Jed - >> That's preposterous.
If you wish. It's also a fact. It's inherent in how the math works. >> If that were true, weather forecasting computer programs would not work. You are correct. You've heard of Lorenz, of course. The programs only work for a very brief time before their results degrade to useless noise, so they are only good before they reach that point - which is PDQ. An interesting thing is that no matter how large or how good your data set is, the same thing happens - unless you add artificial buffering or other programming contrivances to manipulate things towards the results you or your sponsors would like to see. The current attempts to model the ongoing workings of greenhouse gasses as they actually perform in the real world is nothing more than an exercise in computer science and chemistry which probably would only be interesting to academics had such work not come to be abused so badly in this current politically charged situation. There's quite a few other things besides this one that undo AGW, but this is the major deal breaker on the models issue which has driven a large part of the claims in favor of it. - Rick
