簡易檢索 / 詳目顯示

研究生: 李佳諠
Li, Jia-Xuan
論文名稱: 降雨誘發集水區崩塌潛勢與整治績效評估
Assessment of Rainfall-Induced Landslide Potential in Watershed and Mitigation Effectiveness
指導教授: 余騰鐸
Yu, Ting-To
學位類別: 碩士
Master
系所名稱: 工學院 - 自然災害減災及管理國際碩士學位學程
International Master Program on Natural Hazards Mitigation and Management
論文出版年: 2023
畢業學年度: 111
語文別: 英文
論文頁數: 78
中文關鍵詞: 衛星影像判釋邏輯斯迴歸崩塌地空間區位分佈集水區整治績效評估
外文關鍵詞: Satellite image interpretation, Logistic regression, Spatial distribution of landslides, Watershed, Effectiveness evaluation of remediation
相關次數: 點閱:101下載:40
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 台灣地質脆弱加上地形大多以陡峭的山坡為主,河流短且流速急,再因地理位置關係,常受到颱風或梅雨帶來的侵襲,帶來集中性的豪雨,促使集水區區域內災害發生事件之機率不斷提升且日趨嚴重,導致崩塌數量及規模也產生變化。因此針對集水區內因降雨所誘發的崩塌進行研究探討,同時輔以政府每年對集水區域內進行整治工程編列進行後續整治績效評估,以提供相關單位進行區域治理探討等災害預防參考。
    本研究以曾文水庫為研究區域,透過歷年衛星影像及文獻資料之蒐集,進行影像判釋分析,以利獲得歷年崩塌地及自然環境資料,同時也蒐集2017年到2019年所發生的豪雨事件資料,進行統整量化分析,並且運用邏輯斯迴歸建立集水區所發生的崩塌潛勢模式,同時,輔以接受者操作特徵曲線(ROC)判斷其準確性,從中探討集水區內致災因子與降雨誘發坡地崩塌間的關係與空間分布情形。此外,為瞭解政府2017年到2019年對集水區區域內所進行的整治工程績效評估,則是先將集水區進行子集水區劃分,接著利用常態化差異植生指標(NDVI)進行植生變遷狀況監測,來獲得長時期有經整治工程治理之下與自然復育下(未整治)之每期影像崩塌地面積,進而計算出整治前後之崩壞比來判斷整治績效。
    結果顯示,本研究以高程、坡度、坡向、地質、距水系距離、距斷層距離及雨量作為本次致災因子,並接續進行坡地崩塌模式之降雨誘發坡地崩塌潛勢評估模式建立,其利用邏輯斯迴歸建置後之總體訓練結果皆達到70%以上之正確率及ROC驗證結果有0.8以上,表示本研究之預測結果有有一定的準確性,以及可從中發現本研究三場降雨事件下集水區崩塌空間區位分佈皆在高程345~357m、坡度百分之7~8、距水系距離100~149m、256~331m及地質多是長枝坑層為主。此外,透過2017年~2019年蒐集之整治工程,進行各子集水區崩壞比比較結果,可得已整治區域與未整治區域之崩壞比比較起來各方面地質、地形及植被皆較穩定,加上經有整治工程之集水區,可承受雨量皆有明顯提升,甚至在整治過後皆無發生崩塌。

    Taiwan's geological vulnerability, coupled with its predominantly steep terrain characterized by abrupt slopes, short rivers, and swift currents, along with its geographical location, frequently subjects it to the impacts of typhoons or monsoon rains, resulting in concentrated heavy rainfall. This has led to a continuous and increasingly severe rise in the probability of disaster events within the watershed area. Consequently, the occurrence of landslides, both in terms of frequency and magnitude, has undergone variations. Therefore, conducting a study on rainfall-induced landslides within the watershed area, supplemented by the annual allocation of government-driven watershed management projects, is essential for subsequent assessment of the effectiveness of these interventions. This provides pertinent authorities with valuable insights for regional governance discussions and disaster prevention references, making it a necessary endeavor.
    This study chosen the Zengwen Reservoir as the research area. Through the collection of historical satellite imagery and literature data, image interpretation analysis is conducted to acquire data on historical landslides and the natural environment. Additionally, data on heavy rainfall events occurring from 2017 to 2019 is gathered for integrated quantitative analysis. The study employs logistic regression to establish a landslide susceptibility model for the watershed. Simultaneously, the study utilizes the Receiver Operating Characteristic (ROC) curve to assess the accuracy of the model, thereby investigating the relationship and spatial distribution between disaster-inducing factors within the watershed and rainfall-triggered slope failures. Furthermore, to understand the performance evaluation of government-led remediation projects within the watershed area from 2017 to 2019, the watershed was subdivided into sub-watersheds. Subsequently, the Normalized Difference Vegetation Index (NDVI) was employed for monitoring vegetation changes, aiming to obtain the area of landslide occurrence in each image before and after prolonged remediation efforts and natural recovery (pre-remediation). This data allowed the calculation of the pre- and post-remediation landslide ratios to assess the effectiveness of the remediation measures.
    The results indicate that this study select the elevation, slope, aspect, geology, distance to water system, distance to fault line, and rainfall as the causal factors for the disaster. Subsequently, a rainfall-induced slope landslide susceptibility assessment model was developed using logistic regression. The overall training results after the establishment of the logistic regression model achieved an accuracy of over 70% and the ROC validation results were above 0.8, demonstrating a satisfactory performance. The predictive outcomes of this study exhibit a certain level of accuracy, and it can be inferred that within the context of the three rainfall events studied, the spatial distribution of landslides within the watershed primarily occurs at elevations ranging from 345 to 357 meters, slopes of 7 to 8 percent, distances to water systems between 100 to 149 meters and 256 to 331 meters, with the predominant geological formation being the Changchih-Keng Formation. Furthermore, by examining the results of the landslide ratio comparison among various sub-watersheds based on the collected data from 2017 to 2019 regarding the remediation projects, it can be observed that the areas subjected to remediation exhibit greater stability in terms of geological, topographical, and vegetative factors compared to the untreated regions. Additionally, the watersheds that have undergone remediation have demonstrated an increased capacity to withstand rainfall, with no occurrences of landslides observed after the implementation of the remediation projects.

    摘要 I Abstract III Acknowledgement V Table of Contents VI List of Figures VIII List of Tables X Chapter 1 Introduction 1 1.1 Preface 1 1.2 Research Objectives 2 1.3 Research Process and Framework 3 Chapter 2 Literature Review 5 2.1 Relevant Studies on the Application of Change Point Analysis 5 2.2 Relevant Studies on Rainfall-Induced Slope Failures 6 2.3 Relevant Studies on Watershed Management Performance Evaluation 8 2.3.1 The causes of disasters in watersheds are as follows 8 2.3.2 The purpose of disasters in watersheds are as follows 9 2.3.3 Evaluation Methods for Watershed Management Effectiveness 11 Chapter 3 Research Area and Methodology 12 3.1 Study Area 12 3.1.1 Topography and Terrain 12 3.1.2 Geology and Rock Types 13 3.1.3 Hydrology and Climate 14 3.2 Geographic Information System (GIS) 16 3.3 Satellite Image Data Processing and Analysis 17 3.3.1 Image Fusion and Image Rectification 18 3.3.2 Change Detection and change point interpretation 24 3.3.3 Variation Detection and anomaly point interpretation 25 3.4 Logistic Regression Model Analysis 26 Chapter 4 Results and Discussion 27 4.1 Satellite Image Data Processing and Analysis Results 27 4.2 Analysis of Landslide Factors 33 4.2.1 Natural Environmental Factors 33 4.2.2 Rainfall Factors 40 4.2.3 Correlation Analysis of Landslide Factors 43 4.3 Analysis of Rainfall-Induced Landslide Susceptibility Assessment 45 4.3.1 Model Development Using Logistic Regression Analysis 45 4.3.2 Analysis of Model Performance 48 4.3.3 Investigation of the Spatial Distribution of Slope Failures in the Study Area 49 4.4 Evaluation of Landslide Remediation Effectiveness 51 4.4.1 Data Collection of the Watershed Remediation Projects in the Zengwen Reservoir Basin from 2017 to 2019 52 4.4.2 Subdivision of Sub-watersheds 53 4.4.3 Calculation of the Landslide Reduction Ratio for Each Sub-watershed 56 Chapter 5 Conclusion and Suggestion 71 5.1 Conclusion 71 5.2 Suggestion 73 References 74

