Quantifying uncertainties in tropical cyclone wind hazard assessment due to synthetic track stochastic variability for Southeast Asia

Abstract

Extreme tropical cyclones (TCs) present serious threat to Southeast Asian communities, causing widespread loss of lives and property damages. This study investigates the uncertainties in probabilistic TC wind hazard assessment due to stochastic variability in the underlying synthetic TC track ensembles. We focus on landfall TCs in the Philippines and Vietnam using multiple realisations from a single synthetic TC generation model. It is found that the maximum 3-s gusts in the Philippines are 233.3 km/h and 301.6 km/h under 100-year and 500-year return periods (RP). The coefficient of variance (CV) for the ensemble means falls within 0.05–0.20 over the main landmass. In Vietnam, the maximum gusts under 100-year and 500-year RPs are 207.6 km/h and 286.1 km/h, respectively, with a CV range of 0.01–0.13. Large uncertainties mainly concentrate in high TC activity zones where the peak gusts from individual synthetic event set can be up to 20% higher than the ensemble outputs. Low TC activity zones are also prone to high ensemble spread in model outputs under short RPs, although these uncertainties tend to decrease as RP increases. The probabilistic hazard model is shown to be most sensitive to TC landfall frequency and maximum intensity. Influence from other track parameters, such as TC size and forward speed, can become important for long exposure of TC trajectories. Such insights enable better understanding on the key drivers of sensitivities in the probabilistic hazard maps. Our study complements the conventional hazard assessment using synthetic TC tracks by explicitly assessing the uncertainties from the underlying stochastic event catalogue. The calculated CVs arising from synthetic track stochastic variability can be compared and/or combined with uncertainty measures on the underlying TC hazard model. This ensures proper quantification and communication of full model uncertainties for more rigorous TC risk assessment and in subsequent risk mitigation.