[elephant]

I remain sceptical about the computer models. Basically, forecasts of weather 100 years from now are unlikely to be accurate, given that the best we can manage repeatably is a five-day weather forecast, and weather models are much more complex than environmental models.

Any computer model needs validation against known events - previous weather patterns. The warming up of the earth about 700ad that lasted to 1300ad, with temperatures in England similar to the Loire Valley today, and only trivial human CO2 emmission levels, followed by a change to colder temperatures, would be a good test. If a computer model tracked that period accurately, it would have some validity. Then of course on to Thames "Frost Fairs" in England from 1683 (Thames froze over for two months) which occured spradically until 1814, when England was colder. This was during the early industrial revolution, when CO2 levels were already rising.

Going back further, we have so-called "false ice ages", one only 95 thousand years ago. A "false" ice age is a sudden sharp drop into extreme glacial conditions that doesn't last, but ends after a few thousand years instead of persisting for ten's of thousands of years. This sudden drop from present temperatures into glacial conditions can take as little as 50 years. If the computer model is incomplete it may fail to predict such an event, so in 50 years time we could have a kilometer of ice on top of Glasgow, falling sea levels, and barren and arid tropics, instead of balmy weather and high sea levels. The ice would last for typically five thousand years and destroy human civilisation.

We are presently in a warm spell in the middle of an ice age, and the ice is expected to return one day. A warm (and hence wet) planet supports more life than a cold arid planet. It might be that global warming should be a target.