GaWx

Whereas winter -NAOs have been much less frequent in recent decades than they had been, summer -NAOs have done the opposite and have increased markedly. % of summer months with NAO<0: increased 1950-97: June: 50% July: 42% Aug: 46% 1998-2023 June: 73% July: 65% Aug: 73% % of winter months with NAO<0: decreased 1950-97: Dec: 42% Jan: 48% Feb: 46% 1998-2022/3: Dec: 36% Jan: 20% Feb: 27% So, whereas summer and winter -NAO month frequencies were about the same during 1950-97, summer -NAO months since 1998 have been 2.5 times as frequent as winter -NAO months! Thus, having a summer -NAO (at least based on recent decades) doesn't at all increase the chance for a subsequent winter's -NAO. https://www.cpc.ncep.noaa.gov/products/precip/CWlink/pna/norm.nao.monthly.b5001.current.ascii.table

The 6Z Euro hour 72 is slightly E of 0Z 78 and similar to yesterday's 12Z at 90.

Not the same, but you can bet on weather via weather futures and options: https://www.cmegroup.com/trading/weather/files/weather-futures-and-options-fact-card.pdf

Last 5 EPS runs (0Z and 12Z only): # of members with CONUS H landfall 0Z 9/6: 5 (10%) 12Z 9/5: 2 (4%) 0Z 9/5: 4 (8%) 12Z 9/4: 1 (2%) 0Z 9/4: 2 (4%) So, although still a small %, the latest run has the highest yet.

The 0Z UKMET is a bit stronger. But more importantly, it is recurving 200 miles E of the 12Z run thus implying a significantly lower risk of landfall to the CONUS. Whereas two runs ago it was in the SE Bahamas and the last run was just N of the SE Bahamas, this run is ~250 miles NE of those islands: TROPICAL STORM LEE ANALYSED POSITION : 13.8N 42.1W ATCF IDENTIFIER : AL132023 LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 0000UTC 06.09.2023 0 13.8N 42.1W 1008 27 1200UTC 06.09.2023 12 14.1N 44.7W 1006 37 0000UTC 07.09.2023 24 15.3N 46.9W 1004 41 1200UTC 07.09.2023 36 16.4N 49.5W 1003 34 0000UTC 08.09.2023 48 17.0N 52.2W 1003 35 1200UTC 08.09.2023 60 17.7N 54.9W 1002 42 0000UTC 09.09.2023 72 18.4N 57.9W 1000 45 1200UTC 09.09.2023 84 18.8N 60.2W 1000 48 0000UTC 10.09.2023 96 19.8N 62.1W 998 53 1200UTC 10.09.2023 108 20.5N 64.2W 998 55 0000UTC 11.09.2023 120 20.8N 65.6W 997 54 1200UTC 11.09.2023 132 21.2N 67.0W 996 52 0000UTC 12.09.2023 144 22.0N 67.7W 993 54 1200UTC 12.09.2023 156 23.3N 68.0W 987 51 0000UTC 13.09.2023 168 24.4N 68.9W 977 56

The 12Z Korean has a 940 mb MH at 276 just S of Long Island that then slams Suffolk County 9/16-7: https://weather.us/model-charts/korea/2023090512/usa/sea-level-pressure/20230917-0000z.html

The 12Z UKMET is 250++ miles west of all of the 12Z GEFS members except for that one on the left, which it is close to.

Eyeballing it tells me the new Euro ONI peak is near or just under +2.25 for NDJ vs +2.40 on the prior run. Feb is ~+2.05. So, although still well within super peak strength, it did drop slightly. CFS dropped more than slightly vs a few weeks ago but it is an inferior model.

12Z UKMET 162-168 shows start of recurve but much closer to SE than other models implying a potential problem for NC north after run ends considering that it is recurving at least 5 degrees (~300 miles) further west than GFS: scroll down to see each map SLP 162 https://weather.us/model-charts/gbr/2023090512/middle-america/sea-level-pressure/20230912-0600z.html SLP 168 https://weather.us/model-charts/gbr/2023090512/middle-america/sea-level-pressure/20230912-1200z.html H5 at 162 https://weather.us/model-charts/gbr/2023090512/north-america/geopotential-height-500hpa/20230912-0600z.html H5 at 168 https://weather.us/model-charts/gbr/2023090512/north-america/geopotential-height-500hpa/20230912-1200z.html Edited

12z Euro at hour 180 ~75miles SSW of 0Z 192 with 925 mb but still recurving (along 67W).

That looks like a -NAO being predicted by this Euro run for DJF averaged out based on a just over 1 mb AN SLP at Iceland and a just over 1 mb BN SLP in the Azores. Do you or does anyone else have the 2M temp for DJF? Or can someone provide me a link please?