    Aleotti, P. (2018). "A warning system for rainfall-induced shallow landslides based on the probability of failure on DEM." Engineering Geology, 237, 215-223.

    Crosta, G. B., Frattini, P. (2008). "Rainfall-induced landslides and debris flows." Hydrological Processes: An International Journal, 22(3), 473-477.

    Du, Y. J. (2015). "Remediation Efficiency Assessment of Small Catchment Scale - A case study in the Tseng Wen Reservoir." Master's Thesis in Department of Resources Engineering, National Cheng Kung University.

    Forkel, M., Carvalhais, N., Verbesselt, J., Mahecha, M. D., Neigh, C. S. R., Reichstein, M. (2013), "Trend Change Detection in NDVI Time Series: Effects of Inter-Annual Variability and Methodology," Remote Sensing, 5(5), 2113-2144.

    Ghaderpour, E., Vujadinovic, T. (2020), "Change Detection within Remotely Sensed Satellite Image Time Series via Spectral Analysis," Remote Sensing, 12(23), 4001.

    Guzzetti, F., Peruccacci, S., Rossi, M., & Stark, C. P. (2008). "The rainfall intensity–duration control of shallow landslides and debris flows: An update." Landslides, 5(1), 3-17.

    He, X. J., Hsueh, Y. C., Chiang, M. C., Lai, J. S. (2013). "Evaluation and Monitoring of Conservation and Management Effectiveness in the Slope Areas of Zengwen Reservoir Watershed after Typhoon Morakot." Journal of Chung Hsing Engineering, Issue 118, pp. 21-30.

    Huang, F., Chen, J., Liu, W., Huang, J., Hong, H., Chen, W. (2022). "Regional rainfall-induced landslide hazard warning based on landslide susceptibility mapping and a critical rainfall threshold." Geomorphology, Elsevier.

    Huang, T.-H. (2003). "Characteristics of Rains Triggering Debris Flows in the Watershed of Chenyoulen Stream." Doctoral dissertation, Department of Hydraulic and Ocean Engineering, National Cheng Kung University.

    Inglada, J., Vincent, A., Arias, M., Tardy, B., Morin, D., Rodes, I. (2017), "Operational High Resolution Land Cover Map Production at the Country Scale Using Satellite Image Time Series," Remote Sensing, 9(1), 95.

    ISRM (1981). "Rock Characterization, Testing and Monitoring." Pergamon Press.

    Li, X. C. (2010). "The Analysis of Potential in Heavy Rainfall-induced Landslides Affected by the Attenuation Pattern of Ground Disturbing after a Major Seismic Event." Doctoral Dissertation in Department of Resources Engineering, National Cheng Kung University.

    Lin, B. X., Hsiao, C. Y., Hsu, C. K., Ji, S. Y. (2012). "Assessment of High-Precision Airborne LiDAR Scanning Technology on the Effectiveness of Watershed Treatment in Shimen Reservoir." Journal of Chung Hsing Engineering, Issue 114, pp. 23-33.

