FORECASTING OF LEE WAVES OVER COMPLEX TERRAIN BY MEANS OF A NON-HYROSTATIC FORECAST MODEL

Abstract

The examples show very impressively the possibilities offered by a high resolving non hydrostatic model for forecasting lee waves. Even if after the first test calculations it is too early to make definite statements on the quality of these forecasts, the results are astonishingly good in a comparison with the measured and calculated conditions. Position, horizontal extension and vertical structure of the ascending motions and downdraughts over the Alps in areas with high mountains and deep valleys are correctly described. Howevel, in comparison to the measurements, the vertical speeds are too low. This deficiency results from the fact that the actual structure of the relief cannot yet be corectly described even with a 2.5 kilometer grid. In reality, the mountains are still slightly higher, the slopes and flanks of the mountains are steeper and the valleys are deeper than in the model. For numerical experiments this will not cause a problem, however for the purpose of operation and routine forecasts we will have to content ourselves for the time being with model versions with mesh widths of 7 and in 2000 or 2001 of 2.5 kilometers, as even the fastest computers in operation are too slow for higher resolutions. From the beginning of 2000 the DWD will be making numerical forecasts of the vertical wind speed over the Alps available via the self briefing system pc_met. with the new cartographic representations and the selected cross sections across the Alps completely new possibilities will be opened up for the planning and performance of lee wave flights. The forecasts, however, will be beneficial to the entire aviation as the LM also offers new possibilities for a more accurate forecast of windward and lee effects and of mountain wave induced turbulence.