Climate indicator: sea ice
About 15% of the world’s ocean surface is covered with ice for parts of the year, mainly in polar regions. Sea ice plays an important role in the global climate system, and the ice both influences and is influenced by global warming. Sea ice is an important climate indicator because it responds to climate-related changes in the atmosphere and the ocean. The new record lows in the extent of sea ice in the Arctic signal that the earth is approaching a climatic tipping point. Sea ice in the Arctic is melting at an accelerating rate, much faster than the climate models could predict, whereas the extent of sea ice around Antarctica is stable or increasing slightly, though the regional variations are large.
The extent of sea ice in the northern hemisphere varies with the season. Normally it covers an area of 5-15 million km2, corresponding to 0.5-1.5 the area of Europe. The total volume of ice is calculated at 13 000 km3 in the summer and 16 500 km3 at the end of winter. Formation of sea ice in the winter is essentially counterbalanced by melting in the summer, aided by factors such as the East Greenland Current, which transports 2 000-3 000 km3 of drift ice out of the Arctic Ocean annually. The extent of the ice is smallest in September and largest in March.
It is impossible to give any precise figures on ice extent prior to 1979, but some descriptions of the ice extent can be found in historic records and old maps. Since 1978, sea ice has been monitored round the clock from satellites fitted with passive microwave instrumentation. Each image covers 25-50 km, making the resolution rather low. Changes in snow cover, water on top of the ice, and its salinity and temperature can also lead to misinterpretation of its extent.
Nonetheless, these long time series provide unique data, robust enough for analysis of long-term trends and deviations from ”normal”, and several different sets of observations can be compared and used analyses. Since the beginning of the 1990s, analyses are based on SAR data (synthetic aperture radar data), yielding ice charts with greater detail and more precision.
Sea ice thickness was previously measured by drilling holes at specific sites along the Arctic coast. After 1958, measurements using sonar have been taken sporadically from submarines, and as of the 1990s, also from permanent installations. Sea ice was measured from satellites fitted with radars starting in 1993, and with lasers starting in 2003. The Norwegian Polar Institute also monitors ice with the ”EM-bird”, an instrument that employs electromagnetic induction to calculate ice thickness. Other ice measurement methods include buoys in the ice that calculate mass balance over large geographic areas, and sonars mounted on arrays of buoys that measure ice mass balance continuously over long spans of time. This information helps validate results obtained using satellite data and models of sea ice.
In recent years, the sea ice in the Arctic has decreased significantly in both extent and thickness. Over the period 2001-2011, the average sea ice extent in September was 5.49 million km2, which is 22% less than the 7.04 million km2 average for the period 1979-2000. The ice extent in September has decreased considerably in recent years, reaching a record low extent of 3.41 million km2 in the middle of September 2012. The trend over the past decade (2001-2011) is a loss of 191 000 km2 per year (which corresponds to a loss of 27% per decade relative to the average extent for 1979-2000. The extent of the ice has not declined as much in the winter season, but even there we see a reduction. The March average for the period 2001-2011 was 15.04 million km2, or 4.5% below the 1979-2000 average of 15.75 million km2. 
When the ice reached its maximum extent in the middle of March 2013 it covered 15.13 million km2: that was the sixth lowest reading since satellite measurements began. The smallest maximum ever was noted in March 2011.
The IPCC summary of available knowledge indicates that the total extent of sea ice decreased by 3.5-4.1% per decade between 1979 and 2012, while the amount of sea ice that has survived at least one summer decreased by 9.4-13.6% per decade. Changes in ice extent are almost certainly more rapid than previously, and the fastest changes of all are occurring in summer and autumn. There is documentation suggesting that the ice in the Arctic Ocean has become 1.3 to 2.3 metres thinner (on average) between 1980 and 2008. In some parts of the Arctic, this means that ice-free conditions last up to two months longer. These pan-Arctic trends are apparent also in waters near Norway, the Barentshavet og Framstredet. Changes in ice extent are almost certainly more rapid than previously, and the fastest changes of all are occurring in summer and autumn. There is documentation suggesting that the ice in the Arctic Ocean has become 1.3 to 2.3 metres thinner (on average) between 1980 and 2008. In some parts of the Arctic, this means that ice-free conditions last up to two months longer. These pan-Arctic trends are apparent also in waters near Norway, the MOSJ show a distinct negative tendency in ice extent throughout the monitoring period, and reveal substantial variability in the thickness of ice passing through Fram Strait, although both the variability and the thickness have decreased in recent years.
