An interesting look at the IPCC

John Christy, Professor of Atmosphere Science at the University of Alabama and a contributor to the IPCC reports has an interesting take on its work.

First, he notes the politics behind the IPCC:

At an IPCC Lead Authors' meeting in New Zealand, I well remember a conversation over lunch with three Europeans, unknown to me but who served as authors on other chapters. I sat at their table because it was convenient.

After introducing myself, I sat in silence as their discussion continued, which boiled down to this: "We must write this report so strongly that it will convince the US to sign the Kyoto Protocol."

Politics, at least for a few of the Lead Authors, was very much part and parcel of the process.

And this about skepticism:

Scepticism, a hallmark of science, is frowned upon. (I suspect the IPCC bureaucracy cringes whenever I'm identified as an IPCC Lead Author.)

The signature statement of the 2007 IPCC report may be paraphrased as this: "We are 90% confident that most of the warming in the past 50 years is due to humans."

We are not told here that this assertion is based on computer model output, not direct observation. The simple fact is we don't have thermometers marked with "this much is human-caused" and "this much is natural".

So, I would have written this conclusion as "Our climate models are incapable of reproducing the last 50 years of surface temperatures without a push from how we think greenhouse gases influence the climate. Other processes may also account for much of this change."

And, lastly about the models used by the IPCC:

Our group is one of the few that builds a variety of climate datasets from scratch for tests just like this.

Since we build the datasets here, we have an urge to be sceptical about arguments-from-authority in favour of the real, though imperfect, observations.

In these model vs data comparisons, we find gross inconsistencies - hence I am sceptical of our ability to claim cause and effect about both past and future climate states.

Mother Nature is incredibly complex, and to think we mortals are so clever and so perceptive that we can create computer code that accurately reproduces the millions of processes that determine climate is hubris (think of predicting the complexities of clouds).

Of all scientists, climate scientists should be the most humble. Our cousins in the one-to-five-day weather prediction business learned this long ago, partly because they were held accountable for their predictions every day.

Answering the question about how much warming has occurred because of increases in greenhouse gases and what we may expect in the future still holds enormous uncertainty, in my view.