To: Nadine Carroll who wrote (226605 ) 4/10/2007 10:04:13 AM From: neolib Read Replies (2) | Respond to of 281500 Has anybody got any model runs from 10 years ago that can be checked against empirical data? Just asking. The current models do a pretty good fit for the last 150 years when used with empirical data for the input forcing functions (like CO2). That is a very important point, and one the bashers don't like to discuss. The models are actually quite good. The uncertanties in forward predictions are related more to assumptions about the forcing inputs. We can't say for sure what CO2 production and CH4 production will do. Hansen, the well known NASA scientist who helped start public awareness of global warming in 1988 published a graph based on global climate models, which although the model was not as complex as current ones, did an excellent job of predicting what has happened since. So that is an example from 19 years ago looking at predictive abilities of these models. See here:giss.nasa.gov Note the very amusing example of a basher, one Pat Michaels who testified to congress a decade after Hansen and tried to "debunk" Hansen by modifying his graph to leave out two of the three model runs, including the one (line B) which Hansen had stated was most likely. The three lines were of course, for different CO2 input projections (and also included the cooling effects of a volcano, which bizarrely enough did happen!). In defense of Pat, I will note that he was also using data from two sources (balloons and sateliites) which had the time indicated cooling, but these well known problems were resolved, which was a good piece of science in itself. Anyway, I repeat, the models are pretty good, but there is more wiggle room reguarding what the inputs will be. Further, IF the severe predictions come to pass, THEN the models have more problems BECAUSE they are operating further away from the currently calibrated baselines. But they should still be pretty good for the next century, unless all Hell breaks loose.Would you happen to know how climatologists verify modern temp data to eliminate such political effects, or the effects of urbanization on temp reporting? The urban heat problem has been fairly well checked out and is not an issue. A little googling got me here, which gives a good discussion:realclimate.org