This is the worst bust I have ever experienced by far

[Bob Chill]

I have gone against the models and been wrong and still learn and modify my approach. Honestly though, if anyone with skill on here simply monitors, let's say, 3 snowstorm potential for the Ric-Balt corridor per year for 5 years you would have the results to compare to the predictions and that would be a decent based to compare the outcomes to the predictions.

 

If you only went back 3 years it would mean dc-balt has a 1 in 9 chance at getting anything meaningful and ric is the bullseye 50% of the time. 

[87storms]

The days of the Met are dying. in 20 years, the models will blow the mets out and the best Mets will be the ones who learn how to translate the models the best. You will see alot more weenie blogs of people boasting they are better than pro Mets

 

i thought about that and while it's most likely true that accurate predictions will be made in advance by computers, there may always be a human need to predict further out than the computers can go.  it might all be relative.

[winterymix]

not to digress too much but from discussion of yesterday's storm, but it reminds me of the Nov. 1987 storm where we managed to get heavy snow and high precip.rates overcoming marginal surface temperatures.  IMagine these were different set ups but any key differences that explain the differing outcomes?

Excellent question, I was wondering the same.

If I recall, that '87 system trudged across from west to east as well.

There was good snow in the big cities.

[WxUSAF]

I don't follow DT, I'm not an idiot lol. I'm just saying a lot of people on here were only focused on the GFS/NAM porno runs just as DT was ranting and raving about the euro. I'm just saying there was not a lot of discussion of what the other models were showing.

Just my observation of reading this forum in the days leading up to the storm.

Are you kidding?  There was tons of discussion on the Euro.  I think most people didn't go "all in" with this storm until the Euro caved to the GFS and brought the storm farther north and got significant precip into DC/Balt.  Please recall that King Euro had DCA getting well over 1" of QPF on the 12z Tuesday run.  People were worried about the low-level warmth on the Euro, but most of us discounted it because of the Euro's BL warm bias.  The Euro DID verify too warm in the BL along I-95.  

 

If the post count is higher on the GFS/NAM runs, it's probably just because those are free runs that everybody can access versus the 6-10 folks on here who pay to have Euro access and then post about it.  

[87storms]

this storm busted on temps a lot more than it busted on precip imo.  didn't dca still get 1" of precip.  i just don't buy into the "it was a lack of precip rates that did us in" theory.  i'm sure it hurt, but it shouldn't have hurt as much as it did.  basically, the ocean waters were too warm and there was too much of a low level easterly component.  that's why the further you traveled from the atlantic the better chances that rates could outdo transition to rain.  i'm not even sure we had all snow tuesday night even during some moderate bands.  it was still a sloppy mess.

[WxUSAF]

this storm busted on temps a lot more than it busted on precip imo.  didn't dca still get 1" of precip.  i just don't buy into the "it was a lack of precip rates that did us in" theory.  i'm sure it hurt, but it shouldn't have hurt as much as it did.  basically, the ocean waters were too warm and there was too much of a low level easterly component.  that's all she wrote for us.  

GFS predicted temps to be 1-2C for the big cities during the morning and afternoon.  Temps were 1-2C.  Not sure that's a bust.  NAM was cold, Euro was warm.  If we had a sounding at 18z we could verify the temps between the surface and 850mb...but we don't, so we can't.  

[winterymix]

... I had rented 5 movies......

The legend of JI grows larger.

[Bob Chill]

GFS predicted temps to be 1-2C for the big cities during the morning and afternoon.  Temps were 1-2C.  Not sure that's a bust.  NAM was cold, Euro was warm.  If we had a sounding at 18z we could verify the temps between the surface and 850mb...but we don't, so we can't.  

 

I think the only meaningful model bust was the evolution from the waa to the ccb. That caused the marginal temps to really show their face. I'm not very far from 95 and the column overhead supported snow for much longer than any rain. It just never came down hard enough. My surface temp was spot on. I knew it wasn't going to get below freezing and not go above 35. That's exactly what happened. 

