Comparative Review of Hydrological and Hydraulic Models for Urban Flood Simulation

Abstract

Urban development has significantly impacted hydrology in urban areas. With increasing impervious surfaces such as paved roads and rooftops, the rate of infiltration decreases while runoff enters storm drains much faster. Such phenomena lead to higher risks of urban flooding in compact catchments due to complex interactions between pipes, road networks, surface runoff, urban structures, and people. Hydrological and hydraulic models are among the key tools in urban flood risk assessment, drainage design, inundation mapping, and evaluating mitigation strategies. However, models differ substantially in terms of data input, numerical techniques, calibration procedure, performance evaluation metrics, and computational effort required for simulations. In this review, one-dimensional, two-dimensional, and hydrological-hydraulic coupling models are analyzed, such as SWMM, HEC-HMS, HEC-RAS, MIKE, InfoWorks ICM, TUFLOW, Delft3D FM, SWAT, and LID models. A PRISMA-like search was conducted for publications within January 2003 and May 2025. Models were evaluated based on their consistency criteria including dimensionality, numerical technique, input data, calibration procedures, performance metrics, computational complexity, software availability, and applicability for urban drainage and surface flooding modeling. According to the findings, while one-dimensional models successfully describe sewer network and hydrograph simulation, they are not efficient in modeling two-dimensional flood distribution. Coupled one- and two-dimensional models represent sewer-surface interaction more completely, but they need detailed infrastructure data and specialist expertise. The paper presents a model-selection framework and identifies research needs in uncertainty quantification, real-time data integration, machine-learning surrogates, GPU acceleration, cellular automata, stronger 1D-2D coupling, and socio-economic impact assessment.