Karolygate V

IT IS almost inconceivable what it must feel like to wake up one morning and discover that you have lost $300,000. Not spent it on women and pleasure, as the old joke goes, but actually wasted it – left it on the tram or mistaken it for Sorbent in the dark of a bush night, that sort of thing. You would feel pretty low, that’s for sure, and it would only be human nature to first concede screwing up and then, by small degrees, invent little excuses and explanations to mitigate the crushing guilt of blowing all that dough.

The three-year climate study led by the Parkville Asylum’s Joelle Gergis, sorcerer’s apprentice to David Karoly, had cost precisely that sum when, on June 6 of this year, a co-author alerted his colleagues to the fact that they might just as well have sent that taxpayer cash down the pipes to Werribee. The bringer of sad tidings was Raphael Neukom, and nuke ’em he did with a very explicit explanation of the paper’s flaws. The entire letter is worth reading (and all the others in the same folio of correspondence) but the bits underlined are the most relevant to this post:

Subject: Mistake in the Australasian TT paper

Date: Wednesday, 6 June 2012 9:46AM

From: Raphael Neukom <neukom@giub.unibe.ch>

To: Joelle Gergis <jgergis@unimelb.edu.au>, David Karoly

<dkaroly@unimelb.edu . au>

Conversation: Mistake in the Australasian TT paper

Hi Joelle and David,

As just discussed with joelle on skype, I found a mistake in our paper in journal of climate today . It is related to the proxy screening, so it is a delicate issue. In the paper we write that we do the correlation analysis for the screening based on detrended {instrumental and proxy) data, but in reality we did not use detrended data.


The origin of the mistake is that at the stage when we were writing the paper my approaches have already evolved and I had made the proxy selection for the SH reconstruction based on detrended data. I therefore had in my mind that we had done the same for Australasia months ago and was very negligent not to check this carefully.


Using detrended data would only select very few proxy records that would not allow a reasonable reconstruction. I think it is basically justifiable to do the screening without detrending but changing these words may cause troubles.


Fortunately we have not received the proofs yet. So my suggestion is to write to the editor, explain the mistake and ask for permission to correct the error, if necessary via sending it out to review again.


I apologize for the mistake and the troubles it may cause and hope that we can find a good way to correct it.


David your advice on this would be very much appreciated.


Thanks a lot and best regards

Raphi

Let all that technical palaver go through to the keeper; there are plenty of science bloggers who get heavily into that sort of thing and it is to their sites the curious should turn for enlightenment. At the Billabong the fascination has been observing the evolution of an excuse. Follow the trail from the climate comrades’ emergency Skype session to the pages of national press and see how things work with grant-funded warmists.

First, Raphi concedes he was “very negligent” to use the wrong figures, a mistake he admits is at the root of the team’s problems. Yet he is desperate to be helpful, even to the point of embracing cognitive dissonance. Sticking with what has been submitted to the learned, peer-reviewed Climate Journal would not represent “a reasonable reconstruction”, he states in one sentence, only to suggest in the very next that it would be “basically justifiable” to bluff it out and leave things as they are.

As all the FOI emails from Melbourne University demonstrate, quite a bit of thought goes into satisfying reporters’ curiosity, which fortunately for the unsettled scientists was rather limited. Whatever conversations took place, whichever strategies were adopted, by the time Bernard Lane is explaining the paper’s flaw to readers of the The Australian on June 13, the explanation has evolved into something radically different. Indeed, it seems to have no relevance whatsoever to the reality of Rafi’s mea culpa email.

First, Karoly tells Lane that either of two methods might have been used, no doubt failing to mention one of them cannot produce a “reasonable reconstruction.”

Second, in explaining how such a mistake came to happen, Karoly makes no mention of his “very negligent” comrade’s oversight. Rather, he blames the rather more convenient, and previously unmentioned, ghost in the machine:

Professor Karoly said the data would be reanalysed using the year-to-year variations only. A switch in the computer code was wrongly set to include the long-term trend and this went unnoticed.

How the hapless Rafi feels about being dismissed as a humble and inanimate “switch” we will never know. What would be good to find out is how much knowledge the $300,000 poured into such a schermozzle actually purchased, where the money went and how many delusions were peddled to the public as part of a post-failure snow job.