- Very little winter sea ice in the Arctic, trend of the past few months continues
- Slow summer melting in the Antarctic prevents above-average ice loss
- AWI researchers survey the distribution and origins of the platelet ice below the fast ice in Atka Bay
The Arctic: More of the same
In the first month of 2025, the sea-ice satellite data from the Arctic held no positive surprises. In many regions of the Arctic Ocean, so little new ice was formed in January that the deficit from previous months could not be offset. The mean sea-ice extent was 13.19 million square kilometres and, with only 4,900 square kilometres more than number two, the third-lowest mean value for January since the beginning of satellite observations. To date, the only lower January sea-ice extents on the Arctic Ocean were in January 2017 (13.19 million square kilometres) and January 2018 (13.12 million square kilometres). Moreover, for the first time in five years (2019 – 2024) the monthly mean value was below the trend line, which denotes a decline in Arctic sea ice of 2.6 percent per decade (Figure 1).
Figure 1: Development of mean sea-ice extent in the Arctic for the month of January. The light blue line represents the long-term decline in sea ice. After five years above the trend line, the monthly mean value fell below the line in January 2025.
Figure 2: Map of the mean sea-ice concentration in the Arctic in January 2025. Ice-free areas in the eastern Hudson Bay, northern Barents Sea, and Sea of Okhotsk are clearly recognisable. The bright green line marks the mean January sea-ice extent for the years 1981 – 2010.
Figure 3: Difference in the mean position of the ice margin in January 2025, compared to the long-term mean for the years 2003 – 2014. Regions marked in blue have/had more Arctic sea ice than the reference period; those marked in red have/had less.
Further, the regions with markedly little sea ice were unchanged from the previous month (Figures 2 and 3). In the mean sea-ice concentration map, extensive ice-free areas can be seen – in the eastern Hudson Bay, northern Barents Sea, and Sea of Okhotsk. According to our satellite data, the Canadian sector of Hudson Bay still wasn’t completely frozen over on 31 January.
These sea-ice observations correspond to reports that January 2025 was the warmest January since the beginning of record-keeping. In some parts of the Arctic, the monthly mean temperature in January was 6 degrees or more above the long-term mean for the reference period 1971 to 2000. The air masses were particularly warm over northeast and northwest Canada, Alaska, northern Greenland, and the Sea of Okhotsk (Figure 4).
Figure 4: Air temperature anomalies in January 2025 compared to the long-term January mean for 1971 – 2000. Based on daily mean temperatures measured at an altitude of ca. 760 metres.
Strong winds produce extensive polynya off the northern coast of Greenland
In the Wandel Sea off the northern coast of Greenland, from the last week of January offshore winds with peak speeds of up to 131 kilometres per hour pushed apart the pack ice to such an extent that, for the second time in the past ten years, a narrow, 500-kilometre-long coastal polynya formed (Video 1). At this time, air temperatures over the Wandel Sea were near the freezing point.
In February 2018, experts had observed a similar interplay of atmosphere, sea ice and ocean in the same region. In a related study, they concluded e.g. that the warm winds were due to a sudden warming of the atmosphere at an altitude of 10 to 50 kilometres, and that said winds had to be intense and sustained in order to drive the (predominantly thick) pack ice so far from the coast.
Video 1: This animation of sea-ice movements on the Arctic Ocean shows how, in the first days of February, a long, channel-like polynya formed off the northern coast of Greenland. This was chiefly due to powerful offshore winds and air temperatures near the freezing point. (GIF: Lars Kaleschke, AWI)
The Antarctic: Conditions slightly improved
In the Antarctic, the summer sea-ice melting continued in January. If we compare the sea-ice situation at the end of the month with the winter sea-ice extent in September 2024, only a quarter of the region is still covered with pack ice. In the eastern Weddell Sea, the sea ice has now completely melted or drifted to the north or west in response to wind and currents. The satellites also picked up very little sea ice in the Ross Sea. The exception: ice floes that, as narrow strips of fast ice, are frozen to the calving front of the Ross Ice Shelf, and extensive floe fields far off the coast of Marie Byrd Land (Figure 6).
Figure 6: Mean sea-ice concentration in the Antarctic in January 2025. The turquoise line indicates the mean position of the ice margin from 1981 – 2010 and illustrates the unusually large remaining ice areas in the northern Weddell Sea and off the coast of Marie Byrd Land (western Ross Sea), as well as the lack of sea ice in the Bellingshausen Sea.
Figure 7: Development of mean sea-ice extent in the Antarctic for the month of January. The light blue line represents the long-term decline in sea ice. After eight years below the trend line, for the first time the monthly mean value was above the line again in January 2025.
Figure 8: Difference in the mean position of the ice margin in January 2025, compared to the long-term mean for the years 2003 – 2014. Regions marked in blue had more Antarctic sea ice than the reference period; those marked in red had less.
In January, the mean sea-ice extent on the Southern Ocean was 4.67 million square kilometres. As such, although lower than the long-term mean for the period 1981 – 2010 (4.95 million square kilometres), it was still within the span of the minima and maxima during the same period. Since the beginning of satellite observations, there have been 19 Januaries in which the mean monthly value for Antarctic sea ice was lower than that in January 2025. As such, after a winter characterised by extremely little sea ice, the situation has now improved somewhat (Figure 7).
Nevertheless, it’s worth taking a closer look at the ice distribution (Figure 8). If we compare the position of the ice margin in January 2025 with the long-term mean for the period 2003 – 2014, we can see that at the beginning of the year, particularly in the northern Weddell Sea, in the Bellingshausen Sea, the Amundsen Sea, and off the coast of East Antarctica, there was significantly less pack ice than in the reference period. In contrast, there were more ice floes off the coast of Marie Byrd Land and in some parts of the Weddell Sea. Local winds and ocean currents can produce such differences in sea-ice distribution.
Platelet ice: A small but significant difference
With the breakup of the sea ice in Atka Bay, January also marked the end of the sea-ice monitoring season for the overwintering team at the German Antarctic research station Neumayer III. From June to December, members of the team venture onto Atka Bay by snowmobile once a month as part of the long-term research project “Antarctic Fast Ice Network. At six sites, they drill holes in the sea ice, measure the ice thickness, collect ice samples and, at fixed locations, measure the seawater’s temperature and salinity down to the seafloor, up to 250 metres below (Figure 9).
In November, when the first summer guests begin to arrive at the station, the sea-ice programme is then extended, since not all Antarctic sea ice is created equal: there is a small but significant difference between floes that drift far out on the open ocean, and the coastal sea ice that often freezes onto the calving fronts of major ice shelves in winter. On the underside of the latter, generally referred to as fast ice, metre-thick layers of delicate ice crystals – platelet ice – can form under the right conditions (Figure 10 and Video 2).
Figure 9: Since 2010, members of the overwintering team have surveyed the sea ice in Atka Bay at regular intervals. Accordingly, the experts at the AWI’s sea-ice physics section have a good overview of the sea ice’s qualities: on average, it is 2 metres thick, topped by 80 centimetres of snow cover, and has a 4-metre-thick layer of platelet ice on its underside. (Photo: Stefanie Arndt, AWI)
Figure 10: Though delicate, platelet ice can measure up to palm size. It only accumulates below sea ice when supercooled ice-shelf water flows out from beneath the shelf. (Photo: Mario Hoppmann, AWI)
Video 2: Platelet ice can form up to 10-metre-thick layers below the sea ice. This underwater footage, taken in Atka Bay, shows just what these layers look like “up close and personal”. (Video: Centre for Scientific Diving, AWI)
This platelet ice is formed when relatively warm and high-saline seawater flows under the ice shelf and slowly melts away its underside. These melting processes can even result when the seawater has a surface temperature near the freezing point (-1.85 degrees Celsius), as the freezing point drops in line with water depth (e.g. to -2.5 degrees Celsius at a depth of 800 metres).
Due to the melting on the underside of the ice shelf, a mixture of (not briny) meltwater and high-saline seawater is formed. Referred to as ice-shelf water, its temperature lies below the freezing point at the water’s surface, yet its density is lower than that of the surrounding seawater. As a result, it slowly rises to the underside of the ice shelf, with the water pressure gradually declining in the process. As this happens, the water’s freezing point automatically rises.
