RU848789

Very interesting that the NWS bumped snowfall back up for areas N of 276/195, including CNJ/NNJ/NEPA/NYC, with their 4 am forecast, as per the map below; it's fairly similar to the Channel 7 forecast. Their 1" line moved from just south of 78 to about 80, except for NENJ (Union/Essex/Hudson) and NYC/LI, where <1" is forecast, and their 2" line is now from about Frenchtown to Sea Bright and the 3" line is along 276/195 (was 10-15 miles south of there); south of that 3" line very little change. The NWS discussion doesn't have the usual "model discussion" section at 4 am. In contrast, TWC has the 1" line along 276/195 with <1" north of there and the 3" line from about Wilmington to AC, as do others, while News12NJ is between the two camps with their 1" line from about Frenchtown to Sea Bright and their 3" line from Philly to LBI with 3-6" south of there. Bust potential is pretty high, IMO, as the global models at 0Z last night and 6Z this morning continue to show much less snow than this across the board, as all of you know, with generally <1" N of 276/195 and even only an inch or two for the Phillly/SNJ area; they must be leaning more towards the meso/CAM models, like the NAM3K, HRRR, RAP, SRERs, etc., which show amounts similar to the NWS map (or even more). From a meteorological perspective, this is a fascinating conundrum to have these two camps so far apart, but from a "client" perspective it's very frustrating to have such low confidence in the NWS forecast.

Long range RAP anyone? Hey, if we're going to grasp at straws, at least let them be good ones, like this and the long range HRRR, lol.

Yes it does. Nice to at least have the NAM/GFS and even the long range HRRR on the snowier side; maybe some of the others will want to join the party at 18Z.

While this is obviously fantasy land as it contains a 2nd storm 14 days out, the first storm is now just 7 days away, which is transitioning from fantasy land to something that could have potential, as per the 24 hr snowfall map below - not that we'll get that much snow, but that perhaps there's a decent chance of a significant snowstorm next weekend.

Can you elaborate? Is this related to the conversation Walt, Mike Gorse and I were having over precip/snowfall discrepancies and the time-lagged NBM issue? I can't tell what you've highlighted. Thanks.

SIAP (been playing at a disc golf tourney all day), but the NWS updated their regional snowmap (I love these as they show Mt. Holly and nearby adjacent areas). Clearly backed off for snow north of 276/195 in EPA/CNJ, especially N of 78 which is <1" but beefed snowfall up a bit for Philly/SNJ and south of there. Still lots of uncertainty and bust potential in both directions.

NAM moved a bit north not a lot (same for the NAM3K), but the bigger thing is it's the snowiest model and it did improve. ICON looks the same at 18Z and 12Z, but the HRRR, which did well with the two Dec storms, but did terribly with the SNJ storm yesterday, is pretty snowy. The RGEM is slightly less snowy with very little snow for Philly-AC even and northward. SREFs look great as always, lol.

wow, even Philly gets screwed with essentially nada

it's probably over for more than 2" for CNJ, but I'd still be happy with 1-2". There's still time for minor shifts back north to make that happen, as we're not that far off. CMC still has 1-2", GFS/RGEM have an inch, and NAM/ICON have an inch or so for the southern half of CNJ, plus there's the 2-4" on the SREFs. If we hadn't seen snowier solutions over the past few days and were magically transported to where we are now with no knowledge of the past, we might look at where we are now and say, hey, some minor to moderate changes get CNJ and maybe even NYC into at least a couple of inches of snow. The models don't "remember" previous runs and are independent events with new data every time (as far as I know).

