January Mid/Long Range Disco 3: The great recovery or shut the blinds?

[osfan24]

I keep seeing GEFS snow maps where I get multiple inches and I haven't had multiple flakes yet. Those maps are so meaningless.

[psuhoffman]

6 minutes ago, WxUSAF said:

Yeah…GEFS isn’t as good for us with this 12z run. Slight can kick, but more importantly it keeps the ridge axis well off the west coast. That teleconnects to more SE ridge and hence has a cold and dry, warm and wet sorta pattern. 

Is it a can kick or is it just showing the details more clearly as it gets closer and there is less washing out due to variance and timing differences. This was always how I figured it would actually look given the gefs longwave pattern. Other than one run where it temporarily caved to the eps I never really liked what it was offering. 

[Heisy]

Is it a can kick or is it just showing the details more clearly as it gets closer and there is less washing out due to variance and timing differences. This was always how I figured it would actually look given the gefs longwave pattern. Other than one run where it temporarily caved to the eps I never really liked what it was offering. 

Right the ensembles start picking up on individual shortwaves better as we get closer. Luckily cmc and euro look more similar to each other at end of run vs gfs.

[brooklynwx99]

13 minutes ago, SnowenOutThere said:

Did a met just post the hour 366 GFS snowfall map? We really are desperate for anything.

snowfall means at range are somewhat useful for determining what periods have increased snowfall potential. was just posting to show that there is still decent potential late in the month due to the all the cold air on this side of the globe 

it’s not rocket science. there isn’t much else to post anyway, all of this stuff is like 10 days away

[jayyy]

I keep seeing GEFS snow maps where I get multiple inches and I haven't had multiple flakes yet. Those maps are so meaningless.

They’re not meaningless, they just shouldn’t be relied upon. Ensembles have dozens of individual members - some of which show decent hits for some of these systems, which is why the mean / average shows a few inches. The problem is… we keep ending up verifying the snowless members of these ensembles.

No model is meant to be fully leaned on, we know this. They are simply hypothetical outcomes and tools to use in conjunction with one another to assess the most probable outcome.

If the GEFS mean shows 3”, but the EPS and CMC show 0”, it’s likely we end up at or near 0”. If the other models agree or show something similar, perhaps the GEFS mean is onto something. The issue is people find a model with a desirable outcome and hug it. They also don’t understand what a “mean” is or what ensembles are actually used for.


.

[brooklynwx99]

6 minutes ago, WxUSAF said:

Lots of cold air available nearby at least, but it’s probably one of those cutter—boundary wave sort of patterns that @psuhoffman is dreading. Crap ton better than we have now though I guess.

boundary waves with lots of available cold in Canada is how you guys cashed in during March 2015, though

it’s definitely possible to score in that kind of pattern. just need to hope there’s a good cold press, which is certainly possible here

[SnowenOutThere]

1 minute ago, brooklynwx99 said:

snowfall means at range are somewhat useful for determining what periods have increased snowfall potential. was just posting to show that there is still decent potential due to the all the cold air on this side of the globe 

it’s not rocket science 

I'm just confused because typically anytime further than 5 days out snow maps shouldn't really be put ahead of other map types to my limited knowledge. So is it a good thing that the GEFS's snow mean got better but H5 looks worse because at this range I would think that wouldn't be the case. 

[jayyy]

boundary waves with lots of available cold in Canada is how you guys cashed in during March 2015, though
it’s definitely possible to score in that kind of pattern. just need to hope there’s a good cold press, which is certainly possible here

I actually don’t mind W/E overrunning type events. We tend to do well and even overperform with these types of setups. Give me a cold dome & some solid CAD with a trailing SLP coming from the Tennessee valley and I’ll gladly roll the dice. I cringe more when I hear the words phasing, redevelopment, ridge / trough axis, etc. That’s when you know we need things to be near perfect (particularly with a hostile PAC)

The problem is that the PAC is not allowing any cold air to remain entrenched. There’s also virtually 0 snowpack to our N/NW to draw any cold air into the NW flank of storms that are otherwise taking nice tracks but still giving us cold rain.

