dc.description.abstract |
The development of more reliable and accurate flood-prone zone maps is a flood management strategy that depends on historical rainfall data. In Lilongwe City, Malawi, severe river floods during 2018 and 2019 inflicted significant damage on urban residential areas. Several studies have demonstrated that climate change coupled with anthropogenic activities such as urbanisation has led to an increase in rainfall runoff, which causes these floods. To support flood management in Lilongwe City, this study aimed at employing hydrological and hydraulic modelling combined with GIS to develop flood-prone zone maps based on varying extreme rainfall return periods in consideration with current rainfall patterns. HEC-HMS and HEC-RAS models were utilized, with satellite-based CHIRPS precipitation data for rainfall frequency analysis addressing observed rainfall data scarcity in the Lilongwe River’s watershed. Observed and satellite rainfall data were bias-corrected and used to construct intensity-duration-frequency (IDF) curves, essential for flood modelling. Using the Gumbel distribution, probable maximum rainfall intensities for 5-, 10-, 20-, 50-, and 100-year return periods were produced and used in estimating peak discharges of 1434.7 m³/s, 1800.7 m³/s, 2117.3 m³/s, 2574.4 m³/s, and 2949.5 m³/s, respectively. The model validation for streamflow achieved an NSE of 0.71 and an RMSE of 0.69. The findings of this study indicate that flood models and GIS effectively determined the production of flood extent maps for different return periods based on satellite historical precipitation data. Therefore, planning authorities can leverage GIS and remote sensing with modelling techniques to update IDF curves, which can enhance the accuracy of extreme rainfall predictions that may trigger flooding, particularly in urban areas like Lilongwe. |
en_US |