- The Arctic: Warm sea surface temperatures and weak winds partly explain the low sea-ice extent at winter’s end.
- ESA sea-ice experiment in northern Canada: Polar 6 and other research aircraft will directly match satellites’ routes so as to validate the satellite sea-ice monitoring data with their own.
- The Antarctic: In the first month of winter, the sea-ice extent increased by nearly 2 million square kilometres. A new climate study reveals the causes of the sudden sea-ice decline from 2015 to 2017.
The Arctic: Second-lowest winter maximum since the beginning of satellite observation
The month of March marks the end of winter in the Arctic, and with it, a central turning point in the sea-ice statistics – and this year was no exception. According to our satellite data, the Arctic sea ice reached its winter maximum (greatest winter extent) on Friday, 13 March 2026. At 14.48 million square kilometres, it was the second-lowest winter maximum since the beginning of satellite observation in 1979. By comparison: from 1981 to 2010, the mean winter maximum value was 15.61 million square kilometres. As such, compared to the long-term mean there were more than a million square kilometres less ice-covered ocean in March 2026 – an area larger than of Germany and France combined. The lowest-ever winter maximum was reached in 2025: on 21 March 2025, satellites recorded a maximum winter extent of 14.45 million square kilometres – only 26,000 square kilometres less than this year’s figure.
“The low sea-ice maximum this winter is hardly surprising if we bear in mind that the major marginal zones of the Arctic Ocean have been ice-free for the past few months. This is particularly true for the Barents Sea, southern Fram Strait, and Sea of Okhotsk. Accordingly, this year’s winter maximum confirms the long-term decline in Arctic sea ice,” says Dr Klaus Grosfeld, a climate expert at the Alfred Wegener Institute and co-founder of the Sea Ice Portal (Figures 1, 2 & 3).
Compared to the long-term mean, March began with a markedly low sea-ice extent of 14.11 million square kilometres, before the total amount of sea-ice-covered waters rebounded – in one smaller and two larger increments (Figure 2). In the second half of the month, the sea-ice extent decreased by nearly 500,000 square kilometres. Nevertheless, the sea-ice extent curve no longer dipped below the range of minima and maxima for the period 1981 – 2010, hovering along its lower edge instead (Figure 2).
Figure 1: In March 2026, the monthly mean sea-ice extent in the Arctic was just below the trend line for March, confirming the long-term decline in Arctic sea ice.
Figure 2: Map of sea-ice extent development, seasonal cycle. The curve for the year 2026 (sky blue line) gradually rose in the first half of March. But after reaching its winter maximum on 13 March, it declined again just as quickly: in the second half of the month, the extent dropped by nearly 500,000 square kilometres in a matter of days.
Figure 3: Difference in the mean position of the ice margin in March 2026, compared to the long-term mean for the years 2003 – 2014. Regions marked in blue had more Arctic sea ice in the third month of 2026 than in the reference period; those marked in red had less.
Identifying the causes: Warm waters, weak winds
Unusually high sea surface temperatures could have been one cause of the low sea-ice extent in the northern Barents Sea, the Sea of Okhotsk, southern Fram Strait and adjacent Labrador Sea: in all these regions, the mean sea surface temperature in March was 2 to 3 degrees Celsius above the long-term mean for the reference period 1971 – 2000 (Figure 4).
This winter, the wind- and current-driven motion and distribution of the Arctic sea ice had little influence on the sea-ice extent.
From January to March 2026, there was little wind in the Arctic and therefore less pack-ice motion. The sea ice that was newly formed during that time was basically still at exactly the same spot at the end of winter
As a rule, weaker ice drift on the Arctic Ocean is good news for the sea-ice cover on the Siberian marginal seas. “I predict that the sea-ice cover in the Laptev Sea will break up somewhat later this spring. Due to the weak winds from December 2025 to March 2026, less sea ice was transported out of coastal waters. Therefore, the sea ice near the coast is a few weeks older and somewhat thicker than in windier years, which should allow it to withstand the spring sunshine a bit longer,” says Thomas Krumpen.
Figure 4: Arctic sea-surface temperature anomalies in March 2026 compared to the reference period 1971 – 2000. Particularly in the Labrador Sea, Barents Sea, eastern Bering Sea, and the Sea of Okhotsk, the sea surface was unusually warm.
