• Hello, please take a minute to check out our awesome content, contributed by the wonderful members of our community. We hope you'll add your own thoughts and opinions by making a free account!

Wintry January 23rd-27th 2026

NickyBGuarantee said:
Well if we keep going south that could be a problem. I was not in the camp that this south trend would hold… I would still typically favor going north around hour 90 but dang models are consistently going south .. yall wanted a dominate high pressure .. well it’s coming and it might be screwing the upper south before too long..

Might have to eat my words about this thing going north
Click to expand...
I think we’ll see a correction north once we get inside 72 hours, but until then I don’t really see anything to stop a south trend as long as models are spitting out 1048, 1050+ highs. I do suspect that those are overdone a bit, but even if we have a 1044 mb, there’s gonna be a limit to how far north this system goes.
 
blueheronNC said:
AI folding, wow. I guess it has no “precedent” in the dataset when it’s been so long. Do tail events get underpredicted by AI when it’s not part of the trained scope of outcomes?
Click to expand...
Truly rare events in the training set are things like intense cyclones that are globally rare.

CAD/HP situations like these are incredibly common around the globe, so how rare it is HERE is not relevant to the model training. The model doesn't see 'snow event in North Carolina' or 'Strong HP pushing across the plains' in training, but rather just 'snow event' or 'strong HP'.
 
bouncycorn said:
Truly rare events in the training set are things like intense cyclones that are globally rare.

CAD/HP situations like these are incredibly common around the globe, so how rare it is HERE is not relevant to the model training. The model doesn't see 'snow event in North Carolina' or 'Strong HP pushing across the plains' in training, but rather just 'snow event' or 'strong HP'.
Click to expand...
In other words, the AIFS and other AI models aren't emulating historical weather events, they are implicitly emulating the PDEs that govern the atmosphere.
 
BKWRN29 said:
No it won’t hold. This thing will make many changes before we get a grasp on the location. Bottom line, someone’s gonna get something.
Click to expand...
Of course, no storm details are set in stone thus far out, but, I think we can have a little more confidence in this set up that there may not be drastic changes. For one, we are not dealing with a few degrees of tilt in the trough axis to drastically affect the surface like that last “storm”. And two, it is not a phased system where we have to worry if two packets of energy fuse at the last second.
 
blueheronNC said:
AI folding, wow. I guess it has no “precedent” in the dataset when it’s been so long. Do tail events get underpredicted by AI when it’s not part of the trained scope of outcomes?
Click to expand...
What I’ve seen today is every model coalescing more around a consensus and shaving fringes off of the confidence interval of what could happen. With a small south shift right now with that consensus. It’s a nudge towards consensus but idk if I would call this a “fold”, it’s holding on to the structure of what it sees well
 
Edward Nygma said:
I wish I could recall the storm, but I know for a fact there is a storm where the GSO RAOB had a warm nose aloft but was saturated with very cold near-surface temps below -12c and GSO was reporting snow not FZRA despite a classic FZRA sounding. For some reason December 2002 is a guess at GSO, but I'd have to check obs and soundings. Point being to @Webberweather53 suggestion, it definitely has happened with a very cold low-level air mass before in NC.
Click to expand...

Yeah that was the storm! Dr Lackmann talked about this in my synoptic class nearly a decade ago and it was because the rain that fell into the cold nose froze so fast it shattered, and the shattered remnants of the now ice nuclei recombined back into small snow dendrites before they hit the ground. I’ve seen this happen a few times over the years
 

Webberweather53 said:
Yeah that was the storm! Dr Lackmann talked about this in my synoptic class nearly a decade ago and it was because the rain that fell into the cold nose froze so fast it shattered, and the shattered remnants of the now ice nuclei recombined back into small snow dendrites before they hit the ground. I’ve seen this happen a few times over the years
Click to expand...
Synoptic is now taught by a different professor (took it lest semester) but Lackmann did a guest lecture here and he knows his winter weather stuff
 
rburrel2 said:
@bouncycorn could the ai models be doing poorly bc this set up is so anomalous? And since they only have data back to 1980, they just don’t have tons of data points on how this plays out?
Click to expand...
I just answered this above in more detail, but the short answer is “no”
 
Does that freezing rain ever change over to just rain in Ga by the time its done or just hours and hours of freezing rain?
 
bouncycorn said:
In other words, the AIFS and other AI models aren't emulating historical weather events, they are implicitly emulating the PDEs that govern the atmosphere.
Click to expand...
Max Velocity says that the models will have a tough time in the south because it doesn't usually snow in the south. Because it's so rare, they can't figure it out well.
 
ChattaVOL said:
Do we care about timing at all on this?

GFS has this coming in around 12 hours earlier than the EURO.


Sent from my iPhone using Tapatalk
Click to expand...
BAM weather said that the slower Euro would give it time to gain strength and come north. Doesn’t look like that’s being depicted now. The euro has been know to hold energy too long in the south west too. Don’t know if that’s in play with this or not.
 
Ross said:
the cold high, and please bear with me, vertically is shaped like a shield volcano. so think of the structure of the cold air shaped like this. this is stupid but it's the only way i could think of how to describe it
View attachment 185848
this doesn't change. that low level cold bleeding in will always be a feature. zr is always going to be around with this setup- no way to get rid of it, it's just pinning down where that zone is.
Click to expand...
Is the zr going to be Macon south, Griffin south, Atl. south, Gainesville, south. Chattanooga south?? Most folks want snow, some crazy people want sleet, and hardly anyone, except the innocent, or deranged want zr, lol. Something wintery is becoming ineluctable, but we aren't there yet. and zr is always part of the mix.
 
You must log in or register to reply here.
Back
Top