基于统计和数值模拟的黄土滑坡地形因子敏感性分析

Sensitivity analysis of loess landslide terrain based on statistic and numerical simulation

斜坡地表形态对滑坡的孕育起着重要控制作用,地形参数的提取及地形地貌对滑坡的影响是区域滑坡灾害评价的必要过程。文中利用最小流域和精密河网将研究区划分为31418个自然斜坡单元,利用ArcGIS区域统计功能提取斜坡单元的地形参数,分析研究区斜坡的坡体形态分布规律。依据区内292处滑坡调查点的资料,统计分析不同坡体形态下的滑坡发生概率,结合数值模拟分析结果,确定黄土斜坡发生滑坡的坡型、坡度、坡高及坡向敏感性。分析结果表明:研究区凸型斜坡发生滑坡的概率最高;随着斜坡坡度和坡高增大,发生滑坡的概率增大;阳坡发生滑坡的概率明显高于其他坡向的边坡。

Slope terrain plays an important role in loess landslide.In this study,the large research area is divided into 31418 slope units according to the watersheds and steam networks by the Geographic Information System（GIS）.All the terrain parameters of the slope units can be got by Zonal Statistics of Spatial Analyst Tools in GIS and the topographic distribution of slope units will be shown.The probability distributions of landslides at ranges of different profile curvatures,slope,slope high and aspect aree discussed using parameters of all slope units and 293 loess landslides in the study area.It is found out that the probability of loess landslides would increase with the profile curvatures,slope and high.