How Much has the Arctic Warmed?

There is a lot of discussion in the wider community of how much the Arctic has warmed in recent decades, and how much the Arctic has warmed relative to the rest of the globe. The recent paper by Mika Rantanen and co-authors

¹

has highlighted how the answer to this question strongly depends on 1) change over what time period and 2) what definition of the “Arctic” is used.

The question of “change over what time period” is not unique to the Arctic but is a decision that has to be made for all climate or environmental change studies.

However, it may come as a surprise to you that is no widely agreed upon definition of the “Arctic”. Rather, many different definitions are or have been used, depending on purpose and, to a large extent, past practice. All are arbitrary to one extent or another, and while I have a preferred definition for my work (coming out of atmospheric science, I usually think of “Arctic” as poleward of 60N), others have perfectly good reasons for using something else.

The purpose of this post is not to try and suggest what time period is best or “what is the Arctic”. Rather, here I want to stress the regional variation in change, because this is very important regardless of the “when” and “where”. Fig. 1 shows the total change in annual average temperature in the past half century using the ERA5 reanalysis. As you can see, based on this analysis, by far the greatest warming has been in the area of Svalbard east toward to the Russian-owned Novaya Zemlya and Franz Josef Land archipelagos, where temperatures have warmed 5°C to 7°C in the past 50 years. Large areas show more than 3°C warming over the Arctic Ocean and adjacent land areas, all the direct result of longer open water seasons and thinning sea ice. At the other end of the change range, some areas in southern Siberia, southeast Alaska mainland into the Yukon Territory and southern portions of the Northwest Territories in Canada have warmed by less that 2°C over the same time. In fact, just within the Alaska mainland, the total change has varied from less than 1°C in the Southeast Interior to more than 4°C on the western North Slope. Sticking (arbitrarily) to all areas north of 60N, the annual temperature change in the past 50 years varies by a factor of ~7.

Fig 1. Total change in average annual temperature 1972-2021 using ERA5 reanalysis based on Theil-Sen regression,

While it’s straightforward and sometimes useful to come up with a single number to express change in the Arctic, it’s very important to keep in mind that impacts to ecosystems, people and cultures respond most immediately to changes (or lack thereof) at much smaller scales, and regional variability in change is a feature of the Arctic just as much as it is elsewhere.

1. Rantanen, Mika, et al. "The Arctic has warmed nearly four times faster than the globe since 1979." Communications Earth & Environment 3.1 (2022): 1-10.

---

Technical details: This analysis is based on the ERA5 reanalysis data from ECMWF/Copernicus and used the most current version of the reanalysis available as of this writing (Nov 2022). Theil-Sen regression is used in because it is less influenced by extremes in the data compared to ordinary least squares regression and so provides a more robust trend estimate, though with air temperatures the difference are usually minimal.