Late Winter Chill

It's getting rather late in the winter for harsh cold in much of Alaska (excepting perhaps the North Slope where it's more typical), but that didn't stop the thermometer from dropping to -42°F at Fairbanks airport on Monday morning.  This is the coldest of the winter so far, and the first -40° of the season in the Golden Heart City.  It's also the first time since 2007 that such cold has occurred so late in the winter.

In keeping with my last post, temperatures aloft were significantly below normal on Monday morning above Fairbanks: about -20°C at 850mb and -38°C at 500mb.  Cold air was circulating around a very intense mid-atmosphere low pressure system just to the west of Canada's Arctic Archipelago - see below.  This is a remarkable turnaround from the record high pressure that occurred in the same area just two weeks ago.

 

Here's the Fairbanks sounding from 3am AKST on Monday.  Even though winds were out of the west (and even slightly south of west in the lower troposphere), the airmass had an Arctic rather than southerly origin.

Checking in on that notorious cold spot, Chicken, they also saw -40° in the last few days.  No surprise there, but what is surprising is that Chicken has produced more -40° readings this month than in any other February: 14 of them so far (although with an 8am obs time, it's not actually 14 different nights).  Warmer air has arrived now, and the next few days will remain warmer, but there's still a chance it may be the coldest February on record for Chicken (but with data only back to 1997).  However, for the winter as a whole, the temperature is running near normal, with the -40° count approaching a typical two dozen.

Is Alaska Cold Home-Made?

My post from a couple of weeks ago about cold on the North Slope elicited a comment from a reader about the origin of the very low temperatures.  Specifically, the comment asked whether it was accurate to suggest that the cold was imported from farther north in the Arctic; my post stated that high pressure was acting to "funnel very cold air into northern Alaska".  Surely, the reader asked, Alaska's cold develops in situ when there's no competing warm influx to prevent it.  After all, there's not much severe cold over the Arctic Ocean these days, so how can the North Slope cold have its origin farther north?

The answer to this question is that both ideas have validity.  Valley-level surface cold across interior and northern Alaska certainly is generated locally when conditions are right (clear skies, calm winds), but it's also true that surface conditions are powerfully influenced by the thermal quality of the large-scale air mass.  When it's warm aloft, temperatures at the surface are usually warm and are never severely cold; but when it's cold aloft, then the surface is usually very cold and is rarely warmer than normal.

The overall correlation can be illustrated with a scatter plot of December-February 850mb temperatures and daily minimum temperatures at Fairbanks:

Slightly more than half of the variance in daily minimum temperatures is explained by temperatures aloft (850mb is around 1200-1400m above sea level).  Taking -40° as a benchmark for severe winter cold, this only happens when 850mb temperatures are below normal, and more often 850mb temperatures are significantly below normal.

The correlation with temperatures higher up in the atmosphere is smaller, of course, but it's still significant.

Similarly, the recent cold spell at Umiat (as low as -56°F 10 days ago) was associated with colder than normal conditions aloft.  The chart below shows the rather close correspondence of surface and 850mb temperatures in the past several weeks - but note that I've used upper-air data from the nearest sounding site at Utqiaġvik, about 170 miles to the northwest.

Here are the scatter plots of Umiat daily minimum temperatures and Utqiaġvik upper-air temperatures:

I also looked at the correlations in terms of daily departure from normal, rather than absolute temperatures - see below.  As it turns out, it makes very little difference in this analysis.

To summarize, historical data show that about half of the variance of winter daily temperature is explained by the temperature of the air aloft, which in turn is largely dictated by the source region of the air mass.  So while it's fair to say that Alaska's cold can be greatly amplified in situ - for example under a strong inversion with clear skies and calm winds - the coldest episodes are never just home-made; you need both cold air aloft AND favorable surface conditions to take the thermometer down to its lowest levels.

Arctic High Pressure

Harsh cold is persisting at many interior locations and also across the Brooks Range and North Slope, with numerous -40°F to -50°F readings this morning, and some even colder.  Notable low temperatures in the past couple of days include:

-49°F at Galena in the western interior

-56°F at the Birch Creek RAWS just above the Yukon Flats

-46°F at Anaktuvuk Pass, 2100' elevation in the Brooks Range

-58°F on the upper Sagavanirktok River

The latter is apparently the coldest observed in Alaska so far this winter; Chicken reached -57°F a couple of days ago.

Severe wind chill continues to afflict Arctic coastal sites, and the breeze even picked up at Umiat today: around 6am, a mean hourly wind speed of 11 mph combined with a temperature of -52°F.  This turns out to be the highest wind speed on record at Umiat with a temperature below -50°F (with hourly data back to 2007).

