Last week we reported on breathless world wide coverage (e.g. in the Washington Post) of a study published in the prestigious journal Nature that claimed, based on a novel statistical methodology, that the oceans were warming faster than previously thought. There have enough instances where epic alarmist findings are later found to be wrong that one naturally expects the same sequel, and in this case the pattern held. British mathematician Nicholas Lewis has found that the authors messed up their trend calculations: the new method actually shows the oceans are warming a bit slower than previously thought. The authors have in fact acknowledged making a mistake and are preparing a correction. Don’t hold your breath for the media to do the same though. Just remember this basic rule of climate change journalism: (a) the science is settled and (b) it’s always worse than the scientists thought.
The study authors, led by a Princeton oceanographer, commendably introduced a new way to measure warming of the oceans by using gas concentrations in the atmosphere, which change as the oceans absorb and release oxygen and CO2 depending on temperatures. Less commendably they didn’t describe their methods very clearly and an alert reader named Nicholas Lewis noticed that their warming trend didn’t match their data. When he confirmed his hunch using their data he sent the lead author a query, but heard nothing back. So he published a blog article and soon he and other sleuths figured out that the authors hadn’t done what they claimed they did with their data. Had they done the trend calculation in the usual way their warming estimate would be on the low end of current measures, not above the high end as they originally claimed.
After a few days of silence the lead author posted a cryptic statement on her website acknowledging that their method “underestimated the uncertainties.” She thanked Lewis for bringing it to their attention and promised a clarification, though so far nothing has appeared.
The worst aspect of this episode is the behaviour of Nature and its network of media megaphones who took a new, untested set of results and blared it worldwide in a blast of alarmism. Had the authors’ results showed relatively slow warming of the oceans, matching the records from other measurement systems, there wouldn’t have been any headlines. The cynical summary of alarmist local journalism is “if it bleeds it leads.” Anyone tempted to think that top research journals and science journalists work to a higher standard need to let go their naivete. Any study that lets them declare “worse than we thought!” will get the worldwide headlines.