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 1:30 PM ET, 12/15/2010

Weekend snow threat down, not out

By Wes Junker

The potential for a significant snowstorm this weekend is slipping away but there remains enough uncertainty to keep a close eye on the models and forecasts.

During the past two days Jason and I have discussed how the phasing between the northern and southern jet streams had to be almost perfect for us to get a major snowstorm this weekend. The models at that time were in two distinct camps, one favoring a strong storm near enough to the coast to produce a significant storm and another with the low too far out to sea to give us snow.

Because of the consistency of the GFS model in forecasting a decent snow event for 5 straight runs, Jason even brought out the Snow Lover's Crystal Ball (SLCB) giving the event a 30% chance of producing an inch of snow. But I'm putting it back in the closet as I think those odds are just below 30%.

Keep reading to learn why I've put away, at least temporarily, the SLCB...

Alas, last night's higher resolution operational models, the GFS and European, showed no jet stream phasing until the storm got too far into the Atlantic to produce much snow. Note in the two images below how far out to sea both models were with the storm (surface low).

Also, the UKMET (a model from the United Kingdom), which had been a proponent of a more wound-up coastal storm last night, also shifted into the out to sea camp. Collectively, the suite of high resolution models appear to be heading towards a consensus away from a snowstorm.

ukmet-121910.gif
Last night's 7:00 PM runs runs of the GFS (left) and the European right showing the 500 mb pattern (the shading) and the mean sea level pressure pattern. The GFS map is valid at 1:00 PM on Sunday and the European model plot is valid at 7:00 PM Sunday evening. Source: Raleigh Weather

However, the latest (12z) GFS run does give us a little snow as it did trend back a little closer to the coast. So, an out to sea track is not a done deal. But even with its slightly more favorable look, the 12Z GFS still has the low development a little too far east for a solid snow hit. Taken literally, it hints that lighter snows could impact the region Saturday night into Sunday with a chance of more substantial snows toward southeast Virginia.

For those snow lovers that need a stronger glimmer of hope to keep them interested and excited: a number of ensemble members from this morning's GFS model run did still show a storm as shown in the figure below. Remember, ensemble members are run to give an idea of what might be possible if the operational model's initial conditions are a little different than were used in the operational model run. This figure probably does an adequate job of assessing the potential for a storm this weekend though I'd probably lower the probabilities a little in deference to the solutions from the other operational models.

ensembles-121910.gif

Display of 12 individual GFS members showing the surface pressure, 1000-500 mb thickness (an indicator of the mean temperature of the layer) and precipitation forecast for Sunday at 7 a.m. Source: Penn State (See this image bigger.)

My subjective guess is that the probability of getting over an inch of snow is just 20 to 30 percent with the odds of getting a major snowstorm are even less. However, the forecast certainly is not cast in stone.

By Wes Junker  | December 15, 2010; 1:30 PM ET
Categories:  Latest, Winter Storms  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: Snow threat up, 1 or 2 inches possible tomorrow
Next: What are your snow predictions?

Comments

The latest European model also has the storm out to sea -- this supports Wes's "downgrade" of this threat...

Posted by: Jason-CapitalWeatherGang | December 15, 2010 1:55 PM | Report abuse

C'est la vie.

Posted by: Rcmorgan | December 15, 2010 2:05 PM | Report abuse

Ouch...had a bad feeling about this one as a snow lover, but the explanations are very cool to read...love these columns and thanks for the update Jason!

I'm still holding out some form of hope for a big snow as I have no plans this weekend.

Posted by: parksndc | December 15, 2010 2:09 PM | Report abuse

Don't make me break out the Bing Crosby, my secret weapon in the white Christmas arsenal :-p

Posted by: TheAnalyst | December 15, 2010 2:11 PM | Report abuse

simply, :(

Posted by: MaxPowerDJ | December 15, 2010 2:13 PM | Report abuse

Not unusual 4 us 2 miss big storms, last yr everything came 2gether prefectly, so the odds on having it happen this yr r slim. I would say it's usually about a 25% chance on any coastal storm being the prefect storm. Guess we got spoiled last yr. Storm tracks can b subject change, so maybe this 1 will track back E.

Posted by: VaTechBob | December 15, 2010 2:19 PM | Report abuse

If you want a more bullish (pro-snow) analysis for the weekend, Joe Lundberg at AccuWeather shares an interesting perspective: http://www.accuweather.com/blogs/Lundberg/story/43067/the-weekend-sto.asp

Posted by: Jason-CapitalWeatherGang | December 15, 2010 2:45 PM | Report abuse

this storm is still, what, 4 days away? we may not want to be in the sweet spot as those things have a away of changing. if all the model runs kept dumping tons of snow on us i'd be worried this thing would bust. as i mentioned on that last thread, one thing that's highly likely is that where the sweet spot is now in the models won't be where it ends up in actuality.

Posted by: walter-in-fallschurch | December 15, 2010 3:12 PM | Report abuse

Whether it misses us or not, the storm is guaranteed to generate a week of cold dry windy weather afterward. Thanks a lot Ms La Nina and Mr NAO.

Posted by: eric654 | December 15, 2010 3:18 PM | Report abuse

I am writing this message 12/15/10 at 3:40pm. Does this model include waht's on tap for this Thursday and Friday? Thanks.

