A Systematic Review of Real-time Urban Flood Forecasting Model in Malaysia and Indonesia -Current Modelling and Challenge

Abstract

Several metropolitan areas in tropical Southeast Asia, mainly in Malaysia and Indonesia have lately been witnessing unprecedentedly severe flash floods owing to unexpected climate change. The fast water flooding has caused extraordinarily serious harm to urban populations and social facilities. In addition, urban Southeast Asia generally has insufficient capacity in drainage systems, complex land use patterns, and a largely susceptible population in confined urban regions. To lower the urban flood risk and strengthen the resilience of vulnerable urban populations, it has been of fundamental relevance to create real-time urban flood forecasting systems for flood disaster prevention agencies and the urban public. This review examined the state-of-the-art models of real-time forecasting systems for urban flash floods in Malaysia and Indonesia. The real-time system primarily comprises the following subsystems, i.e., rainfall forecasting, drainage system modeling, and inundation area mapping. This review described the current urban flood forecasting modeling for rainfall forecasting, physical-process-based hydraulic models for flood inundation prediction, and data-driven artificial intelligence (AI) models for the real-time forecasting system. The analysis found that urban flood forecasting modeling based on data-driven AI models is the most applied in many metropolitan locations in Malaysia and Indonesia. The analysis also evaluated the existing potential of data-driven AI models for real-time forecasting systems as well as the challenges towards it