established in one to two year and fully recovered within five years); in Australia, cables crossing seagrass bed were placed in narrow slit trenches (400 millimetres wide), which were late replanted with seagrass removed from the route prior to installation; in the Puge Sound in Washington State in the USA, cables were installed in conduits drilled unde a seagrass bed. Mangroves are reported to have recovered within two to seve months, and physical disturbance of sandy coasts subject to high-energy wave an tide action is reported to be removed within days or weeks. Where burial has no been possible, it has sometimes been necessary to impose exclusion zones and t monitor such zones (as between the North and South Islands of New Zealand (Carte et al., 2009)).
Further disturbance will occur if a cable failure occurs. Areas of cable failure ar likely to have already been disturbed by the activity that caused the cable failure Normally, the cable will have to be brought to the surface for repair. This will involv the use of a grapnel dragged across the seabed, unless a remotely operated robo submarine can be used. Reburial of the cable may involve agitating the sediment i which it has been buried. This disturbance will mobilise the sediment over a strip u to 5 metres wide. Fibre-optic cables have a design life of 20-25 years, after whic the cable will need to be lifted and replaced, with a recurrence of the disturbance although there is also the possibility of leaving them in place for use for purposes o scientific research (Carter et al., 2009; Burnett et al., 2014).
Evaluating the impact on marine animals and plants of this disturbance is not easy since the area affected, though long, is narrow. In general, the verdict is that th seabed around a buried cable will have returned to its normal situation within a most four years. In waters over 1,000-1,500 metres deep (where burial is unusual) no significant disturbance of the marine environment has been noted, although an repairs will disturb the plants and animals that may grow on the cable. Such growt is common on exposed cables in shallow calm water, but is limited in water depth greater than 2000 metres, where biodiversity and macrofaunal abundance are muc reduced (Carter et al., 2009). Some noise disturbance may be caused by the proces of laying cables, but this is not significantly more than would be caused by ordinar shipping (OSPAR, 2008).
© 2016 United Nations  

1.3 Threats to communications cables from the marine environment
Soon after transoceanic communications cables were laid, problems wer experienced from impacts of the marine environment on the cables: specifically submarine earthquakes and landslides breaking the cables (Milne, 1897). However around 70 per cent of all cable failures are associated with external impacts cause by fishing and shipping in water depths of less than 200 metres (Carter et al., 2009).
Nevertheless, the risks of damage through catastrophic geological events (includin those triggered by storms) are real, and some aspects of such risks are probabl growing (see the discussion of the effects of climate change on storms in Chapter 5) The most recent major events have been near the Taiwan Province of China. On 2 December 2006, an earthquake occurred at the south end of the island. Thi triggered multiple submarine landslides. The landslides and subsequent turbidit currents travelled over 330 kilometres and caused 19 breaks in seven cable systems Damage was located in water depths to 4,000 metres. The cable repair work involved 11 repair vessels and took 49 days. The result was a major disruption o services in the whole region: the internet connections for China, Japan, Philippines Singapore and Viet Nam were seriously impaired. Banking, airline bookings, emai and other services were either stopped or delayed and financial markets and genera commerce were disrupted (Detecon, 2013; Carter et al., 2014).
Three years later, Typhoon Morakot hit the island of Taiwan Province of China, on  August 2009. Three metres of rain fell on the central mountains, causing muc erosion. The sediment carried into sea caused several submarine landslides whic broke a number of cables. The level of disruption was shorter and less serious tha in 2006. This case is particularly significant, however, because it was the result of a extreme weather event. Given the consensus that climate change is causing th poleward migration of storms, areas that have previously been spared this kind o event are more likely in future to suffer from such storms. This is likely to increas the chances of submarine landslides, since an instability will be introduced into area where it has not previously been generated (Carter et al., 2012).
The seas off East Asia present a combination of a very dense network of submarin communications cables (see the diagrammatic map in http://submarine-cable-map 2014.telegeography.com/) and an area of unstable geology. The scale of disruptio that might be caused, either by a geological incident or by a vessel, can be envisage by considering the Straits of Malacca. Fourteen of the 37 main submarine cables i the Western Pacific run through this narrow strait. These cables represent virtuall the entire data connection between Asia, India, the Middle East and Europe. I addition, it is one of the busiest shipping routes worldwide. This drastically increase the likelihood of disruptions by anchors and other manmade hazards. Suc disruptions unfortunately do happen regularly (Detecon, 2013). This, and th situation on the Isthmus of Suez, is one of the main attractions in a submarine cabl route from the Pacific to the Atlantic around the north of either the American or th Eurasian continent. There is further a risk from deliberate human interference, bu statistically this is a rare event (Burnett et al., 2014).
The International Cable Protection Committee Ltd. (ICPC) is a non-profit organization
© 2016 United Nations  

