Discussion-OBS slippery plowable moderate to high impact snowstorm with many delays near I95 and inland btwn 10AM Sun 1/19-4AM Mon Jan 20. R/S line uncertain near and east of I95 - parts of LI may be limited snowfall of around 1"?

[wishcast_hater]

55 minutes ago, RockawayRowdies said:

 

sign me up.

Where can I find this package for Dutchess County? Apparently I am under a Winter Storm Watch for 7" or more.  

 

[MJO812]

[Allsnow]

Wow gfs 

 

 

[Franklin0529]

What a model war!!! Who's gonna win?! 

 

[Franklin0529]

Geez, GFS with a bomb

[EWR757]

55 minutes ago, Roger Smith said:

Is there any easy answer to the question, what is the difference between an "A.I." weather model and the regular models? At first glance, it would see that all weather models are "A.I."  

Tough one...I'll try (correct me if I'm wrong).

"Regular" weather models rely on very complicated and time consuming math and physics to setup an initial state of the atmopshere at each grid point, then compute using physical laws the future state of the atmosphere.

"AI/ML/DL" weather models use the initial state of the atmosphere determined by the above model (I think in all cases the ECMWF dataset) but then instead of using physical laws to determine the state of the atmosphere, similar patterns are found from previous data, and "experiences" of the AI.  Think "analogs", but it is more complicated than that...perhaps "statistical physics" is a better analogy.

The AI model can be run in a matter of minutes on a desktop, whereas the "regular" models need supercomputers and take a long time to run.  The AI model grid spacing is currently very large...100 km or more, so they don't have the detail the "regular" models have, especially in the short-term.  However, in the medium and long-range they seem very promising.

In the past and today, meteorologists know which models behave well and which behave poorly, which models are initialized better than others, etc.  They could perform a scientific analysis based on their understanding of how the model works and their meteorological expertise.  With AI, that is a lot harder if not impossible (I think).

 

https://deepmind.google/discover/blog/graphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting/

 

https://www.nature.com/articles/s41586-024-07744-y

[jdj5211]

[RockawayRowdies]

7 minutes ago, wishcast_hater said:

Where can I find this package for Dutchess County? Apparently I am under a Winter Storm Watch for 7" or more.  

 

https://www.weather.gov/okx/winter

 

 

[Allsnow]

Gfs with another tick east 

[Allsnow]

10-1 

 

6-8 for nyc 

[NJwx85]

GFS has been very steady. I think it has the best handle of all the guidance.

[brooklynwx99]

GFS is a smash hit. PVA is oriented a hair more zonally but is also more consolidated. equals a colder and stronger solution

[jm1220]

If we could tick the GFS 25 miles SE everyone would win. RGEM-no comment.

[Joe4alb]

see that white spot on the South fork of LI... yeah that's me! F my life  lol

[NJwx85]

1 minute ago, jm1220 said:

If we could tick the GFS 25 miles SE everyone would win. RGEM-no comment.

You've been on here long enough to know that not everyone can win in a setup like this.

 

[NJwx85]

2 minutes ago, Joe4alb said:

see that white spot on the South fork of LI... yeah that's me! F my life  lol

That's what you get for living in the middle of the ocean. 

[ILoveWinter]

So the GFS has an 8 or 9 inch crush job in the City and the RGEM has basically nothing and we are less than 2 days away. Just goes to show you how difficult it is to forecast snowfall for coastal areas of the northeast.

[steve392]

19 minutes ago, BoulderWX said:

What are we looking at timing wise for the onset of the precipitation? 

News 12 had it from around 11am to 9pm.

[Brian5671]

1 minute ago, ILoveWinter said:

So the GFS has an 8 or 9 inch crush job in the City and the RGEM has basically nothing and we are less than 2 days away. Just goes to show you how difficult it is to forecast snowfall for coastal areas of the northeast.

massive bust for one of them coming up.

[Allsnow]

2 minutes ago, ILoveWinter said:

So the GFS has an 8 or 9 inch crush job in the City and the RGEM has basically nothing and we are less than 2 days away. Just goes to show you how difficult it is to forecast snowfall for coastal areas of the northeast.

Don’t forget the euro with just 2-4 

[jm1220]

1 minute ago, NJwx85 said:

You've been on here long enough to know that not everyone can win in a setup like this.

 

There’s no way there can be a GFS like outcome shifted 25-50mi southeast or even the numerous Euro outcomes that would be a solid hit everywhere? I’m as aware as anyone that I could be dealing with a mainly rain event here and the RGEM barely budged but that’s ridiculous.

