January Medium/Long Range Discussion

[psuhoffman]

9 minutes ago, NorthArlington101 said:

The one thing about this storm working out is now we have to hold the GFS in high regard for a few weeks at least. Heck, the long range HRRR did well.

Something to remember, all the models have flaws and use different equations to compensate and do their best to make up for our inherent limitations.  We have not advanced to the point where we can model the atmosphere with 100% accuracy.  We compensate for that using different methods in the different models to limit the errors.  That means that each model will be wrong to some degree at range, and one part of the skill is picking out how they are likely wrong.  The other thing is each synoptic setup is different with different variables being most important in influencing a storms outcome and so one model could have an advantage over another based on its specific physics.  That model might not always be the one that is the "best" on average in the long run.   

 

In this last setup the key issues the models were getting wrong, even the GFS just to a lesser degree, was the interaction and spacing between the lead NS wave and the trailing wave on the front.  For a while guidance had the timing off and the NS wave right on top of the southern wave which essentially flattened and washed it out.   At range the models all took turns seeing the setup but they all lost it in the medium range.  They had the timing between the two wrong.  Something about the GFS allowed it to catch on a little sooner than the other globals in seeing the separation that would allow the southern wave to amplify.  

 

The GFS isnt a clown model.  Its not as good as the euro in the long run but its one of the 3 best global models and worth taking into consideration in every situation.  But every situation will be different and we have to try to judge which model is handling it better.  In that regard I actually felt more comfortable making general long range forecasts 15 years ago than I do now.  When the models had a much lower resolution and before we started updating them every couple years...they had more consistent biases and erros IMO.  I was able to look at them and identify more often what they were each doing wrong based on their known error biases.  Now, with the increased resolution and rapid updates these errors seem less consistent to me.  It's a lot harder IMO to know exactly how to adjust for the model errors.  Yes they are better in that they are each more likely to be closer to reality on any given run...but in a way the ability to adjust for their errors, and they will still have them, is lower for me at least.  

 

That said...looking ahead to this next system, I think the GFS is playing into one of its still prominent known errors of too much cold and suppressing the wave a little too much.   Looking at the upper level flow that wave should be pulling northward with the energy coming through the ohio valley and tracking to our NW...not getting squashed off to the SE.  The Euro/GGEM looks closer to what the pattern supports IMO.  But like I said before...its a lot harder imo to make that statement with any confidence since the models are more jumpy having different errors with each situation lately.  

[NorthArlington101]

Something to remember, all the models have flaws and use different equations to compensate and do their best to make up for our inherent limitations.  We have not advanced to the point where we can model the atmosphere with 100% accuracy.  We compensate for that using different methods in the different models to limit the errors.  That means that each model will be wrong to some degree at range, and one part of the skill is picking out how they are likely wrong.  The other thing is each synoptic setup is different with different variables being most important in influencing a storms outcome and so one model could have an advantage over another based on its specific physics.  That model might not always be the one that is the "best" on average in the long run.   
 
In this last setup the key issues the models were getting wrong, even the GFS just to a lesser degree, was the interaction and spacing between the lead NS wave and the trailing wave on the front.  For a while guidance had the timing off and the NS wave right on top of the southern wave which essentially flattened and washed it out.   At range the models all took turns seeing the setup but they all lost it in the medium range.  They had the timing between the two wrong.  Something about the GFS allowed it to catch on a little sooner than the other globals in seeing the separation that would allow the southern wave to amplify.  
 
The GFS isnt a clown model.  Its not as good as the euro in the long run but its one of the 3 best global models and worth taking into consideration in every situation.  But every situation will be different and we have to try to judge which model is handling it better.  In that regard I actually felt more comfortable making general long range forecasts 15 years ago than I do now.  When the models had a much lower resolution and before we started updating them every couple years...they had more consistent biases and erros IMO.  I was able to look at them and identify more often what they were each doing wrong based on their known error biases.  Now, with the increased resolution and rapid updates these errors seem less consistent to me.  It's a lot harder IMO to know exactly how to adjust for the model errors.  Yes they are better in that they are each more likely to be closer to reality on any given run...but in a way the ability to adjust for their errors, and they will still have them, is lower for me at least.  
 
That said...looking ahead to this next system, I think the GFS is playing into one of its still prominent known errors of too much cold and suppressing the wave a little too much.   Looking at the upper level flow that wave should be pulling northward with the energy coming through the ohio valley and tracking to our NW...not getting squashed off to the SE.  The Euro/GGEM looks closer to what the pattern supports IMO.  But like I said before...its a lot harder imo to make that statement with any confidence since the models are more jumpy having different errors with each situation lately.  

