January Medium/Long Range: A snowy January ahead?

[mitchnick]

1 hour ago, Heisy said:

Nice hit on the euro Ai for around the 20th @mitchnick


.

18z AI looks completely different around that time frame this run than the 12z run. Like I said, it's jumpy too post 240hrs.

[stormtracker]

GFS kinda interesting at 228

[stormtracker]

50/50 in almost the perfect spot for the 20th and watch it cuts.  Anyway, signal is there.  Models have all kinds of ways to get there

[stormtracker]

Verbatim...rain.  Wish we had that H5 look for this weekend

Thank God it's gonna be different with the 6z run

[Ji]

50/50 in almost the perfect spot for the 20th and watch it cuts.  Anyway, signal is there.  Models have all kinds of ways to get there

What are you doing?

[stormtracker]

Man...moisture bomb.  Like i said, verbatim, shitty rainstorm, changing to frozen at the end.  Get those temps right tho and it's a beatdown.  Tracks over nyc

[WEATHER53]

Looking like double digit negative departures for at least another week.  

[stormtracker]

2 minutes ago, Ji said:

What are you doing?

Looking for a snowstorm in fantasy land.  It's not like we're chasing anything else.

[mitchnick]

1 minute ago, stormtracker said:

Man...moisture bomb.  Like i said, verbatim, shitty rainstorm, changing to frozen at the end.  Get those temps right tho and it's a beatdown.  Tracks over nyc

Accumulating snows with it.

[mitchnick]

It's credible this year because it screws NE.

[stormtracker]

2 minutes ago, mitchnick said:

Accumulating snows with it.

yup..4-8

[mitchnick]

1 minute ago, stormtracker said:

yup..4-8

That's our MLK surprise.

[Ji]

Looking for a snowstorm in fantasy land.  It's not like we're chasing anything else.

Pbp at 11 days out. You’re going to be wrecked by Tuesday

[stormtracker]

Just now, Ji said:

Pbp at 11 days out. You’re going to be wrecked by Tuesday

It was 4 posts

[bncho]

1 minute ago, stormtracker said:

yup..4-8

Just your average backend 4-8 inches of snow from a cutter after 1-2 inches of torrential 35 degree rain.

[Bob Chill]

1 minute ago, stormtracker said:

It was 4 posts

The guy who's wrecked by Columbus day is worried about your 4 posts and the hidden meaning behind them, friend. 

[Terpeast]

This 6-10” storm we just had was a cutter and rain for us 10-ish days out. 

[psuhoffman]

47 minutes ago, Bob Chill said:

It's pretty interesting really. If we stay cold d10+ it will be for different reasons. So we currently have a pretty epic/classic block period that looks to relax into *potentially* a +AO and deeply negative EPO. It's like a highlight reel of the large scale longwave patterns that bring us cold temps and maybe ok snow lol. 

Historically, a progressive -EPO isn't something we want for snowfall but the 2013-15 stretch is either a clue that it's more favorable nowadays or it was lucky AF and a return to regularly scheduled programming of warm/wet -> cold/dry is more likely lol. 

I think it depends on other variables. An epo alone still won’t do it. Especially if the epo is linked to a wpo ridge with a full latitude central pac ridge. That leads to too much SER. The key to 2014 and 2015 were other variables going our way. In both the pacific had a trough under the ridge. We have that now although super long range guidance wants to establish a canonical Nina look but I don’t buy it until it gets inside 15 days. It’s been stuck at day 17-20 all year. 
 

2014 there was a ridge from the epo over the pole that depressed the trough in the east. 2015 a TPV lobe got stuck in Quebec doing the same. Those were critical factors in making those years work.  We seem to have some of those same factors on long range guidance. My interest in the pattern goes beyond the epo but there are other more critical similarities to 2014-2015 as well. Those matter more imo than the epo.  

[psuhoffman]

9 hours ago, stormtracker said:

Verbatim...rain.  Wish we had that H5 look for this weekend

Thank God it's gonna be different with the 6z run

The storm that just fringed me to the south was rain on all guidance at this range. 

[IronTy]

JB says MJO flips and goes hostile for the East in wks 3/4 and stays that way through February.  Looks like winter's almost over. With some luck I'll get to climo if I can get another inch or so out of this weekend event.  

[psuhoffman]

Just now, IronTy said:

JB says MJO flips and goes hostile for the East in wks 3/4 and stays that was through February.  Looks like winter's almost over. With some luck I'll get to climo if I can get another inch or so out of this weekend event.  

Why should anyone care what that clown thinks. He is a joke. I’m glad he thinks it’s gonna be bad. 

