Letter To The Editor: ‘Some Thoughts On Global Warming Simulations’

By RICHARD NEBEL
Los Alamos
 
Since everybody seems to be putting in their two cents on global warming, here’s mine. 
 
The equations which describe climate change are a coupled set of hyperbolic partial differential equations. Hyperbolic partial differential equations have the property that all solutions diverge (popularly known as the “butterfly effect”). 
 
So how far ahead might one “rigorously” predict the climate? 
 
About 5-10 days, or about the same length of time as the weather. In short, these simulations are running all over the Lyapunov numbers (which measures the rate at which solutions diverge) so rigorously your solutions will turn to crap in a few days.

Now this is not to imply that such simulations are without value. Even though solutions may diverge, they are still bounded. The boundedness comes in from the fact that there is a bounded amount of energy in the atmosphere, oceans, etc. 

 
One can argue, and reasonably so, that these simulation solutions are a sample solution that gets the major parts of the physics correct. And I’m also not criticizing the people who are doing the simulations. They are doing the best they can with what they have.

So here’s the question: How much better are these simulations than the simple global balances that predicted global warming back in the 1970s?
 

Sure, things that come out of big computers with lots of fancy graphics look a lot more impressive to the public. And they can show all kinds of detail. But how much of that detail is new physics and how much is “chocolate coated manure”? And how do you tell the difference?

Finally, one of the first things I was taught about scientific research is that one should not become emotionally attached to the results. The reason for that advice is that we all have a tendency to see what we want to see in the results instead of what is really there. It seems to me that people on all sides of the climate change issue are so passionate that it would make it extremely difficult to do an objective piece of work. And that is sad.

CSTsiteisloaded