Probabilistic assessment of concurrent tornado and storm‐related flash flood events

Abstract

The concurrent occurrence of extreme events can significantly impact societies and infrastructure systems, especially when the instructions provided to the exposed communities demand conflicting responses. In this study, a probabilistic assessment of concurrent tornado and storm-related flash flood (TORFF) events is performed across southern Canada. We quantify the interdependencies between tornadoes and flash floods (extreme precipitation as proxy) using ground-based and reanalysis datasets. Windspeed values corresponding to tornado events, categorized based on the recorded Fujita rating, are derived through a resampling approach. The TORFF events are clustered and the corresponding bivariate probability distributions of the resampled windspeed and associated precipitation are characterized based on Copula framework. The individual and joint return periods of concurrent tornadoes and flash floods are then assessed under the AND (when both variables exceed predefined thresholds), OR (when either one of the two variables exceeds predefined thresholds), and conditional hazard scenarios across Canada. Results show positive strong dependencies between resampled windspeed and associated precipitation in Saskatchewan, and weaker dependencies followed by Alberta, Manitoba, Ontario, and Quebec. The Saskatchewan region shows the highest risk underestimation considering the independence scenario compared to the other regions. Higher precipitation is also expected during extreme windspeed, as observed in the conditional assessment of precipitation given windspeed. This study provides insights for more robust recurrence interval estimation for tornadoes and flash floods to aid in emergency planning of evacuation decision-making process.