| Popis: |
Two approaches are proposed for updating the geophysical characteristics of the surface (land use, albedo, leaf index, fraction of absorbed photosynthetically active radiation) in the Weather Research and Forecasting (WRF) numerical model for the territory of Belarus: updating the monthly averages based on modern Earth remote sensing databases GLASS (Global Land Surface Satellite), GLC2019 (Global Land Cover, 2019); daily update based on operational MODIS satellite composite products. To estimate the impact of the initial geophysical data on the quality of numerical prediction of surface temperature, a number of numerical experiments were carried out to predict various synoptic situations in the summer period. To assess the influence of on the quality of the numerical prediction of the surface temperature, a number of numerical experiments were performed to predict various synoptic situations in the summer period. A correction factor for the land surface albedo in the WRF model was calculated, which makes it possible to reduce the root-mean-square error of temperature forecast for the lead time of +12, +24, +36 and +48 h by 0.30 ºС, 0.10 ºС, 0.15 ºС and 0.16 ºC respectively. In numerical experiments the initialization of the WRF model using operational satellite products had the most positive effect on the surface temperature forecast for nighttime periods: for the lead time of +24 and +48 h the standard error decreased by 0.11 ºС and 0.14 ºС respectively. |