I agree that it's possible although a steep climb for the next 3.5 weeks to ~+2.2 would be needed. I do see that ERSST caught up to OISST for August as a while vs it being slightly cooler than OISST for many months. So, that ups thise chances somewhat. *Edited for correction*

Even with the 12Z UKMET being 150 miles ENE of its 0Z run at the end, it is at hour 168 still a whopping 350 miles WSW of the 12Z GFS at 168.

-12Z UKMET initialized this at 13.4N, 39.1W vs the Best Track's 12Z of 12.2N, 39.6W. So, the 12Z's initialization is 90 miles NNE of the actual position. -12Z UKMET is stronger and is at hour 156 ~150 miles ENE of the 0Z's hour 168. At 168, the 12Z is N of the SE Bahamas moving WNW at only 5 mph vs the 0Z being in the SE Bahamas: TROPICAL DEPRESSION 95L ANALYSED POSITION : 13.4N 39.1W ATCF IDENTIFIER : AL952023 LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 1200UTC 05.09.2023 0 13.4N 39.1W 1011 24 0000UTC 06.09.2023 12 13.8N 42.0W 1009 25 1200UTC 06.09.2023 24 14.2N 44.7W 1007 31 0000UTC 07.09.2023 36 15.3N 46.7W 1007 34 1200UTC 07.09.2023 48 16.4N 49.0W 1006 35 0000UTC 08.09.2023 60 17.5N 52.0W 1006 31 1200UTC 08.09.2023 72 18.5N 55.1W 1007 32 0000UTC 09.09.2023 84 19.4N 58.4W 1006 30 1200UTC 09.09.2023 96 19.9N 61.2W 1005 39 0000UTC 10.09.2023 108 20.6N 63.6W 1003 46 1200UTC 10.09.2023 120 21.4N 66.5W 1002 50 0000UTC 11.09.2023 132 21.9N 68.6W 998 52 1200UTC 11.09.2023 144 22.4N 70.2W 998 50 0000UTC 12.09.2023 156 23.9N 71.0W 996 60 1200UTC 12.09.2023 168 24.2N 72.0W 996 46

I remember it and the Euro both being W of the other models and thus those two doing the best for Irma in FL overall once within 5 or so days of landfall. Actually for Ian last year for both the FL and SC landfalls, the UKMET was further right rather than further left of the model consensus and it did the best. Also, for Idalia, the UK was right of the GFS/CMC and did better. So, I don't know whether or not it still has a too far W bias. Maybe it still does in the longer range out in the MDR though.

.NEW TROPICAL DEPRESSION FORMS IN THE CENTRAL TROPICAL ATLANTIC... ...FORECAST TO BECOME A POWERFUL HURRICANE BY THE END OF THE WEEK... 11:00 AM AST Tue Sep 5 Location: 12.5°N 40.2°W Moving: WNW at 15 mph Min pressure: 1008 mb Max sustained: 35 mph 0Z UKMET initialized 190 miles due east of this

The 0Z UKMET initialized at 12.5N, 37.4W. The Best Track had 95L at 11.2N, 36.2W at 0Z; 11.5N, 37.5W at 6Z; and at 12.2N, 39.6W at 12Z. So, the 0Z UK initialized 1.3 further N and 1.2 further W or 120 miles NW of the Best Track position. However, 95L has moved about 3:1 W by N from 0Z to 12Z rather than moving the 1:1 of straight NW. Thus, the 12Z Best Track position is 150 miles W of the 0Z UK initialization. Translation: the 12Z UKMET track may very well initialize nearly due W vs its 0Z initialization. We'll see whether or not that translates to the 12Z UK track being even further W to SW of the already far W to SW 0Z run vs the other models, but that's quite possible. *Edited 2nd paragraph

1. The 0Z GFS has 95L as a MH on 9/13 100 miles W of Bermuda followed by a big hit on Nova Scotia 9/14-5. 2. The 12Z UKMET was the westernmost major 12Z model run by 300-350 miles. Despite that, the 0Z is ~200 miles WSW of the 12Z. It is also weaker. So, at 168, the 0Z UKMET is a whopping 500 miles WSW of the 0Z GFS and 450 miles W of the 0Z ICON in the SE Bahamas moving slowly WNW: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 24 HOURS FORECAST POSITION AT T+ 24 : 13.3N 41.7W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 0000UTC 06.09.2023 24 13.3N 41.7W 1010 23 1200UTC 06.09.2023 36 13.9N 44.5W 1007 31 0000UTC 07.09.2023 48 15.0N 46.5W 1007 35 1200UTC 07.09.2023 60 16.1N 48.9W 1006 34 0000UTC 08.09.2023 72 17.0N 52.0W 1006 32 1200UTC 08.09.2023 84 17.9N 55.3W 1006 35 0000UTC 09.09.2023 96 18.9N 58.6W 1006 32 1200UTC 09.09.2023 108 19.3N 62.1W 1006 38 0000UTC 10.09.2023 120 20.3N 65.1W 1005 40 1200UTC 10.09.2023 132 21.2N 67.9W 1004 41 0000UTC 11.09.2023 144 21.7N 69.4W 1005 38 1200UTC 11.09.2023 156 22.1N 71.5W 1004 36 0000UTC 12.09.2023 168 22.7N 72.9W 1004 34

