Clues from a warmer Arctic 6,000 years ago suggest the region could lose a million square kilometers more winter sea ice than scientists previously expected.
The loss of Arctic sea ice has been linked to colder winters and heavier snowfall in the UK, the rest of northern Europe, and North America. The ice regulates how much heat and moisture the ocean releases into the atmosphere, so changes in its cover can influence weather far beyond the polar region.
Climate models disagree most about how much ice will cover the Arctic in winter, making its future particularly difficult to predict. Using evidence from a warmer Arctic 6,000 years ago, researchers have now revised those projections downward by about one million square kilometers, or roughly 386,000 square miles. The findings, published in Nature Geoscience, come after the Arctic recorded its lowest maximum sea ice extent since modern measurements began in March 2025, a record tied again this year.
Ancient algae sharpen Arctic ice forecasts
Researchers from the University of Cambridge and the Centro de Investigación en Matematicas (CIMAT) in Mexico compiled the largest-ever database of a key substance produced by algae living in sea ice. Called IP25, this biomarker was measured in ocean sediment cores dating to the mid-Holocene, when subtle shifts in Earth’s orbit contributed to warmer Arctic conditions and reduced ice cover. Alongside IP25, the team measured complementary biomarkers from organisms that live in open water. Together, these traces allowed them to estimate how much sea ice covered the ocean thousands of years ago.
“The scientific community has amassed lots of paleoclimate records over recent decades, but it’s remained a challenge to directly link these records to future projections, since there is uncertainty at both ends,” said lead author Crystal Fu, from Cambridge’s Scott Polar Research Institute.
By comparing their reconstruction with simulations of the mid-Holocene from 27 climate models, the researchers could assess how closely each model reproduced the ancient ice conditions. They used a statistical method called Bayesian weighting to give greater influence to models that matched the evidence more closely. Those weights could then be applied to simulations of the recent past and future because the same physical processes govern sea ice across different climates.
Although the mid-Holocene warming had a different cause from today’s warming, Fu said the ice would respond in the same way. “It doesn’t matter if the warming is natural variation or caused by human activity: if it warms, sea ice is going to melt,” she said.
Less winter ice than previously projected
In the historical simulations, the weighting removed the models’ collective tendency to overestimate winter sea ice, while the range of uncertainty in the projections narrowed by as much as 85%.
“When we apply the model weights to historical simulations, we completely eliminated the positive bias,” said Fu. “The future projections similarly end up being lower than previous estimates, by about a million square kilometers.”
Applying the revised weights to precipitation projections also suggested that snowfall over southeast Greenland and rainfall in the Mediterranean may be underestimated. The researchers described these as hints that effects beyond sea ice could be larger than expected, with possible implications for the Greenland ice sheet’s stability and future flood risk in southern Europe.
“By using evidence from the past, we can now produce more reliable maps of where Arctic winter sea ice is likely to disappear,” said co-author Marco Aquino-López, of CIMAT. “These give us a much clearer picture of what the Arctic might look like in the coming decades, and how much uncertainty remains.”
Reference: “Palaeoclimate constraints narrow uncertainty in future Arctic winter sea ice projections” by C. Y. Fu, M. B. Osman and M. A. Aquino-López, 16 October 2026, Nature Geoscience.
DOI: 10.1038/s41561-026-02123-w
The research was supported in part by the US National Science Foundation and the Cambridge Trust. Crystal Fu is a Member of Girton College, Cambridge.
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.