2024-2025 La Nina

[BlizzardWx]

One thing I've had on my mind lately is that it would do all of us well to get the bigger picture of "analogs" and not get bogged down in extraneous details. I think seasonal forecasting skill would increase if we could do more to tie the specific anomalous features of a year to the type and magnitude of forcing they will produce. As a hypothetical example, "If la nina focuses at 150W it would favor a standing ridge at x longitude. But because of the west Pacific Warm pool it may be shifted further west...etc" That at least ties real features to real background forcing, which IMO is all we can really do at a seasonal level. I think if we can consider each individual feature, like throwing pebbles in a pond and seeing where the waves go or intersect, we might get a better idea of how things will constructively or destructively interfere, which forcing will dominate if they are competing, and so on. In borderline snow climates all we can really do is try to figure out if the background state will be reasonably supportive for winter weather or not. 

 

And to be clear, I do see some of you doing what I am saying we should do. I think my broader point is that it's very unlikely you will ever have a truly good analog year, and so although analogs can be useful with context, I think we'd be better off looking at specific pieces of analog years instead of the years as a whole. Anyway, that's my opinion. If I am wrong, that's ok too

 

[Terpeast]

7 minutes ago, BlizzardWx said:

One thing I've had on my mind lately is that it would do all of us well to get the bigger picture of "analogs" and not get bogged down in extraneous details. I think seasonal forecasting skill would increase if we could do more to tie the specific anomalous features of a year to the type and magnitude of forcing they will produce. As a hypothetical example, "If la nina focuses at 150W it would favor a standing ridge at x latitude. But because of the west Pacific Warm pool it may be shifted further west...etc" That at least ties real features to real background forcing, which IMO is all we can really do at a seasonal level. I think if we can consider each individual feature, like throwing pebbles in a pond and seeing where the waves go or intersect, we might get a better idea of how things will constructively or destructively interfere, which forcing will dominate if they are competing, and so on. In borderline snow climates all we can really do is try to figure out if the background state will be reasonably supportive for winter weather or not. 

 

And to be clear, I do see some of you doing what I am saying we should do. I think my broader point is that it's very unlikely you will ever have a truly good analog year, and so although analogs can be useful with context, I think we'd be better off looking at specific pieces of analog years instead of the years as a whole. Anyway, that's my opinion. If I am wrong, that's ok too

 

Agreed. I’m actually experimenting with AI and machine learning to enable this sort of pattern recognition and use that to predict where the main features are going to be on a seasonal time scale.

[BlizzardWx]

2 minutes ago, Terpeast said:

Agreed. I’m actually experimenting with AI and machine learning to enable this sort of pattern recognition and use that to predict where the main features are going to be on a seasonal time scale.

This is excellent news! I really think we as a community will be able to produce significantly better outlooks in the future if we do a better job controlling the noise in analogs. I'm looking forward to seeing what you come up with. 

[GaWx]

1 hour ago, brooklynwx99 said:

2020-21 and 2010-11 are pretty good analogs... you don't have to think the winter will be as cold/snowy as those years for them to be viable

they both match up well with ONI, they both have a more hybrid/basin-wide orientation rather than a EP or Modoki event, the PDO was negative (strongly so in 2010-11), and the QBO is a good match as well. solar isn't a good match, but it's not an end-all-be-all for analogs. I feel like it gets overstated a bit, if anything

they have enough similarities to what we'll be seeing for them to be included. those years get evened out when you have winters like 1999-00 and 2022-23 being weighed heavily

Close Niña orientation and QBO notwithstanding:

1. Dec of 2010 barely missed having the strongest Dec -NAO since 1950 with only 1963 and 2009 slightly stronger. In contrast and though it of course is possible, there’s currently no indication that Dec of 2024 will have a -NAO (sub -0.25) at all. This shows the opposite, a very strong +NAO fwiw:

 

2. With the exact opposite SSN vs 2010, this strong +NAO prog for Dec is very believable.

[brooklynwx99]

22 minutes ago, GaWx said:

Close Niña orientation and QBO notwithstanding:

1. Dec of 2010 barely missed having the strongest Dec -NAO since 1950 with only 1963 and 2009 slightly stronger. In contrast and though it of course is possible, there’s currently no indication that Dec of 2024 will have a -NAO (sub -0.25) at all. This shows the opposite, a very strong +NAO fwiw:

