Network News

X My Profile
View More Activity
The new Washington
Post Weather website
Jump to CWG's
Latest Full Forecast
Outside now? Radar, temps
and more: Weather Wall
Follow us on Twitter (@capitalweather) and become a fan on Facebook
Posted at 1:15 PM ET, 01/21/2011

Potential storm Tues-Wed: more snow or rain?

By Wes Junker

The models are converging toward the possibility of a winter storm Tuesday into Wednesday. The cold air mass settling over the region now now is setting the stage for wintry precipitation. However, like most such storms projected so far into the future, there is a lot of uncertainty about the storm's track and intensity.

Next accumulating snow chance: Tuesday and Wednesday
Probability of more than 1" of snow: 30-40%
Probability of more than 4": 10-20%

A number of models last night forecast the storm to start as snow but then change to rain as the storm takes on an inland track. But some model runs have suggested more of a coastal track and mostly snow. The majority of the models suggest this storm will have quite a bit of precipitation with it. How much of the precipitation will be snow, sleet or rain is very much in question.

TECHNICAL DISCUSSION

Last night's European (0z) and GFS (06z) models had very similar solutions. Because the two operational models are so similar, I've decided to only show the GFS solution (see to the right, below). Both the GFS and Euro developed and closed off a 500 mb (upper level, 18,000 feet) low to the southwest and then tracked the 500 mb low almost right over the city. Usually for an all snow event you would like the low to close off later and track a little farther south and east than those models were forecasting.

gfs-0125.jpg
Last night's GFS (6z) model forecasts of surface pressures and 850 temperatures (top panels) and 500 mb heights and vorticity (spin at around 18000 ft, bottom panels). Valid times 7 p.m. Tuesday (left panels) and 7 a.m. Wednesday (right panels)

Their forecast 500 mb low track (bottom panels to right) helped pull a very intense surface low (top panels to right) far enough west to bring in warm air off the ocean. Note on the top right hand panel that green colors overspread the Washington area indicating that temperatures at around 5000 ft warm to above freezing. Although such a track allows the 850 temperatures on both the GFS and Euro to rise above freezing, a slight eastward shift in the track could produce a snowier solution rather than the current one which still offers several inches of snow before mixing and changing to rain.

There remains considerable uncertainty about the storm track and most likely type of precipitation. Why am I being so wishy-washy when the operational models last night in such good agreement about a changeover with the the storm?

ensembles-1-25.jpg
Spaghetti diagram showing ensemble members valid 1 a.m. Wednesday morning.

The ensemble members, shown to the right, continue to show considerable uncertainty about the development and timing of the 500 mb trough (dip in the jet stream at 18,000 feet) that helps drive the development of the system. Remember that ensemble members use the same model with slightly different initial conditions and are used as a gauge of how much uncertainty there is in a solution.

If you compare the dip in the white line (amongst the green) which is the control run (that has the same initial conditions as the operational run but with lower resolution) to the various green lines (the various ensemble members), you can see how variable the positioning of the trough is - the main player in the development of our potential storm. Some are much faster with the system, others slower. The strength, timing and orientation of that trough will determine the ultimate track and development of the storm. The earlier the upper low closes off and takes on a negative tilt (which directs the storm inland, bringing in milder air off the ocean) the more likely it will be for the storm to produce more rain than snow. The later it takes on a negative tilt, the better the chances for a snowstorm.

gfs-012611.jpg
This morning's GFS 850 mb height and temperature forecast valid at 1 a.m. Wednesday and the total precipitation simulated over our area by the GFS .

The morning's GFS has shifted toward a snowier solution but requires an initial low pressure system in the Ohio Valley to reform into another much stronger low near the coast. Such a storm evolution, sometimes referred to as a Miller B development, often puts us near the dry slot. In fact, that has happened several times already this winter. But right now the model is forecasting the coastal low to form far enough south to produce significant precipitation over our region. And importantly, the upper level low at 850 mb (at 5000 feet) (shown above, left side panel) takes a perfect snow track just to our east such that cold air stays in place and model ends up giving us 8-10 inches of snow (right side panel indicates about .8-1.0 liquid equivalent precipitation).

I suggest readers not get too excited yet about today's GFS solution. It is liable to change on the next run.

The latest Euro run (12z) indicates a period of snow at the onset, but tracks the rapidly intensifying low far enough inland such that a good deal of the precipitation falls as rain, except toward the mountains (along and west of I-81). It would suggest the precipitation might change back to the snow before ending once the low passes to our north.

Any of the model solutions offered above are still possible as the ensemble members are showing significant spread in their handling of the upper trough that will determine our fate. The one thing that looks likely is that there will be a storm. Beyond that it's all guess work.

In summary, there is potential for a significant winter storm somewhere in the mid-Atlantic region. How much snow and winter weather it might bring is still up in the air. My own assessment is that a snow to rain scenario seems more likely than all snow at this time. Over the weekend, I'll take another look at the storm. Hopefully by then the situation will be clearer.

By Wes Junker  | January 21, 2011; 1:15 PM ET
Categories:  Latest, Snow Lover's Crystal Ball  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: AM round-up: last night's "storm", cold & more
Next: PM Update: Super cold tonight

Comments

Wes: Nice explanation, thanks! Can you comment on how the 12Z Euro looks today? It should be coming in about now, I think.

Posted by: Kevin-CapitalWeatherGang | January 21, 2011 1:26 PM | Report abuse

@Kevin

We added in some discussion of the Euro. It would suggest to snow to rain. Pretty intense inland track.

Posted by: Jason-CapitalWeatherGang | January 21, 2011 1:29 PM | Report abuse

Given this news:
http://voices.washingtonpost.com/fasterforward/2011/01/duke_nukem_forever_ships_may_3.html#comments

I, personally, am preparing for a snowpocalypse. Also, an umbrella to turn aside the rain of droppings from airborne ovines might be a good idea...

Posted by: wiredog | January 21, 2011 1:29 PM | Report abuse

Kevin, if anything the euro has gone with a warmer run than last night's run. The GFS and its ensembles are outliers compared to the GGEM, UKMET and Euro. It will be interesting to see what the euro ensemble mean does.

Posted by: wjunker | January 21, 2011 1:33 PM | Report abuse

wes,
on the previous thread several posters and even SteveT of CWG said there was something wrong with the 12z GFS output and that we should basically discount it. what say you?

Posted by: walter-in-fallschurch | January 21, 2011 1:34 PM | Report abuse

Walter, I think it is less likely than the a look like the previous runs or than the euro but I'm not good enough to completely toss is out. Right now this looks more like a snow to rain event than a pure snow event but there is enough uncertainly not to get too caught up in the details this early.

Posted by: wjunker | January 21, 2011 1:45 PM | Report abuse

wes,
to continue that thought, i saw you said not to get too excited about the run because it will probably change with the next run. but is that just due to the normal variations from run to run? or because something was "wrong" with the run?

Posted by: walter-in-fallschurch | January 21, 2011 1:46 PM | Report abuse

What gets me...these storms keep targeting my Tuesday night dances...while TominMichiganPark keeps complaining about "not enough rain".

At least if it goes over to all rain by Tuesday evening the roads aren't as slippery and the buses will run on time...though even rainy weather can keep dance crowds down.

Posted by: Bombo47jea | January 21, 2011 1:52 PM | Report abuse

Walter, possibly both. As Steve noted, the model develops a secondary low due to latent heat release that probably isn't very realistic. That's what Steve was alluding to. There also is the run to run variability due to chaos which you can see when looking the spaghetti diagrams.

Posted by: wjunker | January 21, 2011 1:54 PM | Report abuse

I think it will rain on the coastal plain
and snow on the valley and applechains

Posted by: pvogel88 | January 21, 2011 1:56 PM | Report abuse

well HM from accuwx has us in the "Big Daddy" snow storm on his map for next week. But of course he's been promising BIG snow 5 days out on every other storm around here since late november :/

http://vortex.accuweather.com/adc2004/pub/includes/columns/margusity/2011/590x443_01211744_severe1.png

Hype. Wait till monday to get excited.

