February Medium/Long Range Thread

[dailylurker]

1 minute ago, WeatherQ said:

Looks like Millville, Ocean View, the DE/MD Beaches and the lower DelMarVa are gonna be the winners, eh? 

I'm headed down there on Wednesday morning or before it starts. Lots of places to stay. I'm thinking oceanfront. 

[MillvilleWx]

3 minutes ago, WeatherQ said:

Looks like Millville, Ocean View, the DE/MD Beaches and the lower DelMarVa are gonna be the winners, eh? 

You guessed it! 

[Solution Man]

Just now, adelphi_sky said:

My son and I snow chased at Ocean City the last time they got a foot. We drove back to DC the morning after and there was no snow on the ground. Had fun though. Got a cheap hotel room with an atrium covered pool area.

Cool man, that’s something you both will never forget 

[Amped]

1 minute ago, TSSN+ said:

Just looks more phased there if anything 

Yes but it's too early and prevents the southern stream from developing. You want lower heights over MO/AR and higher heights over WV/PA to force a slower more northerly track.

[Lowershoresadness]

3 minutes ago, MillvilleWx said:

You guessed it! 

winners?  it smells like manure down here

[mitchnick]

AI east again

[nw baltimore wx]

On 11/15/2024 at 8:44 AM, nw baltimore wx said:

Tiebreaker (SBY): 16.5”

 

On 11/15/2024 at 12:07 PM, MillvilleWx said:

If SBY gets shelled like that, I will be visiting my parents at some point at the shore haha 

 

On 11/15/2024 at 2:37 PM, nw baltimore wx said:

The winter of @CAPE and @Lowershoresadness!!  If so, I'll be doing road trips both west and east.

I’m not giving up on this storm, but it is a Nina.

[anotherman]

AI east again

I would like to not see anymore posts like this.

[Jebman]

14 minutes ago, WeatherQ said:

Looks like Millville, Ocean View, the DE/MD Beaches and the lower DelMarVa are gonna be the winners, eh? 

I hope MillvilleWx gets 70 inches of snow from this. 

[AlexD1990]

9 minutes ago, dailylurker said:

I'm headed down there on Wednesday morning or before it starts. Lots of places to stay. I'm thinking oceanfront. 

Snow on the beach/boardwalk in Rehoboth is a special experience. We drove into town for the last storm, and the whole city was beautiful. There's a big pile of sand by one of the beach entrances that I climbed with my dog at the height of the storm, and you could see all of Rehoboth Ave, the bandstand etc lit up with heavy snow falling. Purple parrot restaurant does snow parties too, fun locals vibe 

7 minutes ago, MillvilleWx said:

You guessed it! 

I'm still nervous that even we get shafted with another shift tho.

[psuhoffman]

34 minutes ago, SnowenOutThere said:

When do you think the models may start to reflect that (if they do at all)? Additionally, does the tilt of the upper level trough not matter in this case?

It does but it’s unusual for a closed low that far northwest of that amplitude to produce a wave that escapes south of us. There is a feedback to this. With a closed upper level low they far NW the surface should amplify enough to feedback with some height rises in front.  There is a reason there are no examples of a big SE VA snowstorm with this look.  None. Zero zip. 
 

on the other hand reliable upper level data only goes back so far. Maybe it’s a once a century type thing and we’re about to see the first instance of it. 

[Terpeast]

1 minute ago, psuhoffman said:

It does but it’s unusual for a closed low that far northwest of that amplitude to produce a wave that escapes south of us. There is a feedback to this. With a closed upper level low they far NW the surface should amplify enough to feedback with some height rises in front.  There is a reason there are no examples of a big SE VA snowstorm with this look.  None. Zero zip. 
 

on the other hand reliable upper level data only goes back so far. Maybe it’s a once a century type thing and we’re about to see the first instance of it. 

One would think models would start correcting more NW for that reason, but they’ve been going the other way instead. Maybe it’s still too soon for a NW correction.

[mikem81]

1 minute ago, psuhoffman said:

It does but it’s unusual for a closed low that far northwest of that amplitude to produce a wave that escapes south of us. There is a feedback to this. With a closed upper level low they far NW the surface should amplify enough to feedback with some height rises in front.  There is a reason there are no examples of a big SE VA snowstorm with this look.  None. Zero zip. 
 

on the other hand reliable upper level data only goes back so far. Maybe it’s a once a century type thing and we’re about to see the first instance of it. 

