The Washington PostDemocracy Dies in Darkness

It’s increasingly difficult to get snow in D.C.

Despite the astonishing snowstorms of 2009-10, as well as two other winters in recent decades, Washington is not a snow town. It gets snow every year, but winter generally isn’t too kind to resident snow lovers.

Besides those once in a great while blockbuster winters, the story of late is one of underachieving snowfalls. It’s often a struggle for snow to accumulate in the city and the immediate surrounds, as well as those to the south and east.

While it has never been super snowy all the time, plentiful anecdotal evidence says the region is not seeing the same type of winter it did in decades past. And data support this notion, at least for D.C.

Accumulation events over time

Among the more striking trends can be seen in the simple “greater than 1 inch snowfall” category.

Events reaching even this relatively small mark have dwindled from an average over six per winter in the beginning of the historical record to about three today. Now, the number of winters with what once was the norm of six are generally few and far between.

Similar trends are seen in most of the lower to mid-scale rankings of storm snowfall accumulation. Examining the frequency of snow events that produce at least 0.1 inch, which includes a good deal more storms that are mostly rain, we see a decline from around nine per year to six per year in the same period noted above.

Rising cold season temperatures

The biggest problem for a snow lover locally is that D.C.’s average wintertime temperatures often barely support snow. This is something you can appreciate when seeing that snowfall averages quickly drop to near nothing heading towards southeast Virginia and southern Maryland while they increase substantially to the north and west (see below).

And temperatures have been rising over time. In the graph below, and the series to follow, the focus is on 30-year periods overlapping each other – all within the modern snowfall records dating back to the late 1880s. The warming is unmistakable.

In the above chart, the focus is on months that make up meteorological winter (Dec.-Feb.). This is when the lion-share of our snow falls in and around D.C. A trend towards warmer temperatures is seen throughout the record, which includes a change in D.C.’s observing site from 24th and M St. to Reagan National Airport in the 1940s.

Comparing today’s temperatures to those of the very cold period surrounding the turn of the 20th century, the differences are stark. As perhaps the best example, December’s average at that point in time was about equal to January’s now. It may seem like a small difference, but keep in mind that December is historically the warmest month of winter. There can also be a substantial difference in expected snowfall over that small difference in temperature.

In marginal situations – when D.C. is near the rain-snow line, temperature matters. Remember Snowquester? How about December 10 last year? The average Dec.-Feb. temperature average is about 4 degrees higher now than it was at the turn of of the last century. It matters in many storms, even the big ones in prime-time for cold and snow.

Winter liquid precipitation pretty steady

Changes in precipitation over the winter months throughout the last 100-plus years are considerably less noticeable than those in the temperature category.

Given the fairly straight trend-lines above, the main takeaway might be remarkable consistency in winter precipitation over time. (This is not something you see too often, looking at weather data in particular. All months are tightly grouped, with only fairly insignificant difference in the overall winter precipitation total.)

So, it’s not a drought causing the local snowfall decline. The Gulf of Mexico and Atlantic are still helping kick up doozies of storms that dump quite regular precipitation over the area in winter.

Losing snow on the margins

The overall story when it comes to D.C. snow appears to be that many of our events are now producing less snow than they once might have.

This is especially noticeable early and late in the season, as already marginal temperature averages slowly warm to a less hospitable direction for snow. Changes on the edges of the winter season have been especially noticeable over time.

While January and February have only seen minor fluctuations in mean snowfall over the historical period, December and March are down a good deal compared to what we once saw. March, in particular, appears in real trouble as a snow month.

Even further on the margins, April once was reliable in producing at least a little snow. Not these days. November is still hanging on, largely because it has had a history of a random storm that surprises everyone from time to time. In a few more years, November’s snowfall average for the last 30 years will go down towards zero, as the fluke 11.5 inches of Veteran’s Day 1987 drops out of the record.

Looking at a whole season average back in that 1894-1903 time period, D.C. was getting about 22 inches of snow per cold season. After a gradual decline, it bounced back as high as 18.4 inches in the 1954-1983 period. We’re currently closing in on 14 inches ending last winter.  That’s a hefty loss of annual snow in just over 100 years.

Mid-winter still performs, with caveats

While January and February are less impacted when it comes to change in snowfall compared to other months, the months both show great variability between high and low totals, and it seems we’ve been low more often than not.

Consider 20 of the last 30 Januaries had snowfall below the 1981-2010 average of 5.6 inches. 22 of the last 30 Februaries fell below the average of 5.7 inches for that month.

A January with no snow accumulation is unusual, and that idea has not changed much over time. Yet, it’s a month often dominated by the most extreme cold of winter, with snow risks on the edges of cold outbreaks.

The January median is also considerably lower than the monthly average, showing a good portion of our Januaries produce less snow than we might “expect.” The occasional blockbuster snowstorm raises the average, but the median remains consistently at a lower level. This is often a truism across D.C. snow climatology.

In February, the trend is quite obvious when examining the frequency of months that get less than one inch. From 1894-1923 it happened four times. In the most recent 30 years it happened 14 times, and the upward movement has been basically continuous. Removing the extreme outliers of 2003 (28.7″) and 2010 (32.1″), February would currently average about three inches of snow compared to an actual almost twice that.

Variability has always been a part of the D.C. snow game. While smaller events ensure our regular seasonal (over?)anticipation of snow days, it’s the big ones that heavily skew our averages. The news is much less bleak there.

Snow lover’s silver lining

Though it may seem we’re on our way to having a snow climatology akin to the Piedmont of North Carolina, not all hope is lost for city snow lovers.

Related: Does the “less snow, more blizzards” global warming theory hold up in Washington, D.C.?

Whatever we might be losing in the small to mid-sized storms, we don’t appear to be doing similar with historic events.

10 inch or greater storms, as well as categories down to 5 inches or so, are generally rather steady or even slightly up over time. They’re not necessarily coming with the same consistency in smaller winter-to-winter stretches, but they’re still coming, and possibly in larger groupings in big winters.

The last three decades, or since the 1980s, the city has recorded six of its current top 20 snowstorms. That’s as high as any other period in history of snow around the city, and up from decades prior that often featured more in the way of regular smaller events.

And of course, 2009-2010 is the only winter in D.C. with three 10 inch snowstorms. Down but not out. Not yet at least.