that facilitates the exchange of technical, legal and environmental informatio concerning submarine cable installation, maintenance and protection. It has ove 150 members representing telecommunication and power companies, governmen agencies and scientific organizations from more than 50 countries, and encourage cooperation with other users of the seabed. It is thus the main forum in which issue about the protection of these submarine cable connections, vital to globa commerce, are being discussed.
1.4 Information and capacity-building gaps
A large body of knowledge already exists about the construction and operation o submarine communication cables, including how to survey environmentall acceptable routes and allow for the submarine geology. Coastal States need acces to these skills to decide on safe locations and to take account of areas of potentia geological change and disruption, or (at least) to negotiate successfully wit commercial undertakings planning to install cables.
As with many other uses of the marine environment that involve uses of the seabe within their jurisdictions that may prevent or limit other legitimate uses of the sea States need to have the capacities, in taking decisions on submarine cables, fo resolving the conflicting demands of these uses with the other parties involved.
2. Submarine power cables
2.1 The nature and magnitude of submarine power cables
The number and extent of submarine cables carrying power rather tha communications are much less significant, both in terms of their impact on th marine environment and in their importance to the world economy. They ar essentially of only local interest.
Most of the world’s submarine power cables are found in the waters around Europe The cables fall into one of two classes, depending on whether the electricity i carried as direct current (DC) or alternating current (AC). The choice depends o several factors, including the length of the submarine cable and the transmissio capacity needed: DC cables are preferred for longer distances and highe transmission capacities. DC cables can be either monopolar (when the curren returns through the sea water) or bipolar (when the cable has two components wit opposite polarities). Because monopolar DC cables tend to produce electrolysis they are now rarely used for major projects.
© 2016 United Nations  

United States
Ay Gulf o pe Mexico
Brazil
Sout Atlanti Ocean
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
Figure 1. Diagrammatic map of transatlantic submarine cables. Source: Telegeography, 2014.
© 2016 United Nations  

Last updated on November 9, 2014
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.
Figure 2. Diagrammatic map of transpacific submarine cables. Source: Telegeography, 2014.
The AC cables include those between the mainland of Germany and its island o Heligoland, between Italy and its island of Sicily, between Spain and Morocco between Sweden and the Danish island of Bornholm and, outside Europe, betwee the islands of Cebu, Negros and Panay in the Philippines. The DC cables includ cables linking the Danish islands of Lolland, Falster and Zealand to Germany Denmark to Norway, Denmark to Sweden, Estonia to Finland, Finland to Sweden France to the United Kingdom, Germany to Sweden, the Italian mainland to its islan of Sardinia and to the French island of Corsica, the Netherlands to Norway (at 58 kilometres, this is the longest submarine power cable in the world), the Netherland to the United Kingdom, Northern Ireland to Scotland in the United Kingdom of Grea Britain and Northern Ireland and the mainland of Sweden to its island of Gotland Outside Europe, there are DC cables linking the mainland of Australia to its island o Tasmania, the mainland of Canada to its Vancouver Island, Honshu to Shikoku i Japan, the North Island to the South Island in New Zealand and Leyte to Luzon in th Philippines.’ As can be seen, all these cables (with the exception of th Netherlands/Norway cable) cross fairly narrow stretches of water. They play  locally important part in managing electricity supply, enabling surpluses in on country or area to be transferred to another, or to enable an island to benefit fro the economies of scale in power generation through a link to power stations serving
1 This list has been compiled from a variety of sources.
© 2016 United Nations  

