2026-2027 Record-breaking El Nino

[bluewave]

14 minutes ago, Terpeast said:

The PDO drop back is unfortunate. It’s now around where it was at this time last year, and there is still some time for the PDO to rise back up to near neutral once the marine heat wave area cools off with the season change.

But I think +PDO territory is off the table, and the clock is ticking for us to even get back to neutral. 

The seasonal forecast models continue to underestimate the mid-latitude 500mb ridging pattern leading to this persistent -PDO.

 

[michsnowfreak]

3 hours ago, LibertyBell said:

Did you mean December 1992 Chris? That storm was far more historic than this one.

Like 1983 it won't stop a historic hot summer for next year.

 

Yes but 1983 had an exceptional endless hot summer from June through September so the el nino did not make the summer cooler.

1982-83 and 2015-16 had some similarities. 

If anything it would be less snow than 1997-98 as that late March event was a complete fluke and the airports got less snow. 2001-02 had even less snow. At the coast latitude doesn't really matter unless you're north of 42N you have the potential to be completely shut out of snow. Both 1982-83 and 2015-16 were far better for snow than 1997-98 was with much more late season blocking. In April 1983 JFK had their latest snowfall on record.

 

Again, completely different climate on the east coast vs Great Lakes. The least days with measurable snow on record at Detroit was in 1889-90, when just 15 days recorded measurable snow. But thats still 15 days where snowfall was measured. Even Detroit and Boston, which average near identical snowfall, are completely different climates, but I know further south in NYC its more extreme. All that said, I don't see how you could ever expect a snowless winter in NYC without everything going completely wrong, especially in a wet pattern. I mean, easily rain would dominate over snow. But snowless would be an extraordinary feat. 

[GaWx]

 After having finally gotten back to slightly AN since late Aug (that was 1st time since way back in early May), note the recent drop and how cold it was yesterday N of 80N:

 

 2026 having it this cold N of 80N on Sep 28th is per this site technically the coldest this early in the season in 30 years, quite notable! I say “technically” though because it only barely beat 2011, 2004, and 2000. 

 Regardless, the most interesting comparison is to 2012-25, which were all much warmer/AN on Sep 28th:

1958-2002 mean: -11C
26: -13C
25: -7C
24: -9C
23: -8C
22: -7C
21: -8C
20: -5C
19: -5C
18: -5C
17: -8C
16: -7C
15: -8C
14: -7C
13: -9C
12: -9C
2012-25 mean: -7C

So, regarding Sep 28th, 2026 is:
1. 4-8C colder than every year since 2012
2. 6C colder than the mean of 2012-2025

 Yes, I’m mainly only talking in this post about a point in time and 80N+ is a small area in relation to the N hemisphere (only 1.5%). But it is a very important 1.5% due to Arctic ice. Also, that area is just coming off the coldest late May-early Aug back to 1958 by a good margin, implying the possibility that this overall cooling up there will have legs.

 I’ll keep following this to see if there’s cooling trend follow-through up there going ahead as this is quite interesting, especially due to its implications on Arctic ice since GW had lead to the greatest warming by a good margin in the Arctic. Is it possible that the massive warming up there is done at least for awhile? Could there be a new cycle that’s just started?

@Stormchaserchuck1@bluewave@Wild Weather Monger

[40/70 Benchmark]

1 hour ago, GaWx said:

 After having finally gotten back to slightly AN since late Aug (that was 1st time since way back in early May), note the recent drop and how cold it was yesterday N of 80N:

 

 2026 having it this cold N of 80N on Sep 28th is per this site technically the coldest this early in the season in 30 years, quite notable! I say “technically” though because it only barely beat 2011, 2004, and 2000. 

 Regardless, the most interesting comparison is to 2012-25, which were all much warmer/AN on Sep 28th:

1958-2002 mean: -11C
26: -13C
25: -7C
24: -9C
23: -8C
22: -7C
21: -8C
20: -5C
19: -5C
18: -5C
17: -8C
16: -7C
15: -8C
14: -7C
13: -9C
12: -9C
2012-25 mean: -7C

So, regarding Sep 28th, 2026 is:
1. 4-8C colder than every year since 2012
2. 6C colder than the mean of 2012-2025

 Yes, I’m mainly only talking in this post about a point in time and 80N+ is a small area in relation to the N hemisphere (only 1.5%). But it is a very important 1.5% due to Arctic ice. Also, that area is just coming off the coldest late May-early Aug back to 1958 by a good margin, implying the possibility that this overall cooling up there will have legs.

