January Storm Term Threat Discussions (Day 3 - Day 7)

[CAPE]

5 minutes ago, leesburg 04 said:

Today feels like a "talk ourselves out of it" kinda day

I am still trying to talk myself into the early week event. Its not working. 

[Quasievil]

13 minutes ago, leesburg 04 said:

Today feels like a "talk ourselves out of it" kinda day

Yeah. It's a we've been down this road before kind of thing, right? So many Charlie Brown moments.

[BristowWx]

19 minutes ago, CAPE said:

I am still trying to talk myself into the early week event. Its not working. 

A verified WSW event of 5 non slant stick inches would be all I could hope for at my terdshow location.  Not a lot to ask for. 

[Heisy]

1 hour ago, yoda said:

Where is this from?  I dont see it on here... and i've never heard of that before with any model

Tony, rain shadow. One of the best Philly Mets 

[yoda]

Just now, Wentzadelphia said:

Tony, rain shadow. One of the best Philly Mets 

Seems really weird though... I politely disagree with that assessment

[Ji]

I've seen the gfs give us a big storm at 12z only to take it away at 18z....one of the many reasons I complain last so much

Sent from my SM-G981V using Tapatalk

[Ji]

So ICON went even more strung out....essentially a nothing burger. UKIE is a nothing burger. And Euro is in that camp? Euro needs to find better friends....it doesn't need that sort of negativity.

Icon has a good ull but it's crushed by a strong ull feature in Canada. Gfs had that feature way north

Sent from my SM-G981V using Tapatalk

[frd]

2 minutes ago, Ji said:

I've seen the gfs give us a big storm at 12z only to take it away at 18z....one of the many reasons I complain last so much

Sent from my SM-G981V using Tapatalk
 

It is simply incredible when you look at the GFS and this morning's European even the GFS ensembles and the huge differences at 120 hours between them. Sometimes folks will mention to GFS has been consistent for 3 days in a row but even that means little.  consistency does not necessarily mean accuracy eventually the GFS flipped and went to the European solution I believe this was a few years ago

[Heisy]

1 minute ago, yoda said:

Seems really weird though... I politely disagree with that assessment

It’s not that simple as the above met posted, idk was just throwing it out there. Doesnt  mean the gfs is wrong would like to have the euro/Ukie on board very soon 

 

50 minutes ago, LeesburgWx said:

In December, GFS led the way, right? I am concerned seeing only the GFS and maybe Canadian giving us a big storm. Anyone extrapolating the NAM yet?

Idk up here it was atrocious. Was way too south. So was the euro though at times. There were runs like 3 days out that had Philly getting the max snow with Bgm sniffing virga 

[Quasievil]

Just now, frd said:

It is simply incredible when you look at the GFS and this morning's European even the GFS ensembles and the huge differences I 120 hours between them. 

Well, if either caves more to each other, let's hope they settle on the CMC variant.

[Ji]

It is simply incredible when you look at the GFS and this morning's European even the GFS ensembles and the huge differences at 120 hours between them. Sometimes folks will mention to GFS has been consistent for 3 days in a row but even that means little.  consistency does not necessarily mean accuracy eventually the GFS flipped and went to the European solution I believe this was a few years ago

Yep being consistent means nothing when it doesn't agree with a superior model. It's heartbreaking but I think we all know what's going to happen

Sent from my SM-G981V using Tapatalk

[mappy]

That’s a lot of precip that falls as mostly snow IMBY. All in. 

[Ralph Wiggum]

NAM at range (all caveats apply)  has the boundary for the ull to.follow much farther N imo and is reflected at the surface with a weak precip shield stretching from the slp in the Southwest all the way to upper PA. Or am I way off base here?

[dallen7908]

 

EPS members snow/sleet totals (inches) for the last 12 runs (College Park) through 00 UT Feb 3

90th percentile: 9 15 12 15 12 16 18 12 13 10 10 11

80th percentile: 6 10 11 9 10 12 13 9 11 7 6 8

70th percentile: 6 6 9 8 7 11 11 8 9 6 3 6

60th percentile: 4 5 8 6 6 7 9 7 7 4 2 5

50th percentile: 3 4 6 5 4 6 7 6 6 2 1 4 

40th percentile: 2 3 4 4 4 5 6 5 6 2 0.8 4

30th percentile: 2 1 4 3 3 4 5 4 5 0.9 0.6 3 

20th percentile: 0.7 0.6 3 2 2 3 3 4 3 0.5 0.4 2

10th percentile: 0.5 0.3 1 1 1 1 3 3 1 0 0.1 1

[clskinsfan]

For those saying the Euro doesn't usually fail inside 5 days. It failed miserably with the December storm inside 3 days. It is the best model for sure. But it isn't perfect.

[high risk]

1 hour ago, WxUSAF said:

I know it used to do that for the 6z and 18z runs.  Thats why back in the day, the 6z and 18z runs could burp out some real outlier solutions.  But I thought that changed long ago.  @high risk?

 

1 hour ago, LP08 said:

I’d bet dollar to donuts that the gfs doesn’t use it’s 6hr forecast for its next run.  It injests new data prior to every run.

       

     Both are true.    Yes, the GFS ingests new data for every cycle, but you have to use something as the first guess, and that is the 6-hr forecast from the previous GFS cycle.     You can't just insert obs onto a blank slate - there have to be initial values for all prognostic fields at all grid point at all levels, and you adjust those initial values based on the observations to get a solution that is balanced.    You can't just use blunt force with a data observation and try to just perform straight interpolation of its value in the horizontal and vertical.

