Temperature Update

I'm away from the office this week, without access to my usual tools, so here's just a simple update on temperatures since spring for several interior and northern locations.  At Fairbanks, McGrath, and Bettles, the very persistent and unusual warmth of winter and spring finally came to an end in mid-May, and temperatures have bounced around either side of normal since then.  Of the 3 locations, Bettles has been coolest relative to normal, with marginally below-normal temperatures on average since June 1.

Up in Barrow, a few below-normal days started showing up in mid-June, but it has remained generally warmer than normal this summer; and in Kotzebue the summer has been very warm so far (4.4°F above normal since June 1).  The warmth in the northwest is closely related to the extraordinary warmth of the ocean surface in the Bering and Chukchi Seas (see map below); until this large ocean area cools off, there won't be much in the way of below-normal temperatures in western Alaska.  Unsurprisingly the CPC's latest seasonal forecast has a very high probability of warmer than normal conditions in western Alaska for August through October (also shown below).

Wetter than 2014 in the Hills

I don't want to belabor the "wet" theme too much, but the amount of rain that has fallen in the Fairbanks area in recent weeks is quite remarkable.  I noticed this morning that most of the SNOTEL sites in the hills above Fairbanks have observed more rainfall so far this summer than in the same period of 2014, which was of course the wettest summer on record at Fairbanks airport.

Here's a chart showing that most of the SNOTEL sites had fallen well behind 2014 by the end of June - despite a wet June this year - but this year's cumulative precipitation has surged ahead in the past week.

At the Munson Ridge site, elevation 3100', the total rainfall since June 1 is now the highest on record (1982-present); a remarkable 15.6" has fallen so far this summer (see data here).  Of this, 6.5" fell in the past week, which makes this the wettest week in the history of the observing site.

Nowhere else in Alaska has reported over 15" of rain so far this summer, and relatively few places in the lower 48 have been wetter - see the map below, based on NWS (combined gauge and radar) data.  The national "winner" (airport and COOP stations only) is Fort Myers, FL, with 21.9" of rain since June 1 - and that's a good deal wetter than normal even for a southwest Florida summer (I used to live there - it's wet).

The maps below show the mean 500mb height anomaly (top) and westerly wind anomaly (bottom) since June 1.  The very unusual pressure gradient near the date line has produced a zone of strongly enhanced westerly flow centered near Wrangel Island and extending eastward over northern Alaska and the Beaufort Sea; this area normally only sees weak westerly winds at this season (around 4-6 m s^-1 on average)

The strong westerly flow across northern Alaska no doubt goes some way to explaining the wet weather in Alaska's interior, although as we noted earlier 2014 was different; the map below shows the height anomaly for the same period in 2014.

[Update Sunday morning] Here are river gauges from the Chena River near Two Rivers, and the Tanana River at Fairbanks and Nenana.

The Chena River flood control system was activated on Wednesday to prevent water rising too high in Fairbanks.  Here are webcam views from downtown Fairbanks (Friday evening) and from Nenana (yesterday evening):

More Rain On The Way

Following swiftly on the heels of last week's hot spell, wet weather returned to Fairbanks earlier this week, with well over 24 hours of continuous rainfall on Sunday and Monday; and more rain is on the way.  The total rainfall since June 1 is up to 5.51", which is well over twice the normal amount and is the 4th highest on record for the summer to-date.  The number of days with at least 0.25" of rain this summer is tied with 2014 for first place (9 days), and the same is true for number of days with 0.1" of rain or more (14 days, tied with 1949).

The two charts below are updated versions of the charts I showed a month ago.  While this summer is not as wet as the record deluge in 2014, it's still very unusual.

An update of the 500mb height anomaly map in the earlier post shows that the last month has brought a rather similar pattern to that of early June: low pressure over the Arctic and high pressure from the Bering Sea to south of the Alaska Peninsula.  Broadly speaking, this translates into enhanced westerly flow compared to normal and - importantly - reduced dry southerly flow.

