Snow cover extent on January 1, 2013. (NOAA from AccuWeather)

Too much is being made of recent snow cover records established in the U.S. and northern hemisphere.

On Wednesday, AccuWeather headlined an article: “January 1 Snow Coverage Sets New Record for US”. This record is meaningless as the dataset is only 10 years old.

Today, the blog WattsUpWithThat (WUWT) draws attention to a new record established in December for northern hemisphere snow cover to rebut a claim snowfall will decline in Britain from global warming. But a single data point (averaged over a hemisphere) says little to nothing about the future course of snowfall in a specific area.

Let’s take a deeper look at both of these records and what they mean or don’t mean.

January 1 U.S. snowfall record

Yes, on January 1, 67 percent of Lower 48 had at least a little snow on the ground, more than any other January 1 on record. But said record is only 10 years long, dating back to 2004. It’s impossible to say anything about the significance of a record consisting of only 10 data points.

The most we can say is that recent weather over the U.S. has been sufficiently cold for snow to accumulate over a large part of the country. Much more so than, say, last year when just 20 percent of the U.S. had snow on the same date - a so-called “record low”.

December northern hemisphere snowfall record

Data from the Rutgers University Global Snow Lab show that, averaged over December, snow covered the most area on record in 2012, dating back to 1966.

Northern hemisphere snow cover difference from average December 1966-2012. December 2012 had the most snow cover on record. (Rutgers University Global Snow Lab)

This record is certainly more meaningful than the U.S. January 1 record. The time series is much longer, applies to a larger area and covers an entire month (rather than a single day).

Yet WUWT inappropriately uses the record to poke fun at a prediction by Dr. David Viner of the University of East Anglia’s Climate Research Unit that on Britain: within a few years winter snowfall will become “a very rare and exciting event”.

I don’t disagree with WUWT that Viner’s statement was over the top. While Britain may well experience a decline in snow frequency from global warming, the decline is likely to manifest itself over decades rather than “a few years”.

But WUWT shouldn’t (mis)use northern hemisphere snow cover data from a given year in a given month to make this argument. At any point, just because there is snow covering record-breaking parts of the northern hemisphere doesn’t necessarily mean there is snow in Britain. Take for example, this record-setting December 2012 - a WUWT reader in the UK comments: we haven’t got much snow (other than in the far north of Scotland) just a load of floods and the second wettest UK since records began in the 1950′s.

It turns out, in places where temperatures are cold enough for it to snow (due to increases in evaporation and precipitation), global warming may actually increase winter snow cover.

(WUWT indirectly makes this point in its explanation for the December record assuming you link the record low Arctic sea ice with warm temperatures: Increased evaporation combined with more heat loss in the Arctic due to a record low amount of Arctic sea ice is the likely cause.)

But in milder climates like Britain, continued warming should increasingly favor rain compared to snow.

Thus, neither this latest December snow cover record nor the prediction Britain may experience less snow over time is inconsistent with global warming ideas.


Does snow cover extent debunk global warming?