"Tony Cox" wrote:
I claim to understand the _modeling_ process, not the
details of their models. Modeling any phenomena is a
complex business.

If you know the difference between an initial-value problem and a boundary-
value problem then you understand the difference between climate modeling
and weather modeling. If not, then it is because you have not had training
in a physical science, differential equations, and numerical analysis.

To make an analogy using aviation content, predicting weather is like
predicting the details of the formation of turbulent flow over an airframe.
It's very sensitive to initial conditions. On the other hand, predicting
climate is like predicting the stall angle of a wing. The specific details
of the turbulent flow are not vital to determining the answer - because the
stall angle is a macro averaged effect.

Another analogy sometimes used is boiling a pot of water over a fire.
Predicting weather is like predicting when and where a bubble will form in
that pot. Predicting climate is like predicting how long the water will
take to boil off - or (trivially) what the average temperature of the water
is during the boil.

A good model makes predictions that can be tested.
One that I know of -- predicting increasing temperatures
in the stratosphere -- it has apparently failed. The others involve
climate variation which can't be measured until after
the proponents have conveniently retired. This
doesn't give one much confidence.

Um, Dr. James Hansen is not yet retired. How do you account for the
correlation between observations and the temperature changes his model
predicted some 20 years ago?