Generally speaking though, I don’t mind boundary setups. I just don’t like that being the only way we can score.


.

[brooklynwx99]

4 minutes ago, SnowenOutThere said:

I'm just confused because typically anytime further than 5 days out snow maps shouldn't really be put ahead of other map types to my limited knowledge. So is it a good thing that the GEFS's snow mean got better but H5 looks worse because at this range I would think that wouldn't be the case. 

it’s just that the 500mb maps are also a mean. there is likely a lot of variation within each of the individual members, and this upcoming pattern is a huge improvement due to lots of available cold in Canada

so, there is the potential for snow if we can get the boundary far enough south. pretty big if, but it’s 100 times better than the pattern we have now… the upcoming pattern at least has a legit connection to Siberia. both the 12z GEFS and GEPS agree on that. there are also some hints of transient blocking

[WinterWxLuvr]

Jan 21-22

Jan 26-27

Feb 1-2

Feb 11-12

Feb 19-20

Snow

Lets see how my personal ensemble does.

[Heisy]

I'm just confused because typically anytime further than 5 days out snow maps shouldn't really be put ahead of other map types to my limited knowledge. So is it a good thing that the GEFS's snow mean got better but H5 looks worse because at this range I would think that wouldn't be the case. 

Well the snowfall mean went up because 2 members have snowfall with one having like 40” of snow in the region.


.

[jayyy]

Jan 21-22
Jan 26-27
Feb 1-2
Feb 11-12
Feb 19-20
Snow
Lets see how my personal ensemble does.

I actually like the 25th. Storm #2 after #1 (likely) cuts west and (hopefully) provides us with a temporary 50/50 to push the boundary south. Another domino effect type storm, but that seems to be the best we can hope for right now with this threat.


.

[brooklynwx99]

@SnowenOutThere here are the GEFS individual members at 360 hours. notice the spread on these. some are ice cold with some -NAOs, some are torches, and most lie somewhere in between. most of these are serviceable, though, and most have a ridge in AK to get cold air into Canada and the US, which is the most important part

a decent portion of those actually have strong Scandinavian ridging, which is a precursor to -NAOs

[olafminesaw]

2 minutes ago, SnowenOutThere said:

I'm just confused because typically anytime further than 5 days out snow maps shouldn't really be put ahead of other map types to my limited knowledge. So is it a good thing that the GEFS's snow mean got better but H5 looks worse because at this range I would think that wouldn't be the case. 

I think in general people over-estimate the predictive power of an H5 pattern on surface outcomes. Especially for a smoothed out mean. We can only say, x features are correlated with Y outcome, but exact placement, strength, interaction between them is an inexact science at best, even if the output is largely correct.

Snow maps help to show where cold and moisture is linking up broadly, but are prone to random variation. The addition of the 100+ members of the new EPS will be a big help in this respect, assuming they are sufficiently perturbed. Temp and moisture anomaly maps together can get you a little more detail on how it's all coming together, but don't show how individual waves may potentially play out. All this together is what makes d5+ forecasts a complete crapshoot.

[WxUSAF]

32 minutes ago, psuhoffman said:

Is it a can kick or is it just showing the details more clearly as it gets closer and there is less washing out due to variance and timing differences. This was always how I figured it would actually look given the gefs longwave pattern. Other than one run where it temporarily caved to the eps I never really liked what it was offering. 

It’s like a 12-24hr can kick, which could just be it “seeing” another shortwave in western Canada. I don’t love the look the GEFS has, but still better than now. Has some potential at least. Let’s see if the EPS holds.

[Heisy]

GEFS vs eps. Hopefully just a blip on the GEFS




.

