MGorse

4 hours ago, kickingupastorm said:

Winter Storm Watch down graded to a Winter Weather Advisory. What a joke. Ardmore, Montgomery Country.

Technically it is upgraded. Watches cannot be downgraded, just upgraded or cancelled. Just FYI.

1 minute ago, Kevin Reilly said:

Thought blue areas were requiring 4” T remembering from distant past off the top of my head?

The blue area is 5 inches. Increased from 4 inches several years ago. The red area is 6 inches. This is for warning criteria. 

10 minutes ago, yankeex777 said:

Does kuchera take into account cold temps aloft?

I came across this explanation: The Kuchera ratio can be especially useful when temperatures are close to freezing, as it can correctly reduce snowfall estimates below 10:1. However, the Kuchera ratio is not perfect, as the true snow-to-liquid ratio (SLR) depends on many factors, including cloud and precipitation physics.

7 minutes ago, anthonyweather said:

Hrrr is way north

Quite the wide swath of snow as well, although it is still showing a zone of heavier snow near and south of Philly. 

3 minutes ago, penndotguy said:

Nope I’m not doing it, outside of 5  days, ahh who am I kidding I might peek at it 

You know you want to look at it. 

25 minutes ago, MillvilleWx said:

I believe the GFS runs on the FV3 core and is fairly hi-res compared to what it used to be. RGEM runs at a resolution similar to the NAM with the HRDPS the Canadian version of the hi-res 2.5km, although it has some significant issues with a multitude of parameters. This is a great question for the MDL people since they have more of the background. 

From what I have read, the NWS chose the FV3 as the new GFS's (and GEFS) dynamical core in part because it uses less computer resources than other options.

Just measured 0.4 inches. Just some flurries now and temperature down to 31. 

Snowing pretty good here and it did not take long for the grass, mulch and roofs to get a coating of snow. Temperature down to 32. 

12 minutes ago, RU848789 said:

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.  

I had to search some but hopefully this is what you are looking for…

https://vlab.noaa.gov/documents/6609493/7858320/NBMv4.1NWPMatrix.pdf

2 hours ago, RU848789 said:

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.  

The NBM is time lagged. If there is a notable pattern change in the guidance and even a trend then the NBM will take time to catch up. Closer in time, the NBM incorporates the high resolution guidance which can lead to a noticeable change in the output. 
 

More info about the NBM is available here: 

https://vlab.noaa.gov/web/mdl/nbm

27 minutes ago, Ralph Wiggum said:

We were about to get NAM'D good but the run ended

 

Just extrapolate it. 

5 minutes ago, Kevin Reilly said:

Looks like a bit of frontogenesis forcing hmm... could be a quick 3" pop for places in Delaware County centered on Newtown Square to Media.

Yup something to watch as the NAM has an area of 850 mb FGEN strengthening some as it shifts east. 

7 minutes ago, tristateweatherFB said:

You should hire taller people. 

I see what you did there. 

3 minutes ago, wdrag said:

Sure.  You can do all the Severe.  My fingers don’t work as quickly as yours and my severe detection is not up to snuff   Keep me posted 

And you were not the only one at PHI to push the envelope on alerting. We have made some progress on that but still work to be done. Not an easy task either when staffing keeps changing. 

1 minute ago, RU848789 said:

Enough said, thanks.  My one comment for the author of that AFD would be to perhaps address the NBM/precip inconsistency, especially given how often the NBM is invoked by the NWS as a major aid in forecasts.  

That inconsistency has been noticed by some NWS field offices, but I am not aware of where that stands. In my opinion if something like this NBM stuff is referenced in things like AFDs then some added info should be included to provide more context of that data. The deterministic NBM is bias corrected (60 days I believe) which results in a lag especially during pattern changes. Still not all that sure why a blend of models has a bias correction applied. 