a much bigger area. The links between Denmark, Norway and Sweden play a important role in the common power policy of those three States.
2.2 Environmental impacts of submarine power cables
The disturbance of the marine environment caused by the installation of a powe cable will usually be larger than that for a communications cable, simply because th cable will be larger, in order to carry the current. However, neither the physica disturbance nor the associated noise is likely to have more than a temporary effect.
The other two aspects that have given rise to concern are heat and electromagneti fields. There are few empirical studies of heat emitted from submarine powe cables. AC cables are theoretically likely to emit more heat than DC cables Calculations for the cable from the Australian mainland through the Bass Strait t Tasmania suggested that the external surface temperature of the cable would reac about 30°C-35°C. The seabed surface temperature directly overlying the cable wa expected to rise by a few degrees Celsius at a burial depth of 1.2 metres. Reading taken at a Danish wind farm in 2005 showed that, for a 132 kV cable, the highes temperature recorded closest to the cable between March and September wa 17.7°C. German authorities have set a precautionary standard for new cables suc that the cables should not raise the temperature at a depth of 20 cm in the seabe by more than 22C. This can be achieved by burying the cables at an appropriat depth (OSPAR, 2008).
Concerns have been raised about the effects of the electromagnetic fields generate by the electric current flowing along submarine power cables, since some fish an marine mammals have been shown to be sensitive to either electric fields o magnetic fields. The World Health Organization, however, concluded in 2005 tha “none of the studies performed to date to assess the impact of undersea cables o migratory fish (e.g. salmon and eels) and [on] all the relatively immobile faun inhabiting the sea floor (e.g. molluscs), have found any substantial behavioural o biological impact” (WHO, 2005). A literature survey in 2006 reached a simila conclusion (Acres, 2006), and nothing had emerged by the 2010 European Unio report on the implementation of the EU Marine Strategy Directive to cast doubt o those conclusions (Tasker et al., 2010).
2.3 Knowledge and capacity-building gaps
As with communications cables, coastal States need to have access to the skills t locate submarine power cables in a safe and environmentally acceptable way, and t evaluate the economic and social benefits of introducing such links.
3. Submarine Pipelines
3.1 The nature and magnitude of submarine pipelines
Submarine pipelines are used for transporting three main substances: gas, oil and
© 2016 United Nations  

