[t vago]

Oh, and here's a Newsweek article about one of the co-authors of the IPCC report that you must believe in, Neil, because the IPCC report forms the basis for UN involvement with AGW.

Another money quote:

But here's the rub: water vapor is not well understood. Models, for instance, assume that a warmer atmosphere would hold more water vapor, but it wouldn't necessarily, says [Alfred P. Sloan Professor of Meteorology at the Massachusetts Institute of Technology Richard] Lindzen. Another wild card is the role of clouds in regulating humidity. Cumulus clouds draw moist air from the surface and carry it skyward. Some of the moisture falls back to the ground as rain, and what's left over is taken high up in the atmosphere, where it freezes into cirrus clouds. These clouds drift hundreds of miles raining ice particles into the lower atmosphere; these evaporate and raise humidity. But how much? Lindzen asserts that as the atmosphere warms, cumulus clouds will produce rain more efficiently, thereby leaving less for humidity-causing cirrus clouds. The result would be drier air. Rather than amplifying the greenhouse effect of carbon dioxide, this would counteract it.

And another money quote:

Even if scientists understood climate perfectly, the models would still contain another type of error inherent in the way computers do the calculations. In an ideal world, models would account for everything, down to each molecule of water. In practice, compromises are made. The Hadley Centre's model, for instance, dices the atmosphere into 250-kilometer squares, and then crunches equations that describe scientists' best approximation of the atmosphere's aggregate behavior. Making the squares smaller would reduce error, but it's expensive: shrink the squares to 125km, and the calculation balloons 16-fold. Even so, much of what goes on at the scale of clouds is lost.