Beyond the Single-Hazard Framework: A Proof-of-Concept Study of Multihazard Worst-Case Scenarios from Ensemble Forecasts of Tropical Cyclone Hagibis (2019)

Abstract Ensemble forecasting is a powerful tool for supporting informed decision-making in managing multihazard risks associated with tropical cyclones (TCs). Although TC ensemble forecasts are widely used in operational weather prediction systems, their potential for disaster prediction and management has not been fully exploited. Here, we propose a novel, efficient, and practical method to extract meaningful multihazard worst-case scenario (MHWCS) from a 1000-member ensemble TC forecast while preserving full physical consistency. We performed the ensemble atmospheric forecasting of TC Hagibis (2019) using the Japan Meteorological Agency’s (JMA) 5-km nonhydrostatic model. The simulated atmospheric predictions are subsequently used as inputs to JMA’s operational flood forecast model and a parametric storm surge model. These models estimate river flooding, storm surge, and wind hazard intensities in Tokyo. By incorporating forecast uncertainties, we demonstrate that Pareto-optimal ensemble members objectively represent the meaningful combinations of multiple hazards, corresponding to MHWCS. These solutions illustrate complex trade-offs among competing hazard components across various forecast locations. While some identified MHWCS pose severe risks for a single hazard type or location, others present moderately high risks across multiple hazards and locations. This diversity in potential scenarios requires risk managers to prepare multiple response strategies for both imminent risks and postdisaster management. Our findings further underscore the importance of evaluating Pareto-optimal ensemble members to assist forecasters and risk managers in understanding how combinations of TC meteorological variables—such as track, translation speed, size, intensity, and rainfall—shape worst-case scenarios. Significance Statement This work aims to maximize the benefits of ensemble tropical cyclone (TC) forecasts by introducing a novel multihazard worst-case scenario (MHWCS) framework. By examining a large ensemble forecast of river flooding, storm surge, and wind hazards due to TC Hagibis (2019) in Tokyo, this approach objectively identifies MHWCS that maximizes multihazard intensities. The study also highlights the importance of evaluating forecast meteorological conditions present in the MHWCS. The proposed approach will enable forecasters to efficiently select a subset of ensemble TC forecasts that can predict potential worst-case conditions due to multiple hazards and serve as a valuable tool for emergency responders to prepare for multihazard scenarios effectively.