water. Submarine gas and oil pipelines fall into three groups: intra-field pipelines which are used to bring the oil or gas from well-heads to a point within the operatin field for collection, processing and onward transport; export pipelines, whic transport the gas and oil to land; and transport pipelines, which have no necessar connection with the operating field, but transport gas or oil between two places o land. The last category is often included with the export pipelines. The intra-fiel and export pipelines are discussed in Chapter 21 as part of the processes o extracting the oil and gas. This section is concerned only with the transpor pipelines. In general, what is said about submarine pipelines in Chapter 21 applies t gas and oil transport pipelines.
Submarine transport pipelines are used mainly for the transport of gas and ar located predominantly around the Mediterranean and the Baltic and North Seas Many have been created since 2000. In the Mediterranean, the earliest gas pipelin was the Trans-Mediterranean Pipeline, built in 1983 to link Algeria and the Italia mainland, via Sicily. This was followed in 1996 by the Maghreb-Europe Pipeline t link Morocco and Spain across the Strait of Gibraltar. Subsequent Mediterranea pipelines are: the Greenstream Pipeline, built in 2004 between Libya and Sicily, th interconnector built in 2007 between Greece and Turkey, the link completed in 200 between Arish in Egypt and Ashkelon in Israel (which has been out of service sinc 2012), and the Medgaz Pipeline built in 2011 between Algeria and Spain. Furthe north, a link was built between Scotland and Northern Ireland in the United Kingdo in 1996. An interconnector was built between Belgium and the United Kingdom i 1998. The Balgazand/Bacton Line (BBL) connected the Netherlands and the Unite Kingdom in 2006. Finally, the Nord Stream Pipeline was completed in 2011 and 201 through the Baltic, between Vyborg in the Russian Federation and Kiel in Germany This is the longest gas transport pipeline in the world (1,222 kilometres in length) Issues about its environmental impact bulked large in the negotiations leading to it construction, and particular problems were encountered over munitions dumped i the Baltic at the end of the Second World War (see Chapter 24 (Solid wast disposal)).2 There are also a number of gas pipelines linking Norwegian ga production to its export markets. The Norwegian upstream gas transportatio system has been developed from the 1970s, and continues to develop, to cater fo the transportation of natural gas produced on the Norwegian continental shelf Norwegian domestic consumption of natural gas is limited. Almost all the ga produced is exported (101,000 million standard cubic metres) to European ga markets through landing terminals in Belgium, France, Germany and the Unite Kingdom. The pipeline network in 2014 forms a 7,980-kilometre integrate transportation system, transporting gas from nearly 60 offshore fields and thre large gas processing plans on the Norwegian mainland, to European gas markets The latest main addition to the system is the Langeled Pipeline, opened in 2007 which goes from the onshore processing plant in Norway for the Ormen Lange ga field to the United Kingdom, via a riser platform at the Sleipner field.
Outside Western Europe and the Mediterranean, there is a gas pipeline linking th Russian Federation and Turkey across the South-Eastern corner of the Black Sea, and
? This list has been compiled from a variety of sources.
© 2016 United Nations 1 

one linking the island of Sakhalin to the mainland of the Russian Federation in th North-West Pacific. Oil transport pipelines exist between Indonesia and Singapor across the Strait of Malacca, and in China, linking the island of Hainan to Hong Kong. Generally, these submarine transport pipelines have been built and financed by oi and gas operators (including national oil and gas companies), sometimes i consortiums with national gas distribution undertakings.
3.2 Environmental impacts of oil and gas pipelines
The environmental impacts of intra-field and export submarine pipelines ar discussed in Chapter 21 (Offshore hydrocarbon industries). The impacts of oil an gas submarine transport pipelines are essentially the same.
3.3 Submarine water pipelines
Because of the high cost and maintenance difficulties, submarine pipelines are onl used to supply small islands close to continents or larger islands where the natura water supplies of the islands are insufficient for their needs. The supply of water t Singapore from Malaysia is the only significant international example (PUB, 2014) Domestic examples include: China (where Xiamen Island receives some of its wate from the mainland through 2.3 kilometres of submarine pipelines), Fiji (wher several small islands with tourism resorts are supplied through submarine pipelines) Malaysia (where Penang receives some of its water supply from the Malaysia mainland through 3.5 kilometres of submarine pipelines), the Seychelles (where fiv small islands are supplied through submarine pipelines of up to 5 kilometres i length) and, most significantly, in Hong Kong, China (where water is supplied t some of the islands, including the densely populated Hong Kong Island, from th Chinese mainland, through 1.3 kilometres of submarine pipelines) (UNESCO, 1991).
3.4 Knowledge and capacity-building gaps
For oil and gas transport pipelines, the requirements are likely to arise from th overall planning of the exploitation of hydrocarbon reserves and the provision of ga services. The comments in Chapter 21 on this subject are therefore relevant.
For submarine water pipelines, the essential questions will be linked to the plannin and implementation of freshwater supply services. Questions of access t information and the necessary skills need to be addressed in that context. As wit the laying of submarine communication cables, in taking decisions on submarin water pipelines within their jurisdictions, States need to have the capacities fo resolving the conflicting demands of these uses.
3 woe . :  This information has also been compiled from a variety of sources.
© 2016 United Nations 1 

