GaWx

I use the conservative UKMET much more for position than for strength. It was the best by a good margin for Ian and one of the best for Irma and Idalia once within a few days of landfall. For strength, it is the trend rather than the actuals that are usually more informative.

The 12Z JMA, admittedly inferior for the tropics, has an obviously worrisome position at 192 in the Caribbean just S of DR.

When looking at the 12Z EPS 228 vs 0Z EPS 240, the consensus is similarly keeping the risk to the corridor from the NE Caribbean to the Bahamas pretty low (~10% based on ~5 of 51 hitting Bahamas) and to the region well east that includes Bermuda significantly higher. The bulk of the members cross into the box that covers 22-32N, 60-70W. (Bermuda is near 32N, 65W.) Hardly any are (aiming) west of 75W. Verbatim, this run suggests very low risk to the CONUS/Canada, similar to the prior run. There are once again many very powerful hurricanes on this run.

UKMET 0Z 168: 0000UTC 10.09.2023 168 19.5N 62.1W 1005 32 UKMET 12Z 156: 0000UTC 10.09.2023 156 20.6N 64.0W 1002 49 So, 12Z UK at 156 is a bit WNW of 0Z at 168 and stronger. Here's the complete 12Z UK run: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 72 HOURS FORECAST POSITION AT T+ 72 : 14.3N 44.1W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 1200UTC 06.09.2023 72 14.3N 44.1W 1009 27 0000UTC 07.09.2023 84 14.7N 46.4W 1008 36 1200UTC 07.09.2023 96 15.7N 48.7W 1007 34 0000UTC 08.09.2023 108 16.4N 51.6W 1006 30 1200UTC 08.09.2023 120 17.2N 54.8W 1005 36 0000UTC 09.09.2023 132 18.1N 58.1W 1004 36 1200UTC 09.09.2023 144 19.3N 61.4W 1003 44 0000UTC 10.09.2023 156 20.6N 64.0W 1002 49 1200UTC 10.09.2023 168 21.5N 67.3W 1000 54 Looking at H5, there is a pretty strong persistent E coast trough and stationary TX strong (594 dm) ridge. The trough is stronger than it is on the other models. If that were to verify, that would imo likely recurve it away from the SE US like the model consensus is still suggesting. However, from NC Outer Banks north wouldn't be as clearcut. Again, this is strictly per the 12Z UKMET, not a forecast. H5: scroll down https://weather.us/model-charts/gbr/2023090312/north-america/geopotential-height-500hpa/20230910-1200z.html

Well, actually, the Nino 3 anomaly was slightly warmer than 3.4 in December. So, I guess it can be considered east-based to start. But Nino 3.4 was barely warmer in Jan and significantly warmer in Feb leading to a higher DJF in 3.4. I thought you were calling 1972-3 east-based, overall. If not, I misunderstood you. Nino 1+2 averaged cooler than both, especially in Jan and Feb.

1. Why are you calling 1972-3 east-based? I see per ERSST that DJF anomalies averaged higher in 3.4 than in any other region. They're slightly higher than Nino 3 and significantly higher than 1+2. Nino 4 anomalies are much lower than the other three regions. So, it certainly isn't west-based. But with 3.4 having the highest, wouldn't this be considered central rather than east based? The most central region has the highest anomalies. 2. I assume you agree that east-based doesn't always mean mild E US. 1976-7 had the highest anomalies of the four regions in Nino 3 and that was a frigid winter. I do realize it being weak helped its chances.

After two runs without TCG being noted on the UKMET, the new run (0Z) has TCG. So, together with a much stronger CMC vs 12Z, there is some reversal of recent runs' weakening trends despite the 0Z GFS being similar to its recent weaker runs. 0Z UKMET: moving WNW at 15 just N of Leewards at 168: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 84 HOURS FORECAST POSITION AT T+ 84 : 14.8N 44.0W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 1200UTC 06.09.2023 84 14.8N 44.0W 1010 24 0000UTC 07.09.2023 96 15.7N 45.9W 1009 27 1200UTC 07.09.2023 108 16.4N 47.8W 1008 29 0000UTC 08.09.2023 120 17.2N 50.6W 1008 26 1200UTC 08.09.2023 132 17.8N 53.7W 1008 29 0000UTC 09.09.2023 144 18.3N 56.3W 1007 27 1200UTC 09.09.2023 156 18.6N 59.6W 1007 31 0000UTC 10.09.2023 168 19.5N 62.1W 1005 32

The 12Z EPS at 228 is similar to the 0Z EPS at 240 with <20% of members threatening the NE Caribbean, Bahamas, or CONUS. Of those that do, two appear to be cat 4-5 with a 923 and a 933 as of 240 both moving WNW near/toward the Bahamas. Bermuda is still threatened by a good # and there are once again many Hs.

