Climate Change Models

Realclimate has a supposed FAQ post on global climate models that I find disappointing. I get the impression that these models are like the large macro forecasting models that economists build, but which have a bad reputation for accuracy and have been sneered at at least since I was a grad student in 1980, though they, like climate change models, were pushed as sophisticated science at one time. The macro models tried to put in lots of different interactions and calibrate them with data from lots of sectors of the economy, and then were fine-tuned by nudging things to make the numbers come out to be plausible. It turns out that mindless autoregressions-- basically, extrapolation without theory-- works just as well.

I'd like to see a presentation of a climate change model that laid out its assumptions.