The Regional Atmospheric Modeling System (RAMS) developed at Colorado State University (CSU) was used to investigate the influence of the large-scale monsoonal winds and the mesoscale local circulations on the diurnal precipitation pattern over Kenya. Three basic experiments were performed. In the first control experiment (CONTROL),RAMS was initialized using observational data (''variable initialization'') from the global analyzed ECMWF 2.5 degrees X 2.5 degrees data of 0000 UTC 14 April 1985. The model was integrated forward in time for 24 h to simulate the large-scale Bow fields over Kenya. Full physics including moist convection were implemented in the model. The model outputs were validated against available observations in order to determine the ability of the model in replicating the synoptic climatology prevailing over the study domain. In the second experiment (MESO), the model simulation was started from an atmosphere at rest in order to exclude the large-scale flow from the model runs. This experiment used a horizontally homogeneous procedure to initialize the model. Moist convection was allowed to occur in the mesoscale simulations. The third experiment (SYNO) excluded topography and land-water contrast data from the simulation. This was done in order to suppress the thermally induced local mesoscale circulations from the model runs so as to isolate and determine the role played by the synoptic scale monsoonal winds on the generation of precipitation over Kenya. Topography was considered a local forcing that significantly modifies the large scale temperature, moisture, and flow pattern. Comparison was made between grid-averaged convective precipitation generated in the CONTROL, MESO, and SYNO simulations in order to identify the contribution of the synoptic-scale monsoonal Bow and mesoscale circulations on the rainfall over Kenya. Additional sensitivity experiments were performed to test the impact of topography and large water bodies on the precipitation over the country. The results showed the following. More active convection developed in regions where the large-scale monsoonal winds in the lower troposphere converged with the local mesoscale circulations embedded in the large-scale Bow. The large-scale flow advected substantially more moisture into the study domain as compared to the mesoscale flow alone. This illustrated the fact that monsoonal winds transport moisture from the Indian Ocean and advect it inland over the country. The large-scale monsoonal flow controlled the locations and movement of the convergence/precipitating zones over the country since the deep and active convective zones shifted in the direction of the prevailing low-level large-scale monsoonal flow. Topography has a significant impact on the diurnal precipitation pattern over the country. It generates anabatic-katabatic winds with a strong diurnal cycle; it modifies the large-scale temperature, moisture, and wind flow patterns over the country; and it also inhibits a substantial amount of the moisture-laden monsoonal winds from reaching farther inland over the country. The large water bodies (Lake Victoria, Lake Turkana and the Indian Ocean) generate strong lake-land, sea-land-breeze circulations with an intense diurnal cycle that contributes to the overall precipitation pattern over the country. It was concluded that the large-scale monsoonal winds and the mesoscale circulations are both crucial components in the realization of rainfall in Kenya during a wet season. The study similarly revealed the ability of the high-resolution RAMS model to replicate realistic meteorological fields for both large-scale and mesoscale weather systems in an equatorial regime.
