Multiscale Simulation of Moist Global Atmospheric Flows
Collaborative Institutional Lead(s):
This proposal seeks to advance theoretical methodologies and their efficient implementations for very-high-resolution non-hydrostatic simulation of the Earth's atmosphere general circulation. The key objective of the proposed work is a universally accurate representation of moist processes on structured grids and unstructured meshes. With higher resolution models becoming available, traditional numerical approaches face new demands. Compressible dynamics are universally valid across the entire range of spatial and temporal scales (i.e., from small-scale turbulence to planetary circulations) but impose computational limitations that are difficult to overcome. Current community efforts focus on alternative forms of the governing equations for modeling large-scale dry atmospheric motions. Efficient numerical simulation of moist processes in very-high-resolution cloud-resolving general circulation models remains an uncharted territory.
There is significant experience in modeling moist processes at the opposite limits of the spatial scales involved (i.e., small-scale non-hydrostatic versus large-scale hydrostatic dynamics and thermodynamics). A practical approach, suitable for multiscale simulation of weather and climate, that combines experiences from large-scale and small-scale dynamics calls for unification of moist thermodynamics. We propose to develop methodologies suitable for moist thermodynamics across the range of scales, aiming at very-high-resolution non-hydrostatic simulation of atmospheric general circulation. Theoretical developments will be supported by practical implementations and sensitivity studies using idealized and realistic high-resolution simulations of moist atmospheric global flows. In numerical implementations of the proposed theoretical developments, we will rely on the existing EULAG model (see www.mmm.ucar.edu/eulag) and its newly developed unstructured-mesh modeling framework. We also propose to investigate how the flexibility of mesh topology and the refinement offered by unstructured discretization can benefit the representation of key physical processes. Overall, the project will benefit the Earth Modeling Systems that are being developed at the University Corporation for Atmospheric Research/National Center for Atmospheric Research, the Department of Energy, the National Oceanic and Atmospheric Administration, and the European Center for Medium-range Weather Forecast.
This project builds upon existing collaborations between scientists involved in the proposal. Drs. Smolarkiewicz, Szmelter, and Wedi have a long history of collaborative research documented by joint publications. Drs. Grabowski and Smolarkiewicz are involved with a proposal recently submitted by Szmelter to UK's National Environmental Research Council. Both Grabowski and Smolarkiewicz have a long history of collaboration with Professor Malinowski and PhD student supervision at the University of Warsaw. Professor Klein is a leading European applied mathematician involved in theoretical aspects of atmospheric dynamics and their modeling. This project will establish close ties with his group and the German project Metstrom the Professor Klein is heading (see http://metstroem.mi.fu-berlin.de/).