The climatological mean wind field over the Baltic Sea has been simulated with a mesoscale atmospheric boundary layer model developed at the Department of Meteorology at the Uppsala University (MIUU) (Sandström, private communications). The geostrophic wind distribution was determined from wind data of the light house Ölands Södra Grund. To compare the results of the model with measurements at Näsudden on Gotland (Sweden) WASP has been used to calculate the climatological mean wind field at Näsudden with input data of the MIUU model. This showed that there are enormous differences for different wind directions in the climatological wind field over the western Baltic Sea. With the aim to get further information of this phenomenom, a detailed comparison of the wind climate of Näsudden, Ölands södra Grund and Almagrundet (also a light house in the Baltic Sea) was made regarding the four seasons and the whole year. This analysis showed that the geostrophic wind over the Baltic Sea cannot be regarded as constant in magnitude and direction, but there has to be a weak trough mainly in winter over Sweden and the Baltic Sea (regarding the main pressure field). This phenomenom is probably caused by thermal effects as well as by the Scandinavian mountains.
In the second part WASP simulations for Näsudden have been made with three different sets of input data (Ölands Södra Grund, Almagrundet and results of the MIUU model). At the comparison of the simulations and the measurements, large differences appeared in the wind field, in the mean monthly wind speed resp. in the yearly mean wind speed and in the mean energy density. The deviations of the mean energy density calculated by WASP were much bigger than the deviations of the mean wind speed. With the same simulated mean wind speed, deviations up to more than 30 % appeared between the different simulations depending on the frequency distribution.
The yearly mean wind speed as well as the yearly mean energy density of the three simulations were compared. The result of the MIUU model combined with WASP was best with almost the same mean wind speed (but it can also be a coincidence). The two other results (simulations only with WASP with input data from the light houses) were about 0.5 m/s too high. As regards the yearly mean energy density the simulation with input data from Almagrundet was the best, but the yearly mean energy density was always overestimated with 16—28 %. It has been showed that the yearly mean wind speeds calculated as the mean of all 12 months and as the mean of January, April, July and October are almost exactly the same. Also the mean energy densities calculated as the mean of all 12 months and as the mean of the four months are about the same.
The biggest problem for the MIUU model simulations is the determination of the ’right’ geostrophic wind field.
Arkivexemplar i Uppsala universitets arkiv