2026-2027 Record-breaking El Nino

[40/70 Benchmark]

1 hour ago, GaWx said:

 This is not only the coldest CFSv2 forecast for JFM made in summer for the SE US to C Plains part of the U.S. in El Niño, it is by a good margin THE coldest JFM forecast for this area that I could find made from any season and in any ENSO back to the start of the CFSv2 in 2012! This is extreme, folks!

That doesn't look like prohibitive warmth in SNE, either....like 2004-2005 warmth probably. ... +1-+3F. Give me that all day long if it comes with tons of moisture....I'll take that over last year in a heartbeat. 

2004-2005 was slightly above average in terms of temps, but only 1995 and 2014 saw higher total seasonal snowfall IMBY.

[40/70 Benchmark]

1 hour ago, GaWx said:

 This is not only the coldest CFSv2 forecast for JFM made in summer for the SE US to C Plains part of the U.S. in El Niño, it is by a good margin THE coldest JFM forecast for this area that I could find made from any season and in any ENSO back to the start of the CFSv2 in 2012! This is extreme, folks!

That CFS plot looks like 2009-2010 IMO.

[bluewave]

12 hours ago, bncho said:

Not to be a downer or anything but is it possible/likely that the models that show the favorable Feb/Mar pattern are just falling back on what a normal El Niño would generally do? This is not a normal El Niño. I think @bluewave has touched on this in seasons past about how seasonal models are susceptible to this? Just worried for something like Feb 2024 to happen.

Seasonal model forecasts for super El Niños usually have warmer temperatures across Canada and the Northern tier states into the Northeast. So in that general sense they usually end up correct. But they haven’t been correct with the magnitude of the warmth.

Spots that wound up with +10 or more months have only been forecast to be something like +1 to 3 or maybe +2 to +4. Plus each of the 3 super El Niños since the late 90s had other variations in regard to where the ridge axis was located vs forecast.

1997-1998 was our only east based event of that era with a strong +PDO and no Maritime Continent forcing. But it started out cooler in the East during December and got progressively warmer during January and February. So it didn’t follow the general El Niño progression. It also had a strong -AO in January that linked up with the Southeast ridge even with such a strong STJ.

The models had the stock El Niño composite for December 2015 as late as November 2023 with warmer north and cooler south. We wound up with the warmest December on record in the East greater than +13 in spots which were forecast to be something like +1 to +3. Strong +PDO that winter and a Maritime Cintinent influence in December

2023-2024 was forecast to be only something like +1 to 3 for places  which verified more than +10. The event  had a -PDO with Maritime Continent influence.

This event is stronger than all three of those events. So the models probably won’t start showing the actual December pattern until into the first week of December. The January pattern probably won’t begin to show up until closer to New Years. And the actual February pattern until that month starts also.

 

 

 

 

[mitchnick]

5 hours ago, GaWx said:

 This is not only the coldest CFSv2 forecast for JFM made in summer for the SE US to C Plains part of the U.S. in El Niño, it is by a good margin THE coldest JFM forecast for this area that I could find made from any season and in any ENSO back to the start of the CFSv2 in 2012! This is extreme, folks!

Glad you posted it and not me! Lol

[mitchnick]

6 hours ago, GaWx said:

 This is not only the coldest CFSv2 forecast for JFM made in summer for the SE US to C Plains part of the U.S. in El Niño, it is by a good margin THE coldest JFM forecast for this area that I could find made from any season and in any ENSO back to the start of the CFSv2 in 2012! This is extreme, folks!

Looking at the Cfs2 past March, it warms up North America pretty fast getting you to this early forecast for May. So the relative extreme heat from this Niño does get distributed thru the conus/planet, but not until late spring into summer and beyond. Just a thought.

[GaWx]

 Yesterday’s NOAA weekly release for relative 3.4 came in at +1.8, which is the warmest early August on record back to 1982:
 

29JUL2026         3.2        2.1        1.5        0.2

05AUG2026        3.3        2.4        1.8        0.3

 ————————

How does this compare to 1997?

06AUG1997         3.9        2.3        1.6        0.3

——————

How does this +1.8 for 3.4 compare to others for weeks centered near August 5th?

2026 +1.8

2023 +0.4

2015 +1.1

2012 +0.7

2009 +0.4

2006 +0.2

2004 +0.7

2002 +0.7

1997 +1.6

1994 +0.8

1993 +0.4

1991  +0.7

1990 +0.4

1987 +1.2

1986 +0.3

1982 +1.0

 

 The earliest any of these reached +1.8 is August 20, 1997.

