Urban Flood Assessment and Resilience Strategies: A PRISMA-Based Systematic Review Supported by Bibliometric Analysis

Abstract

Urban flooding has intensified with rapid urban growth, expanding impervious surfaces, and more frequent extreme rainfall events linked to climate change. This study applies a PRISMA-based systematic review with supporting bibliometric analysis to evaluate recent advances in urban flood research. From 489 records retrieved from the Web of Science Core Collection, 207 peer-reviewed studies were selected after systematic screening. Hydrological and hydraulic models such as SWMM and HEC-RAS remain the dominant tools (n = 92), followed by GIS-based analysis (n = 78), remote sensing (n = 64), and machine learning approaches (n = 56). Four major research themes emerged: flood modeling and simulation, flood mapping and detection, environmental and climatic drivers, and planning and resilience strategies. Research has shifted from conventional runoff simulation toward integrated geospatial and AI-driven approaches, particularly machine learning and deep learning for flood prediction and susceptibility assessment. Land-use change, drainage limitations, and climate-driven rainfall extremes consistently appear as the main drivers of urban flooding. Research output is concentrated in China, the United Kingdom, and Europe, while rapidly urbanizing regions in Africa and South America remain poorly represented. Planning and governance studies are still limited compared to technical modeling research, creating a gap between flood science and urban policy implementation. Persistent challenges include methodological inconsistency, uncertainty assessment, limited socio-economic integration, and weak translation of modeling outputs into climate-resilient planning strategies.