全球变化知识资源中心

Stochastic method in extreme meteorology such as typhoon involves the use of statistical models to represent the climatological characteristics of the typhoon and physical models consisting mainly of genesis, intensity, tracks and wind field. This paper presents an MCS (Monte Carlo Simulation) based on models for genesis, intensity, tracks and wind field to estimate the extreme wind speed of future typhoons. This proposed method calculates GPIs (Genesis Potential Index) using ERA-interim data to consider the effects of climatological factors on the genesis of typhoons which are modelled based on the correlations between GPIs and past typhoon genesis locations. The intensity of typhoons is modelled using SST (Sea Surface Temperature) and OOR (Oceanic Occupation Ratio). The typhoon tracks are determined through an Auto Regression method. The wind field is estimated using a modified Batts' model, and Global Circulation Models (GCM) is adopted to estimate future extreme wind speeds. In order to reduce the inherent uncertainty, the reliability ensemble averaging is applied. The comparison of the wind speed estimated for Korean peninsula with those of other researchers reveals that the proposed method can more reasonably predict the maximum wind speed used to determine structural design speed considering regional and climatic factors.