Impact of the improved urban fraction on rainfall simulation using the WRF model over Kampala, Uganda

<p>Urbanization affects the initiation and intensification of convective activities by changing local meteorological variables, which alters the atmosphere's convective processes. Therefore, proper urban surface information is required to model the energy partitioning pattern and its contrast with neighboring grid cells. In this study, the mesoscale weather research and forecasting (WRF) model is configured with satellite-derived urban fraction for optimal rainfall simulation and to evaluate its impact on the simulated rainfall over Kampala, Uganda. The WRF urban parameter values associated with the considered urban fraction are adjusted based on the literature reviews. The satellite-derived urban fraction represents the more realistic extent and intensity of the urban class with a more representative urban fraction. Three different simulations are performed to distil the impact of changing urban fractions as well as of adjusting urban parameters: (1) DUF_DUP, which uses the default urban fraction and default urban parameter values, (2) DUF_AUP, which uses the default urban fraction with adjusted urban parameter values, and (3) SUF_AUP, which uses the satellite-derived urban fraction and adjusted urban parameter values. A single extreme rainfall event, which caused a flood hazard in Kampala on 25 June 2012, is used for all three simulations. The simulated peak rainfall and its spatial distribution over the Kampala catchment are evaluated using observed rainfall data from gauging stations and satellite data from Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS). Results indicate that the simulated rainfall is overestimated compared to CHIRPS and underestimated when comparing gridcell values with gauging station records. However, the SUF_AUP simulation shows relatively better results with a lower absolute relative error score compared to the other two simulations. Compared with the default urban fraction, the satellite-derived urban fraction represents the more realistic urban extent and intensity. As a result, SUF_AUP results in a more realistic rainfall simulation compared to when using the default urban fraction. Rainfall analysis for both 24-hour and 2-hour indicates that the presence of an urban landscape alters both the structure and propagation of high-intensity rainfall over the city, mainly due to the impact of the urban landscape on the different meteorological variables leading to modifying mechanisms associated with rainfall.</p><p><em>Keywords: rainfall, Default urban fraction, Kampala, urban parameter, Updated urban fraction, and WRF model</em></p>