Despite the hot air, the Antarctic is not warming up

Junk science, the brain candy of the new millenium...

So it predictably made headlines across the world last week when a new study, from a team led by Professor Eric Steig, claimed to prove that the Antarctic has been heating up after all. As on similar occasions in the past, all the usual supporters of the cause were called in to whoop up its historic importance. The paper was published in Nature and heavily promoted by the BBC. This, crowed journalists such as Newsweek's Sharon Begley, would really be one in the eye for the "deniers" and "contrarians".

But then a good many experts began to examine just what new evidence had been used to justify this dramatic finding. It turned out that it was produced by a computer model based on combining the satellite evidence since 1979 with temperature readings from surface weather stations.

The problem with Antarctica, though, is that has so few weather stations. So what the computer had been programmed to do, by a formula not yet revealed, was to estimate the data those missing weather stations would have come up with if they had existed. In other words, while confirming that the satellite data have indeed shown the Antarctic as cooling since 1979, the study relied ultimately on pure guesswork, to show that in the past 50 years the continent has warmed – by just one degree Fahrenheit.

One of the first to express astonishment was Dr Kenneth Trenberth, a senior scientist with the UN's Intergovernmental Panel on Climate Change (IPCC) and a convinced believer in global warming, who wryly observed "it is hard to make data where none exists". A disbelieving Ross Hayes, an atmospheric scientist who has often visited the Antarctic for Nasa, sent Professor Steig a caustic email ending: "with statistics you can make numbers go to any conclusion you want. It saddens me to see members of the scientific community do this for media coverage."