Monday, 21 June 2004
The purpose of this study was to find suitable combinations of parameters in the ECMWF model to predict the probability of foehn events in the subgrid-scale Wipp Valley. Pressure gradient, wind speed and potential temperature from analysis and forecast data of ECMWF-T511 were investigated and compared to foehn observations made during and after MAP. The chosen gridpoints are located across the gap of the Alpine crest of model orography, which is a coarse approximation to the larger gap into which the real Wipp Valley is embedded.
The best single predictor for analysis data is the pressure difference between upstream and downstream gridpoints. With a pressure difference > 1.8hPa, the probability of foehn detection is about 85%. Potential temperature and wind speed only improve the detection when they were combined with each other. Then, especially onset and end of a foehn period are well detected. In addition, some further improvement is gained by combining pressure difference and wind velocity.
Also shown will be the evolution of the forecast skill with increasing lead time.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner