Mapping the Scientific Evolution of Flood Risk Modelling with GIS: A Bibliometric and Thematic Analysis (1987–2024)

Abstract

Flood risk modeling using Geographic Information Systems (GIS) has gained increasing global significance amid intensifying climate change and extreme weather events. This study presents a comprehensive bibliometric and thematic analysis of GIS-based flood risk modeling research published between 1987 and 2024. A total of 3,946 records were retrieved from the Scopus database and analyzed using the Bibliometrix R package to examine publication trends, authorship patterns, institutional collaboration, and thematic evolution.

Results indicate a 15.08% annual growth rate in scientific output, with a pronounced surge after 2010 driven by advances in remote sensing technologies, global policy frameworks such as the Sendai Framework, and the rise of climate adaptation initiatives. China and India dominate in publication volume, whereas Italy and Malaysia show higher citation impacts, reflecting methodological depth and innovation. Thematic evolution analysis reveals a shift from traditional floodplain mapping to emerging themes such as AI-driven flood forecasting, real-time hydrological monitoring, and resilience assessment. Despite this progress, significant gaps persist in socio-economic integration, uncertainty quantification, and model interpretability. Based on these identified gaps, this study highlights the potential of integrating Machine Learning (ML), explainable AI techniques (e.g., SHAP), and Conformal Prediction (CP) as emerging technologies to enhance the predictive power, transparency, and confidence estimation of GIS-based flood risk models.