Hasn't this been out for a while already?

In your example the system's chaotic features can be neglected as you increase the length scale. I haven't seen anything that shows that this applies to modeling of the earth... Also chaotic systems become increasingly difficult to predict as the time scale increases... especially global trends. The problem with climate is that we are dealing with high order non-linear differential equations with very many terms. The ignorance of even a few of these terms can completely invalidate any predictions of future events and trends. In the book Chaos Making a New Science it mentions that many attempts at climate models have found only one stable solution and that is a frozen earth that reflects all energy out. It mentions that people who tried could not find a "normal temperature" for the earth. Warming trends and ice ages could just be products of the chaotic system we live in... sure adding CO2 might raise temperatures in the short span of a few decades but what will be the result in a few hundred years? Also, I remember hearing somewhere that water vapor contributes to the greenhouse effect by many orders of magnitude more than CO2... does anybody know how much truth is in this?