Early Landslide Identification And Deformation Monitoring In Ledu District Based On Radar Remote Sensing

Landslide,as one of the geological disasters that threaten the most human beings in nature,seriously restricts the development of society and economy.By clarifying the past temporal and spatial evolution of landslides,we can understand the current state of landslides and predict the future evolution trend of landslides,so as to prevent or reduce the losses caused by landslides.The Interferometric Synthetic Aperture Radar(InSAR)and pixel offset tracking(POT)method deeply couple the phase information and amplitude information of radar images.They can complement each other to retrieve the entire deformation process of the landslide.Therefore,taking Ledu District in the northeast margin of Qinghai-Tibet Plateau as an example,this paper used different Sentinel-1A data to study the SAR topographic visibility of Ledu District based on radar remote sensing.The early identification of unstable slope in Ledu District was studied systematically,and the surface deformation monitoring and analysis of two typical landslides in Ledu District were respectively conducted.Our study aims to provide scientific reference for local governments to make further in-depth observations or take effective preventive measures.The main conclusions are as follows:(1)Based on the geometric parameters of Digital Elevation Model(DEM)and Sentinel-1A data,the topographic visibility of SAR images in Ledu District was studied.The results showed that the sentinel-1A satellite’s ascending and descending orbit data are more sensitive to the monitoring of east-facing and west-facing slopes,respectively.(2)Based on the Stacking-InSAR method,the surface deformation results of Ledu District from 2018 to 2020 were obtained,which verified the ability of the Stacking-InSAR method for early identification of regional unstable slopes.Based on the surface deformation results,8 unstable slopes were delineated in Ledu District.(3)For the typical loess landslide in Ledu District,the geomorphic changes and pre-and post-failure spatiotemporal evolution of Jiangou landslide were deeply studied by means of visual interpretation,unmanned aerial vehicle(UAV)observation and field investigations.The results show that Jiangou landslide is always in creep state before it occurs.After the occurrence of the Jiangou landslide,the remaining landslide materials were washed into the Jiangou River and formed the dammed lake,which made the landslide still have potential sliding risks.Through further research and analysis,it is found that the pre-and post-failure spatiotemporal evolutionary patterns of the landslide are different,which are progressive failure mode and retrogressive failure mode respectively.In addition,we found that the slope adjacent to Jiangou landslide is in an unstable state.(4)Using the SBAS-InSAR method and the improved POT method,the surface deformation velocities before and at the occurrence the first landslide in Gaojiawan were retrieved.The study showed that the fault may have affected the stability of the landslide,which in turn affected Zhangjiazhuang tunnel operation.In addition,the Gaojiawan landslide has experienced long-term creep,and while studying the post-second landslide’s failure stability,the acceleration trend was identified via time series analysis,which can be used as a precursor signal for landslide disaster warning.This is consistent with the news report that the China Railway Qinghai-Tibet Group monitored the warning event in the Zhangjiazhuang tunnel,indicating that the deformation of the landslide had a certain impact on the tunnel.