茂县滑坡形变的Sentinel-1数据分析

Analysis of Maoxian landslide deformation by using Sentinel-1 data

为了获取2017年6月24日四川省茂县特大滑坡滑前的形变信息并分析其诱发因素,该文利用覆盖滑坡区域滑前40景Sentinel-1数据,采用合成孔径雷达干涉点目标分析(IPTA)技术,解算出该滑坡滑前2014年10月9日至2017年6月19日期间形变的年速率及形变的时间序列,并与降雨数据进行了对比分析。对于滑坡区植被覆盖严重的问题,该文利用永久散射体目标(PSC)对影像配准窗口大小及过采样因子不敏感的特性来识别PS点。分析结果表明,茂县滑坡在滑动前两年多时间内具有明显的形变,最大形变速率达到4(cm·a-1),而且在滑坡发生前半个月内形变有突然增大的现象,推测该滑坡滑前持续的强降雨是导致滑坡失稳的主要诱因。

An extraordinary landside occurred in Maoxian,Sichuan province on June 24 th,2017.In order to obtain the pre-slide deformation and induce the trigger factor of this landslide,40 scenes Sentinel-1 data from October 9 th,2014 to June 19 th,2017 were involved.And deformation velocity and pre-slide deformation time series were achieved with synthetic aperture radar interferometric point target analysis(IPTA)technique,which was compared with the pre-slide rainfall data.According to the problems of the heavy vegetation coverage in landslide area,persistent scatterer candidates(PSC)were identified based on the fact that strong scatterers are less sensitive to the window size and oversampling factor in the coregistration procedure.Analytical results indicated that the landslide had obvious deformation in two years before this landslide event.The maximum deformation rate reached 4(cm ·a-1).Moreover,an abrupt increase deformation was monitored in half a month before this landslide event,which can be inferred that the continuous heavy rainfall before landslide event was the main trigger factor of the landslide.