Wednesday, February 28, 2024

Calendar of Cold

It's back to cold(er than normal) in Alaska again, with a brisk -31°F observed at Fairbanks airport this morning, and -40°F in colder spots to the north, including Bettles.  Wind chills are unpleasant on the North Slope too, but as noted by reader Mike, Arctic Alaska is very much still locked in winter.  The seasonal minimum in temperature occurs a lot later for the North Slope - and indeed for many of Alaska's coastal locations - than for the interior.

For a visual perspective on the timing of the seasonal minimum, here's a map based on ERA5 1991-2020 data.

According to this analysis, the seasonal temperature cycle bottoms out before the turn of the year for a few areas in western and interior Alaska, but most of the interior sees its coldest "normals" in the first half of January.  Cold peaks later for the North Slope - the first half of February for the eastern North Slope - and the highly maritime climate of the North Pacific sees a seasonal minimum even later, in late February or even early March.  The pronounced lag over the ice-free Pacific is related to the very large heat capacity of the ocean's upper layers, as well as the ample cloud cover that prevents a strengthening sun from bringing warmth to the ocean surface.

I'm a little surprised by the early seasonal minimum for coastal western Alaska, where I would have expected sea ice to keep temperatures suppressed to a later date.  The ERA5 results are confirmed by the NCEI 1991-2020 daily normals: Nome has its lowest normal temperature around January 12, and Bethel around January 9.  For comparison, Fairbanks also bottoms out around January 12.  I don't know why the west coast doesn't have more of a maritime influence in this aspect of the climate.

But a couple of caveats are worth noting in this analysis.  First, I'm using harmonic functions to smooth the seasonal temperature cycle through the year, and the details of the method (e.g. the number of functions) can affect the timing of the estimated seasonal minimum.  Second, the sample size of 30 years also implies some uncertainty in the date of the minimum, because the random timing of major historical cold and warm episodes will affect the details of the calculated cycle.  It would be interesting to quantify the uncertainty with some statistical experiments.

On a broader scale, the map below shows that the European side of the Arctic Ocean has a later seasonal minimum than the Pacific side: it even extends past March 15 for a small area near Iceland.

And expanding the domain into the mid-latitudes, we see that the eastern North Pacific has the latest seasonal minimum of anywhere in the hemisphere: as late as March 29 according to this analysis.

Another question I can't answer is why the eastern North Pacific has so much more lag than the western North Pacific.  Perhaps it's as simple as the western ocean being influenced by continental air from the vast land mass of Asia, which warms up quickly in late winter.

Wednesday, February 21, 2024

More Context for Cold

Another round of downslope/chinook warming brought another widespread thaw to the Tanana River valley and Fairbanks-land yesterday.  Temperatures above 40°F were widespread, including 45°F at Fairbanks airport, the highest February temperature in 20 years.  The year-to-date average temperature in Fairbanks is now only 2.2°F below normal.

Looking back again at the late January and early February cold snap, I was curious to see how the lower atmosphere (not surface) temperatures compared to past cold spells on various time scales.  However, with NWS weather balloons no longer going up during severe cold, there's some missing data for the recent event; so I extracted data from the ERA5 reanalysis.  I looked at Fairbanks 850mb temperatures and 1000-500mb thickness, with the latter being an excellent measure of the average temperature of the lower half of the atmosphere.

Here's the annual (winter) minimum for 3-day average 1000-500mb thickness:

On this time scale, the recent event was one of the coldest since 2000, but it was nowhere near as cold as the more extreme events of earlier decades.  The complete absence of seriously cold events after 2000 is quite remarkable, and suggests that the cold tail of the distribution has been dramatically curtailed since then.

Below is another perspective, showing annual minimum values of 3-day 850mb temperature (y-axis) versus 1000-500mb thickness (x-axis).  The values are quite highly correlated, of course, but there's some variability in the extent to which cold is concentrated at lower atmosphere levels like 850mb.  On a 3-day basis, the recent event was relatively less unusual in terms of 850mb temperature, with several other events being colder in recent years.

