Integrated flood modelling and risk assessment in urban areas: A review on applications, strengths, limitations and future research directions

Global scale. The purpose of this study is to provide a comprehensive global assessment of urban flood modelling by: (i) critically reviewing the most widely used flood models in urban settings; (ii) synthesizing their operational mechanisms, including the integration of diverse data types and validation techniques; and (iii) evaluating each model's strengths and limitations in simulating flood dynamics and assessing urban flood susceptibility. Furthermore, the paper establishes a framework for selecting acceptable modelling methodologies for successful flood risk management in real-world urban scenarios. Hydraulic-hydrological models, and cloud-based geospatial platforms have been widely applied in flood modelling and risk and vulnerability assessment. Despite these advancements, accurate flood modelling remains a challenge due to limitations in input data quality. Among earth observation tools, radar satellite data was identified as the most effective due to its reliability under cloudy and rainy conditions. Enhancing model accuracy and validation remains possible through the integration of both optical and radar data with hydraulic and hydrological models. For example, radar backscatter intensity can be used to estimate flood depths. However, key research gaps remain, notably, the integration of high-resolution climate projections and socio-economic factors into flood risk models, and the application of modelling tools in poorly planned urban areas to assess real-time changes in land use following flood events. • Rising urban flood risks driven by climate change and urbanization. • Hydraulic models boost accuracy in flood prediction and control. • Remote sensing and GIS enhance flood risk assessments. • Machine learning and cloud computing optimize flood forecasting. • There is a need to incorporate socio-economic factors into future flood models