Down to 994 at 192.

Yep although it still is the strongest of the 12Z runs at 144. Also, the 0Z Euro had similar weakening 132-156 followed by restrengthening. I'll be looking to see if the 12Z is similar.

12Z Euro is strongest of the 12Z models as of 120 and pretty close to earlier runs.

The 12Z model consensus is weaker/delayed on TCG. Not only does the GFS have very little early in the run (GFS TCG is delayed til the Caribbean), the CMC is similar. Plus the last few ICONs have been weaker than recent days. Also, the last two UK (goes out 168) have no TC after many in a row with it. Trend? Mean further west if there is ever TCG?

NDJ

The CFS ONI peak has come back way down from +2.15 just a few weeks ago to ~+1.80-1.85.

For this same AEW, I count 27 12Z EPS members (53%) with a H. That is more than the 0Z and is as high as any EPS I can recall for any as yet to develop system in the MDR this season to date. So, at a minimum, this looks likely to be a big ACE producer giving us pretty satellite pics. Edited

From the 12Z UKMET hour 168: Surface showing TC 300 miles E of Leewards (moving WNW at 15): scroll down https://weather.us/model-charts/gbr/2023090112/americas/sea-level-pressure/20230908-1200z.html -------------------------- H5: In addition to being a threat to the NE Caribbean, the 12Z UKMET doesn't yet look like anything close to a guarantee to not hit the CONUS although the best bet this far out is a miss based on 75%+ of E MDR systems doing just that: scroll down https://weather.us/model-charts/gbr/2023090112/americas/geopotential-height-500hpa/20230908-1200z.html

It won't be +1.2. It will be either +1.0 or +1.1. August OISST is ~+1.3, which translates to ERSST likely being slightly cooler than that (~+1.25?). June was 0.84 and July was +1.00. So, if Aug ERSST turns out to be +1.25, that would mean JJA average of +1.03, which would round down to +1.0 for the table. But if Aug ERSST were to come in at +1.31+, that would give a JJA of +1.05+, which would be reported as +1.1 in the table. So, that's why I say +1.0 or +1.1. To get a +1.2, you'd need a +1.6 for Aug, which isn't happening.

-Usually the best bet for an E MDR system since 75%+ don't hit the US although nowhere near a safe bet at least yet imo -12Z UKMET: still has TCG but unlike the prior run having TCG on Sunday, this waits til Wednesday (9/6). With the delay, it is notably weaker but is still in a potentially dangerous position at 168 (9/8) 300 miles E of the Leewards moving WNW at 15 mph: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 120 HOURS FORECAST POSITION AT T+120 : 14.2N 44.1W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 1200UTC 06.09.2023 120 14.2N 44.1W 1010 27 0000UTC 07.09.2023 132 14.1N 46.7W 1009 29 1200UTC 07.09.2023 144 14.9N 49.0W 1007 37 0000UTC 08.09.2023 156 15.4N 51.8W 1007 31 1200UTC 08.09.2023 168 16.1N 54.9W 1007 34

Actually, this new CANSIPS run, despite its -NAO, isn't very cold. Where do you see "very cold"? It is for the bulk of the E US (other than the slightly AN in upper Lakes to NNE) near normal (after adjusting to 1991-2020 instead of 1981-2010 baseline), which is oddly enough slightly warmer than the prior run (which was slightly BN in the SE): Of course, near normal would seem very cold relative to most recent winters. I'll take the colder change although I'm hoping for actual BN here in the SE as El Niño provides the best chance for that.

I'm trying to understand the terminology better. At 12:12Z on 8/30, the linked graph for Tampa shows a 6.64 foot tide (red line) that appears to be using MLLW as a reference. It also shows a 1.433 predicted (blue line, which is the astronomical) tide then. In addition, when one clicks on "observed - predicted", it then displays a purple line that was at its max of 5.21 feet at 12:12Z. That's 6.64 ft minus 1.433 ft. So, that appears to be the maximum extra water level caused by the storm. Are you saying that the maximum "observed - predicted" is not the storm surge? If not, what is that called? https://tidesandcurrents.noaa.gov/waterlevels.html?id=8726607&bdate=20230829&edate=20230831&units=standard&timezone=GMT&interval=6