Figure 5: AWI sea-ice physicist Dr Thomas Krumpen has specialised in sea-ice drift. By drawing on satellite data, he can determine how the pack ice is redistributed throughout the Arctic, or how much sea ice is transported to Fram Strait by the transpolar drift. Photo: Alfred Wegener Institute / Esther Horvath
Polar 6: Aerial campaign mirrors satellites’ paths
At the end of March, he and his AWI colleague Luisa Wagner packed their bags: the two experts on sea-ice surveying will take part in an extraordinary ESA (European Space Agency) monitoring campaign in northeast Canada. As part of the international “Copernicus Expansion Missions Sea Ice Experiments”, they will spend the second week of April on board the research aeroplane Polar 6, which they’ll use to survey the seasonal sea ice on and around Cambridge Bay, a small town on the southern tip of Victoria Island, part of the Canadian Arctic Archipelago. With the aid of their colleagues Arttu Jutila (Finnish Meteorological Institute) and Richard Kelly (University of Waterloo, Canada), they will monitor the ice and snow thickness, as well as the surface properties of the sea ice, with the AWI’s sea-ice thickness sensor, the EM-Bird. What sets the campaign apart: not only will the flyovers correspond with those of various satellites; Polar 6 and two other research aircraft will also match the satellites’ courses – and each of the three will use a different method to record the properties of the sea ice.
“If it all goes to plan, then all three aircraft will essentially fly below the satellites, while colleagues from other research centres take readings directly on the surface of the ice,” says Luisa Wagner, describing the international research project (Figure 6). “The parallel readings taken at the ocean’s surface, from the air and from space will all be needed to generate the datasets we’ll then use to refine certain algorithms. In turn, these algorithms are to ultimately allow us to derive sea-ice thickness from the satellites’ radar data,” she adds.
Figure 6: AWI remote sensing expert Luisa Wagner is planning and coordinating the flights of the research aircraft Polar 6 during the ongoing sea-ice monitoring campaign in northern Canada, Greenland and Svalbard. Photos: Alfred Wegener Institute / Thomas Krumpen
Figure 7: On the map, the stations of this year’s spring survey campaign are indicated in red, while the regions in which the team will measure the sea ice from the air are indicated in yellow. Map: Google Earth, modified by Luisa Wagner
Applying one-of-a-kind observational technologies
The flights in cooperation with the ESA are a real highlight of this year’s spring aerial survey campaign with the research aircraft Polar 6 (IceBird sea-ice monitoring programme). The campaign will take Luisa Wagner and her team from Inuvik on the Canadian border of the Beaufort Sea to Cambridge Bay, Eureka and Station Nord in northern Greenland, to Svalbard (Figure 7). “In this way, we’ll be continuing the aircraft-supported long-term observations of the Arctic sea ice that the AWI has regularly conducted in the North American Arctic and northern Greenland for more than 30 years. Large-scale campaigns like this one not only allow us to identify changes in the sea ice and to distinguish between extreme events and natural fluctuation. On the basis of the data gathered, we can also investigate how ice growth is changing in an ever-warmer world,” she explains.
The PhD candidate is not only planning and coordinating the flights; she will also gather data for her own dissertation. According to the 27-year-old remote sensing expert at the AWI: “I’m concentrating on our snow radar, the only one of its kind in the world. I plan to use its data to derive the snow’s thickness and properties, so that I can then investigate how, for instance, changes in the sea ice’s surface topography affect snow distribution.”
What kind of changes in the surface could these be? “We know that the Arctic pack ice now less frequently forms pressure ridges. That means the sea-ice cover is generally flatter and smoother. When the ice has these properties, the wind can distribute snow more broadly. But does that mean the snow cover has become thinner and more homogenous? Using the snow radar data, my goal is to answer that question and identify potential interrelations,” Luisa Wagner explains.
The Antarctic: New sea ice covers nearly 2 million square kilometres
In March 2026, the extent of the Antarctic sea ice grew by nearly 2 million square kilometres. On the last day of March, it was 4.95 million square kilometres, just 340,000 square kilometres under the long-term mean for the reference period 1981 – 2010 for this date. The monthly mean sea-ice extent was 3.79 million square kilometres, putting it in 19th place in the time series and roughly 200,000 square kilometres below the trend line (Figures 8 & 9).