Data from ice cores, tree rings and lake-bottom sediments show that there is less sea ice cover in the Arctic now than at any time over the past 1450 years
At no time in the past 1450 years has there been a greater loss of sea ice than what we have observed since the beginning the 1990s. The study also shows that the ice reached its greatest extent in 1912, in the final phase of the “Little Ice Age”. The study’s reconstruction reveals no clear relationship between historic ice coverage and atmospheric temperature; rather it appears that sea ice extent is the result of complex interactions between atmospheric circulation and ocean currents – confirming what we see today.
Every year since 2007, the observed minimum sea ice extent has been considerably under the minimum predicted by climate models, on the order of two standard deviations below the average extent for 1979-2000. The models have thus been unable to predict this rapid melting. Every year after 2007, the minimum extent of sea ice has been lower than was ever observed before 2007. The low ice coverage has lasted through October, when ice normally starts to form and ice extent increases. In addition to the record low ice extent in summer, sea ice coverage hit a record low in January 2011. Hudson Bay did not freeze over until the middle of January – about a month later than usual. At that time, the Labrador Sea was still essentially ice-free. Low ice coverage in September is a consequence of little ice at the beginning of the summer season. The record low sea ice extent in September of 2012 was on the same order of magnitude as what the IPCC, in 2007, predicted would not be seen until 2070. SWIPA adjusted the projections to show this amount of ice in 2040. The climate models must still be improved so they can describe changes that have already been observed, and provide more accurate projections for the future. Currently available climate models show an ice-free Arctic Ocean in summer by the end of this century – probably even within the next 30 to 40 years.
Sea ice in the Arctic is currently dominated by relatively young ice; lots of first-year and less multiyear ice. Over the past few years, the oldest multiyear ice (older than 5 years) has been disappearing rapidly, particularly in the Canadian Arctic, and most of this oldest ice is now gone. The summers of 2008 and 2009 were less extreme than the summer of 2007 in terms of loss of ice, and relatively little ice was transported out of the Arctic Basin during the winters of 2009-2010 and 2010-2011. During these years, the amount of young multiyear ice (2-3 years old) increased. This shows how annual variations also influence ice conditions, in combination with the overall downward trend. The Arctic will probably become increasingly dominated by first-year ice. Multiyear ice will probably disappear quickly and in “surges” rather than gradually.
The waters of the Barents Sea and around Svalbard are characterised by seasonal formation of sea ice that melts away during the summer (seasonal sea ice). Landfast ice forms in the fjords. In northeastern Svalbard, some ice can survive an entire summer, becoming first-year ice and ultimately multiyear, provided it survives several summers. Some multiyear ice can drift into the Barents Sea. There has been little ice around Jan Mayen in recent years, but drift ice is occasionally present. Ice conditions in this area vary substantially from year to year, but the overall trend throughout the measurement period is negative. In Fram Strait, sea ice consists mainly of drift ice from the Arctic Ocean moving southward with the East Greenland Current along the east coast of Greenland. Some of this drift ice is old multiyear ice that can be several metres thick and quite hard owing to packing and formation of pressure ridges. Ice drift velocity in the Arctic Ocean has increased, mainly due to lower ice density and thinner ice. Monitoring data presented in MOSJ indicate a huge loss of multiyear ice through Fram Strait since 2005, after a relatively stable period in the 1990s.
Sea ice melting in the Arctic is caused by a combination of global warming and positive feedback mechanisms that speed up the melt rate as the ice gets thinner and the ice cover shrinks. These effects come in addition to the natural variations in sea ice cover on different time scales.
The extent of Antarctic sea ice is stable or increasing slightly, albeit with large regional differences. The ice extent in Antarctica reached a record high in 2013, when it was greater than at any time for which measurements from satellites are available. Satellite data from Antarctica show that the average ice extent has increased slightly in the area as a whole (ranging from 1.2 to 1.8% per decade between 1979 and 2012), but that the regional differences are large. Around the Antarctic Peninsula, the sea ice cover is shrinking. In the Bellingshausen Sea, the ice cover has retreated, with about 5.3% less ice cover per decade in recent years, an effect that correlates closely with rising temperature over the Antarctic Peninsula. West of the Antarctic Peninsula, the ice season is now almost 90 days shorter, and multiyear ice has disappeared. These changes are accelerating. The trends in sea ice in the south can be linked to ice movement patterns; wind-driven changes in the advection of ice appear to be the most important factor. It has also been suggested that fresh water from the melting ice sheet plays a role.