Here's this morning's GFS model forecast for rainfall in the next 7 days in Fairbanks.  It sure looks a lot more like August than July; but it's good news for the fire situation.

Heat Wave Review

As we look back at last week's hot spell, it's clear that it was historically significant for northern Alaska, as Thursday's 86°F in Kuparuk set a new all-time heat record for any Alaska location near the Arctic coast.  As we noted earlier, the 85°F in Deadhorse also broke the all-time record for the Deadhorse/Prudhoe Bay combined climate history (1969-present).  However, the heat wave may not long be remembered by residents of interior Alaska, as it was not particularly unusual there.  Fairbanks reached 88°F on 2 consecutive days, which is something that has happened 21 times before in the 86 years of NWS/Weather Bureau observations.  If the 88°F is the hottest temperature of the year, as seems very likely, it will be only 1 degree above what's typical for the annual maximum temperature.

Looking at 850mb temperatures, the Fairbanks soundings recorded a peak of 16.0°C, which is equal to the average maximum for the year.  This is a little lower than the GFS model forecast was indicating, so the air mass didn't quite heat up as much as expected.  Additionally, the surface temperature came a degree or two short of what might be expected based on the observed 850mb temperature, and this could be the effect of moist soil as I speculated earlier.  Humidity was high, with the heat index reaching 87°F at the airport after the rain on Thursday afternoon; the heat index only gets this high about once every 5 years, on average.

Here's a map of the maximum temperatures observed during the hot spell, courtesy of the xmACIS2 interface at http://xmacis.rcc-acis.org.  With one exception, all of the reports above 90°F are from RAWS stations, which are known to have a high bias in warm, sunny weather, because of inadequate ventilation of the thermometer.  The highest plausible non-RAWS temperature is the 92°F reported on July 14 by the Two Rivers COOP observer, about 25 miles east of Fairbanks.

It's interesting to look at the evolution of the temperature profile as measured by the Fairbanks soundings during the hot spell; the animation below shows the twice-daily soundings from Tuesday morning through yesterday morning.  The steady warming and then cooling of the lower troposphere is seen very clearly between 800 and 900mb, which is above the near-surface layer that undergoes rapid diurnal changes.  Notice the light winds throughout the column near the peak of the warmth and the strengthening westerly flow that brought an end to the event.

Courtesy of Environment Canada, here's the 500mb analysis from Wednesday afternoon, when the upper-level ridge was nearing its peak intensity.  Notice the proximity of the ridge to the Arctic coast and the low pressure far to the north over the Arctic Ocean; this is reflected in the sea-level pressure map (also below), which shows a pressure gradient from south to north across the North Slope.  The pressure gradient allowed south winds to prevail over eastern North Slope locations on both Wednesday and Thursday, thereby preventing the sea breeze from coming inland and bringing cooler air to places like Deadhorse and Kuparuk.

The 850mb temperature analysis from Wednesday afternoon (first map below) shows that the warmest air in an absolute sense was located over the interior, but compared to normal (second map below) the warmth was most anomalous well to the north of Alaska over the Beaufort Sea.  Again it was the unusual pressure gradient over northern Alaska and the southern Arctic Ocean that allowed warm air to flow north unhindered, and this is the key reason why the Arctic coast temperature record was broken.

Finally, to answer a question from reader Mike, here's a chart that shows the frequency of reaching specific thresholds of warmth in Fairbanks on or after dates in July and August.  We're already down to a chance of only 30% that 85°F will be observed again this year, and with a cool and very wet forecast the odds are even lower than that.  There's even a modest chance that Fairbanks won't get out of the 70s again; both last year and 2014 failed to see a temperature higher than 80°F for the rest of the year after this date.

As far as chilly nights are concerned, the median date for first 40°F at Fairbanks airport is August 18, but of course it could happen any time at colder outlying locations.

