Yes, it turns out that if you put something off long enough, you won't need to do it after all.

I knew that if I procrastinated long enough, someone else would comment on skill in climate models, and they have, much better than I could. The Climate Science weblog, run by Dr. Roger Pielke’s group at Colorado State, is an excellent discussion on the climate topics of the day by real scientists with facts and no hysteria, even when they disagree with each other. Where else on the Web are you going to get that? (you'll have to scroll way down to get to the skill discussion)

One thing that is very clear and that the IPCC and other various climate commissions (AKA “grant money furnaces”) even admit is that the global models currently being used to influence public policy show very little skill in predicting (or projecting- see Pielke again for a discussion on whether there’s even a difference) regional effects. And yet it is on the REGIONAL scale that any political or social adjustments must be implemented. There is no “one-size-fits-all” global approach to dealing with climate change. Some areas may experience wetter conditions and have to manage floods; some will have to find a way around increasing desertification. Low-lying nations may have to deal with rising sea levels. Some regions will have warmer temperatures and longer growing seasons; some may have to deal with the opposite. Change is the normal state of the physical world. We live on a dynamic planet that goes through its own cycles in the atmosphere and lithosphere without regard to the convenience of the creatures that inhabit it. Those who insist that we have any real control over these are fooling themselves (or running for public office, or trying to get more funding, or any two of these conditions). Dealing with climate change should be like dealing with life in a tectonically active area, i.e. you plan for the worst and get out of nature’s way as much as possible to minimize your losses. It should be even easier, really, because these changes won’t happen as suddenly as an earthquake or tsunami, so the costs can be spread over a longer period of time. But to make realistic and effective plans, you must have reasonable projections as to what may happen, which brings us back to the models again.

Right now the climate modeling community, or at least the ones that get all the press, seem to be stuck on global models that emphasize atmospheric CO₂, especially anthropogenic CO₂ as a major climate forcing mechanism. Using their current measures of “skill”, they will only be brought into closer agreement with each other, not necessarily with reality. It’s comparable to the history behind the determination of the charge of an electron as recounted by Richard Feynman in a commencement address at Caltech in 1974 called “Cargo Cult Science”. The speech contains many valuable insights about scientific integrity that should be required reading for all scientists, and especially climate scientists these days. It has been reprinted in the book “Surely You’re Joking, Mr. Feynman”. Anyway, and I’m recalling this from memory, between 1908 and 1917 Robert Millikan measured the charge on an electron using an experiment with falling oil drops which turned out to give an incorrect result because of an error in some other value he used in his calculations. The interesting part is that when other physicists started trying to determine this charge, they would get different numbers but would throw out the ones that were too far off from Millikan’s because they figured that something must be wrong with them. The correct charge was finally found and agreed upon, but it took a while due to this herd phenomenon. Climate modeling seems to be falling into the same trap only with more serious consequences all around if policy makers are really going to use the predications made by these models to enact laws and regulations. Bringing all the models in agreement with each other just means they could be equally wrong, not that there’s any additional insight into the climate system. By the time the future rolls around and the predictions can be checked, those responsible for the models and their results will have gone to that big Research Institute in the sky, or at least retired. Another concern I have is HOW these models are brought into agreement. By “tweaking” certain elements until the desired result is achieved? How does this advance our understanding of climate physics again? Why should I believe that they really do represent reality? I really wish someone would tell me.