 I’ll keep following this to see if there’s cooling trend follow-through up there going ahead as this is quite interesting, especially due to its implications on Arctic ice since GW had lead to the greatest warming by a good margin in the Arctic. Is it possible that the massive warming up there is done at least for awhile? Could there be a new cycle that’s just started?

@Stormchaserchuck1@bluewave@Wild Weather Monger

2004 was a weak El Nino...any other warm ENSO years on that list?

[GaWx]

4 minutes ago, 40/70 Benchmark said:

2004 was a weak El Nino...any other warm ENSO years on that list?

2011 was moderate La Niña while 2000 was weak La Niña. Also, both were coming off of a prior year’s Niña.

[40/70 Benchmark]

8 minutes ago, GaWx said:

2011 was moderate La Niña while 2000 was weak La Niña. Also, both were coming off of a prior year’s Niña.

I know that.....I mean the entire list of cold arctic years.

[raindancewx]

We're up to 3.8 inches of rain as of a minute ago for the month. Should pass September 2013 (3.97") to become the wettest month for Albuquerque since July 1930 in the next two hours. Above 4.45" (July 1930), its easily the wettest month in over 100 years. Real shot at passing July 1930 too with ~36 hours of rain/month left as Polo makes landfall two.

Sept rain/temp patterns nationally don't really reflect correlations to El Nino or the PDO, although PDO is irrelevant for US temps in Sept.

 

[GaWx]

8 minutes ago, 40/70 Benchmark said:

I know that.....I mean the entire list of cold arctic years.

I’ll check. But first tell me this: do you mean around the current time (say, the same day (Sep 28th) or late Sept through Oct) or do you mean also later?

[raindancewx]

The PDO tends to decay toward whatever 1.2 is in October, so I do expect it reverse toward neutral fairly rapidly in the coming weeks. But I never really expected it to sustain positive, and that still looks correct. PDO is a pretty weak signal in December nationally. The lightest blue/green is 0.1 r, not r-squared. But PDO is a bigger deal in Oct-Nov and Jan-Feb. The only important tendency with a -PDO December is Western Canada will be cold (probably not flooded with Pacific air).

[40/70 Benchmark]

3 minutes ago, GaWx said:

I’ll check. But first tell me this: do you mean around the current time (say, the same day (Sep 28th) or late Sept through Oct) or do you mean also later?

Good question...I guess the year in the aggregate?

[40/70 Benchmark]

12 minutes ago, raindancewx said:

The PDO tends to decay toward whatever 1.2 is in October, so I do expect it reverse toward neutral fairly rapidly in the coming weeks. But I never really expected it to sustain positive, and that still looks correct. PDO is a pretty weak signal in December nationally. The lightest blue/green is 0.1 r, not r-squared. But PDO is a bigger deal in Oct-Nov and Jan-Feb. The only important tendency with a -PDO December is Western Canada will be cold (probably not flooded with Pacific air).

Yes, I have always been with you on that. Interesting you note the weak signal nationally because some of the seasonal guidance has been hinting that December maybe colder than January and February in the NE.

[40/70 Benchmark]

[GaWx]

56 minutes ago, 40/70 Benchmark said:

Good question...I guess the year in the aggregate?

Coldest 1958-2023 at 80+N for El Niño for aggregate period Sep 28th-Dec 31st:

 

1987: coldest

 

1976:2nd coldest

 

1968: 3rd coldest

 

Others that were also cold overall but not top 3 cold:

1979:

 

1977:

 

1963:

1958:

[40/70 Benchmark]

15 minutes ago, GaWx said:

Coldest 1958-2023 at 80+N for El Niño for aggregate period Sep 28th-Dec 31st:

1987: coldest

 

1976:2nd coldest

 

 

1968: 3rd coldest

 

Others that were also cold overall but not top 3 cold:

1979:

 

1977:

 

1963:

1958:

1963 and 1968 (both -PDO) are the only El Nino events of appreciable intensity....1958, 1976 and 1977, like 2004, were weak.

[GaWx]

14 minutes ago, 40/70 Benchmark said:

1963 is the only El Nino of appreciable intensity....1958, 1976 and 1977, like 2004, were weak....1968 was borderline weak/mod.

What about the very cold 1987, which peaked higher (+1.3) although early (JAS/ASO)? Are you not counting it due to early peak?

[GaWx]

22 minutes ago, 40/70 Benchmark said:

1963 and 1968 (both -PDO) are the only El Nino events of appreciable intensity....1958, 1976 and 1977, like 2004, were weak.