Micrometeorology in Fairbanks

Yesterday's hot weather was interrupted in the middle of the afternoon at Fairbanks airport by a quick thunderstorm that dropped almost half an inch of rain.  It was a small and short-lived pulse-type storm, but it happened to occur right around the time of the balloon sounding release, and there were some fortuitous and interesting observations as a result.

The zoomed-in radar image below shows the storm when it was most intense, at about 2:45pm AKDT (click to enlarge).  The red dot and marker indicate the approximate midpoint of the airport; note that the ASOS is located farther southwest, near the southwest end of the airport, and right under the path of the storm as it moved from west to east.  However, the balloon launching site is located near the western perimeter road, to the left of the marker, and on the north side of the storm.  The ASOS reported 0.42" of rain from the storm, while the upper air site measured only 0.10".  I've included below a map that Brian made a couple of years ago showing the locations of the two observing sites.

The simple animation below shows the brief life span of the storm; it moved right over the southwest end of the airport and Fairbanks' official climate site, but just grazed the upper air site.

Now take a look at the sounding from the balloon that was released at 3:03pm, which was just when the storm was moving out and decaying.  Rain-cooled air is evident at the surface, with a temperature of only 69°F, but a very high dewpoint of 66°F.  This is actually the highest dewpoint ever measured at the surface in a Fairbanks sounding (a marginally higher dewpoint was measured in June 2002 just above the surface).

Above the surface we see that the air was much warmer and very much drier, with the temperature rising to 83°F only 35m above the ground.  Clearly the air just above the surface was undisturbed by rainfall; it represents the ambient environment in which the storm formed.  I conclude that the rain had already ceased at the upper air site, and warm dry air had moved back in aloft from the immediate surroundings, but rain-cooled air was still in place at the sheltered sounding release site (which is surrounded by trees).

The 14°F inversion between the surface and 35m aloft is a remarkable occurrence, because it's rare to see any inversion in the Fairbanks sounding at the afternoon observation time in summer.  The sun-warmed surface is nearly always the warmest level in the troposphere, and less than 5% of afternoon soundings have an inversion in June and July.  (This post from 2 years ago showed the typical morning and evening temperature profiles through the year.)  Amazingly, yesterday's inversion was the strongest ever observed at any afternoon sounding from April through September.  Obviously it was a very unusual set of circumstances that produced such a shallow layer of cool air at the surface, and it happened at exactly the right time to be measured by the sounding.  It also didn't last long, because the ASOS temperature was back up to 84°F by 4pm and 88°F by 5pm.  Remarkably, the Fairbanks ASOS reached 88°F twice in one day, with a temporary dip down to 73°F during the brief heavy rain.

Arctic Coast Heat Record

While Fairbanks reached a very unusual 88°F yesterday, tying a daily record, the really unusual heat was up on the North Slope.  A remarkable 85°F was reported at the Deadhorse airport (ASOS) and also by the COOP observer in Kuparuk, about 30 miles to the west of the Deadhorse/Prudhoe Bay area.  This is a new all-time high temperature record for both sites and is higher than any temperature in the historical record for any location near the Arctic coast in Alaska.  Umiat has exceeded 90°F in the past, but they are much farther inland.

The chart below shows the annual maximum temperature for the combined period of record from Deadhorse and - prior to 1999 - from Prudhoe Bay.  The two sites are about 5 miles apart, with Prudhoe Bay being closer to the water, so the question arises as to whether the Prudhoe Bay location is cooler in summer.  If so, then it's not quite fair to compare yesterday's Deadhorse record to the old Prudhoe Bay data.  Unfortunately there is almost no overlap in the data from the two sites, so a direct comparison is not possible; however, I was able to compare both sites to the Kuparuk historical data - read on below.