[40/70 Benchmark]

3 hours ago, bluewave said:

Seasonal model forecasts for super El Niños usually have warmer temperatures across Canada and the Northern tier states into the Northeast. So in that general sense they usually end up correct. But they haven’t been correct with the magnitude of the warmth.

Spots that wound up with +10 or more months have only been forecast to be something like +1 to 3 or maybe +2 to +4. Plus each of the 3 super El Niños since the late 90s had other variations in regard to where the ridge axis was located vs forecast.

1997-1998 was our only east based event of that era with a strong +PDO and no Maritime Continent forcing. But it started out cooler in the East during December and got progressively warmer during January and February. So it didn’t follow the general El Niño progression. It also had a strong -AO in January that linked up with the Southeast ridge even with such a strong STJ.

The models had the stock El Niño composite for December 2015 as late as November 2023 with warmer north and cooler south. We wound up with the warmest December on record in the East greater than +13 in spots which were forecast to be something like +1 to +3. Strong +PDO that winter and a Maritime Cintinent influence in December

2023-2024 was forecast to be only something like +1 to 3 for places  which verified more than +10. The event  had a -PDO with Maritime Continent influence.

This event is stronger than all three of those events. So the models probably won’t start showing the actual December pattern until into the first week of December. The January pattern probably won’t begin to show up until closer to New Years. And the actual February pattern until that month starts also.

 

 

 

 

1997 was easterly QBO/Low-ascending solar, so it isn't that surprising that the polar domain was more disturbed early on.

[40/70 Benchmark]

1 hour ago, mitchnick said:

Looking at the Cfs2 past March, it warms up North America pretty fast getting you to this early forecast for May. So the relative extreme heat from this Niño does get distributed thru the conus/planet, but not until late spring into summer and beyond. Just a thought.

That quick warm up in early spring is another parallel to 2009-2010.

[PhiEaglesfan712]

7 hours ago, 40/70 Benchmark said:

That CFS plot looks like 2009-2010 IMO.

Yeah, it's time to start paying attention. When there is a cold signal this strong during a robust el nino, it's trying to tell you something. Just look at 2009-10, which had a very strong cold signal there in the fall (of 2009).

Just now, 40/70 Benchmark said:

That quick warm up in early spring is another parallel to 2009-2010.

That one came with a backlash la nina. The 2009-10 strong el nino formed and dissipated quickly, which allowed a transition to a strong la nina. I wonder if the upcoming el nino peaks early and gives way to the development of a strong la nina in 2027-28. If it does, then expect a record warm summer in the East. Backlash la nina years, like 1973, 1988, and 2010, have produced record warm summers at PHL.

[40/70 Benchmark]

8 minutes ago, PhiEaglesfan712 said:

Yeah, it's time to start paying attention. When there is a cold signal this strong during a robust el nino, it's trying to tell you something. Just look at 2009-10, which had a very strong cold signal there in the fall (of 2009).

That one came with a backlash la nina. The 2009-10 strong el nino formed and dissipated quickly, which allowed a transition to a strong la nina. I wonder if the upcoming el nino peaks early and gives way to the development of a strong la nina in 2027-28. If it does, then expect a record warm summer in the East. Backlash la nina years, like 1973, 1988, and 2010, have produced record warm summers at PHL.

I would be careful with buying into it full-stock....I feel like these Modoki like solutions are overdoing it in the seasonal mean because they are undergoing the magnitude of the warmer interludes early on. Rather, I interpret those solutions as validation for my conceptualization of how the second half will evolve.

[40/70 Benchmark]

The westerly QBO and the descending solar are two red flags for that kind of an outcome in the seasonal mean, neither of which were present in 2009....not to mention nearly 20 years of CC.

[GaWx]

39 minutes ago, 40/70 Benchmark said:

The westerly QBO and the descending solar are two red flags for that kind of an outcome in the seasonal mean, neither of which were present in 2009....not to mention nearly 20 years of CC.

1982-3 and 2015-6 are two super strong ones with both a +QBO and descending solar fwiw.

1957-8 was a super strong with a +QBO and was barely after peak.

[40/70 Benchmark]

16 minutes ago, GaWx said:

1982-3 and 2015-6 are two super strong ones with both a +QBO and descending solar fwiw.

