The warmer than normal winter is expected to continue into at least the first two weeks of March and will probably last through the month. That doesn’t mean there won’t be quick shots of colder than normal weather but that there is unlikely to be sustained cold. And snowstorms will still probably be hard to come by.

While snow chances will be suppressed due to the lack of persistent cold air, the average snowfall during March is so low (1.3, 1.9 and 2.8 for Reagan National, BWI and Dulles airports respectively) that it only takes one snowstorm to put you over the top. Therefore, I’m much more confident about temperatures averaging above normal through March than the outlook for snow. My call for less than average snowfall is just about as likely to be wrong as right.


On January 30, I posted I thought the pattern would change to a better one for cold and snow after the first week of February for at least two weeks as the models were forecasting ridging in western North America (a positive PNA pattern) and the Arctic Oscillation (AO) to change from positive to negative.

My February outlook for snow was for near but slightly below normal amounts with temperatures averaging near normal. That has not panned out, even though last Sunday’s storm did produce a significant snowstorm as close as Charlottesville. Now the AO has joined it cousin the North Atlantic Oscillation (NAO) and switched back to positive.

In February, even during the period when the AO was negative, the NAO was positive which really made it hard to get and hold cold air across the South. That, and the lack of snow cover, probably killed chances for any sustained cold.

The first (top) shows the differences in the look of the pattern when the NAO is in its positive mode versus negative one. Note that during the positive phase there is usually a strong low near Iceland and there is also often another cyclonic vortex just west of Greenland near Baffin Bay where the top schematic shows cold (the blue area). One or both of these upper level lows has been in place almost the entire winter. The positive mode of the oscillation usually keeps the Southeast and mid-Atlantic states warmer than normal.

By contrast, the negative mode (the bottom panel) generally has a ridge extending northward into Greenland with only a weak low or even a high over Iceland. It also has big dip in the jet stream over the eastern U.S. that delivers cold, more frequent Nor’easters and coastal lows increasing our chances for snow. This year, the NAO has been positive pretty much the entire winter.


The positive mode of the NAO also keeps the jet stream farther to the north putting us on the warm side of most storms. La Nina years also have the same storm track tendency driving low pressure centers toward the Ohio Valley or Great Lakes before passing us to the north.

When both the AO and NAO are positive like they are forecast to be in the coming weeks, that favors warmer than normal temperatures across our region (these warmer than normal temperatures usually stretching westward into the Plains).

Source: Appinsys

Regular readers will probably recall that only one 4” or greater snowstorm at DCA has occurred with a positive AO in a La Nina year as strong as this one since 1950. Remember that La Nina years also usually have most storms tracking to our north. Or, sometimes like what happened Sunday, they track too far south to impact us much because the northern stream overwhelms the southern stream.

In any event, our snow chances historically drop off considerably after the first two week of March even without having to fight against a positive AO and NAO in a La Nina year.

The models also suggest that the ridge over the western U.S. during the positive phase of the Pacific North American pattern will be shifting west. When the ridge moves off the West Coast and a trough develops over the western U.S,. then you have a classic negative or reverse PNA pattern. That usually leads to flow from the southwest into the East and a storm track that ends up to our north giving us warmer than normal temperatures. It also decreases our chances for getting a 4 inch or greater snowfall event.

The two maps below are composite maps of the temperature and precipitation anomalies across the U.S. during a La Nina when the PNA index is negative and the AO is positive during March. Most of the southern two thirds of the country end up being warmer than normal and the Ohio Valley into the Northeast ends up wetter than normal, especially across the Ohio Valley. Based on how this year has played out so far, I suspect that the Northern Plains region will end up warmer than shown on the composite.

March composites of temperature (left) and precipitation (right) departures from normal for March with a La Nina, negative PNA and positive AO

CFS2 forecasts of the March temperature and precipitation departures from normal

The weekly forecasts from the same model did a poor job during in mid-February when indicating enhanced chances of getting snow. So the model certainly can be wrong. However, for our area this winter, the model still was correct in predicting that February would average warmer than normal. This winter the model has done a decent job forecasting the temperature anomalies across our area. The CFS2 is generally predicting a wetter than normal March across the same general area (i.e. the Ohio Valley) as the composites of La Nina years shown above.

The Climate Prediction Center (CPC) March outlook looks very much the same as the CFS2 forecasts. All the above argues that there is no reason to depart from persistence (i.e. the same warmer weather pattern continuing).

Bottom line: March looks to average warmer than normal for our area which favors less snow rather than more snow. Are we guaranteed that we’ll get less snow than normal? Not at all. One fluke storm could put all three airports above normal.