两种概率模型在滑坡敏感性分析中的应用及比较——以甘肃省天水市罗玉沟为例  

Application and comparison of two probability models in landslide susceptibility analysis-A case study in Luoyu Valley,Tianshui

作　　者：李亚军[1] 郭鹏[1] 赵洪涛[2] 陈冠[1] 曾润强[1] 乔良[1] 

机构地区：[1]兰州大学西部环境教育部重点实验室,兰州730000 [2]甘肃省科学院自然灾害防治研究所,兰州730000

出　　处：《干旱区资源与环境》2013年第2期198-202,共5页Journal of Arid Land Resources and Environment

基　　金：国家基金委地球科学部主任基金项目(批准号:41040005)资助

摘　　要：滑坡敏感性分析是对一个地区孕灾禀赋的评价,在缺少时间序列观测资料地区,敏感性分析往往比危险性分析更具科学性。CF指数与信息量模型是敏感性分析中的两种概率模型,文中运用以上两种模型对天水市罗玉沟流域滑坡作了敏感性分析,在对模拟结果进行验证的基础上得出,两种模型都表现出了较好的模拟结果,但较之CF指数,信息量模型具有更高的分析精度。根据完整滑坡数据生成了研究区的敏感性分布图,结果显示研究区敏感性出现了明显的空间分异格局:流域中下游敏感性高而上游敏感性低,主沟南岸敏感性高而北岸敏感性低。最后结合研究区水文地质条件,分析了这种格局的形成原因。Known to assess the hazard-pregnant capability of a given area,landslides susceptibility analysis seems more efficient than hazard analysis when the necessary data of time sequences are inadequate.CF and Information model,two probability models were used to carry out landslide susceptibility analysis in Luoyu Valley,Tianshui.Based on the validation of the results,both model shows good performances,while the later seems more accurate.A susceptibility map was generated which shows distinct spatial distribution mode of landslide susceptibility： the middle and lower reaches of the catchment appear to be more susceptible than the upper reaches and the south side of the main valley is less susceptible than the north.At last,a conceptual analysis of the mode of spatial distribution is given.

关 键 词：CF指数 信息量模型 敏感性 滑坡 GIS 

分 类 号：P642.22[天文地球—工程地质学]