The ULL actually

appears to consolidate earlier at better on the more recent EURO runs. Don’t really understand how this is translating into the offshore results. I can see the GFS result with the ULL never

really getting its act together until it’s too late but the EURO h5 seems weird to produce what it is showing 

[aldie 22]

3 minutes ago, Terpeast said:

One would think models would start correcting more NW for that reason, but they’ve been going the other way instead. Maybe it’s still too soon for a NW correction.

I mean they've gone from gangbuster to nutbuster in one run who knows 0z tonight could be rocking 

[mikem81]

Just now, mikem81 said:

The ULL actually

appears to consolidate earlier at better on the more recent EURO runs. Don’t really understand how this is translating into the offshore results. I can see the GFS result with the ULL never

really getting its act together until it’s too late but the EURO h5 seems weird to produce what it is showing 

If someone can look at h5 from yesterdays 12z euro and compare to todays and somehow explain why the surface is so Much worse today is really appreciate it. 
 

https://www.tropicaltidbits.com/analysis/models/?model=ec-fast&region=us&pkg=z500a&runtime=2025021512&fh=120

https://www.tropicaltidbits.com/analysis/models/?model=ec-fast&region=us&pkg=z500_vort&runtime=2025021612&fh=96

[Terpeast]

2 minutes ago, mikem81 said:

If someone can look at h5 from yesterdays 12z euro and compare to todays and somehow explain why the surface is so Much worse today is really appreciate it. 
 

https://www.tropicaltidbits.com/analysis/models/?model=ec-fast&region=us&pkg=z500a&runtime=2025021512&fh=120

https://www.tropicaltidbits.com/analysis/models/?model=ec-fast&region=us&pkg=z500_vort&runtime=2025021612&fh=96

On both, go back 24 hr. You’ll see that the SS sw doesn’t dig as much as before

[adelphi_sky]

9 minutes ago, psuhoffman said:

It does but it’s unusual for a closed low that far northwest of that amplitude to produce a wave that escapes south of us. There is a feedback to this. With a closed upper level low they far NW the surface should amplify enough to feedback with some height rises in front.  There is a reason there are no examples of a big SE VA snowstorm with this look.  None. Zero zip. 
 

on the other hand reliable upper level data only goes back so far. Maybe it’s a once a century type thing and we’re about to see the first instance of it. 

A once in a century rug pull. That sounds about right for the area. lol

[psuhoffman]

26 minutes ago, WEATHER53 said:

Thanks however I read this as “it is but it isn’t”. More  ambiguity. 

Respectfully I didn’t mean to be ambiguous. First of all in this setup some criticism is warranted. The models were pretty bad across the board regardless of outcome on this event.
 

But in general what I mean is this… we know we don’t have the ability to model the atmosphere completely correctly. We have not come close to completing that code. In some cases we know where we are lacking and write bias corrections into the math to try to compensate for our insufficient physics.   

So we have multiple model simulation tools like the euro gfs uk gem that all try their best to model the atmosphere. We have their ensembles where we run permutations to see a spread of likely errors.  Then we use all of them together holistically to try to clean clues to what’s happening. No one of them is meant to be a rip and read forecast.  If you look at each model run at long range only by itself they are next to useless.  If you take all of it together and look for trends and clues factoring in all the model evidence together they are more useful (still not perfect) at being one piece of making a forecast. 
 

Even then they shouldn’t be the whole piece. Look at the crusade I’m on right now where the models are spitting out results that don’t align with my expectations based on the larger pattern drivers so I’m skeptical of their output. You should factor that stuff in also!  Analog guidance and knowing your climo is part of the forecast puzzle as well. 

[WxMan1]

 

What irks me about all of this is here we are, 2025, and the global models still struggle consistently locking in with a system for 5+ days out during the winter. At least, one that's going to involve snow for the DMV. 

 

Rain storms April through November? No problem. It always seems that the models are generally in better agreement. Whether It's split flow, northern and southern stream contributions, amplified waves or not. Same sampling issues, or lack thereof, that we have with winter storm tracks here in the East. Can't figure it out, unless it's just anecdotal on my part.

