Abstract: An adaptive grid technique has been applied to a revised version of NCAR/NASA's next generation dynamical core for climate and weather research. This hydrostatic so-called Lin-Rood dynamics package is based on a conservative and monotonic finite volume discretization in flux form. The adaptive model design utilizes a spherical adaptive-grid library which is based on a cache-efficient block-structured data layout. The dynamical core is run in two configurations: the full 3D hydrostatic dynamical core on the sphere and the corresponding 2D shallow water model. This shallow water setup serves as an ideal testbed for the horizontal discretization and the 2D adaptive-mesh strategy. The static and dynamic adaptations are tested using a standard shallow water test suite and two idealized 3D baroclinic wave test cases.

Keywords: advection, atmospheric modeling, spherical adaptive grid, adaptive blocks, parallel computing, adaptive mesh refinement

Complete paper. This paper appears in ECMWF Proc. Recent Developments in Numerical Methods for Atmospheric and Climate Modeling (2004), pp. 233-250.

Related work

• My work in modeling climate, atmospheric, and space systems: papers and an overview of research.
• My work in parallel computing: papers, and an overview of research.
• General parallel computing: an explanation of parallel computing, a tutorial, Parallel Computing 101, and a list of resources

Copyright © 2005-2017 Quentin F. Stout