Glut of hot years a coincidence? Fat chance

by Catherine Brahic, NewScientist, December 17, 2008

Thirteen of the hottest years since records of global temperatures began in 1880 have clustered in the last 17 years. It is tempting – and it sure makes good headlines – to blame it on climate change. But does science support such a claim?

According to new statistical research, it does. The recent glut of unusually hot years is incredibly unlikely to happen in a stable climate.

Eduardo Zorita of Germany's Institute for Coastal Research and colleagues calculated the probability of this happening in a range of scenarios.

A key consideration is that the weather one year is not independent of the weather the year before. If it were, the odds of having any given temperature would be the same each year, and the likelihood of getting a such a 17-year cluster would be tiny – on the order of 1 in 10 trillion.

Natural memory

"An anomalous warm year tends to be followed by a warm year," says Zorita, because of the way oceans store heat and release it slowly. "A devil's advocate could argue that the clustering of warmest years at the end of the record could be simply due to chance, since the climate system has a natural memory."

However, even when Zorita included this natural feedback in his model, but excluded global warming, the odds of observing the cluster of record-breaking years was still about 1 in 10,000.

"We cannot ascribe the anomaly to any particular physical factor, like anthropogenic greenhouse gases," says Zorita. "But our conclusions are consistent with those of the fourth IPCC report," which states there is a very high probability that human emissions are causing global warming.

Journal reference (not yet accessible): Geophysical Research Letters (DOI: 10.1029/2008GL036228, in press).

Link to article: http://www.newscientist.com/article/dn16292-glut-of-hot-years-a-coincidence-fat-chance.html