Thursday, October 20, 2011

BEST, Muller, and the Issue of Peer Review

My previous post notwithstanding, there is some reason to criticize BEST for announcing their results before peer review and publication. You wonder what their hurry is (well, you don't have to wonder too much), because they did this before when Richard Muller testified before Congress based on only 2% of their data. Muller does come off as a bit of a loose cannon trying to get attention stir things up.

And he's been not just wrong, but unfair. He clearly misrepresented the phrase "Mike's trick to hide the decline," right here in his by-now-famous lecture, where he says (of pre-1961 proxy temperatures measured by northern high-latitude tree rings)
"Is this unreliable? No. How do we know? Well, we don't know."
That's simply wrong, and extremely unfair--paleoclimatologists have done a lot of work establishing the correlation between tree ring properties and 1880-1960 temperatures--see D'Arrigo et al 2008 for a great review, and Skeptical Science for a very good summary. Muller implies that that this was just made up, which perpetuated the attacks on the hockey stick, especially since, just two minutes earlier in the same lecture, he held himself up as an expert with "prestige." No study that came out today is going to undo that blight, which I think is why Gavin Schmidt wrote today
"The only thing I would take issue with is the description of Muller as 'brave'. It isn't brave to slander people in many talks (including Jim) without having done the work required to justify such a claim. I do give him some credit with following the data as they emerged from their analysis (mostly from Rohde), but it would have been foolish not to. But this notion of a 'prodigal scientist' being braver than the people who worked out the answer before is not tenable."
Anyway, I really meant to write about peer review, but got sidetracked. It's the nature of our times, propelled by (of course) the Internet/Web, that results are coming out earlier and earlier in the scientific process. This happened most recently with the OPERA superluminal neutrinos, where the collaboration can't even agree if they should submit a paper, but it's been happening for years with the physics arXiv of preprints. Physicists certainly key off what they read there and hear in seminars, and increasingly PIOs send out releases before a paper actually appears -- and, of course, bloggers have ruined everything.

I suspect the community review model is going to replace traditional peer-review for a field like climate science that is so contentious and important. In one sense the world needs to know ASAP what projects like BEST conclude -- there really isn't much time left to argue about things (if any). I'm not sure that was their motivation, though. On the other hand, the issue is clearly being diluted by every person who can type with questions about the science, so they can't all be allowed into a community review. Whatever is left in the world that the Internet hasn't changed already, it will.

And let's not fool ourselves: if and when the BEST papers do appear in a peer-reviewed journal, the Anthony Watts of the world aren't going to then accept them. Instead they'll make the by-now usual claims that the editors are biased, or in favor of world domination, and it's all a big conspiracy and the publication of the BEST papers just proves it all the more, and BEST did it to ensure future grant money and whatever. They simply aren't amenable to evidence, period. Read the comments on today's post by Watts -- they're pathetic. Hilarious, but deeply pathetic.

Red Sox fans are never going to respect the Yankees no matter how many World Series the Yankees win. And that's what the attitude to climate change is these days, for most of the public. Listen to an AM sports talk show, and then read the comments on a skeptic blog. There is very little difference.

No comments: