I spoke Thursday with Steve Zubrick, the science and operations officer at the NWS forecast office in Sterling, and we talked again about Baltimore's curious history of snowfalls on Dec. 5. It seems he, too, was intrigued by the phenomenon. And he was not alone.
As faithful WeatherBlog readers will know, Baltimore has recorded at least a trace of snow on six of the last eight Dec. 5ths. When I asked Steve about the phenomenon in an email a while back, he expressed doubt that a real statistical analysis of the phenomenon would find anything more than coincidence at work.
But he queried the weather service computers and found that snowstorms of an inch or more have occurred on Dec. 5 more than any other date in December - nine of them since daily snowfall records began for the city in 1892. There is a similar spike for snow on Dec. 5 at Washington's Reagan National Airport, although it's not the snowiest December date there.
That stirred some speculation at the Sterling office. Forecaster Jared Klein noted that early December is about when the first arctic cold outbreaks begin the reach Maryland, and wondered whether the Dec. 5 snow spike is "more a factor of when the first cold outbreak makes it possible to snow."
At a Winter Weather Workshop for media types Thursday in Sterling, Zubrick revealed that he had been intrigued enough to run the question by some heavyweights in the meteorological statistics community.
The first was Harry R. Glahn, director of the NWS's Meteorological Development Lab. Glahn took the 117 years of data and ran it through a series of statistical tests. Finally focusing on the December numbers for Baltimore, he said the math "leads me to conclude that, looking at the data and finding one spike in December of [greater than or equal to] 9 days is not unusual."
He called it a "singularity," and added that, "while at first blush there may be something unusual about December 5, it is nebulous. I might say, it is unusual not to find something unusual in a set of data." But, he added, "Much more work would have to be carefully done to confirm a Dec. 5 abnormality. It might make a good MS thesis topic for some student, or even a Ph.D."
Glahn forwarded the question to Ian Jolliffe, a professor emeritus at Aberdeen University in Scotland, whom Zubrick described as a "world reknowned statistician and atmospheric scientist." Jolliffe agreed, saying there is "clearly no evidence of anything but a chance occurrence."
"I'm pretty convinced by this," Zubrick said. Me, too.
Still, Glahn cautioned against dismissing such oddities too quickly.
"In your original email," Glahn told Zubrick, "you said it is absurd to think there may be something happening that could cause such an 'abnormality.' I wouldn't be too sure about that. There are extra-terrestrial events that could conceivably cause something like this; meteor showers come to mind. But most such things wouldn't be tied to specific earth calendar days," he said.
He reminded us all not to be too quick to dismiss apparent oddities: "Sometimes, someone noticing something unusual leads to new understandings."
(PHOTOS: Top, Laurel Park, Dec. 5, 2009, by Jerry Dzierwinski, Maryland Jockey Club. Bottom: Sun Photo, Baltimore, Dec, 5, 2002, by Algerina Perna)