Posted by: rusty6 | December 15, 2010 3:39 PM | Report abuse

Thanks for the link, Jason. Interesting.

Posted by: Rcmorgan | December 15, 2010 3:40 PM | Report abuse

The GFS ensemble from 12Z this morning has notably increased the the odds for snow this weekend from yesterday's runs. A direct readout from the ensemble output places the DC Metro region having a 30-50% chance for .5" water equivalent (~ 5-6" snow), and has increased the odds for up to 1.0" (~10-12") from 00% to 20%.

It's still a difficult call, but perhaps the SLCB is about to make a reappearance.

See: http://cms.met.psu.edu/sref/ensembles/
Select: on left, select "GEFS", Date (201215), Cycle (12 ==> run from 7AM), parameter (POP240.5 for .5 in and 241.00 for 1"); top is probability, bottom is spaghetti of area encompassed by each ensemble member with .5" and 1", respectively. Sorry about this, we'll try to get the actual figures posted as a simple one step link ASAP.

Posted by: SteveT-CapitalWeatherGang | December 15, 2010 3:41 PM | Report abuse

Walter...very unscientific but I appreicate and agree :)

Posted by: parksndc | December 15, 2010 3:44 PM | Report abuse

Walter and parksndc

Even if the models forecast the basic features of a storm, it's virtually impossible to get the small scale detail of a "sweet spot" right (if there is one), even in forecasts less than 12-24 hours in advance - let alone 4-5 days in advance. For all intents and purposes its solely a matter of chance (unpredictable).

Posted by: SteveT-CapitalWeatherGang | December 15, 2010 4:17 PM | Report abuse

Posted by: SteveT-CapitalWeatherGang | December 15, 2010 4:20 PM | Report abuse

Sorry I've missed much of the discussion. Despite the gfs ensembles. I'm not ready to change my probabilities. The European model and its ensemble members are emphatic that the system will go out to sea. If anything they have done the opposite of the GFS and its ensembles and are out to sea. Looks to me like the 1*Z gfs operational also is going with a suppressed look until the low is well offshore. For us to get a storm, especially a decent one, it's still a thread the needle situation.

Posted by: wjunker | December 15, 2010 5:13 PM | Report abuse

Posted by: wjunker | December 15, 2010 5:26 PM | Report abuse

Walter, your backward logic might work once in a long while, but that's not the way to bet. In fact, I wouldn't mind sitting with you at a poker table. :)

By the way, the global models have been exceptionally accurate over the last 30 days, with the 5-day northern hemispheric 500 mb height scores being 93% for EC and 92% for UK and GFS. That doesn't necessarily mean a lot with regard to this system, of course.

Posted by: imback | December 15, 2010 5:31 PM | Report abuse

I had trouble posting this comment. The 18Z gfs has come in east of the 12Z version and now has the precipitation to our east as it doesn't phase the system until it is well east of the coast.

Posted by: wjunker | December 15, 2010 5:38 PM | Report abuse

imback,
i'd probably enjoy sitting at a poker table w/you too (but would rather play chess), but i warn you i'm much more logical when it comes to things not snow or redskin related...

wes,
you're talking about sunday's potential storm, right? what you just said about 18z being even further east means NO PRECIP AT ALL for us on sunday?

Posted by: walter-in-fallschurch | December 15, 2010 6:17 PM | Report abuse

Walter, that's what the operational 18Z says. I don't think that changes my odds one 18Z esnemble member still has a big hit and a couple of others look like the gfs earlier today. That's one reason I tend to sometimes be slow changing the probabilities based on one run. It's sort of like NHC being slow to move their tracks based on one cycle of progs as they might shift back the next run.

I was really worried about my probabilities when I saw the 12Z gfs ensemble mean which was quite wet but then the euro and it's ensemble mean were way east. Still despite my rather tepid feelings for the event, it's far from over and won't be until probably sometime tomorrow. Until then, my probabilities are just educated guesses.

Posted by: wjunker | December 15, 2010 7:37 PM | Report abuse

The discussion above is manifestation of the fact there are uncertainties (level of confidence) in estimates of uncertainties.

For the record, my reference to the 12Z ensembles in comment above was a direct readout of the raw model output, which serves as guidance (not the final prediction) to forecasters.

There are few like Wes with the background, experience, and expertise to translate model guidance to an actual ("value added") forecast, including use of ensemble output. I'm most certainly not amongst those few.

Posted by: ensemblemean | December 15, 2010 8:36 PM | Report abuse

Steve, Thanks for the nice complement, I didn't realize you had made it on the blog.

The 00Z NAM continues the trend of teh 18Z gfs towards too little too late for our area this weekend. That's just an update of the models and doesn't change the probabilities mentioned in the original article.

Posted by: wjunker | December 15, 2010 10:06 PM | Report abuse

The last three runs of the gfs have thrown in the towel concerning the weekend snow threat.

Posted by: wjunker | December 16, 2010 9:02 AM | Report abuse

When you say 30% odds do you really mean a 30% probability? They are different quantities so if you are referring to a probability can you state it as such? (odds= probability/(1-probability)).

Posted by: merglerm | December 16, 2010 5:31 PM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company