Remember in late January an article in Nature was published concluding Antarctic warming over the past 50 years was more extensive than previously thought?

A chap called Ryan O. has got to the bottom of the numbers. The conclusion is that the statistical analysis is flawed, and the results do not stand up. Steve Mcintyre, on Climate Audit has published blogs here and here explaining and enlarging on aspects of the findings.

I wonder if Nature will publish these findings? I can say with near certainty that the press will not give the same prominence to this as the original findings. There will be no announcement on the BBC, nor will msnbc be announcing that “Flawed research undermines climate change consensus”. It is not just a media bias in the Global Warming Climate Change debate, but simply that counter-news is rarely reported. So every minor bit of medical research is shown as news, but rarely do we get updates that the results have been overturned, despite it happening 80% of the time. Yet in scientific areas that rely on statistical analysis this happens all the time. In studying economics I found in many areas, such as the Theory of Demand, The Phillips Curve and on the Monetarist / Keynesian debates on the 1960s to 1980s, papers were constantly being overturned.

For those who believe that predictive ability is the sign of good science James M. Taylor, senior fellow for environment policy at The Heartland Institute, should be commended. Quoted in OPedNews on January 24^th 2009, three days after the article was published.

“I would be quite wary of assigning much value to this article. Raw temperature data and a number of studies over many years have determined that Antarctica is cooling. Now we have a single article, reliant on subjective data interpretation from well-known global warming alarmists, saying the opposite.

“For a long time now, Antarctic cooling has been a stone in the shoe of global warming alarmists. Now, conveniently, those who regularly blog on an alarmist Web site claim they have ‘statistically smoothed’ the data to show Antarctica is warming, even though surface temperature stations show a significant, long-term cooling trend.

“The article appears to argue that due to incredibly bad luck, many temperature stations scattered throughout the continent are located in random, isolated pockets of cooling that defy the overall warming trend. The odds of this being the case are quite remote, and the theory is notably short on reliable evidence. Adding to the dubious nature of the study’s conclusion is the authors’ self-interest in silencing an embarrassing mountain of raw temperature data that contradict the authors’ global warming theory.

Taylor points to the results contradicting established data. When that happens the review process should ask searching questions. Something seems to have gone amiss with the review process at Nature, despite them having taken 11 months to review the paper.