Yeah, I think my original post came off a bit wrong — I mean we’ll be hearing a lot of justification of (maybe) unlikely outcomes based on the success of the GFS and others this time around. I recognize that each model has it strengths/weaknesses and they’ll shine brighter in some circumstances than others. What this storm does from my perspective is that for the next few years, I imagine the next time the GFS is on an island, people will cite this, haha.

[benjammin]

On to Cincinnati. 

[Imgoinhungry]

4 minutes ago, benjammin said:

On to Cincinnati. 

the low is that far west?

[psuhoffman]

4 minutes ago, NorthArlington101 said:

Yeah, I think my original post came off wrong a bit wrong — I mean we’ll be hearing a lot of justification of (maybe) unlikely outcomes based on the success of the GFS and others this time around. I recognize that each model has it strengths/weaknesses and they’ll shine brighter in some circumstances than others. What this storm does from my perspective is that for the next few years, I imagine the next time the GFS is on an island, people will cite this, haha.

I wasn't trying to be critical of your post, just adding my own thoughts too it.  In the future people would do well to realize the GFS wasn't completely on an island.  Other guidance did see the setup in previous runs and had just lost the look temporarily.  Additionally the GFS was only about 12 hours ahead of the curve with the other models and they were trending along with it just a cycle or two behind the curve.  If any one model is totally and completely on an island with no support at all its wrong the vast majority of the time. 

[Bob Chill]

Pretty much every long timer here knows that our "good" winter periods are usually very compressed and storms come in bunches. No idea why. Just how it is.  Don't have any worthwhile thoughts about late week except we might have kicked off the first of a "storm bunch" today. If that's really possible, Fri may end up similar to today with how models did in the mid range. 

 

If you go thru history, our compacted storm periods can happen in any type of enso and can even happen in the face of an abysmal winter. Wx is weird like that. We haven't had a heater since Feb-Mar 2015. Would be kinda nice to have another but that's just me. 

[psuhoffman]

7 minutes ago, Imgoinhungry said:

the low is that far west?

I think he was referring to the Bill Belichick press conference from years ago

[Deck Pic]

euro is a nice 3-6" event Thursday night

[psuhoffman]

Euro is 994mb about 75 miles east of OC at 96 hrs. 

[Imgoinhungry]

3 minutes ago, psuhoffman said:

I think he was referring to the Bill Belichick press conference from years ago

lol, phew!

[Eskimo Joe]

6 minutes ago, psuhoffman said:

Euro is 994mb about 75 miles east of OC at 96 hrs. 

Good. We snow.

[frd]

6 minutes ago, psuhoffman said:

Euro is 994mb about 75 miles east of OC at 96 hrs. 

[mappy]

^^ I can't miss out on that! 

[Eskimo Joe]

Euro is nice. Everybody wins.

[LeesburgWx]

Guys, this storm is only 84-90 hours out… 

[frd]

 

Brrr........    maybe  too.  Clipper time 

 

 

[SnowenOutThere]

17 minutes ago, psuhoffman said:

I wasn't trying to be critical of your post, just adding my own thoughts too it.  In the future people would do well to realize the GFS wasn't completely on an island.  Other guidance did see the setup in previous runs and had just lost the look temporarily.  Additionally the GFS was only about 12 hours ahead of the curve with the other models and they were trending along with it just a cycle or two behind the curve.  If any one model is totally and completely on an island with no support at all its wrong the vast majority of the time. 

 

Yeah, after the GFS and the CRAS both showed it far before the others. So the lesson is simple, the CRAS is a better model than the EURO. lol

[Bob Chill]

Damn. Right on time with the Euro. 72-96 hours too. Close to the range where the general idea locks in. Perfect. 

[SnowenOutThere]

1 minute ago, LeesburgWx said:

Guys, this storm is only 84-90 hours out… 

Long range RGEM time soon

[psuhoffman]

3 minutes ago, mappy said:

^^ I can't miss out on that! 

This next system has a lot more room to lift north up the coast.  It's ahead of the trough axis instead of behind it.  This last system was cut off and had just enough space to amplify but it was never going to be able to gain latitude given the NS configuration.  This next system, if it has enough amplitude, will lift north.  The risk to missing it IMO would be if the system is simply too weak to amplify a storm in time.  If the southern wave isnt strong enough to initiate a heathy storm in time it could get squashed out.  I don't favor that scenario at this time...but that would be the risk.  If the southern wave is anything close to as amplified...even half...as this last system it will be a healthy storm that will affect places further north than this last one did.  

