Wednesday, November 25, 2009

Hackers, Hockey Sticks, and Hot Air

By now most have heard about the hacking of emails at the East Anglia Climate Research Unit, with revelations of questionable conduct by Michael Mann, Phil Jones, and others, including apparent efforts to cover up embarrassing facts and manipulate refereeing processes at journals. There have been piles of posts all over, and many people spouting off about this, with global warming skeptics pronouncing this to be the definitive proof that they are right (or at least not as unprofessional as some of those arguing for anthropogenic global warming). I note the response of those at the CRU. I also note that they should indeed be embarrassed at some of which has been uncovered.

It is indeed unfortunate that the scientific discussion has degenerated into such a slime pit of personalistic attacks and exaggerations. The skeptics are on the attack now, but the other crowd has attacked as well, and much of this is truly internecine. Thus, two of the major contenders and players in the emails are Michael Mann and Patrick Michaels. They were both in the same department at the University of Virginia for some time and were both there when the great controversy over the "hockey stick" erupted (neither are there now). This was a diagram first cooked up by Mann that showed rapid recent warming compared to earlier periods, which received lots of publicity, including in one IPCC report. It turned out the original study was flawed in terms of both data and statistical methodology. Some of the more pathetic floundering in the emails surrounds this. However, it must be noted that more recent data and better statistical methodology has largely confirmed the hockey stick result. We have been in a period of amazingly exceptionally rapid warming recently.

I will also note that this hack job may well be revenge for an attack from the other side on Michaels that was overdone. This was triggered by a Paul Krugman column on May 27, 2006, in which he attacked Michaels for testimony he gave a Senate committee about other testimony by James Hanson. While I agreed that Michaels could be criticized for some of what he said (just as I think Mann and crew can be criticized for some of what they have said), I thought at the time and said so in various places that the attack on him was overdone. One can read Brad DeLong's account of Krugman's argument, Michaels' own defense of himself, and Hansen's take on it. This stuff has been going on for some time.


Brenda Rosser said...

Well I must admit that I was very skeptical about the skeptics. If only they would be skeptical with their own money and refuse funding from vested interests!

I've been searching the leadked archive for words like: 'alarmed', 'alarming', 'catastrophic'. All if get are a group of folks sharing info with each other; sometimes expressing concern about fraud on the skeptics side of the equation.

There's nothing that I've read so far that justifies the beat up that this story is getting.

(a search engine at this link, for those who want to check for themselves)

Tim Lambert said...

Pat Michaels is a liar and a fraud, but I think it is a bit much to suggest that he was responsible for the theft of the emails from CRU.

Barkley Rosser said...


I can see that it could appear that I suggested that. This was not my intention, although certainly people who agree with him (or even more so, as his position is more nuanced than most realize) probably did so.

gordon said...

Since it's a big issue (particularly on the eve of the Copenhagen meeting) I'll take the liberty of putting up a slightly edited version of a comment I made yesterday at Economist's View.

"[A previous commenter linked to] a CBS News blog called "Taking Liberties", which reproduces text from (I suppose) the leaked material. The [section he refers to] is a lament by a programmer who is wrestling with old and undocumented code and data which has obscure problems, again undocumented.

"The blog post finishes up: "The irony of this situation is that most of us expect science to be conducted in the open, without unpublished secret data, hidden agendas, and computer programs of dubious reliability".

"Yes, but modelling and science have an interesting relationship, much like modelling and economics. Climate models can run, I'm told, to millions of lines of code. Maintaining and updating and documenting that kind of monster to a good standard needs a standing army. Few research institutes are likely to be able to muster the resources.

"If anything, what seems to have come out so far might indicate that we have been relying too much on models.

"As far as upsetting the global warming theory wholesale, I don't know. Logically, if flawed models don't really offer credible support for the theory, discovering the models’ flaws doesn't destroy it either. It might damage the credibility of the modellers, but that’s not the same thing.

"Before the GFC, there were some economists and others who worried about a US housing bubble, and predicted that when it burst, bad things would happen. Some of them have claimed "I told you so" since the collapse. But I don't know of anybody who not only predicted the bursting of the housing bubble but also how that would push over a huge and (with hindsight) unstable structure of deriviatives and insurance, leaving banks all over the world suddenly unable to meet their obligations. I could be wrong, but I can't think of anybody who successfully predicted in such detail.

"So were the Cassandras of the housing bubble actually wrong? Were policymakers right to ignore them, and can those policymakers now say that because those Cassandras didn't predict the entire course of the GFC in detail, they were in the wrong and deserved to be ignored? Is there a useful analogy here?"

Barkley Rosser said...


Not to pat myself on the back too much, but prior to the big crash I gave several plenary addresses that ended in me pounding my fist on tables and screaming about the complex interconnectedness of the derivatives markets based on the housing bubble and how when it blew things would get really bad. I tended to be much lower key on all this in print, including blogs, but I did so here at a crucial moment. So...

You make an important point about the runup to Copenhagen. Indeed, I think that is what is really behind this. The moneyed interests that want Copenhagen to fail probably were ultimately behind the hack job. But, Copenhagen is going to be at best a wet squib, if not a total bomb, thanks to the molasses of the US Senate.

But, in the end in science, facts do prevail, and we have seen serious temp runups with little good alternative explanation than human input, all the huffing and puffing about hockey sticks aside, which in the end has to do with what happened a thousand years ago.