Earth observation and flood modelling to support flood hazard mitigation in Somalia

In recent decades, the planet has witnessed a significant increase in hydrometeorological extreme events, such as floods and droughts, causing devastating impacts on populations and infrastructure. This thesis focuses on the Juba River in Somalia, a region particularly vulnerable to flooding due to its climatic conditions and lack of adequate hydrological data. The main objective of this research is to develop a reliable hydraulic model of the Juba River using HEC-RAS software, a tool widely used for hydraulic modeling of river systems. The research methodology involves the analysis of observed data, the use of alternative models such as GEOGloWS and HBV to simulate flows, and the application of bias correction techniques to improve the accuracy of the simulations. The results obtained allow the identification of the most vulnerable areas in case of flooding, providing a valuable tool for risk assessment and the planning of mitigation measures. In addition, the study explores alternative models to HEC-RAS, such as SynxFlow and FastFlood, to assess their potential in reducing computational costs while maintaining an acceptable level of accuracy.