ガーナ国周辺地域におけるWRFによる気象予報試行

Abstract

A trial meteorological forecasting over Ghana was conducted with the use of Weather Research and Forecasting model (WRF) and data assimilation system (WRFDA). Three di- mensional data assimilation (3DVAR) was applied to a flash flood event that occurred at Accra Ghana on 3rd June 2015 for better rainfall forecasts in this study. It is found that the horizontal scale of the convective systems that spawn a flash flood was about 1000 km. SYNOP and SONDE data were used for our assimilation. The assimilation of SYNOP data provides a quantitatively positive feedback on the rainfall forecast. On the other hand, the location of the predicted rainfall distribution is shifted to westward when SONDE data is assimilated to the initial condition. SYNOP data mostly improved moisture conditions, whereas SONDE data overestimated zonal winds in this case. While SYNOP data used in this study does not include Ghanaian area, the results suggests that there is one of a possi- bility for an improvement of rainfall forecasts when Ghanaian SYNOP data is assimilated.