Posted by: KRUZ | January 21, 2011 2:04 PM | Report abuse

wes,
thanks for the replies. i am really not liking the idea of a snow-to-rain event where the rain washes away all the snow. if i interpreted it correctly, i think the previous GFS did a snow-to-rain-to-snow prediction - with all the heavy precip being rain... that was no good. that's what i liked so much about the 12z run - it was less precip, but all snow... i guess i'll just wait for the next 20 runs... (like i have a choice!).

bombo,
hahaha... i knew the tuesday timing would irritate you, but we've got to take these snow chances when we can get them.

Posted by: walter-in-fallschurch | January 21, 2011 2:12 PM | Report abuse

If it washes my car I'll be happy.
If I get to use my new shovel I'll be even happier.
Will hyperventilate over the weekend on this flickering snow prospect. Maybe it will help keep me warm.

Posted by: FIREDRAGON47 | January 21, 2011 2:33 PM | Report abuse

Is there a link to an average annual snowfall map in the area on the site? I found one on http://www.erh.noaa.gov/er/lwx/Historic_Events/md-snow-avg.gif but not sure how accurate it is. It shows places in Northeastern Maryland past Bmore averaging less than DC as well as the WV panhandle receving less than northern Baltimore county. Thoughts?

Posted by: lobp | January 21, 2011 2:39 PM | Report abuse

Any guesstimate as to the onset of this storm? Planning to travel by air on Tuesday.

Posted by: griffin1108 | January 21, 2011 2:45 PM | Report abuse

griffin1108, there is way too much uncertainty about the evolution of the system to offer a guess at when it will start.

Posted by: wjunker | January 21, 2011 2:51 PM | Report abuse

you know, nothing makes my afternoon quite like a SLCB. Now let's hope it stays all snow!

Posted by: bachaney | January 21, 2011 3:17 PM | Report abuse

Gang - going waaay back, has there ever been a report of getting a HECS in consecutive winters in DC? This would be our equivalent of the 2004-2005 hurricane seasons.

Posted by: DullesARC | January 21, 2011 3:20 PM | Report abuse

Wes,
Thanks for the discussion and explanation. I learn more with each one posted.

Posted by: mscofield1 | January 21, 2011 3:40 PM | Report abuse

My Mother is flying into BWI Wednesday. Any indication that this storm will be big enough to make it difficult for her to get into DC via metro?

Posted by: DC607 | January 21, 2011 3:48 PM | Report abuse

I noticed that the NAO and AO are transitioning from a negative to a positive state around this storm's time period. I heard that a lot of winter storms occur when these two teleconnections go through a transitioning phase (switching from negative to positive). CWG or anyone else, why do you think that happpens? Thanks!

Posted by: Yellowboy | January 21, 2011 4:05 PM | Report abuse

RAIN!!!!

Posted by: SouthsideFFX | January 21, 2011 4:22 PM | Report abuse

My Mother is flying into BWI Wednesday. Any indication that this storm will be big enough to make it difficult for her to get into DC via metro?

Posted by: DC607 | January 21, 2011 3:48 PM | Report abuse

_______________________________

Yes. There are also indications that it won't be big enough.

Posted by: AsymptoticUnlimited | January 21, 2011 4:36 PM | Report abuse

Yellowboy,

I've asked the same question before and never got a response. Hopefully someone in the gang has some more insight regarding the correlation between east coast cyclogenesis and the predicted NAO/AO phase change.

Posted by: triplephaser | January 21, 2011 4:44 PM | Report abuse

I still want to know why it is that in a a typical weather year...many snow chances, but nearly every one disappears as the "real" time weather closes in. It seems the models show potential for snow the long term, but for the actual event, predicted amounts diminish and diminish and then the dreaded dry slot hits, of course, usually over Manassas. Why is it that it never happens the opposite way? Why don't the models EVER show nothing, nothing, nothing and then SURPRISE!!!! There's a bunch of snow right here, right now? What's up with that?

Posted by: manassasmissy | January 21, 2011 5:15 PM | Report abuse

I believe the all levels are going to be well below freezing when the precip starts. We will probably start out with at least a few inches of snow.

The big danger may be if it warms up aloft and the low levels lag, someone in our general area could get a significant ice storm.

It may not be a pure rain vs. snow situation. It could be multiple precipitation forms during the course of the storm.