 If I had added another El Niño to the list, it would have been 1991, which did average slightly BN Sep 28th-Dec 31st and thus is the coldest strong to very strong Nino at 80+ N for 9/28-12/31 averaged out:

[40/70 Benchmark]

16 minutes ago, GaWx said:

What about the very cold 1987, which peaked higher (+1.3) although early (JAS/ASO)? Are you not counting it due to early peak?

Yea, missed that one.

[PhiEaglesfan712]

8 hours ago, LibertyBell said:

Did you mean December 1992 Chris? That storm was far more historic than this one.

I thought the historic storm was in March 1993, not in December 1992. I always remember the winter of 1992-93 this way: (1) my youngest sister being born (December 29, 1992), then (2) her baptism (late January 1993), and finally the snowstorm (mid-March 1993).

I don't remember the storm happening first (in December 1992). I found this storm (which was right before my sister was born), but it appears further south: https://products.climate.ncsu.edu/weather/winter/event/?e=128

[OSUmetstud]

7 hours ago, GaWx said:

 After having finally gotten back to slightly AN since late Aug (that was 1st time since way back in early May), note the recent drop and how cold it was yesterday N of 80N:

 

 2026 having it this cold N of 80N on Sep 28th is per this site technically the coldest this early in the season in 30 years, quite notable! I say “technically” though because it only barely beat 2011, 2004, and 2000. 

 Regardless, the most interesting comparison is to 2012-25, which were all much warmer/AN on Sep 28th:

1958-2002 mean: -11C
26: -13C
25: -7C
24: -9C
23: -8C
22: -7C
21: -8C
20: -5C
19: -5C
18: -5C
17: -8C
16: -7C
15: -8C
14: -7C
13: -9C
12: -9C
2012-25 mean: -7C

So, regarding Sep 28th, 2026 is:
1. 4-8C colder than every year since 2012
2. 6C colder than the mean of 2012-2025

 Yes, I’m mainly only talking in this post about a point in time and 80N+ is a small area in relation to the N hemisphere (only 1.5%). But it is a very important 1.5% due to Arctic ice. Also, that area is just coming off the coldest late May-early Aug back to 1958 by a good margin, implying the possibility that this overall cooling up there will have legs.

 I’ll keep following this to see if there’s cooling trend follow-through up there going ahead as this is quite interesting, especially due to its implications on Arctic ice since GW had lead to the greatest warming by a good margin in the Arctic. Is it possible that the massive warming up there is done at least for awhile? Could there be a new cycle that’s just started?

@Stormchaserchuck1@bluewave@Wild Weather Monger

Its a data bug. The reanalysis data up there for July and August is near normal.

https://x.com/WorldClimateSvc/status/2075198146641051650

 

[40/70 Benchmark]

3 hours ago, PhiEaglesfan712 said:

I thought the historic storm was in March 1993, not in December 1992. I always remember the winter of 1992-93 this way: (1) my youngest sister being born (December 29, 1992), then (2) her baptism (late January 1993), and finally the snowstorm (mid-March 1993).

I don't remember the storm happening first (in December 1992). I found this storm (which was right before my sister was born), but it appears further south: https://products.climate.ncsu.edu/weather/winter/event/?e=128

There was a historic storm during both months....obvioously March was the larger scope, but that 92 nor' easter was easy for the NE...historic CF and up to 4' of snow in the Berkshires and 3' in the Worcester hills.

[GaWx]

4 hours ago, OSUmetstud said:

Its a data bug. The reanalysis data up there for July and August is near normal.

https://x.com/WorldClimateSvc/status/2075198146641051650

 

Thanks for posting about this.

 After reading your post, I decided it would be worthwhile to check out another forum (which includes other pro-mets):

https://forum.arctic-sea-ice.net/index.php

 I read hundreds of posts in their “The 2026 melting season” thread. The first post I saw mentioning the unusually cold DMI 80+ N temperatures was on 5/25/26 (post #268) on the following page (p. 6) by very experienced pro-met Niall Dollard:

https://forum.arctic-sea-ice.net/index.php/topic,4507.250.html

 Niall said, “I suspect we will see a rapid increase on the DMI N80 graph in the next day or two, as that deep low advects warmth into the general DMI N80 area near the pole. 

Unusually, the daily mean has been below the 1958-2002 mean for much of the month of May.”

————————
  Then go to p. 9 and look at post #474 on 7/16/26 from pro-met Glen Koehler in here:

https://forum.arctic-sea-ice.net/index.php/topic,4507.450.html

Glen said: “The DMI 80N values are unprecedented.  I looked at every annual chart for 1958-2026, and no other year has such a large and prolonged period of below 2C temperature during June and July.”  
 
and

”The 2026 DMI 80N chart is so anomalous that it makes me wonder if a sensor or data processing change is affecting the 80N chart in summer 2026.  Anoother possibility is that the slushification of the ice near the North pole is creating a different relationship between ice and 2M air temperature.”