Looking at the surface analysis from 3pm yesterday, we see the same Arctic high pressure system that I highlighted in my last two posts - except now the high has expanded over an enormous area from the Chukchi Sea to northwestern Canada and across to Greenland.  It's really a sight to behold: click to enlarge.

The all-time record highest MSLP over the Arctic Ocean is 1069mb (per ERA5 data since 1950), so this is not far off.  It's a rare anomaly, and it's responsible not just for the cold in Alaska, but for outbreaks of severe cold far to the south in the lower 48 and also in Europe.

Frigid in the North

With high pressure over the Arctic Ocean continuing to funnel very cold air into northern Alaska, surface conditions have become downright frigid across the interior and eastern North Slope, with widespread -40s and some -50s today.  Umiat reached -55°F this morning according to the RAWS instrument, and saw a high of only -47°F this afternoon.

But at least there's no significant wind in Umiat.  From Deadhorse all the way across to the Canadian coast, wind chill values are -70°F or lower, with stiff offshore breezes creating seriously nasty conditions.

The lowest temperatures occurred in the sheltered valleys of the interior North Slope to the south and southwest of Umiat; satellite measurements suggest some spots may have dropped below -65°F.  Here's a satellite-observed temperature map courtesy of Twitter user wrighthydromet (click to enlarge).

 

 

Our favorite wind chill site, Howard Pass, also had a very extreme episode of wind chill yesterday, with sustained winds over 50mph in conjunction with temperatures below -40°F.  This is the first time in 7 years that the Howard Pass thermometer has dropped below -40°F, although missing data is an issue.

Back in December I looked at the strong inverse relationship between wind and temperature at Howard Pass.  Remarkably, -40°F or lower has only ever occurred with a wind speed of 37mph or higher, and the median wind is 48mph at such low temperatures.  The chart below zooms in on the low-temperature portion of the hourly distribution of temperature and wind.

It's interesting to observe that while yesterday's cold blast was a typical wind chill episode at Howard Pass, the temperature didn't recover much last night as the winds quickly died down to almost nothing.  Consequently, today's temperatures were easily the coldest on record for low wind speeds; this morning it was a full 10°F colder than previously observed at Howard Pass with a wind speed below 5mph.  This illustrates the intensity of the cold air mass over northern Alaska.

Here's a surface analysis chart for 3am yesterday; notice the very strong pressure gradient over northwestern Alaska, an obvious prerequisite for these extreme wind chill episodes.

Cold

Daily gains in daylight and solar strength are heralding the transition to late winter across Alaska, but temperatures plunged to their lowest levels of the season so far across parts of the interior and north today.  Umiat and Arctic Village reached -50°F on opposite sites of the Brooks Range, and a remarkable -55°F was reported by a co-operative observer far to the south, near Tanacross in the upper Tanana River valley.  Here's a look at minimum temperatures through 4pm (click to enlarge):

Wind chill has been nasty too; Fairbanks was reporting a steady breeze even as the temperature dropped into the -20s last night (bottoming out at -28°F).

Here's part of the cause of the cold outbreak: an intense high pressure system over the Arctic Ocean to the north of Alaska.

The estimated central MSLP of 1059 mb at 3pm yesterday is towards the upper end of the climatological history; for example, ERA5 data says that MSLP has reached 1060 mb at 80°N 180°W only once in the last 50 years (January 2013).  The anomaly is related to an unusual weather pattern across the mid-high latitudes of the Northern Hemisphere: the Arctic Oscillation is strongly negative, which means the usual circumpolar westerly flow is weak and disrupted in the upper atmosphere, allowing cold air to spill south from the Arctic to certain parts of the lower latitudes.

Here's a simple animation of a webcam view at Arctic Village today, showing thin layers of ice fog moving around, as is typical on very cold days.

Cooling in March

A couple of weeks ago I compared the new 1991-2020 temperature normals to the previous climate period of 1981-2010, and one of the interesting features that stood out was the widespread change to slightly cooler conditions in March.  With the exception of Utqiaġvik, which has of course seen rampant warming, all of the climate sites I looked at showed a cooling trend in March; and March is the only month with a statewide decrease in the 30-year normal temperature.

Here's a map of the March temperature difference according to ERA5:

The cooling is quite widespread in the southern half of mainland Alaska, but the magnitude of the temperature difference is small: rarely more than 1°F.  Contrast this with the much larger warming that occurred in the neighboring months of February and April, particularly across western Alaska.