 

2. With the exact opposite SSN vs 2010, this strong +NAO prog for Dec is very believable.

yes, I do not think that there will be -NAO blocking of that magnitude. out of the snowy winter scenarios, 2013-14 with poleward AK ridging makes more sense than 2010-11

[mitchnick]

2 hours ago, Terpeast said:

Notice it happened more frequently before the monster -pdo/-enso background state after 2015. Since that year, it only happened once. So i’m not sure if this pattern will hold this time. If it does, I think its random chance, like flipping a coin and getting tails 5x in a row and then getting heads on the 6th try. 

But if we DO get a cold February, I’m willing to eat my words and concede that you’re onto something. 

Cfs2 says grab the catchup and mustard. 

[Stormchaserchuck1]

It's not a coincidence that now that the ENSO subsurface has neutralized, the -PNA will take a hit for the first time in a while.. no big central-east Pacific ridges for 15 days according to 12z GEFS

The N. Pacific pattern has been following central-ENSO subsurface very closely since mid February

[George001]

5 hours ago, brooklynwx99 said:

2020-21 and 2010-11 are pretty good analogs... you don't have to think the winter will be as cold/snowy as those years for them to be viable

they both match up well with ONI, they both have a more hybrid/basin-wide orientation rather than a EP or Modoki event, the PDO was negative (strongly so in 2010-11), and the QBO is a good match as well. solar isn't a good match, but it's not an end-all-be-all for analogs. I feel like it gets overstated a bit, if anything

they have enough similarities to what we'll be seeing for them to be included. those years get evened out when you have winters like 1999-00 and 2022-23 being weighed heavily

Yep. The best way to do this is to go in with no expectations about the winter, and try to be as objective as possible. It’s really not easy to do so unfortunately. I’m not very good at it myself, it’s hard to not be biased when you heavily prefer a certain type of weather. Both 10-11 and 22-23 have value as analogs, but opposite winter outcomes. I’m not convinced the outcome of the winter matters when assessing analog validity. There is just too much randomness (luck) involved in individual outcomes to be able to properly evaluate whether a winter succeeded/failed due to just a couple lucky/unlucky breaks or the large scale pattern being a certain way. Also, some of the things being touted as a strong signal for a bad winter (+NAO) vs good winter (-NAO) are overstated especially for more northern posters. The nao r^2 correlation to temps for my area for example is -.2, which is very weak. Can’t just say it will be a +NAO or -NAO and draw all kinds of conclusions about what the winter will or will not look like based on that. 

[bluewave]

On 9/28/2024 at 12:10 PM, binbisso said:

@bluewave you were monitoring the mjo in october for clues about this coming winter. With October starting in phase 1 and ending in the COD mid month does this give you any hints regarding winter 

It’s based on the MJO RMM amplitude and VP anomalies in phase 5-6 during La Niña Octobers since 2010. The stronger October MJO responses had better winters in NYC relative to the other La Ninas in the multiyear event group. Those stronger MJO 5-6 Octobers also had the more robust peak La Niña ONIs from fall into winter of the multiyear groups.  All the La Niña winters around NYC since 11-12 have averaged warmer than normal. But even the warmer than average 17-18 and 20-21 winters had better snowfall than the other years in the multiyear group.

In addition, we haven’t had any cold neutral ENSOs in the -0.1 to -0.5 peak range with a strong -PDO since 2010. So not sure how the relationship would work for any year with the official CPC ONI not getting colder than -0.5. The other thing to consider is that we don’t know how long this relationship will continue to work it.

So I will update this post with some early ideas for the coming winter after we see how the MJO verifies for the month of October. 

http://www.bom.gov.au/clim_data/IDCKGEM000/rmm.74toRealtime.txt

https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php

 

Multiyear La Niña groupings below with peak MJO 5-6 strength and the peak ONI from the fall into winter…….Strongest October MJO 5-6 and La Niña ONI leading to best NYC winters of the multiyear group bolded…

 

October 2022…Phase 6 peak +2.177…..ONI….-1.0

October 2021….Phase 5 peak +2.087….ONI….-1.0

October 2020…Phase 5 peak +2.818….ONI…..-1.3

 

October 2017….Phase 6 peak +3.353….ONI….-1.0

October 2016….Phase 5 peak +1.800….ONI….-0.7

 

October 2011….Phase 6 peak +1.674….ONI….-1.1

October 2010….Phase 5 peak +2.880…ONI…..-1.6 
 

[40/70 Benchmark]

6 hours ago, mitchnick said:

Cfs2 says grab the catchup and mustard. 

