A rapid series of low-pressure systems can create weather conditions that make it difficult for ice to recover, affecting Arctic coastlines and the global climate system.
Sea ice in the Arctic can recover from a single storm, but repeated cyclones can make that recovery much harder. When several low-pressure systems sweep through the same area in quick succession, they can prolong warm, turbulent conditions that break apart ice and prevent newly opened gaps from freezing over.
An international team led by the Alfred Wegener Institute has now systematically examined this phenomenon using decades of weather and satellite observations. Their study, published in Nature Communications, describes how this process, known as ‘cyclone clustering,’ affects Arctic sea ice and may increase risks along Arctic coastlines.
‘Cyclone clustering’ refers to several low-pressure systems passing rapidly through the same region one after another, creating storms that are stronger and longer lasting than ‘normal’ storms. Similar sequences have already caused serious impacts elsewhere.
“In Europe, such events have frequently led to significant damage in the past due to high winds, heavy rainfall or high tide levels along the coast,” says Dr Lars Aue, lead author of the study from the Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI).
To determine how often these clusters occur in the Arctic and what they do to sea ice, the researchers analyzed satellite and weather records covering 1979 through 2024. They modified an existing weather tracking algorithm so it could identify Arctic cyclones arriving in rapid succession, then compared satellite images of the ice before and after each cluster.
“We adapted an algorithm for tracking weather systems so that it could detect rapidly consecutive cyclones in the Arctic,” explains Lars Aue. “We were then able to compare the state of the sea ice on satellite images before and after a cluster had passed through.”
Clustered storms keep ice gaps open longer
When cyclones pass over Arctic sea ice, they can fracture the frozen surface and push ice floes into motion, causing them to drift and collide. In winter, the resulting openings would normally freeze again within several days. A cluster of storms can prolong the disturbance, however, delaying or preventing those gaps from closing.
The effects change with the seasons. Cyclone clusters tend to be weaker during warmer months, but they can still substantially alter sea ice locally by driving floes northward and compressing them against existing ice, reducing the total area covered.
The storms can also bring warmer air that melts ice from above and draw warmer seawater upward from deeper layers, heating the surface ocean. Over the past two decades, cyclones have contributed to sea ice loss particularly near the end of the melting season.
The size of a cyclone cluster varies by region and season, averaging about 2.5 storms. Its effect on sea ice also differs considerably from place to place. In some regions, the impact is amplified by as little as 1.5 times, while elsewhere it can reach seven times the effect of an isolated storm.
“On average across the entire Arctic, cyclone clusters reduce Arctic sea ice cover by about twice as much as conventional, isolated low-pressure systems. This condition also lasts about two and a half times longer,” says Lars Aue.
Thinner ice is becoming more vulnerable
The researchers also found that sea ice losses associated with cyclone clusters have increased significantly in recent decades. One likely reason is that Arctic ice is becoming thinner and more mobile as the planet warms, leaving it less able to withstand repeated storms.
During summer, increasingly warm Arctic Ocean water may intensify the problem further.
“We could be entering a self-enforcing feedback loop: the weaker the sea ice becomes, the more cyclone clusters can reduce its extent, which in turn weakens the sea ice,” says Lars Aue. “Currently, we are using climate projections to investigate whether the intensification of the effects of cyclone clusters that we have seen in recent decades could continue in the future.”
Retreating ice leaves Arctic coasts exposed
The findings show how strongly short-term weather can influence Arctic sea ice and why understanding cyclone clusters matters beyond the ice itself. Arctic coastal communities face growing exposure to severe weather as cyclone activity increases because sea ice normally helps shield shorelines from waves.
As that protective ice retreats while permafrost continues to thaw, Arctic coasts become increasingly vulnerable to storm damage.
“Until now, there has been no systematic analysis of the impacts of cyclone clusters in the Arctic. With our work, we are filling a knowledge gap and lay an important foundation for projections that can depict the future of the Arctic sea ice – a central component of the global climate system – as accurate as possible.”
Reference: “On the relevance of serial cyclone clustering for Arctic sea ice” by Lars Aue, Sofie Tiedeck, Peter Finocchio, Timo Vihma, Petteri Uotila, Gunnar Spreen and Annette Rinke, 5 August 2026, Nature Communications.
DOI: 10.1038/s41467-026-76245-5
This research was supported in part by a grant NSF PHY-2309135 to the Kavli Institute for Theoretical Physics (KITP). Open Access funding enabled and organized by Projekt DEAL.
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
8 Comments
“A rapid series of low-pressure systems can create weather conditions that make it difficult for ice to recover, …”
There is little doubt in my mind that it isn’t something that can’t be remedied by some very cold weather. However, we are in an interglacial. The last major Pleistocene glacial advance in North American, commonly called the Wisconsin, reached its maximum more than 20,000 years ago; the ice sheets were at least a mile deep, if not 2 miles in the primary zone of accumulation.
The Holocene, the current warm epoch is defined as starting about 11,800 years ago. While it isn’t warming uniformly or monotonically, clearly, Earth is warmer than it was 12 millennia ago. Why would anyone be surprised if sea ice is also declining?
“To determine how often these clusters occur in the Arctic and what they do to sea ice, the researchers analyzed satellite and weather records covering 1979 through 2024.”
