One of the lowest maximum ice extents in 30 years
This year, Arctic sea ice reached its maximum extent at only the very start of spring, on 21 March 2014. Arctic sea ice extent for March 2014 averaged 14.8 million km2. This area is only 300,000 km2 greater than that of March 2006, the record year when ice extent shrank to its lowest winter maximum since the beginning of satellite observation. In fact, the winter of 2014 should be considered one of low sea ice cover, as it ranks fourth or fifth lowest in the satellite record. Ice extent in March was some 730,000 km2 below the benchmark average calculated for the period 1981-2010. March 2014 also confirms the slow decrease in winter ice extent, at a rate of 2.6% per decade, that has been observed since satellite records began. More recently, the figures for 2014 show the continuation of the decrease in the March sea ice extent that has been observed since 2012, the year when particularly high ice coverage in the Bering Sea and Baffin Bay areas could have suggested a relative recovery in winter ice extent.
Maximum winter Arctic sea ice extent on the 21th mars 2014 - © NOAA
In line with what has been systematically observed since the start of the millennium, the Arctic sea ice extent in 2014 reached its peak value in March. The winter maximum is generally reached during or even at the end of March – the latest maximum occurence was recorded in 2010, on 31 March – but this has not always been the case. In the 1980s and 1990s, the winter maximum was sometimes reached as early as mid-February, six weeks earlier than the latest recorded maximum. As with sea ice extent, there ice an overall trend in the date of the winter maximum which occurs later and later although with considerable year to year variability. The reasons for this trend have yet to be determined but may partly reside in the fact that winds promoting the expansion of sea ice cover produce more visible effects when the ice extent is low, by enabling the late coverage of ice-free areas by drifting ice.
The main difference between the sea ice distribution at the maximum extent in 2014 and its average over the reference period 1981-2010 is essentially due to reduced ice coverage in the Barents Sea and Sea of Okhotsk. In the Barents Sea, this anomaly is broadly consistent with the persistent decline in the ice extent observed in this area not only in recent decades, thanks to satellite observations, but also over the much longer period of historical records spanning the past four centuries. Another noteworthy event that contributed to winter 2014 being one of low ice coverage compared to previous years was the decline in ice extent in the Bering Sea, an area where the maximum sea ice extent has remained above average for many years. In January and February 2014, northerly winds, which are usually responsible for cold temperatures and the southward expansion of the ice pack in this region, gave way to easterly winds, which limited ocean surface freezing and reduced the southward expansion of the ice. This exceptionally low ice coverage may explain the early break-up of the ice pack in the Bering Sea this spring, which left fishing communities and off-shore oil companies struggling with the rapid destabilization of what is essential base for their activities.
Average March monthly winter sea ice extent in the Arctic since 1978 - © NSIDC
Winds and, more generally, atmospheric variability play a significant role in the interannual variations of the winter sea ice extent. As in the example of the Bering Sea mentioned above, winds not only influence the drift and expansion of the ice in the peripheral areas of the ice pack but they also transport air masses of varying temperatures from the continent or the neighbouring Pacific and Atlantic Oceans. For example, the rapid increase in ice extent that was observed during the second week of March can be explained by the strongly positive Arctic Oscillation, the latter being the dominant mode of wintertime atmospheric variability in these regions. This mode predominantly affects the Atlantic sector of the Arctic and a positive phase of the Arctic Oscillation results in winds that cause significant export of ice into the Barents Sea. This episode helped to limit the expected decrease in maximum ice extent in the Barents Sea this year. At the same time, the Arctic Oscillation largely controls Arctic air temperatures, with a positive phase leading to colder temperatures in the North Atlantic sector and over Siberia and Eurasia. Similarly, the low sea ice extent in the Bering Sea in winter 2014 could no doubt be linked to the atmospheric variability over the Pacific. It is nevertheless important to bear in mind that in addition to these year to year changes in the ice distribution that are primarily governed by the atmosphere, slower changes also take place over periods of one to several decades, which are suspected to be linked to the oceanic variability.
Climate condition variation in Northern hemisphere with apositive (left) atmospheric pressures indice (Arctic Oscillation indice, AOi) or negative AOi (right) - © J. Wallace, University of Washington
Variability of the AOi (in black) and the Northern-Atlantic Oscillation indice (in pink) since 1950 - © http://la.climatologie.free.fr/ao/AO.htm
The percentage of multiyear ice as a fraction in the Arctic ice pack at the end of the winter rose from 30% in 2013 to 43% in 2014. This increase comes after the summer minimum ice extent in September 2013 remained above the record low of September 2012, suggesting that, in 2013, a larger quantity of ice survived the summer melt and continued to age. These figures, which could be a positive sign of a replenishing Arctic ice cover, must nevertheless not distract from the ever alarming fact that the oldest ice is gradually disappearing, since the percentage of ice aged five years or more is presently no more than 7%. As multiyear ice represents the thickest ice, the relative increase in the percentage of multiyear ice should lead to an increase in the volume of Arctic sea ice. Such an increase could indeed be deduced from the altimeter observations made by the satellite CryoSat, which have been used since 2010 to estimate the Arctic sea ice thickness. CryoSat measurements showed that the Arctic sea ice volume in October 2013 was 50% greater than in October 2012.
Multilayered sea ice distrition in march 2014 in the Arctic (up). Variability of the different ages sea ice proportions since 1983 (down) - © NSIDC
Evaluating the percentage of multiyear ice in the Arctic sea ice pack is particularly important when it comes to predicting the seasonal behaviour of the ice pack. Improving predictions in this regard, particularly on the annual minimum of the sea ice extent, is a priority goal in Arctic research due to the socio-economic implications. Recent studies have shown that the structure of the pack, especially the proportion of multiyear ice, at the end of winter, can help to predict the summer minimum. However, the latter parameter is also controlled by the summer atmospheric circulation.
An example is the spectacular minimum of September 2007, which was caused by a dipole pattern of atmospheric pressure between Greenland and northern Eurasia. This pattern promoted both the arrival of warm air masses over the Chukchi Sea and the East Siberian Sea and enhanced ice export through the Fram Strait. In addition, the increasing predominance of young ice in the Arctic ice pack is leading to changes in the ice surface state including, in particular, more widspread melt ponds which reduce surface albedo and accelerate surface melting. Assessing the impact of these various mechanisms on the development of the summer ice pack is essential if predictions in this field are to be improved.
Marie-Noëlle Houssais, June 2014
To read more:
Voir le site du Polar view center of Bremen University (Germany)
Voir le site du NSIDC (National Snow and Ice Data Center - USA)
© June 2014 - Le Cercle Polaire - All rights reserved