Monday, July 9, 2012

The Trouble With Models...

Computer models of real-world phenomena all have the same weakness: the value of their output depends completely on the validity of the data and assumptions incorporated in the model.  Business folks are very familiar with this weakness, as it plagues the world of financial forecasting.  Climate models, I (and others) have long argued have the same problem – they're dependent on data and assumptions that have not been verified.

Today comes news of yet another such assumption: how heat is transported from inside the Sun to the outside (where it is radiated to the Earth).  This matters intensely to climate models, as their primary energy input is solar radiation – and if the models use the wrong assumptions, in particular, about variability of solar radiation, then...they'll just be wrong...

No comments:

Post a Comment