That is huge -WPO.

[40/70 Benchmark]

Raindance dropping a hot one on the 2022 analog.

"You'll see a lot of people claiming 2022 as an analog for the winter. But the models continue to show a very warm pattern for the Southwest, with cold dumping into the Plains. In 2022-23, cold alternated between dumping into the West & Plains. That's not supported on the models."

[raindancewx]

WPO was extremely positive in 2022-23. Looks very negative on the Canadian. Models are garbage at this range. But the WPO was very positive in 2022 in October. Doesn't look like it will be very positive this month. Almost impossible to get record heat in CA/NV in Fall or Spring when the WPO is positive.

[40/70 Benchmark]

I don't think it will be as bad as 2022 for the east, either, but I do still think it has utility as an analog.

[GaWx]

On 9/3/2024 at 11:57 AM, GaWx said:

 Fwiw, this CANSIPS map for Aug of 2025 implies a developing Modoki El Niño in time for ‘25-26. Opinions? @PhiEaglesfan712

@snowman19

 For the 2nd month in a row the CANSIPS fwiw has a developing Modoki El Niño (centered near 150-155W in Sept and drifting further W) next summer with it peaking in Sept 2025 or later. It is a little stronger on this run vs last month’s with my rough estimate of +1.0 in Sep in Nino 3.4. On a relative basis, it would probably be more like +0.7. Just food for thought for the winter after next and a reason I’ve been more excited about 2025-6 than 2024-5 for my area’s potential. Now if only the core of the WPAC marine heatwave could migrate E (one can always dream):

[bluewave]

8 hours ago, 40/70 Benchmark said:

I don't think it will be as bad as 2022 for the east, either, but I do still think it has utility as an analog.

The 22-23 winter was so warm in the East due to the Aleutian Ridge axis out near the Dateline coupled with the deep trough in the West. So this allowed the Greenland block based -AO to link up with the Southeast Ridge. While the WPO was positive, it didn’t have as much influence over the pattern for the East as other teleconnections. The 16-17 La Niña was very warm around NYC and the WPO was negative. 49-50 was the warmest La Niña winters of the older era and the WPO was negative also. So in the most general sense the 22-23 winter matched the winter pattern since 15-16 of an Aleutian Ridge and Southeast Ridge pattern. But it had an unusual combination of features not seen before in the same winter. 

I think all this talk of winters following exact past analogs is a bit of a misnomer. Since every new winter has combinations of features not seen before. This is where natural variability come in. So the current pattern since the summer has had some similarities to 2022. But to take that and say every feature going forward will be a carbon copy with no natural variability mixed in is misplaced. Plus exact magnitudes of winter warmth usually aren’t known until we at least see how the early winter verifies. Remember, none of the long range models as of November 2015 were forecasting a +13 December. The really big departures aren’t usually made known until we get into the short to medium range. 

22-23 La Niña +WPO winter warm in the East

16-17 -WPO La Niña winter warm in the East

49-50 -WPO La Nina winter warm in the East

 

 

 

[snowman19]

48 minutes ago, bluewave said:

The 22-23 winter was so warm in the East due to the Aleutian Ridge axis out near the Dateline coupled with the deep trough in the West. So this allowed the Greenland block based -AO to link up with the Southeast Ridge. While the WPO was positive, it didn’t have as much influence over the pattern for the East as other teleconnections. The 16-17 La Niña was very warm around NYC and the WPO was negative. 49-50 was the warmest La Niña winters of the older era and the WPO was negative also. So in the most general sense the 22-23 winter matched the winter pattern since 15-16 of an Aleutian Ridge and Southeast Ridge pattern. But it had an unusual combination of features not seen before in the same winter. 

