How about: the uncertainties in their extrapolation to 400 years beyond the present are probably greater than the changes they are predicting.