[87storms]

Are you kidding?  There was tons of discussion on the Euro.  I think most people didn't go "all in" with this storm until the Euro caved to the GFS and brought the storm farther north and got significant precip into DC/Balt.  Please recall that King Euro had DCA getting well over 1" of QPF on the 12z Tuesday run.  People were worried about the low-level warmth on the Euro, but most of us discounted it because of the Euro's BL warm bias.  The Euro DID verify too warm in the BL along I-95.  

 

If the post count is higher on the GFS/NAM runs, it's probably just because those are free runs that everybody can access versus the 6-10 folks on here who pay to have Euro access and then post about it.  

 

let's face it, we all had weenie goggles on.

[wxmeddler]

It was far earlier than 12-24 hours.  Someone...perhaps wxmeddler...posted an amazing graphic of 3 Euro runs in a row.  I think he posted it Tuesday when DT was having the huge argument on his FB page.  King Euro, Lord of Consistency, had jumped WAY north over a 24 hour period of model runs.  I think those 3 runs were 0z Monday, 12 Monday and 0z Tuesday.  

 

It was 03/12z, 04/0z and 04/12z

[87storms]

GFS predicted temps to be 1-2C for the big cities during the morning and afternoon.  Temps were 1-2C.  Not sure that's a bust.  NAM was cold, Euro was warm.  If we had a sounding at 18z we could verify the temps between the surface and 850mb...but we don't, so we can't.  

 

when i say temps, i mean at all layers.  i don't think it was the precip as much as it was the fact that we had some rain drops (or the wettest snow flakes ever) thrown into the snow almost the entire duration.  i think that hurt us much more at night than surface temps.  hard to accumulate when it's a rain/snow mix.

[WxUSAF]

I think the only meaningful model bust was the evolution from the waa to the ccb. That caused the marginal temps to really show their face. I'm not very far from 95 and the column overhead supported snow for much longer than any rain. It just never came down hard enough. My surface temp was spot on. I knew it wasn't going to get below freezing and not go above 35. That's exactly what happened. 

Bingo.  

 

let's face it, we all had weenie goggles on.

I don't think so.  We figured on heavy rates that would bring accumulating snow to the surface.  That happened in RIC and Midlo-land, just not here.  Crappy rates led to crappy non-accumulating snow, RASN and rain.  

 

It was 03/12z, 04/0z and 04/12z

Perfect, thanks.  So that big jump came between 03/12z and 4/0z which was still more than 48 hours before precip started falling.  

[WxUSAF]

when i say temps, i mean at all layers.  i don't think it was the precip as much as it was the fact that we had some rain drops (or the wettest snow flakes ever) thrown into the snow almost the entire duration.  i think that hurt us much more at night than surface temps.  hard to accumulate when it's a rain/snow mix.

It was RASN because the rates sucked.  Did anyone ever post IAD's 12z sounding from yesterday?  I can't prove it, but I doubt the freezing level was ever higher than 950mb.  That was plenty low if we got the precip rates.  

[DDweatherman]

when i say temps, i mean at all layers.  i don't think it was the precip as much as it was the fact that we had some rain drops (or the wettest snow flakes ever) thrown into the snow almost the entire duration.  i think that hurt us much more at night than surface temps.  hard to accumulate when it's a rain/snow mix.

I find that to be off maybe by the slightest bit. One of the things we have to keep in mind is that early on without checking back to IAD's 0z sounding that the mid levels down to probably 950-975mb were cold enough. Stations and people were sitting at 36-40 degrees and reporting snow right off the bat in the WAA precip area. This indicates the warm layer was shallow and with heavy rates there could have been enough cooling were the heavy rates to sustain themselves. 

 

I don't know about any of the other professionals, but all things considered I didn't like the primary low moving into WV/OH into a very poor antecedent airmass. Typically, that isn't the recipe for success and is usually a hinderance. Many didn't mention that factor and even though it likely didn't do too much, I feel it certainly had an effect. 

[87storms]

I don't think so.  We figured on heavy rates that would bring accumulating snow to the surface.  That happened in RIC and Midlo-land, just not here.  Crappy rates led to crappy non-accumulating snow, RASN and rain.  

