2025-2026 ENSO

[Daniel Boone]

1 hour ago, roardog said:

I’m pretty sure the model was forecasting all of that below average snow because it had a country wide blowtorch. So it was right for the wrong reason pretty much. 

Exactly. We averaged normal here. Model had us projected well below snowfall.

[GaWx]

 Here are the Sept Euro forecasts for DJF 2024-5 back to 2017-8 vs actual

2024-5 prog: avg 4F too warm much of US but SW close; grade: F

2024-5 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2024&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot
 

2023-4 prog: MW 5-8 too cold, NE 3-4 too cold, SE 1-3 too cold, W close; grade D

 

2023-4 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2023&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

2022-3 prog: W MW 3 too warm; E MW and NE 3 too cold, SE 1-2 too cold, W 6 too warm; grade D

2022-3 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2022&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

2021-2 prog: pretty close except 2 too warm W; grade A

 

2021-2 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2021&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

 

 2020-1 prog: 0-4 too warm MW, 3 too warm NE/SE, SC 4-6 too warm, W 1-2 too warm; Grade D

2020-1 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2020&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

 

2019-20 prog: MW 2 too cold, NE close; SE 2-3 too cold, W 0-2 too cold; grade A

2019-20 actual:

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2019&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

 

2018-9 prog: MW 1 too warm, NE close, SE 2-3 too cold, W 2-3 too warm; grade B

2018-9 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2018&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

 

2017-8 prog: MW 3 too warm, NE 2 too warm, SE/W close; Grade C

 

2017-8 actual

https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2017&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot

———————

-GPA: 2.0

-Last 3 poor

-Best 2019-20 and 2021-2; worst 2024-5

-2 A, 1 B, 1 C, 3 D, 1 F

-The Sept ‘25 fcast for 2025-6 is for US as a whole the 2nd warmest to 2024-5 since 2017-8

-NE prog misses: +4, -3.5, -3, 0, +3, 0,  0, +2; so, NE prog was too warm thrice, too cold twice, and close thrice; So, NE prog on avg was only 0.3 F too warm vs actual meaning negligible bias vs actual though avg miss was by ~2 F

-MW prog misses: +4, -6.5, 0, 0, +2, -2, +1, +3; so, MW prog was too warm 4 times, too cold twice, and close twice; So, MW prog on avg was only 0.2 F too warm meaning negligible bias vs actual though avg miss was by ~2 F

-SE prog misses: +4, -2, -1.5, 0, +3, -2.5, -2.5, 0; so, SE prog was too warm twice, too cold 4 times, and close twice; So, SE prog on avg was only 0.2 F too cool vs actual meaning negligible bias vs actual though avg miss was by ~2 F

-W prog misses: +2, 0, +6, +2, +1.5, -1, +2.5, 0; so, W prog was too warm 5 times, too cold once, and close twice; So, W prog on avg was 1.6 F too warm! Thus, there appears to be a notable Euro warm bias for the W for winter at least in Sept progs. Avg miss was ~2 F

[raindancewx]

Some context on La Nina ACE through August. We finished at 39 in the Atlantic. 

30-50 La Nina ACE, through Aug 31 since 1930 () and then final ACE after the slash.

1942 (36.1) / 62.5

1956 (32.5) / 56.7

1964 (39.7) / 153

1998 (44.7) / 181.8

2000 (43.7) / 119.1

2007 (38.0) / 73.9

2010 (45.1) / 165.5

2011 (36.7) / 126.3

2016 (31.5) / 141.3 excluding the Jan 2016 system

2017 (30.4) / 224.9

2020 (43.0) / 179.8

2021 (44.4) / 145.7

Blended final total - 135.9. Ten most recent La Ninas average 146 ACE (2007-2024). October is almost always below 70 ACE even in hyper-active seasons, and Nov is almost always under 35 ACE even in hyper-active seasons. So the quiet September days start to add up pretty quick if you're behind with nothing active. If you throw out 2017 (since we're very unlikely to be anywhere near 175 ACE in Sept), you end up with a map consistent with my research showing the low ACE La Nina hurricane seasons have more frequent cold snaps in the Southwest.

