Skip to main content

Morning Overview

Antarctic sea ice topped out at 17.59 million square kilometers, the third-lowest maximum on record

Sea ice around Antarctica most likely reached its 2026 winter maximum on September 14 at 17.59 million square kilometers, or 6.79 million square miles, according to the National Snow and Ice Data Center in Boulder, Colorado. If that date holds, the center says, it will be the third-lowest Antarctic sea ice maximum in the satellite record that began in 1979.

The four lowest Antarctic maxima in a 48-year record are now the last four winters.

Where 17.59 million sits in the record

The reference point is 2023. NSIDC’s report on that season gave a maximum of 16.96 million square kilometers on September 10, 2023, more than 1 million square kilometers under the previous record low from 1986. The 2026 peak is about 630,000 square kilometers above that floor, which is close to the bottom of the range yet still far under the long-term average, and the third-lowest ranking places 2026 firmly inside the run of low winters that began with 2023.

Last year’s peak gives the near comparison. NSIDC’s 2025 summary recorded 17.81 million square kilometers on September 17, 2025, which was third lowest at the time with 2023 as the record and 2024 second. The 2026 figure is 220,000 square kilometers below it, which pushes 2025 down to fourth. Senior research scientist Ted Scambos of the Cooperative Institute for Research in Environmental Sciences at CU Boulder said in that release that the lengthening trend of lower Antarctic sea ice “poses real concerns regarding stability and melting of the ice sheet,” while noting it may also bring more snowfall over the continent.

The 2025 peak was 900,000 square kilometers below the 1981-2010 average, NSIDC said, with conditions markedly below average in the Indian Ocean and the Bellingshausen Sea and slightly above average in the Ross Sea.

A peak date that could still move

Antarctic sea ice does not peak cleanly. The NSIDC analysis published September 23 describes large fluctuations in extent near the seasonal maximum, which can delay confident identification of the peak, and that is why the wording is “most likely” and “if this maximum date holds.” NSIDC’s 2025 release named the risk explicitly: changing winds or late-season growth could still raise extent, as happened in 2002 and 2017.

The United Kingdom’s Met Office reached the same ranking by a different route. Its September 21 report uses its own data record and its own numbers, a maximum of 18.12 million square kilometers on September 11, and says that if not subsequently exceeded it would be the third-lowest maximum of the satellite era. It also notes that many years have set their maxima after September 20. Two independent records agreeing on third place is the strongest argument that the ranking will survive, though the Met Office’s absolute number is not comparable with NSIDC’s because the two agencies process the satellite data differently.

The arithmetic of the ranking is simple, and it shows how narrow the margin is. Should late-season growth carry extent past 17.81 million square kilometers, the 2025 value, 2026 would slip to fourth. Nothing in the September 23 analysis suggests that is under way, and the Met Office’s reference to late maxima is a standing caution rather than a forecast.

Four low winters and the scientists reading them

The pattern matters more than any single figure for the peak, and the researchers quoted keep returning to it. In Carbon Brief’s account of the season, which also notes that the 2026 extent remains substantially below the decadal averages of the 1980s through the 2010s, Alfred Wegener Institute sea ice researcher Lettie Roach said it is “concerning to see these low values persist” after decades in which Antarctic ice had been relatively steady. She cautioned that deciding whether the continent has undergone a “structural shift” needs more years of data, because natural variability makes attribution to human-caused warming hard.

Clare Eayrs of the Korea Polar Research Institute pointed to the monthly record: “all five of the lowest July extents have occurred since 2022 and all four of the lowest August extents since 2023.” Regionally, the 2026 season swung. The Bellingshausen Sea began the growth season nearly ice-free while the Amundsen Sea ran above average, and by late August the pattern had reversed with a shift in atmospheric pressure.

The same NSIDC analysis covers the opposite pole, where the Arctic minimum of 4.60 million square kilometers on September 12 tied for tenth lowest, and where the last 20 years, 2007 through 2026, are the 20 lowest extents on record. The poles are not behaving alike: the Arctic decline is steady at 11.8 percent per decade against the 1981-2010 baseline, according to the same analysis. Whether four consecutive low Antarctic winters mark a new baseline or a cluster is the question Roach says needs more years to answer.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.


More from Morning Overview