[Heisy]

6z euro Ai, keeps up the signal for around the 20th. As is it verbatim it’s a front end thump to mix or rain, with a huge hit for N&W, but it wouldn’t change much to make this a big dog for everyone. Better push from N/S ahead of the main low. Could also trend worse or all be make believe at this range, but the euro Ai has been pretty damn good at picking up storm threats







.

[IronTy]

15 minutes ago, psuhoffman said:

Why should anyone care what that clown thinks. He is a joke. I’m glad he thinks it’s gonna be bad. 

To be fair he's basing that off the JMA forecasting phases 4/5.  Assuming the JMA is right I agree with him that it'll be hard to overcome that hostile a regime.  Maybe we get something big just as it flips, or maybe the JMA is wrong and it doesn't flip.  We'll know in a couple weeks.  

[Terpeast]

29 minutes ago, IronTy said:

JB says MJO flips and goes hostile for the East in wks 3/4 and stays that way through February.  Looks like winter's almost over. With some luck I'll get to climo if I can get another inch or so out of this weekend event.  

Not so fast... I just saw a new trade wind plot and it appears that trades might want to push W of the dateline well into the eastern half of the MC. If this pans out, that will suppress convection over the MC and keep it westward into the IO (phases 2-3) with a little forcing over the eastern pacific. This is the kind of "split forcing" we've been getting almost the whole winter rather than the dreaded 4-5-6 warm phase forcing.

[Terpeast]

30 minutes ago, Heisy said:

6z euro Ai, keeps up the signal for around the 20th. As is it verbatim it’s a front end thump to mix or rain, with a huge hit for N&W, but it wouldn’t change much to make this a big dog for everyone. Better push from N/S ahead of the main low. Could also trend worse or all be make believe at this range, but the euro Ai has been pretty damn good at picking up storm threats







.

For those here who don't know or aren't sure (or need a reminder)... AIFS is deterministic AI model that uses the operational ECMWF as initial conditions. It is not an ensemble model, nor an ensemble member, which is pretty amazing since ensembles tend to perform better at long leads. AIFS might be picking out good "analogs" and extrapolating what happens after each analog as a composite, and doing an excellent job at that.

[frd]

https://x.com/burgwx/status/1877356736488652849

 

Tomer likes late Jan for Mid Atlantic snowstorm potential 

 

[cbmclean]

1 minute ago, frd said:

https://x.com/burgwx/status/1877356736488652849

 

Tomer likes late Jan for Mid Atlantic snowstorm potential 

 

I'm not a fan of the PR regime in general.  Maybe it's not as hostile if there is a trough underneath the ridge.

[Terpeast]

3 minutes ago, cbmclean said:

I'm not a fan of the PR regime in general.  Maybe it's not as hostile if there is a trough underneath the ridge.

I checked the EPS weekly clusters for Jan 23-Feb 7 (sorry, can't share or I'll get in trouble), and 60% does have a pacific ridge, but with some extension northward into western AK, and some troughiness into the midwest and east. It's not as bad a look as we might expect. 40% is SE ridge.

[LordBaltimore]

13 minutes ago, Terpeast said:

For those here who don't know or aren't sure (or need a reminder)... AIFS is deterministic AI model that uses the operational ECMWF as initial conditions. It is not an ensemble model, nor an ensemble member, which is pretty amazing since ensembles tend to perform better at long leads. AIFS might be picking out good "analogs" and extrapolating what happens after each analog as a composite, and doing an excellent job at that.

I hope for my sake and the sake of a lot of other weather peeps, that the people in charge don't forget that these things are still dependent on physical modeling systems and will be for a very very long time. It's not "AI magically takes raw data and transforms it into forecasts", it's more like the numerical modeling system does the data assimilation - using an ensemble - and then the analysis is used as input for a ML model trained off of ERA5 (also produced with a numerical modeling system). The quality of AI forecasts directly scales based on the quality of the numerical model. 

[mattie g]

1 minute ago, LordBaltimore said:

I hope for my sake and the sake of a lot of other weather peeps, that the people in charge don't forget that these things are still dependent on physical modeling systems and will be for a very very long time. It's not "AI magically takes raw data and transforms it into forecasts", it's more like the numerical modeling system does the data assimilation - using an ensemble - and then the analysis is used as input for a ML model trained off of ERA5 (also produced with a numerical modeling system). The quality of AI forecasts directly scales based on the quality of the numerical model. 

Of course. It's the same for any AI/ML system or use case. What comes out of it is only as good as the data that goes in.