I think all this talk of winters following exact past analogs is a bit of a misnomer. Since every new winter has combinations of features not seen before. This is where natural variability come in. So the current pattern since the summer has had some similarities to 2022. But to take that and say every feature going forward will be a carbon copy with no natural variability mixed in is misplaced. Plus exact magnitudes of winter warmth usually aren’t known until we at least see how the early winter verifies. Remember, none of the long range models as of November 2015 were forecasting a +13 December. The really big departures aren’t usually made known until we get into the short to medium range. 

22-23 La Niña +WPO winter warm in the East

16-17 -WPO La Niña winter warm in the East

49-50 -WPO La Nina winter warm in the East

 

 

 

The problem I see with 13-14 and I know you have discussed it several times is the PDO configuration (basically +PDO). IMO, the North PAC SSTs have to change big time and very quickly for that to become a viable analog. That PDO configuration lead to the -EPO/-WPO and at times very strongly +PNA. The AO/NAO was wall to wall strongly positive; all PAC/+TNH driven. The MJO forcing was also very different, it was also a cold-neutral/La Nada

[40/70 Benchmark]

1 hour ago, bluewave said:

The 22-23 winter was so warm in the East due to the Aleutian Ridge axis out near the Dateline coupled with the deep trough in the West. So this allowed the Greenland block based -AO to link up with the Southeast Ridge. While the WPO was positive, it didn’t have as much influence over the pattern for the East as other teleconnections. The 16-17 La Niña was very warm around NYC and the WPO was negative. 49-50 was the warmest La Niña winters of the older era and the WPO was negative also. So in the most general sense the 22-23 winter matched the winter pattern since 15-16 of an Aleutian Ridge and Southeast Ridge pattern. But it had an unusual combination of features not seen before in the same winter. 

I think all this talk of winters following exact past analogs is a bit of a misnomer. Since every new winter has combinations of features not seen before. This is where natural variability come in. So the current pattern since the summer has had some similarities to 2022. But to take that and say every feature going forward will be a carbon copy with no natural variability mixed in is misplaced. Plus exact magnitudes of winter warmth usually aren’t known until we at least see how the early winter verifies. Remember, none of the long range models as of November 2015 were forecasting a +13 December. The really big departures aren’t usually made known until we get into the short to medium range. 

22-23 La Niña +WPO winter warm in the East

16-17 -WPO La Niña winter warm in the East

49-50 -WPO La Nina winter warm in the East

 

 

 

I think you are confusing my point. I am not claiming that the WPO is the silver bullet that will save winter in the east, but it is one factor. I have told you before that I do not expect the same magnitude of amplification with the trough out west as 2022, either...which probably ties into the WPO to an extent. If you want to bet on a reoccurence of that, then be my guest.

2022 with slighly less amplification out west would have been a good winter for my area...no doubt.

[40/70 Benchmark]

You are never going to see +13 magnitude anomalies in seasonal guidance...I think we all understand that.

And my my money is on above normal in a lot of the east, but just not so extreme...all I am saying.

[40/70 Benchmark]

5 hours ago, GaWx said:

 For the 2nd month in a row the CANSIPS fwiw has a developing Modoki El Niño (centered near 150-155W in Sept and drifting further W) next summer with it peaking in Sept 2025 or later. It is a little stronger on this run vs last month’s with my rough estimate of +1.0 in Sep in Nino 3.4. On a relative basis, it would probably be more like +0.7. Just food for thought for the winter after next and a reason I’ve been more excited about 2025-6 than 2024-5 for my area’s potential. Now if only the core of the WPAC marine heatwave could migrate E (one can always dream):

Yea, textbook.

[40/70 Benchmark]

Man, PSL not updating those temp and precip composites is killling me. Hopefully that is fixed this month before I go to work on this winter.

[bluewave]

21 minutes ago, 40/70 Benchmark said:

I think you are confusing my point. I am not claiming that the WPO is the silver bullet that will save winter in the east, but it is one factor. I have told you before that I do not expect the same magnitude of amplification with the trough out west as 2022, either...which probably ties into the WPO to an extent. If you want to bet on a reoccurence of that, then be my guest.

2022 with slighly less amplification out west would have been a good winter for my area...no doubt.