 

If we look at 3 climate observing sites with noticeable March cooling, we see one key reason for the anomaly: the 1980s were exceptionally warm in March, and that decade has dropped out of the new climate normal period. 

The 1990s were very warm too - notice the complete absence of cold March's in both the 80s and 90s - but then some cold re-emerged after the turn of the century, and Bethel in particular had a spate of cold March's.  Interestingly just a few years ago Bethel's 30-year running average for March was distinctly lower than it is today, so if for example the climate normal period had changed 5 years ago, the new normal would have been more than 2°F colder than the one 10 years before.

Clearly, annual and decadal variability play a large role in the details of the changes from one climate normal period to the next, and this becomes most evident where annual and decadal variance is large compared to the magnitude of the long-term warming trend.

Some readers will have recognized already that the PDO phase is a prime candidate to explain the decadal variability we're considering here - which is to be expected, given that it's the Pacific Decadal Oscillation.  Here's a chart of the March PDO index; notice the remarkably persistent positive phase of the 1980s and the return of some significantly negative values around 2008-2013.

But an obvious question then is - if the PDO is responsible for the March cooling, then why did February and April warm so much?  The PDO tends not to change drastically from month to month, and indeed it was strongly positive in the 1980s in February as well.  However, Bethel's February temperatures (for example) have warmed dramatically in the past two decades compared to the 1980s.

The explanation for this month-to-month contrast is (I believe) that we're dealing with natural variability that has pushed the monthly decadal trends in different directions.  In other words, it just so happens that February and April have recently been warmer than we would have expected based on the long-term trend and the PDO phase; and on the flip side, March has been colder than we would have expected.  If we somehow had another "realization" of recent decades, it could equally well have been the other way around, with relatively subdued warming in February and April and much larger warming in March.

To support this idea that "luck of the draw" with the weather patterns explains the different trends, the maps below show the 2011-2020 average departure from normal for MSLP and 500mb height in March.  This is a relatively cold pattern for southern Alaska, with a ridge axis over the Bering Sea that produces a northerly component to the flow to the east of the ridge.

In contrast, February's of the past decade have tended to see a ridge over the Gulf of Alaska and low pressure over the Arctic Ocean, and this combination brings a lot of warm air up from the southwest.

The decadal pattern for April (see below) is quite different, but low pressure near the Aleutians is another typical signal for warmth in southern Alaska (this is an El Niño-like pattern).  Clearly then the monthly weather patterns of the past decade are aligned with the observed differences in trends: cool in March, very warm in February and April.

 

The pattern analysis works at the other end too, in the 1980s: it just so happens that March's from 1981-1990 had the warmest pattern of the 3 months - see below.  Consequently, March in the 1980s tended to be warmer than expected based on the background temperature baseline and the (very positive) PDO phase.  And so with unusual warmth in 1980s and a cool pattern in the past decade, it's no surprise that the 30-year normal has turned a bit cooler for this month.

Siberian Warmth in 2020

One of the most striking and memorable global climate stories of 2020 was the remarkable and persistent extreme warmth that occurred in Arctic Russia.  The sheer magnitude of the anomaly for the annual mean temperature in north-central Siberia was simply amazing: over 5 standard deviations above the 1981-2010 normal.  Here are maps from the ERA5 reanalysis and from surface station measurements (click to enlarge):

The agreement is very good: every station over a wide area in north-central Russia saw an annual temperature over 3 SD above normal, and 3 sites exceeded 5 SD in the region where ERA5 data highlighted the most significant anomalies.  Of course there was a lot of very anomalous warmth elsewhere around the Northern Hemisphere (and globe) as well, but clearly Russia's Taymyr Peninsula was the epicenter of warmth.

Here's a chart showing the history of annual mean temperature for the 3 sites with over 5SD anomalies; Khatanga was the "winner", with a mean 2020 temperature of -6.4°C compared to a 1981-2010 mean of -12.2°C and standard deviation of only 1.1°C.  February, April, and November were all more than 10°C above normal, 3 more months were over 5°C above normal, and only December was less than 2°C above normal.

 

It's interesting to look at the vertical structure of the temperature anomalies using ERA5 data.  Here's a cross-section around the Arctic at 75°N for the lowest 25% of the atmosphere.  The departure from normal was far greater near the surface than aloft, although the anomaly for the year still exceeded 3°C at 750 mb (about 8000 feet elevation).

Here's the actual annual mean temperature (absolute, not anomaly) for the same cross-section:

And here's the 1981-2010 normal:

 

It's interesting to see that normally the annual mean temperature is approximately constant with height up to about 900mb in the vicinity of the Taymyr Peninsula, and farther east the normal temperature profile shows an inversion over the East Siberian, Chukchi, and Beaufort Seas.  But in 2020 the more well-mixed environment of the Barents Sea extended dramatically farther east than normal (on an annual-mean basis).

