GaWx

Chris, What are your thoughts about what met. Travis Roberts said about the E US winter potential for cold/trough similar to Jan of 2014 in the video (linked by Anthony above) due to the very warm water in the NE Pacific that’s similar to 2014? My concern is that 2014 didn’t also have that very warm water going back to Japan thus meaning it isn’t an apples to apples comparison. I thought that the 2014 warm water was heavily concentrated in the NE Pacific and thus was the primary cause for that Jan 2014 cold E US. But in contrast now, the warm water isn’t just concentrated in the NE Pacific. Related to this, the August 2013 PDO wasn’t nearly as negative with it at -1.56 vs August 2025’s -3.23:

Yeah, I’ve been reading discussions elsewhere regarding CC’s general effect on tropical seasons: fewer storms but stronger extremes. Well, we’ve had the stronger extremes with a whopping 11 cat 5s just since 2016! Compare that to the prior 11 cat 5s occurring over a 27 year period meaning a near tripling of their frequency. But what’s not yet evident is the drop in the avg # of storms as there have been 14+ every year since 2016 with 168 storms 2016-24 or an avg of a whopping 19 storms/year! Compare that to only an avg of 14 storms the prior 9 years and 15 during the 9 years before that: https://en.m.wikipedia.org/wiki/List_of_Category_5_Atlantic_hurricanes

-It’s still rather active in the Gulf just beyond midmonth on the 12Z ensemble runs. -New Euro Weeklies are similar to yesterday with below avg ACE weeks 1-2 followed by a rise to NN. Gulf activity is BN weeks 1-2 rising to NN or above weeks 3-4.

Indeed, August rose substantially from July’s record low of -4.12 to -3.23. It appears to still be in a rising trend for now.

Middle of MDR moving WNW to W as a strengthening TD early next week underneath a stout high:

NEW TROPICAL CYCLONE FORECAST TO DEVELOP AFTER 150 HOURS FORECAST POSITION AT T+150 : 13.9N 32.5W LEAD CENTRAL MAXIMUM WIND VERIFYING TIME TIME POSITION PRESSURE (MB) SPEED (KNOTS) -------------- ---- -------- ------------- ------------- 0000UTC 15.09.2025 156 15.0N 33.7W 1012 27 1200UTC 15.09.2025 168 15.1N 37.9W 1010 33

Any snowy Dec (8”+ ) followed by dud? I’ll check now: Indeed, your idea verifies well! I have to go all of the way back to 1945-6 to find the last snowy Dec not having 16”+ Jan+! Dec had 15.6” and then they had only 12.1” Jan+. -Then I have to go back to 1926-7: 11.7” Dec; 10.6” Jan+ -1912-3: 11.4”; 3.1” -1902-3: 14.4”; 14.3” That’s it going back to 1869-70. Just 4 winters out of 43 (9%) with 8”+ in Dec failed to get 16” Jan+. For the other 113 winters (<8” in Dec): 46 (41%) Jan+ had <16” Jan+ *So, only 1 in 11 Jan+ were <16” when Dec 8”+ vs just over 4 in 10 when Dec 8”+! That’s a pretty strong correlation.* Longterm avg for is Jan+ is 22”. ———— *Edit: I counted 22 Jan+ <10”. 21 of those were when Dec was <8”. So, whereas 21 of the 113 (19%) when Dec <8” had Jan+ with <10”, a mere 1 of the 43 (2%) when Dec 8”+ had Jan+ with <10”! So, 2% vs 19%, meaning an even stronger indication of the correlation between Dec and Jan+ snow at NYC.

Thanks. As I mentioned elsewhere, I think a good portion of the EW lowering in weeks 1-2 is due to the 91L model bust while the powering in week 3 is too far out to be 91L related. Regarding trusting them, I don’t trust any model as there’s too much inherent uncertainty especially as one goes out further. But I do use them as tools. I feel that the EW have done reasonably well enough since 2024 regarding the overall big pic/trends that following them at least for trends is probably somewhat useful.

My thought is that he did pretty well with boldly saying early on the Carolinas would be impacted by Erin. How has he done otherwise as of Sept 7th? Had he made other bold calls?

Is this Gary Lezak stuff?

I can’t make that detailed a call, but that’s the start of prime W Caribbean/Gulf season climowise.