I don’t think anyone is betting early on that  this winter will rival the magnitude of the trough digging into the SW. That was the strongest trough out there of the last decade. But less trough in the SW doesn’t equal a great winter here absent other changes to the pattern. 

[40/70 Benchmark]

27 minutes ago, 40/70 Benchmark said:

You are never going to see +13 magnitude anomalies in seasonal guidance...I think we all understand that.

And my my money is on above normal in a lot of the east, but just not so extreme...all I am saying.

 

4 minutes ago, bluewave said:

I don’t think anyone is betting early on that  this winter will rival the magnitude of the trough digging into the SW. That was the strongest trough out there of the last decade. But less trough in the SW doesn’t equal a great winter here absent other changes to the pattern. 

Great ("here" as in NYC) is your word; not mine. But I do standby my assertion that that could be enough for a decent winter for a much of New England, at least.

[PhiEaglesfan712]

6 hours ago, GaWx said:

 For the 2nd month in a row the CANSIPS fwiw has a developing Modoki El Niño (centered near 150-155W in Sept and drifting further W) next summer with it peaking in Sept 2025 or later. It is a little stronger on this run vs last month’s with my rough estimate of +1.0 in Sep in Nino 3.4. On a relative basis, it would probably be more like +0.7. Just food for thought for the winter after next and a reason I’ve been more excited about 2025-6 than 2024-5 for my area’s potential. Now if only the core of the WPAC marine heatwave could migrate E (one can always dream):

Sorry, I can't see this happening. The only way this happens is if we have anywhere near a 13-14 solution this winter. For that to happen, as @snowman19pointed out, the PDO configuration and the North Pacific SSTs have to change very quickly for this scenario to even be viable. Even then, there's no guarantee even 2026-27 will be an el nino (just look what happened in 14-15, the el nino didn't really take off until the spring).

We're still years away from the next el nino.

[michsnowfreak]

10 minutes ago, 40/70 Benchmark said:

 

Great ("here" as in NYC) is your word; not mine. But I do standby my assertion that that could be enough for a decent winter for a much of New England, at least.

I'm seeing a lot of warm bias in this thread. Not calling anyone out or anything. But even with the acknowledgment that no two winters are identical, thus no analog is perfect, I'm seeing some analogs that have produced good winters immediately get dismissed by some. And make no mistake, the internet is filled with plenty of cold bias too. Just my observation, and I feel you, @raindancewx and @GaWx are very balanced. 

You brought up how seasonal guidance will never show some insane anomalies, which is so true. But on the other hand, we know that CPC will never go cold in a seasonal forecasts (they will in a weekly or monthly if the signal is strong). Go look back at some of their forecasts for cold winters. It's hilarious. 

[mitchnick]

Cansips pretty much held serve from last month's forecast, and looks a lot like the Cfs2 too tempwise.

[snowman19]

4 minutes ago, mitchnick said:

Cansips pretty much held serve from last month's forecast, and looks a lot like the Cfs2 too tempwise.

Big Aleutian ridge/high signal 

[GaWx]

14 minutes ago, michsnowfreak said:

I'm seeing a lot of warm bias in this thread. Not calling anyone out or anything. But even with the acknowledgment that no two winters are identical, thus no analog is perfect, I'm seeing some analogs that have produced good winters immediately get dismissed by some. And make no mistake, the internet is filled with plenty of cold bias too. Just my observation, and I feel you, @raindancewx and @GaWx are very balanced. 

You brought up how seasonal guidance will never show some insane anomalies, which is so true. But on the other hand, we know that CPC will never go cold in a seasonal forecasts (they will in a weekly or monthly if the signal is strong). Go look back at some of their forecasts for cold winters. It's hilarious. 

 Keep in mind that you’re in a location more much less apt (if apt at all) to be warmer than normal in La Niña. Many of us are concentrating more on closer to the E coast. In my case, I naturally also like to talk about my area, which is more highly impacted to the mild side by a dominant SE ridge. I’ve yet to see a seasonal model for this winter not calling for a mild winter and warmer winter than 23-24 in the SE US overall, which is what I’d expect with a strong -PDO and Niña. That’s the main reason I keep looking ahead to 25-6 for a potentially much colder winter down here with the CANSIPS again literally showing a Modoki El Niño.