1957-8 was a super strong with a +QBO and was barely after peak.

Yup...which is why they are my two top polar analogs.

[GaWx]

 Mainly for entertainment: The image below is the TT 2m mean of the last 12 CFSv2 ops for Feb ‘27. It is extremely cold from the C Plains to the SE US with anomalies of -8 to -11 F even vs 1984-2009 climo, which is a bit colder than 1991-2020!
 

 It’s just about as cold there as the frigid Febs of 1978, 1958, and 1905 and is a bit colder than Febs of 2015, 2010, 1968, 1963, 1947, and 1902. So, it is the coldest Feb there since 1978. It isn’t quite as cold as the record coldest, 1895:

[mitchnick]

7 minutes ago, GaWx said:

 Mainly for entertainment: The image below is the TT E US 2m mean of the last 12 CFSv2 for Feb ‘27. It is extremely cold from the C Plains to the SE US with anomalies of -8 to -11 F even vs 1984-2009 climo, which is a bit colder than 1991-2020!
 

 It’s just about as cold there as the frigid Febs of 1978, 1958, and 1905 and is a bit colder than Febs of 2015, 2010, 1968, 1963, 1947, and 1902. It isn’t quite as cold as the record coldest, 1895:

Cansips not too much different in the Conus really.

[40/70 Benchmark]

2 hours ago, mitchnick said:

Cansips not too much different in the Conus really.

That cold in the high plains is NOT stock El Niño.

[MJO812]

I always watch his videos. Very knowledgeable  https://youtu.be/DrAPIw8i0iA?is=e1-O1Q_oXIphw5js

[Jersey Andrew]

Does the cold model forecast for this winter jive in any way with Super El Niño? Are there reasons to believe this winter season goes against the grain and produces widespread cold and snow?

[GaWx]

6 minutes ago, Jersey Andrew said:

Does the cold model forecast for this winter jive in any way with Super El Niño? Are there reasons to believe this winter season goes against the grain and produces widespread cold and snow?

If so, Feb would seem to be the best opportunity per models and analogs with a mild Dec and in between Jan.

[LakePaste25]

8 minutes ago, Jersey Andrew said:

Does the cold model forecast for this winter jive in any way with Super El Niño? Are there reasons to believe this winter season goes against the grain and produces widespread cold and snow?

I don’t believe it. I think there could be some opportunities but otherwise it’s way overdone. 

[brooklynwx99]

so the CFS basically shoves the eastern Canadian ridge into the Davis Strait, lol 

it's an unprecedented super Nino but I'll believe that when I see it

[Stormchaserchuck1]

37 minutes ago, brooklynwx99 said:

so the CFS basically shoves the eastern Canadian ridge into the Davis Strait, lol 

it's an unprecedented super Nino but I'll believe that when I see it

I think the model is overestimating the NAO: We aren't going to get extreme cold in the east with the N. Pacific looking like that. 

[Stormchaserchuck1]

3 hours ago, mitchnick said:

Cansips not too much different in the Conus really.

CPC disregarding it completely. The CANSIPS has looked like the 2 years in a row, and the last 2 Winters did verify like that. I remember in July last year though it had a 0.0 month lead in the Midwest of -4 to -6 for the month and it ended up being like +2 to +4 lol. 

[bncho]

@40/70 Benchmark@BlizzardWx @bluewave @Ephesians2

Thanks for the informative responses! Your points all make sense and continue to help me learn.

[GaWx]

On 8/9/2026 at 8:03 AM, GaWx said:

 Follow-up: Today’s SOI is -10, which makes the current streak at an amazing 91 days, 2nd only to 1998’s 100 days for longest on record back to 1991. I’m predicting that tomorrow’s will be a +SOI, which would mean the streak ends at 91 days.

As expected, today is a +SOI (+2), thus ending the -SOI streak at 91 days. Thus, the 100 day streak of 1998 is safe.

[40/70 Benchmark]

1 hour ago, Stormchaserchuck1 said:

I think the model is overestimating the NAO: We aren't going to get extreme cold in the east with the N. Pacific looking like that. 

Absolutely....you seem to be on the same page that I am. It's taking a valid element of the coming winter and just amplifying it. Like I said a few posts ago....valid pattern for an interval during the second half, but we aren't going to see that in the seasonal mean.