Seasonal Contrasts

As the change of season accelerates at this time of year, I find it interesting to consider where in Alaska the climate varies most over the course of the annual cycle.  It's obvious that summer-winter temperature differences are very much greater in the Alaskan interior than in the maritime south, but what are the details?  And how about precipitation?  Alaska is a much drier place in spring than in summer or autumn (depending on location), but how much does liquid-equivalent precipitation vary from seasonal minimum to seasonal maximum?

Here's a look at the temperature question using monthly ERA5 data: the map shows the difference between the warmest and coldest calendar month of the year, based on 30-year averages.

The annual range varies from less than 20°F in the Aleutians to over 75°F in the Yukon Flats.  As we might expect, the Alaska Range marks a major boundary between the "continental" climate of the interior and the more maritime climate of the south.  However, the southwestern mainland is relatively continental, being more exposed to bitterly cold intrusions in winter.

Looking at Arctic Alaska, the high terrain of the Brooks Range has a reduced annual range, but the North Slope has a thermal range as great as most of the interior, even though there is ocean not far away to the west and to the north.  Of course the maritime influence is very subdued in winter because of ocean ice cover.

We can get a more detailed perspective from the PRISM climate normals, which are available at 1km resolution - but only for the 1981-2010 period.  Click to enlarge the following:

Note that the PRISM data is derived directly from ground-truth observations, unlike the ERA5 reanalysis (which is a model estimate constrained by upper-air and satellite data).  However, the results generally line up well.  I've annotated the PRISM domain-wide minimum and maximum temperature range: again less than 20°F in the Aleutians, and very nearly 80°F at the southern edge of the Yukon Flats (just to the west of Circle, to be precise).

Here's a zoomed-in version to show the details in the interior.  The topographic detail in the PRISM data is a delight to behold.

To analyze the seasonal variability of precipitation, I calculated the ratio of the maximum and minimum monthly normal precipitation:

The results are quite different between the two sources, with PRISM showing a greater contrast between the wettest and driest months nearly everywhere.  This illustrates the fundamental uncertainty around precipitation quantities for much of Alaska, especially in winter.  PRISM's higher ratios for Southeast and western Alaska definitely seem preferable: examining a few individual sites confirms that the wettest-versus-driest ratio generally exceeds 4.0 in these areas.  However, the patchiness of the PRISM results does not seem realistic, especially across the spatially uniform Yukon Flats and North Slope. 

One obvious takeaway of the precipitation results is that seasonal variability is particularly high to the north of the Alaska Range, owing to the drying effect of southerly chinook winds in the cold season.  Fairbanks is in this zone: about 7-8 times as much precipitation falls (per day/week/month) in late summer as in early spring.  Contrast this with locations to the west of the Alaska Range, i.e. the western interior and southwestern mainland, which escape the rain shadow effect; winter and spring precipitation are not as low.  McGrath, for example, sees only about 4 times as much precipitation in late summer as in early spring.

According to the PRISM data, the lowest ratio of wettest to driest month (1.65) is found at a spot on Kodiak Island.  Data from Kodiak airport does confirm that precipitation amounts are relatively uniform through the year, with mid-summer being marginally drier than spring - but it's very wet year round.

In contrast, PRISM reflects the profound summer-winter contrast of precipitation in northern Alaska: as in most of the Arctic, the cold season is very dry simply because the lower atmosphere is so cold.

I'd welcome reader comments on other aspects of the maps that seem noteworthy.