Laboratory of Forest Ecosystem Management, Division of Forest Sciences, Department of Agro–Environmental Sciences, Faculty of Agriculture, Kyushu University
九州大学大学院農学研究院環境農学部門森林環境学
This study aimed to elucidate the differences in the tendency for landslides to occur according to heavy rain types in summer in South Korea. For this, the rainfall intensity–duration (I–D) relationship and the differences in I–D relationship by three representative heavy rain types (convective rainstorm, monsoon rain and typhoon rain) were analyzed statistically, based on the corresponding hourly rainfall data for 478 shallow landslides occurred through South Korea during 1963–2012. In the quantile regression analysis of the I–D relationship for triggering landslides, the regression line of the 50th percentile of each heavy rain type showed notable differences, and convective rainstorms showed the highest rainfall intensity for the same duration, followed by monsoon and typhoon rains. While the difference was remarkable in the rainfall duration range of 4–20 h, it tended to decrease with an increase in the duration, suggesting that the effects of rainfall intensity on landslide occurrence gradually became insignificant due to an increase in soil water content. Rainfall intensity and duration according to heavy rain type revealed a statistically significant difference. Rainfall intensity was the highest in convective rainstorms and the lowest in typhoon rains, whereas rainfall duration was the longest in typhoon rains and the shortest in convective rainstorms. Such a trend is consistent with the characteristics of heavy rain types reported in Korean meteorology studies, clearly indicating that landslides occurring in South Korea were mainly triggered by heavy rains in summer. The results of this study should contribute for establishing landslide warning and evacuation systems when considering the characteristics of heavy rain in the future and suggest the necessity to improve rainfall and landslide prediction techniques during localized convective rainstorms and to provide systematic and proactive warning and evacuation systems.