    

 

 

[WinterWxLuvr]

Just now, high risk said:

 

       

     Both are true.    Yes, the GFS ingests new data for every cycle, but you have to use something as the first guess, and that is the 6-hr forecast from the previous GFS cycle.     You can't just insert obs onto a blank slate - there have to be initial values for all prognostic fields at all grid point at all levels, and you adjust those initial values based on the observations to get a solution that is balanced.    You can't just use blunt force with a data observation and try to just perform straight interpolation of its value in the horizontal and vertical.

    

 

 

But wouldn’t all models do that as well?

[psuhoffman]

2 hours ago, WxUSAF said:

^stronger and more consolidated shortwave. Can’t think that’s bad.

Yes but also slightly less ridging and a more compressed flow in front of it. They probably offset imo. 

[Heisy]

1 minute ago, high risk said:

 

       

     Both are true.    Yes, the GFS ingests new data for every cycle, but you have to use something as the first guess, and that is the 6-hr forecast from the previous GFS cycle.     You can't just insert obs onto a blank slate - there have to be initial values for all prognostic fields at all grid point at all levels, and you adjust those initial values based on the observations to get a solution that is balanced.    You can't just use blunt force with a data observation and try to just perform straight interpolation of its value in the horizontal and vertical.

    

 

 

Would this explain trends to a certain degree? 

[psuhoffman]

1 hour ago, Wentzadelphia said:

6z eps is a little better. Still nothing like GEFS but certainly an improvement

What made you say this?  Just curious because it looks worse then 0z to me.  

[Buddy1987]

Nam looks “diggy” toward the end of its run when comparing to gfs at 90. I’m tempering my expectations as I’m sure most are. Right now threat slightly better for areas to my north and east but I’m far from out of it with para and Canadian looking better region wide.

[high risk]

4 minutes ago, Wentzadelphia said:

Would this explain trends to a certain degree? 

     Maybe?     If, say, the GFS badly initialized the wave off of the west coast at 18z yesterday, it *could* take a few cycles to undo the damage.     For the 00z cycle, it would start with the specification of the wave from the 18z 6h forecast (the first guess).    If there is new data where that feature resides at 00z, it will take that into account, but it will probably only adjust the first guess partially towards the new obs there.      Then the 06z cycle will adjust further.      It depends how much data is available where the feature is and how different it is from the first guess.

     A famous example is with the NWP failures with the 2000 blizzard.     On the morning before the blizzard, the Peachtree City, GA raob had big wind speeds up at jet level, as a huge jet streak rotated through the trough and helped it go negative tilt.       The first guess value of the jet level wind speed  over GA was about 40 (I can't remember the number) knots lower than the ob, and the models all treated the ob as garbage and pretty much didn't adjust the wind speed there at all.      

 

[nj2va]

ICON is a weak POS wave that just slides off the coast. Some very light (less than an inch) snow to CVA.

[LeesburgWx]

We all know what is happening today... Can’t wait for Spring

[LP08]

1 minute ago, nj2va said:

ICON is a weak POS wave that just slides off the coast. Some very light (less than an inch) snow to CVA.

It’s that ULL in Canada that doesn’t allow it to amplify.  At least in my eyes.

[nj2va]

Just now, LP08 said:

It’s that ULL in Canada that doesn’t allow it to amplify.  At least in my eyes.

 

Good catch, yep the placement of that feature looks to be key.  GFS handling it totally differently.

[Heisy]

25 minutes ago, psuhoffman said:

What made you say this?  Just curious because it looks worse then 0z to me.  

I posted later, but I thoughtthe H5 was more organized early on, but turns out the confluence was worst from the lingering 50/50

[psuhoffman]

1 minute ago, Wentzadelphia said:

I posted later, but I thoughtthe H5 was more organized early on, but turns out the confluence was worst from the lingering 50/50

Yea the wave was stronger but the shred factory flow got worse and more then offset. 

[PhineasC]

36 minutes ago, clskinsfan said:

For those saying the Euro doesn't usually fail inside 5 days. It failed miserably with the December storm inside 3 days. It is the best model for sure. But it isn't perfect.

It fails in New England all the time, FWIW. I have seen it cave to the GFS and even the GGEM twice now. It's not the same Euro as in the past.

[Heisy]

20 minutes ago, high risk said:

     Maybe?     If, say, the GFS badly initialized the wave off of the west coast at 18z yesterday, it *could* take a few cycles to undo the damage.     For the 00z cycle, it would start with the specification of the wave from the 18z 6h forecast (the first guess).    If there is new data where that feature resides at 00z, it will take that into account, but it will probably only adjust the first guess partially towards the new obs there.      Then the 06z cycle will adjust further.      It depends how much data is available where the feature is and how different it is from the first guess.

     A famous example is with the NWP failures with the 2000 blizzard.     On the morning before the blizzard, the Peachtree City, GA raob had big wind speeds up at jet level, as a huge jet streak rotated through the trough and helped it go negative tilt.       The first guess value of the jet level wind speed  over GA was about 40 (I can't remember the number) knots lower than the ob, and the models all treated the ob as garbage and pretty much didn't adjust the wind speed there at all.      

 

Thanks, to me it makes sense why we sometimes see the GFS kind of tick towards other models at times, and then it has one run where it completely caves  on th  overall progression of events. I’d  wager good money the 12z gfs ticks south again, but I suppose the big money will be on the 00z run. This also happens to other models though too so there’s just so many factors well see