Looking only at temperature data from July, there is complete historical data at Kuparuk from 1984 to the present, except for 1990 and 2011; so we have a nice overlap with both Prudhoe Bay and Deadhorse, allowing us to see if the temperature differences changed systematically in 1999.  See the chart below.  While there is considerable year-to-year variation in the July temperature difference between Kuparuk and the other two sites, there's almost no systematic difference between Deadhorse and Prudhoe Bay; both sites are slightly cooler than Kuparuk, and Deadhorse is just 0.3°F warmer than Prudhoe Bay on average.

But what about very warm days?  It's possible that even though the mean July temperatures are similar, Deadhorse might get warmer than Prudhoe Bay on hot days.  To look at this, I pulled out the 86 days on which Kuparuk has observed 70°F or higher (in any month); of these days, Prudhoe Bay averaged 2.4°F cooler (44 days) and Deadhorse averaged 2.6°F cooler (42 days), so again there is no significant difference.  For all days on which Kuparuk reached 75°F, Prudhoe Bay averaged 3.8°F cooler (12 days) and Deadhorse averaged 4.1°F cooler (14 days).  I conclude that there is no reason to believe that Deadhorse is more likely to reach high temperature extremes than the old Prudhoe Bay site.  This in turn means that it's fair to compare yesterday's record to the entire historical period at Prudhoe Bay/Deadhorse; and that makes yesterday's record quite significant.

[Update July 18: Kuparuk managed to reach 86°F on Thursday - thanks to Rick Thoman for pointing this out - so they now hold the record for highest temperature near the Arctic coast.]

Heat Wave Imminent

This week is likely to bring the hottest temperatures of the year to the Fairbanks area, as an upper-level ridge intensifies over the state and heats up an already warm air mass.  Here's this morning's sounding from Fairbanks; in the next 48 hours, subsidence under the ridge will warm and dry the lower levels dramatically, and clear skies will allow the sun to do its work.

The current NWS forecast calls for 87°F tomorrow and 86°F on Thursday at Fairbanks airport, and I think that's a conservative prediction.  This morning's GFS MOS forecast has 89°F tomorrow and 90°F on Thursday, which seems about right.

The chart below shows a time-height cross-section of temperature according to the GFS model, showing that 850mb temperatures are expected to reach about 17°C.  If we trace that warm air down to the surface along a dry-adiabatic lapse rate, we get a surface temperature of 88°F at Fairbanks airport, and typically solar heating will boost temperatures a bit beyond that estimate.  For reference, the last time 90°F was reached in Fairbanks was in June 2013; the high was 92°F on the 25th and 26th, but the lapse-rate method gave a temperature of 88°F.  So lower 90s are certainly a possibility this week.

The one factor that might help keep temperatures down a bit is the high level of soil moisture remaining from recent wet weather; the moist ground will not heat up quite as much, and increased evaporation will soak up some energy compared to, say, 2013.  It will certainly be interesting to see if 90°F can be reached - for only the 5th time in the past 20 years.

Update: here's a radar image from Tuesday evening (see comments below):

High Humidity

It has been humid recently in Fairbanks - much more so than usual.  In the first 10 days of the month, the average dewpoint measured at the airport was 55.2°F, which is the highest since at least 1950 for this period of the year.  The dewpoint exceeded 60°F on each of the first 4 days of the month, and it appears Fairbanks has only seen 4 such consecutive days on a handful of occasions in the past (1962, 1990, 1994, 2004), and never earlier than the end of July.  Peak humidity in Fairbanks (in terms of dewpoint) is normally at the end of July.

The humid conditions at the surface are consistent with the presence of a very moist air mass aloft too; the Fairbanks soundings show that the July 1-10 dewpoint was the 3rd highest on record at both 850mb and 700mb.  I think the depth of the moist anomaly suggests that the moisture has been imported to interior Alaska, and the high humidity doesn't just reflect (for example) local evaporation from moist ground.

Is there a long-term trend towards higher humidity in Fairbanks in summer?  Yes, but only slightly.  The charts below show monthly mean temperature and dewpoint since 1950 for June, July, and August.

