Alishan Highway Slope Failure Characteristics and Failure Potential Evaluation

• Main Authors: Shih-Chieh Fang, 方世杰
• Other Authors: Der-Her Lee
• Format: Others
• Language: zh-TW
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/88398316006193187163

博士 === 國立成功大學 === 土木工程學系碩博士班 === 97 === ABSTRACT The Chi-Chi earthquake induced serious disasters of landslide and slope-land collapses in the middle Taiwan. It is very important to find out the locations of highly dangerous slopes and present a series of appropriate prevention and treatments to prevent the second disaster of slopes. This paper presents a detailed investigation of the failure highway slopes in the Alishan area to discuss the influence of typhoon and earthquake on the failure behavior. The site studies were made along the highways of Taiwan 18 (Alishan Highway), Chiayi 162A, Chiayi 122 and Chiayi 169. Results show that the number of slope failure cases decreases obviously in the next year of typhoon, but the number of cases and the volume of slope collapse increase in the next year of earthquake. It is noted that the earthquake will produce a serious secondary hazard. From the investigation in this area, the main stratums of failure slopes induced by earthquake were sandstone and gravel. However, the main stratum of failure slopes caused by typhoon was shale. The objective of this study is also to construct an assessment model to determine the failure potential of slopes. Based on the collected database of failure slopes, multivariate analysis, discriminant analysis and neural network analysis method was used for the susceptibility study. Meanwhile, the analytic results will be plotted as the susceptive figures of failure slopes, which will be compared with the real failure cases by the help of the aero-photographs. According to the results, it may be find the deviation between different assessment methods. The results prove that all these methods have the ability to construct an assessment model to predict landslides, and especially the neural network analysis, the accuracy of the prediction can reached 88.9%. Finally, results also show the main influence factors of slope stability along the highways of Taiwan 18 are the height of slope, the depth of weathered soil, stratum and formation. The third objective of this study is to demonstrate the thermography as an effective tool for fast and accurate diagnostics of rocky slope protecting structures, particularly for the emergent evaluation in a post-event investigation. This enables an engineer to monitor continuously the integrity of the rock mass and fix the problem before it reaches a point where costly failure and widespread damage occur.