Rain developing Wednesday may change to snow in the evening and produce some accumulations but considerable uncertainty still exists on exactly how the storm will play out.
The forecast problem stems from a fairly warm air mass across the region. Temperatures tomorrow are expected to rise into the low 40s across the area by 1 p.m. Meanwhile, precipitation will be arriving between late morning and mid-afternoon with the temperatures falling into the mid-30s by nightfall.
The precipitation should start as rain across the area and then perhaps change to wet snow towards or during rush hour, changing first to our north and west with snow possibly working eastward (slowly) across the region during the evening hours.
The areas most likely to get accumulating snow on roads would be areas to our north and west with elevation. Closer to the city, accumulation is more iffy and would depend on how quickly temperatures fall during the late afternoon and evening, which - in turn -depends on the intensity of the precipitation.
Forecasting accumulations when the temperatures are marginal for snow is always dicey. Because of the uncertainty about the temperatures and precipitation intensity, snowfall amounts - if any - are highly uncertain. There are three possible scenarios:
Scenario 1, the snowier of the three: Rain around midday tomorrow with temperatures starting out in the low 40s but then falling into the mid-30s in far western suburbs by late afternoon and in the city by about 7 p.m. Temperatures get close to, but stay just above freezing throughout the storm.
Moderate to heavy snow leads to accumulations of 2-4 inches in the far western suburbs especially areas with elevation. Roads especially in the higher terrain could become slick. Snowfall intensity is high enough for a sloppy inch or two to accumulate over grassy areas in and around the city. (25 percent likelihood)
Scenario 2 - a compromise: precipitation starts as rain with temperatures running in the 40s but slowly falling during the afternoon and evening . Temperature then would fall into the mid-30s by early evening in the western suburbs, and D.C. and points east by 8 or 9 p.m.
Temperatures would eventualy fall low enough to support some accumulations with moderate snow falling towards the end of the storm. In this scenario, areas north and west of the suburbs could expect 1-3 inches of snow and areas around the city might expect a sloppy inch or less. (35 percent likelihood)
All these scenarios are almost equally likely though we favor scenario 3.
Considerable differences exist between the operation models about how quickly air temperatures will cool once the precipitation starts.
Throughout this event, the NAM model has been the most aggressive model with its precipitation amounts and in cooling the surface temperatures. Those two parameters are closely linked.
In this case, the cooling of the air mass is not due to some cold high pressure center nearby but rather from a combination of tapping slightly cooler air to the north and dynamic cooling from a fairly vigorous upper trough (with a cold pool at high altitudes) crossing the area. The model with a stronger upper trough, heavier precipitation and stronger upward motion is going to be colder than one that has weaker upward motion and less precipitation.
Today’s NAM fits the bill in terms of bringing together the ingredients needed for snow. It is pretty much the basis for scenario 1. The NAM higher horizontal and vertical resolution may be an advantage in discerning how quickly the air mass will cool providing it is not overdoing the precipitation and lifting, but it does have a slight bias for forecasting too much precipitation.
The European model, however, says not so fast to the NAM. It has a much weaker system with less cold air, less precipitation and effectively no snow for much of the area.
The temperature problem
By contrast, according to the GFS, the lowest 5000 feet of the atmosphere are forecast to be above freezing with a surface temperature still hanging at 4.0 degs C (39 F) at 7 p.m. This solution is considerably warmer than the NAM forecast. In such a scenario, rain would still be occurring. Both models seem to be on the two extreme sides of possible spectrum of forecast solutions. Both give D.C. decent precipitation but they differ significantly about temperatures.
The precipitation type uncertainty
Today’s plume diagram for DCA from the SREF model shows the uncertainty about precipitation type (see below). A number of members never change the rain to snow especially the members that predict less than 0.30” of total precipitation. They suggest that the NAM model may be an outlier. However, they also don’t have as much vertical resolution as the NAM which may be a disadvantage when trying to forecast the low level temperatures. The SREFs and the today’s GFS argue that there is lots of uncertainty about the low level temperatures and precipitation type and amounts.
Almost all of our forecast uncertainty about tomorrow’s storm concerns temperatures. I could see any of the tree scenarios panning out and lean towards a GFS-European model compromise.
Beyond Wednesday and Thursday
Another cold shot is expected for the weekend but the chances of a snowstorm have diminished with the operational models no longer depicting a storm.
The models are hinting that next week that a low will lift from the southern Rockies towards the Great Lakes or Ohio Valley Wednesday or Thursday of next week, a track favoring rain rather than snow. Temperatures would rise ahead of the front coupled with the low, and then fall back in its wake.
We’ll take a closer look at the long term once we move beyond Wednesday night’s event.