The other is Cari Kaufman, who builds probability distributions over the results of a climate simulation. (the idea seems to be to extrapolate from simulations actually run with similar but not identical parameters)

I was introduced to the idea of 'emulation' of complex models by Tony O'Hagan a few years back, where you use a Gaussian Process to model what a black box simulation will give across all possible inputs, seeded with actual simulation runs that you performed. (This also helps with active learning, in that you can find the regions of the input space where you're most uncertain what the simulation will give, and then run a simulation with those input parameters.) I believe the first application it saw was also in climate modeling.