————————

Then @TheClimateChanger, who you may recall used to go by “Frivolous” here, posted in #490 (also on p.9) as “Frivolousz21” on 7/18/26 as a reply to Glen:

”It's not from an error. It's just that there has been a relentless PV anomaly and essentially next to no unimpeded solar insolation.  

 It's been pretty crazy.  This might end up the best year for ice retention since 2006.”

————————

 I went through the rest of the thread and there never was agreement on the cause. So, it remains a mystery there.

 I respect your knowledge and appreciate your post on this. But I plan to continue posting these DMI graphs at least until if/when there’s strong agreement in the overall pro-met community with you that 2026 really isn’t as cold relative to past years as the graphs show.

——————

 Today’s daily update had another sharp cooling (~1C) from yesterday keeping it colder than the blue line’s climo:

 I just looked at the model consensus and it is, indeed, solidly BN N of 80N. However, those same models are showing significant warming to AN very quickly. So, I’m expecting it to go back above the blue line within just a couple of days at most.

[so_whats_happening]

On 9/27/2026 at 10:37 AM, bluewave said:

I think that’s why the long range models continue the WWBs and the extreme east based nature of this event right into next June. 

 

It would be fascinating to see happen as it would be a very 1982-83 style end.

[so_whats_happening]

1 hour ago, GaWx said:

Thanks for posting about this.

 After reading your post, I decided it would be worthwhile to check out another forum (which includes other pro-mets):

https://forum.arctic-sea-ice.net/index.php

 I read hundreds of posts in their “The 2026 melting season” thread. The first post I saw mentioning the unusually cold DMI 80+ N temperatures was on 5/25/26 (post #268) on the following page (p. 6) by very experienced pro-met Niall Dollard:

https://forum.arctic-sea-ice.net/index.php/topic,4507.250.html

 Niall said, “I suspect we will see a rapid increase on the DMI N80 graph in the next day or two, as that deep low advects warmth into the general DMI N80 area near the pole. 

Unusually, the daily mean has been below the 1958-2002 mean for much of the month of May.”

————————
  Then go to p. 9 and look at post #474 on 7/16/26 from pro-met Glen Koehler in here:

https://forum.arctic-sea-ice.net/index.php/topic,4507.450.html

Glen said: “The DMI 80N values are unprecedented.  I looked at every annual chart for 1958-2026, and no other year has such a large and prolonged period of below 2C temperature during June and July.”  
 
and

”The 2026 DMI 80N chart is so anomalous that it makes me wonder if a sensor or data processing change is affecting the 80N chart in summer 2026.  Anoother possibility is that the slushification of the ice near the North pole is creating a different relationship between ice and 2M air temperature.”

————————

Then @TheClimateChanger, who you may recall used to go by “Frivolous” here, posted in #490 (also on p.9) as “Frivolousz21” on 7/18/26 as a reply to Glen:

”It's not from an error. It's just that there has been a relentless PV anomaly and essentially next to no unimpeded solar insolation.  

 It's been pretty crazy.  This might end up the best year for ice retention since 2006.”

————————

 I went through the rest of the thread and there never was agreement on the cause. So, it remains a mystery there.

 I respect your knowledge and appreciate your post on this. But I plan to continue posting these DMI graphs at least until if/when there’s strong agreement in the overall pro-met community with you that 2026 really isn’t as cold relative to past years as the graphs show.

——————

 Today’s daily update had another sharp cooling (~1C) from yesterday keeping it colder than the blue line’s climo:

 I just looked at the model consensus and it is, indeed, solidly BN N of 80N. However, those same models are showing significant warming to AN very quickly. So, I’m expecting it to go back above the blue line within just a couple of days at most.

This is something you certainly do not see very often in this warmer world, especially in the confines of the Arctic. It was a solid retention year to say the least. Whether the temps are representative or not the overall pattern would suggest things didn't go haywire. 

It was a shame we started at such a low point from the previous winter which is weird coming out of La Nina to see such low volumes for sea ice.

Will be interesting to see how this plays out going into winter.

[GaWx]

16 minutes ago, so_whats_happening said:

This is something you certainly do not see very often in this warmer world, especially in the confines of the Arctic. It was a solid retention year to say the least. Whether the temps are representative or not the overall pattern would suggest things didn't go haywire. 

It was a shame we started at such a low point from the previous winter which is weird coming out of La Nina to see such low volumes for sea ice.