The OHC has recently been cooling substantially after the prior rather steady period: @snowman19you’ll probably like this trend

That first tstorm gave me only a little rain. But a stronger storm is starting now just before 6PM with very gusty winds, heavy rains, and some nearby CTG lightning. SPECIAL WEATHER STATEMENT NATIONAL WEATHER SERVICE CHARLESTON SC 548 PM EDT SUN SEP 7 2025 GAZ118-119-072215- COASTAL CHATHAM GA-INLAND CHATHAM GA- 548 PM EDT SUN SEP 7 2025 ...A STRONG THUNDERSTORM WILL IMPACT CHATHAM COUNTY UNTIL 615 PM EDT... AT 548 PM EDT, DOPPLER RADAR WAS TRACKING A STRONG THUNDERSTORM OVER HUTCHINSON ISLAND, MOVING SOUTHEAST AT 10 MPH. HAZARD...WIND GUSTS UP TO 50 MPH. SOURCE...RADAR INDICATED. IMPACT...GUSTY WINDS COULD KNOCK DOWN TREE LIMBS AND BLOW AROUND UNSECURED OBJECTS. LOCATIONS IMPACTED INCLUDE... DOWNTOWN SAVANNAH, HUTCHINSON ISLAND, MIDTOWN SAVANNAH, HUNTER ARMY AIRFIELD, WINDSOR FOREST, GARDEN CITY, THUNDERBOLT, SAVANNAH HISTORIC DISTRICT AND SANDFLY. —————— **Edit: I ended up with ~0.90”

MJOwise, alone, 9/7-21/25 is clearly more favorable than 9/7-21/24. But the MJO is just one factor of many. There’s lots of variation of actual outlines within each MJO phase. The MJO is best treated as a guidance tool rather than a crystal ball.

Snowman, the GFS and EPS MJO forecasts (the two best imho) for the next 2 weeks agree that the MJO will remain in the IO and then reverse back to the W Hem/Africa. There’s no MC in sight:

Ray, I graded the Sept ‘24 Euro 2M temp. forecast for DJF 2024-5 an F, the worst Sept winter forecast going back at least to that of Sept of 2017, the furthest back I can find. The Euro averaged ~4F too warm for most of the US! The only area it did well with was the SW. The Sept ‘20 Euro forecast for 20-21 was similarly too warm but not by quite as much (avg of ~3F too warm for the US). OTOH, the Sept forecast for 2023-4 was pretty bad in the other direction with an average miss of ~3F too cold for the US overall. The last good Sept Euro 2M forecasts for the lower 48 as a whole were for 2021-2 and 2019-20. The avg miss in the NE was very small.

3:25PM: I’m getting a thunderstorm now, the first rain here this month.

Barry, I don’t know why JB is saying that. The Sept 2025 MJO (first 7 days plus 2 week forecast through 9/21) looks nothing like 2024! Sept 2025: -GEFS: lower to moderate amplitude 2, 3, 2, 1, 8, which is actually favorable -EPS: lower to moderate amplitude 2, 3, 2, 1, 8, which is actually favorable Sept 1-21 of 2024 (blue): moderate to high amplitude 4, 5, 6 (not usually favorable unlike what we now have) was like night and day vs 2025: Perhaps JB is looking further ahead to late Sept-Oct. Regardless, he wants lots of clicks!

9/7/2015 (see below) was the last 9/7 TWO with no new TCG being considered. However, that was going into a strong El Niño, there was TS Grace then in existence in the E MDR, and the outlooks then went out only 5 rather than 7 days. Also, this TWO failed to see a new TD (which lead to Henri) that formed 9/8-9 in the subtropics: TROPICAL WEATHER OUTLOOK NWS NATIONAL HURRICANE CENTER MIAMI FL 800 AM EDT MON SEP 7 2015 For the North Atlantic...Caribbean Sea and the Gulf of Mexico: The National Hurricane Center is issuing advisories on Tropical Storm Grace, located several hundred miles west-southwest of the Cape Verde Islands. Tropical cyclone formation is not expected during the next 5 days. $$ Forecaster Pasch https://www.nhc.noaa.gov/archive/xgtwo_5day/gtwo_archive.php?current_issuance=201509071437&basin=atlc&fdays=2

1. Indeed. The Sept Euro fcast for the NE DJF anomaly is for an avg of only ~+1.8 F. 2. But that is almost as warm as any Sept fcast for the NE DJF of the last 9 with it just slightly colder than 24-5’s ~+2.0 F and near 22-3, 20-1, and 19-20’s +1.8 F. 3. How has the Sept Euro verified vs actual for La Niña in the NE? I’ll look at NYC: For NYC using 1991-2020 avgs for the 5 Nina winters since 2017-8: -24-5 ended up -1.4 F vs ~+2F Sept Euro fcast or ~3.4 F colder than Euro -22-3 ended up +5F vs Euro ~+1.8F fcast or ~3.2F warmer than Euro -21-2 ended up +1.1F vs Euro ~+0.4F fcast or ~0.7F warmer than Euro -20-1 ended up 0.0F vs Euro +1.8F fcast or 1.8F colder than Euro -17-8 ended up +0.2F vs Euro +0.9F or 0.7F colder than Euro So, for the 5 Niña winters since 17-8, the Sept Euro turned out to be in F: 3.4 too warm, 3.2 too cold, 0.7 too cold, 1.8 too warm, and 0.7 too warm or an avg of 0.4 too warm

The 18Z GEFS is the most active run yet in the W Car/Gulf 9/12-end of the run! If I were living on the Gulf coast, I’d be a bit concerned. As Barry earlier mentioned, this has strong support from the MJO. More about that soon:

