At a recent dinner seminar that I attended, upon hearing that I was a Capital Weather Gang member, an otherwise respectable appearing gentleman asked me how global warming could be occurring in light of the "record-breaking cold winter." I must admit that I was a bit mystified at first, since that claim was not at all consistent with any data I had seen.
I soon discovered, however, that the reports of snowfall records which were being broken in certain northern areas from Wisconsin across southern Canada and into northern New England were being spun into another of the seemingly endless series of politically motivated "proofs" that there is no such thing as global warming. Even the NYTi's normally circumspect science writer Andrew Revkin was drawn in to the hype with an article titled, Skeptics on Human Climate Impact Seize on Cold Spell.
It turns out the cooling claims are wrong on multiple levels.
First of all, as pointed out recently by our own Andrew Freedman and by Dr. Jeff Masters in his blog at Weather Underground ("If global warming is occurring, why was the winter of 2007-2008 so cold and snowy?"), on average, it just wasn't that cold. Any winter which is above average by definition is not cold, and certainly not record-breaking, even if it was cooler than the last seven winters. In fact, globally it was the 16th warmest December-February out of 128 years, putting it in the top 13%, according to NOAA's National Climatic Data Center.
The temperature record in itself should be the end of the story, but "What about all that snow?", you may be asking. Snowfall records crashing from Wisconsin to Maine! Git out the mukluks and the woolly mammoth traps, maw, it's the new Ice Age! It's been observed in other contexts that the view from Inside the Beltway frequently diverges from what is usually perceived as reality in the rest of the world, and a similar effect may be at work in this case. A paltry 4.9" for the season with March rapidly winding down puts this winter into a tie for 12th least snowy since 1899. From that perspective of a snow-starved DC region, it may seem quite plausible that more snow equates with colder temperatures. The chart shows Washington seasonal snowfall totals plotted against average temperatures from December through March. As might be expected, the trend line slopes upward toward higher snowfall with colder temperatures. However, there's a lot of scatter to the data. Statistically speaking, the so-called "R-squared", a measure of how well the line fits the points, indicates that only 26% of the variability of the seasonal snowfall is accounted for by the temperature alone.
The average Washington winter temperature is well above freezing at about 37° (not even including March), so that means much colder than average winters will tend to have more than the usual amount of snow. What about areas that are cold enough that most of their winter precipitation is usually snow to begin with? Madison, Wisconsin is one of those places which set a snowfall record this winter. Their long-term average temperatures are: December 23°, January 17.3°, February 22.6°. Of the three winter months this season, only February was significantly below average at -5.8°. December was only -1.8° and January was very slightly above (+0.1°). For the season as a whole, the average was below the historical level, but not record-breaking.
The real story at Madison was the precipitation: There was more snow because there was more moisture, not because temperatures were exceptionally cold. December's 33.5" of snow was generated from a precipitation total nearly double the long-term average, January's precipitation was also nearly double the average, and February's was close to triple. The chart from NOAA shows much more moisture than usual in the lowest 3 miles of the atmosphere on average for most of the winter between the Ohio River and the Great Lakes. Although Wisconsin was only marginally colder than average overall, the southern portion of the state was in the "Record Wettest" category. For more details on "Why is it snowing so much in Southern Wisconsin this winter?", see an analysis by the NWS Science and Operations Officer in Milwaukee.
Moving eastward, Caribou, Maine set a snowfall record with a total of over 15 feet. What were the temperature differences from average? December was -2.9°, January +3.7°, and February 0.0, so the three main winter months averaged slightly above the long-term mean. Here again, the story was moisture, not temperature, with a few tenths of an inch less than four feet of snow piling up in February on more than double the normal amount of precipitation.
Finally, to the north, the record-breaking snow in Montreal was enough to provoke an eruption of snow rage. The temperature record, however, tells a very different story. Montreal, along with most of southeastern Canada in the heavy-snow zone, averaged above normal (upper 1/3) in the December-February period, according to Environment Canada. Looking at the longer term history for the middle winter month of January, the chart shows that the temperature-snowfall distribution is all over the map. The trend line is almost completely flat, and, to two decimal places, 0% of the variability of the snowfall is explained by the temperature. In fact, the average snowfall for the ten coldest Januaries is 7% lower than the average for the ten warmest Januaries.
The bottom line is that record snow doesn't necessarily imply record cold or even below average cold. The title of the seminar I was attending when I was asked about this "record-breaking cold"? "Negotiating with Bullies and Nut Cases."