This is where I thought AI would help, and maybe to some extent it is. But even the EC AIFS has bounced around with solutions between days 4-7.

 

[TSSN+]

18 minutes ago, mitchnick said:

AI east again

 

[frd]

1 minute ago, WxMan1 said:

 

What irks me about all of this is here we are, 2025, and the global models still struggle consistently locking in with a system for 5+ days out during the winter. At least, one that's going to involve snow for the DMV. 

 

Rain storms April through November? No problem. It always seems that the models are generally in better agreement. Whether It's split flow, northern and southern stream contributions, amplified waves or not. Same sampling issues, or lack thereof, that we have with winter storm tracks here in the East. Can't figure it out, unless it's just anecdotal on my part.

This is where I thought AI would help, and maybe to some extent it is. But even the EC AIFS has bounced around with solutions between days 4-7.

 

Seems winter is the hardest season for model accuracy. Is that correct?  

[Terpeast]

3 minutes ago, Terpeast said:

On both, go back 24 hr. You’ll see that the SS sw doesn’t dig as much as before

Otoh, NAM doesn’t dig that much either but is way more north. So there’s that…

[Jebman]

Damn Lucy orchestrating a once in a century rug pull.

[psuhoffman]

34 minutes ago, Amped said:

Still not a great trend since 12z yesterday, the southern stream looking flatter and more progressive. We need the opposite. 

Hmm if the trend continues the h5 will end up looking closer to the southern snow look. Maybe I’m wrong in expecting the surface to correct and it’s the h5 that will. That’s rare. Usually it’s the other way around. 

[adelphi_sky]

3 minutes ago, psuhoffman said:

Respectfully I didn’t mean to be ambiguous. First of all in this setup some criticism is warranted. The models were pretty bad across the board regardless of outcome on this event.
 

But in general what I mean is this… we know we don’t have the ability to model the atmosphere completely correctly. We have not come close to completing that code. In some cases we know where we are lacking and write bias corrections into the math to try to compensate for our insufficient physics.   

So we have multiple model simulation tools like the euro gfs uk gem that all try their best to model the atmosphere. We have their ensembles where we run permutations to see a spread of likely errors.  Then we use all of them together holistically to try to clean clues to what’s happening. No one of them is meant to be a rip and read forecast.  If you look at each model run at long range only by itself they are next to useless.  If you take all of it together and look for trends and clues factoring in all the model evidence together they are more useful (still not perfect) at being one piece of making a forecast. 
 

Even then they shouldn’t be the whole piece. Look at the crusade I’m on right now where the models are spitting out results that don’t align with my expectations based on the larger pattern drivers so I’m skeptical of their output. You should factor that stuff in also!  Analog guidance and knowing your climo is part of the forecast puzzle as well. 

This is where machine learning can help. But the amount of weather data involved would require a HUGE learning model. It would also be interesting why we more often see snow on a model and wake up to no snow more often than we see no snow and wake up to a warning level event. I guess no precipitation is easier to forecast than none? 

[mitchnick]

4 minutes ago, psuhoffman said:

Hmm if the trend continues the h5 will end up looking closer to the southern snow look. Maybe I’m wrong in expecting the surface to correct and it’s the h5 that will. That’s rare. Usually it’s the other way around. 

Completely losing the ridging out front.

[aldie 22]

7 minutes ago, psuhoffman said:

Hmm if the trend continues the h5 will end up looking closer to the southern snow look. Maybe I’m wrong in expecting the surface to correct and it’s the h5 that will. That’s rare. Usually it’s the other way around. 

White flag moment?

[mikem81]

Bernie Rayno had decent video. Essentially blaming the energy entering the west coast kicking the ULL formation east of yesterdays runs 

[Weather Will]

WB 15Z SREF

[mitchnick]

JB update

This is really pretty interesting. The better the upper air pattern goes with the negatives coming right through the slot, the further out the model pushes the storm. Well at the very least if it’s right I’ll get to see something that I can’t recall seeing. That kind of negative should produce an I 95 blizzard DC to Boston So the call from February 11 remains if I have to change it I’ll change it on Tuesday because until this feature comes out of the Rockies I’ve said too many of these flip flop around. If you’re standing on the ledge don’t jump off yet 

Last edited12:50 PM · Feb 16, 2025 from Pennsylvania, USA