Antarctic sea ice has recorded its third-smallest winter peak extent since satellite observations began 48 years ago, according to new data that underscores persistent low values at both of Earth’s polar regions.
Provisional data released by the US National Snow and Ice Data Center (NSIDC) shows that Antarctic sea ice hit its annual winter peak on September 14, reaching an extent of 17.59 million square kilometers. While the organisation noted that the figures remain preliminary and that large fluctuations in extent are typical of Antarctic sea ice near its seasonal maximum, polar scientists have voiced concern over the ongoing trend of depressed ice levels.
The latest findings arrive as researchers continue to monitor systemic changes across both the Southern and Northern Hemispheres. Meanwhile, at the Earth’s opposite pole, Arctic sea ice reached its annual summer minimum on September 12, ranking as the joint-tenth lowest in the nearly five-decade satellite record. The NSIDC highlights a striking long-term trajectory in the Arctic, noting that the past 20 years have recorded the 20 lowest Arctic sea ice extents in the satellite era.
Polar climate experts emphasize that while the multi-decade decline in the Arctic is clearly attributable to human-caused climate change, the mechanisms driving recent disruptions in the Antarctic remain more complex, blurring the lines between anthropogenic warming and natural climate variability.
Dr. Lettie Roach, a polar climate scientist at the Alfred Wegener Institute in Germany, noted that this year’s Antarctic maximum remained well below the historical seasonal average. Following several decades of relatively stable or even slightly increasing winter sea ice conditions in the Southern Ocean, she described the persistence of these low values as concerning.
"Compared with the Arctic, it’s less clear how recent changes in Antarctic sea ice are attributable to human-caused warming versus natural variability," Dr. Roach explained. "We need more years of observations to better understand whether this is truly a structural shift."
Dr. Clare Eayrs, a postdoctoral researcher at the Korea Polar Research Institute (KOPRI), pointed out that after recording a near-average minimum extent in February, Antarctic sea ice quickly reverted to unusually low winter coverage. Monthly tracking underscores the anomaly: Antarctic sea ice extents for July and August ranked as the fifth and fourth lowest on record, respectively. Notably, all five of the lowest July extents have occurred since 2022, and all four of the lowest August extents have been recorded since 2023.
Dr. Eayrs also highlighted how regional patterns of Antarctic sea ice cover shifted substantially over the course of the growth season. In April, for instance, the Bellingshausen Sea remained almost entirely ice-free, whereas the neighbouring Amundsen Sea held more ice than usual. However, by late August, shifting atmospheric pressure systems and wind patterns completely altered the distribution. The Bellingshausen deficit largely recovered, while ice became unusually scarce in the Amundsen Sea and across vast stretches of East Antarctica.
Arctic Minimum
In the Northern Hemisphere, the Arctic recorded its summer minimum sea ice extent on September 12. At 4.60 million square kilometers, the 2026 minimum ties with 2025, 2010, and 2008 as the tenth-lowest summer sea ice extent in the satellite record.
While this year’s minimum did not break all-time records, Dr. Roach emphasized that it remains significantly lower than any sea ice minimum recorded prior to 2007. Human-caused climate change has fundamentally reduced Arctic sea ice cover and thickness, leaving the region structurally transformed compared to just a few decades ago.
Tracking the health of the Arctic ice pack has recently faced technical hurdles. Since 1979, scientists have relied on a reanalysis system known as the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) to monitor sea ice volume and thickness. However, in March 2026, the National Oceanic and Atmospheric Administration (NOAA) terminated a critical global air pressure dataset that researchers depended upon to generate PIOMAS, forcing a halt to updates for both systems.
Developers of PIOMAS have stated that finding alternative data sources for their reanalysis will require considerable time and effort, leaving uncertainty over whether funding and available datasets will allow the time series to resume.
Compounding these challenges, August brought record-breaking global heatwaves that extended deep into Arctic coastal regions. These high temperatures converged closely with areas experiencing substantial sea ice loss, particularly the Barents-Kara and East Siberian seas.
Dr. Zack Labe, a scientist at Climate Central, noted that the absence of a new record-low minimum this year should not be misinterpreted as a sign of increasing resilience in the Arctic ice pack. Local weather patterns continue to exert a powerful influence on year-to-year extent, even as the multi-decade downward trend remains unequivocal.
Summer weather across the Arctic was heavily influenced by lower atmospheric pressure centered over the region, which introduced cloudier and cooler conditions that suppressed surface melting and delayed the start of the melt season. This was particularly evident across parts of the Beaufort and Chukchi seas, where the melt season did not effectively begin until after early July—more than two weeks later than normal.
Conversely, the Atlantic side of the Arctic experienced a much more extreme summer. Sea ice in the Barents Sea recorded its earliest melt-out on record, driven by unusually warm air and ocean temperatures. Western Siberia, for instance, endured persistent temperatures more than 5 degrees Celsius above the 1981-2010 average throughout much of the summer, extending outward over the Kara Sea and accelerating local ice loss.
As polar researchers navigate both data disruptions and complex meteorological shifts, scientists stress the urgent need for enhanced observations of metrics beyond mere surface extent. Metrics such as ice thickness will be critical for evaluating the true health and structural integrity of the ice pack, particularly during fragmented and volatile years.
