基于多源数据的广域滑坡隐患数字地形分析

Digital terrain analysis of wide-area landslide hazards based on multisource data

浅表层滑坡地质灾害受地形土质含水影响较大,一般具有明显的形变特征。本文首先利用广域合成孔径雷达干涉技术(InSAR)识别异常形变区,并采用机载激光雷达采集获取地面不低于20个点每平方米(pts/m^(2))的点云数据,对点云进行滤波分类生成滤除覆被数字高程模型获取斜坡坡高、坡度、坡向及边坡形态等参量化地形地貌特征,随后利用高分辨率影像辅助提取滑坡隐患疑似区域开展外业现场核验。结果表明:滑坡隐患坡高集中分布在10~20m,坡度分布在25~45°,以北东向直线及阶梯形斜坡为主,植被较为发育。利用激光雷达(LiDAR)能精细探测滑坡隐患林下坡面形态特征,对植被发育区地质灾害早期识别具有工程参考价值。

The geological disasters of shallow surface landslides are significantly affected by the terrain and soil moisture content,and they generally have obvious deformation characteristics.This paper first employed wide-area interferometric synthetic aperture radar(InSAR)to identify abnormal deformation areas,and adopted airborne light detection and ranging(LiDAR)to collect point cloud data of not less than 20 points per square meter(pts/m^(2))on the ground.The point cloud was filtered and classified to generate the digital elevation model of filtered cover.The parameterized terrain and geomorphological features were obtained,including slope height,slope,aspect,and slope shape.Then,high-resolution images were employed to help extract suspected areas of landslide hazards for on-site verification.The results show that the height of landslide hazards is concentrated at 10-20 m,and the slope is distributed at 25°-45°.The northeast straight line and the stepped slope are the main,with well-developed vegetation.The use of LiDAR can accurately detect the slope surface morphological characteristics of forests with landslide hazards,which provides an engineering reference value for early identification of geological hazards in vegetation developed areas.