Once it has reached a critical depth, the supercooled water begins to form tiny ice crystals. The crystals in turn slowly form thin wafers, some coin-sized, others hand-sized. As platelet ice, they accumulate directly below the fast ice (Figure 11). And that means: the amount and distribution of the platelet ice below the fast ice allow us to draw conclusions regarding the interplay between seawater and ice shelf, as well as the current state of the glacier tongue.
Figure 11: This graphic illustrates the formation process of platelet ice. (Graphic: adopted to Mario Hoppmann, AWI)
Figure 12: This time series shows the mean thickness of the sea ice (black line), the snow cover atop it (light blue line), and the platelet ice (dark blue line) below it. In years with especially thick sea ice, icebergs blocked Atka Bay. As a result, the ice cover didn’t break up and drift away in January. (Figure: Stefanie Arndt, AWI)
Sea-ice measurements in Atka Bay: First-ever map of platelet ice distribution
On the basis of the countless sea-ice measurements taken over the years in Atka Bay and a targeted platelet ice measuring campaign in summer 2022/23, sea-ice physicists at the Alfred Wegener Institute have now created the first map of the platelet ice distribution in Atka Bay. The data was partly gathered during hour-long treks using a multi-frequency EM device. Towed over the ice on a sledge, it constantly measures the electromagnetic conductivity of the surrounding sea ice, platelet ice and seawater. Since all three media have distinctive conductivities, the researchers can subsequently filter their data to clearly distinguish between the platelet ice (medium conductivity), sea ice (low conductivity) and water (high conductivity).
We used numerous ice-thickness readings taken at boreholes to validate our electromagnetic data and are now proud to say that large-scale campaigns with our multi-frequency EM approach offer an excellent method for recording the distribution and amount of platelet ice under the fast ice without any interference.
In summer 2022/23, she personally spent days riding a snowmobile to survey 25-kilometre-wide Atka Bay with the measuring sledge behind her (Figures 13 and 14). Later, while analysing the unique data packet, she was in for a surprise: she didn’t find the thickest platelet ice layers on the western edge of the bay as expected, but to the southeast, right by the edge of the Ekström Ice Shelf. “The platelet ice there was more than 10 metres thick in some places and so compact that, when we tried drilling into it, our thickness probe couldn’t reach the bottom,” Mara Neudert recalls (Figure 15).
Figures 13/14: These two figures show the geographic position of Atka Bay (top) and the course taken by Mara Neudert and her colleagues to survey the bay with the measuring sledge in November and December 2022 (bottom). The black dots denote boreholes used for validation readings. The shaded areas indicate where the Ekström Ice Shelf rests on the seafloor and where not. (Figure: Neudert, M. et al. 2024)
Figure 15: This graphic shows the distribution and thickness of the platelet ice below the fast ice of Atka Bay. The brighter red the shade is, the thicker the platelet ice layer is. The especially thick platelet ice under the fast ice at the southeast edge of the bay indicates that ice-shelf water (with temperatures below the surface freezing point) flows out of a cavern below the Ekström Ice Shelf there. This water is what forms platelet-ice crystals. (Figure: Neudert, M. et al. 2024)
Until this discovery, the AWI experts had assumed that the platelet ice in Atka Bay mostly formed below the western ice shelf and was then washed into the bay. But the new distribution map supports the conclusion that the ice-shelf water can also be found under the Ekström Ice Shelf in the eastern part of the bay – at exactly the point where, for a short distance, the massive glacier tongue does not rest on the seafloor. Here, there is a connection to a natural cavity (cavern) below the ice shelf, where comparatively warm seawater can melt the shelf from below.
The AWI’s sea-ice physicists will now use the map of the amount and distribution of platelet ice in Atka Bay in summer 2022/23 as the point of departure for further observations. Moreover, they will soon analyse all ocean data from Atka Bay and combine it with the sea-ice and platelet-ice data so as to identify potential interactions between the ocean, ice shelf and sea ice. They can now also measure the platelet ice from the air – using research aeroplanes and helicopters to tow the electromagnetic measuring device EM-Bird on a cable behind or below them, which makes large-scale measuring campaigns possible. As such, the story of the platelet ice in Atka Bay and the surrounding Weddell Sea region is far from over.
Contact
Author
Sina Löschke (science writer)
Questions?
Write us an email or use the contact form.
Graphics
