Abstract
The typical climate computational workflow involves evaluation and serialization of the remapping weights during the pre-processing step (using ESMF, SCRIP or TempestRemap), which is then consumed by the coupled driver infrastructure (MCT) during simulation to compute coupled solution field projections. However, such a two step process presents several hurdles in terms of the scientific workflow, and impedes research productivity. In order to overcome these limitations, we present a fully integrated infrastructure within E3SM, based on the Mesh Oriented datABase (MOAB) library, which allows for a complete description of the component numerical grids, and solution data used in each submodel. The Fortran compatible interfaces in MOAB, linked with TempestRemap is used in E3SM to enable scalable online remapping strategies without a need for a separate offline step. We demonstrate the computational efficiency of these remapping algorithms in comparison with other state-of-science tools, and present strong scaling results on large-scale machines for computing projection weights between the spectral-element atmosphere (HOMME) and finite-volume discretizations on the polygonal ocean grids (MPAS).