Most climate models predict an essentially ice-free Arctic Ocean within this century, some models as soon as 2040. The accuracy of the model predictions for sea ice is still not good enough, and the models have been unable to recreate the rapid disappearance of sea ice observed since 2005.
Available knowledge indicates that is it is reasonable to assume that sea ice around Antarctica will gradually decrease in extent and thickness.
The amount and the properties of snow on top of sea ice, the occurrence of soot and meltwater ponds on the ice are all factors that affect albedo. Much of the research done today focuses on quantifying how these factors contribute to melting in the Arctic. The role of clouds in global warming is also a focus of research interest, as are attempts to understand how a reduction of sea ice in the Arctic will affect cloud formation. These research results will make climate models more robust and accurate, which is crucial for projecting future developments in sea ice.
Less ice and thinner ice in the Arctic will have many consequences, some of which involve thermodynamic feedback to the climate system. Most of the feedback loops in the ocean–ice–climate system are positive; in other words, they enhance melting and warming. The global climate system is controlled by the energy balance in the ocean and atmosphere, and sea ice plays a crucial role in the climate.
Global ocean and atmospheric circulation is driven by forces that strive to eliminate temperature differences between the poles and the equator, and heat exchange between ocean and atmosphere is an important factor. Since sea ice floats, it “puts a lid” on the ocean, creating a barrier between the atmosphere and ocean. In this way, sea ice strongly influences heat exchange. In winter, the air in polar regions is extremely cold, while the sea is considerably warmer. Sea ice serves as an insulating layer between the ocean and the atmosphere, preventing warmth in the ocean from reaching the atmosphere. About half of the heat exchange in the Arctic Ocean occurs through openings in the ice: leads or polynyaer.
When sea ice thins and ice coverage decreases, more heat from the ocean penetrates through the insulating layer and heats the atmosphere. The ice melts both from above, in contact with warmer air, and from below, in contact with warmer water.
In, on, and under the sea ice, there are entire ecosystems that depend on ice. These ecosystems, and all the ice-associated species that are a part of them, will be affected to a greater or lesser degree now that the Arctic sea ice is diminishing. Some endemic arctic species will probably become extinct if the ice continues to shrink. Species from more southerly latitudes will extend their range northward, and may compete with and displace arctic species. Read more about isavhengige økosystemene og klimaendringer.
At the same time, an ice-free ocean opens up previously ice-bound areas for shipping and exploitation of resources. The international community must soon make decisions concerning the opportunities and risks these changes entail.
- Arctic Monitoring and Assessment Programme (AMAP), 2012. Arctic Climate Issues 2011: Changes in Arctic Snow, Water, Ice and Permafrost. Snow, Water, Ice and Permafrost in the Arctic (SWIPA) 2011 Overview Report.
- Intergovernmental Panel on Climate Change (IPCC) 2013. Fifth assessment report contribution.
- Christophe Kinnard et al. 2011. Reconstructed changes in Arctic sea ice over the past 1,450 years. Nature 479, 509–512. DOI:10.1038/nature10581
- James E. Overland et al. 2011. Warm Arctic—cold continents: climate impacts of the newly open Arctic Sea. Polar Research (30): 15787. DOI:10.3402/polar.v30i0.15787
- Intergovernmental Panel on Climate Change (IPCC) 2007. Fourth assessment report contribution.
- Nalan Koç et.al. 2009. Melting snow and ice: a call for action. Centre for Ice, Climate and Ecosystems, Norwegian Polar Institute. ISBN 978-82-7666-264-1
- Oscar Schofield et.al. 2011. How Do Polar Marine Ecosystems Respond to Rapid Climate Change?. Science (328) 5985. 1520–1523. DOI: 10.1126/science.1185779
- A.J. Turner et al. 2013. Antarctic climate change and the environment: an update. Polar Record. DOI:10.1017/S0032247413000296.