A Study on Landslide Geomorphologic Characteiatic Analysis at Upstream Watershed of Tseng-Wen Reservoir After Typhoon Morakot

• Main Authors: Dong-En Lee, 李東恩
• Other Authors: 陳文福
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/39000765199876502149

碩士 === 國立中興大學 === 水土保持學系所 === 100 === Owing to steep topography and complex geological environment, geologic disasters occur frequently in Taiwan, especially during typhoon seasons. Practically, the 1999 Chi-Chi earthquake (MW=7.6) resulted in a large number of landslides at the Upstream Watershed. In recent years, climate change due to global warming will result in the increases of the frequencies and intensities of storm events; the mass wasting by rainfall also induced debris flow, new and further extensions of landslides, riverbank erosion, and massive sediment deposit in the reservoir watershed. The study area locates on upstream watershed of Tsengwen Reservoir. The study focuses on the landslides induced by the major Typhoon Morakot occurring in 2009.This study used GIS as a tool to map storm-induced landslides from SPOT5 images. Digital elevation model (DEM) was used to extract geomorphic landslide causative factors. SPOT image taken was also used to calculate an environmental factor - NDVI (normalized differential vegetation index). This study analyzes factors including elevation, slope, aspect, distance to rivers, distance to roads, relief, roughness and NDVI. The method of this study is using the Logistic Regression Method to calculate the weight of potential instability factors for landslide and combine GIS technique to build the potential landslide map at Upstream Watershed of Tsengwen Reservoir after Typhoon Morakot. Validation result shows that this model could be used for the prediction of future landslides. The result shows that the landslide area is estimated as 2.21×106 m2 and 5.15 ×106 m2 in 2008 and 2011. In addition, when compared with landslide area, about 3 times more new landslides generated near by distance to rivers. These causative factors were used to build a landslide susceptibility map via logistic regression. In landslide susceptibility map, we found the moderately-high susceptibility index close to the area near by distance to rivers, exposed area, and terrain change tremendously area. Distance to rivers, relief, roughness and NDVI are main factors controlling the landslide potential. For this research, using Logistic regression method can compare differ landslide potential factor and quantified the weight of each factor effectively, and estimating the potential value between factors with landslide. A DEM with better resolution is believed more proper for landslide susceptibility map in the future work. And still need many improvements so as to match the real slope situation after earthquake and typhoon. In the future, we have to continue conserving and managing slopeland, monitoring and counteracting slopeland related disasters to achieve the objectives of soil and water conservation and disaster mitigation.