References
Acres, H. (2006). Literature Review: Potential electromagnetic field (EMF) effects o aquatic fauna associated with submerged electrical cables. Supplement t the Environmental Assessment Certificate (EAC) Application for th Vancouver Island Transmission Reinforcement (VITR) Project. Prepared for B Hydro Environment & Sustainability Engineering, Victoria BC.
Arctic Fibre (2014). www.arcticfibre.com (accessed 10 November 2014).
APM (Alaska Public Media). (2015). “Arctic Fiber Project Delayed Into 2016 (http://www.alaskapublic.org/2014/12/23/arctic-fiber-project-delayed-into 2016/ accessed 10 June 2015).
Burnett, D.R., Beckman, R.C. and Davenport, T.M. (eds.), (2014). Submarine Cables The Handbook of Law and Policy, Nijhoff, Leiden (Netherlands) and Bosto (USA) (ISBN 978-90-04-26032-0).
Carter, L., Burnett, D. Drew, S. Marle, G. Hagadorn, L. Bartlett-McNeil, D., and Irvine N. (2009). Submarine Cables and the Oceans — Connecting the World. UNEP WCMC Biodiversity Series No. 31. ICPC/UNEP/UNEP-WCMC, Cambridg (England.
Carter, L., Milliman, J.D., Talling, P.J., Gavey, R., and Wynn, R.B. (2012). Near synchronous and delayed initiation of long run-out submarine sediment flow from a record-breaking river flood, offshore Taiwan, Geophysical Researc Letters, Volume 39, 12, doi:10.1029/2012GL051172.
Carter, L., Gavey, R. Talling, P.J. and Liu, J.T. (2014). Insights into submarin geohazards from breaks in subsea telecommunication cables. Oceanograph 27(2).
Detecon (2013). Detecon Asia-Pacific Ltd, Economic Impact of Submarine Cabl Disruptions, prepared for Asia-Pacific Economic Cooperation Policy Suppor Unit (Document APEC#213-SE-01.2).
Hecht, J. (2012). Fibre optics to connect Japan to the UK — via the Arctic, Ne Scientist, 2856.
Milne, J. (1897). Sub-Oceanic Changes: Section III, The Geographical Journal, Vol 10(3).
OSPAR (2008). OSPAR Commission, Background Document on potential problem associated with power cables other than those for oil and gas activities London.
PUB (Singapore Public Utilities Board) (2014). The Singapore Water Story Water From Vulnerability to Strength http://www.pub.gov.sg/water/Pages/singaporewaterstory.aspx (accessed 2 October 2014).
© 2016 United Nations 1 

Rauscher, K. F. (2010). ROGUCCI — Reliability of Global Undersea Cabl Communications Infrastructure — Report. IEEE Communications Society, Ne York, USA.
Tasker, M.L., Amundin, M., Andre, M., Hawkins, A., Lang, W., Merck, T. Scholik-Schlomer, A., Teilmann, J., Thomsen, F., Werner, S. and Zakharia, M (2010). Marine Strategy Framework Directive Task Group 11 Report Underwater noise and other forms of energy, Luxembourg.
Telegeography (2013). Is dormant Polarnet project back on the agenda Telegeography (https://www.telegeography.com/products/commsupdate/articles/2013/01 28/is-dormant-polarnet-project-back-on-the-agenda/ accessed 10 Octobe 2014).
Telegeography (2014). Submarine Cable Map 2014. Telegeograph (http://submarine-cable-map-2014.telegeography.com/ accessed 3 September 2014).
Terabit (2014). Terabit Ltd/Submarine Telecoms Forum Inc, Submarine Cable Industry Report, Issue 3 (http://www.terabitconsulting.com/downloads/2014-submarine-cable market-industry-report.pdf accessed 20 August 2014).
UNESCO (1991). United Nations Education Scientific and Cultural Organization Hydrology and Water Resources of Small Islands, A Practical Guide. Studie and Reports on Hydrology No. 49, UNESCO, Paris.
WHO (2005). World Health Organization, Electromagnetic Fields and Public Health  Effects of EMF on the Environment, (http://www.who.int/peh emf/publications/facts/envimpactemf_infosheet.pdf accessed on 2 November 2014). Geneva.
© 2016 United Nations 1