Parallelizing the Spectral Transform Method

The spectral transform method is the standard numerical technique for solving partial differential equations on the sphere in global climate modeling. This paper describes the parallelization and performance of the spectral transform method for solving the nonlinear shallow water equations on the surface of a sphere using a 128-node Intel iPSC/860 hypercube. This work is part of a research program to develop climate models that are capable of much longer and more numerous simulations at a significantly finer resolution than are currently available. Such models are important in understanding the effects of the increasing atmospheric concentrations of greenhouse gases, and the computational requirements are such that massively parallel multiprocessors will be necessary to run such simulations in a reasonable amount of time. 10 refs., 6 figs., 1 tab.