The 1950-2015 rates of change according to linear trend lines are as follows (although the changes are not linear):

	Temperature	Dewpoint
June	+0.39°F/decade	+0.25°F/decade
July	+0.35°F/decade	+0.13°F/decade
August	+0.17°F/decade	-0.02°F/decade

In each month the temperature has risen more than the dewpoint, which means that the relative humidity has fallen as the climate has warmed.  To maintain the same evapotranspiration rates from vegetation, the dewpoint would have to increase faster than the temperature, so despite the slowly rising dewpoint, moisture demands have become greater for vegetation (at least at the Fairbanks airport site).

Cooling After Warm Summer Days

Reader Mike posed a question recently about whether wet conditions in summer allow for cooler temperatures on clear nights after hot days (see the comments for this post).  The idea expressed here is that moist ground would remain cooler during and after a hot day (because of greater evaporation and higher heat capacity), and continued evaporative cooling overnight may also bring temperatures down.  (Of course "night" is a relative term for much of the summer in Fairbanks.)

The 1950-present hourly and daily data from Fairbanks show some support for the hypothesis, although it's not a strong signal; see the chart below.  My analysis looked at all days in June through August with a high temperature at least 10°F above normal, i.e. above 72-84°F depending on the date, and I then took the 25% of these days with the lowest "overnight" (6pm-6am) cloud cover.  For each of these 124 days, I found the precipitation in the previous 90 days and plotted this against the overnight decrease in temperature anomaly (i.e. the decrease from high temperature anomaly to low temperature anomaly).

The main feature of the results, I think, is that overnight cooling is usually moderate or large - and rarely small - when the prior 90 days have been either moderately or very wet (above 3" of rainfall).  The scatter is large when the prior 90 days have been dry - sometimes there is no cooling relative to normal, but sometimes there is a lot.  There is an interesting cluster of large-cooling events at intermediate precipitation levels, but I'm not sure what this means.

Bear in mind that late summer is usually much wetter than early summer in Fairbanks, so most of the points on the left-hand side are from June while most of the points on the right are from July and August.  The overall trend is not just reflecting the date, however, because (a) I've used temperature anomalies rather than temperature, and (b) the normal diurnal temperature range actually gets a bit smaller as summer advances in Fairbanks.

One reason for the high degree of scatter in the chart is that overnight wind speed also has an influence on the rate of cooling; calm winds allow the surface layer to decouple so that enhanced low-level cooling can occur.  The chart below shows the effect of wind speed for all of the warm summer days regardless of cloud cover.

Finally here's a chart showing the influence of cloud cover regardless of wind speed.

June Anomalies

The month of June was very wet in Fairbanks, with 3.29" of rain, the 4th highest total on record (1930-present).  The monthly mean temperature was slightly below the 1981-2010 normal for the first time since last September.

Interestingly, however, the month was not more cloudy than usual in the Fairbanks area - in fact the mean cloud cover at the airport was lower than any other recent year except 2013 (when June was very hot and sunny).  The CRN site located 11 miles NE of town recorded 18.6 MJ/m² of solar radiation, which is slightly below the 2003-2015 average of 19.3 (standard deviation=2.1), so at that site it was somewhat cloudier.  The high rainfall total combined with only modest cloud cover is consistent with the rainfall being unusually convective; thunder was observed on 12 days, which is the highest number of days in June since 2010.

Compare this pattern to what happened two years ago, when June was the wettest on record.  In that month cloud cover was far above normal and thunder was observed on only 2 days.  The maps below show the 500mb height anomalies in the two months.  Back in 2014 there was a low pressure anomaly over the state, but this year there was a ridge to the southwest and a trough over the Arctic Ocean.  Not all wet Junes are created equal in Fairbanks.

The maps below show the anomalies for June at 16 stations across the state.  Except for Annette Island last month, the below-normal temperatures at Fairbanks and Bettles are the first cool anomalies to show up at any of these stations since December.