Further Advancement and Implementations of a Global Framework of Quasi-Uniform Rectangular Grids

University Of Maryland, College Park, College Park MD

Investigators

Abstract

The objective of this project is to further expand the concept of quasi-uniform gridding of the sphere and explore its potential for improvement of various aspects of modeling of the large-scale atmospheric flow, including prediction of hurricanes and downscaling of the regional climate. The quasi-uniform rectangular grids (cubic and octagonal) that will be studied for global circulation models by the PI and his collaborators provide an attractive numerical paradigm because of their capability to efficiently handle the computing resources, high resolution in tropical and equatorial regions, and absence of strong singularities around geographical poles. A global modeling framework employing the concept of quasi-uniform grids has been recently developed and successfully preliminary tested by the PI using numerical infrastructure of the National Center for Environmental Predictions' (NCEP) regional model as a prototype. A set of various expansions and implementations of the developed framework is used in this project. The expansions include several variations of the quasi-uniform grids which will be able to describe the variable resolution; a new, curvilinear formalism that may further improve the efficiency; higher order Arakawa type schemes (4th and 6th order); a new innovative method for treatment of the lower boundary, and merging with NCEP's version of the non-hydrostatic Weather Research and Forecasting (WRF) model. The most important of the implementations include testing of the variable resolution quasi-uniform framework in the medium range forecasting of hurricanes and in the regional climate simulations using data from the Stretched Grid Model Intercomparison Project. The challenging novel ideas for numerical modeling of the large scale atmospheric flow, such as new global grids, new numerical schemes and new approach for treatment of the terrain constitute the intellectual merit of the research. This research may deliver important new pointers for advancement of weather forecasting in general, and particularly the skill of hurricane prediction, by which it will realize the broader impact on the community. In addition, the developed numerical techniques will be transferable to other geophysical disciplines dealing with mapping of the Earth's surface, such as oceanography and seismology.