본 연구에서는 우리나라에 태풍이 내습할 때 먼저 피해를 받는 남해안 지역을 대표하는 Quantile을 제안하고, 각 지점들의 재 현기간에 따른 극치 풍속을 추정하기 위하여 연 최대풍속 자료와 Hosking이 제안한 선형-모멘트 방법(L-moments)을 이용한 지역빈도 해석을 수행하였다. 모든 기상관측 지점에서는 비정상적인 값이 존재하지 않았고 이질성 검정을 통해서 하나의 동질 한 지역을 나타 낼 수 있음을 확인하였다. 또한 적합도 과정을 통해서 Generalized Normal (GNO) 및 Generalized Extreme Value(GEV) 분포를 남해안 지역을 대표하는 빈도분포로 선택하였다. 상대 오차(RB)와 상대 평균제곱근 오차(RRMSE)를 이용하여 두 분포의 안정성을 평가한 결 과, GNO 분포가 GEV 분포보다 더 안정한 것을 알 수 있었다. 마지막으로 남해안 지역을 대표하는 Quantile과 각 지점들의 평균, 중앙 값, 그리고 위치 매개변수를 이용하여 지점들의 극치 풍속을 추정하였다. 본 연구에서 적용한 지역빈도해석이 자료가 부족하거나 계측 되지 않은 지점들에 대한 극치 풍속을 추정하기 위한 방법으로서 도입이 필요하다고 생각된다.

For this study, WRF numerical modeling was performed, using RDAPS information for input data on typhoons affecting the Korean peninsula to produce wind data of 700hPa. RAM numerical modeling was also used to calculate 3-second gusts as the extreme wind speed. After comparing wind speeds at an altitude of 10 m to evaluate the feasibility of WRF numerical modeling, modeled values were found to be similar with measured ones, reflecting change tendencies well. Therefore, the WRF numerical modeling results were verified. As a result of comparing and analyzing these wind speeds, as calculated through RAM numerical modeling, to evaluate applicability for disaster preparedness, change tendencies were observed to be similar between modeled and measured values. In particular, modeled values were slightly higher than measured ones, indicating applicability for the prevention of possible damage due to gales. Our analysis of 3-second gusts during the study period showed a high distribution of 3-second gusts in the southeast region of the Korean peninsula from 2002-2006. The frequency of 3-second gusts increased in the central north region of Korea as time progressed. Our analysis on the characteristics of 3-second gusts during years characterized by El Niño or La Nina showed greater strength during hurricanes that affected the Korean peninsula in El Niño years.

There were 35 typhoons affecting Korean Peninsula from 1999 to 2009(The average annual number of typhoon is 3.18). Among these typhoons, the number of typhoon passing through the Yellow sea, the Southern sea and the East sea were 14, 6 and 15 respectively. Wind speed on the height of 10 m can be finally estimated using the surface roughness after we calculate wind speed on the height of 300 m from the data on the surface of 700 hPa. From the wind speeds on the height of 10 m, we can understand the regional distributions of strong wind speed are very different according to the typhoon tracks. Wind speed range showing the highest frequency is 10~20 m/s(45.69%), below 10 m/s(30.72%) and 20~30 m/s(17.31%) in high order. From the analysis of the wind speed on the hight of 80 m, we can know the number of occurrence of wind speed between 50 and 60 m/s that can affect wind power generation are 104(0.57%) and those of between 60 and 70 m/s that can be considered as extreme wind speed are even 8(0.04%).