Will be interesting to see how this plays out going into winter.

I started following the model consensus of 2m temps for the high Arctic regularly since May. The majority of the time through early August they initialized with a mean of BN fwiw.

[so_whats_happening]

5 minutes ago, GaWx said:

I started following the model consensus of 2m temps for the high Arctic regularly since May. The majority of the time through early August they initialized with a mean of BN fwiw.

Yea 2m temps are tough over the arctic, typically you want to look at 925mb for temp anomalies to get better look at the overall. So average to below average for most of the time period over roughly 75N to the pole. 

[OSUmetstud]

7 hours ago, GaWx said:

Thanks for posting about this.

 After reading your post, I decided it would be worthwhile to check out another forum (which includes other pro-mets):

https://forum.arctic-sea-ice.net/index.php

 I read hundreds of posts in their “The 2026 melting season” thread. The first post I saw mentioning the unusually cold DMI 80+ N temperatures was on 5/25/26 (post #268) on the following page (p. 6) by very experienced pro-met Niall Dollard:

https://forum.arctic-sea-ice.net/index.php/topic,4507.250.html

 Niall said, “I suspect we will see a rapid increase on the DMI N80 graph in the next day or two, as that deep low advects warmth into the general DMI N80 area near the pole. 

Unusually, the daily mean has been below the 1958-2002 mean for much of the month of May.”

————————
  Then go to p. 9 and look at post #474 on 7/16/26 from pro-met Glen Koehler in here:

https://forum.arctic-sea-ice.net/index.php/topic,4507.450.html

Glen said: “The DMI 80N values are unprecedented.  I looked at every annual chart for 1958-2026, and no other year has such a large and prolonged period of below 2C temperature during June and July.”  
 
and

”The 2026 DMI 80N chart is so anomalous that it makes me wonder if a sensor or data processing change is affecting the 80N chart in summer 2026.  Anoother possibility is that the slushification of the ice near the North pole is creating a different relationship between ice and 2M air temperature.”

————————

Then @TheClimateChanger, who you may recall used to go by “Frivolous” here, posted in #490 (also on p.9) as “Frivolousz21” on 7/18/26 as a reply to Glen:

”It's not from an error. It's just that there has been a relentless PV anomaly and essentially next to no unimpeded solar insolation.  

 It's been pretty crazy.  This might end up the best year for ice retention since 2006.”

————————

 I went through the rest of the thread and there never was agreement on the cause. So, it remains a mystery there.

 I respect your knowledge and appreciate your post on this. But I plan to continue posting these DMI graphs at least until if/when there’s strong agreement in the overall pro-met community with you that 2026 really isn’t as cold relative to past years as the graphs show.

——————

 Today’s daily update had another sharp cooling (~1C) from yesterday keeping it colder than the blue line’s climo:

 I just looked at the model consensus and it is, indeed, solidly BN N of 80N. However, those same models are showing significant warming to AN very quickly. So, I’m expecting it to go back above the blue line within just a couple of days at most.

The rather obvious issue is youre comparing the ecmwf operational forecast 2 m temperature all summer to the ERA40 (old climatology) temperature of the reanalysis (apples to oranges). Copernicus/ECMWF climate pulse i posted is comparing the reanalysis anomalies to 1991-2020 ERA5 climatology and clearly by that standard (apples to apples) the summer temperatures are not crazy anomalous. 

[OSUmetstud]

5 hours ago, so_whats_happening said:

Yea 2m temps are tough over the arctic, typically you want to look at 925mb for temp anomalies to get better look at the overall. So average to below average for most of the time period over roughly 75N to the pole. 

This plot is not consistent with 2 m temperature being 2 C below the much older 1958 to 2002 climatology all summer. This area is all ocean/ice. 

[OSUmetstud]

The ecmwf was upgraded on May 12, 2026. What do you notice?

IFS users will also see the introduction of the new NEMO4-SI³ ocean and sea-ice model, bringing substantial improvements in forecasts and analysis over the marine areas, as well as over 40 new ocean and sea ice-related variables. Key improvements include a more accurate representation of how sea ice affects the power of waves and the interactions between waves and ocean currents, often at play in rougher seas. Other improvements for IFS include improved tropical upper-air temperature and wind forecasts, and improvements to temperature and humidity forecasts around the tropopause.

 

https://www.ecmwf.int/en/about/media-centre/news/2026/ifs-cycle-50r1-aifsv2-live

What's likely happening here is actually the new ecmwf 2 m temperatures over the Arctic are more accurate as the near surface inversion is more properly coupled to the ice and the cold ocean. The old 2 m ERA40 climatology is what's too warm.