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 11:20 AM ET, 12/17/2010

On-again, off-again Sunday snow is...

By Dan Stillman

Accumulated precipitation forecast from the GFS model for the 36 hours ending at 1 p.m. Monday. Credit: NCEP.

...more off again than on again. But anyone who's been following our continuing coverage of this weekend's storm threat knows that even less than 48 hours from the storm's potential start would be too early to give up on it completely.

For those who haven't been following so closely, here's the gist of what's been going on: The various models we use to help guide our forecasts have been wavering back and forth for several days now between a coastal storm that passes well out to sea, and one that comes close enough to threaten a major snowfall for the mid-Atlantic into the Northeast. In other words, as our winter weather expert Wes Junker aptly put it yesterday, this storm has been a "huge headache," at least from a forecaster's perspective.

The latest model runs, from this morning and last night, paint a less snowy or snow-less picture with the storm center tracking pretty far off the coast, thus lowering the odds of accumulating snow (1" or more) in the metro area Saturday night into Sunday to about 25%.

Keep reading for more on this "headache" of a storm...

As Wes explained yesterday, what's needed to get the storm to hug and ride up the coast, and bring significant snow to the D.C. area, is a sharp enough dip in the jet stream. Such a dip has appeared in some runs of the various models over the past several days. But looking at the sum total of all models and all runs in recent days through this morning, the overall edge and latest trend goes to the less snowy, too-far-off-the-coast scenario.

Wes agrees: "The preponderance of model guidance shows the storm going out to sea," he said after viewing this morning's model output.

So, we're left with a situation in which the odds are tilted toward only light snow or no snow in the D.C. metro area, with increasing chances (but nothing definite) of accumulating snow heading east/southeast across the Eastern Shore (especially the lower Eastern Shore). But also a situation that still requires careful monitoring.

And a dose of Tylenol for that forecaster's headache, too.

Next snow chance: As we've hopefully made clear above, we'll be watching the weekend storm threat closely in case it starts heading back to the west. In the meantime, we also have our eye on a chance of light snow Monday night into Tuesday from a system that may be similar to the one that brought yesterday's minor accumulation. Stay with us through the weekend for the latest.

By Dan Stillman  | December 17, 2010; 11:20 AM ET
Categories:  Latest, Winter Storms  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Forecast: Continued cold. Any more snow?
Next: One year ago: the eve of Snowpocalypse


yee-haw--that is good news!

Posted by: weatherdude | December 17, 2010 11:30 AM | Report abuse

Oh, the drama! Now it looks like even Boston might not get its snow.

Posted by: KBurchfiel | December 17, 2010 11:32 AM | Report abuse

Weekend storms are the best.

Posted by: FIREDRAGON47 | December 17, 2010 11:57 AM | Report abuse

It sure is a headache not being able to pin things down even now for a weekend snow event.

But, I think a better description of the situation facing forecasters is it being a humbling experience, i.e., coming face to face with the limits of weather science and model capabilities. Especially important is readily acknowledging those limitations and communicating the nature and degree of uncertainties in the forecasts.

Just as importantly is the ability - routinely demonstrated by CWG forecasters - in discriminating the relative degree of confidence from one case to another. Some are "in the bag" up to 5-7 days in advance, and others, like the present situation, remain headache generators to the bitter end.

Posted by: ensemblemean | December 17, 2010 12:10 PM | Report abuse

Let me make the forecast for this weekends storm for you...

OTS, not a flake!

There you go CWG, theres your weekend forecast.

No blocking, no high and the jet stream.

Posted by: KRUZ | December 17, 2010 12:43 PM | Report abuse


There is blocking (see the high over greenland shown here: -- but you're right about no high over southeast Canada - which we like to see for a mid-Atlantic coastal...and jet stream phasing too late.