    Liao, W. W. (2018). "Construction of Vulnerability Assessment Model for Landslide Potential Areas by Using GIS and Logistic Regression." Master's Thesis in Institute of Land Management and Development, Chang Jung Christian University.

    Maturidi, A. M. A. M., Kasim, N., Taib, K. A. (2021). "Rainfall-induced landslide thresholds development by considering different rainfall parameters: A Review." Journal of Ecological Engineering, 22(11), 211-225.

    Pardeshi, S. D., Autade, S. E., Pardeshi, S. S. (2013). "Landslide hazard assessment: recent trends and techniques." SpringerPlus.

    Qiu Haozheng (2004) "Quantitative Research and Statistical Analysis of Social and Behavioral Sciences: Analysis of SPSS Chinese Windows Version Data Analysis Paradigms." Taipei. five south.

    Southern Region Water Resources Office, Water Resources Agency, Ministry of Economic Affairs (2017). "Zengwen Reservoir Watershed Ground Patrol Management and Zengwen and Mudan Reservoir Satellite Monitoring Project in 2017." Report on the Results of the Commissioned Project by Southern Region Water Resources Office, Water Resources Agency, Ministry of Economic Affairs.

    Southern Region Water Resources Office, Water Resources Agency, Ministry of Economic Affairs (2018). "Zengwen Reservoir Watershed Ground Patrol Management and Zengwen, Agongdian and Mudan Reservoir Satellite Monitoring Project in 2018." Report on the Results of the Commissioned Project by Southern Region Water Resources Office, Water Resources Agency, Ministry of Economic Affairs.

    Southern Region Water Resources Office, Water Resources Agency, Ministry of Economic Affairs (2019). "Zengwen Reservoir Watershed Ground Patrol Management and Zengwen, Agongdian and Mudan Reservoir Satellite Monitoring Project in 2019." Report on the Results of the Commissioned Project by Southern Region Water Resources Office, Water Resources Agency, Ministry of Economic Affairs.

    Tsai, K. J., Lee, M. H., Chen, Y. J., Li, J. H. (2019). "The Spatial Trend Analysis of Landslide Disaster Characteristical Variation on Slopeland at Nanhua Reservoir Catchment in Southern Taiwan." Taiwan Mining, 71(4), 16-28.

    Viana, C. M., Girão, I., Rocha, J. (2019), "Long-term satellite image time-series for land use/land cover change detection using refined open source data in a rural region," Remote Sensing, 11(11), 1368.

    Verbesselt, J., Hyndman, R., Zeileis, A., Culvenor, D. (2010), "Phenological change detection while accounting for abrupt and gradual trends in satellite image time series," Remote Sensing of Environment, 114(12), 2970-2980.

    Wen, Z. Y. (2005). "Analysis of Landslide Models Incorporating Earthquake and Typhoon Factors." Master's Thesis in Department of Resources Engineering, National Cheng Kung University.

    Zhu, Q., Z. Haowei, D. Yulin, Xie Xiao.(2019). "A review of major potential landslide hazards analysis." Acta Geodaetica et Cartographica Sinica. 

    Web Rerferences :

    Central Geological Survey, Ministry of Economic Affairs (2023)
    https://www.moeacgs.gov.tw/

    Central Weather Bureau, Ministry of Transportation and Communications (2023)
    https://www.cwb.gov.tw/

    Esri (2023)
    https://www.esri.com/en-us/home

    Forestry Bureau, Council of Agriculture, Executive Yuan (2017~2019)
    https://www.forest.gov.tw/

    Water and Soil Conservation Regulations (2023)
    https://law.moj.gov.tw/LawClass/LawAll.aspx?PCode=M0110001

    Water Resources Agency, Ministry of Economic Affairs (2023)
    https://www.wra.gov.tw/

    Wikipedia (2023)
    https://www.wikipedia.org/

    下載圖示 校內:立即公開
    校外:立即公開
    QR CODE