近年來隨著氣候的變遷,極端氣候出現頻率升高,導致複合型災害發生頻率大大增加。2009年莫拉克颱風侵襲台灣,造成台灣中南部地區災情慘重,如小林村遭受滅村,而陳有蘭溪流域內新山村、神木村、東埔村及頭坑溪,由於莫拉克挾帶豪大降雨量,導致山崩、土石流及洪水侵蝕等嚴重災害發生。 本研究旨在建立陳有蘭溪流域河階地危險性評估,以陳有蘭溪流域內40個具有保全對象之河階地進行分析。利用地理資訊系統(Geographic Information System;GIS),將相關圖層進行套疊,萃取出河階地之攻擊岸長度、距斷層距離、距河道距離、土石流潛勢溪流數量、最小比高、平均坡度、地質、侵蝕溝數量及距後方崩塌地距離等9項參數。藉由層級分析法(Analytic Hierarchy Process;AHP)建立各因子關係階層,經由調查結果:河階地前方潛感因子條件考量(0.352)>河階地本身潛感因子條件考量(0.342)>河階地後方潛感因子條件考量(0.306);保全住戶因子條件考量(0.599)>交通因子條件考量(0.292)>農田因子條件考量(0.109);其顯示專家學者較偏向於重視河階地前方潛感因子條件考量及保全住戶因子條件考量。根據層級分析(AHP)調查結果之權重,進行河階地各因子配分,再將潛感各因子及保全因子分數加總進行河階地危險評估,本研究將危險度分為高危險、中高危險、中危險及低危險4類,經分析結果,高危險河階地有8個、中高危險河階地有14個、低危險河階地有14個及低危險河階地有4個。 另外,探討河階地經由降雨入滲條件下整體地下水滲流行為,以及邊坡滑動潛勢。本文以GEO-SLOPE為分析工具,並以頭坑溪階地I為例,根據莫拉克颱風時雨量進行一般克利金分析及雨場分割設計,以Seep/W滲流分析模式,進行降雨入滲之分析,再以Slope/W邊坡穩定分析模式計算出安全係數。此外,本研究以暴雨頻率年所分析結果,將重現期距雨量帶入模式中,以進而探討河階地在不同降雨條件下的穩定性情形。
In recent years, along with the climate changes, there is an increasing frequency on the extreme weather, and the rate of composite disasters is greatly increased as well. For example, in 2009, when Typhoon Morakot intruded Taiwan, there were severe disasters in both central and southern parts of Taiwan, including incidents as the whole village of Hsiaolin was wiped out, and occurrences of severe disasters, such as landslides, debris flows and flooding erosions due of the extreme heavy rainfalls by Typhoon Morakot to the areas within Chenyoulan River, namely Hsinshan Village, Sheng Mu Village, Dongpu Village and Toukeng Creek. The purpose of this study is to establish hazard assessment for the river terraces along Chenyoulan River, and to make analysis of 40 river terraces with are secured objects within Chenyoulan River. Using Geographic Information System (GIS), to make overlapping on the related graphic layers, and to make extract of 9 parameters for river terraces-length to attack shore, distance away from fault, distance from river channel, number of creeks and streams with possibility of debris flows, height above stream level, average slope degree, geology, number of erosion ditch, and distance away from landslide area behind. Through Analytical Hierarchy Process (AHP), to establish relationship levels for the various factors, and with the finding results as: river terraces front potential factor condition consideration (.352)>river terraces its own potential factor condition consideration(.342)> river terraces back potential factor condition consideration(.306); secured resident factor condition consideration (.599)>traffic factor condition consideration(.292)> farming land factor condition consideration(0.109); this indicates that the expert scholars put more emphasis on the river terraces front potential conditions as well as the secured resident factor condition considerations. According to the weight found by AHP analysis results, to make grading for the various factors of the river terraces, to then work on the hazard assessment for the river terraces with totaling of both potential trend of the various factors as well as the secured factors, moreover, the research divides its critical levels into high, medium-to-high, medium, and low, furthermore, by the analytical results, there are 8 high level river terraces, 14 medium-to-high level river terraces;14 medium level river terraces , and 4 low level river terraces. In addition, to make investigation of the overall ground water seepage behaviors through the conditions of rainfall infiltration, as well as what the propensity on the movement of side slopes are. In the study, using GEO-SLOPE numerical model as analytical tool, taking the example of Toukeng Creek Terrace I, according to the rainfalls condition during the Typhoon Morakot, to make general analysis by Kriging Method as well as rainfall threshold partition design, moreover, using Seep/W infiltration analysis model, to work on the infiltration analysis on the rainfalls during Typhoon Morakot, and to make calculation of safety coefficient by Slope/W or slope stability analysis model. Besides this, with the analytical results on the various frequencies frequency of heavy rainfall, to once again bring different duration of rainfall into the model, in order to further investigate the stability under different rainfall conditions for river terraces.