基于光学影像的滑坡形变监测研究——以2018年Palu地震触发滑坡为例

Monitoring of Landslides Deformation Triggered by the 2018 Palu M W7.5 Earthquake Based on Optical Images

以2018年印度尼西亚Palu地震触发的低坡度滑坡为例,利用Planet、Landsat-8和Sentinel-2等3种光学影像对Gumbasa水渠沿线高含水量区域的地表形变进行监测和精度评估。结果表明:1)3种独立的光学卫星获得的形变场结果具有较好一致性,且Planet获得的形变场结果精度最高;2)沿Gumbasa水渠分布有4个主要形变区,最大形变超过20 m;3)光学影像地表形变监测精度与影像空间分辨率之间并非呈现简单的线性关系。最后,通过滑坡位置、地物属性和水渠的空间位置分析地震触发滑坡的机理。

We investigate the low slope landslides triggered by the 2018 Palu earthquake in Indonesia.We use three kinds of optical images(Planet,Landsat-8 and Sentinel-2)to monitor and evaluate the surface deformation along the Gumbasa canal.The results show that:1)the deformation field results obtained by the three independent optical satellites are of good consistency,and the deformation field results obtained by Planet have the highest accuracy;2)there are four main deformation zones along the Gumbasa canal,and the maximum deformation exceeds 20 m;3)there is not a simple linear relationship between the surface deformation monitoring accuracy of optical images and the spatial resolution of the image.Besides,we analyze the mechanism of earthquake-triggered landslides by the landslides location,ground object properties and the spatial location of the canal.