Wednesday, November 4, 2009

Cooler Heads with Dr Richard Lindzen

Below are embedded Dr Lindzen's talk, Deconstructing Global Warming, with interesting quotes transcribed from each part.

Part 1:

Part 2:

Part 3:
However, Hockfield's [President of MIT] response does reveal the characteristic feature of the current presentation of this issue: namely any and every statement is justified by an appeal to authority rather than by scientific argument.

Part 4:
This is the meat of the presentation.
Once people regard authority as a sufficient argument, then you're free to make any claim you wish, especially if you are in high government office, refer to authority, the authority frequently doesn't say what it is claimed to have said, but any advocate in high office can rest-assured that some authority will come along to assent.

...most claims of evidence of global warming are guilty of something called the prosecutors fallacy. In its simple form it runs as follows, for instance, if A shoots B, with near certainty there will be gunpowder evidence on A's hands. On the other hand, in a legal proceeding this is often flipped, so that it is argued that if C has evidence of gunpowder on his hands, then C shot B.

However, with global warming the line of argument is even sillier.

This rhetorical question is my favourite, it is the most damning criticism:
Whoever heard of science where you have a model, and you test it by running it over and over again?

Part 5:
Now it turns out, data is the way you usually check calculations.

This is where Dr Lindzen compares some hypothesis with satellite measurements. He does this by forcing models with observations and watching the response, then comparing this response to the measured response. Also discusses feedback in the climate system.
By the current logic, or illogic, of science, the fact that all models agree is used to imply that the result is 'robust'.

See this paper for a criticism of this type of model 'verification'.

The bottom line Dr Lindzen comes to after analysing the satellite data:
What we see is the very foundation of the issue of global warming is wrong.

Part 6:
In answer to a question about how many scientists "believe" / "don't believe":
How do you describe these people [scientists] as being for or against, it's meaningless, they're opportunistic.

Whenever I'm asked if I'm a climate skeptic, I always answer no. To the extent possible I am a climate denier. That's because skepticism assumes there is a good a priori case, but you have doubts about it. There isn't even a good a priori case.

I think the thing that bothers me most is the lack of rigorous model validation (defined for example in this AIAA standard). When you're providing analysis for decision makers you really should use the results of a validated model and give a quantitative uncertainty analysis. That enables ethical and responsible risk management, anything else is negligent (in the professional sense if not the criminal).


  1. Lindzen isn't the only one who is dismayed by lack of validation. I got Tufte's Beautiful Evidence book for Christmas, here's a relevant snippet:

    "Controlled trials are prospective: a possible cause-effect relationship is identified, an intervention is made, future outcomes observed. In contrast, evidence for many phenomena -- weather, politics, geology, economics, art history, business -- largely comes from retrospective, nonexperimental observation. In such after-the-fact studies, researchers and presenters have a good many opportunities to decide what counts as relevant evidence, which are also terrific opportunities for cherry-picking."

    Climategate anyone?