재분석 바람장을 이용한 한반도 근해 극치 해상풍 추정

Estimation of Extreme Sea Surface Wind Around the Korean Peninsula Using Reanalysis Data

A long-term sea surface wind is a basic data for offshore construction and design but collecting accurate long-term observed wind data in the seas adjacent to the Korean peninsula (KPsea) is very difficult. These days, as an alternative to observed winds, empirical typhoon vortex model winds and model reanalysis wind data have been widely used in engineering, but these data have limitations in ensuring accuracy in both typhoon and non-typhoon periods. To overcome this problem, this study evaluates the performance of Global Forecast System(GFS) reanalysis wind data, which is known to improve the typhoon wind performance using typhoon relocation and Bogussing technique, and estimates extreme winds for a given return period in the KPsea. The analysis reveals that the GFS winds match well with data from 17 buoys observed in the KPsea from 2009 to 2018. In particular, the evaluation for 35 typhoons that affected Korea shows that GFS winds can simulate realistic typhoon winds in the KPsea and thus can be applied to extreme wind analysis. This study adapts an extreme analysis method based on the general Pareto distribution(GPD) and Peak-Over-Threshold(POT), which are more suitable to short period(10 years) data. The extreme wind speeds for the 50-year return period in the KPsea, estimated using this method, shows that the spatial distribution of extreme winds was similar to that of maximum wind speed induced by typhoons. However, the high occurrence of strong winds in winter suggests that winter winds also play an important role in estimating extreme winds in the KPsea and there is a limit to the method of analyzing extreme winds based only on the empirical typhoon vortex model winds.