Regarding all 9 Euro DJF fcasts issued in Sept since 2017-8 (as far back as can be seen at the Euro site), there wasn’t even a tiny speck of BN anywhere in the lower 48 E of 110W long. The only one that had any BN anywhere in the lower 48 was the 2019-20 fcast, which had a small area of ~-1F in a small part of the Mtn W (which amazingly enough ended up being the only area with BN): Despite that, there were notable areas of actual BN in portions of the lower 48 in all of these DJFs except 2019-20 and 2023-4. So, your statement ECM is now incapable of seeing below normal temps has essentially been true for its Sept fcasts since 2017-8. However, two of its 8 Oct fcasts and one of its Nov fcasts for DJF did have some BN: Oct ‘23, which failed as there was no BN in the lower 48: Oct ‘18, which got the BN in KS right but the rest of this BN area ended up NN or AN and also it missed the widespread BN in NW 1/2 of the US: Nov ‘17 in UP of Mich, which was correct, but it missed the widespread BN in the Midwest, Plains, and N Rockies:

Here are the Sept Euro forecasts for DJF 2024-5 back to 2017-8 vs actual 2024-5 prog: avg 4F too warm much of US but SW close; grade: F 2024-5 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2024&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2023-4 prog: MW 5-8 too cold, NE 3-4 too cold, SE 1-3 too cold, W close; grade D 2023-4 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2023&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2022-3 prog: W MW 3 too warm; E MW and NE 3 too cold, SE 1-2 too cold, W 6 too warm; grade D 2022-3 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2022&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2021-2 prog: pretty close except 2 too warm W; grade A 2021-2 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2021&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2020-1 prog: 0-4 too warm MW, 3 too warm NE/SE, SC 4-6 too warm, W 1-2 too warm; Grade D 2020-1 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2020&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2019-20 prog: MW 2 too cold, NE close; SE 2-3 too cold, W 0-2 too cold; grade A 2019-20 actual: https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2019&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2018-9 prog: MW 1 too warm, NE close, SE 2-3 too cold, W 2-3 too warm; grade B 2018-9 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2018&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot 2017-8 prog: MW 3 too warm, NE 2 too warm, SE/W close; Grade C 2017-8 actual https://psl.noaa.gov/cgi-bin/data/usclimdivs/climdiv.pl?variab=Temperature&type=1&base=7&mon1=12&mon2=2&iy[1]=2017&iy[2]=&iy[3]=&iy[4]=&iy[5]=&iy[6]=&iy[7]=&iy[8]=&iy[9]=&iy[10]=&iy[11]=&iy[12]=&iy[13]=&iy[14]=&iy[15]=&iy[16]=&iy[17]=&iy[18]=&iy[19]=&iy[20]=&iy[21]=&iy[22]=&iy[23]=&iy[24]=&iy[25]=&iy[26]=&iy[27]=&iy[28]=&iy[29]=&iy[30]=&irange1=&irange2=&xlow=-10&xhi=10&xint=1&iunits=1&scale=100&iwhite=1&iswitch=0&Submit=Create+Plot ——————— -GPA: 2.0 -Last 3 poor -Best 2019-20 and 2021-2; worst 2024-5 -2 A, 1 B, 1 C, 3 D, 1 F -The Sept ‘25 fcast for 2025-6 is for US as a whole the 2nd warmest to 2024-5 since 2017-8 -NE prog misses: +4, -3.5, -3, 0, +3, 0, 0, +2; so, NE prog was too warm thrice, too cold twice, and close thrice; So, NE prog on avg was only 0.3 F too warm vs actual meaning negligible bias vs actual though avg miss was by ~2 F -MW prog misses: +4, -6.5, 0, 0, +2, -2, +1, +3; so, MW prog was too warm 4 times, too cold twice, and close twice; So, MW prog on avg was only 0.2 F too warm meaning negligible bias vs actual though avg miss was by ~2 F -SE prog misses: +4, -2, -1.5, 0, +3, -2.5, -2.5, 0; so, SE prog was too warm twice, too cold 4 times, and close twice; So, SE prog on avg was only 0.2 F too cool vs actual meaning negligible bias vs actual though avg miss was by ~2 F -W prog misses: +2, 0, +6, +2, +1.5, -1, +2.5, 0; so, W prog was too warm 5 times, too cold once, and close twice; So, W prog on avg was 1.6 F too warm! Thus, there appears to be a notable Euro warm bias for the W for winter at least in Sept progs. Avg miss was ~2 F

I just found Sept 2M forecasts for Dec and Jan individually though not Sept for DJF: https://charts.ecmwf.int/products/seasonal_system5_standard_2mtm?area=NAME&base_time=202509010000&stats=ensm&valid_time=202510020000 @donsutherland1@mitchnick

In addition: The last few GEFS and EPS runs have forecasted MJO in the relatively favorable phases of 2, 1, and 8 through Sept. 19 and this would probably extend at least through the rest of Sept per longer term model guidance and climo: