GaWx

Latest models are pretty unimpressive. That includes the 12Z UKMET, which has no TCG.

The latest daily WCS PDO fell back some more to -0.90 (for Oct 30th). Per WCS, the PDO will come in near -1.0 for Oct vs ~-2.0 for Sep. Since NOAA has been coming in significantly more negative, that tells me that the NOAA PDO for Oct will probably come in or near the -1.5 to -1.8 range, which would be an impressive rise from the -2.94 of Sept as it would be one of the largest El Niño rises Sep to Oct. (rise likely to be in/near range of 1.14 to 1.44): Other large El Niño Sep to Oct NOAA PDO rises: -1876: 1.08 -1880: 1.79 -1918: 0.90 -1969: 2.02 -1986: 1.03 -2006: 1.08

The new EPS mean maintains after the stronger than avg Nov SPV a drop to a significantly weaker than normal SPV by mid Dec. with a mean of 25 mps (56 mph) vs climo norm of 34 mps (76 mph) for then (normals peak then). The individual members imply a 10% chance for a major SSW in mid Dec., which is significantly higher than normal as the chance for the entire month is only ~10% and is pretty high for this far out. So, that will be something to monitor. If there were to be no SSW, which is the most likely scenario, this does still imply a good chance for a -AO dominating mid to late Dec. and would put any E US mild Dec forecasts in jeopardy. Keep in mind my earlier post that shows that the extended Euro Weeklies have done much better with the easier to predict polar stratosphere than the polar troposphere because there’s little atmospheric chaos way up at strat levels:

I can’t assess personally how the Euro Weeklies have done in general because I just started following their SPV forecasts last Feb. and I only restarted following them ~two weeks ago. They did do well last Feb. Here’s the oldest one I saved (2/6/23), which forecasted the mid Feb major SSW impressively well along with the SPV remaining quite weak into March instead of recovering much: The following graph of Feb 10th forecasts during 2003-22 for 60-90N suggests that the extended portion of the Euro Weeklies likely does much better with the stratosphere vs the troposphere. Note the 60%+ performance (two darkest greens) for above 100 mb (the stratosphere) through the entire six weeks of Weeklies vs significantly worse performance in the mid to lower troposphere (250 mb and lower levels). The explanation is that the strat is easier to predict because of it being so far above the surface, where there isn’t nearly as much atmospheric chaos: Thus, the recent EPS forecasts for a weaker than avg SPV in mid Dec should probably not be dismissed as mere extreme speculation. If we are actually able to get a solid -AO in Dec, then that obviously would increase the chance that the forecasts calling for a mild E US Dec would be out to lunch. I’d be a bit nervous right now if I had a mild E US Dec. Regarding El Niño Decembers since 1950, these 8 had a -AO: 1958, 1963, 1968, 1969, 1976, 1987, 2002, and 2009. So, about 1/3 of them had a -AO (near climo % chance for all Decembers). Which of these 8 were cool to cold in the bulk of the E US? All but 1987, which was slightly AN. Explanation: its -AO was only weak (weakest of the 8) and it was the only one of the 8 without a -NAO (it was weakly +NAO).

The extended Euro ensemble has consistently for many days been forecasting a stronger than average strength SPV in Nov (peaking Nov 11th in near record high territory) followed by a weakening SPV to near neutral late in Nov. For the last week or so, it has also been forecasting the weakening to then progress to a significantly below average strength SPV by mid Dec, which implies a good shot at a -AO then along with a better than average potential in mid Dec for an SSW. But keep in mind that sudden SSWs are uncommon in Dec as peak season doesn’t even start til mid Jan. So, this could very well end up as a -AO/weaker than average SPV without an SSW, which would still be a cold signal for the E US especially if a -NAO could somehow accompany it:

The CFS ONI forecasted peak has been bouncing back recently and is now back up to ~+1.73/+1.75 for NDJ/DJF, vs +1.59/+1.56 in NDJ/DJF as of 10-14+ days ago. The Jan 3.4 forecast is now back up to ~+1.98, which is near its latest forecast for Nino 3. For Jan, it now has 1+2 (which it has been way too cool with for many months) at ~+1.25 and Nino 4 at ~+1.1. So looking strictly at anomalies, the latest CFS has a cross between an EP and CP El Niño this winter. Latest CFS forecast of ONI: peak +1.75 DJF

I follow the SAI and it has since 2014 actually been a reverse indicator of the winter AO vs what it used to indicate. So, if the trend of the last 9 years were to continue into the winter of 2023-4, the AO may actually end up negative despite the lower than average SAI so far this month. Also, we still don’t have the final 2023 SAI. I hope it stays below average based on what has occurred since 2014.

