El Nino 2023-2024

[GaWx]

3 hours ago, GaWx said:

0Z 12/24 (orange) barely makes major SSW status with it a hair under 0 m/s at the end and still dropping. The 0Z CDN ens mean (blue) is at 2 days in a row at a major. But the 0Z GEFS mean never gets below +17.

 Following up on this, the GEPS mean had a major 14 days out yesterday and 13 days out today vs having to wait til 10 days out in advance of the 2/16/23 major. So, its signal is much stronger than then. OTOH, the GEFS mean in advance of 2/16/23 was ~+10 or less from 15 days out before also going major 10 days out. But with it still up at +17, it is a much weaker signal than that of GEPS vs having a stronger signal than GEPS 15-14 days in advance of 2/16/23. 
 
 So, there are mixed signals between the GEFS (less bullish) and GEPS (more bullish) vs last year. Based on my memory, the EPS is stronger than it was then. So, two of the three ensemble means are more bullish currently vs 2/16/23 and one is less bullish. I expect the GEFS to get more bullish on the coming days.

[40/70 Benchmark]

8 hours ago, jbenedet said:

def not.
 

Point is, balance of guidance has the core of the PV towards the mountain west. 

That latest run of the GEFS looks best though for our region. But some caution on that—it’s a bit misleading— you can still see height falls greatest over the SW on that run.

Yes, fair enough. I don't care to have the heart of the cold personally...not to distract from the fact that these SSW events are not fail safe. There is risk.

 

[40/70 Benchmark]

6 hours ago, qg_omega said:

Yes, would take

Well, puff, puff pass on that...I'd rather risk precip type issues.

[40/70 Benchmark]

10 hours ago, raindancewx said:

I'll just leave this here.

Boston, <=1.0" Snow by Dec 31, El Nino
-------------------
1941-42 - 23.9 (Oct-May snow)
1953-54 - 23.6
1957-58 - 44.7
2006-07 - 17.1
2015-16 - 36.1
2018-19 - 27.4
--------------------
Mean: 28.8" (Oct-May Snow)

NYC (Central Park), <=1.0" Snow by Dec 31, El Nino
-------------------
1941-42 - 11.3 (Oct-May snow)
1965-66 - 21.4
1972-73 -  2.8 
1977-78 - 50.7
1986-87 - 23.1
1991-92 - 12.6
1994-95 - 11.8
1997-98 -  5.5
2006-07 - 12.4
2015-16 - 32.8
-------------------
Mean: 18.4" (Oct-May Snow)

Philly, <=1.0" Snow by Dec 31, El Nino
-------------------
1941-42 -  5.1 (Oct-May snow)
1958-59 -  5.1
1965-66 - 27.4
1972-73 -  0.0
1977-78 - 54.9 
1986-87 - 25.7
1991-92 -  4.7
1994-95 -  9.8
1997-98 -  0.8
2004-05 - 30.4
2006-07 - 13.4
2014-15 - 27.0
2015-16 - 27.5
2019-20 -  0.3
-------------------
Mean: 16.6" (Oct-May Snow)

ABQ, 0.0-2.0" Snow by Dec 31, El Nino (1.0" so far)
-------------------
1939-40 -  4.0 (Oct-May snow)
1963-64 - 10.0
1969-70 -  7.1
1977-78 - 11.5
1994-95 -  9.4
2002-03 -  2.8
2009-10 -  4.2
2014-15 - 15.2
------------------
Mean: 8.0" (Oct-May Snow)

No snow on the GFS or Euro for the Northeast cities through 12/31.

This is a great list because you can see the strong dichotomy with respect to the polar domain...the years that made a good come back looked alot like this year. The average is pretty useless IMO....its like taking the ensemble mean when half of the members phase an event and half don't. 

[griteater]

48 minutes ago, 40/70 Benchmark said:

This is a great list because you can see the strong dichotomy with respect to the polar domain...the years that made a good come back looked alot like this year. The average is pretty useless IMO....its like taking the ensemble mean when half of the members phase an event and half don't. 

Or like using the SREF model

[brooklynwx99]

45 minutes ago, 40/70 Benchmark said:

This is a great list because you can see the strong dichotomy with respect to the polar domain...the years that made a good come back looked alot like this year. The average is pretty useless IMO....its like taking the ensemble mean when half of the members phase an event and half don't. 

i’m liking 1957-58 more and more… had a very warm December, especially in the northern Plains. 500mb pattern also putrid. had a SSW and a very blocky Feb as a result. this Nino also started off very east based and migrated west similar to this one 

[40/70 Benchmark]

25 minutes ago, brooklynwx99 said:

i’m liking 1957-58 more and more… had a very warm December, especially in the northern Plains. 500mb pattern also putrid. had a SSW and a very blocky Feb as a result. this Nino also started off very east based and migrated west similar to this one 

Yea, the only reason it wasn't higher on my list is the PDO. I also doubt the NAO finishes THAT negative....but it's great from an ENSO standpoint.