Yes, at one point they did. Though they did initially and correctly suggest a south of west track through the MDR due to ridge placement, that ridge was initially modeled to break down in the west with an advancing ECONUS trough. But that trough was overmodeled and lifted out while the WAR kept building west. Even after Irma was driven into the Leewards, there were more Carolina solutions in the ensembles than S. Florida. The majority did finally come west when the ECONUS trough was handled better. Irma eventually steamrolled the northern coastline of Cuba, and the rest is history. It's just good to keep in mind that though we're used to surprises with the global OPs swinging back and forth, ensembles do swing as well when out far enough in range. To confirm what I think you already know: After aiming Irma for the SE US on the 8/31/17 EPS runs, the EPS mean moved NE for the next two days of runs to where about half recurved offshore the E coast (though the mean being near 76W wasn't nearly as far offshore as today's runs have 95L, which is near 67W and with only one member hitting): Then the next day, 9/3/17 (seven days out), the EPS started trending back to the SW and was back to being centered on FL (just as the 8/31/17 0Z run was) by the 9/5/17 0Z run!

I think that a +2+ Nov OHC is quite possible, especially as regards the 3rd column of the table found here: https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ocean/index/heat_content_index.txt Note the +1.40 of June of 2023 in the table. Now compare to the OHC graph that I posted above. June doesn't even rise past +1.35 at any point during the month on this graph and thus the month as a whole when eyeballing it is only ~+1.25. Thus the table is ~0.15 warmer than the graph. July in the table is +1.02 whereas the graph says it is only ~+0.90. So, the table is ~0.12 warmer than the graph. August in the table is +1.10 vs ~+1.05 on the graph (table ~0.05 warmer). So, although the difference has been shrinking, the table could continue being slightly warmer than what the graph suggests.

Yes, at one point they did. Though they did initially and correctly suggest a south of west track through the MDR due to ridge placement, that ridge was initially modeled to break down in the west with an advancing ECONUS trough. But that trough was overmodeled and lifted out while the WAR kept building west. Even after Irma was driven into the Leewards, there were more Carolina solutions in the ensembles than S. Florida. The majority did finally come west when the ECONUS trough was handled better. Irma eventually steamrolled the northern coastline of Cuba, and the rest is history. It's just good to keep in mind that though we're used to surprises with the global OPs swinging back and forth, ensembles do swing as well when out far enough in range. Below is the Euro suite for 8/31/17 at 12Z, which was when Irma was close to the current longitude of 95L and when it was 10 days from S FL. Note that this was already very threatening to the US with ~3/4 of the members along with the op either threatening the FL/SE coast or going into the GOM. The mean is at 77W at 240 and still moving NW/NNW. In contrast, the furthest W today's 12Z EPS mean got was ~67W and already moving due N with only one member hitting the US: Also, below is a Tweet showing the 0Z 8/31/17 EPS, which was similarly highly threatening to the SE US with many members even into the GOM: My point is that although model consensus means shifted too far north the next day as you showed, the two EPS runs were very bullish for a US hit at the 10 day point compared to the very bearish EPS runs today for 95L. Now granted, Irma was already a TS/H on 8/31/17 vs 95L being only an Invest.

Good point. I'm currently near 75% for a miss of the CONUS or pretty near the climo based chance for it to not hit. If I were to go strictly by the EPS, I'd be near 95%. But in deference to some of the other models and the extra uncertainty present before a TC has even formed, I'm not nearly that high at least for the time being.

OHC is keeping up its steady rise. It is as of ~Aug 30th near +1.30 and not far from the highest so far this El Niño:

The 12Z EPS is similar to recent runs with track consensus but is even stronger when averaging the members. I don't recall ever seeing a stronger average for an as yet to form TC.

The inferior JMA (12Z), unlike the worrisome run from 24 hours ago going into the Caribbean to just S of the DR, is ~500 miles NE of that run and is recurving near 65W.