Crighton hates computers
February 21, 2007
Michael Crighton, the SciFi writer and science gadfly, talked about his disagreement with establishment views on global warming on The Charlie Rose Show recently. He does not believe there will be a catastrophic increase in global temperatures. However, he supports nearly all efforts to reduce CO2 ouput, so he’s not a shill for energy companies. It appears he actually has a problem with computer models of non-linear systems, e.g. the weather. It is difficult to predict the weather, and extremely difficult to predict the global climate 100 years from now. The weather is a chaotic system where small changes can have a large impact on a prediction. Scientists make simple approximations of weather systems, including many strange assumptions and constant parameters, so they can build and run computer climate models. The simplifying approximations lose all of the tiny complexities in the real world. In non-linear systems, seemingly tiny complexities can have large impacts on predictions; therefore, dropping them makes the model fairly useless. Economic models have the same problem: they simplify real economic systems into computer models, but the models turn out to be wrong a lot. Crighton argues that these simple computer models of complex non-linear systems are useless and the predictions are junk.
Crighton is right: anyone who says the climate will be X degrees warmer in 100 years is full of shit. No model can accurately predict the temperature 100 (or even 10) years from now. Crighton is also wrong: these models can reveal qualitative relationships between different natural forces. Models can show that doubling the amount of CO2 increases climate temperature somewhat, glaciers will melt when ocean temperatures reach some tipping point, and sea levels will rise a bit when glaciers melt. The models can’t tell us exactly how many inches the sea will rise, but it can suggest a relationship between CO2 and higher sea levels. This is enough to tell us that we are on the wrong path and should move to cleaner energy, though it doesn’t tell us exactly when the world will end. So the world should definitely reduce carbon, but there’s no reason to panic. Someone will figure out how to eat carbon out of the air with genetically engineered bacteria. Problem solved.