2020 had the mid Sept/late Oct/Feb severe cold, Feb-Apr 2022 (and 2023) was very cold locally. Feb 2011 very cold. Jan 2008 is very cold. Nov-Jan 2000-01 is very cold. Dec 2011 very cold locally. 

September should be quite clarifying for which years in the group are the good matches. The Summer temperature pattern was fairly close to some kind of blend of 2007, 2020, 2021

[snowman19]

There is growing evidence that we may see a big shakeup and complete pattern flip by October to much warmer than normal and possibly much drier than normal as well. @Stormchaserchuck1 warned of this weeks ago….-PDO/-ENSO October correlation. The -IOD is magnifying those effects
 

 

 

 

 

 

 

 

 

[Stormchaserchuck1]

snowman19 has been pretty accurate with his predictions. Don't underestimate him. 

[bluewave]

11 hours ago, roardog said:

I’m pretty sure the model was forecasting all of that below average snow because it had a country wide blowtorch. So it was right for the wrong reason pretty much. 

But the storm track matched the warm winter pattern it was forecasting. The composites below are for the 20 days that .20+ of precipitation fell from Philly to Boston. The average temperature at Philly to NYC on these days was in the low 40s. So it was too warm for much snow with the storm track to the west of the I-95 corridor.

The Euro seasonal precipitation forecast had above average near the Eastern Great Lakes matching the warmer storm track which verified. Plus there was major flooding in areas like West Virginia along this storm track. 

The Southeast ridge on the days when the wettest storms occurred was even more amplified than what the general seasonal forecast from the Euro was. Plus the warm departure on the heaviest storm days was significantly warmer than the seasonal temperature forecast.

So in effect a warmer version of the seasonal forecast did occur through the storm track. The Pacific Jet was too strong and warm for much snowfall from Philly to Boston relative to the means. This has been a common theme leading to the record low snowfall near I-95 since 2018-2019.

 

 

[snowman19]

This is a good point….Given the background climate change/AGW, have we entered the “new” -AMO cycle, which would explain the anemic Atlantic hurricane season? In the past, before CC really kicked in, cold North Atlantic SSTs (like we have now) was a staple of -AMO cycles. Just food for thought

[SnoSki14]

3 hours ago, Stormchaserchuck1 said:

snowman19 has been pretty accurate with his predictions. Don't underestimate him. 

Not exactly groundbreaking to predict AN temps and BN snows in winter in our new CC era climate. 

That's been the general theme since 2015. 

[SnoSki14]

17 hours ago, michsnowfreak said:

Pretty big fail last winter.

Because the usual seasonal forecasts that relied on past tools have become much more varied in our new CC era. 

There's too many variables that are interfering with one another and are producing outcomes we've never seen before. 

[mitchnick]

3 hours ago, snowman19 said:

There is growing evidence that we may see a big shakeup and complete pattern flip by October to much warmer than normal and possibly much drier than normal as well. @Stormchaserchuck1 warned of this weeks ago….-PDO/-ENSO October correlation. The -IOD is magnifying those effects
 

 

 

 

 

 

 

 

 

One thing to keep in mind. Not that anyone should discount the Eps, but the AI EPS has done much better with the cooler air in the last month+ as I mentioned a few pages back and that BAMWX also noted.

Anyone with a Wxbell or Weathermodels subscription, do they offer long range AI EPS? Tia

[snowman19]

4 minutes ago, mitchnick said:

One thing to keep in mind. Not that anyone should discount the Eps, but the AI EPS has done much better with the cooler air in the last month+ as I mentioned a few pages back and that BAMWX also noted.

