Discussion of “utility and drawbacks of traditional approaches”

I have three comments on the topics which were touched upon this morning by other people concerning the predictive power of closure and other models. I think the situation is slightly better than suggested by various people. There is one particularly good success. You know that the inverse cascade which takes place in twodimensional incompressible flow turbulence was predicted first on the basis of absolute equilibrium models, and DIA and DIA-related models by Bob Kraichnan [1], and before there was any serious evidence that it existed. The evidence came later in numerical form [2,3,4]. I'm sure one can find other successful closure-based predictions but this one is especially prominent.