降雨诱发浅层滑坡危险性的快速评估模型及应用  

Fast assessment model for rainfall-induced shallow landslide hazard and application

作　　者：郭子正 何俊 黄达 周永强[2] 朱宇航 GUO Zizheng;HE Jun;HUANG Da;ZHOU Yongqiang;ZHU Yuhang(School of Civil and Transportation Engineering,Hebei University of Technology,Tianjin 300401,China;State Key Laboratory of Geomechanics and Geotechnical Engineering,Institute of Rock and Soil Mechanics,Chinese Academy of Sciences,Wuhan,Hubei 430071,China;Faculty of Engineering,China University of Geosciences(Wuhan),Wuhan,Hubei 430074,China)

机构地区：[1]河北工业大学土木与交通学院,天津300401 [2]中国科学院武汉岩土力学研究所岩土力学与工程国家重点实验室,湖北武汉430071 [3]中国地质大学(武汉)工程学院,湖北武汉430074

出　　处：《岩石力学与工程学报》2023年第5期1188-1201,共14页Chinese Journal of Rock Mechanics and Engineering

基　　金：国家自然科学基金资助项目(41972297);河北省自然科学基金资助项目(D2022202005)。

摘　　要：现有能考虑降雨作用的确定性模型,在区域尺度进行滑坡危险性评估时存在计算成本高、无法解决岩土体参数的不确定性等弊端。针对这一问题,提出一种可用于降雨诱发浅层滑坡危险性快速评估的模型FSLAM(fast shallow landslide assessment model)。该模型能够同时考虑前期有效降雨和诱发事件降雨来计算坡体地下水位,并且使用随机参数计算破坏概率以反映滑坡危险性。均值各向同性斜坡模型的参数敏感性分析结果表明,土壤黏聚力、内摩擦角以及植被根系黏聚力是模型最为重要的输入参数。以2016年台风“鲇鱼”诱发的浙江省温州市滑坡为例,进行了FSLAM模型在区域滑坡危险性评估中的工程应用。结果显示FSLAM模型能够准确捕捉降雨对于浅层滑坡破坏概率的影响,受试工作者特征曲线表征的精度达到了76.4%。使用随机参数作为输入能有效减小岩土体参数的不确定性,并提高模型的计算精度。由于使用了简化的SCS-CN方法来计算地下水垂直流,而非复杂的Richards方程,FSLAM模型在区域尺度上的计算效率比同类型的TRIGRS模型提高了25倍,且后者精度仅为前者的69.8%。The existing deterministic models that can consider rainfall effects have disadvantages such as high computational cost and inability to solve the uncertainty of geotechnical parameters when assessing regional-scale landslide hazard.To address this problem,the fast shallow landslide assessment model(FSLAM)was proposed,which can be used for rapid assessment of rainfall-induced shallow landslide hazard.The FSLAM model can consider both antecedent effective and event rainfall conditions to calculate slope groundwater and stochastic parameters were used in the model to calculate probability of failure to reflect landslide hazard.A homogeneous slope was used to do sensitivity analysis of parameters of the model.The results showed that soil cohesion,friction angle,and vegetation root cohesion were the most important input parameters of the model.The engineering application of the FSLAM model for regional landslide hazard assessment was carried out by taking the landslides induced by typhoon Megi in Wenzhou City of Zhejiang Province in 2016 as an example.The analysis results indicated that the FSLAM model can accurately capture the effect of rainfall on the probability of failure of shallow landslides,and the accuracy of receiver operating characteristic curve reached 76.4%.The model can effectively reduce the uncertainty of geotechnical parameters and improve the accuracy by using stochastic parameters.Since the simplified SCS-CN method was used to calculate the vertical flow of groundwater instead of the complex Richards equation,the computational efficiency at the regional scale of the FSLAM model is 25 times better than the TRIGRS model,and the accuracy of the TRIGRS was only 69.8%.

关 键 词：边坡工程 降雨 浅层滑坡 确定型模型 危险性评估 随机参数 破坏概率 

分 类 号：P642[天文地球—工程地质学]