Autumn Threatens

The past 12 days or so have been significantly cooler than normal statewide, and southwestern and south-central Alaska have been very chilly and wet, as unusual low pressure has prevailed in the eastern Bering Sea.  The 500mb height anomaly over western Alaska in the last two weeks looks remarkably similar to the pattern in May:

Model-derived 850mb temperatures (about 5000 feet above sea level) recently hovered not far above record low levels for the time of year over Bristol Bay; the chart below is taken from the vicinity of Dillingham.

Rainfall has been abundant and locally excessive in southwestern and south-central Alaska:

Remarkably, Anchorage is currently tied in first place for the wettest July on record (data back to 1952), and this is the 4th year out of the last 5 to see more than 4 inches of rain in July - with a long-term median of only about 1.7 inches.  In 75 years of data, 4 of the 8 Julys with over 4" of rain have occurred since 2022.

McGrath is close to a record wet July as well, with 5.78" so far versus the 2024 record of 6.16".  Bethel's 4.49" so far this month is the highest since before 1930.

The Yentna and Talkeetna rivers are both in flood:

With all the rain and cloud recently, temperatures have trended down rather than up since early June in Anchorage:

Looking ahead, a ridge is now building over eastern Siberia, and that means cold air is dropping south over the state and will be reinforced early next week.  The latest ensemble-mean ECMWF forecast for Monday shows very unusual cold over the Brooks Range, perhaps threatening records for the month of July.  There will certainly be more snow; the Toolik Lake webcam is already showing patchy snow accumulation this morning.

Here's a lower-atmosphere temperature chart for a central Brooks Range location, with the forecast portion on the right.

Intense El Niño Under Way

I'm sure all readers of this blog are well aware that a major El Niño is expected to prevail in the coming months; the media coverage has been widespread.  However, it's worth posting a few comments of my own, including some Alaska perspective.

The first thing to observe is that this is not shaping up to be a run-of-the-mill El Niño episode; it is going to be an extremely strong event - almost certainly one of the strongest on record, and quite possibly the strongest in many decades (more than a century).  This assessment is not just a long-range forecast, because the event is already well under way: sea surface temperatures are already far above normal across the central and eastern equatorial Pacific.  Here's a recent map of the SST departure from the 1991-2020 normal:

The central Pacific Niño3.4 region is the most widely used region for tracking El Niño SSTs, and the departure from normal is already about +2.2°C:

An anomaly of +2°C is already in the upper echelon of historical El Niño events, and it's easily a record for the time of year, because El Niño's strength tends to peak in Northern Hemisphere winter.  In previous events of recent decades, the earliest the Niño3.4 anomaly reached +2.2°C was in October (1997 and 2015).

So how strong do we expect this El Niño to become?  The chart below shows a progression of forecasts from the Copernicus seasonal models in recent months.

Three things stand out in this graphic:

- The forecasts have been escalating with each update

- The forecasts have generally lagged behind the observed rate of warming in the Niño3.4 region (black line)

- The latest update, issued this month, raised the bar more significantly than in any previous update.  The multi-model predicted Niño3.4 SST anomaly now exceeds +3.5°C, far above the previous monthly-average record of +2.7°C in November 2015 (according to NOAA CPC).

Given that the planet is warming, we should arguably remove the background warming for a fair comparison with previous events.  Using the relative Niño index instead, the multi-model forecast expects a peak monthly value of +3.1°C, compared to a record of +2.7°C in January 1983; so we're still looking at a record event even after "detrending".

It's hard to overstate how significant the impacts of this El Niño will be on weather and climate patterns, especially in the tropics; droughts and floods will be directly linked to the altered atmospheric circulation.  In Alaska, the most reliable effect of a robust El Niño is to increase the strength of the Aleutian Low in winter, and especially in late winter, bringing an increased tendency for southerly and chinook winds across most of the mainland - and therefore widespread above-average temperatures (particularly in eastern Alaska).  Winter snowfall tends to be below-normal to the north of the Alaska Range.

