Risk analysis of dams is inherently complex due to the interaction of multiple hazards, both natural and anthropogenic, which often act in combination to exacerbate failure potential. Conventional approaches typically assess hazards independently and aggregate risks, overlooking nonlinear interactions and cascading effects. In dam cascade systems, failure of an upstream dam can trigger successive breaches, leading to catastrophic flooding downstream. This proposal seeks to develop a comprehensive methodology for multi-hazard risk assessment of dams in India, integrating artificial intelligence (AI), machine learning, and fuzzy frameworks to capture uncertainty and hazard interdependencies. The methodology will be structured in three tiers: qualitative screening to classify dams by risk level, semi-quantitative analysis to identify failure modes and surveillance needs, and quantitative analysis to evaluate high-risk dams with detailed modelling. Breach simulations will be performed for selected dams and cascade systems under current and future climate scenarios, incorporating CMIP6 projections of extreme precipitation and land-use changes. The framework will also address river morphological changes, sediment transport, and debris flows, which significantly influence flood defence performance. Outputs will include risk indices, GIS-based flood zonation maps, and guidelines for mitigation measures, both structural and non-structural. The research emphasizes capacity building through workshops and training, alongside dissemination of findings via reports, publications, and case studies. By advancing science-based methodologies and integrating climate change considerations, the project aims to strengthen dam safety management, reduce vulnerability of downstream communities, and provide actionable strategies for sustainable water infrastructure resilience.
INTERNATIONAL CENTRE OF EXCELLENCE FOR DAMS
Indian Institute of Science, Bengaluru
Indian Institute of Science, Bengaluru