In a geologic context, 45 years is a blink of an eye. It is long enough to be considered “climate,” instead of weather, which is considered to currently be experiencing an interglacial! Arctic ice hit an annual low in 2012, and may be experiencing a recent warming on top of the long-term climate. It is “much ado about nothing.” It is akin to reading tea leaves and pretending that one can actually discern long-term changes foretold by drinking a single cup of tea.
“The storms can also bring warmer air that melts ice from above and draw warmer seawater upward from deeper layers, …”
I think that this claim needs some explanation. How does warm surface air draw warmer seawater from the depths? The warm surface air can contribute heat energy to supply the latent heat of melting energy necessary to melt the ice, but in the wake of the ice is left relatively fresh brine, which is less dense than seawater, and, therefore, caps the seawater. My mind struggles to understand the claim.
“How does warm surface air draw warmer seawater from the depths?” – Clyde
“[D]raw” is a plural verb, “air” is a singular noun. Apparently you had forgotten the sentence opener, “The storms” before you finished reading the sentence. The storms draw warmer seawater from the depths in a process known as upwelling via their winds. I know you’re familiar with upwelling because you invoked it in a comment on another article. Upwelling can bring up cold, dense seawater despite the cap of less dense water along the west coast of South America, for example, and you don’t seem to have a problem with that.
You are so focused on correcting my grammar that you overlook the fact that you didn’t really answer my question and you contradicted yourself. The storms (subject of the sentence) are characterized by their winds. Yes, those winds can result in up-welling if they have a general trend that displaces the surface meltwater. You acknowledge that the “Upwelling (sic) can bring up COLD, dense seawater.” However, my question was, how can “The storms … draw WARMER seawater upward from deeper layers?” when even in the mixed layer above the thermocline, the water generally gets colder with depth during the melting season. During the middle of Winter, the air might be -40 degrees and any open water will freeze very quickly. During the melting season, which is really the concern, the surface water can get well above freezing. The authors even acknowledge that “Cyclone clusters tend to be weaker during warmer months”. Another question raised by this discussion is how effective are cyclones in inducing up-welling compared to trade or prevailing winds such as are common on the west coasts of North and South America?
Lol. Since when was South America in the Arctic? I was introducing the concept of upwelling using a location more readers are likely to have heard of it occurring in, and at the same time showing an example of how powerful wind can be.
When ice is present, the water immediately underneath is fresher and extremely cold, at the freezing point (the ice is, after, insulating the water from temperatures that are often much colder), followed by a halocline, and underneath that, warmer, saltier water from the Atlantic. You’re not going to get a thermocline until the ice is gone, and that’s not the topic of this paper.
And it’s upwelling, not up-welling.
“Since when was South America in the Arctic?”
It isn’t, nor is up-welling at Monterey Bay the Arctic. That is the point. The better known examples don’t fit the model of cyclonic winds being responsible. Do you have an alternative well-known example of cyclonic winds causing up-welling?
When I write “upwelling” here it gets flagged and Windows OS gives me a choice of “up-welling” or “up welling.” Do you not get a similar notice? Are you using a Mac? The American Heritage Dictionary recognizes “upwelling” and NOAA and Wikipedia use that form. I’m just trying to go with the flow. I’ll gladly conform to your convention.
Minimum Sea Ice Extent has declined roughly 2.5 million km2 between 1979 and 2025, or about 54,000 km2/year. Assuming 100% of the Arctic Ocean’s 14 million km2 surface area was ice-covered during the summer at the peak of the last glacial and taking your notion that the current decline is simply a continuation of the post-glacial decline, we arrive at the absurd notion that the end of the last glacial maximum was 177 years ago, rather than the generally accepted 19-20 thousand years ago. In other words, it’s declining 108x faster than a naive linear extrapolation, and from that alone we can disregard your claim.
But you’ve also conveniently ignored that there other data besides from satellite, including proxy records related to temperature, the remains of species that can’t live with ice, or conversely, can’t live without, and historical records. Recent efforts to assimilate such data into Arctic sea ice reconstructions over the last two millennia naturally have large uncertainties, but are still clear enough to show that from 750 and 1820 CE there was a long term increase in sea ice, directly the opposite of putative long-term post-glacial decline, and significant increases due to large volcanic eruptions and the Little Ice Age. These pale in comparison to the current decline. See Brennan, M. K., and G. J. Hakim, 2022: Reconstructing Arctic Sea Ice over the Common Era Using Data Assimilation. J. Climate, 35, 1231–1247
Unfortunately, we don’t have good data on what the Arctic ice coverage was like prior to the imaging satellite era and we have to rely on proxies that are of unknown veracity. There are some anecdotal reports from whaling captains suggesting that there were times in the past when the coverage was much less then now. Recent data ( https://i0.wp.com/wattsupwiththat.com/wp-content/uploads/2026/08/Arctic-August-27-2026-768×371-2.png ) suggest that the annual minimum has been static for about the last 2 decades, after a notable decline. Extrapolation is always fraught with risk leading to unreliable estimates.
Because the Arctic basin is fixed in area, comparing area at present to much longer back in time, we are left with the impression that the coverage was at 100% for a long time. However, the reality is that when the ice filled all the available area, it just got thicker when it got colder. In other words, today, the percentage of the area of the Arctic basin is a convenient reference. However, it tells us little about the volume of ice during colder regimes.