Posted by: Jason-CapitalWeatherGang | December 17, 2010 12:50 PM | Report abuse

Allright Jason - the enormous amount of moisture heading into CA this weekend - I see a high qpf of 14 inches of liquid in the Sierras (holy *&^*@!) has go to go somewhere. Thats a lot of energy...a Christmas storm has been popping in and out of the GFS for a few days now. What happens to that moisture and atmospheric dynamic when it gets to the East Coast? NAO should still be negative, right?

Posted by: DullesARC | December 17, 2010 1:07 PM | Report abuse


When we're already so cold, high pressure over southeast Canada (helpful to lock the cold air in place) isn't necessarily a prerequisite for a big storm here. Last winter we saw that we don't always need strong high pressure to the north to get a major snow. But as Jason mentions above, the jet stream phasing too late may prove to be the reason this storm is ultimately a miss, if that ends up being the case. -Dan, CWG

Posted by: CapitalWeatherGang | December 17, 2010 1:08 PM | Report abuse


Alot of people have said we dont need that high pressure over Canada to get a BIG storm ALA 2009. But if I may id like to respectfully remind everyone that this is NOT 2009!

Posted by: KRUZ | December 17, 2010 1:23 PM | Report abuse

New Euro even further out to sea.. Two runs in a row with no precip north of NC/VA border. Back to pessimism winning in D.C. winter -- all is back to normal! ;-)

Posted by: Ian-CapitalWeatherGang | December 17, 2010 1:32 PM | Report abuse

What a mess--especially with Christmas so close--and the continuing anomaly of never having had more than 5 inches of snow on Christmas, but over a foot [in 1987] on Veterans' Day!

However I still have two or three dances this weekend & Tuesday before a Christmas weekend nearly devoid of dance activity, and most people I know out of town.

Obviously the ideal time for a coast-hugging major "coastal", with 8-16 inch accumulations expected here, would be Dec. 23-24, though such a "coastal" would really create issues at the airports around here as people try to leave town for other destinations. However we would be GUARANTEED a white Christmas, unlike last year when it started melting before Dec. 24.

Posted by: Bombo47jea | December 17, 2010 1:37 PM | Report abuse

KRUZ - I'm not following your's not 2009, correct. It's also not 2008, 2007, 2011 (yet), 2001, and 2050. What's your point? Just b/c it's 2009 means you now need a HP over the ole' Maple Leaf?

Posted by: parksndc | December 17, 2010 1:46 PM | Report abuse

Hi CWG. You might want to tell your colleagues at the Washington Post that they should not link this story to the banner that says more snow is possible this weekend!

Posted by: erbele | December 17, 2010 1:55 PM | Report abuse


What i meant is the 09-10 winter was a one in a decade type winter. And unlike the 09-10 winter, THIS winter we are in a moderate La Nina.

Posted by: KRUZ | December 17, 2010 1:56 PM | Report abuse

Liking this Euro run.

Posted by: ThinkSpring | December 17, 2010 2:02 PM | Report abuse

Let me add (with tongue firmly embedded in cheek), as the odds apparently continue to decline for a weekend snow storm, the opportunity for a surprise snow storm are continually increasing!

Posted by: ensemblemean | December 17, 2010 2:02 PM | Report abuse

Thanks Kruz but if you have been following...we're all pretty well aware of that.

Posted by: parksndc | December 17, 2010 2:11 PM | Report abuse

The wavering of the models prompts me to ask the following question. How do meteorologists determine, or calculate, the "accuracy" of various models? It seems to me one could say that regardless of what happens this weekend, the models will be both right and wrong...depending on which runs you look at. I suppose that one looks at the different run times and determines accuracy based on time to event -i.e. at 48 hours out, it was accurate x% of the time? And is there some official criteria for determining that the weather matches the predictions? For instance, if a model says we will get a high of 32-36 and we get 37, how is that calculated?

Posted by: amaranthpa | December 17, 2010 2:15 PM | Report abuse

I am also curious about the winter forecast. You seem to have nailed it so far with a colder-than-normal December. When will warmer-than-normal temps predicted kick in?

