Adaptive flood risk management: A decision support system integrating deep learning, digital twins, and economic risk assessment.

Floods are among the most destructive climate-related disasters, with their frequency and intensity increasing due to climate change and urbanization.

As insurance withdrawal from high-risk areas grows, responsibility for risk management is shifting toward businesses and local authorities. This study, developed within the framework of the European Union Floods Directive, presents an integrated, AI-powered decision support system that combines ConvLSTM-based flood forecasting, economic vulnerability modelling, digital twin simulations, and predictive analytics.

Applied first to the coastal city of Nice (France) and then to more than one hundred public and private sites, the framework improves the accuracy of risk assessments and supports data-driven adaptation investment planning.

The findings also highlight significant governance and behavioural differences between public and private stakeholders, underscoring that effective flood resilience requires not only better data and technology, but also coordinated decision-making across institutional and economic actors.

Global Environmental Change, 95: 103069: https://doi.org/10.1016/j.gloenvcha.2025.103069