Accurate D.C. snowfall totals? Minnesota invention went missing

It turns out measuring precisely how much snow fell during Snowzilla is not as easy as it looks.

One of the more interesting subtexts in last weekend's epic Snowzilla in Washington, D.C. involves questions about the accuracy of officially reported snowfall totals at Reagan National Airport.

It turns out the standard National Weather Service method for measuring snowfall was not followed at Reagan National. Weather observers there literally lost one of their measuring tools, a snow board, in the deep snow.

Angela Fitz at The Washington Post's Capital Weather Gang elaborates.

It’s not that 17.8 inches of snow wasn’t enough.

But the number that will go down in the history books as Washington’s official total — recorded at Reagan National Airport — is downright paltry compared with some other spots in the region, raising the question: Why the disparity?

The reason, it turns out, may be partly due to the improvised technique used by a small team of weather observers at the airport who lost their snow-measuring device to the elements midway through the blizzard. It was buried by the very snow it was supposed to measure.

The mix-up, some say, may have kept the blizzard of 2016 from breaking into the region’s top three snowstorms on record, based on accumulations.

It has also brought into focus the mechanics of how snow depth is counted and has prompted the National Weather Service to announce it will look into the procedures used at the airport.

Since the snow board was lost, observers at Regan National were not able to clear the snow every six hours and add the totals per standard NWS procedure.

The resulting readings likely under-counted totals by at least 2 inches. Again, Capital Weather Gang:

The timeline of snowfall on Saturday suggests something was askew at National.

At 8 p.m. Saturday, 17.8 inches of snow had been recorded there. Although snow continued to fall until midnight, that was the final measurement that the airport submitted to the National Weather Service.

The measurement reflects a 0.3-inch increase in the three hours between 5 and 8 p.m., during which time light to moderate snowfall was being reported at the airport. It also reflects no additional snow accumulation in the four hours after 8 p.m. Significantly higher preliminary snowfall totals were reported at the region’s two other airports. Dulles International recorded 29.3 inches, while Thurgood Marshall-BWI Airport recorded 29.2 inches. Teams at both of those airports used snow boards.

But without the use of a snow board, the measurement is questionable. Jim Lee, the meteorologist in charge at the National Weather Service in Sterling, Va., said he agrees that if the snow hasn’t been measured by the guidelines, the totals are “perishable.”

Leffler, also the author of the Weather Service snow measurement guidelines, said that depending on the density of the snow, the loss of the snow board may have kept the total 10 or even 20 percent too low. That adjustment would push National’s total past 20 inches.

National's total: several inches too low?

To give an idea of how compaction may have affected snowfall totals at Reagan National during Snowzilla, take a look at the image below.

It's a model output of forecast snowfall totals, with compaction figured in. I published this in Updraft several days before the storm hit last week. Notice how measuring snowfall after model estimated compaction would have likely reduced a 20 inch snowfall totals by several inches.

Snow boards: A Minnesota invention

It turns out the basic National Weather Service snow board technique for measuring snowfall was developed and refined right here in a Minnesota backyard, and adopted by NWS nationally around 2002.

The idea is a simple one: The weight of additional falling snow causes the snow already on the ground to compact as it falls. The resulting reading ends up lower than what actually fell.

The use of a snow board, and clearing the snow away every six hours and adding the measured totals gives a more accurate measurement of how much accumulation actually occurred.

I spoke this week with Pete Boulay from the Minnesota State Climatology Office about the development of the snow board.

Pete is largely credited with the advent of using snowboards nationally by NWS. His tests here in Minnesota during the winter or 2001-2002 were some of the basis for adopting his techniques for snowfall measurement by NWS nationwide.

I asked Pete about the development of what some call, the "Boulay Board." Here are some excepts form our email exchange about measuring snowfall at DCA during Snowzilla.

I see what happened now. They lost the board! That's why I deploy blue flags at my snow board sites (I have two sites, one at home and one here at work.) The board can be easy to lose if it is not marked since it is a white board of expanded PVC. As far as I know the National Weather Service has adopted the boards that we tested here across the county. I'm not sure if that is the same board that was used at DC or not since the article mentions a wooden board painted white. I agree with the article that NWS standard measurements at the large airports use 6 hour observations for snowfall.

Measuring large snow events are very tricky, due to the wind and blowing and drifting involved. Also, airports are typically wide open areas where blowing and drifting is more severe. In the end an observer does the best they can. I often people to get a board and a ruler and give snow measurement a try to get an idea how it is done.

The National Weather Service (I can't recall if it was Duluth or the Twin Cities National Weather that began to use them) I wound up the getting a call from the engineering department of the National Weather Service in Washington DC and they really liked the sound of them. They tested the boards in 2002 and wound up deploying them.

In 2002, the National Weather Service tested the snowboard and ran a test in a cold chamber with a sample and there were no negative results after subjecting the board to -50 F.  They wound up deploying 6,000 boards.

The real credit for the expanded PVC board idea goes to the MNDNR sign shop. We were looking for boards to use for the Snow Rules! north shore snow study and it was the sign shop that suggested using that material and it worked great!  (we also had heavy duty metal rulers made in tenths of an inch in increments and they last a lifetime!)

By the way, the first snow board I tested is still in use 15 years later.

It will be interesting to see if the NWS adjusts snowfall totals higher at Reagan National based on the likely compaction of snow measured without the missing snow board.