The 7-day and 14-day charts below show quite similar findings on these time scales, with the recent event being a bit more anomalous overall on a 14-day basis, as I noted before; but it was still not in the least unusual compared to earlier decades.

The two "granddaddy" cold spells of 1989 and 1999 stand out clearly here; those two events were much more extreme than anything in recent years, and also worse than anything that happened in the otherwise colder 1950s, 60s, and 70s.

Finally, it's interesting to observe that two quite extreme cold spells occurred in the last 20 years when we look at a 30-day average: late February and early/mid March of 2007, and January of 2012.

I hadn't previously realized how anomalous that 2007 event was, but in fact ERA5 shows the event having the lowest 30-day average temperature since 1950, by both metrics (thickness and 850mb temp above Fairbanks).  Given that it was so late in the winter, the surface temperatures were nowhere near as low as they would have been a few weeks earlier; but nevertheless March 2007 was the second coldest on record in Fairbanks, and 27 of 31 days had a low of -10°F or lower (the highest such number for the month).

Here's a look at the 500mb height pattern for the 30-day period: what a classic!

The 500mb height anomaly:

And the 850mb temperature anomaly, according to the older (but still basically reliable) NCEP reanalysis:

Friday, February 16, 2024

The Pendulum Swings

In the world of weather, extremes tend to beget extremes, and sometimes in the same place after only a short interval.  Alaska has seen a wholesale reversal from the cold of late January and early February to very unusual warmth in the past few days.

Wednesday was quite extreme, with chinook winds producing widespread 40s and some 50s downstream of the Alaska Range:

Nenana reached 52°F, the highest temperature in February since 1943.  Only February 24, 1943, had a higher temperature there in February: 54°F, and on that same day in 1943 Fairbanks reached 50°F, its record high for the month.

The Fairbanks sounding on Wednesday afternoon reported a remarkable temperature of 9.4°C about 1500 feet above ground.  There are only a few historical events with a temperature that high at any level in a February sounding from Fairbanks, the most recent being in 2010.  It's not surprising then that the Little Chena Ridge SNOTEL site above Fairbanks (2000' elevation) reached 50°F, and Teuchet Creek (1640') reached 49°F.

There'll be no prizes for guessing the mid-atmosphere flow pattern giving rise to the heat: a ridge over southeastern Alaska, with strong southerly flow coming up all the way from the sub-tropics.  Here are two different views of the 500mb height (pressure) pattern on Wednesday.

The deep trough far to the south of western Alaska was a key feature here, because it allowed that southerly flow to tap into very warm air from a low latitude.  That deep trough in turn was connected to a strong west-east jet stream across the central Pacific to the north of Hawaii: winds in excess of 80 m/s, or 180mph.  And this feature - an enhanced jet stream heading east into the southern US states - is very typical of El Niño.  So you can thank (or blame) El Niño for the big warm-up in Alaska.

Tuesday, February 13, 2024

January Climate Data

Taking a look back at January now that all the usual climate data is available, it was a month with a lot of variability, both in space and time.  Broadly speaking, the first half was mild and the second half was cold for the state as a whole, but as usual that mostly reflects the large mainland area, not peripheral regions like the southeast panhandle.  Here's the UAF statewide temperature index since November 1; despite the very unusual cold in the last 10 days of January, the average index value since November 1 is still positive (warmer than normal).

There was also a major temperature gradient between interior Alaska versus western Alaska and the North Slope:

The epicenter of cold was in the Porcupine River drainage of the northeast, with the region around Old Crow in the Yukon being particularly cold.  Old Crow itself reported -50°F or lower on 9 days and reached -61°F on the 26th; it was the coldest calendar month since 2009.  In contrast, Utqiaġvik had its 8th warmest January on record.

