Dry and Moist Idealized Experiments With a Two-Dimensional Spectral Element Model
Abstract
Numerical models used for mesoscale weather forecasting can be assembled into two groups depending on the approach used to solve the governing Navier-Stokes equations. In the first group, the equations are kept in the differential form and both temporal and spatial derivatives are approximated using finite differences. In the second group are methods based on an integral form, less frequently used in mesoscale forecasting, which includes spectral, pseudospectral, finite-element, spectral element and finite-volume models. Advances in high performance computing have led to substantial increase in the number of computational cores and to a smaller extent improved speed per core. The spectral element (SE) method has the potential to meet this changing and challenging computational paradigm. This method provides a flexible platform, which supports unstructured grids and provides flexibility to adjust the accuracy of the dynamical core with a simple change of a control parameter. The purpose of this paper is to assess the strengths and weaknesses of a SE model through the systematic exploration of the parameter space and validate simulation results of two idealized mesoscale phenomena: a linear, hydrostatic mountain wave and mid-latitude squall line.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2011
- Accession Number
- ADA547220
Entities
People
- Francis Giraldo
- James D. Doyle
- Sasa Gabersek
Organizations
- Naval Postgraduate School