Here's an animation of the temperature anomaly cross-section on a monthly basis.  Again, the sheer persistence of the warmth in north-central Russia is remarkable.

 

What causes can we identify for the amazing 2020 anomaly?  Extraordinary conditions began to emerge early in the year in association with a strongly positive phase of the Arctic Oscillation that swept warm Atlantic air eastward across Siberia and the Russian Arctic.  Warm westerly flow therefore dominated the first three months of the year in association with the AO, and then as winter gave way to spring, the pattern shifted in such a way that very warm air continued to flow into the region; this new setup involved a trough over western Russia and a ridge over central Siberia.  See below for January-March and April-May 500mb height anomaly maps.

 

With the arrival of summer, the circulation pattern shifted again, and yet again it favored unusual warmth along Russia's Arctic coast, but by this time I think the feedback associated with reduced sea ice and soil moisture was at least as important for maintaining the warm anomaly.  Owing to the winter and spring warmth, Arctic sea ice was much thinner than normal near Russia's coastline, and it quickly broke up over the East Siberian Sea in June and July, removing a key cooling mechanism that would normally be present near the coast until much later in the summer.

Another factor that drove unusual warmth through the summer was the prevalence of dry soils caused by the early loss of snowpack and greatly enhanced drying under warm conditions - and also exacerbated by high pressure and reduced rainfall over central and eastern Siberia.  Wildfire activity was widespread and intense.  Here's a late August soil moisture map from ERA5.

As summer ended, the impetus towards unusual warmth was then reinforced with new vigor into the autumn months as sea ice was far below normal across most of the Arctic; the Arctic-wide ice extent almost set a new record low in 2020.  The excess heat available from open ocean rather than ice-covered ocean has been a major contributor to enhanced warming trends in the Arctic during autumn over the last two decades.

 

Finally, to top it all off, the Arctic pattern returned to a strongly positive AO phase in November, bringing a renewed warm westerly flow to the Russian Arctic.  Khatanga saw its warmest month of the year in terms of departure from normal, with a massive +11.3°C anomaly: the November average temperature was -13.3°C versus a 1981-2010 normal of -24.5°C.  December then brought cooler northerly flows, but the month was still warmer than normal.

New Climate Normals - Temperature

Once every 10 years, the world of climate science transitions to a new 30-year normal period for the purpose of climate monitoring, and 2020 marked the end of the most recent normal period.  The choice of a 30-year window is nothing more than conventional practice; here's a bit of background from the WMO (click to enlarge).

Based on ERA5 gridded data, here's a look at how Alaska's annual average temperature changed between 1981-2010 and 1991-2020:

Most of Alaska to the north of about 65°N warmed by more than 1°F, and of course the greatest change was seen on the North Slope - and especially in Alaska's northwestern Arctic territory.  Substantial warming of more than 1°F also occurred in the southern half of the Alaska Panhandle, according to this data.

Here's how the changes look in the wider hemispheric context north of 45°N.  Alaska's Arctic warming was only a small portion of a wide swath of 1.5-2°F warming that stretched across the Russian side of the Arctic to the Barents Sea.

The ERA5 temperature changes are broadly consistent with data from surface observing sites across Alaska.  According to NOAA/NCEI data, the state warmed 0.8°F as a whole, and climate observations include changes of +2.0°F at Utqiaġvik, +0.3°F at Fairbanks, and +0.7°F at Anchorage.  Interestingly, however, Juneau warmed only 0.2°F, considerably less than ERA5 suggests.  It would be worth looking at other sites in the southeast, but it's likely that ERA5 has difficulty representing valley-level temperatures in the complex topography of Alaska's southern coast.

Here's a series of charts to illustrate the month-by-month changes in the temperature normals at a variety of sites.  All have the same vertical scale, and the series goes roughly from north to south.  Click to enlarge.

 

In addition to the well-known autumn warming that is most dramatic in the Arctic (where it is linked to ice loss), it's interesting to see a large temperature increase at Alaska's western sites in February and April - although not in March.  Presumably year-to-year variability is to blame for the March outlier, because after all we're really just taking the difference between two decades: the 1980s and the 2010s.  Taking the late winter period as a whole (February-April), I'd wager that the notable warming is closely linked to loss of Bering Sea ice in just the last few years; 2018 and 2019 in particular had extremely low ice extent, and temperatures were far above normal at the end of winter.

Finally, here's the month-to-month chart for Alaska as a whole.