The American humorist, Kin Hubbard, once said, “Don’t knock the weather. If it didn’t change once in a while, nine out of ten people couldn’t start a conversation.” In this federal city, with its vaunted halls of Congress, the CIA, the FBI, the Federal Reserve, and many other secretive or semi-secretive agencies, this saying may have more meaning than anywhere else—especially at the water cooler and in car pools.
During the cooler months, Washington’s weather does change more than “once in a while.” Day-to-day temperature fluctuations often exceed 30 degrees or more. And during the warmer months, although large-scale changes are unlikely, sudden - sometimes unpredictable - changes can occur, completely transforming the character of a summer day, if only for a few hours.
Some even say “if you don’t like the weather now, just wait a few minutes,” a phrase seemingly embedded in the genes of people east of the Rockies,. But it’s that very changeability that many believe makes the mid-Atlantic climate the invigorating one that it is.
Whether it’s extreme heat, extreme cold, blizzards, hurricanes, severe thunderstorms, tornadoes, droughts, floods, or just plain high winds, Washington experiences an incredible range of weather conditions. But is D.C. weather really more difficult to forecast than elsewhere or is that just an urban myth, perpetuated by mean-spirited cynics?
Today’s post will focus on some specific forecast challenges in D.C. Tomorrow, we’ll compare D.C. to other locations across the nation. Keep reading to learn about some specific forecasting challenges the region contends with...
Local meteorologists: what’s their take?
Topper Shutt (WUSA Channel 9), for one, believes there’s at least a grain of truth in the idea that because the District lies between the mountains and the bay (and the ocean), forecasting is sometimes a challenge. Maybe not as challenging as say, Boston, but still a challenge, he believes.
Shutt frequently mentions the local “microclimates,” such as the “heat island” around Reagan National Airport (DCA), the cooler summer daytime temperatures near the bay, etc. He is careful, however, not to regularly blame forecasting dilemmas on our geographic position, although he does believe that, absent the mountains and the ocean, Midwest forecasters aren’t quite as challenged as they are here.
Chris Strong, of the National Weather Service (NWS) Baltimore/Washington Forecast Office in Sterling, VA, reiterates the forecasting difficulties sometimes encountered with the various D.C. microclimates. But from his perspective, with the exception of sudden pop-up storms, summer weather is generally the most stable and predictable of the entire year. It therefore presents the lowest likelihood of a “busted” forecast.
When asked what aspects of D.C.’s weather are easiest to forecast, Bob Ryan (WJLA, Channel 7) says that humidity changes and prolonged, large-impact weather events are generally the easiest. The smaller-scale, usually warm-season events are the hardest, he believes, due to the difficulty in pinpointing exact storm locations with significant warning time.
Nevertheless, Ryan recognizes and has reason to know about the occasional forecasting dilemmas associated with certain large-scale weather events. Originally from the Boston area, he says he well remembers the winter storms that dumped heavy snow on the north side of town - but heavy rain on the south side, an experience he and other forecasters are thoroughly familiar with here in the D.C. area.
Washington frequently finds itself on the western edge of storms riding up the East Coast, on the rain-snow (and sometimes ice) dividing line. This can result in remarkable weather variations over the geographic range of the metro region, creating all kinds of forecasting headaches.
December 2010’s Christmas weekend storm that crushed the coast but just clipped Washington, is a case in point. (As was the storm that produced several inches in central Virginia last week, but nothing in the District, to a lesser extent).
December 25-26, 2010: Beach blizzard, beltway bust
The Dec. 26, 2010 storm was not of the snow or rain type. It was more of a snow or no- snow event as it appeared on computer models about a week before Christmas. Soon, the models seemed to converge on a Christmas, or day-after -Christmas, storm. A low pressure system skirting across the South would emerge off the North Carolina coast and ride up the coastline, greatly intensifying in the process.
This is a classic nor’easter setup for the D.C. area and much of the Northeast. However, on Dec. 21, Capital Weather Gang (CWG) wrote “the big question is how much moisture and precipitation will, in fact, reach the area on Christmas Day [or thereafter].”
Based on years of experience with NOAA’s Hydrometeorological Prediction Center (HPC), Wes Junker, CWG’s winter weather expert, has seen this scenario many times. The storm had a very tight “gradient,” he says, meaning that the “heavy snow-no snow” areas were quite close together, relatively speaking.
Until the storm explosively deepened off the Carolina coast, the models continued to waffle back and forth regarding the westward extent of the storm’s Atlantic moisture plume. But early on Christmas Day, there was enough consensus for appreciable snow to reach D.C.’s western suburbs that the NWS issued a “winter storm watch” and later a “winter storm warning” for 5 inches or more of snow.
In reality, almost the entire metro area was left high and dry. The eastern boundary of meaningful snowfall set up on the western shores of the Chesapeake Bay. One to two feet of snow fell from eastern North Carolina to New England, including more than a foot in Ocean City, Md., and almost two feet in Atlantic City and New York City. Even Annapolis got a few inches.
Unfortunately, modern technology, as good as it is, could not simulate the exact location of where the storm’s expansive moisture field would cut off - of where that tight “gradient” would be most apparent -even 12 hours before the storm hit.
But one can also look at it another way: one week before Christmas, computer models accurately predicted that a significant snowstorm would hit major mid-Atlantic metro areas 7-8 days later. In actuality, the storm did form and found its mark at about the predicted time, even if just skirting one particular area (which happened to be D.C.), by less than one and one-half degrees of longitude - or about 70 miles
Please be sure to come back and read the follow-on piece to this tomorrow where I’ll discuss some of the big forecasting challenges in other parts of the country, and how they compare...