Friday
Dec312010
by 
Bishop Hill
Bishop Hill Revkin on Steig and O'Donnell
Dec 31, 2010 Climate: Surface Journals Andy Revkin looks at O'Donnell et al's recent improvement on/rebuttal of Steig et al, the 2009 paper that suggested that the whole of the Antarctic was warming. Revkin sees the appearance of the rebuttal as reinforcing his faith in the peer review process, a marked contrast to the views of Richard Smith, which were discussed here a couple of days back.
Reader Comments (17)
Steig "The results appear to confirm all of the key results in our 2009 Nature paper, notably the significant warming in West Antarctica, with especially strong warming — very widespread and including all of West Antarctica and much of East Antarctica — in spring."
I don't remember the headline reading anything like "Spring causes Antarctic warming"
Lewis "we show no statistically significant warming for the continent as a whole over 1957-2006"
How does that appear to confirm Steig, even to Steig, or the team? Seems they're just papering over the cracks with ready sound bites for the faithful.
Funny how SM's comments about an 88 page critical review have to "ring hollow" for Andy's trumpeting of the scientific method & peer review to stand on solid ground. I'm more inclined to take SM's word that....
"The gauntlet that had to be run shows that practices in climate science journals remain unchanged despite Climategate."
The façade is failing all over the place, I read a resident physicist climate attack dog on one particular forum claiming the AO was only discovered because they subtracted the effects of Co2, and therefore the AO is man made - this is the level of twisted logic true believers can conjure up.
As A. Said at WUWT, when they start claiming warming causes cold we must draw a line in the sand - this far and no more!
The original Steig paper used statistical methods to exaggerate local warming over the entire continent of Antarctica. This was not caught out at peer-review but the paper was given wide publicity by Nature. Bloggers caught these errors as soon as the paper was available, but Steig refused to take them seriously. One peer-reviewer then generated 88 pages of comments and played obstruction - for close to a year, and the paper was published in spite of this obstructionism.
Only Andy Revkin can spin this in favor of the team.
The latest project seems to be building up the scientific credibility of peer review in climate science. How heroic.
The thing that mystifies me about Revkin is that any of the doubters think him open-minded.
If the Steig peer review had been any good he wouldn't have been able to publish his paper. Anyone with any experience in marketing, engineering or any of many disciplines that use statistics knows that "interpolating" means you don't know what the actual data is so you've guessed it. You can guess scientifically, but it's still a guess and has no value whatsoever in proving anything. Steig interpolated his data for the vast majority of the Antarctic.
Having read the Revkin post and not being a scientist. Does the phrase "the results are artifacts" actually mean "the results are WRONG" ?
Just a question out of interest...
Perhaps if Revkin were to read Chapter 5 of the Bish's Hocky Stick Illusion, where Nature Magazine plays Lucy holding the football to M&M's Charlie Brown, Steve Mc's barbed comments about scientific publishiing and "Big Science" (my words) wouldn't ring so hollow.
Geronimo,
Agreed. I just commented at Jeff Id's that Steig made two mistakes. First, he screwed the stats badly. Second, he used crappy data from stations with grossly inadequate geographic coverage to try to make definitive statements about temp trends. I think the 2d mistake is the more important one. Which makes arguing about trends or confirmations of papers pointless.
@ Mactheknife
In a nutshell - yes. Ryan O'Donnell, being the nice chap he so obviously is and in an effort to cool the tiresome climate wars, termed the paper an improvement to S09. Unfortunately, the RC boys jumped on this as a way out and are spinning their little tops off trying to show that the new paper merely confirms theirs. The truth is it does no such thing. It well and truly kicks S09 into the long grass. The peanut galleries and the echo chambers may well fall for their spin. But real scientists and real statisticians now know the truth.
@LC
Thanks. I got the feeling having read the comments on the differences highlighted between the papers that "artifact" was a euphemism for "wrong". It seemed that the authors were a little miffed at the RC crew misinterpreting the results...probably not for the first time tho' eh ?
How about Ryan O's reply in the comments being followed 1 minute later by Steig's "rebuttal"?
And the completely unprovoked comment about Steve McIntyre, ".....the dominant figure among those who have attempted to take ownership of an implicit scientific trait, skepticism, and make it a personal badge of honor"...........
Revkin does seem to occasionally try to offer a level playing field for ideas, but these slants, always in favor of the consensus keepers, shows that is indeed a tool........
"The thing that mystifies me about Revkin is that any of the doubters think him open-minded."
Revkin, like Trenberth*, shows occasional signs of having a brilliant and open mind. Unfortunately, he's been influenced by threats of "the big cutoff," so he now consistently adheres to the party line.
* Trenberth was the one who said of Steig's study: "I remain somewhat skeptical. It is hard to make data where none exist." That was the most incisive comment about Steig 2009 made at the time. Amazing.
"artifact" (or "artefact") means that the answer was manufactured. It is usually used in similar contexts to mean that the result is simply a product introduced by the method used, rather than being present in the data. So it does mean wrong, but in a particular and reasonably precise sense.
I commented on Dot Earth through most of 2008, and I found Andy Revkin to have a immense curiosity and intellectual integrity. I would have thought he'd have figured out the meaning of ClimateGate by now.
But this post of his? He'll regret it soon enough.
====================================
Yes,
Revkin's has managed to squeeze in two unprovoked statements against McIntyre. That is how he fends off the "big cutoff"?
Any hints as to why Revkin will regret this post kim?
Comment 33:
We are cooling, Shub, for how long even kim doesn't know.
==============
I don't know TC, but I worry about a trap. The evidence needs to be presented, but perhaps not by Steve McIntyre. Were I editor I'd assign Revkin. Hey, there's a sharp guy.
==================
88 pages of review comments AND response. The review questions was one page with "how far does the wild goose fly?"
Sorry. It was there so I said it.