HR: 0830h
AN: H51E-1137    [PDF]
TI: Quantitative Prediction Models of Landslide Dependent on Lithology
AU: Kim, W
EM: kwy@kigam.re.kr
AF: Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, Daejeon, 305-350 Korea, Republic of
AU: * Chae, B
EM: bgchae@kigam.re.kr
AF: Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, Daejeon, 305-350 Korea, Republic of
AU: Cho, Y
EM: choych@kigam.re.kr
AF: Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, Daejeon, 305-350 Korea, Republic of
AU: Lee, C
EM: colee@kigam.re.kr
AF: Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, Daejeon, 305-350 Korea, Republic of
AU: Kim, K
EM: kks@kigam.re.kr
AF: Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, Daejeon, 305-350 Korea, Republic of
AU: Choi, Y
EM: yschoi@kigam.re.kr
AF: Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, Daejeon, 305-350 Korea, Republic of
AB: The most common type of landslide in Korea is debris flow taken place on natural terrain covered by thin unconsolidated weathered rock or soil. The landslides are initiated by regolith failure at shallow depth. KIGAM research team has performed a research project to develop a quantitative prediction method of landslide. Detailed survey of landslides was conducted for 198 landslides among 1,600 landslides in a region of metamorphic rocks and granites, and 1,500 landslides in sedimentary rocks and volcanic rocks regions. The landslides were investigated at the scale of both 1:50,000 and 1:5,000 to obtain landslide triggering factors. The items investigated for each landslide are lithology, weathering degree, discontinuity, type of landslide, slope geometry, cross section of landslide, triggering position, length, width, depth and volume. Fourteen physical and mechanical properties of soil samples were tested for 139 specimens of undisturbed samples. In order to select landslide triggering factors, the investigated and tested items were analyzed using the logistic regression method. Seven factors from metamorphic rocks and granites were chosen as the triggering factors of landslides. Five factors were also chosen as the landslide inducing factors from sedimentary rocks and volcanic rocks. The logistic regression method produced two models of logit equations that can calculate absolute landslide probability. Because the computation results of logit equations present as type of percentage, it is easy for the common people to understand the meaning of landslide probability. Verification of the landslide prediction models showed more than 90% of prediction accuracy. Finally eight sheets of quantitative landslide probability map were produced based on the landslide prediction models at the scale of 1:5,000.
DE: 1625 Geomorphology and weathering (1824, 1886)
SC: Hydrology [H]
MN: 2003 Fall Meeting