Tim,
all climate change predictions are based on computer generated extrapolations of past events, modified by applying a range of variable factors.
The finance sector does the same: recent evidence suggests that they really are not very good at it. So why should we believe the climatologists any more than the financial pundits?
The climate models have got an extremely long record of accuracy, remember that it is now 50 years since scientists issued the first warnings.
If you look at the predictions made by Wallace Broecker's 1974 model and the recorded observations, I would say the correlation seems pretty good. So I would say yes, 40 years of observations matching predictions does make the models trustworthy.