Here's a look at monthly temperature anomalies from August through April in the last three "super" El Niño events: 1982-83 on the left, 1997-98 in the center, and 2015-16 at right.  Click to enlarge.

The only consistent feature across all three events is the tendency for late winter and spring warmth, from February through April.  Interestingly, December and January were cold across northern Alaska in two of the three winters, and October was very cold in 1982 and 1997 (5th and 7th coldest Octobers on record statewide, respectively).  2015-16 was much warmer overall than the previous two episodes, and in fact the extended cold season and spring of 2015-16 (e.g. October through May) was the warmest on record for Alaska.

Here are time series charts of Fairbanks temperatures in the three winters, relative to a modern normal:

It's an interesting question as to which of the three most recent "super" El Niños is most similar to the current one.  Based on SSTs in the tropical Pacific, recent conditions look a lot like 2015, but the negative PDO phase that continues to persist is more like 1982 than 2015 (and is certainly very unlike 1997).  Notice in the figure below the warmth east of Japan and the relatively cool waters near Alaska in 1982 and 2026.  There are a few other intriguing similarities to 1982 that I won't get into; but of course no two years are exactly alike, and this likely record-breaking El Niño will undoubtedly have some surprises in store.  

June Climate Data

Last year at this time I commented on the unusual high-latitude circulation pattern in June 2025, when a broad but very strong circumpolar circulation extended up into the stratosphere.  Low pressure in the Bering Sea, related to the strong jet stream aloft, produced a wet and windy June for Aleutian and Bering Sea communities last year.

Interestingly, there were some similarities in the flow pattern last month - June 2026 - with a strong polar vortex extending to the stratosphere.  However, this year the vortex was much more compact and intense, residing over the central Arctic Ocean between Alaska and the Pole.  The following figure shows the rank of the June average 500mb height; the mid-atmosphere pressure was the lowest on record for June in the high Arctic, but strong ridges of record high pressure occurred to the south over central Russia, the western Bering Sea, and eastern Arctic Canada.

The coherent vertical structure of the vortex is evident when we look at 100mb height (pressure), above 50,000 feet.

And down at sea level: record low sea-level pressure under the center of the vortex.

A strong polar vortex is a "cold core" feature, so temperatures were unusually low - in fact, record low for the month of June over the sea ice to the north of the Chukchi Sea (based on ERA5 data since 1950):

What relevance did this have for Alaska in June?  With the vortex being very much more compact than last year, the outcome was very different.  In fact, the weather pattern specific to Alaska was in some ways opposite to last June; low pressure was confined to the north, and the Bering Sea saw above-normal MSLP instead of last year's stormy pattern.  As a consequence, it was much drier and sunnier than normal over western Alaska and the Bering region:

Temperatures were also partly inverted from last June, with a relatively cool month in the north and warm in the southwest, approximately opposite of last year:

With the Arctic vortex extending a lobe down across northern Alaska (between the Bering and Canadian ridges), it was chilly on the North Slope last month, with the lowest June temperature since 2018.  Somewhat remarkably - considering the pronounced warming trend - it was the 7th consecutive month with mean temperature below the 1991-2020 normal for the North Slope climate division.

One major similarity between this year and last year, however, was the extensive lightning; June 2026 ranks as 3rd highest for lightning strikes in the month, and last year saw the most strikes on record for June (data since 2012 - and excluding ALDN-recorded strikes in northwestern Canada).

Comparing Pacific Ocean SSTs between the two years, we see that the zone of unusual warmth extending east from Japan is modestly reduced in size and intensity from last year.  There's also more cool water this year in the Gulf of Alaska; but the overall North Pacific configuration is quite similar and remains indicative of a negative PDO phase.

The big difference, of course, is farther south, and it is an enormous difference: strong El Niño conditions are already established in the equatorial Pacific, and we're well on the way to one of the strongest El Niños on record - and possibly the strongest in more than a century.  More on that important story in another post.

