A semi-implicit forecast model using the flux form of the primitive equations
✍ Scribed by D. J. Gauntlett; L. M. Leslie; D. R. Hincksman
- Publisher
- John Wiley and Sons
- Year
- 1976
- Tongue
- English
- Weight
- 889 KB
- Volume
- 102
- Category
- Article
- ISSN
- 0035-9009
No coin nor oath required. For personal study only.
✦ Synopsis
Abstract
As part of an attempt to implement more efficient integration procedures, a semi‐implicit time differencing procedure has been evaluated within the context of a southern hemisphere, six‐level primitive equations model. The model employs the flux form of the primitive equations and includes processes such as convection, vertical and horizontal diffusion, and orography.
Comparative explicit/semi‐implicit integrations show only marginal meteorological differences. However, the semi‐implicit scheme allows a timestep about five times as large. This represents a net saving factor of approximately 2.5 for integrations involving no orography. In instances where high orography is included net efficiency advantages may be degraded significantly when using the flux form of the primitive equations unless special numerical procedures are adopted during the solution of key model equations.
On the basis of these successful results, the semi‐implicit scheme is now being used in the separate development of a limited‐area prediction model to provide routine forecasts for the Australasian region.
📜 SIMILAR VOLUMES
The Lagrangian-averaged Navier-Stokes alpha (LANS-a) model is a turbulence parameterization that has been shown to capture some of the most important features of high resolution ocean modeling at lower resolution. The LANS-a model improves turbulence statistics with an additional nonlinear term and
## Abstract We give a linear‐stability analysis of the two‐time‐level semi‐implicit discretization of the adiabatic fully‐compressible equations on an __f__‐plane. Previous work has shown that the scheme is stable with respect to perturbations to a hydrostatic and isothermal basic state if the same
A semi-implicit, semi-lagrangian algorithm suitable for the simulation of the dry, adiabatic, nonhydrostatic atmospheric dynamics is introduced and analysed. Height is used as vertical coordinate, without the customary terrain following normalization, thus resulting in a stable, robust, and efficien