[Brian5671]

Just now, Allsnow said:

Don’t forget the euro with just 2-4 

maybe that's the compromise-they meet in the middle

[psv88]

GFS has slowly trended to the euro and now brings the 6” snow line well east and into Suffolk county. 
 

RGEM is entirely on its own

[EWR757]

There ain't no way it will be cold enough close to the close for a lot of snow.

For the record (and get this notarized), the 18Z SUN temperature at EWR will be 37 F or higher.   The arctic air is going to take a long time to move southeast with developing cyclogenisis.  

[psv88]

GFS/Euro/ukie/NAM V RGEM

 

 

[jm1220]

2 minutes ago, EWR757 said:

There ain't no way it will be cold enough close to the close for a lot of snow.

For the record (and get this notarized), the 18Z SUN temperature at EWR will be 37 F or higher.   The arctic air is going to take a long time to move southeast with developing cyclogenisis.  

Depends on where the cyclogenesis happens. If it happens over NJ like the RGEM certainly. If it’s offshore like the other models, the N flow will pull the cold air in. It’s an arctic airmass being tapped. 

[weathermedic]

Great AFD from OKX on the upcoming storm:

Then all eyes turn to a wave of low pressure moving across the lower
Appalachians Sunday morning, then ENE to the Mid Atlantic coast
later Sunday. Uncertainty remains high with mid level and the sfc
low track / position, and thus high uncertainty around location
specific temperature profiles. And this leads to high uncertainty
around the derived liquid and frozen precip amounts. One big
question right off the bat is how much sfc cold air can get all the
way into the coastal plain for late Sunday morning as the wave of
low pressure to our south begins to spin up. Therefore some light
rain or a light wintry mix at onset towards mid day / early
afternoon for at least the eastern 1/3, maybe even the eastern 1/2
of the region. Further west profiles look cold enough for mainly all
snow, even at the onset. The higher PoPs and steadiest precip looks
to take place for the late afternoon and into the evening. The
system overall is quite progressive and this will keep liquid
equivalent amounts / QPF in check overall. However, at least half
inch liquid precip amounts are suggested by the majority of the high
res guidance, mainly from 18z Sun to 06z Mon. Thus a complicated
situation with the higher res guidance and even the global guidance
now having more precip over the colder mid level baroclinic zone
with arctic air attempting to press in from the west. The 500 mb
vorticity track is suggestive of a wintry mix and climatologically
speaking when you have the 500 mb vort bisect the region you often
don`t have an all snow scenario for the coast. However, the system
will be quite progressive and that offsets things some as the region
looks to be completely on the cold side of the upper level jet
dynamics. How much the cold air gets in determines precip type
across eastern most areas and snow ratios further inland. This
ultimately will determines liquid and subsequent snowfall totals.
With all this said a Winter Storm Watch has been hoisted for most
northern and northwester potions of the area. Thinking warning
criteria snow is looking more likely across NW portions of the area,
especially into most of the Hudson Valley and interior NE NJ, and
across northern sections of Fairfield, New Haven, and Middlesex
counties in CT for the potential for 5 to 8 inches of snow through
Sunday evening. Elsewhere there is the potential for advisory level
snow, and held off with any watches further east and southeast
across the area with an increasing threat of mixing with rain, or
perhaps just plain rain at the onset before a changeover to snow
towards evening.

The steadier precip shuts off fairly quickly towards midnight, say
(04-06z Mon) as the system moves quickly NE. Thus any precip / snow
tapers quickly for the overnight period. The shortwave feature lifts
further NE later at night and clearing should begin during the pre-
dawn hours, or at least towards daybreak from SW to NE. In any
event, the arctic air completely presses in towards the Monday
morning commute. Thus any moisture on surfaces and roadways will
freeze solid as temperatures plummet mainly into the teens and lower
20s and wind chills likely getting into the single digits to around
10 above.

[BxEngine]

5 minutes ago, EWR757 said:

There ain't no way it will be cold enough close to the close for a lot of snow.

For the record (and get this notarized), the 18Z SUN temperature at EWR will be 37 F or higher.   The arctic air is going to take a long time to move southeast with developing cyclogenisis.  

You can believe that as your opinion, but the gfs is nowhere near 37 at that time. 

[Stormlover74]

So many absolutes for being 48 hours out of things that can or can't happen. Let's remember the gfs had nothing 48 hours ago. The nam nada yesterday

[hudsonvalley21]