Did you see the key part of the Mt. Holly AFD where they talk about their reasoning for going bullish with the snow for CNJ/NNJ? Wish they'd come on one of these boards and clearly show what they mean by "looking at the overall atmosphere aloft including jet dynamics and mesoscale features." The biggest challenge that still persists at this time is with regard to the track of the surface low and where it will be as it crosses through the area. While the majority of ensemble guidance has shown a southern shift over the past 12-24 hours with QPF, looking at the overall atmosphere aloft including jet dynamics and mesoscale features, it seems that the overall consensus of the track seems to be too far south. Depending on the strength of the system as it approaches and confluence that lingers over the Northeast US, will simply determine how far north the snow shield reaches. Have started to see some waffling back and forth with the 12Z guidance, so don`t completely buy the entire southern shift. Unfortunately, there is still a lot of uncertainty amongst global guidance, where the greatest potential for higher-level impacts remains over the southern-half of the forecast area.

Surprised at this, but more hopeful now, but also a bit skeptical...we'll see soon...see the part in bold, in particular for amounts, but to summarize here: 1-2" N of 80, 2-4" from 80 down to 276/195, 4-6" south of 276/195 with lollipops to 6-8 where the best banding sets up in SEPA/SNJ. I could certainly live with this. Edit: here's the NWS map for NWS-Philly and parts of adjacent offices. Nice. Area Forecast Discussion National Weather Service Mount Holly NJ 335 PM EST Fri Jan 3 2025 .SHORT TERM /SATURDAY NIGHT THROUGH MONDAY NIGHT/... Our attention then turns to the winter storm potential that will impact the Mid-Atlantic region late Sunday night through Monday, possibly lingering into Monday night. A potent shortwave ejecting out of the Rocky Mountain region will develop into a closed low as it slides across the Central Plains. This closed low will undergo cyclogenesis and lead to a surface low developing while tracking through the Central US on Sunday night. The surface low will continue to track due east crossing somewhere between the southern Delmarva and eastern North Carolina by Monday afternoon. The biggest challenge that still persists at this time is with regard to the track of the surface low and where it will be as it crosses through the area. While the majority of ensemble guidance has shown a southern shift over the past 12-24 hours with QPF, looking at the overall atmosphere aloft including jet dynamics and mesoscale features, it seems that the overall consensus of the track seems to be too far south. Depending on the strength of the system as it approaches and confluence that lingers over the Northeast US, will simply determine how far north the snow shield reaches. Have started to see some waffling back and forth with the 12Z guidance, so don`t completely buy the entire southern shift. Unfortunately, there is still a lot of uncertainty amongst global guidance, where the greatest potential for higher-level impacts remains over the southern-half of the forecast area. With all this in mind, the forecast this afternoon features a solution that incorporates warning level snowfall for areas around Philadelphia and points south. The highest snowfall amounts will end up being located within a frontogenetic band where SLRs will be locally higher under an area of enhanced snowfall. The confidence is rather low on where this FGEN band will set-up over the region though. Based on latest 700-850 mb frontogenesis progs, the greatest likelihood will be somewhere between the Philly metro and central Delaware. In fact, some of the hi-res guidance that has come in this afternoon shows a warm nose between 850-925 mb that creeps northward during the day on Monday into portions of the Eastern Shore of Maryland and southern Delaware. This would result in some potential mixing issues in the form of sleet and/or freezing rain which will drastically undercut snow totals, with the heaviest snow totals located just north of the mixing line. Upon discussion with neighboring offices and given the overall uncertainty, have opted to not issue any Winter Weather headlines with this forecast package. In terms of snowfall amounts, our initial forecast is for a 4-6 inch snowfall event for areas on a line from I-76 to I-195 and points south, including the Philadelphia metro. Within this area, will be a localized maxima of 6-8+ inches depending on where the frontogenetic banding sets up. Between the I-80 and I-76/I-195 corridor, expecting totals in the 2-4 inch range with totals less than 2 inches to the north of I-80. There will be a sharp northern gradient to the snowfall due to confluence and dry air off to the north, so some of these totals may be slightly too high.

Thanks!! I tried looking through that website before and just now and maybe I'm just dense, but I couldn't find anything on the time lag nature of the inputs into the NBM, but I'll certainly take your word for it. A time lag definitely explains how the NBM output at a particular time can look more like the output from the models from 6 hours earlier rather than the models that run somewhat concurrently with that time, assuming that the NBM is not including those concurrent runs. And yes I do realize the shorter term/high res guidance does start getting incorporated as one nears an event.

