>The scientists went on to try 300 configurations of rain, pollution, and heat flows—something they can do as gods of their own digital earth—before matching th

>The scientists went on to try 300 configurations of rain, pollution, and heat flows—something they can do as gods of their own digital earth—before matching the model to history. Having worked with complex geologic (not climate) models in the past, this is the part that gives me pause. The scientists have set out as a first principle to find warming and then are working backwards and, ostensibly, everything is legitimate because model results are calibrated to previous real data. But here's the problem. When you have complex, chaotic models with hundreds of freely tweekable parameters, it's fairly easy to fit data and simultaneously get any future prediction you (or your bias) want. Even when individually each variable is within a reasonable range. We arguably have proof of this happening - climate models have been consistently overpredicting dT for what, 10 years now? There is an institutional problem with climate science, in that there is no real penalty for being wrong, and in fact the more aggressive results probably get the most attention. Because the modeling is so far removed from any sort of experimental or observational data collection, I think the field is dangerously susceptible to institutional bias.