January Med/Long Range Discussion

[BTRWx]

I admit I rarely even look at the numerical stuff when it comes to these things. Until this year I mostly just referred to things as ridges/troughs in various places. I have spent more time with NAO/AO numbers because I think that's the primary driver around here for big events but even there somewhat limited.  I'm more into studying the patterns themselves on maps.. been a weirdo about that for years, though at this point I'd say I'm pretty skilled at pattern recognition because of it. 

 

I'm not even really sure what the typical +PNA bounds are now that I think about it. I could certainly see where it's shifted a bit east but counts etc.  I don't even feel like the +PNA is good for a whole lot other than cold delivery and even there it's not often the best cold source ever. In strong Ninos it does seem to be a main source for typically brief cold shots, especially the first half of winter, at least based on my somewhat quick looks.

 

As WxUSAF and you talk about further below if you are looking for an idealized ridge axis during a storm you generally look to the Rockies. You can almost draw a line from Scandinavia area over the polar region to the desert southwest or so and the more 500mb ridging you can get in that whole zone the better generally. But as mentioned the other day we often look for perfect patterns when most of our storms come from significantly less.

 

I guess my original statement was more that it seems in stronger Ninos the low anomaly is so close to the coast it often torches that airmass with both PAC air and downsloping. When I said +PNA is atypical I was thinking more in the frame of stopping that from happening. I've looked at enough to think that it must mostly be a transient feature though the Feb composite does back the GOA low west a bit leaving that region more in normal heights.  Of course ENSO strength isn't the be all anyway though in a Nino of this strength I think anyone not focusing on it first is doing themselves a disservice (cough cough, cohen, cough cough).

Should there be an analog thread so we can return to it as needed? Just a thought.

[Amped]

18z GEFS aren't anywhere close to where they were 3 days ago for the storm on the 16th. If it happens now it's out of luck.

EDIT: or the EPS

[Ian]

Should there be an analog thread so we can return to it as needed? Just a thought.

Probably not a bad idea. Analog/pattern recognition/etc. I'm nearing final part of a neat personal project that will fit nicely.

[clskinsfan]

I would like to learn more about the Ensembles. I know that they are run with a base run and then "perturbed" to get the various members. I am wondering what is perturbed? Is it specific strengths of HP's and LP's....Anyway I guess what I am looking for is a link to understand the Ensembles a little more.

[psuhoffman]

18z GEFS aren't anywhere close to where they were 3 days ago for the storm on the 16th. If it happens now it's out of luck.

EDIT: or the EPS

Just pray the gefs is wrong. Huge fight between the ggem/eps and gefs right now. I kinda like where the gem/eps are going with a split flow stj and decent trough in the east towards the end while the gfs is straight up screaming "no winter for you". It reloads the ridge over the lakes and we go back into the nov-dec cycle. If that happens it's probably game over.

ETA: gefs is also headed the wrong way with the chance of a storm before the torch reloads. Mean snowfall is back down near an inch after a couple runs in the 3-4" range a couple days ago. I'd bet on the euro if a gun was to my head but I would feel a lot better if the gefs would stop showing that total nuke to our winter in the long range.

[BTRWx]

I would like to learn more about the Ensembles. I know that they are run with a base run and then "perturbed" to get the various members. I am wondering what is perturbed? Is it specific strengths of HP's and LP's....Anyway I guess what I am looking for is a link to understand the Ensembles a little more.

As much as I use numerical models, my go to map for ensembles is almost always the means. Here is a nice source for non-subscribers: http://www.weatheronline.co.uk/cgi-bin/expertcharts?LANG=en&MENU=0000000000&CONT=namk&MODELL=ecmwfens&MODELLTYP=2&BASE=201601041200&VAR=t85p&HH=144&ARCHIV=0&ZOOM=0&PERIOD=&WMO=

[Scarlet Pimpernel]

Just pray the gefs is wrong. Huge fight between the ggem/eps and gefs right now. I kinda like where the gem/eps are going with a split flow stj and decent trough in the east towards the end while the gfs is straight up screaming "no winter for you". It reloads the ridge over the lakes and we go back into the nov-dec cycle. If that happens it's probably game over.

ETA: gefs is also headed the wrong way with the chance of a storm before the torch reloads. Mean snowfall is back down near an inch after a couple runs in the 3-4" range a couple days ago. I'd bet on the euro if a gun was to my head but I would feel a lot better if the gefs would stop showing that total nuke to our winter in the long range.

I was just about to mention that the 18Z GEFS in the LR is, to be very diplomatic about it, very uhhh, uninspiring! It's a veritable train wreck toward the end to be honest. Don't like the past several runs of the GEFS showing this evolution now. Ironically, the 18Z deterministic GFS isn't half bad.

