In my post about Keith Shine's contribution to the All-party Climate Change Group briefing on AR5 I said that Prof Shine had failed to discuss the divergence between model and observational estimates of climate sensitivity. In fact the report on which I had based this statement (and which I quoted in the original piece) misrepresented what Shine said. There is an audio file available here (H/T Richard Betts) which shows that a significant chunk of the talk was in fact given over to a consideration of the two main approaches to estimating climate sensitivity. Shine described the pros and cons of the various approaches as follows:
...state-of-the-art climate models, which are our embodiment of the laws of physics as applied to the atmosphere...
...a mixture of observations and simple models. Now you might think using observations were a better way of doing things but the problem is that there are a lot of things going on at the same time in the climate system - the internal variability, the aerosols - and there is also a lot of debate about how you should do your statistics to estimate the climate sensitivity.
I'm not sure I detect any question marks over the reliability of the models. No mentions of fudge either. And the remarks about debates over statistics are grossly misleading. The IPCC has been engaging in PR rather than science: showing a series of studies that all use flat priors in ECS is equivalent to repeated shouting "We already know that climate sensitivity is high". It's propaganda, not science.
So I'm not sure that Prof Shine is giving the politicians a fair assessment here. It looks more like an attempt to pooh-pooh the observational estimates.
Still, if you think this is bad, wait until you see the next post.