The chart above from DMI shows why Arctic ice melted more slowly than average. This year the daily mean temperature never went above freezing, From now on it will only grow colder.  The daily minimum may have occured already on Sept. 7 with an ice extent of 4.9M km2.

August is significant because it is the last month that NH ice extent declines.  Those familar with the datasets know that March monthly average is taken as the annual maximum ice extent, and September monthly average serves as the annual minimum. Importantly both months are neutral, i.e. the ice extent values are nearly the same at the month start and end.  All the melting occurs April to August, all the freezing October to February.

[ It reminds me what a Saskatchewan grain farmer told me: “Around here we only have AAA farmers”. I asked what he meant and he replied: “April to August and then Arizona.”]

The arctic ice extents are now reported through September 11, 2026, and as noted previously the wavy polar vortex had hampered ice formation with incursions of warmer southern air into the Arctic circle.  This factor receded in May and June, with July extents closing the gap with the averages. Melting ceased after Sept. 7 and is now ~5.0M km2. The Northern Sea Route (NSR) goes through the Russian shelf seas of Laptev, East Siberian, and Chukchi seas on the way to Bering Strait in Beaufort Sea.

The image from  September 11 shows the Arctic Ocean core is still solid, while the Eurasian NSR now has open water on the left vertical side, including passages through Laptev (top left) and Chukchi (bottom Left).  As usual in August, Hudson Bay (bottom right) is full open water, as is Baffin Bay (middle right).  At this point Canadian Archipelago has opened up the eastern end of the NW passage, but has extensive old ice as well as growing additional fast ice in Beaufort sea (bottom center). Canadian Ice Service latest conditions along the Alaskan coast include an ice warning for ships:

The chart below shows the 20-year August averages for Arctic ice extents, along with 2026, 2025 and 2007 as well as SII v.4. Note that on average ice extents hover around 4.6M km2 ending this period. In this period MASIE was well above average throughout currently hovering around 5.0 M km2, and comparable to 2025. SII v.4 continues to greatly underestimate Arctic ice extent, presently ~0.35M km2 lower than MASIE. All of these are higher than 2007.

The table below shows the distibution of ice extents on day 254 across regions of the Arctic ocean.

Region	2026254	Day 254 ave.	2026-Ave.	2007254	2026-2007
(0) Northern_Hemisphere	4948149	4643810	304340	4349612	598537
(1) Beaufort_Sea	659822	502590	157231	599679	60143
(2) Chukchi_Sea	344007	216432	127575	74733	269274
(3) East_Siberian_Sea	277773	292941	-15168	311	277462
(4) Laptev_Sea	213926	155123	58803	247496	-33569
(5) Kara_Sea	18	33322	-33304	62274	-62255
(6) Barents_Sea	0	14658	-14658	7384	-7384
(7) Greenland_Sea	192766	176959	15807	324789	-132023
(8) Baffin_Bay_Gulf_of_St._Lawrence	39619	28358	11261	21406	18213
(9) Canadian_Archipelago	302503	273010	29493	210083	92420
(10) Hudson_Bay	16	6972	-6956	16552	-16536
(11) Central_Arctic	2916558	2942564	-26006	2783651	132907

The table shows an overall surplus of 304k km2 or 7%, and more than a half wadham greater than 2007. Many regions are close to or above the 20-year average. The seas of Laptev, Chukchi and Beaufort are in surplus. The only sizeable deficit is in  Kara Sea. Bering and Okhotsk seas are left off the list since they are open water now as usual.

 

Illustration by Eleanor Lutz shows Earth’s seasonal climate changes. If played in full screen, the four corners present views from top, bottom and sides. It is a visual representation of scientific datasets measuring ice and snow extents.