Posted by: frontieradjust | January 21, 2011 5:15 PM | Report abuse

I think you have the transitions reversed.
I lifted this from an abstract "Results of the statistical analysis suggest that cool-season NE precipitation tends to be enhanced during positive-to-negative NAO and negative-to-positive PNA transitions, and suppressed during negative-to-positive NAO and positive-to-negative PNA transitions. " I think the reason precipitation is enhanced when the nao phase changes from positive to negative is simply the fact that the storm is a monster and helps establish the blocking high.

Posted by: wjunker | January 21, 2011 5:23 PM | Report abuse

The new GFS run is very cold, thus changeover might not even be an issue.

Posted by: frontieradjust | January 21, 2011 5:25 PM | Report abuse

manassasmissy, I think we have selective memory, the dec 19th blizzard last year didn't show up until around 72 hrs before the event. The same held for the Jan 31 snow event.

the 18Z gfs held onto a similar solution to the 12Z version, plenty cold enough for snow and a fairly suppressed storm track. It continues to be an outlier compared to the other model guidance.

Posted by: wjunker | January 21, 2011 5:28 PM | Report abuse

Most of the time when the storm track is iffy, go with the snow 2 rain event. Right now, this is looking like the typical snow 2 rain storm. On the average, it snows 18% of the time & sleets/fr. rain/rain 82% in this area. When in doubt go with with more rain than snow. Hopefully things will swing more 2 a snow event, but as of now, I'm not expecting a big snow. Wouldn't b surpised if it's another snow bust.

Posted by: VaTechBob | January 21, 2011 5:30 PM | Report abuse

SouthsideFFX - Not sure why you like rain more than snow...they both have challanges...just a question.
Wes - Thanks for your awesome detailed analysis once again. I think your analysis is just the right temperment given what we understand today. At this point, I'm assuming we have a very little chance of more than an inch of snow and likely will see mostly rain. That being said, given the consistent pattern of storms this winter, I would assume it'll be little precip, once again.

Posted by: parksndc | January 21, 2011 5:32 PM | Report abuse

"I believe the all levels are going to be well below freezing when the precip starts. We will probably start out with at least a few inches of snow.

The big danger may be if it warms up aloft and the low levels lag, someone in our general area could get a significant ice storm."

Ugh, then whatever snow falls could have a thick sheet of ice over top of it. If that happens, either hope for a very fast warm up or be ready for some very tough shoveling, scraping, and plowing, even if snow accumulation is only minor.

Posted by: mkarns | January 21, 2011 5:45 PM | Report abuse

Yellowboy, more on the going from a negative to a positive nao. Kocin and Uccellini have shown that a number of major snowstorms have the nao rise during them. I suspect that the weakening of the block allows the storm to come farther north than when the block is very strongly negative. When the noa is strongly negative the storm track is more likely to be suppressed. Sorry for the confusion as there were two different studies. One concerning precipitation and another concerning major snowstorms.

Posted by: wjunker | January 21, 2011 6:57 PM | Report abuse

At the present time it looks like 1-2", then rain. This time the L needs 2 move farther E. Present track brings in 2 much warm air aloft. Of course, models aren't that good more than 2 days out & even then its not a done deal.

Posted by: VaTechBob | January 21, 2011 8:16 PM | Report abuse

I have a general question about the models. Would models be more accurate if there was an infinite amount of atmospheric data. If it were possible to have all of the atmospheric information from every point on Earth would the models be closer to an absolute certainty or would they still be inherently unreliable?

Posted by: jbernard703 | January 21, 2011 9:16 PM | Report abuse

jbernard703, I'm not a modeling expert but do think increasing the number of observations would improve the accuracy. That said, it's cost prohibitive to have observations at every point on the globe sO i don't think deterministic forecasts will ever be accurate at the longer time ranges in my lifetime or yours

Posted by: wjunker | January 21, 2011 10:54 PM | Report abuse

If an infinite number of models had an infinite number of computers, one of them would type the complete works of Shakespeare....

Yet it still wouldn't snow in DC.

Posted by: AsymptoticUnlimited | January 21, 2011 10:57 PM | Report abuse

Wes-
To make up for this year's lack of snow, I am just hoping for a WARM Spring that starts early with lots of rain....making our skin nice and dewy....who needs that facial???

Posted by: moo1 | January 21, 2011 11:03 PM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company