Posted by: weatherdude | December 17, 2010 2:21 PM | Report abuse

Last winter was a once-in-a-lifetime as far as snow goes around here, not once-in-a-decade...

Posted by: weatherdude | December 17, 2010 2:26 PM | Report abuse

12z GFS ensembles still sorta mixed (operational is the top left image) -- fair number imply at least a flake or two. Not sure I buy it but...

Posted by: Ian-CapitalWeatherGang | December 17, 2010 2:33 PM | Report abuse


Just curious. If the right conditions do appear after all *fingers crossed* and we see the dip we need to bring the storm ashore, how much precip would we be looking at?

Posted by: Alexandria2009 | December 17, 2010 3:20 PM | Report abuse


If we got a big enough dip, we could get into a 2-4"+ situation...but odds of that less than 10%.

Posted by: Jason-CapitalWeatherGang | December 17, 2010 4:18 PM | Report abuse


Yes-- the NAO looks to be negative for a while longer -- which gives a fighting chance for a storm. I've seen the 12/25 GFS storm but it's still in fantasy timeframe. Wouldn't it be fun though if we did get a Christmas storm?

Posted by: Jason-CapitalWeatherGang | December 17, 2010 4:24 PM | Report abuse


Model verification is a specialty area of meteorology that I don't know a ton about. But scientists do this work and assign skill scores to the models. Steve Tracton or Wes could probably tell you more about it.

Posted by: Jason-CapitalWeatherGang | December 17, 2010 4:28 PM | Report abuse


Some of the long-range analyses suggest warming in early January as the negative NAO reverses.

Posted by: Jason-CapitalWeatherGang | December 17, 2010 4:31 PM | Report abuse

More RAIN please to wash this remiander of crap away.... Snow and the street treatments to keep them clear just creates potholes and craters. The salt and other chemical agents crap are corrosive and sticks to everything (including your car), and to top it all off, driving and walking in the stuff can be trecherous.

Posted by: rkayblock1 | December 17, 2010 5:09 PM | Report abuse

There are lots of ways to measure performance. You can do correlations between the observed and forecast fields, use the root mean square error. Hre's a web page that shows the different methods of verification that can be used. It's pretty mathmatical but you can get the idea that verification is indeed done regularly by the National Weather Service and other government forecast services around the world.

Posted by: wjunker | December 17, 2010 5:11 PM | Report abuse

To echo @erbele, the consensus over the past few days has been that this weekend's coastal low will miss us to the east - and it appears that rather little has changed in the past 12 hours to alter that assessment, so why the link from the homepage touting a possible weekend storm?

Posted by: prickles1009 | December 17, 2010 6:31 PM | Report abuse

As I mentioned in a comment a couple days ago, when the issue of verification also came up, I'll be doing a post on this subject in the near future with the intent of conveying key principles, concepts, results and value without getting into mathematical esoterica.

Posted by: ensemblemean | December 17, 2010 7:46 PM | Report abuse

As of this writing (DEC. 17TH 2010), we think the most likely scenario is the following:

Saturday/Saturday night…
- Light snow, turning to heavier snow later on into the Virginia, Washington D.C., Baltimore, and New York areas.

Sunday/Sunday night…
- Snow increases in intensity and quickly swells into Snowmageddon. Lake effect goes into full swing with half of the storm sitting stationary over the Northeastern states, while half sits over the Atlantic ocean gathering more strength.

-Storm remains stationary straddling the eastcoast and Atlantic ocean which translates into about 12 feet of snow fall between Monday and the end of the week.

Posted by: tedleski | December 17, 2010 9:27 PM | Report abuse

prickles1009 and erbele

I agree. I've asked the local home page editors to remove mention of snow. Earlier today though, we still had a slight chance in the that's why it was there...

Posted by: Jason-CapitalWeatherGang | December 17, 2010 10:34 PM | Report abuse


Oh that would be fun :)

Posted by: Meana1213 | December 17, 2010 11:58 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company