Just a minor shift south over the last 18 hours for the ICON.

A couple of folks on 33andrain, including one met, have said that the NBM is time lagged, i.e., the 07Z NBM would only include model inputs through 0z and not the 06Z models. That would at least explain why the NBM snowfall for 07Z is greater than the 01Z NBM for areas N of Philly and much greater than the 06Z models would suggest, since they're generally much lower than the 0Z models were. To me, that makes the NBM a much less useful tool, if that's correct, as it would always be 6 hours behind. @MGorse- sorry to tag you again, but can you clarify if this is the case, i.e., what time model inputs go into a particular NBM run? I'd also be curious (see the quoted post) if the 12Z WPC QPF output is simply the 07Z NBM QPF field (they're identical), which is based on 0Z model data - that seems to be very old data, then.

Would love to know how the NBM increased snowfall north of 276/195 up through 80 and even 84 at 07Z vs 01Z when every single model I have access to (GFS, Euro, ICON, and the ensembles for these) showed a significant decrease in snowfall for this area at 06Z vs 12Z. Only one model at 6Z, the ICON, has NYC at >1" (and the 0Z UK and CMC also had NYC at <1"), yet the NBM has NYC going from 2.4" to 3.5" from 01Z to 07Z. @wdrag? I know we had a conversation yesterday morning about inconsistencies with the NBM (between its high snowfall vs. the WPC low precip at the time), but this seems even worse. Edit: well, the 07Z NBM clearly thinks there will be 0.2-0.3" QPF for NNJ/NYC, as its QPF estimate is identical to the latest WPC QPF from 12Z today, below...

Ya never know. My guess is 75% of the area under advisories gets <1", but that a few locations that get sustained banding get up to 2", but those are very hard to predict.

Wasn't sure where to post this HRRR comparison of 18Z and 0Z for tomorrow's little storm in SNJ/SEPA/DE, but in the vein of model mayhem, here's reason #993 to not trust the HRRR beyond 24 hours, lol...

Every pro here and on 33andrain and elsewhere is scratching their heads over what to make of all of this now that the Euro joined the UK as misses for our forum, meaning I have no shot of figuring anything useful out at this time. Even using the NBM Model Blend is probably a bad idea when there are two very distinct camps like this. One can blend them and get a moderate snowfall for our area, but the outcome is just as likely now to be one camp or the other. FWIW, the NBM is about the same as it was at 18Z. I don't envy forecasters for this one. At this point I would probably just punt and wait for better model consensus and use the NBM which still says 1-3"/2-4" for CNJ (less north and more south), but I'd note that <1" (or even nada) and 3-6" are still on the table. Crazy.

I still like 1-3/2-4" for CNJ, which is what we're still seeing for the NBM, which actually got a little snowier for CNJ at 01Z vs. 19Z. Of course, a blend will miss if one of the outliers is on to something and can be terribly wrong, but major outliers rarely verify, so it'll usually at least be in the ballpark, unless we see some huge shift by all the models, which has never happened before 3+ days before an event, lol.

Reason #993 to not trust the HRRR beyond 24 hours, lol..

Just when I thought we were going to have consensus on a snow party with the ICON/GFS, the CMC brings us back to reality and then the UK impersonates Dr. No.

Holy crap, we just got NAM-ed by the 0Z GFS, lol. Stepped away for 30 min and boom!

Why? It was an outlier at 18Z north, so coming back a little bit south is no surprise; it's still likely going to be the furthest north of the 0Z models.

I'll take that, thanks for guaranteeing that outcome, lol.

11.25" - not a top 10 storm, accumulation-wise, but a top 5 storm, experience-wise, partly because of the surprise element and partly because of the incredible 3-4" per hour rates for most of 2 hours. Different setup and I doubt we'll see that kind of gradient on both sides of a very narrow band.