wpostServer: http://css.washingtonpost.com/wpost

The Post Most: Local

Posted at 11:00 PM ET, 03/03/2013

Snowquester live blog: Tracking the models, March 3 evening edition

11:00 p.m. update: We’ve reviewed the latest model data, and here are our updated D.C. area snow accumulation chances for between Tuesday night and Wednesday night:

Chance of 2” or more: 70-80 percent
Chance of 5” or more: 45-60 percent
Chance of 10” or more: 10-30 percent

The low end of the ranges apply to D.C./I-95 and points east, while the higher end of the ranges apply to areas west of D.C. and I-95 (where the accumulation potential is greater due to colder temperatures).

These are slightly skewed more toward higher chances for higher accumulations than our forecast from earlier today. The main change this evening is that one model we look at (the NAM) trended the precipitation bullseye of the storm south of the area (in central Virginia) instead of directly over the D.C. area as seen earlier today. Had that model kept the bullseye over the D.C. area we likely would have increased the odds for higher-end accumulations even more, and still may depending what tomorrow’s data brings.

The sweet spot for snow is looking more and more like west-central Virginia (e.g., Charlottesville, Culpeper) where chances for double-digit snow totals are even higher.

As far as storm timing here in the D.C. area, light precipitation could start in the form of rain or a mix Tuesday evening. Sometime around 3-6 a.m. (timing subject to change) is when we would likely see a transition to mainly snow and significant accumulation to begin (if the storm is strong enough).

We’ll resume full coverage early tomorrow, and may tweet an update or two on late models.

(Dan Stillman and Jason Samenow created this forecast update.)

Keep reading for previous live-blog updates...

Overview: The odds of a high impact storm (we’re calling it Snowquester) have grown today, and we’re closing in on a range where confidence is increasing. As noted earlier, we think there’s a 70-80% chance of at least 1 inch of snow, a 45-60% of at least 5 inches, and lower but plausible odds of at least 10 inches.

While tonight’s model results roll out, we’ll keep you updated on the latest data they show. The entries should not be interpreted as our forecast. Instead, we aim to provide insight into the process of model watching. Once we get past the main early set of model data (around 11:00 p.m.), we’ll provide a short update to our forecast.


21z SREF ensemble mean liquid equivalent precip for 24 hrs ending late Wed night. (NCEP)
8:45 p.m. update: First up, the Short Range Ensemble Forecast (SREF) mean. Basically a hold from prior runs for most of us. Lots of precipitation (~1-1.5” liquid equivalent over the area) on Wednesday. See the current run for a 24-hour period versus the previous. Also important from this set of ensembles, the key mid-level feature is in more or less the same spot as other model guidance today. Temperatures also appear as cold or maybe a touch colder than prior runs.

The North American Model (NAM), already running, is up next...

9:10 p.m. update: The NAM is now out to 30 hours (or near midnight Monday evening). The NAM runs slow, so we’ve got a bit more till the storm impacts the area on the model. What we look for is continuity run to run at this point. For the most part, it seems quite similar to prior runs that ended up quite moisture filled and snowy. (mid-levels: current/previous | surface: current/previous).

Jason made a point in the comment section worth repeating. In many storms, a liquid to snow ratio of about 10:1 is average (10 inches snow, for 1 inch rain). In this case, we expect those numbers to be lower most or all spots due to the “mild” surface temperatures near or above freezing and the late-season nature of the storm.

9:30 p.m. update: By late evening Tuesday, the NAM shows the early stages of precipitation nearing the area. It appears that on this run the low is in about the same position as the prior run of the NAM thus far. We’ll be getting to the meat shortly.

9:38 p.m. update: By the near-dawn hours of Wednesday, the new NAM indicates the coastal low pressure has taken over nicely. If anything is noticeable thus far, there is slightly more precipitation associated with the low than the prior run. At the same time, very heavy snow is falling just to our south. The run may end up a hair south of the last one.


0z NAM simulated radar for Wednesday afternoon shows moderate to heavy snow over the area. (NCEP)
9:48 update: Definitely a noticeable shift south with the heaviest precipitation on this run of the NAM (current / prior). One run does not a trend make though. The NAM is known for “bouncing around” at this range. Still, in many ways this aligns it closer with the mid-day Euro, which was still a good hit for most of the region, but kept the edge close enough to feel it’s not a sealed deal. The big snow totals are likely to be focused S/SW of D.C. on this run of the NAM. Simulated snow maps along shortly...

9:56 p.m. update: Below is the NAM’s simulated snow depth as the storm ends. Talk about a tricky forecast locally if it was right! (note: these maps are really more for fun than serious totals).
Snow depth in the D.C. area and broader region based on computer algorithms from the 0z NAM. See our 10:1 ratio note below. Most of these images are based on that type of ratio. The map is unquestionably overdone on the high end and probably underdone on the low end. (Americanwx.com)

10:15 p.m. update: While we wait for the 0z Global Forecast System (GFS) to come out, starting around 10:30 p.m., want to reiterate that this run (as well as a lot of guidance) favors places like Charlottesville, Culpeper, and other locations throughout central and northwest Virginia. We currently believe central Va. looks like a sweet spot for totals. However, small shifts can mean big changes as seen by the tight precipitation gradient.

One thing we saw with the NAM is that the timing for impact is still expected to be after midnight on Tuesday, with the near-dawn timeframe favored for the beginning of heavier activity. This is about when heavier rates change things over to snow in much of the modeling.

10:25 p.m. update: CWG’s Brian Jackson also shared some other ways to look at possible snowfall from the NAM, which indicate more snow than shown on the map below (DCA/BWI). Additionally, an alternate version of the NAM shows precipitation further north than the main NAM. Worth noting: The NAM’s strength is not generally snowstorms at this range.

10:40 p.m. update: The GFS is out to 30 hours (precip/pressure map). It runs quick, so we’ll know its results soon. So far it looks pretty similar to the last run, but it’s early to project for our region.

10:50 p.m. update: The GFS is out through the storm. It shows a major hit across the area (hrs: 54 | 60 | 66 | 72 | 78). Simulated snow maps behind a pay wall I have access to show much of the area with around 1 foot+ of snow (but a sharp gradient to the east, and the same caveats as before about snow maps). In other words, those types of totals may be too high given that we expect temperatures hovering near or just above freezing, so snow may not stick and pile up as fast.

10:55 p.m. update: A comparison of the past two runs of the GFS total liquid output is shown below. Both runs show significant late-season snowstorms for much of the area.


Related

Major winter storm eyes Washington, D.C. and Mid-Atlantic for midweek: Snowquester

Snowquester: March 6 Washington D.C. winter storm name

By  |  11:00 PM ET, 03/03/2013

 
Read what others are saying
     

    © 2011 The Washington Post Company