 

i just didn't think we would need to rely so much on rates.  if that was the case, then i'm not sure why there was such a widespread prediction of snow unless we thought the entire area would be banded nonstop.  

[Ian]

Maybe DT has biased your memory. I can't recall a single solitary model run where DC was the QPF bullseye. Every model had the bullseye SW of DC on a very consistent basis. DC had lots of QPF on most runs, no doubt there (even on King Euro), but was never the bullseye. Maybe the Euro continued to show RIC getting 2-4" of snow while the NAM didn't, but the American models never bullseyed DC.

GFS bullseyed us once or twice a few days out. But it also mainly rained on us the day prior.

[wxmeddler]

I don't think so.  We figured on heavy rates that would bring accumulating snow to the surface.  That happened in RIC and Midlo-land, just not here.  Crappy rates led to crappy non-accumulating snow, RASN and rain.  

Agreed, DCA, IAD, BWI all got 35-50% of their reasonably expected totals, and if you go back and look the gfs, nam, euro, cmc/ggem all had similar precip totals around 2". In the end. IAD got .92", DCA: 1.03" , BWI: .75"

[DDweatherman]

i just didn't think we would need to rely so much on rates.  if that was the case, then i'm not sure there was such a widespread prediction of snow unless we thought the entire area would be banded nonstop.  

And thats just the thing, a nice localized CCB was anticipated for many areas. Yesterday the precipitation was far to "bandy" with the lulls allowing saturation and too much of a moist adiabatic LR. The lack of precip in those on the edge of good bands can partially be thanks to subsidence that was partially a factor in between some of the bands, despite the easterly flow. 

[87storms]

I find that to be off maybe by the slightest bit. One of the things we have to keep in mind is that early on without checking back to IAD's 0z sounding that the mid levels down to probably 950-975mb were cold enough. Stations and people were sitting at 36-40 degrees and reporting snow right off the bat in the WAA precip area. This indicates the warm layer was shallow and with heavy rates there could have been enough cooling were the heavy rates to sustain themselves. 

 

I don't know about any of the other professionals, but all things considered I didn't like the primary low moving into WV/OH into a very poor antecedent airmass. Typically, that isn't the recipe for success and is usually a hinderance. Many didn't mention that factor and even though it likely didn't do too much, I feel it certainly had an effect. 

 

i agree, i never like the primary being that far north.  i remember seeing cincinnati sitting at rain and 39 degrees on tuesday afternoon.  that didn't seem to me to be a good sign.  

 

as for temps, even when the rates were good i feel like there was still some slop mixed in.  there was a large chunk of area expected to get snow, so i guess the mets thought the entire area was going to get major rates.

[gymengineer]

Agreed, DCA, IAD, BWI all got 35-50% of their reasonably expected totals, and if you go back and look the gfs, nam, euro, cmc/ggem all had similar precip totals around 2". In the end. IAD got .92", DCA: 1.03" , BWI: .75"

I think that's a bit of an exaggeration. The NAM had 1.75-2"+ towards the end while the GFS was usually in the 1.5-2" range. The Euro held pretty steady at 0.75-1.25" for the three airports for the last few runs. The Euro was actually not that far in the end.

[winterymix]

I'm glad Wes said this because just as a weenie with very little cred I feel like there's been too much self-doubt among those more experienced over snow-drunkenness or wishcasting or whatever. JUST speaking about this board, when even the most skeptical forecasters overcome innate conservatism, and when mets outside the region w/o emotional investment are thinking the storm will perform, I feel like it's unfair -- and all too easy -- to pin it on wishcasting. Too much attention to model forecasts, and the models themselves, are other questions, and all psychology aside, more nuanced ones. I trust folks like Wes and others here to not just blindly "see what we want" and will continue to do so. :-)

There is always selection bias with the models and data. The NAM (arguably too cold) was giving a forecasted snow sounding in the big cities.