2 hours ago, RU848789 said:

Interesting that the NBM snowfall forecast bumped up for 07Z vs 01Z last night and yet the NWS-Philly is clearly downplaying the event potential north of Philly, as per the part I bolded below, when they often follow the NBM for snow events.  Maybe they're going by the WPC precip forecast and not the NBM snowfall forecast...

@MGorse- any comments on this apparent disconnect?  Just trying to figure out what you guys are thinking, thanks...

National Weather Service Mount Holly NJ

401 AM EST Thu Jan 2 2025

LONG TERM /SUNDAY THROUGH WEDNESDAY/...

Guidance from a synoptic standpoint is depicting a Miller B
nor`easter to develop with an upper level trough swinging a
surface low from west to east across the KY/TN area before
redeveloping as a low pressure system offshore Monday. Ensemble
guidance is split into 2 different clusters with the variance
generally explained by how strong the upper level ridge will be
over Quebec. This boils down into two scenarios, one with a more
northern track bringing snow towards the RDG/TTN/PHL, and one
with a more southern track keeping the snow more over the DC
area. The more northern stream track relies on a weaker high and
thus develops a bit more cyclogensis over our region thus
lifting the band of 2-4 inches of snow further north, whereas in
the southern track, the high is stronger and displaces the
surface low to the south more leading to most of the area seeing
little in the way of snow.

01Z/02 NBM generally is taking a blend of the two scenarios thus
leads to probabilities similar to the 13z/01 NBM with snowfall
greater than 1 inch is 60 to 80 percent across southern New
Jersey, the Delaware Valley including Philadelphia, and the
eastern shores of Maryland and Delaware while the probability of
snowfall greater than 3 inches is 40 to 50 percent. North of
Philadelphia, probabilities lower to 30 to 50 percent for
greater than 1 inch of snow and 20 to 30 percent for greater
than 3 inches of snow.

For the time behind will continue to show a blend of the two
tracks but anticipate that the gradient of snowfall will sharpen
quickly somewhere north of Philadelphia over the coming runs
depending on how guidance handles the upper level ridge. While
it is still too soon to tell how much snow will fall, the trend
looks to be for advisory level snowfall for Delmarva, sub-
advisory amounts for southeast New Jersey and the Delaware
Valley, and minimal amounts north of Philadelphia.

 The NBM is not the greatest guidance many times and I will leave it at that.

1 hour ago, wdrag said:

Agree 100%.  NWS general field office approach differs from NHC.  It's definitely not my approach to well advertised multi modeling.  

NWS was pretty happy to see me go, at least at field level where I was generally pushing the envelope to get alerting going. I thought LWX was advanced in leading these winter situations as NHC in tropical. 

Automation will end this personal - institution conservative approach...  it has to be coming as progress is made.  These AMWX discussions will become moot I think in a few years, at least inside 7 days...  presuming the monies are available for research, platform improvement and processing.  Then I can sleep later.

Right now, reliably accurate interpreting voices are needed to get planning considerations on the table, sooner than saying nothing. I

You want to come back? We are still short staffed! 

1 minute ago, RedSky said:

Hopefully the line intensifies and the ball thing blows away and becomes a drone

Perhaps the ball would then be reported as a massive hailstone. 

Nothing like a Spring evening thunderstorm, oh wait it is 12/31.  

Happy New Year everyone! 

It is like a Spring night out there.

2 hours ago, JTA66 said:

Ah, I see what you did there…the old, double-reverse jinx 

Umm, sure. 

46 minutes ago, Mikeymac5306 said:

Holy Hanna 18Z GFS!!!! 

The 00z run will probably have nothing. 

Just lovely out there this morning. 
 

I hope everyone had a nice Christmas. 

14 minutes ago, RU848789 said:

Any reports from Philly, proper, yet?  Haven't noticed any in this thread.  Thanks.  

Have not really seen any yet.

12 minutes ago, BBasile said:

Why bother asking?

Because you said 'heavy snow' and I was curious about the visibility. Just wanted to double check. Carry on...