[bluewave]

1 hour ago, snowman19 said:

The problem I see with 13-14 and I know you have discussed it several times is the PDO configuration (basically +PDO). IMO, the North PAC SSTs have to change big time and very quickly for that to become a viable analog. That PDO configuration lead to the -EPO/-WPO and at times very strongly +PNA. The AO/NAO was wall to wall strongly positive; all PAC/+TNH driven. The MJO forcing was also very different, it was also a cold-neutral/La Nada

The 13-14 winter was one of what I call the big 5 of modern times. The other winters were  95-96, 09-10, 10-11 and 14-15. 95-96 still stands as the greatest winter in NYC Metro for wall to wall cold and snow from November into April. 09-10 was the greatest winter for snowfall from DC to Philly. 10-11 featured the greatest 33 day snowfall around NYC Metro at just over 60” from late December into late January. 13-14 was one of the best for snow combined with cold in the Great Lakes. 14-15 was tops for Boston snowfall. 

The 95-96 winter was before the big global temperature jump in 97-98. So NYC Metro was still able to register their snowiest winter with parts of Long Island recording over 90”. The extended cold from November into April hasn’t been rivaled either as the world has warmed. 

The next big global temperature rise to a new higher baseline occurred  in 15-16 and no winters since then have been able to rival 09-10 in the mid-Atlantic. Same goes for 10-11 around NYC and 13-14 for the Great Lakes and 14-15 near Boston.

Now we have seen one of the greatest global temperature rises over the last year or so. These big temperature rises make it unlikely that we could see one of these big 5 repeat again in this much warmer climate. So we can probably retire those 5 years as analogs for the new warmer climate. 

[40/70 Benchmark]

19 minutes ago, michsnowfreak said:

I'm seeing a lot of warm bias in this thread. Not calling anyone out or anything. But even with the acknowledgment that no two winters are identical, thus no analog is perfect, I'm seeing some analogs that have produced good winters immediately get dismissed by some. And make no mistake, the internet is filled with plenty of cold bias too. Just my observation, and I feel you, @raindancewx and @GaWx are very balanced. 

You brought up how seasonal guidance will never show some insane anomalies, which is so true. But on the other hand, we know that CPC will never go cold in a seasonal forecasts (they will in a weekly or monthly if the signal is strong). Go look back at some of their forecasts for cold winters. It's hilarious. 

I appreciate that because my work has definitely had a cold/snow bias, so I do make an effort to demonstrate that it isn't a conscious initiative or anything. While it is fair to say that I have some work to do in that regard, I am willing to bet that 99% of forecasters out there have been too cold and snowy in the east over the past several years. Even raindance, as exceptional as he has been and clearly the best in my mind, went very snowy in Maine for I think 2020-2021 and they ended up porked.

[jm1220]

6 minutes ago, GaWx said:

 Keep in mind that you’re in a location more much less apt (if apt at all) to be warmer than normal in La Niña. Many of us are concentrating more on closer to the E coast. In my case, I naturally also like to talk about my area, which is more highly impacted to the mild side by a dominant SE ridge. I’ve yet to see a seasonal model for this winter not calling for a mild winter and warmer winter than last 23-24 in the SE US overall, which is what I’d expect with a strong -PDO and Niña. That’s the main reason I keep looking ahead to 25-6 for a potentially much colder winter down here with the CANSIPS again literally showing a Modoki El Niño.

I’m pessimistic myself for anyone south of I-90/I-84 until that boiling SST blob east of Japan goes away and we stop getting the constant MJO 4-6 feedback loop. That will keep the same general conditions going. We had a strong Nino and about all that did was add a lot more moisture to the Nina general pattern. One of the constant storms managed to be snow IMBY in Feb, otherwise it was a near or record wet winter. Otherwise the Pacific jet kept raging and everyone was warm not just the East. We need some mechanism to encourage +PNA in the right location again and mute the SE Ridge. Maybe the long term AMO declining can help on the Atlantic side. But it also doesn’t do any good to have Greenland blocks that link up to an insanely amped SE ridge. Hopefully 22-23 was an outlier anomaly but the same general crap pattern remains that favors the West/N Plains/Lakes/N New England. Hopefully it’s something that can reverse soon, doesn’t at all look like it’ll be this upcoming winter.