基于物理过程模型的降雨型浅层滑坡易发性研究

项目名称： 基于物理过程模型的降雨型浅层滑坡易发性研究

项目编号： No.51208461

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 建筑环境与结构工程学科

项目作者： 赵宇

作者单位： 浙江大学

项目金额： 25万元

中文摘要： 我国东南部地区松散沉积土边坡分布最广，与人类活动相关度最大，每年来袭的台风暴雨都会引起大量的浅层滑坡。为了更加合理的利用人力、物力、财力资源，提高滑坡治理的效率，应用地理信息系统（GIS）预测浅层滑坡时间、空间信息的研究得到大力发展。本项目在GIS框架下，从浅层滑坡的机理出发，综合考虑地下水位、土体强度指标等关键参数的时变效应，建立浅层滑坡的物理过程模型，通过无限边坡模型来评价边坡的稳定性，针对大范围区域开展浅层滑坡易发性研究。预测土层厚度时：采用不同分辨率和精度的DEM数据，实现预测模型精度的优化；计算边坡安全系数时：通过室内试验得到滑坡发生前后土体强度指标和渗透系数等参数的变化规律，对已滑动区域进行相应的参数时变修正。在此基础上提出基于物理过程模型的浅层滑坡实时预报模型，以气象站降水数据为输入，考虑地下水位的时变效应，实现实时输出滑坡安全系数分布图，用于指导浅层滑坡灾害治理和监测。

中文关键词： 降雨型滑坡；GIS；逻辑回归；物理过程模型；滑坡风险分区

英文摘要： In the southeast of China, loose deposits slopes are most widely distributed, which is also mostly connected with human activities. Shallow landslides frequently occurred within these slopes in typhoon seasons. Recently, GIS is widely used in order to provide efficient preventative measures. Within the framework of GIS, the applicants will develop a process-based model to calculate the pressure head, predict the soil thickness and evaluate the factor of safety of each pixel using infinite slope equation. During the soil thickness predicting, DEM with different resolutions will be used to optimize the precision of the predicting models. Laboratory tests will be conducted to obtain time-dependent strength parameters of loose deposits. During the calculation of FS, different soil strength and permeability coefficient will be used for slid slopes and first-time landslide respectively. After the improvements, a real-time process-based model will be proposed to produce real-time hazard maps for shallow landsliding using only hourly update rainfall data. The methodology and results of study will be used to provide advice for shallow landslide mitigation.

英文关键词： Rainfall-induced landslide；GIS；Logistic regression；Process-based model；hazard zonation