Figure 8: The monthly mean Antarctic sea-ice extent for March 2026 was also just below the trend line for the month. But in comparison to the four previous years, there was significantly more pack ice in the Antarctic in March 2026.
Figure 9: Comparison of Antarctic sea-ice concentration on the first and last day of March 2026. In the course of the month, new sea ice particularly formed in the Ross Sea and eastern Weddell Sea.
The sea-ice In the Antarctic, the sea-ice distribution, and therefore also the position of the sea-ice margin, were chiefly influenced by winds in the course of the summer. “From November 2025 to January 2026, unusually strong westerlies developed, especially over the eastern Weddell Sea. They drove the pack ice to the west and northwest. As a result, ice-free patches formed comparatively early in summer in the eastern Weddell Sea. At the same time, sea ice accumulated in the north-west Weddell Sea, where, due to the markedly weak winds in the marginal ice zone, it didn’t spread further,” Thomas Krumpen explains (Figures 10, 11 & 12).
Figure 10: Difference in the mean position of the ice margin in March 2026, compared to the long-term mean for the years 2003 – 2014. Regions marked in blue had more Antarctic sea ice in the third month of 2026 than in the reference period; those marked in red had less.
Figure 11: The map shows in which regions of the Southern Ocean the Antarctic sea ice moved faster (red) or slower (blue) from November 2025 to January 2026 in comparison to the period 2015 – 2025. The unusually weak motion on the outer edge of the Weddell Sea, and the ice transport from the eastern Weddell Sea to the west and northwest, can be clearly recognised. Graphic: Alfred Wegener Institute / Thomas Krumpen
Figure 12: The marginal ice zone marks the transition from sea ice to open ocean. The sea ice melts here, while the surface water is characterised by waves and swell. This photo shows participants in the current Polarstern expedition engaged in fieldwork in the marginal ice zone of the northwest Weddell Sea. We’ll tell you more about their findings in the next sea-ice update. Photo: Alfred Wegener Institute / Christian Haas
New study: Warmer deep water is intensifying sea-ice melting in the Antarctic
Based on the available data, roughly ten years ago the climate system in the Antarctic must have changed fundamentally: since the winter of 2015, the amount of pack ice has no longer increased year after year, as it did from 2008 to 2015. Instead, in the span of eighteen months the winter sea-ice extent dropped to a record low and has been subject to considerable fluctuation ever since. Drawing on thousands of data points from the Southern Ocean and the atmosphere, researchers from the University of Gothenburg and the Alfred Wegener Institute have now determined the cause of this change.
“Back then, there was a protective layer of cold water under the sea ice in the Antarctic. It prevented the warming deep water from rising and melting the sea ice from below,” explains oceanographer and first author Theo Spira. As a PhD student at Sweden’s University of Gothenburg, he led the work behind the new study, though he is now pursuing research at the AWI.
“Then, in the winter of 2015, there were unusually powerful winds over the Southern Ocean. They churned up the surface water to such a great depth that the warm deep water made its way to the surface and the sea ice melted. These processes led to a fundamentally new form of interaction between the sea ice and ocean,” says Theo Spira.
This fundamental shift was preceded by gradual changes in the water-mass layering of the Southern Ocean: as has since been confirmed, the protective layer of cold surface water became thinner from 2005 to 2015. In addition, from 2009 to 2015 the water at a depth of more than 300 metres became roughly 0.15 degrees Celsius warmer and more saline. Due to these two processes, it became harder for this water to mix with the surface water. It essentially shielded the surface water against heat from the deep.
But in the winter of 2015, this trend suddenly reversed. The salinity of the water below 300 metres declined, which was conducive to mixing with other water masses. The thermal shield collapsed. “There is every indication that these changes to water-mass layering in the upper water column of the Southern Ocean are what triggered the dramatic sea-ice decline after the winter of 2015,” write the experts in their new study, which was published as an open-access article in the journal Nature Climate Change this March.
Figure 13: AWI climate expert Theo Spira’s research focuses on how the vertical layering of water masses in the Southern Ocean, related thermal flows, and further interactions between the ocean and ice impact the development of Antarctic sea ice and other climate parameters. He was first author of the new study. Photo: Private
Graphics