The OHC anomaly has finally been rewarming recently although only gradually to just over +1.0 C, i.e. nothing to write home about and still well below the mid June and late August peaks:

The re-drop on the WCS PDO chart continues and it was at -0.81 yesterday after hitting a high of ~-0.15 on 10/21. The momentum will probably take it down at least to ~-1.00. The drop over the last week is no surprise following the tremendous 20 day rise of 2.45 from way down at -2.60. This is a correction of sorts. What I’ll be watching for is where equilibrium is reached and ends the current drop. Hopefully this won’t be much below -1.00 and it resumes a rise, which is supported by the current strong El Niño. Just a reminder: the NOAA monthlies have been more negative than what these WCS charts have been showing. NOAA daily is probably ~~-1.25 to -1.50 now:

So, even more reason to forecast either a neutral NAO or +NAO for this winter overall. (I’m still leaning -NAO in Feb). But what about a -AO this winter? Since 1979-80, we’ve had more -AO winters without a -NAO (9) than -AO/-NAO winters (6). Five of the last 14 El Niño winters have had a -AO (1979-80, 1986-7, 1997-8, 2002-3, 2009-10) vs only two of the last 14 having a -NAO (1986-7, 2009-10). What I feel confident about is a +PNA this winter. A whopping 12 of the last 14 El Niño winters had a +PNA.

Based on the last 44 winters, there have been for unknown reason(s) many more -AO (defined as sub -0.50) winters than -NAO (defined as -0.25) winters. These 9 winters had a -AO but no -NAO: 1979-80, 1985-6, 1997-8, 2000-1, 2002-3, 2003-4, 2005-6, 2012-3, 2022-3 At the same time, no winters since 1979-80 have had a -NAO but no -AO. Since 1979-80, there have been 15 -AO winters but only 6-NAO winters. OTOH during 1950-1 through 1978-9 there were 15 -AO winters and 16 -NAO winters. Only two of these winters had a -AO and no -NAO: 1950-1 and 1952-3. And three had a -NAO but no -AO: 1958-9, 1963-4, and 1970-1. So, something has changed over the last 44 years to make it way more difficult to get a -NAO winter than a -AO winter despite an increase in summer -NAOs. So, based on this, I’m giving it a significantly higher chance that 2023-4 will have a -AO vs a -NAO. During the last 44 winters, only these 6 had a -NAO: 1984-5: (NAO -0.70) (moderate La Niña) headed down in cycle 21 with limited sunspots averaging only 20/month; this was 20 months before the minimum of Sept 1986 1986-7: (NAO -0.30) (moderate El Niño) very early in cycle 22 four months after the Sept 86 minimum with very limited sunspots of only 6/month 1995-6: (NAO -0.62) (moderate La Niña) headed down in cycle 22 with limited sunspots of only 12/month nine months before Sept 1996 minimum 2009-10: (NAO -1.67) (strong El Niño) one year into cycle 24 at 13 months after the Dec 2008 minimum just starting to head upward but with still limited sunspots of only 22/month 2010-11: (NAO -0.68) (strong La Niña) two years into cycle 24 at 25 months after the Dec 2008 minimum headed further up but still low at 33/month 2020-1: (NAO -0.42) (moderate La Niña) early into cycle 25 at 13 months after the Dec of 2019 minimum just starting to head up but still with limited sunspots of only 14/month So, there has been no winter of the last 44 with a -NAO averaged out over the winter along with DJF sunspots above 33/month. With us now in an active part of the sunspot cycle, that’s even more reason for me to feel it is going to be a challenge to get a -NAO this coming winter. So, a -AO combined with a neutral NAO or +NAO looks to me more likely this winter than a -AO/-NAO. However, even with that, I’m leaning toward a -NAO Feb.

I don’t subscribe. This is from the link posted above. After such a sharp rise, the current decline back to -0.7 from -0.2 is totally expected (reverse dead-cat bounce). What’s not known is how much further this reverse dead-cat will go. Per the current trajectory, it going back below -1 at least for a short period wouldn’t surprise me. Also, the monthly NOAA PDOs have been a good bit more neg than the ones calculated from WCS dailies. If there were a NOAA daily reported, I’d guess it would be ~-1.25 to -1.5. Regardless, Oct will almost certainly come in much less negative than Sep on both. I’m wild guessing ~-1.75 for Oct NOAA vs -2.94 in Sep. That would mean a rise of ~1.19, which is quite significant. We’ll see whether the overall momentum of rising continues into Nov. I’m leaning that way based on the huge rise (on the posted graph) earlier this month along with it being a strong Nino. My hope as a cold preferrer is that we end up with a DJF PDO higher than -1 and preferably higher than -0.75 (I’m talking about how NOAA table measures it). Analogous El Niño SE winters following sharp Sep to Nov rises are 1876-7 (cold), 1880-1 (cool), 1884-5 (cool), 1969-70 (cold), 1986-7 (normal), and 2002-3 (cool). It is important that the rise continues through at least Nov.

https://s2s.worldclimateservice.com/climatepanel/

It was -.70 yesterday per below though the NOAA PDO runs more negative and is thus likely ~-1.25:

I guess it depends on one’s definition of atmospheric coupling.

Not only has the SAI not shown a positive correlation to a winter -AO during the last 10 years, it has actually shown a significant negative correlation during that period!

