Arctic warming cold winter hypothesis loses one year

And it also loses a study, but then it gains two…

We’ll just admit a small prejudice. Are gut feelings allowed in science?

When it is about the climatological theory the hypothesis doesn’t feel strong. When it comes down to empirical evidence it is simply about counting recent winters.

The first of these two new studies is the fourth to suggest a link between Arctic warming and associated sea ice decline – and increased chance of American and Eurasian cold spells during the northern hemisphere winter, or at least increased snowfall.

The theory and the evidence

This study was conducted by the Georgia Institute of Technology and the Chinese Academy of Sciences and was published in PNAS two days ago. Just two days prior it was already backed up by an independent publication in Geophysical Research Letters, which concludes the northern hemisphere warming has been asymmetric over the last decade, with spring, summer and autumn warming more than winter. Parts of eastern North America [that’s where your snowstorms end up when you combine a negative Arctic Oscillation with La Niña winds along the Rockies from the northwest] and northern Eurasia, that study says, have actually experienced cooler winters.

To sum the PNAS study up: that one thinks especially at the onset of the winter season snowfall is likely to increase as a consequence of Arctic warming, as more open water in the Arctic Ocean means more evaporation (especially when the cold sets in over open water – that creates an unstable atmosphere with convection) and therefore more precipitation, usually in the form of snow. That does makes sense – more snowfall over the Arctic Ocean, or over the tundra. But rather than snow depth to actually increase snow cover, the winds will have to carry that extra snow down south…

To suggest a route the study also names an atmospheric blocking pattern that would resemble a negative Arctic Oscillation, but would in fact be something else. We promise to look into it later on. But first some of our concerns.

As a regular you may have read our thoughts on Arctic warming versus cold winters at inhabited latitudes. We’re not convinced.

It hurts your ‘probability sense’ to have to work with a positive that requires a double negative – it simply has a smaller chance than a given positive number, like normal winter cooling as the sun starts to decline. Especially the route where a positive AO is required to generate a negative AO, simply feels a bit too fabricated.

But perhaps that was the big climate lesson two days ago: don’t be too quick to work with statistical assumptions if out there you have empirical data.

But what’s the theory without the evidence

However if we focus on the data we can simply conclude the boreal winter of 2011-2012 did not support the row of evidence of previous winters. After again record sea ice loss in the Arctic there wasn’t a single December snow flake in Europe – and (as we had also predicted) there were also fewer snowstorm outbreaks to the US. The Februari 2012 cold spike in Europe was indeed caused by a negative Arctic Oscillation – but that happened during a positive North Atlantic Oscillation – and shows no connection to the published hypothesis anyway (too late in season, no Atlantic blockades).

And then there was this third new study, published yesterday, also in Geophysical Research Letters and conducted by the Royal Netherlands Meteorological Institute KNMI. Their scientists have especially looked into the development of West-European cold spells.

The other theory and the other evidence

If you stick to a statistical definition of cold spell (below the 10% quantile of the winter temperature distribution) these will simply become a lot milder they say. Their model study suggests in the 21^st century future climate these cold spells will be about 5 degrees Celsius warmer.

“An important contributor to the projected future decrease of temperature variance is shown to be the reduction of the mean zonal temperature gradient, the contrast between land and sea.”

And when you leave it to the KNMI it is not just theory and model calculations. Their empirical evidence tells the same story. It is not just that winters are getting warmer; the winter cold is also declining. And for those who actually enjoy cold spells – for instance for iceskating – that is a substantially more important definition of winter climate change.