Before presenting the MASIE and SII results for July, a note about a strange thing in today’s Sea Ice Index report. I have sent a note to them requesting an explanation for why the values have been altered from those in the dataset just two days ago. When attempting to add into my spreadsheets the final two July days, I noticed that all the previous values were now different. Exploring further, going back to beginning of 2024 all values had changed, some showing larger extents and many showing smaller ice extents than previous recorded.
For 2024 the new values added ice extents with the average day gaining slightly (47k km2). But in 2025 so far, the average day lost (-57k km2) compared to the values two days ago. Curiously, since March 14, 2025 all days had lower values at a daily rate of -75k km2. In sum, the altered values in 2025 removed ~11M km2 of ice extents so far, and 10M km2 of that since March 14. In the report below, I excluded the altered SII values awaiting news from NSIDC.
After a sub-par March maximum, by end of May 2025 Arctic ice closed the gap with the 19-year average. Then in June the gap reopened and in July the melting pace matched the average, abeit four days in advance of average. The chart shows the July Arctic ice extents on average decline from 9.7M to 6.9M km2. MASIE started July ~5M km2 in deficit to average and ended the month ~4M km2 down, continuing to melt about four days in advance of the average decline. SII matched MASIE the first half of July, then tracked slightly lower the second half.
The regional distribution of ice extents is shown in the table below. (Bering and Okhotsk seas are excluded since both are now virtually open water.)
| Region | 2025212 | Day 212 | 2025-Ave. | 2020212 | 2025-2020 |
| (0) Northern_Hemisphere | 6555733 | 6941055 | -385322 | 5880746 | 674988 |
| (1) Beaufort_Sea | 944231 | 793206 | 151025 | 875454 | 68777 |
| (2) Chukchi_Sea | 621236 | 555019 | 66217 | 533748 | 87488 |
| (3) East_Siberian_Sea | 683122 | 751512 | -68390 | 329453 | 353669 |
| (4) Laptev_Sea | 329581 | 370847 | -41266 | 61979 | 267602 |
| (5) Kara_Sea | 32436 | 166826 | -134390 | 95539 | -63103 |
| (6) Barents_Sea | 1131 | 29555 | -28424 | 23940 | -22808 |
| (7) Greenland_Sea | 228078 | 296681 | -68603 | 282403 | -54325 |
| (8) Baffin_Bay_Gulf_of_St._Lawrence | 117170 | 150751 | -33581 | 35368 | 81801 |
| (9) Canadian_Archipelago | 460908 | 547942 | -87034 | 515499 | -54592 |
| (10) Hudson_Bay | 73633 | 139798 | -66165 | 92861 | -19228 |
| (11) Central_Arctic | 3062678 | 3137162 | -74483 | 3033706.07 | 28972 |
The table shows most regions in deficit with Kara the largest, and Canadian Archipelago and Central Arctic also sizable. Hudson Bay and Greenland Sea will lose the rest of their ice in upcoming weeks. Surpluses in Beaufort and Chukchi offset about 220k km2 of losses elsewhere.
Why is this important? All the claims of global climate emergency depend on dangerously higher temperatures, lower sea ice, and rising sea levels. The lack of additional warming prior to 2023 El Nino is documented in a post NH and Tropics Lead UAH Temps Lower May 2025.
The lack of acceleration in sea levels along coastlines has been discussed also. See Observed vs. Imagined Sea Levels 2023 Update
Also, a longer term perspective is informative:


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 Arctic ice extents and NH snow cover.



One comment