[raindancewx]

57-58 had the QBO in the opposite phase of this year (+7 in Dec on the dataset), along with the PDO opposite. It's been fine as a match for December, but you can get lots of matches to this type of look in an El Nino December.

The main issue with 1957-58 isn't even those indexes - it's the fact that it was followed by another El Nino. That implies completely different processes at play in late winter that we won't see this year. The Indian Ocean Dipole was flipping hard toward a look that leads El Nino last year, and now it is doing the opposite.

More generally, the stratospheric event in 1958 was much later in January than what is expected to happen this year. The deeper you get into Jan-Feb, the more correlated the +PDO is to severe Eastern cold and hardly any years are above 1957-58 in that regard. I don't really think the SSW even added that much - it would have gotten severely cold regardless.

 

[40/70 Benchmark]

1 hour ago, raindancewx said:

57-58 had the QBO in the opposite phase of this year (+7 in Dec on the dataset), along with the PDO opposite. It's been fine as a match for December, but you can get lots of matches to this type of look in an El Nino December.

The main issue with 1957-58 isn't even those indexes - it's the fact that it was followed by another El Nino. That implies completely different processes at play in late winter that we won't see this year. The Indian Ocean Dipole was flipping hard toward a look that leads El Nino last year, and now it is doing the opposite.

More generally, the stratospheric event in 1958 was much later in January than what is expected to happen this year. The deeper you get into Jan-Feb, the more correlated the +PDO is to severe Eastern cold and hardly any years are above 1957-58 in that regard. I don't really think the SSW even added that much - it would have gotten severely cold regardless.

 

It definitely won't be as cold in the east due to the PDO, agreed, but that doesn't necessarily preclude a good snow-stretch. I get that the PDO is most highly correlated to temp late in the season, but it's also unwise to ignore what will be a very disturbed polar domain. Most of the shitty snow years in your data set had a very strong PV and given that we won't see a record trough down to Baja like we did last year, that will matter. 

[snowman19]

The main difference these days between El Niños and La Ninas in December is the location of the +10 or greater departures. During El Niños the warmest departures are in the north with the +PNA. They shift to the south during the La Ninas with the -PNA. Getting 3 Decembers since 2015 with any section of the US going higher than +10 is pretty extreme. +10 or greater departure months used to be very rare. The +10 at some stations last January in the Northeast was a 4th month which was during a La Niña. A 5th month was the +10 in the Southeast in February 2018. January 2020 which was neutral came close in some locations.
 
This gives further support to the MJO emerging in the IO phases like the models are showing:

[GaWx]

 The model consensus continues to go in the direction of a split major SSW in early January. Almost all of the El Niño years with them had following the splits significant periods of notable cold and/or winter storms, including some wintry precip quite far south (like N Florida).

[GaWx]

The MJO progs from the model consensus of early Dec were much too weak for late Dec: near circle (top image) vs actual well outside the circle (bottom image):

 

[Typhoon Tip]

2 hours ago, GaWx said:

 The model consensus continues to go in the direction of a split major SSW in early January. Almost all of the El Niño years with them had following the splits significant periods of notable cold and/or winter storms, including some wintry precip quite far south (like N Florida).

I don't have a problem with any of this.  I don't typically use/reliant upon the Euro sources but the GEFs have been flagging significant intrusive warm plumes in the 2 and 5 hpa ( I tend to use the temperatures and Wave 1 anomalies).  They haven't been exactly coherent in suggesting warm bursts precede correlated down welling through the systemic PV, but my experience in the past is that there's these modeled false entries into an event before the real one takes place - sort of thing...   I looked this morning and out around 300+ hours, there is now a ceiling thermal event through the entire depth, from 5 hpa all the way down to 70 hpa.   This method of observing warm intrusion in the 2 .. 5 hpa levels, followed by emergence successively down through 30 .. 50 .. 70 hpa ( those are pressure-altitude coordinates ),  has proved useful in early detection in the past.

