A few weeks ago, there was a big media frenzy over a new paper by Roy Spencer and Danny Braswell, published in the journal Remote Sensing. The paper claimed to show, via some satellite data, that the climate models used by the IPCC are way off the mark. Big news, right? However, it turned out that Spencer had, for the umpteenth time, botched his statistics. To summarize, Spencer and Braswell 1) compared a 10 year period of data with 100 year periods in the models, instead of breaking the 100 years into 10 year periods, 2) didn’t put error bars on the data or the model output, and 3) didn’t plot some of the models that did a better job at reproducing the data. If they had done all this the right way, they would have seen that the models do a decent job, although some are better than others.
Another criticism of Spencer and Braswell’s paper was their choice of venue, the open-access journal, Remote Sensing. Now, some of these for-profit, open-access journals are a bit shady, and routinely publish things that should have been rejected with prejudice, but it’s usually hard to tell until they have been around a few years. The bigger worry was that Remote Sensing hasn’t published a lot of climate science in the past. In such a case, the editors handling the manuscript probably aren’t climate specialists, and may not know who the best reviewers would be. Often, authors are allowed to suggest some good reviewers, and if the editors don’t know who would be better, or that the suggested reviewers are all buddies of the authors, they might just ask the suggested reviewers.
It turns out that this was very likely the case here.
The editor of Remote Sensing, Wolfgang Wagner, has now published an editorial in which he 1) describes how the peer review process failed in this case, and 2) announces his resignation in an attempt to save the reputation of the journal. Here’s the money quote.
In hindsight, it is possible to see why the review process of the paper by Spencer and Braswell did not fulfill its aim. The managing editor of Remote Sensing selected three senior scientists from renowned US universities, each of them having an impressive publication record. Their reviews had an apparently good technical standard and suggested one “major revision”, one “minor revision” and one “accept as is”. The authors revised their paper according to the comments made by the reviewers and, consequently, the editorial board member who handled this paper accepted the paper (and could in fact not have done otherwise). Therefore, from a purely formal point of view, there were no errors with the review process. But, as the case presents itself now, the editorial team unintentionally selected three reviewers who probably share some climate sceptic notions of the authors. This selection by itself does not mean that the review process for this paper was wrong. In science, diversity and controversy are essential to progress and therefore it is important that different opinions are heard and openly discussed. Therefore editors should take special care that minority views are not suppressed, meaning that it certainly would not be correct to reject all controversial papers already during the review process. If a paper presents interesting scientific arguments, even if controversial, it should be published and responded to in the open literature. This was my initial response after having become aware of this particular case. So why, after a more careful study of the pro and contra arguments, have I changed my initial view? The problem is that comparable studies published by other authors have already been refuted in open discussions and to some extend also in the literature (cf. ), a fact which was ignored by Spencer and Braswell in their paper and, unfortunately, not picked up by the reviewers. In other words, the problem I see with the paper by Spencer and Braswell is not that it declared a minority view (which was later unfortunately much exaggerated by the public media) but that it essentially ignored the scientific arguments of its opponents. This latter point was missed in the review process, explaining why I perceive this paper to be fundamentally flawed and therefore wrongly accepted by the journal. This regrettably brought me to the decision to resign as Editor-in-Chief―to make clear that the journal Remote Sensing takes the review process very seriously.
UPDATE 1: Read more about this from Joe Romm (and others linked from his blog) and Leo Hickman at The Guardian. The BBC has now picked this up, and makes the same kind of points I did about how things work when you submit your work to an off-topic journal. Media Matters has now picked it up. More comments on Ars Technica. Peter Gleick has a blog post up about this at the Forbes magazine site. You may remember that the original media frenzy about S&B’s paper was started with a Forbes blog written by James Taylor, from the Heartland Institute. Interestingly, the Forbes homepage has bumped Peter’s blog off in favor of less viewed entries already. Could they be sensitive to the fact that they are guilty of letting some non-scientist shill for a right-wing “think-tank” interpret climate science for their readers? John Nielsen-Gammon (Texas State Climatologist) has similar insights to mine about the likely course of the review process in this case.
UPDATE 2: Andrew Dessler has just published a paper that exposes certain aspects of Spencer and Braswell’s paper that, well, ticked me off, for one thing. It’s a bombshell.