Southern Hemisphere Forecasting Experiments with a Semi-Spectral Model

M. E. Voice Australian Numerical Meteorology Research Centre, Melbourne, Australia 3001

Search for other papers by M. E. Voice in
Current site
Google Scholar
PubMed
Close
and
B. G. Hunt Australian Numerical Meteorology Research Centre, Melbourne, Australia 3001

Search for other papers by B. G. Hunt in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

A semi-spectral model formulation has been tested for short-term numerical weather prediction in the Southern Hemisphere: The model configuration consisted of 5 levels in the vertical with 15 zonal waves and 30 grid points from pole to equator. Several two-day forecasts were verified and examined for synoptic accuracy and spectral performance. The semi-spectral technique produced credible and useful NWP results, competitive with standard grid point methods. An extended integration to 8 days demonstrated the stability of the model. The same model with a higher vertical resolution of 18 levels was briefly tested. Comparison with the 5-level version revealed only small improvements in forecasting accuracy and highlighted data limitations in the Southern Hemisphere.

Abstract

A semi-spectral model formulation has been tested for short-term numerical weather prediction in the Southern Hemisphere: The model configuration consisted of 5 levels in the vertical with 15 zonal waves and 30 grid points from pole to equator. Several two-day forecasts were verified and examined for synoptic accuracy and spectral performance. The semi-spectral technique produced credible and useful NWP results, competitive with standard grid point methods. An extended integration to 8 days demonstrated the stability of the model. The same model with a higher vertical resolution of 18 levels was briefly tested. Comparison with the 5-level version revealed only small improvements in forecasting accuracy and highlighted data limitations in the Southern Hemisphere.

Save