Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/33452
標題: Establish Assessment Model for the Landslide Potential Caused by Rainfall Event
降雨引發邊坡崩塌潛勢評估模式之建構
作者: Wu, Chun-Hung
吳俊鋐
關鍵字: landslide potential model
崩塌潛勢評估模式
landslide threshold
崩塌門檻值
出版社: 水土保持學系
摘要: The study focuses on establishing the assessment model for the landslide potential caused by rainfall events. The study suggests to assess the initiating causes and the inherent causes individually during the landslide potential assessment. During the research process, the study assesses the landslide potential values without considering the rainfall and earthquake impacts, and combines the landslide potential values with the rainfall impact as the landslide potential value under rainfall events. The study selects eight assessment factors as the landslide potential basis. The weight of the assessment factors is obtained by using Analytic Hierarchy Process (AHP) and the assessment standard is based on the landslide cases in serious disaster region, includes Nantou County, Wanrung Township in Hualien County, Heping Township in Taichung County, and Alisan Township in Chiayi County, caused by the Typhoon Toraij in 2001. The study use the rainfall quantity to quantify the influence of the rainfall on landslide. Based on the 175 landslide cases from the 7th, 8th, 14th,and 20th Province Highway, the minimum threshold value to induce the landslide under rainfall event is 8.93. The mean value and the standard deviation of the 175 landslide cases assessment value are 12.97 and 1.77 respectively. According to the the geology, terrain, rainfall characteristic of the landside cases, the assessment value of the landslide cases is zoned for different landslide types. The landslide cases from Mingder reservoir watershed and serious disaster region during Typhoon toraji and Shihmen reservoir watershed during Typhoon Aere are the verification cases. The study use kriging method to simulate the rainfall distribution during rainfall event. According the minimum threshold value to induce the landslide under rainfall event, the correctness of the landslide potential model on serious disaster region exceeds 80%. But the correctness of the landslide potential model on Shihmen reservoir watershed is approximate 60%. According to the 10058 landslide cases during Typhoon Toraji and Typhoon Aere, 75.2% of the landslide potential values centralizes in 8.93 to 12.
本研究主旨在建構降雨情況之崩塌潛勢評估模式,本模式在建構過程中將降雨誘發因子與崩塌相關之環境潛在因子分離,並建議採用兩段式評估法進行建構,先評估未降雨未地震自然情況下之邊坡崩塌潛勢,再將自然情況崩塌評估值與降雨事件對邊坡崩塌之影響結合,方為降雨情況下之崩塌潛勢評估值。 於模式建構過程中,本研究將崩塌現象分為六評估面八細部評估因子進行評估,並以多評準決策層級分析法評定各評估面之權重,以2001年桃芝颱風於南投縣、台中縣和平鄉、花蓮縣萬榮鄉及嘉義縣阿里山鄉之崩塌案例為依據,給定細部評估因子之評分方式,至此建立自然情況崩塌潛勢評估模式。 在降雨對邊坡影響之量化上,本研究考量降雨事件代表性及實用性,以臨前降雨量及降雨事件降雨累積量為主要考量,並以北橫公路、中橫公路、台14線及南橫公路等共175處邊坡崩塌案例進行模式建構,模式估算結果發現降雨情況下引發邊坡崩塌之最小崩塌門檻值為8.93,平均崩塌門檻值為12.97,標準偏差值為1.77,本研究並以過往崩塌案例分佈特性區分為崩塌頻發區、崩塌集中區及邊坡暫態穩定區而新竹縣五峰鄉桃山村土場部落及民都有部落於艾利颱風所引發之崩塌案例,以本研究之降雨情況崩塌模式評估值亦落於崩塌集中區之內。 在案例驗證上,本研究採用克力金法模擬研究區於桃芝颱風時之降雨分佈, 並在研究區中崩塌率差異大之村落及土石流多之村落進行驗證,桃芝颱風事件於南投縣及台中縣挑選八個村落共1103處崩塌案例,艾利颱風於石門水庫集水區挑選五個村落共909處崩塌案例,經以本研究所建構之模式評定降雨情況崩塌潛勢評估值分佈,若以降雨情況最小崩塌門檻值8.93為界,則本研究對南投縣及台中縣八個村落之預測準確率約可達80%以上,而在石門水庫集水區之預測準確率則在60%左右,預測準確率下降之原因在於本研究對石門水庫集水區之坡腳掏刷崩塌現象預測率不足。 相較於過往崩塌潛勢評估模式之建構,本模式對降雨誘發因子對邊坡崩塌之影響深入探討,本研究認為過往崩塌潛勢評估模式未能在崩塌時間點上突破,主因在於誘發因子在崩塌潛勢評估模式中之角色混淆所導致,本研究採用兩段式評估法及點對點評估方式嘗試獲得降雨情況邊坡可能發生崩塌現象之最小門檻值及可能發生之時間點。
URI: http://hdl.handle.net/11455/33452
Appears in Collections:水土保持學系

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.