Joint Probabilistic Wind–Rainfall Model for Tropical Cyclone Hazard Characterization

Abstract

Often, rainfall-induced flooding has been the greatest factor influencing property loss and loss of life in major tropical cyclone events. However, a lack of an extensive historical rainfall record has long served as an obstacle to furthering the understanding of tropical cyclone–related rainfall. Advances have been made in the last few decades in rainfall observational techniques, allowing for the development of robust statistical tropical cyclone rainfall models. However, these current models are unable to capture statistical variability in rainfall intensities (i.e., only mean rainfall rates can be predicted) or they cannot simulate rainfall over land. A new tropical cyclone rainfall model is developed herein using satellite rainfall observations, in which the tropical cyclone rainfall is modeled through a Weibull distribution conditioned upon maximum surface wind speed and sea surface temperature (SST). Satellites are able to capture rainfall observations in the most extreme events, unlike surface rain gauges, which tend to fail under high winds, thereby allowing for the consideration of the rainfall climatology of stronger tropical cyclone events. The rainfall model developed herein was then coupled with state-of-the-art probabilistic tropical cyclone models and future projected climate change scenarios, allowing for the concomitant characterization of tropical cyclone hazards (wind speed, rain rate, and storm size) and the assessment of possible changes in the tropical cyclone hazards in a future climate state. In results obtained from tropical cyclone simulations performed herein under a postulated future climate scenario, in which no technology or policies have been implemented to reduce greenhouse gas emissions, tropical cyclones are projected to intensify. The intensification seen in the simulation results is accompanied by a reduction in size, meaning the magnitude of the extreme wind and rainfall hazards are projected to increase, whereas the affected area is projected to decrease. Additionally, this simulated intensification implies events that are considered extreme in the current climate will occur more frequently in the future climate scenario, paving the way for possible exceedance of critical design values.