[pazzo83]

2 hours ago, psuhoffman said:

did you buy a snowblower?  if so we know the problem now...

lol no - i haven't even bought a shovel yet.  I know how this works

[WinterWxLuvr]

30 minutes ago, Heisy said:

Well the snowfall mean went up because 2 members have snowfall with one having like 40” of snow in the region.


.

Therein lies the problem with a snowfall mean. Useless product IMO

[BristowWx]

I heard on the news the latest first snow ever in NYC is 29 Ian.  What’s our latest snow at DCA?  I know someone knows this.  

[psuhoffman]

22 minutes ago, Heisy said:

GEFS vs eps. Hopefully just a blip on the GEFS




.

The eps is how you imagined prom night going at the start of senior year. Gefs is how it actually went. 

[anotherman]

1 minute ago, BristowWx said:

I heard on the news the latest first snow ever in NYC is 29 Ian.  What’s our latest snow at DCA?  I know someone knows this.  

It was in Ian's article yesterday.  Sometime in late February.

[anotherman]

• New York City: This winter has gone the ninth longest without measurable snow. Its latest first measurable snow is Jan. 29.
• Philadelphia: This winter has gone the 12th longest without measurable snow. Its latest first measurable snow is Feb. 3, although no snow fell in the winter of 1972-73.
• Atlantic City: This winter has gone the 13th longest without measurable snow. Its latest first measurable snow is Feb. 16.
• Baltimore: This winter has gone the 11th longest without measurable snow. Its latest first measurable snow is Feb. 21.
• Washington: This winter has gone the 16th longest without measurable snow. Its latest first measurable snow is Feb. 23.

[WxUSAF]

2 minutes ago, psuhoffman said:

The eps is how you imagined prom night going at the start of senior year. Gefs is how it actually went. 

EPS pattern is NOICE

[psuhoffman]

1 minute ago, WxUSAF said:

EPS pattern is NOICE

Careful someone will be along soon to tell us because there isn’t an epo ridge extending to Moscow it won’t be cold enough. 

[psuhoffman]

@CAPE I’m just kidding. Your concerns and preferences are legit for your location and recent history! 

[WxUSAF]

2 minutes ago, psuhoffman said:

Careful someone will be along soon to tell us because there isn’t an epo ridge extending to Moscow it won’t be cold enough. 

No defined 50-50 and not a 5SD -NAO, so maybe we’ll get lucky with some front end sleet? Congrats Burlington.

[psuhoffman]

19 minutes ago, WinterWxLuvr said:

Therein lies the problem with a snowfall mean. Useless product IMO

Imo they are all just tools. None of them are useless but just like you can’t use a hammer alone to build a house no one of these products alone is sufficient to make a forecast.  In this case the mean is skewed by outliers. But a quick look at the probabilities output identifies that. No harm to foul. Trying to use one product to base a forecast is user error. But if you look at all the evidence, including historical reference, it can be useful. We’ve known at least 10 days in advance of each pattern change this year going back to October. We even knew by 5 days out that the snow threat was zilch in most cases. I think guidance has been pretty good, it’s just not been what we wanted!  

[psuhoffman]

6 minutes ago, WxUSAF said:

No defined 50-50 and not a 5SD -NAO, so maybe we’ll get lucky with some front end sleet? Congrats Burlington.

Look all we need is a -4 AO NAO EPO with a +4 PNA and a TPV exactly over 50/50 and a 1055 high over Montreal with a split flow and strong STJ.  
 

Then we might get 2-4” before the flip to rain. 

[Heisy]

Careful someone will be along soon to tell us because there isn’t an epo ridge extending to Moscow it won’t be cold enough. 

Fine I’ll do it. Don’t hate me please. Anyone find it at least a little concerning that H5 looks great on EPS, but there’s little if any members showing snow in the next 15 days?


.

[Rtd208]

17 minutes ago, psuhoffman said:

The eps is how you imagined prom night going at the start of senior year. Gefs is how it actually went. 

Cold, dry and premature?