Adaptive Atmospheric Model Development on Unstructured Grids
The numerical modeling of atmospheric flows and its influence on atmospheric climate variability plays a fundamental role in the physical understanding of climate changes. Interannual variations of the planetary wave patterns and anomalies of the quasi-stationary circulation structure are expressions of atmospheric climate variability. Such variations of the large-scale planetary wave patterns are caused by complex non-linear dynamical processes operating on a wide range of spatial and temporal scales. The successful modeling of these multi-scale interaction processes are one key to improve existing weather- and climate models.
PLASMA - Parallel adaptive model of the atmosphere
With advances in computer architectures (e.g., parallel computing) and numerical analysis over the past decade, the construction of more accurate and efficient global models has become possible. In 2001, two groups at the Alfred Wegener Institute for Polar and Marine Research, namely the Atmospheric Circulation group in Potsdam and the Computing and Data Centre in Bremerhaven, together with the Technische Universität München have established a close cooperation to develop a parallel adaptive model of the atmosphere (PLASMA).
- PLASMA - Parallel LArge-scape Self-adaptive Model of the Atmosphere
- supported by German climate research program (DEKLIM), Federal Ministry of Education and Research of Germany (BMBF)
A barotropic model version has been developed, which is based on the spherical shallow water equations and is an application of a Lagrange-Galerkin method (Finte element method + semi-Lagrangian method). The model adapts the unstructured triangular grid, maintained by the grid generator amatos, at every time step according to a physical error indicator. This method offers the dynamical flexibility to resolve sharp orographic gradient as well as to follow transient dynamical structures, saving grid points at the same time in another region. The arising large linear systems are solved by the parallel solver interface FoSSI. Thus, compared to uniform grid experiments the number of grid points can be reduced significantly. Experimental convergence is shown by means of steady-state and unsteady analytical solutions. PLASMA yields satisfactory results for quasi standard experiments, that is the Rossby-Haurwitz wave and zonal flows over an isolated mountain. Work is going on to extend this approach to a hydrostatic baroclinic atmospheric model.
Läuter, M., Handorf, D., Rakowsky, N., Behrens, J., Frickenhaus, S., Best, M., Dethloff, K., Hiller, W. (2007). A parallel adaptive barotropic model of the atmosphere, Journal of Computational Physics, Volume 223, Issue 2, 609-628
Läuter, M., Handorf, D., Dethloff, K. (2005). Unsteady analytical solutions of the spherical shallow water equations, Journal of Computational Physics, 210, 535-553
Läuter, M. (2004). Großräumige Zirkulationsstrukturen in einem nichtlinearen adaptiven
Atmosphärenmodell, Berichte zur Polar- und Meeresforschung, 497, Dissertation,
Behrens, J., Rakowsky, N., Hiller, W., Handorf, D., Läuter, M., Päpke, J., Dethloff, K. (2005). amatos: Parallel adaptive mesh generator for atmospheric and oceanic simulation, Ocean Modelling 10, 171-183