Precipitation was highly variable too:

The contrast between a wet northern Panhandle (e.g. record snows in Juneau) and dry southern Panhandle was significant.  There was also major variability on a scale too small to be picked up by the ERA5 data: Rick Thoman notes that Anchorage airport actually had its 10th wettest January since 1954, and yet every other site in the region had below-normal precipitation.  Here's Rick's monthly summary:

Wind was well above normal for nearly all of western and northern Alaska.  It's a bit surprising to see above-normal wind with unusual cold for the northeastern interior; I would expect a cold January to be relatively calm (i.e. stronger inversions).

Here's the complex monthly-mean pressure pattern that gave rise to the major climate anomalies: low pressure over Chukotka drew warm air up over the Bering Sea and western Alaska, but high pressure centered over northwestern Canada allowed cold air to dominate at the surface for most of the Alaskan interior.

Looking more broadly across the Arctic, January was a mixed bag there too: much warmer than normal in Canada (but not breaking monthly records), but colder than normal in northern Scandinavia.

Friday, February 9, 2024

Follow-up on Cold

A couple of quick items of follow-up regarding the recent cold snap.  First, here's a look at temperatures in the hills above Fairbanks, using data from several of the SNOTEL sites (not all of them have been reporting data lately).  The two distinct waves of cold are evident - the first culminating in the January 27 cold, and the second reaching its nadir 6 days later, on February 2.

The second cold wave was colder at valley level, with Fairbanks seeing its first daily-mean temperatures below -40° since 2017.  But some of the SNOTEL sites were no colder in the second round, and so the temperature inversion was stronger.  The climatological average inversion at this time of year is about 10-11°F between Fairbanks and most of these sites, although Little Chena Ridge runs a few degrees warmer (as in this event).  So the inversion was roughly normal in the second cold wave, but was certainly weaker than normal on January 27.

We can't confirm the temperature profile for February 2, because the NWS balloon sounding auto-launchers are automatically disabled below -40°, but the 3pm January 27 Fairbanks sounding confirms a total lack of inversion in the lowest 1500 feet or so.

Rick Thoman comments and explains on his blog:

Rick also comments that late January may have been the coldest in the hills since February 1999, based on data from Keystone Ridge and the Fairbanks soundings.  Certainly the -40° on Keystone Ridge on January 27 was the lowest minimum temperature since February 5, 1999, but it's worth mentioning also late January 2012, which saw a high temperature below -30°F on Keystone Ridge (not seen in the recent event).

Looking at the SNOTEL historical data, we get the following comparison for coldest daily mean temperature (°F) in the two years (2024 vs 2012):

Eagle Summit: -30.0 (2024) vs -38.5 (2012)

Munson Ridge  -35.5 vs -33.5

Mt Ryan  -31.5 vs -33.5

Upper Nome Creek  -39.5 vs -39.0

Little Chena Ridge  -32.0 vs -32.0

With the exception of Eagle Summit (which was much colder in 2012 and also colder in other recent years), I'd say it's a toss-up: the recent cold was comparable to 2012 at elevation.

Rick's copious notes from January 2012 (the coldest month statewide since pre-1925) are worth perusing:

Monday, February 5, 2024

Cold in Context

Now that the cold snap is in the past, there will be lots of opportunity for post-mortem analysis when I have a chance (and when I ditch the nasty bug I'm suffering from).  For now, I just wanted to follow-up with a final number on the weekly temperature index that I used last week to show the lack of Fairbanks-Bettles severe cold in recent decades:

Here's the updated chart:

The recent steep uptrend was indeed broken, but on a 7-day time scale the coldest week was not at all unusual by historical standards.

However, the cold spell was more notable in the end for its longevity than its severity at any one time, and so there's a more significant cold anomaly on a 14-day time scale:

Here we see a most remarkable absence of 14-day cold in recent years - almost as if the climate never recovered from the record-warm winter of 2015-16 (during the last very strong El Niño!).  The lack of variability since 2016-17 is quite amazing.

But the cold snap just ended is a significant change.  Only 15 winters since 1950 had a colder 14-day period for these two climate observing sites.