The SREF as has been said was demonstrably showing snow. The 2 meter temperature charts for most if not all of the models were demonstrating

a north-moving surface freezing level throughout much of the storm. (BIG RED FLAG) I don't remember any of the models showing the warm marine layer mixing

in so demonstrably. Who knew the Virginia convection would heavily reduce precipitation to the north? Was it modeled?

The selection bias is manifest by how an expert weighs the various elements of guidance. As has been pointed out, this storm was at a delicate

boundary layer temperature equilibrium.

Just before the storm, the 0Z model suite as well as real time radar screamed THIS IS THE REAL DEAL.

How can it be explained to the public in a forecast that a big snow seems to be on the way but there are

hints that it could fall apart? Almost every good storm has model outliers and guidance that can raise doubts.

The trick is to know how to weigh the outliers and obviously even the most skilled experts can't thread that needle every time.

[87storms]

And thats just the thing, a nice localized CCB was anticipated for many areas. Yesterday the precipitation was far to "bandy" with the lulls allowing saturation and too much of a moist adiabatic LR. The lack of precip in those on the edge of good bands can partially be thanks to subsidence that was partially a factor in between some of the bands, despite the easterly flow. 

 

this all goes back to my theory that we need things simple here to get snow.  whenever it gets complicated and needy is when we run into problems it seems.  this storm was too high maintenance for our area.

[87storms]

It was RASN because the rates sucked.  Did anyone ever post IAD's 12z sounding from yesterday?  I can't prove it, but I doubt the freezing level was ever higher than 950mb.  That was plenty low if we got the precip rates.  

 

i went into this storm thinking we wouldn't need to rely so much on rates to succeed.  i figured moderate snow would be enough, which we did get at times.  maybe we didn't start out heavy enough.  if i remember feb 87 correctly, that storm became heavy quickly, but i also don't remember it being quite as sloppy as this one.  i drove out to centreville yesterday and even past that (no idea where but there was a panera) and even the snow there was one big slushy mess.  

[WEATHER53]

this all goes back to my theory that we need things simple here to get snow.  whenever it gets complicated and needy is when we run into problems it seems.  this storm was too high maintenance for our area.

I agree. Low had to move a bit se then west to east near VA/NC border and then develop, too many issues while it's in it's genesis.  Take a organized, miderate intensity low, move it over Atlanta, thru central SC, and off coast between Norfolk and northenr outer banks with a cold air source to the n-nw and you have got the best shot for good snow in DC.

[DDweatherman]

this all goes back to my theory that we need things simple here to get snow.  whenever it gets complicated and needy is when we run into problems it seems.  this storm was too high maintenance for our area.

That is partially correct, our "secondary" so to say was in a good spot for most cases, though the precip shield was not large enough for us to truly cash in thanks to the energy transfer and such a strong h5 energy tucking the storm into the coast. Miller A's are better for that simple reason. Having to rely on a solid h85 transfer is not the greatest of hopes, and 2/10/10 was an evolution different than anticipated on what, analogously speaking, was not overly different in evolution. The difference was obviously the rate of cyclogenesis, a better antecedent airmass (in addition to cold air being albedo'd back thanks to significant snow cover), as well as a larger mass of precipitation. The initial precipitation dropped less snow than anticipated in that first piece in that event as well, but the area was the beneficiary of a deformation zone that snowed itself out right over central Maryland. 

 

A complex set up? Not fully. It's one that could have worked, however the cons outweighed the pros if you were looking to get a NESIS event (and thats assuming it was earlier on in the winter). Edit to say not a significant NESIS event, even one that was low on the scale. Personally, I don't prefer these set ups. 

[Bob Chill]

I think all aspects of the bust are covered. It was marginal all the way. Everything looked like it was going our way but it just evolved a bit different. 

 

The column supported snow for folks almost at sea level in an near dca. Further into md notsomuch. 

 

It could have been a 6"+ pastejob for dca if it ever got fast and furious and held for 6-10 hours. It didn't get going fast and furious. 

 

No 2 storms are alike and if we got the same type of setup in the future, it could easily end up being a pastejob. It will be doubted until the last flake falls but it's absolutely possible.