Hail Reports

The Fairbanks area saw some hefty showers and thunderstorms yesterday and last Friday, and hail was reported around town on both days.  The storm on Friday afternoon was isolated but quite intense, with three-quarter inch hail reported.  Here's a radar image from 3pm, when the storm's intensity peaked as it moved from northwest to southeast over the north side of town (click to enlarge):

Another cell dropped pea-sized hail at the NWS office on the UAF campus a little while later.

Yesterday saw a larger and more organized storm track into town from the northeast, bringing both hail (up to half-inch reported) and damaging wind gusts.  The radar reflectivity reached an impressive 65 dbZ over quite a lengthy track.

Courtesy of the NWS on Twitter/X, this was the view to the northeast from the airport at the same time as the radar image above:

There was an impressive display of mammatus clouds after the storm passed.  These clouds form as pockets of cold air, loaded with ice crystals, sink out of the high-level cloud shield that forms in the wake of a mature thunderstorm.

Here's a simple animation of the storm evolution on radar:

As of yesterday, Fairbanks is up to 7 days so far this season with thunder reported at the airport.  That's the same as last year at this date, and it's a little ahead of the long-term normal of 4-5 days.  The normal for the whole season is 9-10 days.

Statewide lightning activity has resumed climbing since Friday, and the year-to-date strike count total has nearly reached the median for the seasonal total.

The Lower Yukon region is now well above the 2012-2025 record.

Meanwhile, there's a heat advisory in effect for the interior North Slope, and the NWS forecast for Umiat is currently calling for 87°F tomorrow.  The RAWS thermometer there will no doubt be a few degrees too high as usual, so it will be interesting to see if we can get a 90°F reading - it would be the first since 2013.

[Update: it reached 83°F, a typical value for the summer's highest temperature in Umiat.]

Fire Suppression

The multi-day spate of widespread lightning activity in the second half of June produced a 9-day total of about 57,000 cloud-to-ground lightning strikes, according to the Alaska Fire Service's detection network.  It was a substantial wave of activity, but not all that unusual in the context of historical data since 2012; four other years had larger 9-day totals (2025, 2015, 2019, 2022), with the record being last year's 69,000 strikes that occurred just a few days earlier on the calendar.

As noted previously, the lightning outbreak was unusual for the level of activity in the southwest and northern interior regions.  The Lower Yukon zone, in particular, saw a very intense onslaught compared to the climatological normal, and the year-to-date activity there is already tied with the 2019 record for the entire season.

Here's the actual strike density in absolute terms for the month of June:

The northwestern and southeastern interior were relatively inactive, with the most notable shortfall being in the Koyukuk and Upper Kobuk zone:

It has been something of a surprise to see wildfire activity remain very subdued despite all the lightning.  There have been fire starts, of course, but they haven't grown; statewide acreage is only 29,000 acres, a mere 10% of the climatological median.  The chart below shows that the disparity between the number of fire starts and fire acreage is very unusual; no other year since 1993 had so little acreage with more than 100 new fires in June.

Similarly, the fire acreage relative to the number of lightning strikes is extremely low:

Although last year had more fire, it was qualitatively similar: somewhat subdued fire growth relative to the extensive lightning activity.  At least one key factor is that both years had a wet spring across most of the fire-prone interior:

Contrast this with 2022, for example, which saw very efficient fire growth (a lot of acres per fire and per lightning strike) after a dry (but not warm) spring:

Of course the weather in June also makes a huge difference for the fate of lightning-sparked fires; 2004 and 2013 were extremely warm in June, and fire growth was rapid.  Last month had some warm days, but the second half of the month was near-normal for Fairbanks temperature overall; and there's been a substantial amount of rain in many areas in the past two weeks.  Obviously when lightning is accompanied by widespread and significant rain - as opposed to "dry" thunderstorms - the fire activity is more often suppressed.

Not surprisingly, there is a correlation between June fire acreage and the rest of the season (fires tend to keep burning), but the weather is the key factor.  For instance, 2009 was extremely warm and dry in July, producing big fire activity after a slow start.  Hopefully we won't see anything similar this year.