• 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 Machine Learning Mauler 1/30-2/1

Allan Huffman


f649dc7c94d59bc9d2ab45cb8e2d7ac9.png



Sent from my iPhone using Tapatalk
 
FFC Discussion - https://mesonet.agron.iastate.edu/wx/afos/p.php?pil=AFDFFC&e=202601281943

Snippet

"Our major sticking points for the forecast will be the initiation of
precipitation and precipitation potential on the backside of the
surface low. Several model members, those which are trending slower,
bring more moisture in ahead of the system and generally lead to a
wetter first half. The models which move quicker end up developing a
stronger low which creates a more dynamic system leading to back-end
snow potential as the main hazard (though the initial push may
trigger snow showers in the mountains and isolated flurries
elsewhere). At this point, given the snow coverage across the TN
valley, the colder and faster solution seems to have the slight edge
in this situation. The result from this would be a quicker push of
cold air and the development of snow across the north side of the
surface low in northeast Georgia and western South Carolina. As the
system moves eastward/northeastward, strong northwest flow in the
low levels will drive wind gusts >30mph for much of the area.
Combining this with the 20:1 snow ratios and we could see some
"surprise" snowbands with isolated accumulations even in the dry
slot.

At this time the probability of at least 1" of snow in the ATL metro
sits between 10 and 40% with higher probabilities north and east.
This is likely an attempt to represent the patchy coverage with most
areas getting little to nothing and a lucky few receiving a good
quick snow."
 
Just catching up on the models runs overnight, I like where CLT is for a nice several inch snow event. But I don't buy the crazy high numbers the gfs is spitting out, and I think ratios tend to be overdone. Generally speaking, I think qpf is too high on op models and like to go with the Euro EPS for it. Whatever it says, it's usually close to that. Right now, my guess is the ULL doesn't produce huge totals, the coastal really doesn't give 77 west that much precip if at all (it never seems to in my recollection), and the precip ends sooner than modeled every time west to east. Therefore right now, I'd expect around .3 inches or so from the ULL. So a 3-4 inch storm seems reasonable, and the best we've seen since 2018 I believe.

That drying is what you have to look for though out west. I think Raleigh area ends up better due to the coastal.
I’m typically in agreement with you on the ratios and not expecting them to be much more that 10:1. However I do think this one of the rare exceptions that up to 20:1 is real possibility back our way. I’m basing that on what I’m seeing in the forecast soundings with all models and smart mets that I respect like Webb and Brad P are talking about it as well.
 
New National Blend of Models. It usually runs a cycle behind so we might see the totals trimmed a little for the next update.
1770141600-bhVFQZlUQEo.png
I have honestly never seen the NBM this high so widespread in the Carolinas. It’s interesting that Brad P often sights this in his video posts and forecasts
 
pulling inspiration from NWS local offices?

You can check my previous winter storm forecasts over the last decade ish here if you are curious (& yeah I’ve had some clunkers along the way like all of us), but the most snow I have personally ever predicted in central-eastern NC is 5-10” right before the Jan 2018 storm. Just shows you how anomalous this event could be…

We certainly may challenge this forecast “record” of mine tomorrow evening!

https://www.webberweather.com/winter-storm-forecast-verification.html
 
You can check my previous winter storm forecasts over the last decade ish here if you are curious (& yeah I’ve had some clunkers along the way like all of us), but the most snow I have personally ever predicted in central-eastern NC is 5-10” right before the Jan 2018 storm. Just shows you how anomalous this event could be…

We certainly may challenge this forecast “record” of mine tomorrow evening!

https://www.webberweather.com/winter-storm-forecast-verification.html
I was just jesting a bit of course - but when an NWS office prints something like 3-10" range, I always chuckle
looking forward to your map on this storm!
 
42c930a701f43b0f6c787c9873a78ef5.jpg

I don’t think I saw it posted earlier, here’s the 12z Euro Kuchera


Sent from my iPhone using Tapatalk
Truly Murphy-to-Manteo. 100% of NC in accumulating snowfall. Same seems to be true for SC, which is even crazier!
 
You can check my previous winter storm forecasts over the last decade ish here if you are curious (& yeah I’ve had some clunkers along the way like all of us), but the most snow I have personally ever predicted in central-eastern NC is 5-10” right before the Jan 2018 storm. Just shows you how anomalous this event could be…

We certainly may challenge this forecast “record” of mine tomorrow evening!

https://www.webberweather.com/winter-storm-forecast-verification.html
also shows you how much we've sucked lol
 
Can someone explain what this means in simple terms for my future reference.
The colors on the east coast are drifting south meaning the trough is + tilted so there wouldn't be any lift. Basically we all end up cold and dry. Luckily this model is awful until about 24 hours out tops
 
The colors on the east coast are drifting south meaning the trough is + tilted so there wouldn't be any lift. Basically we all end up cold and dry. Luckily this model is awful until about 24 hours out tops
I remember this in the last storm chase, we saw that on a few models and we were worried about qpf. Thanks for the explanation.
 
Back
Top