Wednesday, March 29, 2017

End in Sight

A major warm-up is under way at last across the eastern interior, as the flow has finally gone around to the south; Eagle is up to 46°F this evening, and Fairbanks is expected to break the freezing mark by Friday at the latest.  This will prevent a significant record from being broken: Fairbanks has never seen a March without the temperature rising above freezing at some point in the month.  The closest was in 1997, when the month's high temperature was only 34°F.

[Update March 31: the temperature has risen above freezing at Fairbanks airport as of 11:30am AKDT.  This is the 3rd latest appearance of the first above-freezing temperature in March.  In 2006 it occurred at 1pm on March 31, and in 2007 the thaw began at 2pm on March 31.]

The last time that above-freezing conditions were observed either at the surface or in the air column above Fairbanks was on February 26, so the period of having a continuously sub-freezing atmospheric column will end at about 30 days.  The longest such period this winter was 46 days, from mid-November through December 30, and this was considerably longer than any of the past 3 winters.  The chart below shows the length and ending date of each winter's longest stretch of sub-freezing conditions (both surface and aloft) back to 1948-49.  (Click to enlarge the chart.)

It's a funny coincidence that 4 of the last 5 winters have seen the longest sub-freezing spell come to an end very close to the end of the year (December 27-31).  Reader Eric first pointed out the similarity in timing of cold and warm spells between this winter and last (see here), and in fact a distinct warm-up can be identified near the turn of the year in each of the past 5 winters.  We'll call it the New Year's Thaw and I'll look at it more closely in a future post.

Saturday, March 25, 2017

Cold and Dry

Wintry cold continues to hold on tenaciously across interior Alaska, at least during the nighttime hours.  Here we are after the spring equinox, and temperatures in the -30s Fahrenheit are still occurring overnight in the usual cold spots.  This morning the Kanuti Lake SCAN site reported -34°F, and the Salcha RAWS hit -31°F.

In Fairbanks-land, Goldstream Creek reported -27°F this morning, and the airport thermometer registered -18°F.  Remarkably, every night but one so far this month has seen a low of -13°F (-25°C) or colder at Fairbanks airport.  The 22 nights at -15°F or colder is tied with 2007 for most on record in the month of March, and the NWS expects another one tonight.

Here's a chart showing the last two weeks of 2m temperature measurements from UAF's Smith Lake site: sub-minus 20 every night, and daily temperature traces indicative of generally clear skies.  So far this month more than two-thirds of the sky condition measurements at Fairbanks airport have indicated either clear skies or "few" clouds (25% coverage or less).

Total snowfall this month in Fairbanks has amounted to only 1.7", but this in itself is not very unusual: the 1981-2010 median is only 3.2" for the month.  April is usually even drier, with a normal liquid-equivalent precipitation of only 0.1".  March and April are the driest months of the year, climatologically speaking, in Fairbanks, and a few days ago Brian showed that this is also the driest time of year across most of interior northwestern North America:

Why is late winter generally a sunny and very dry time in interior Alaska?  Part of the reason is that the air is very dry, and that's because evaporation from the earth's surface is much reduced in the Northern Hemisphere at this time of year: ocean surface temperatures are near their coolest, ice cover is near its peak on both oceans and fresh water bodies, and forests are still mostly dormant.  The lifetime of water vapor in the atmosphere is only about 9 days on average, so precipitation must decrease in tandem with evaporation rates on a hemispheric scale.

The graphic below shows an estimate of how the total atmospheric water vapor content (precipitable water) varies through the year in the Northern Hemisphere: the values shown in the chart are the west-east averages for each month and each line of latitude.  It's interesting to see the earlier summer peak of vapor content in the Arctic compared to the tropics, as solar heating drops off more rapidly in late summer at high latitudes.

Here's a similar chart, but for a single line of longitude running approximately through Fairbanks (148°W).  Over Alaska and the oceans to the south, the lowest normal value of precipitable water occurs in March.

A second reason for dry weather in Fairbanks in March and April is that there's simply not much storminess to generate clouds and precipitation.  The absence of fronts and storms is a more localized factor than hemispheric dryness, because other parts of the hemisphere (indeed of the USA) have plenty of storminess and wet weather in spring.

In the vicinity of Alaska, storminess is reduced in the spring because the North Pacific Ocean is cool, leading to reduced north-south temperature gradients and lower potential energy for storms.  Another reason is that high pressure tends to build over the Arctic basin in late winter and spring.  The chart below show the time-latitude cross-section of sea-level pressure along 148°W; notice the prominent high pressure zone around 75-80°N from February through April.  At the latitude of Fairbanks (65°N) the pressure is lowest in October and then increases into late winter; and while there's much more to the weather pattern than just MSLP, this does illustrate the overall tendency towards generally more clear and dry weather as the winter advances.

Tuesday, March 21, 2017

Arctic to Mid-Latitude Connections

Yesterday the scientific journal Nature published a brief but helpful summary of the state of research concerning the linkage between climate changes in the Arctic and mid-latitudes.  For context, bear in mind that a popular hypothesis in recent years has been the idea that enhanced warming in the Arctic slows the jet stream and therefore creates more blocking and extreme weather at mid-latitudes.  However, the real world may not have such a simple chain of causality; the article emphasizes how much there is yet to learn on this subject.  Here's a key quote:

"...the connection is not one-way from the Arctic to the mid-latitudes but also works in reverse, and observations and climate models give differing estimates of the extent to which mid-latitude climate is influenced by Arctic warming".  Also: "we have a poor grasp of how much of the Arctic amplification is caused by warm and moist transport from the mid-latitudes to the Arctic".

Update: see comments below for links to the workshop website, containing much more information.

Monday, March 20, 2017

New Blog

Regular visitors may be interested to learn that our own reader Eric has started his own blog on interior Alaska's weather and climate; I expect I will link to it occasionally or even repost some material.  Take a look here:

On another note, the webcam from Dawson (Yukon) reveals that the Yukon River has at long last frozen over completely next to the town.  There's been no lack of cold weather this winter, but an unusual ice jam upstream allowed an open lead to remain in the middle of the river throughout the entire winter until now.  It looks like the ice road to West Dawson may finally have opened just in time for the onset of spring.

March 1:

March 7:

March 17:

Saturday, March 18, 2017

Winter Hangs On

The sun is shining for long hours in central Alaska now, but temperatures have remained decidedly wintry of late.  This morning the thermometer registered -25°F at Fairbanks airport, which is where it has been on average for the first 18 days of the month (for overnight lows).  The overall mean temperature so far this month is 17°F below normal.

The 500mb height anomaly map for March 1-16 reveals the pattern that we would expect to find in association with unusual cold in much of Alaska: a big ridge over the Bering Sea and relatively low heights over Southeast Alaska.
The sea-level pressure map (below) shows the surface-level ridge extending across the interior into western Canada, providing the clear sky conditions that are conducive to enhanced overnight cooling.  The ridge has also shut off the moist westerly flow and brought an end to the snowy pattern.

It's interesting to examine the subtle contrast between this month's pressure pattern and that of November through February, see below.  For most of the winter the ridge was centered farther to the south, and the Arctic trough extended down to the Chukchi Sea, so the flow into Alaska was both wetter and warmer (relative to normal).
The map below shows the average MSLP pattern during 10 strong La Niña winters; notice the strong similarity to this winter's pattern.  This is somewhat remarkable to me, because this winter's La Niña episode was very weak in intensity - and yet the circulation pattern over the Aleutians strongly resembled a classical La Niña pattern.  This seems unlikely to be a coincidence, and La Niña probably bears some of the blame for the current cold as well.

For completeness, the maps below show the typical El Niño pattern and the MSLP anomaly from last winter (2015-2016).

Monday, March 13, 2017

North Pacific Blob Update

Hot off the press this evening: a new post on the Alaska "Blob Tracker" blog, showing interesting similarities of recent events with those that followed the 1997-1998 El Niño:

Wednesday, March 8, 2017

Cold Nights

Abnormally cold conditions have returned to interior Alaska again, with widespread -40s at night in the past week or so, and a few -50°F readings.  The Salcha RAWS hit -51°F last night and has fallen below -40°F every day this month except one.  Here's a sample of the locations that dropped below -40° last night:

In Fairbanks the minimum temperature of -39°F this morning is the coldest observed so late in the season since 1964, although -37°F was seen on March 15 only 2 years ago.  Assuming today's high temperature is 0°F or less, which appears likely, this will go down as the coldest day of the winter in terms of standard deviations below normal.  But of course the contrast between today and the January 17-19 cold snap is like night and day: the number of hours of sunshine has doubled since then.

Here's an updated chart of daily mean temperature anomalies for Fairbanks; the mean temperature since November 1 is now 1.7°F below the 1981-2010 normal.

Saturday, March 4, 2017

Record Cold in Arctic Canada

[Update March 5] The 1000-500mb thickness dropped still further today, reaching 4619m at 1200 GMT over westernmost Victoria Island.  This appears to be the coldest air mass since March 1996 in the general vicinity (90-180°W).

Also of note: yesterday's all-time low temperature at Mould Bay is the first all-time low at a long-term observing site north of 70°N since March 18, 2008, when Danmarkshavn in northeast Greenland set its all-time low of -49°F.  (This is based on NOAA's GHCN data, which is reasonably but not entirely comprehensive.) [End of Update]

In recent months and years we've often discussed the extraordinary and persistent warmth around the Arctic Ocean, so it makes a change to consider an episode of extreme cold for once.  This morning the temperature dropped to -66°F at Mould Bay on Prince Patrick Island, which is north of Banks Island and is the westernmost of Canada's Queen Elizabeth Islands; Mould Bay is at a latitude of 76°N.  I've marked the location on a couple of the maps below.

Remarkably, today's low temperature of -66°F sets a new all-time record low for Mould Bay, which has more-or-less continuous weather records since 1948.  The previous record was -65°F, in 1967 and 1987.  The last time -60°F was observed at Mould Bay was in 2004.  Here's a chart of the minimum temperature each winter since 1948-49; note that some winters are missing (including 2003-2004) because the daily data are incomplete in those years.

It's interesting to note that today's cold at Mould Bay is not far from an all-time record low for all of Canada's high Arctic region (north of 70°N).  According to GHCN data, only two locations, Eureka and Lake Hazen (both on Ellesmere Island) have been colder, at -68°F and -69°F respectively.

The circulation feature responsible for the cold weather today is a vortex of extremely cold air that has been hovering over the Queen Elizabeth Islands for several days.  Back on Monday the coldest air was over Ellesmere Island, but the vortex has drifted southwestward and the cold has intensified considerably in the past few days.  Bear in mind that of course there is still very little solar heating at this time of year north of 75°N - the sun rose in Mould Bay only 3 weeks ago and reached only 8° above the horizon today.

The map below shows the GFS model's estimate of 850mb temperature today at 1800 GMT or about noon local time in Mould Bay, when the surface temperature was reported as -64°F.  A substantial area of sub-minus 40° air at 850mb is evident from Prince Patrick Island southward over half of Banks Island; this is an extremely cold air mass.

Here's a map of 1000-500mb thickness at the same time; the thickness is an excellent measure of the mean temperature in the lower half of the atmosphere.  The gridpoint minimum of 4634m is extremely low by historical standards; according to NCEP reanalysis data, this is the lowest 1000-500mb thickness value anywhere in the Northern Hemisphere since March 2011.  For comparison, and again according to reanalysis data, the all-time lowest 1000-500mb thickness value over Alaska was 4598m on 27 January 1989 (towards the end of that month's deep cold snap).

The map below shows the reanalysis-era minimum in 1000-500mb thickness; we see that most of the Arctic basin has seen thickness values below 4650m at one time or another.  However, the second map below shows that only a few selected areas have seen sub-4650m thickness since January 2000.  In other words, the kind of cold we're seeing today in northwest Canada has become rare in the past couple of decades; this is a really extreme and notable event.  It's remarkable and very interesting to see this event occur at the end of what has otherwise been a winter of record-breaking warmth in the Arctic.

Thursday, March 2, 2017

Snow Water Equivalent

Following on from the last post, here's a quick look at the seasonal maximum in snow water equivalent in Fairbanks, as measured at the airport and at the Snotel site (off College Road and not far from the Fairgrounds, according to Rick Thoman).  The automated Snotel instrument monitors snow water content via a pressure pad, whereas the airport measurement is done manually as part of the daily observing routine.  The two time series track together pretty well, although there was a big discrepancy in 2013-14.

The overall downward trend since the early 1990s is rather noticeable, although of course the interannual variance is very large.  So far this winter the SWE has reached 5.2" at the Snotel site; again we see that while this is above normal, it's not highly unusual compared to the long-term climate of Fairbanks.