InSAR Tropospheric Delay Correction And Landslide Monitoring In High Mountain Valley

Interferometric Synthetic Aperture Radar(In SAR)can monitor surface deformation remotely and for a long period of time,filling the gap of traditional geohazard investigation means that it is difficult to detect and investigate hidden hazard points.In recent years,In SAR technology has been widely used in the field of landslide monitoring,and has gradually become one of the common means of landslide monitoring.However,tropospheric delay is an important error that needs to be weakened by In SAR.For different geographical environments,a reasonable selection of tropospheric delay correction methods is needed to improve the accuracy of In SAR deformation monitoring.This study selects a typical high mountain canyon area in China,and carries out In SAR tropospheric delay correction research based on ascending and descending Sentinel-1A image data,aiming to improve the accuracy of In SAR deformation measurement in high mountain canyon area and better serve the work of landslide potential hazard identification,monitoring and analysis.The main research works and contents of the paper are as follows:(1)Three single correction methods of In SAR tropospheric delay,Linear,Generic Atmospheric Correction Online Service for In SAR(GACOS)and ERA5,were used to correct the original In SAR interferogram for tropospheric delay.The results show that the standard deviation of the interferometric phase,the semi-variance function and the correlation between the interferometric phase and the elevation of the three single correction methods were used to evaluate the effectiveness of the tropospheric delay correction.The results show that,in terms of tropospheric delay noise mitigation for the interferograms in the selected study area,the Linear method has the best tropospheric delay correction among the three single correction methods,with the GACOS and ERA5 methods being the next most effective.(2)To further improve the In SAR tropospheric delay correction,a joint In SAR tropospheric delay correction method is proposed in this study on the basis of a single correction method.The method combines the Linear,GACOS and ERA5 methods to correct the tropospheric delay with the standard deviation of the phase in the nondeformed region of the interferogram,the sill value of the semi-variance function and the out-of-correlation distance as constraints.In comparison with the GNSS deformation monitoring results,the joint correction method improves the accuracy of In SAR deformation monitoring for ascending and descending tracks by 26% and 6%respectively.The joint In SAR tropospheric delay correction method has been demonstrated to maximize the accuracy of In SAR deformation monitoring in a small area covered by GNSS monitoring points,and its tropospheric delay correction capability is better than the three single correction methods.(3)Combining the In SAR deformation results of ascending and descending tracks corrected by the joint In SAR tropospheric delay correction method,optical images and3 D topography,and applying them to the identification of landslides on both sides of the Jinsha River basin in the Ahai reservoir area,five typical landslide areas were finally identified,and one of the typical landslides was selected for analysis and discussion of its spatial and temporal evolution characteristics.