Anyone with a Wxbell or Weathermodels subscription, do they offer long range AI EPS? Tia

There is very good support this time for a big shakeup in the pattern by October. Projected tropical forcing/MJO, -ENSO, -PDO, -IOD. I am becoming increasingly confident that October is very likely to be warmer to much warmer than normal….also potentially much drier than normal as well

[PhiEaglesfan712]

27 minutes ago, snowman19 said:

There is very good support this time for a big shakeup in the pattern by October. Projected tropical forcing/MJO, -ENSO, -PDO, -IOD. I am becoming increasingly confident that October is very likely to be warmer to much warmer than normal….also potentially much drier than normal as well

Octobers have been almost always warmer than average since 2010. Funny enough, the only 2 that were below average were 2011 and 2022, and both of those winters were snowless blowtorches.

[mitchnick]

40 minutes ago, snowman19 said:

There is very good support this time for a big shakeup in the pattern by October. Projected tropical forcing/MJO, -ENSO, -PDO, -IOD. I am becoming increasingly confident that October is very likely to be warmer to much warmer than normal….also potentially much drier than normal as well

That may the case, but I'm curious what the AI EPS shows since it's done better in the medium range than the Eps. That's all.

[chris21]

16 hours ago, bluewave said:

The much below average snowfall forecast from the seasonal models run during the fall of 2024 from Philly to Boston did work out. They even got the very low snowfall correct for Chicago.

That’s what most on here follow anyway.

A continuation of the warm storm tracks through the Great Lakes theme. 
 

Huge fail here snowfall wise in the southern apps. Had about average snow here, 43 inches.

[40/70 Benchmark]

20 hours ago, MJO812 said:

We are only in September.  Winter outlooks by the models and people for the past few years haven't been good. 

Grain of salt this far out

I thought mine from last year was okay...not great, but not bad.

[40/70 Benchmark]

17 hours ago, donsutherland1 said:

It was. Notice how similar the map looks to last September's outlook for the past winter. I suspect that the similarity reveals that the expected ENSO state plays a disproportionate role in the seasonal forecasts.

Yes, this is what I have always said about climate models...they are biased towards stock ENSO.

[mitchnick]

52 minutes ago, snowman19 said:

There is very good support this time for a big shakeup in the pattern by October. Projected tropical forcing/MJO, -ENSO, -PDO, -IOD. I am becoming increasingly confident that October is very likely to be warmer to much warmer than normal….also potentially much drier than normal as well

Since you mentioned MJO, you made me look! Lol

Looks like all modeling has recently (day?) shifted back away from 4-5-6.

https://www.stormsurf.com/page2/links/mjo_phase.html

[40/70 Benchmark]

16 hours ago, bluewave said:

The much below average snowfall forecast from the seasonal models run during the fall of 2024 from Philly to Boston did work out. They even got the very low snowfall correct for Chicago.

That’s what most on here follow anyway.

A continuation of the warm storm tracks through the Great Lakes theme. 
 

Did NOT get the pattern right, though.....suprised someone as meticulous as you would omit that.

[40/70 Benchmark]

4 hours ago, bluewave said:

But the storm track matched the warm winter pattern it was forecasting. The composites below are for the 20 days that .20+ of precipitation fell from Philly to Boston. The average temperature at Philly to NYC on these days was in the low 40s. So it was too warm for much snow with the storm track to the west of the I-95 corridor.

The Euro seasonal precipitation forecast had above average near the Eastern Great Lakes matching the warmer storm track which verified. Plus there was major flooding in areas like West Virginia along this storm track. 

The Southeast ridge on the days when the wettest storms occurred was even more amplified than what the general seasonal forecast from the Euro was. Plus the warm departure on the heaviest storm days was significantly warmer than the seasonal temperature forecast.

So in effect a warmer version of the seasonal forecast did occur through the storm track. The Pacific Jet was too strong and warm for much snowfall from Philly to Boston relative to the means. This has been a common theme leading to the record low snowfall near I-95 since 2018-2019.

 

 

It still got the pattern wrong, though...which is why it was too warm. I wouldn't consider it a win for those that forecasted a big winter for the mid atlantic during the 2015-2016 season, despite the fact that snowfall worked out. I'm not arguing your point about storm track, but the EURO seasonal was still a tremendously flawed forecast.  I am willing to bet if we had less of a neg NAO last season, the northeast would have done better via more overrunning and SWFE snowfall, kind of like what we saw in February. I understand your point about increased reliance on Benchmark tracks in the NE, but some of that is due to SE Canada being warmer relative to other areas independent of CC. We saw that shift some last year and if se Canada is appreciably cold again, which I suspect that it will be, I would bet on more snowfall.

