News detail view

Changed ice-drift patterns explain the surprising sea-ice development

In order to understand why the summer melting of the Arctic was so moderate this year, we need to turn our attention to the major ice movements in the Arctic Ocean and look back to the year 2022.

“For the period from 2022 to 2024, our ice-drift data shows an unusually pronounced Beaufort Gyre, which transported multiyear sea ice in particular clockwise from the Canadian Arctic and over the North Pole, toward Fram Strait,” explains AWI sea-ice expert Dr Thomas Krumpen. At the same time, the sea-ice migration from the Russian marginal seas toward Fram Strait, commonly known as the Transpolar Drift, simply didn’t happen. “Winds and currents kept the sea ice formed in the Laptev Sea trapped in the eastern Arctic Ocean for more than two years, which allowed it to survive the summer and gradually grow thicker,” says Thomas Krumpen (Figure 2).

Toward the end of 2024, these atypical wind and drift conditions then reversed: the Beaufort Gyre grew substantially weaker, while powerful offshore winds over the Laptev Sea reactivated the Transpolar Drift. As Thomas Krumpen puts it: “The ice drift essentially switched back to its normal mode”. As a result, the thick multiyear pack ice that had accumulated before the Laptev Sea drifted into the Central Arctic. From there, part of the ice drifted toward the northern coasts of Greenland and Canada, where it will likely continue to grow; the remainder drifted toward Fram Strait, where it will be exported into the North Atlantic.

“We’ve essentially observed a massive redistribution of comparatively thick multiyear pack ice in the Arctic Ocean since the end of 2024, which was only possible because the Beaufort Gyre had previously been intensified by a natural fluctuation. There was a similar development in the early 1990s and we’re currently assessing the atmospheric causes of this unusual drift pattern,” Thomas Krumpen relates.

Three reasons why the projected record-low minimum never happened

The redistribution of the old, thick ice from the Laptev Sea was a major contributing factor to the fact that the expected record-low sea-ice minimum didn’t come to pass. But other factors were also important – including the winds over the Arctic Ocean. This summer, they were relatively weak and also came from the wrong direction to drive large amounts of pack ice into the warmer marginal zones of the Arctic Ocean, where they would have rapidly melted. In addition, air temperatures in the Central Arctic remained fairly moderate through the end of July. Consequently, much less sea ice melted than Thomas Krumpen and his colleagues had anticipated.

Alternating ice-drift patterns in the Central Arctic and a summer sea-ice extent far from the record-low minimum match the findings of a new modelling study recently released in the journal Geophysical Research Letters. In the study, British researchers show that the delayed decline in Arctic sea ice over the past 20 years can likely be attributed to natural fluctuations in the climate system. They surmise that the fluctuations’ stabilising effect could continue for another five to ten years. At the same time, they make exceptionally rapid sea-ice decline in future years more probable.

The CONTRASTS expedition: The many facets of the melting season

How the thickness and geographic position of the Arctic pack ice affect its chances of surviving the summer can be seen in the initial outcomes of CONTRASTS, this year’s summer expedition with the research icebreaker Polarstern. The participating researchers had the chance to deploy monitoring devices on three separate ice floes in the northern Fram Strait, and to document their melting behaviour and all relevant environmental parameters, longer than ever before – both above and below the ice.

“Our data impressively shows that the melting season isn’t a linear process, as is often assumed. Instead, its progression is greatly influenced by individual weather events,” says Dr Marcel Nicolaus, cruise leader for the CONTRASTS expedition and a sea-ice physicist at the AWI. “For instance, it makes a considerable difference whether the precipitation is rain that falls on the sea ice’s surface, producing extensive melting in just one day, or whether it’s snow that falls on the ice, making its surface lighter and, in the next step, automatically colder,” he explains.

The researchers observed rain and moist air particularly in the eastern part of the northern Fram Strait – waters that are directly influenced by the inflow of air and water masses from the Atlantic (Figure 6). According to Marcel Nicolaus: “One day in August, we recorded an air temperature of 13 degrees Celsius at an altitude of 300 metres. The heat signal most likely came from the heatwave in Northern Europe. That being said, it only reached the Arctic at high altitudes and not at the ocean’s surface, where it could have come into contact with the ice.”

In contrast, the pack ice to the west was unaffected by heat from the North Atlantic and more of it survived the melting season (Figure 7). When the first snow fell on the ice in mid-September, the ocean’s surface and ice cooled surprisingly quickly. “Within a week, all patches of open water were covered with new sea ice. As such, the summer was gone virtually overnight,” reported Marcel Nicolaus from on board the research icebreaker Polarstern on 18 September. There, he’s coordinating the sea-ice observations for the follow-up expedition to CONTRASTS.

He’s also using his time on board to analyse the sea-ice data gathered during CONTRASTS. “I get the impression that the sea-ice melting was initially chiefly driven by what went on at the surface, and therefore by the air temperatures and sunlight. The ocean only became a factor once the ice concentration had dropped and the areas of open water were large enough to absorb sunlight. That’s when the ocean began gnawing away the ice at the floes’ edges. The process was much more intense in the east than in the west, where we had relatively thick, seamless ice cover throughout the expedition, so the ocean had no chance to do so. Consequently, one critical factor for the scale of summer melting is whether the ocean is covered by a thin but seamless layer of ice that reflects the incoming radiation, or whether there are patches of open water that absorb that energy,” the sea-ice physicist explains.

Spotlight on the winter sea-ice extent

The broad range of natural fluctuations in the Arctic climate system and the slowed sea-ice decline in summer raise a new question for the AWI’s sea-ice physicists: What do the differing progressions of the winter and summer time series tell us about feedbacks between the sea ice and climate change?

“Whereas, based on summer data, the sea-ice decline has stagnated for the past 15 years, with the extent actually growing slightly, the winter extent has been steadily declining for decades,” says AWI sea-ice physicist Dr Luisa von Albedyll. “However, the level of decline in winter is comparatively low and we don’t see the major outliers – phases characterised by dramatic ice loss – that dominate the development in summer.” (Figure 8)

These extreme events (‘rapid ice loss events’) occur when natural fluctuations produce the same effect as the climate-based downward trend in sea-ice extent. “Under these conditions, atypical atmospheric or oceanic circulation patterns can become pivotal forces and produce rapid and intensive melting, which happened in the period from 2007 to 2012. In turn, the circulation patterns themselves are subject to the substantial natural fluctuations in the climate system,” Luisa von Albedyll adds.

However, a recently released modelling study shows that these extreme events will be more likely, even in winter, if the ice volume continues to decline. Ice volume and ice thickness are two essential keys to understanding how quickly the sea ice reacts to external influences.

The Antarctic: Sea-ice extent reaches a winter maximum of 18.06 million square kilometres

While the Arctic sea-ice extent reached its summer minimum in the first half of September, the Antarctic sea ice reached its winter maximum at mid-month – more precisely, on 18 September according to our data, when it measured 18.06 million square kilometres, putting it in third place on the list of the lowest extents. The monthly mean sea-ice extent in the Antarctic was 17.84 million square kilometres – clearly above the September mean in 2024, but still more than 500,000 square kilometres below the trend line (Figures 9 & 10).

More sea ice than in September 2023 and 2024 could above all be found in the eastern Weddell Sea and adjacent waters to the east, and in the Ross Sea. In contrast, satellites registered less sea ice than the previous year off the coast of East Antarctica, and in the sector of the Bellingshausen and Amundsen Seas (Figure 11).