Late July 2026 Arctic Sea Ice Update
A Tale of Two Hemispheres
As July draws to a close, the waning solar heating is increasingly obvious at high latitudes and this will accelerate in the coming weeks. However, because of the thermal lag of the oceans, Arctic sea ice continues to melt into September. This summer, sea ice conditions between the Atlantic and Pacific sides of the Arctic have diverged even more than most recent years.
**August 6: reader R. James pointed out that I had an error in the Atlantic side sea ice extent rankings. This has now been double checked and corrected.**
Arctic-wide sea ice
The spatial distribution of sea ice around the Arctic at the end of July was broadly similar to last year at this time (Fig. 1) and total extent was only about 1 percent higher on July 30 than the same date in 2025. However, there were some clear differences.
The ice edge was well north of the 1981-2020 median in NSIDC analysis across the Atlantic sector of the Arctic but also in the eastern Canadian Arctic and in the seas north of Siberia. Sea ice extent on the Pacific side of the Arctic was relatively high last year but this year it’s is even higher.

The July 30 Arctic-wide sea ice extent of 6.62 million km² was 14 percent below the 1991-2020 median in the National Snow and Ice Data Center analysis. Ice extent during July declined 29 percent (Fig. 2), which is typical for the past 25 years.

The July 30th Arctic-wide sea extent rankings (since 1979) were similar across the major analyses:
6th lowest National Snow and Ice Data Center (NSIDC) Sea Ice Index and EUMETSAT OSI SAF Sea Ice Index
8th lowest Japanese Aerospace Exploration Agency (JAXA)
10th lowest U.S. National Ice Center/NSIDC Multisensor Analyzed Sea Ice Extent (MASIE)
Regional extent
Regional sea ice extent (Fig. 3) shows about as extreme a split as is possible between the Atlantic and Pacific sides of the Arctic.
The Atlantic side of the Arctic has the lowest in the satellite era for this time of year. The combined ice extent for Greenland and Barents Seas is 30 percent below the previous lowest July 30 extent, which occurred in 2021. The combined ice extent in the Greenland, Barents and Kara Seas is also the lowest of record, about four percent lower than last year. In the Central Arctic basin, open water north of the Franz Josef Land archipelago extended to about 84°N, just 400 miles (650 km) from the North Pole. Note in Fig. 3 the Barents Sea rank of 6th lowest is misleading: there is virtually no sea ice remaining there, and this has happened repeatedly by late July since 2010. Hudson Bay and the Canadian Arctic archipelago also have the lowest sea ice extent since 1979.
On the Pacific side, it’s a completely different story. The end of July sea ice extent around Alaska, with little open water north of 70°N, looks like it’s taken from the 1990s. In the Chukchi Sea, ice extent at the end of July was 8 percent higher than any year since 2000. Additionally, the ice extent decrease during July 2026 was the lowest since 2003. In the Beaufort Sea, sea ice extent is highest for the end of July since 2003, though it’s only a little bit higher than 2006 and last year.

An important 21st century milestone of the summer sea ice melt season is the opening of the high latitude shipping routes. Analysis by the U.S. National Ice Center (Fig. 4) shows that as of July 31 both the Northern Sea Route, the northeast Atlantic-to-Pacific Ocean connection across the north coast of Russia, and the Northwest Passage, the western Atlantic-to-Pacific Ocean connection through the Canadian Arctic are both still closed to non-ice-hardened vessels. In recent years these routes have occasionally been open by the end of July but that is not yet the norm.

Sea ice thickness
Modeled sea ice thickness and volume from the Danish Meteorological Institute show noteworthy difference between the end of July this year and the same time last year. Just glancing at Fig. 5, it’s clear that thicker ice extends farther north from northern Canada this year (left) than last year (right). Total ice volume (the little inserts), which earlier this had been at or near the lowest on record is now appreciably higher than last year and effective tied with 2024 for second lowest.



Absolutely fascinating. Any theories?