[anthonymm]

4 hours ago, snowman19 said:

There is growing evidence that we may see a big shakeup and complete pattern flip by October to much warmer than normal and possibly much drier than normal as well. @Stormchaserchuck1 warned of this weeks ago….-PDO/-ENSO October correlation. The -IOD is magnifying those effects
 

 

 

 

 

 

 

 

 

ECM is now incapable of seeing below normal temps. It blowtorches everything >30 days out by default now. Garbage warm biased model

[GaWx]

11 minutes ago, anthonymm said:

ECM is now incapable of seeing below normal temps. It blowtorches everything >30 days out by default now. Garbage warm biased model

 Regarding all 9 Euro DJF fcasts issued in Sept since 2017-8 (as far back as can be seen at the Euro site), there wasn’t even a tiny speck of BN anywhere in the lower 48 E of 110W long. The only one that had any BN anywhere in the lower 48 was the 2019-20 fcast, which had a small area of ~-1F in a small part of the Mtn W (which amazingly enough ended up being the only area with BN):

  
 Despite that, there were notable areas of actual BN in portions of the lower 48 in all of these DJFs except 2019-20 and 2023-4. So, your statement ECM is now incapable of seeing below normal temps has essentially been true for its Sept fcasts since 2017-8.

 However, two of its 8 Oct fcasts and one of its Nov fcasts for DJF did have some BN:

 

Oct ‘23, which failed as there was no BN in the lower 48:

Oct ‘18, which got the BN in KS right but the rest of this BN area ended up NN or AN and also it missed the widespread BN in NW 1/2 of the US:

Nov ‘17 in UP of Mich, which was correct, but it missed the widespread BN in the Midwest, Plains, and N Rockies:

[michsnowfreak]

18 hours ago, bluewave said:

The much below average snowfall forecast from the seasonal models run during the fall of 2024 from Philly to Boston did work out. They even got the very low snowfall correct for Chicago.

That’s what most on here follow anyway.

A continuation of the warm storm tracks through the Great Lakes theme. 
 

If most here follow model longterm seasonal snowfall forecasts over anything else, we have a problem . A longterm snowfall model forecast is probably the least accurate thing out there. Temp/precipitation bad enough....but snowfall? Model snow maps are to be taken with a grain of salt 48 hours out, much less 7 months out.

[michsnowfreak]

17 hours ago, roardog said:

I’m pretty sure the model was forecasting all of that below average snow because it had a country wide blowtorch. So it was right for the wrong reason pretty much. 

1000%. I can't speak for the east coast, but here in Michigan it absolutely would have snowed more synoptically if it wasnt as cold and dry as it ended up. It snowed nearly every day for long stretches much of winter, yet we finished over a foot below normal. That is as anti "new climate" as you can get. 

[michsnowfreak]

2 hours ago, 40/70 Benchmark said:

Yes, this is what I have always said about climate models...they are biased towards stock ENSO.

They are very biased towards enso and also biased towards a warmer climate. That is why periods of cold are never seen far out, you will see them grow colder as the timeframe nears. 

[michsnowfreak]

3 hours ago, snowman19 said:

There is very good support this time for a big shakeup in the pattern by October. Projected tropical forcing/MJO, -ENSO, -PDO, -IOD. I am becoming increasingly confident that October is very likely to be warmer to much warmer than normal….also potentially much drier than normal as well

Not sure about MUCH warmer than normal, but overall warmer than normal wouldn't surprise me at all. It actually fits perfectly with your expected nina. A mild Fall with a sharp turn colder in late November or early December is classic nina.

The caveat of course is that "warmer than normal" comes with rapidly cooling averages. Not sure how far south this applies, but for here at least, even a warm October is far from hot. A few days of low humidity 80s and a few frosts normally happen in warm Octobers.