[Ian]

GEFS usually does OK or at least it's not as noticeably second fiddle compared to the op battle with the Euro. That said I've seen plenty of talk that it doesn't seem to handle tropical region forcing well. Just from watching here euro seems to handle the southern stream better overall. GEFS MJO forecasts of late haven't been very good compared to euro's either. I'm not sure that means a lot for the key cold factors but going back to Dec in late Jan would be fairly unusual I think. Even the bad Ninos were largely cool and wet for Feb. Think a relaxation and reload (or main show) is more likely than not. Questions are probably how deep we get into a lesser pattern in between and if we are ever going to time something. And if there will ever be a coastal.

[psuhoffman]

I was just about to mention that the 18Z GEFS in the LR is, to be very diplomatic about it, very uhhh, uninspiring! It's a veritable train wreck toward the end to be honest. Don't like the past several runs of the GEFS showing this evolution now. Ironically, the 18Z deterministic GFS isn't half bad.

You can't sugar coat it the gefs is saying we reload the previous pattern. Probably not as warm jan style but not much hope for snow in the pattern shown and its a stable hard to get out of pattern as we saw already. Good news is thw gefs are pretty much on an island with that look with the jma can sips and eps especially the para in a better look although I've heard the old euro weeklies are bad too. A lot of mixed signals. Confidence is low.

[psuhoffman]

GEFS usually does OK or at least it's not as noticeably second fiddle compared to the op battle with the Euro. That said I've seen plenty of talk that it doesn't seem to handle tropical region forcing well. Just from watching here euro seems to handle the southern stream better overall. GEFS MJO forecasts of late haven't been very good compared to euro's either. I'm not sure that means a lot for the key cold factors but going back to Dec in late Jan would be fairly unusual I think. Even the bad Ninos were largely cool and wet for Feb. Think a relaxation and reload (or main show) is more likely than not. Questions are probably how deep we get into a lesser pattern in between and if we are ever going to time something. And if there will ever be a coastal.

Agree with all of this 100%

[Bob Chill]

Interestingly, when you dig into the GEFS members, the nasty + anomaly in a spot we don't want to see isn't really "favored" as much as it might be skewed by 4 extreme members. 

 

[Bob Chill]

 

 

When you talk about the axis, are you talking about the center of the heights? Can you post a map to illustrate?

 

 

 

Go back to my earlier post about the 12z gefs @ 276. That's a good storm look. H5 and vort panels center the dip in heights right around the MS river with energy at the base. This is a common way for us to get storms. LP forms out in front of the trough and tracks underneath us. It's not super simple by just identifying a trough axis and saying "there's 100% chance of a good storm". It's an important piece of the puzzle though.

 

When the trough axis is too far east (like right now), storms form but they are out in the ocean. Too far west and we get a free car wash. 

[Ian]

Agree with all of this 100%

The EPS doesn't look that great to me by the end either, but it's not quite what we just saw... In some ways the GEFS solution makes more sense too, at least out west and with the STJ. Weeklies did shift the precip anomlies north out there with time though as well. It's definitely a lot trickier of a pattern.. that's a plus I guess? By this time in Dec it was already pretty clear we were going to run at records. Something coming in late Jan (if it does) that isn't favorable you just hope it doesn't last long as that could snag a good chunk of our prime climo for big events.

[SnowGoose69]

Interestingly, when you dig into the GEFS members, the nasty + anomaly in a spot we don't want to see isn't really "favored" as much as it might be skewed by 4 extreme members.

Yeah it's been this way for days too, skewed by a few members it seems whenever I look at the individuals

[WinterWxLuvr]

Interestingly, when you dig into the GEFS members, the nasty + anomaly in a spot we don't want to see isn't really "favored" as much as it might be skewed by 4 extreme members.

And therein lies the problem with ensembles IMO. Not sure exactly how you would remove outliers, but looking at individuals closely might give a more realistic representation.

I looked tonight at the ensembles from hours 168-240. Probably should have started at 120, but the range I looked at have several members with solutions that are at least not psychologically shattering.

I think the next ten still offer at least a glimmer.

[SnowGoose69]

And therein lies the problem with ensembles IMO. Not sure exactly how you would remove outliers, but looking at individuals closely might give a more realistic representation.

I looked tonight at the ensembles from hours 168-240. Probably should have started at 120, but the range I looked at have several members with solutions that are at least not psychologically shattering.

I think the next ten still offer at least a glimmer.

The best example of ensembles not being useful unless you dig deep was the 2/6/10 storm or no storm for NYC. The GFS ensemble mean was like .80 for NYC but that was because 2 members had like 2 inches liquid and skewed the crap out of the mean. Of course the other lesson there is don't use ensembles inside 72-96 hours

[Ian]

And therein lies the problem with ensembles IMO. Not sure exactly how you would remove outliers, but looking at individuals closely might give a more realistic representation.

I looked tonight at the ensembles from hours 168-240. Probably should have started at 120, but the range I looked at have several members with solutions that are at least not psychologically shattering.

I think the next ten still offer at least a glimmer.

The ensembles need like 10,000 members. But it's still a preferable guide to use their mean at any range. At least they tend to stay somewhat steady compared to the op.

