After today’s warmth and rain, the pattern is expected to transition to a cooler, drier one.  Temperatures during the week between Christmas and New Year’s should average near  normal. The pattern looks like a dry one, but a few models have hinted at a chance of some light precipitation next Monday or Tuesday as an arctic front crosses the area. We’ll be watching that possibility over the next several days, but right now it does not look like a big deal.

While the next week appears more or less quiet, the first week of January looks less benign. The pattern is one that favors cold and in the past has sometimes delivered temperatures well below normal. The pattern usually yields precipitation a little below normal to near normal as storms either pass to our north or well to our south. However, the pattern still is the most favorable one so far this year for getting an inch of snow (a 30 to 40 percent chance) in D.C.

Why are you expecting the period between Christmas and New Years to be seasonably cool and dry?

Into today, our air had been coming straight from the tropics; that’s why it’s been so mild. But, starting tonight, Canada becomes the source region of our air masses hence the reason temperatures will be closer to normal. Storms passing to our north should allow daytime temperatures to keep from really tanking, rising into the 40s occasionally (maybe even the 50s Sunday).

The dryness is also related to this new flow regime from the northwest. Weather systems will be entering the U.S. from Canada with low pressure systems along the polar jet stream tracking to our north. Our usual source of moisture, the southern jet stream, looks pretty inactive. Any precipitation (Christmas Eve and New Year’s Eve flurries?), therefore, should be light.

The plume diagram for Sterling, Virginia from last night’s ensemble of GFS model simulations (the GEFS) illustrates the dryness of the pattern. (Remember each ensemble run is just a model with its initial conditions tweaked to get an idea of the predictability of the pattern.)

Accumulated precipitation over the next week for multiple simulations of the GFS model. Precipitation is accumulating when lines rise vertically (e.g. today’s rain) and it is dry when the lines are flat (e.g.. tomorrow through Dec. 30). Precipitation type is indicated by color.

The ensembles are in terrific agreement that it will be dry (all the lines are flat and parallel to the x-axis) from Christmas to December 30. Even beyond that dry spell, note how the gray lines at the bottom – which signify precipitation – are mostly only indicating light amounts. The ensembles (the colored lines) are almost equally divided between those giving some precipitation (rising a bit) and those that are bone dry (flat) on December 30 and 31.

What makes you think January 1-7 will be cold?

The 11-15 day average upper level weather pattern (composite mean 500 mb anomaly) forecasts of the European and GFS ensemble systems are in terrific agreement (see below). The jet stream bulges northward where the red areas (ridges, above normal heights) are located on the map, and dips southward where the blue areas are found (troughs, below normal heights). The combination of the red area (above normal heights) over western Canada extending so far north with the blue area (below normal heights) over the eastern U.S. extending so far south usually leads to a southward plunge of arctic air.


The composite patterns above can be compared to 5-day mean patterns from the past. The 10 most similar cases (analogs) were identified. The 5-day mean surface temperature anomalies (differences from normal) can then be ascertained for those ten 5-day periods to get a feel for what the pattern means in terms of our temperatures. The temperatures during those 10 analog periods averaged about 5 degrees below normal across our area (see below). The average maximum for the week is 43 and the minimum sits at 29.


More importantly, a number of the analogs were associated with cold outbreaks that lowered the night time temperatures into the teens for at least one day. My colleague Matt Rogers notes that a couple of the analogs, notably January 2009 and 2003, had the night time temperatures drop into the lower teens and upper single digits. It’s too early to forecast such cold temperatures but is worth noting that in the past such a pattern has, on occasion, led to notable cold outbreaks.

What makes you a little more bullish for snow than you’ve been?

1) Temperatures should average colder than normal
2) The 12 independent analogs identified by the two mean patterns above were associated with four storms that produced an inch or more of snow at Reagan National with one of the storms producing over 7 inches.
3) Our climatology for getting a 1 inch or greater storm at DCA during the week is about 30%; it’s on the rise compared to the week of Christmas, when it’s 20%.
4) The Arctic Oscillation is forecast to flip to negative and the Pacific North American pattern is predicted to be positive (see below). Such a combination is the most favorable one for getting a decent snow in our area

Although the NAO and PNA may be favorable, the pattern across Canada just east of Hudson Bay is still less than ideal for getting a significant snowstorm. Most of our big storms have above normal heights across that area and below normal heights to the south of Nova Scotia. That, and the fact that the pattern still favors lows diving out of Canada towards the Great Lakes always raises the specter of just enough warming prior to the storm to mess things up. Still, colder than normal temperatures with a negative AO and positive PNA should give snow lovers some hope. I expect the models to spit out forecasts of storms during the first week of January but won’t take any of those forecasts seriously until this weekend at the earliest.