.png){kind=icon}
.png){kind=icon}
Return
Multi-event temporal assessment of urban flood hazard and risk in Hyderabad, integrating GIS-based multi-criteria analysis, statistical linkages, and machine-learning robustness evaluation
.png){kind=icon}
PRE
AI
R
H
A
M
S
F
DOI：10.1007/s10668-026-07993-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Rapid urban growth is increasingly turning flood risk from a consequence of isolated rainfall events into a reflection of long-term structural changes in city landscapes. This study assesses urban flood hazard and risk in Hyderabad, India, through a multi-event analysis covering five major floods between 2000 and 2024. A GIS-based Analytical Hierarchy Process (AHP) framework was employed to map flood hazard by combining rainfall, land use and land cover (LULC), topography, drainage, and hydrological data. To test the framework’s robustness, models such as Random Forest (RF), Extreme Gradient Boosting (XGB), and Support Vector Machine (SVM) were developed using the same factors and served as benchmarks. The machine-learning (ML) results closely matched the AHP hazard classifications, confirming the framework’s reliability. Validation using around 500 citizen-reported flood observations showed a strong spatial match with modeled flood-prone areas, bolstering confidence in the findings. By combining flood hazard data with population exposure and proximity to critical infrastructure, the study identifies zones where physical hazard and socio-economic vulnerability intersect. Results show that while extreme rainfall remains the trigger of floods, urban expansion, loss of waterbodies, increased impervious surfaces, and drainage issues are the key structural drivers of flood risk. High-risk zones have significantly grown across the city, especially in rapidly urbanizing regions. The findings underscore the importance of risk-based land-use planning, protecting and restoring waterbodies, improving drainage infrastructure, and making equitable investments to build long-term urban resilience. The framework is designed to be scalable and adaptable to other rapidly growing cities facing similar flood risks.
Keywords:
Flood hazard mapping
Land use transformations
Machine learning validation interpretability
Socio-economic vulnerability
Sustainable cities
Citizen science validation
Journal
IF:
4.2
Papers:
945
Citations:
2.3W
Organization
Cited Papers