[psuhoffman]

7 minutes ago, Bob Chill said:

Damn. Right on time with the Euro. 72-96 hours too. Close to the range where the general idea locks in. Perfect. 

just a thought... we had that weird fairly short one week period in early December 2013 but other then that didnt the "winter" really set in right about now in 2014 also?  The current pattern has a lot of similarities to that year.  We all know we also got really lucky that year on top of the pattern so I am NOT saying to expect that kind of outcome...just saying seeing similar type waves showing up right now maybe shouldnt be shocking with the pattern similarities.  

[SomeguyfromTakomaPark]

Euro has highs in the 20s Friday and Saturday.  Snow on the ground + actual cold. 

[DDweatherman]

The GFS says pretty much nothing for late week, Euro this time the one with snow. Should note the GFS has more support in all honesty (ICON, UKMET)

[jaydreb]

3 minutes ago, DDweatherman said:

The GFS says pretty much nothing for late week, Euro this time the one with snow. Should note the GFS has more support in all honesty (ICON, UKMET)

GEFS is better than GFS OP.  

[psuhoffman]

 

Now before people freak out I just wanted to post a period with a somewhat similar pattern look.  The WAR, full latitude trough from Siberia into the eastern US with a severely stretched PV, and EPO poleward ridge being the 3 significant similar longwave features to what all 3 global ensembles are showing coming up.  There were a lot of boundary waves during this period before the big March storm.  They mostly affected just NW of 95 with a lot of mixed events along 95...but patterns are similar not identical.  You could easily adjust some of those storms south 50 miles and suddenly it was a very snowy period in DC as well as Hagerstown!

 

Secondly, about the elephant in the room...March 93...I mentioned this in the zoom chat last night...this is the kind of look, full latitude eastern trough with a full latitude western ridge that COULD lead to that kind of thing.  The key would be to get enough of the energy from a wave to dig and amplify in the base of the trough.  But what is way more likely is a lot of weaker waves.  We get this kind of setup occasionally but only VERY VERY VERY rarely does it produce the big bomb scenario.  So just because this is the kind of setup that COULD lead to that...doesn't mean its likely.  Just possibly.  Slightly possible.  The idea of boundary waves similar to what we saw during that period in 1993 (and 2014 just a little further south that year) is valid though.  That was my main point was just to show some past similar periods to what guidance is spitting out now.    

[high risk]

2 minutes ago, psuhoffman said:

 

Now before people freak out I just wanted to post a period with a somewhat similar pattern look.  The WAR, full latitude trough from Siberia into the eastern US with a severely stretched PV, and EPO poleward ridge being the 3 significant similar longwave features to what all 3 global ensembles are showing coming up.  There were a lot of boundary waves during this period before the big March storm.  They mostly affected just NW of 95 with a lot of mixed events along 95...but patterns are similar not identical.  You could easily adjust some of those storms south 50 miles and suddenly it was a very snowy period in DC as well as Hagerstown!

 

Secondly, about the elephant in the room...March 93...I mentioned this in the zoom chat last night...this is the kind of look, full latitude eastern trough with a full latitude western ridge that COULD lead to that kind of thing.  The key would be to get enough of the energy from a wave to dig and amplify in the base of the trough.  But what is way more likely is a lot of weaker waves.  We get this kind of setup occasionally but only VERY VERY VERY rarely does it produce the big bomb scenario.  So just because this is the kind of setup that COULD lead to that...doesn't mean its likely.  Just possibly.  Slightly possible.  The idea of boundary waves similar to what we saw during that period in 1993 (and 2014 just a little further south that year) is valid though.  That was my main point was just to show some past similar periods to what guidance is spitting out now.    

     

              Superstorm '22 is a lock.    Got it!        

 

[psuhoffman]

1 minute ago, high risk said:

     

              Superstorm '22 is a lock.    Got it!        

 

glad I made my point clear!!!

[psuhoffman]

Honestly we probably wouldn't want a triple phased bomb anyways...most of the time they are going to cut even more than 1993 did and produce more rain than frozen precip along 95.  

[mappy]

28 minutes ago, psuhoffman said:

This next system has a lot more room to lift north up the coast.  It's ahead of the trough axis instead of behind it.  This last system was cut off and had just enough space to amplify but it was never going to be able to gain latitude given the NS configuration.  This next system, if it has enough amplitude, will lift north.  The risk to missing it IMO would be if the system is simply too weak to amplify a storm in time.  If the southern wave isnt strong enough to initiate a heathy storm in time it could get squashed out.  I don't favor that scenario at this time...but that would be the risk.  If the southern wave is anything close to as amplified...even half...as this last system it will be a healthy storm that will affect places further north than this last one did.  

I appreciate your words, but sometimes i just want to hear "its a lock, get ready".  

[LeesburgWx]

So if Thursday/Friday works out, what are we looking at potential wise for amounts?