The Washington PostDemocracy Dies in Darkness

A white St. Patrick’s Day? Sunday night to Monday snow potential is promising, amounts uncertain

Canadian model shows mixed bag of precipitation moving into region late Sunday. But cold  air from high pressure  to the north will be drawn into the storm, likely changing precipitation to snow Sunday night. (Environment Canada)

Computer models continue to agree that the Washington area will see wintry weather Sunday night into Monday.  Most of the models suggest that the precipitation starts as rain Sunday afternoon and then transitions to sleet and snow Sunday night.  A little bit more snow could fall Monday.   How quickly the rain transitions to snow will play a significant role in determining whether we get any significant accumulations in the city.  This is another complicated and difficult snow forecast.

Key points:

  • Models seem to target 8 p.m. Sunday night to around 4 a.m. Monday morning as the period when the heaviest precipitation will fall;  a second wave of snow may come in Monday during the day
  • Monday morning’s commute could well be messy
  • We think the odds of disruptive snow is more likely in the Sunday night wave because snow has a better chance to accumulate at night this time of year; snow that falls during the possible second round during the day Monday may have trouble accumulating unless it is heavy (as solar energy is strong this time of year)
  • The best chance of significant accumulating snow will be in D.C.’s colder suburbs to north and west – basically west of line from Manassas to Fairfax to Rockville to Columbia
  • Inside the Beltway and to the south and east, accumulation will be heavily reliant on two factors: 1) how quickly temperatures drop below freezing Sunday night and 2) the development of heavy bands.  If the snow is light and/or temperatures are slow to drop, accumulation will be limited and mostly on grassy areas.
  • Total accumulations will probably be elevation dependent; that is, hills and mountainous areas west and northwest of the city above 1,000 feet have the best chance to see in excess of 4 inches of snow

Accumulation prospects:

  • Chance of at least 1 inch: 40 percent south and east of the Beltway, 50 percent inside the Beltway, 60 percent Manassas to Fairfax to Rockville to Columbia, 70 percent Leesburg to Frederick
  • Chance of at least 4 inches: 10 percent south and east of the Beltway, 15-20 percent inside the Beltway, 30 percent Manassas to Fairfax to Rockville to Columbia, 40 percent Leesburg to Frederick
  • What’s the high-end of possible accumulations? 8-12 inches.  What’s the low-end? Bare ground, conversational snow.

Note: We will develop a snowfall map tomorrow, Saturday.

Technical discussion by Wes Junker

Most of the models now have the storm taking place in two separate phases, one with warm advection/overunning associated with the initial surface low Sunday night and a second possible shot with the upper low on Monday.  Unfortunately, the models differ on how they handle these features

Last night’s GFS and European have the temperatures too warm for much accumulation from the initial shot of precipitation Sunday afternoon and evening and then only have light to moderate precipitation as the upper system crosses the area on Monday.   However,  this morning’s GFS looks more favorable for getting some accumulation Sunday night as it crashes temperatures after dark.   However, it now only predicts light precipitation with the upper low Monday .

What happens with the upper low is somewhat of a wild card.  Today’s NAM keeps the precipitation to our south.  Last night’s GFS and Euro gave us a second round of precipitation.   This latter shot might yield  accumulations as temperatures are forecast to be below freezing.  However,   if we remain in the the upper 20s with only light precipitation falling, it probably would be difficult to get accumulations on the roads unless the precipitation were to fall heavily. That’s a possibility, but during March relying on intense precipitation rates with a shearing (weakening) upper trough often fails to deliver.

The 09Z SREF diagram (below) showing the various ensemble member forecasts of the temperatures during the storm illustrate how much uncertainty there is about how quickly the temperatures in the city  will fall below freezing.  Remember that ensemble members are just model forecasts with slightly perturbed initial conditions or models having their physics tweaked a bit.  The purpose is to get a feel for the uncertainty of the forecast.

Note there are several very warm members due to those members really wrapping the initial low up and tracking it to the north putting us in the warm sector.  Last night’s NAM had similar look but was not as warm as those members.  They  appear to be outliers.

The dark line on the graph (below) indicates the mean of all members is still 33 degrees at 2 a.m. Monday morning (6z) but it may be too heavily weighted by the 3 very warm outliers.  Still,  the median of the colder grouping of lines still keeps the temperatures above freezing until 11 p.m. (3z) Sunday night.  That agrees well with the timing of the latest NAM model.  Where the SREF differs from the NAM is most members do not have as pronounced of a warm layer as the NAM and a large number of them deliver additional precipitation with the upper low.  The latter is lacking on the latest NAM.  The bulk of the SREFS are more supportive of the latest GFS than the NAM but they may be a little slower cooling the temperatures (at 2 meters).

The last two NAM runs are very bearish in terms of accumulating snow.  This morning’s NAM model generally has a mixed mess with the initial shot of precipitation Sunday afternoon and night.  The model holds the second round of precipitation associated with the upper low to our south

The NAM run has a elevated layer where temperatures stay above freezing suggesting once the precipitation changes from rain to frozen precipitation, the bulk of it might fall as sleet.    Note even at 2 a.m. Monday morning the NAM still is predicting sleet rather than snow across most of the area (see below). The NAM is probably on the warm side of the envelope of possible solutions but some permutation similar to it is possible.

Today’s GFS has a more southerly track with a less wound up low initially than the NAM and it therefore has a much colder look and snowier look than the NAM.  By 8 p.m. Sunday (0z), it has snow developing across the area and snow is falling across the entire area by 2 a.m. Monday (6z) with temperatures below freezing (see below).  It would suggest 0.50″ or so of precipitation would fall as snow.  The one caveat to mention about the GFS is that it brought in the low level cold air little too quickly into D.C. and points south during the March 3 snowstorm   Still it would argue for some accumulation even in the city.  It only predicts light snow during the day suggesting the bulk of the accumulation would occur Sunday night into early Monday.

A GFS scenario looks more likely than a NAM one but the solution still easily could fall somewhere between the two.  Therefore, trying to predict snowfall accumulations this early is little better than a crap shoot.

Interestingly, the Canadian model has backed off its idea of a heavy mess of precipitation – with substantial snow north and west of the city –  and has moved to lighter but colder solution favoring light snow accumulations across the region.

Just in: the European model cools the atmosphere relatively quickly Sunday night,  changing rain to snow, and suggests several inches of possible accumulation.

So what do we know?   Some winter weather is likely Sunday night possibly into Monday.  Accumulating snow is possible but still is far from certain for locations in and around the city.  The best chances for getting several inches of accumulating snow is north and west of the city.  However, even the latter is not etched in stone.