A multi-hazard cyclone risk framework to enable resilience-informed decision

Abstract

Tropical cyclones (TCs) represent one of the most destructive natural hazards, generating extreme winds, storm surge, waves, and coastal flooding that threaten major urban centers across the eastern United States and Canada. These hazards are projected to intensify under climate change as rising sea levels, increasing atmospheric moisture, and shifting storm tracks alter the frequency and severity of these cyclonic events. Traditionally, high-fidelity numerical models have been used to simulate TC-induced hazards such as storm surge and waves; however, their substantial computational demands limit their applicability for real-time predictions or probabilistic risk assessments. To address this challenge, this thesis develops novel hybrid data-driven models capable of efficiently simulating TC-induced hazards, including wind, storm surge and waves. These hazard models are then applied to perform a probabilistic risk assessment for both current and future climates. Specifically, large ensembles of synthetic tracks are generated to capture a broad range of storm trajectories, incorporating projected changes in storm characteristics and sea-level rise. These synthetic storms are then coupled with the hazard modules to perform probabilistic hazard analyses for both current and future climate scenarios. The resulting probabilistic hazard simulations are then combined with exposure and vulnerability modules to quantify TCs-induced coastal risk. Finally, the resulting risk products are incorporated into a deep reinforcement learning (DRL)-based optimization framework to identify adaptive and cost-effective protection strategies under changing climate conditions. The resulting findings and modeling tools enable effective capabilities for emergency preparedness, coastal resilience planning, and evidence-based decision-making to enhance long-term risk management across vulnerable Canadian coastal regions.

Date	7 Apr 2026
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Reda Snaiki (Supervisor)