2023’s ASO SOI looks to come in near -11. To compare: 2018: -4 2015: -19 2014: -8 2009: -5 2006: -12 2004: -4 2002: -10 1997: -16 1994: -12 1991: -12 1987: -10 1986: -2 1982: -21 1979: -2 1977: -11 1976: -7 1972: -11 1969: -9 1968: -1 1965: -12 1963: -8 1958: +1 1957: -6 1953: -10 1951: -10 - The six moderate El Niños since 1951-2 averaged -7 with a range of -1 to -12. - The 10 weak ones also averaged -7 (with a range of +1 to -12). - The four strong ones averaged -8 (range -5 to -12). - The five super-strong averaged -16 (range -11 to -21). Based on the above: whereas 2023’s -11 ASO SOI isn’t “representative” of a super-strong, it is easily representative of strong (actually border between strong and super) and thus solidly translates to that form of atmospheric coupling during ASO imho.

Tammy has quickly become tropical again and is now a TS.

The significantly warm biased Euro, especially in spring, has been noted before. However, the Australian (BoM) average bias of zero during the 10 years prior to 2023 in May runs’ ONI peak forecast had also been noted though it was also mentioned that there were rather significant swings from too warm to too cool. So, whereas BoM average bias was near zero during 2013-22, the accuracy wasn’t necessarily anything to write home about. It had a pretty good sized average error but it was just that that error was evenly balanced between too warm and too cool. This year of course will end up way too warm barring an unforeseen sudden upcoming rapid warming.

Here’s why I feel the SAI, which is supposed to predict the winter AO, has for some unknown reason(s) not come through during the last 9 years in general: SAI per this graph (in msq): top 4 (mean 11.5) -2018 SAI 14: winter AO +0.2 -2016 SAI 14: winter AO +1.0 (2nd strongest +AO since 2014) -2015 SAI 13.5 winter AO 0.0 -2019 SAI 12.5 winter AO +2.1 (strongest +AO since 2014) The two winters since 2014 with the strongest -AO by far had these SAIs and AOs: -2020 SAI 11 -1.8 -2022 SAI 10 -0.6 So, the two most -AO winters since 2014 (2020 and 2022) were both preceded by a below mean SAI. Also, none of the top 4 SAIs of the last 10 had a -AO winter and two of them were the strongest +AOs since 2014! So, when I said that the SAI in recent years could be used as a reverse indicator, this is the evidence. AO monthly: https://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/monthly.ao.index.b50.current.ascii.table

Of the 6 listed El Nino autumns with large PDO rise, 5 of the 6 subsequent winters were cold in the E US: 1876-7, 1880-1, 1884-5, 1969-70, 2002-3 The only one that wasn’t was 1986-7, which was near normal. So, there have been no mild E US El Niño winters on record back to the 1870s that had a preceding rapidly rising autumn PDO. Also, the projected Oct 2023 PNA looks to favor a chilly E US, especially SE, per El Niño analogs. OTOH, strong +IOD El Niño autumns have generally been the opposite with mild to near normal subsequent winters dominating in the E US. So, in summary, we have mixed indicators.

Update: My rough guess based on Oct 1-24 is for the WCS Oct PDO to end up near -1.05. Based on that, NOAA’s Oct PDO should come in somewhere within -1.4 to -2.1 with the best guess near -1.7. This would mean a substantial rise from Sept’s -2.94 along with the prospect for another substantial rise for the Nov PDO, when there’s a chance NOAA could even come in >-1.0 based on current trends. If it did, that would mean a rise of 1.95+, the largest NOAA PDO rise from Sept to Nov for the last 55 El Niños going back to the 1850s: Largest NOAA PDO increases Niño Sep to Nov 1986: 1.71 1969: 1.65 1876: 1.44 1884: 1.42 1880: 1.37 2002: 1.29 NOAA monthly PDO back to 1854: https://www.ncei.noaa.gov/pub/data/cmb/ersst/v5/index/ersst.v5.pdo.dat

Indeed, a strong vortex is on the way for early to mid Nov. However, the Euro weeklies continue to suggest a drop to neutral near the end of Nov along with a continued fall to a little weaker than average in early Dec:

The good news is that the SAI hasn’t been successfully predicting the winter AO ever since I discovered it ~10 years ago. It may have even become a contra-indicator, which if true would mean that bottom feeder SAI would actually favor a -AO winter.

DMI, which is apparently the same thing per this: Dipole Mode Index (DMI) Intensity of the IOD is represented by anomalous SST gradient between the western equatorial Indian Ocean (50E-70E and 10S-10N) and the south eastern equatorial Indian Ocean (90E-110E and 10S-0N). This gradient is named as Dipole Mode Index (DMI). When the DMI is positive then, the phenomenon is refereed as the positive IOD and when it is negative, it is refereed as negative IOD. As the original provider has changed how they update the index, we have decided to calculate it here using the HadISST1.1 SST dataset for the entire time period. Climatology is currently 1981-2010. Time Interval: Monthly Time Coverage: 1870 to present Update Status: Periodically updated https://psl.noaa.gov/gcos_wgsp/Timeseries/Data/dmi.had.long.data