@Latimer The model ensemble aims to provide a projection of the response of the climate to the change in forcings. Thus it is not a direct estimate of the evolution of the climate itself as the observed climate is the combination of the response to the forcings and a chaotic component corresponding to the unforced response (i.e. weather noise – stuff like ENSO). The latter is chaotic and hence is inherently unpredictable; all we can do is simulate it instead. Thus while the ensemble mean is the projection most likely to match the observations, there is no good reason to expect them to lie any closer than the spread of the model runs (which show the range of things considered plausible given the assumptions of the model). So if you see a model-observation comparison that doesn’t give an indication of the model spread, then there is a good chance it is bogus.
↧