By the way .. this is coming from the American model. It's way up at the 2 hpa. The 5 hpa also is highly anomalously displaced.  I've never seen these temperature wells displaced so S-E as to collocate over the Iberian Peninsula ... What ever causing this, it is suggesting that there is no PV over the regions N of 66.5N at these levels by 300+ hours. 

At a minimum ... this intuitively precedes and/or indicates active "breaking down" of PV. Which probably infers an era of extensive blocking nodes around the 50 to 70 N lower Ferral latitudes.  Checking the distant ( 300+ hour range) height distribution, at the 100 hpa (which tickles the top of the tropopause) the PV has fractured into 3 distinct SPV nodes.  These are all compelling arguments for SSW and at lest indirectly argue that the very important behavioral correlation of downward propagating warm plume(s) is under way.  Without this itl/bold statement, these warm bursts have almost N/S correlation to the Arctic Oscillation.

I also want to caution that the AO is forecast to switch to negative mode well prior to the SSW time-lagged correlation.  This is sometimes true of other years in history whence SSW that were downward propagating successfully ... took place in an already erstwhile/ongoing negative AO biased winter.  This would make it somewhat of a challenge trying to parse out exactly what is what in a the total manifold of forcing.  cross that bridge..

 

[LibertyBell]

On 12/23/2023 at 9:30 AM, jbenedet said:

100%. The foundation of NWP is chaos theory. I don’t believe most in here understand this, or they think they do but in reality, they don’t at all. They’ve never seen it demonstrated, explicitly. 

This stuff pisses me off to no end because the physical equations are robust; it is rock solid theory. It’s the future telling that causes the significant error bars; especially 5 days + out. It’s the addiction of wanting to see the future which has people abusing the modeling tools available to the general public and entrenching in the masses minds that atmospheric science is more BS than macroeconomics.

Taken together, observations are far more powerful (skill-wise) than hypotheticals built on representations of the global atmosphere 2 weeks+ out which are then used to make regional forecasts. No Skill! In other words, more work analyzing does NOT translate to more accurate forecasting results. It just means more jargon and BS explanations that dupe the naive and ignorant, “this guy must know something.” It also means forecasting accuracy is likely lower because it is more clouded, staring at variables as if they are knowns, when in reality they are unknowns because at future time points they are very poorly resolved or not resolved at all.

When it comes to complexity, most often, the simplest explanations are the most accurate ones. 

The question then becomes, should these tools actually be available to the general public?

 

[LibertyBell]

On 12/23/2023 at 9:46 AM, 40/70 Benchmark said:

 

Ah yes......genius. Don't ever issue a forecast, and you can claim to have "nailed it", while telling everyone who did why they missed it. Lol 

In all seriousness, I agree with a lot of what you are saying and I don't expect anyone to spend the time that I do...truthfully it isn't necessary, but it helps me to learn. All I am saying is that if you are going to be critical of others, at least go on record with something yourself. I will also say, these variable are know when looking in hindsight, which is the value of analogs.

also we can use ensembles when there's so many unknown variables, because ensembles can forecast trends, so rather than looking at it one dimensionally you can tweak the same model in different ways to see what outcomes are more likely than others.

 

[LibertyBell]

On 12/23/2023 at 10:23 AM, jbenedet said:

Occam’s Razor.

A “simple” hypothesis is one that doesn’t make unnecessary assumptions, not one that is conceptually simple.

 

I like elegant better, but yes it means the same thing.

The more variables you have, the more that can go wrong.

Occam's Razor is backed up by Murphy's Law

"What can go wrong will go wrong."

 

[LibertyBell]

On 12/23/2023 at 10:33 AM, roardog said:

A lot of the El Niño winters that you guys on the east coast love had this pattern in December. Maybe it’s a little warmer than December 1965 for instance since the planet is warmer overall but the general idea with the warmest anomalies being across the north(especially northern plains) is pretty classic Nino December. Your ‘02/‘09 cold Nino Decembers are pretty rare especially for a Nino that’s technically almost super.

It's more likely to be a one big storm winter, but in general this is correct.

 

[GaWx]

47 minutes ago, Typhoon Tip said:

I also want to caution that the AO is forecast to switch to negative mode well prior to the SSW time-lagged correlation.  This is sometimes true of other years in history whence SSW that were downward propagating successfully ... took place in an already erstwhile/ongoing negative AO biased winter.  This would make it somewhat of a challenge trying to parse out exactly what is what in a the total manifold of forcing.  cross that bridge..

 

 Agreed 100% about the difficulty in estimating what would have occurred without a SSW should one actually occur.

[LibertyBell]

On 12/24/2023 at 9:52 AM, GaWx said:

 Here is a screenshot of a list of major SSWs broken down by split vs displacement: noteworthy splits include 1/30/58 (El Nino), 1/8/68, 2/2/73 (El Nino), 1/9/77 (El Nino though not listed), 1/2/85, 12/8/87 (El Nino), 1/18/03 (El Nino), 2/9/10 (El Nino), and 1/2/19 (El Nino; too recent to be listed here but was a split). Note that there have been 7 split SSWs since 1958 during El Nino that were early enough to impact met. winter and all 7 were notable in terms of subsequent cold and/or historic E US winter storms:

 

what really stands out is how many we got in the 00s decade-- although that one in Jan 2002 didn't do much here, there was a big snowstorm in the Carolinas

[LibertyBell]

On 12/24/2023 at 10:17 AM, 40/70 Benchmark said:

Why would you want the PV lobe over NE? I have no interest in January 2004.

that was a very snowy winter though, at least the first half of it

 

[40/70 Benchmark]

22 minutes ago, LibertyBell said:

that was a very snowy winter though, at least the first half of it

 

Not for me.

[stadiumwave]

12z GFS

 

[GaWx]

1 hour ago, LibertyBell said:

what really stands out is how many we got in the 00s decade-- although that one in Jan 2002 didn't do much here, there was a big snowstorm in the Carolinas

That Atlanta to Carolinas major snowstorm was on Jan 2-3 and the SSW was on Jan 2nd. So, it’s likely that the snowstorm occurring had little to do with the SSW.

[GaWx]

2 hours ago, stadiumwave said:

12z GFS

 

1. That looks like a major SSW (with a split) on the 12Z GFS. The 0Z was right at 0.

2. The 0Z CDN ens mean dropped way down to -16 vs -3 yesterday (see below). A -16 for an ensemble mean 14 days out is extremely impressive! Anything -15 or under would be a record breaker for late Dec through mid Jan and there have been many in early Jan, alone.

3. The 0Z GEFS mean dropped way down to +7 vs +17 yesterday (see below):
 

[griteater]

12z EPS Mean from Allan Huffman

[GaWx]

51 minutes ago, griteater said:

12z EPS Mean from Allan Huffman

 So, the 12Z EPS is implying a most likely major SSW (when winds reverse) centered on Jan 6th. Per the 0Z, the most likely reversal dates appear to be for Jan 5th-8th. The 0Z was the first 0Z EPS with the mean touching 0 m/s. The lowest prior to that had been yesterday’s +2.5. And now this 12Z drops it to -5! A Jan 5-8 SSW date would suggest the period of greatest potential cooling from it on the E US would start ~Jan 20th and go at least well into Feb meaning covering a large portion of the best climo period for E US winter storms, especially.

[40/70 Benchmark]

41 minutes ago, GaWx said:

 So, the 12Z EPS is implying a most likely major SSW (when winds reverse) centered on Jan 6th. Per the 0Z, the most likely reversal dates appear to be for Jan 5th-8th. The 0Z was the first 0Z EPS with the mean touching 0 m/s. The lowest prior to that had been yesterday’s +2.5. And now this 12Z drops it to -5! A Jan 5-8 SSW date would suggest the period of greatest potential cooling from it on the E US would start ~Jan 20th and go at least well into Feb meaning covering a large portion of the best climo period for E US winter storms, especially.

This evolution would be absolute chef's kiss for verification purposes.

[raindancewx]

First major window for a colder/stormier pattern to me is coming up in mid-January. Take the period below and add 45 days - that's my guess for timing. Roughly Jan 7-21. After that, add another 45 days, Feb 21-Mar 7. If you blow those two periods in the East, I think it's just another shitty winter. As the El Nino collapses in Feb-Apr, there should be some extraordinarily powerful storms moving through the US as the balance of power in the Pacific changes. The cold period shown followed multiple storm days here, which looks fairly likely again in the Jan 3-7 period I outlined a week ago using the Bering Sea Rule. We'll see soon enough.

[griteater]

Notable drop on tonight's GEFS Ext 00z run (10mb zonal winds)

 

[GaWx]

4 minutes ago, griteater said:

Notable drop on tonight's GEFS Ext 00z run (10mb zonal winds)

 

 The # of members with a major during just Jan 5-12 went from the 0Z 12/24 GEFS’ one (3%) to the 0Z 12/25 GEFS’ 11 (36%)! I’m getting very close to the @40/70 Benchmarklevel of confidence that there will be one in early Jan.