GaWx

Great question. I haven’t investigated Euro AI ensemble precip/temp maps closely, but a “quick and dirty” check doesn’t reveal to me any obvious algo issues with those. So, my focus on these posts remains fully with the snow maps.

Hey Carvers and others, WxBell definitely didn’t fix the Euro AI ensemble snow maps algos. These two members (#10 and #22) with multiple inches of snow way out in the Gulf are from the hour 84 6Z Euro AIFS, when temps are near normal (70s):

Due to much colder 2 week forecasts vs how it looked on Friday, natural gas is now up an incredible 19%!! If anywhere near this rise were to hold til the close, this would likely be the largest % daily rise in years:

Forecasted indices for 1/24-6: 1. PNA rises from neutral to moderate + 2. AO strong - 3. NAO moderate - 4. EPO rises from a very strong - to moderate -; despite the moderation, the lagging from the prior very strong -EPO in combo with a rising PNA and with the help of a -AO/-NAO is why extreme cold will be dropping into the C/E US. 5. WPO pretty strong - 6. MJO: GEFS has 7 while EPS has mainly 8 So, overall the indices are just about as good as it gets for cold and thus storm potential. —— Edit: Due to the huge amount of blocking, this time there may not be the typical NW trend as we get closer. In other words, the models a few days out may already be closer to reality assuming there actually is a storm.

The #s are out. We just had the 3rd largest 2 day drop in winter of the EPO into negative territory on record back to 1948: 2026 01 14 88.01 2026 01 15 -92.49 2026 01 16 -254.34 So, it dropped 342 from 1/14 to 1/16. The only two larger 2 day winter drops are not much higher: the 367 of 2/9-11/1996 and the 364 of 1/16-18/1961. As I said, this isn’t a measure of the strongest EPO ridges. Rather, it is a measure of the speed in establishing a new strong EPO ridge.

I do need to emphasize that I’m only talking about the algos that produce the WxBell EPS AI snow maps having a major issue. Regarding the qpf and other maps, I have no evidence to suggest these are off. So, I’m saying that it’s possible that it’s only the snow maps that have a major problem with regard to EPS AI.

Carver and others, The WxBell Euro AIFS ens snow maps have a flaw so major that the 12Z run (just for one example) had one member (#38) 300 miles offshore in the Gulf getting a foot of snow within a 6 hour period while the temp was 65F+! You read that right, a 2”/hour snowstorm with temps in the 60s. These maps are so flawed than even Joe Bastardi, himself, noticed the same problem a few weeks ago! If even Joe noted this, well….use these at your own risk. The regular EPS snow maps are otoh legit:

If it is looks like it’s insane, quacks like it’s insane, it really is insane. That’s because the WxBell Euro AIFS ens snow maps have a flaw so major that the 12Z run (just for one example) had one member 300 miles offshore in the Gulf getting a foot of snow within a 6 hour period while the temp was 65F+! You read that right, a 2”/hour snowstorm with temps in the 60s. These maps are so flawed than even Joe Bastardi, himself, noticed the same problem a few weeks ago! If even Joe noted this, well….use these at your own risk. The regular EPS snow maps are otoh legit:

I’m not sure. But near the bottom was the regular Euro as it was last to trend NW with the track. It kept the snow centered near the coast for so long.

The WxBell Euro AI ensemble snow maps have to be using faulty algos. Here’s proof: From today’s 12Z Euro AI ens, this is member #38’s 6 hour snow map for hours 270-276 showing a blob of a foot of snow in the Gulf 300 miles from the N Gulf coast meaning an avg of 2”/hour over 6 hours lmao: That right there is proof enough that the snow algos are screwed up bigtime. But fwiw, I’ll also show the temp anomalies for hours 270 and 276 for that member: Member #38 temp 2m anom hour 270: Member #38 temp 2m anom hour 276: Those temp maps show that the coldest it is during hours 270-6 where it has that snow blob is only ~8 BN. The SST there now is ~77F. That means that the air temp during the supposed snow is likely no colder than ~65F. There’s no way it could snow a foot way out in the Gulf, much less while air temps are 65F+. Conclusion: There is a major algo issue with these Euro AI ensemble snow maps. Thus, I’d be very wary about using them. Even snow weenie Bastardi criticized them. EPS ens snow maps are fine, but not Euro AI ens snow maps. @donsutherland1

Yep, for the main snow not starting for 6-7 days out this may be the heaviest I’ve seen on an EPS mean in NC even after taking ~1” off in some areas for today; also there’s quite the falloff S of NC as ZR seems to take over

Swainsboro switched from rain to snow at the 11AM hourly:SWAINSBORO LGT SNOW 34 34 100 W5

1-1.25” Marianna, FL, just SW of GA/FL/AL corner!1042 AM SNOW MARIANNA 30.78N 85.24W 01/18/2026 M1.3 INCH JACKSON FL PUBLIC ABOUT 1 TO 1.25 INCHES OF SNOW MEASURED AS A FINAL ACCUMULATION.

Has a lot of it melted away already? I read someone else said the Macon camera showed it “gone”, but I didn’t know exactly what he meant. That seemed awfully fast if it did.

At Smithville, GA, 10 miles S of Americus, 2-3” reported by EM:1002 AM SNOW SMITHVILLE 31.90N 84.26W 01/18/2026 E3.0 INCH LEE GA EMERGENCY MNGR 2 TO 3 INCHES OF SNOW REPORTED IN SMITHVILLE BY LEE COUNTY EM. SOME ACCUMULATION ON DIRT ROADS WITH ICY PATCHES ON SPOTS IN PAVED ROADS.