El Niño and Bering Sea Storms

With memories of last year's ex-typhoon Halong still fresh in many minds, it's worth asking the question, "What about this year?"  Does the major El Niño that is now occurring affect the risk of a similar outcome during this autumn's storm season on Alaska's Bering Sea coast?

It's well-known that El Niño tends to strengthen the "Aleutian Low" in winter as a response to major alterations in tropical patterns of convection/rainfall.  Looking at the strongest El Niño events of the past, this effect is evident from November on, but not in September and October.  The following maps show the average 500mb height (pressure) departure from normal during the strongest El Niño conditions for each month in autumn and early winter:

Looking at historical storm track data from ERA5 reanalysis (processed data courtesy of the University of Manitoba), there are some modest differences between El Niño and La Niña.  The maps below show the number of autumn storm tracks passing within each grid cell, for storms with a central pressure of 980mb or lower (a modest threshold - many of these are run-of-the-mill storms, not intense storms).  Note that here I'm defining El Niño and La Niña based on tercile thresholds from the Multivariate ENSO Index.  Click to enlarge:

The main feature that stands out to me is an indication that the Seward Peninsula and Norton Sound region tends to see more autumn storms during El Niño than during La Niña.  Farther south, the storm density is generally similar for the southern Y-K Delta and around Bristol Bay, and the Gulf of Alaska clearly sees more autumn storms during La Niña.

The classic El Niño enhancement of storm activity emerges in December through February, but by then the typical storm track is south of the Y-K Delta region, so the winter storm frequency is not notably enhanced for Alaska's west coast.

Looking more closely at the autumn results from the Seward Peninsula to the northern Y-K Delta, the increase in storm frequency during El Niño is mostly found in November and then continues into December, as we would expect from the 500mb maps.  Here's a table showing the number of 980mb or lower storms that passed through a box over the region:

Here's the box I used for the analysis:

Another way of looking at the ENSO sensitivity is that this box has seen 11 storms below 970mb during the El Niño tercile, but only 3 such storms during La Niña (since 1950).  The three strongest autumn storms in the El Niño set dropped below 960mb in this box and occurred in the Novembers of 1979, 2009, and 2015 (the latter being a very strong El Niño), whereas the lowest MSLP in a La Niña autumn (963mb) was last year's ex-typhoon Halong.  Ex-typhoon Merbok (September 2022) is in 3rd place (964mb) for a La Niña autumn storm.

It's interesting to observe that 3 of the 4 strongest storms in the Norton Sound box during La Niña have occurred in the last few years: Halong, Merbok, and another strong storm in mid-November last year.  In this context, it's worth bringing up again the remarkable warmth that has prevailed in the northwestern North Pacific in recent years, as discussed post-Halong here.  It's not unusual to see a warm northwest Pacific during La Niña, and especially after a multi-year La Niña, but the warmth of recent years has been exceptionally pronounced.

Checking in on current North Pacific SSTs, extreme warmth is still persisting from northern Japan to the waters south of the Aleutians.  This is uncommon during El Niño, when the circulation pattern tends to drive cooling to the east of Japan, i.e. a more positive PDO phase.  The combination of a strong El Niño with a significantly negative PDO phase is highly unusual.

I don't want to draw any strong conclusions from this brief analysis, but it does seem there are a couple of reasons to suggest the odds of west coast storm activity may be higher than usual this autumn.  In the first place, if extremely warm ocean temperatures persist to the south of the Aleutians, then decaying typhoons would be expected to be more vigorous as they enter the Bering Sea in general.  September and October are the peak months for decaying typhoon potential.

And then looking ahead to November and December, historical data suggests that El Niño does favor an increased risk of strong early winter storms for the central west coast of Alaska, i.e. around the Norton Sound region.  However, I haven't attempted to assess the statistical significance of this result, and certainly the last few years demonstrate amply that you don't need an El Niño to get a damaging west coast storm.

Autumn Clouds

Following up briefly on last week's post, I took a closer look at cloud cover and solar radiation data from the Fairbanks area, seeking to document the remarkable lack of sunshine in the last few months.  All lines of evidence indicate that it has, indeed, been persistently and unusually cloudy in Fairbanks since mid-August.  Here's a chart showing the daily solar radiation measured since summer at the high-quality CRN site just to the northeast of town:

Of course it's typical for rain and cloud to increase across much of Alaska in August (see this post), but the change this year was more pronounced than usual for the interior, and the weather remained generally cloudier than normal all the way through autumn.

For the 90 day period ending November 10, the CRN instrument measured less radiation than in any other year, with data going back to 2002.  Interestingly, the last 3 years have seen the lowest solar energy in this short history, so it's a persistent anomaly that I strongly suspect is related to the remarkable North Pacific warmth of the last few years.

ERA5 reanalysis data provides a much longer history of estimated solar radiation for the same window, and it sends a similar message:

Of course most of the solar radiation in this autumn window occurs in the first month or so, while the sun still has some strength, so it's worth looking at cloud observations from the airport ASOS for a more evenly distributed perspective on cloudiness.  Here's the result: this was the cloudiest autumn since at least the mid-1950s, if the cloud cover observations are to be believed (and there are certainly inconsistencies in the observing method over time).

To get a sense of the spatial extent of the recurring anomaly, here are solar radiation anomaly maps for the August-October window back to 2020, the last relatively sunny autumn in Fairbanks.

Again, the analysis for the whole seasonal period is focused on the earlier part of the window (mainly August) because of the annual cycle.  However, the last 5 years have also been cloudier than normal in each month separately, so it's not just a late summer change.  The maps below also show a similar trend in the Arctic - at least according to the ERA5 model data.  It's not good news for autumn aurora watchers, or for those who enjoy clear and crisp weather at that time of year.

Autumn Climate Data

Autumn seems like a long time ago for much of Alaska, but it's worth making a mention of the autumn climate anomalies, if only for future reference.  I'll use "autumn" here to refer to the standard climatological season of September through November.

Statewide all three months were slightly warmer than the 1991-2020 average, and the North Slope was considerably warmer than the 30-year baseline, but overall it was a slightly cooler autumn than the last two years (but nowhere near as cold as 2021, when November was very cold).  The North Slope was the only region with a large departure from normal temperature, although parts of Southeast were relatively cool according to the consensus of ERA5 and NCEI data:

More significantly perhaps, September through November was the driest such period in Alaska since 2016, according to NCEI, and nearly all western and southern coastal areas encountered this dryness.  All three months were drier than average on a statewide basis, and that's the first string of three consecutive dry months (relative to normal) since spring 2022.  And yet in contrast, the central interior was very wet, with Fairbanks observing its wettest autumn since 2017; only five autumns have been wetter since 1930.  The wet weather occurred mostly in October.

November was easily the driest month of the autumn, as the Aleutian ridge of October shifted north and took up a dominant position over the Bering Sea and western Alaska - see below.

With Pacific storms held at bay, winds were lighter than normal for most of western Alaska in November, and autumn wind overall wasn't dramatically different from normal for most of the state.  That's a big difference from summer, which was exceptionally windy in the western half of the state.

The calm weather in the Bering Sea allowed SSTs to return to near-normal there, eliminating the cool anomaly that developed back in the summer.  However, exceptional and widespread warmth persisted to the south of the Aleutians, and the Gulf of Alaska remained quite cool throughout autumn.

As a reflection of the extreme temperature anomaly differences across the North Pacific, the PDO index became extremely negative throughout the autumn.  Before the past decade or so of exceptional warmth in the North Pacific and in the Arctic, it was commonly observed that (at least southern) Alaska had a strong tendency for unusual cold during pronounced negative PDO episodes, but we don't see the same robust connection these days.  Consider the map below, showing the average temperature departure from trend in 10 past years with a significantly negative PDO index in autumn:

Compare this to the temperature anomaly map for autumn 2024:

The general orientation of the anomaly patterns is approximately the same, but the 2024 map is considerably warmer overall.  I think this illustrates that even with a favorable SST configuration, cold in Alaska these days tends to be very muted by historical standards.

Getting Chilly

As we move into the last week of September, snow levels are dropping on the mountains, and autumn colors are fading along with the daylight.  But freezing temperatures have been unusually absent and/or mild so far for much of the interior: Fairbanks airport hasn't even dropped below 35°F yet, and Bettles just had its first sub-32°F reading yesterday morning (along with a trace of snow).  This lack of cold is almost unprecedented for the time of year.

The most rapid pace of temperature declines (based on the climatological normal) occurs in October for most of the valley-level interior, as snow cover usually becomes established during the month; this greatly aids the development of persistent temperature inversions, and shuts off the warming effect of the waning sun.  Early-mid October is also the time of most rapid cooling for coastal southern Alaska, where snow cover isn't yet common at low elevation.

Here's a map of the date of "peak cooling" based on ERA5 1991-2020 data:

Note that this is based on smoothing the progression of daily normal values with a harmonic function; the choice of method will affect the result to some extent.

The sharp gradient over the Bering Sea is interesting.  It seems that where ice cover is common by mid-winter, the maximum cooling is delayed, because (I suppose) greater cooling tends to develop after at least some ice is in place.  But in more southerly regions, where ice cover is unusual, peak cooling occurs in mid-October, similar to the Gulf of Alaska.

Here's an equivalent map for the peak rate of spring warming:

The west-east contrast across the interior is more striking in spring, as warming picks up rapidly in the east while it's held back by the ice-covered waters near the West Coast.

First Snow in Fairbanks

The first accumulating snow fell in Fairbanks today - right on time according to the calendar.  The median date since 1930 is October 2nd.  But these were also the first flakes to be reported in the air at the airport, and that typically happens a few days earlier: September 26 on average.

Its getting white outside our office this morning. ❄️A weather front moving east across the Interior will bring areas of light snow ranging from one half to 2 inches before the snow ends tonight. How much snow have you seen so far? Have a nice day out there. #akwx pic.twitter.com/3AezK9w2JX

— NWS Fairbanks (@NWSFairbanks) October 3, 2023

This morning's Fairbanks sounding showed a temperature profile entirely below freezing, and it's also the first time that has happened so far this season.  This is also close to normal.  The odds of any particular 3am sounding being entirely below freezing in Fairbanks rises approximately as follows:

10%   September 20

25%   October 1

50%   October 10

75%   October 27

85-90%   November 15

It's more uncommon to see a sub-freezing profile in the 3pm sounding at the beginning of October, because there's still a modicum of solar heating to produce a modest diurnal temperature cycle.  Today's 3pm sounding reported exactly 0°C a small distance above the ground.  But the historical frequency of a fully sub-freezing afternoon profile rises quickly as that solar heating gets snuffed out:

10%   October 3

25%   October 9

50%   October 18

75%   November 1

85-90%   November 15

Appealing views from the Nenana webcam today:

Record August Warmth

Just a quick note to re-emphasize how extraordinarily unusual the warmth of the last 6 weeks has been for the central and eastern interior, and particularly in terms of daily minimum temperatures.

Here's the leaderboard for average daily low temperatures in Fairbanks from July 16 through the end of August:

2023   55.2°F

2016   52.7°F

1990   52.5°F

2004   52.3°F

1974   52.3°F

2017   52.2°F

2021   52.0°F

This year stands out like a sore thumb.  The departure from normal represents a 3.3 standard deviation anomaly compared to the 1991-2020 distribution.  This is an extreme climate event, with a "return period" of many hundreds of years assuming a stable climate.

Accordingly, it's perhaps no surprise to see that there is *still* very little color in the foliage, based on Nenana and Cleary Summit webcam views from yesterday afternoon:

Here's a chart of Fairbanks airport daily minimum temperatures since June 1:

Looking at monthly statistics for August, Eagle had its warmest August on record (1905 is warmer in the books, but the minimum temperatures look much too high).  Northway was also the warmest on record, while Fairbanks was 3rd warmest overall, but warmest for daily low temperatures.  Gulkana also had its warmest August for daily low temperatures.

But change is in the air: snow was falling in the hills above Fairbanks yesterday morning.  Here's a view from Wickersham Dome, before the sky cleared out later in the day.

Snow at 2600 feet north of Fairbanks this morning! @NWSFairbanks @Climatologist49 pic.twitter.com/pNCsUXk0YO

— Adam Gill (@adam_gillwx) September 2, 2023

Late Summer Nights

Reader Gary alerted me to the fact that so far there's very little autumn color in the trees around Fairbanks, and this is indeed unusual for the date.  Prime season for autumn foliage is typically from late August to early-mid September in the central interior.  Webcam photos from today confirm the report:

(Goldstream Rd at Old Steese Hwy)

(View from Cleary Summit on the Steese Hwy)

(Nenana)

Unusually warm weather is undoubtedly the main cause for the anomaly, and in particular it's the absence of cool nights, I think.  Check out how persistently warm the daily minimum temperatures have been since mid-July.

Remarkably, there has only been one day with a low temperature below 50°F since July 9, and that was a mere 47°F on Sunday morning.  Compare this to the 1991-2020 normal, which dropped below 50°F two weeks ago and is already down to 45°F.

No previous year (since 1930) had less than 3 nights below 50°F by this point in the season, so the absence of overnight chill is unprecedented in the modern climate record.  Not surprisingly, then, the average minimum temperature since mid-July is easily the highest on record:

Besides the dramatic long-term trend, it's also interesting to see the loss of sub-50°F values on this chart in the last 10 years; it seems it was only a matter of time before a new high mark was reached in the series.

In contrast to the persistently mild nights, daytime temperatures have dropped off quite sharply in Fairbanks as the warm and dry fire-weather pattern gave way to cloud, rain, and high humidity.

It's the abundant cloud cover that is now keeping temperatures elevated at night, because the air aloft isn't particularly warm any more.

For those who look forward to the change of season, a glance to the north reveals the future: today saw the first snow of the season at Toolik Lake.

First snow of the year @Toolik pic.twitter.com/0iu2OU2D0F

— Dr. Amanda Young (@arctic_ubetula) August 23, 2023