The Arctic October Transition
The return of snow and ice
October is a transition month in and around the Arctic. With rapidly decreasing solar heating, the accumulated warmth of summer dissipates. As temperatures drop, sea ice expands and the winter snowpack is established over most of the Arctic that isn’t open ocean.
IMS basics
The National Ice Center (NIC) creates a daily unified snow and ice cover analysis known as the Interactive Multisensor Snow and Ice Mapping System (IMS). Unlike most large scale climate/environmental data products, the IMS is created with experts “in the loop” and this usually results in better analyses than “machine only” algorithms.
There are few things to keep in mind with the IMS. First, snow and ice cover is analyzed as either “yes or no” and fresh water ice is included in the analysis. Also, sea ice depiction does not follow the passive microwave convention of using 15 percent concentration ice as the delimiter between ice and no ice, but rather 40 percent. At larger scales this difference is usually not significant. The product is currently available in several horizontal resolutions, 24 km, 4 km and 1 km. Because of the importance of terrain in and around Alaska I generally use the 4 km for monitoring purposes.
October 2024 transitions
Figures 1 show the 4km resolution IMS analyses at the start, middle and near the end of October 2024. As you expect, overall there has been a significant increase in sea ice extent and snow cover. However, there are areas of ice and snow going “the wrong way”. For example, between October 1 and October 15, since along the fat northeast Russian coast retreated a bit to the northwest: for an area west of Cape Dezhnev this was lowest ice concentration of the year. On the October 15 analysis, snow cover is indicated over parts of the Nordic and western Russian Arctic, but that was largely gone by October 27 due to much above normal temperatures in the region the second half of October.
From the climate perspective we’d like to have an idea of how the current snow and ice extent compares to a long term baseline. Daily 4 km analyses have been available since early 2004, so we can construct an early 21st century estimate of “normal, which is shown in Fig. 2 for for October 27.
Comparing this to this year, we find that in the past 20 years there typically was a larger area of snow cover in the European Arctic than we have this year, while snow cover is more extensive in the northwest North American sector than typical for the past 20 years. The lack of ice this year in the Beaufort Sea and in the Canadian Arctic is also obvious.
Alaska and vicinity
Figure 3 shows the same October 27 analysis as in Fig. 1 but zoomed in to Alaska and vicinity. As an experiment, I added hatching to show where the 2024 ice and snow category is different from the median category shown in Fig. 2. So for example, in the Beaufort Sea north of the Alaska coast, the hatching shows that in the past 20 years ice was analyzed in the region currently shown as water. On the other hand, several areas in the Russian far northeast without snow cover this year typically have snow by this point in the season.
We can of course plot the difference in categories directly, and I show this in Fig. 4. This makes clear that the snow cover across, say, both the southern Yukon Territory and southwest Alaska is not typical for October 27, while the lack of snow cover over the upper MacKenzie River valley is unusual for the date.