 

One thing I don't see mentioned much was the proximity of the ull. That thing was a beast and can overcome many adverse things irt temps between surface and 950. It had no problem doing it in RIC and they were surrounded on all sides by terrible thermals if you wanted snow. We needed the slp to explode a bit futher east and let dynamics knock out the sketchy part of our column up this way. Didn't happen this time. I'll take my chances with that setup at 500 anytime though. 

 

One thing a beastly ull can do on a consistent basis is produce snow when surrounded by air that doesn't support snow. There are a zillion examples of this. 

[wxmeddler]

I think that's a bit of an exaggeration. The NAM had 1.75-2"+ towards the end while the GFS was usually in the 1.5-2" range. The Euro held pretty steady at 0.75-1.25" for the three airports for the last few runs. The Euro was actually not that far in the end.

Noted, and accepted.

[DDweatherman]

I think all aspects of the bust are covered. It was marginal all the way. Everything looked like it was going our way but it just evolved a bit different. 

 

The column supported snow for folks almost at sea level in an near dca. Further into md notsomuch. 

 

It could have been a 6"+ pastejob for dca if it ever got fast and furious and held for 6-10 hours. It didn't get going fast and furious. 

 

No 2 storms are alike and if we got the same type of setup in the future, it could easily end up being a pastejob. It will be doubted until the last flake falls but it's absolutely possible.

 

One thing I don't see mentioned much was the proximity of the ull. That thing was a beast and can overcome many adverse things irt temps between surface and 950. It had no problem doing it in RIC and they were surrounded on all sides by terrible thermals if you wanted snow. We needed the slp to explode a bit futher east and let dynamics knock out the sketchy part of our column up this way. Didn't happen this time. I'll take my chances with that setup at 500 anytime though. 

 

One thing a beastly ull can do on a consistent basis is produce snow when surrounded by air that doesn't support snow. There are a zillion examples of this. 

Absolutely, you see it all the time on the backside in systems in the deep south (i.e. MS/AL). For those areas, obviously the effects are far less reaching (in terms of miles), but do serious damage in crashing the column. Nevertheless, this airmass is better then theirs and I find that the h5 was a beauty indeed, but I didn't like the necessary transfer of the upper levels. More times than not yesterday would have gotten it done for us, especially without the inital LP center heading into the Apps region. It will most certainly be doubted in the future, but the set up despite being marginal is better than some of the complex things we attempt to succeed with. 

[MN Transplant]

We actually had the interesting situation where the GFS was better further out, and then the Euro handled the sensible weather conditions better once it finally figured out the 500 track within 48 hours.

[ORH_wxman]

Will...those temps you showed DID verify too warm for most of the area.  The 35F isotherm was pretty near I-95 for much of the day.  I think the GFS had the surface temps the best, with DCA and BWI between 1-2C for most of the precip.  NAM was obviously too cold and WAY too wet.  It was just the heavy precip never got here.  NAM, GFS and Euro were all too wet.  DCA ended with 1.02", BWI with 0.75" and IAD with 0.92".  GFS and Euro were closest (NAM was about 2 times those numbers), but they were still too wet.  Whether the dynamic situation near RIC robbed us of moisture or the storm tracked and developed too differently, I think that was bigger issue.  We knew surface temps would be above freezing, but figured with heavy precip, the profile would go isothermal right near 32F (much as it did on 1/26/2011) and we'd get plastered.  That is in fact exactly what happened west of RIC to folks like Midlo.  

 

 

Yeah I'm not surprised...they are biased warm...the worry though was that it didn't show 35F temps for I-95 that would have verified more like 32-33F...it showed 36-37F that verified as 35F during lunchtime.

 

At any rate, the precip was definitely an issue. The really intense bands and rates (like 0.15" per hour in the bucket kind of stuff) never seemed to materialize over the DC/BWI area save maybe some fleeting moments.

 

As is often the case, there was probably no single factor late led to the huge bust, but rather a few of them conspiring in exactly the wrong way.