Here's the 500mb height anomaly for the 13 days that had average temperature below -30°F for Fairbanks and Bettles combined.

This too is interesting, because the mid-atmosphere anomaly is right over the northern interior; and while we might expect that to be true, in fact significant cold in Fairbanks is more often associated with a trough over western Canada.  This too will be a topic for follow-up.

Friday, February 2, 2024

Deep Cold

Today's high temperature (so far) of -39°F -38°F at Fairbanks airport is the coldest in 7 years, and it's the coldest February day since 1999.  Since 1930, only two years have seen a daily high temperature this cold in Fairbanks in February: 1947 and 1999.  It also happened in 1910, according to the Ag Farm data.

As for the low temperature, -49°F this morning was also the coldest in Fairbanks since 2017; but there's a chance -50°F could be reached before midnight, with the 5pm temperature already back down to -45°F.  [Update: it was reached]

Unfortunately there's no sounding data to look at the vertical profile of temperature, because the automated balloon launcher is disabled when the temperature reaches -40°.  This is a great pity, in my view, and a loss to science.  If the automated launcher can't operate, then perhaps some hardy soul could be persuaded to launch a balloon manually, as in the old days - like 7 years ago.  Here's the sounding from 3pm on January 18, 2017, the last date Fairbanks saw a high temperature below -40°F (the lowest levels aren't included on the plot, but the data was recorded.)

The 2017 cold snap was much shorter than the current one, but was briefly severe, with more widespread -50s than any day in the current cold spell.  There's a series of posts in the blog archives, e.g.

Here's a splendid photo published to Twitter today by Cody Moore, showing ice fog filling the valley from a vantage point on UAF's West Ridge:

Visibility at the airport has been 1/4 - 1/2 mile for most of the day.  Water vapor emissions from the urban environment are to blame for the fog, with a substantial contribution from the power plant plumes; here's another photo, this time from UAF GINA:

The FAA webcam on Ester Dome provides another perspective on just how localized the fog is: skies were gloriously clear across most of the interior today.

A simple animation gives a sense of the moisture flowing from the power plants at left towards the Tanana River valley at right.

Thursday, February 1, 2024

Cold Snow

There's been a lot of interesting weather across Alaska in recent days - as demonstrated by the sudden uptick in blog posts - and here's another one that merits attention.

Back on Sunday, as the upper level trough pivoted over western Alaska, heavy snow developed in the frontal zone over Anchorage, even as temperatures at the surface remained uncharacteristically low.  Sunday's high temperature was only +2°F even while prolonged snowfall occurred in the afternoon and through the overnight hours, leading to an accumulation of 9.9" by midnight and another 6.7" the next day (adjacent to the airport).

Such a large snowfall at such a low temperature is unprecedented in the Anchorage climate data back to 1953, and not by a small margin: check out the following chart showing the joint distribution of daily high temperatures and daily snowfall.

The explanation for this outcome involves the rare combination of strong mid-atmospheric lift along the frontal zone, combined with an unusually cold airmass at the surface.  Here's the Anchorage airport sounding at 3pm on Sunday:

Notice the cold layer at the surface, with northerly winds of over 30mph at about 2500' above ground.  But higher up the winds were out of the southeast and south, and the sounding shows a deep layer of saturated (humid) air near or just below -10°C, which is a temperature conducive to growth of dendritic snow crystals.

Here's a map of 850mb temperatures (shaded) at the same time, showing the extremely cold air to the north and west, and much warmer air to the southeast of Anchorage; the snow occurred in the strong gradient between the two.

And the 500mb height analysis at the same time:

It's interesting to note that Fairbanks and Bettles also saw accumulating snow to an unusual degree for such cold weather.  In Fairbanks the daily total was only 0.6" - also on Sunday - but nevertheless this tied the record for most snow with such a cold high temperature (-27°F).  In Bettles 4.9" fell on Monday, with a high temperature of -13°F.