[WinterWxLuvr]

Go back to my earlier post about the 12z gefs @ 276. That's a good storm look. H5 and vort panels center the dip in heights right around the MS river with energy at the base. This is a common way for us to get storms. LP forms out in front of the trough and tracks underneath us. It's not super simple by just identifying a trough axis and saying "there's 100% chance of a good storm". It's an important piece of the puzzle though.

When the trough axis is too far east (like right now), storms form but they are out in the ocean. Too far west and we get a free car wash.

I've never agreed more with a post more than this. I've said the same in the past, perhaps not as far west as you. You can't have a trough axis over Atlanta and get much, maybe a clipper (too far north more likely). We need to be on the up side of the trough IMO. Plus, the gulf seems to be cut off with a deep east trough.

[clskinsfan]

As much as I use numerical models, my go to map for ensembles is almost always the means. Here is a nice source for non-subscribers: http://www.weatheronline.co.uk/cgi-bin/expertcharts?LANG=en&MENU=0000000000&CONT=namk&MODELL=ecmwfens&MODELLTYP=2&BASE=201601041200&VAR=t85p&HH=144&ARCHIV=0&ZOOM=0&PERIOD=&WMO=

 

Thanks. I am familiar with reading them and have access to them. I am more interested in how they work. 

[packbacker]

Go back to my earlier post about the 12z gefs @ 276. That's a good storm look. H5 and vort panels center the dip in heights right around the MS river with energy at the base. This is a common way for us to get storms. LP forms out in front of the trough and tracks underneath us. It's not super simple by just identifying a trough axis and saying "there's 100% chance of a good storm". It's an important piece of the puzzle though.

 

When the trough axis is too far east (like right now), storms form but they are out in the ocean. Too far west and we get a free car wash. 

 

When you look at the strong -AO Jan mod+ nino's you get what the EPS/Canadian are spitting out in the extended, question is how much help do we get from EPO and how long does -AO last...GEFS does make you pause though.

[WinterWxLuvr]

The ensembles need like 10,000 members. But it's still a preferable guide to use their mean at any range. At least they tend to stay somewhat steady compared to the op.

I don't have access to Euro members so I'm stuck with the gfs. Regardless, I'm picking 3 or 4 of the best looking members and going with them .

[SnowGoose69]

I've never agreed more with a post more than this. I've said the same in the past, perhaps not as far west as you. You can't have a trough axis over Atlanta and get much, maybe a clipper (too far north more likely). We need to be on the up side of the trough IMO. Plus, the gulf seems to be cut off with a deep east trough.

In general too the further south you are in the eastern US the more you want the trof to be broad, a sharp trof for places like MEM CLT ATL BNA LIT OKC DFW usually isn't good for winter storms because either the system is going to amplify and cut north or it will shear out. Further north the broad trof meridional flow can be bad because it often means there is a gargantuan high over the OH Valley and everything will be suppressed

[mitchnick]

I don't have access to Euro members so I'm stuck with the gfs. Regardless, I'm picking 3 or 4 of the best looking members and going with them .

That's the method I tried to stick with when I was in a bar looking to meet women. It didn't work for me then and I don't think it will work for you now. Lol

[Scarlet Pimpernel]

Interestingly, when you dig into the GEFS members, the nasty + anomaly in a spot we don't want to see isn't really "favored" as much as it might be skewed by 4 extreme members. 

 

 

Interesting...yeah, definitely 4 real stinkers in that lot (I'm looking at you, P002, P005, P007 and P009!).  But there are a couple of others that are not great either.  Likewise, some really good looks in there too.

[BristowWx]

Interesting...yeah, definitely 4 real stinkers in that lot (I'm looking at you, P002, P005, P007 and P009!).  But there are a couple of others that are not great either.  Likewise, some really good looks in there too.

P010 would be sweet. At least a bad pattern is not locked in. I can't believe how hard this is this year. I mean we can't score a couple inches. Some sleet. ZR. Nothing?

[Ian]

I don't have access to Euro members so I'm stuck with the gfs. Regardless, I'm picking 3 or 4 of the best looking members and going with them .

I don't look at the members too often. Then again I'm kind of a lazy forecaster and I only tend to consistently look this deep into the future during winter (or before chasecation lol). It would be fun to track each member individually run to run like we do with the ops, tho I guess maybe they are randomly perturbed each time-- not up on those details.

[Fozz]

P010 would be sweet. At least a bad pattern is not locked in. I can't believe how hard this is this year. I mean we can't score a couple inches. Some sleet. ZR. Nothing?

 

I don't think a strong El Nino tends to have much of any nickel and dime events.

[BristowWx]

I don't think a strong El Nino tends to have much of any nickel and dime events.

Yeah. Two things I am not a fan of are coastals and transfers....both tend to disappoint. I would much prefer an overrunning event with good CAD.

[Ian]

I don't think a strong El Nino tends to have much of any nickel and dime events.

The post 1950 strong nino winters have mostly been rather paltry event-wise but still a number of smaller events in there. 57-58 had quite a few.. 10 accumulation events at DC. Tho it was also a snowy winter at this point.

[WinterWxLuvr]

Where is our raging stj?