From NWS CHS: "A MAXIMUM TIDE LEVEL OF 3.47 FT MHHW (9.23 FT MLLW) WAS OBSERVED AT THE CHARLESTON HARBOR TIDE GAGE." From the first image linked below, there was a maximum surge of 2.38' at CHS at 8:12 PM on August 30th that coincided exactly with high tide to produce the 9.23 ft tide (in red)/CHS battery flooding: https://tidesandcurrents.noaa.gov/waterlevels.html?id=8665530&units=standard&bdate=20230829&edate=20230831&timezone=GMT&datum=MLLW&interval=6&action= This was made worse by the center being ~50 miles closer to CHS at high tide vs where models had projected it then. Thus the strongest perpendicular SE winds came in and peaked just before and at high tide instead of being much lighter then, which would have been the case had the center been 50 miles further away. OTOH, it being 50 miles closer to CHS at high tide meant 50 miles N of SAV meaning W (offshore) winds at Ft. Pulaski instead of very close to SAV. That result was a major factor that lead to the maximum surge at Ft. Pulaski of 2.7' occurring much earlier at 1:36 PM on August 30th, which was luckily near LOW tide when winds were nearly perpendicular to the coast (ESE winds). Soon after that, winds shifted to SSE (no longer perpendicular and thus not pushing as much water in) and eventually to WSW at the time of high tide literally pushing the high water out. The result was that at the 8:24 PM high tide, the storm surge had shrunk to nearly nothing as per this linked image, which meant at Ft. Pulaski only an 8.95 ft tide vs a 9.4 ft tide (both in red) at the high tide the evening before. So, while CHS was having their worst flooding, Ft. Pulaski and the nearby islands/lowlands around SAV were already back to normal! https://tidesandcurrents.noaa.gov/waterlevels.html?id=8670870&bdate=20230829&edate=20230831&units=standard&timezone=GMT&interval=6 ---------- Edit: My power was out ~10 hours (5PM 8/30-3AM 8/31). I measured under 1" of rain. Closer to the coast got less than well inland, which was well predicted by the models. The airport got ~1.5". Just 50 miles inland from SAV despite the fast storm movement (20 mph), Candler, Bulloch, and Screven counties had flooding from very heavy rainfall as the remnants of the very wet NW eyewall hit that area. Statesboro, Metter, and Newington were deluged with 7-8"! That extended into Allendale, Hampton, and Colleton counties. Some of that scraped CHS (airport well inland), which got 3". There were 3 confirmed EF0 tornadoes, two in the CHS metro (including the one linked to earlier ITT that caused that car to lift in the air) and one in Liberty County, GA. No injuries and minimal damage, fortunately.

Well, I looked at the site linked below and it appears that Tampa may have had a 5.2 ft storm surge at 8:12 AM on August 30th! If this were to verified as accurate, that may be the highest surge at least since the 5 ft from Gladys of 1968! That's pretty amazing considering how far offshore it was and that it landfalled way up in the Big Bend! Can anyone verify this data? https://tidesandcurrents.noaa.gov/waterlevels.html?id=8726607&bdate=20230829&edate=20230831&units=standard&timezone=GMT&interval=6

The typically conservative UKMET (0Z) once again is pretty strong with this E MDR lemon AEW that about all models now develop: NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 54 HOURS FORECAST POSITION AT T+ 54 : 10.9N 22.9W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 1200UTC 03.09.2023 60 11.4N 24.8W 1008 37 0000UTC 04.09.2023 72 10.8N 27.6W 1008 32 1200UTC 04.09.2023 84 11.6N 31.6W 1008 31 0000UTC 05.09.2023 96 12.3N 35.1W 1006 34 1200UTC 05.09.2023 108 13.0N 39.3W 1005 35 0000UTC 06.09.2023 120 13.7N 42.6W 1005 31 1200UTC 06.09.2023 132 13.9N 45.5W 1001 37 0000UTC 07.09.2023 144 14.5N 48.1W 1000 39 1200UTC 07.09.2023 156 15.3N 50.6W 997 51 0000UTC 08.09.2023 168 16.2N 53.3W 997 47

Due mainly to Harold and Idalia, much of the Gulf has cooled ~1C over the last two weeks: Aug 17th near record warmth dominated by 31-32C (~88-90F): August 30th: still very warm but much of it cooled ~1C to mainly 30-31C (86-88F) with a small area affected by both storms cooling nearly 2C (32 to 30) near 25-26N, 84-85W:

Great point. Does anyone know what the official Tampa Bay storm surge was?