Monitoring And Analyzing Displacement Of The Daguangbao Landslide By Multi-Temporal InSAR

China has become one of the most frequent landslide disasters in the world due to its own topography,landform and climate conditions.Great casualties and economic losses are caused by landslides in China every year.The deformation monitoring and analysis before landslide occurrence can provide important reference and decisionmaking basis for landslide disaster early warning.On May 12,2008,Wenchuan County,Aba Tibetan and Qiang Autonomous Prefecture,Sichuan Province,suffered a huge earthquake with a magnitude of 8.0Ms.The earthquake not only caused huge casualties,damage to buildings and property losses,but also caused great damage to the local ecological environment and geological structure,resulting in more than 60,000 landslides and landslides.The Daguangbao landslide,located in Gaochuan Township,Anxian County,Mianyang City,Sichuan Province,is the largest co-seismic landslide triggered by the Wenchuan earthquake and one of the largest giant landslides in the world in recent decades.The landslide also formed a large number of loose deposits and unstable slopes,and the surrounding earthquakes are frequent,concentrated rainfall,easy to induce landslides;The geological hazards caused by strong earthquakes and their long-term aftereffects are an important topic of current geological disaster prevention.In order to identify,judge and evaluate potential landslide hazards from the source,it is necessary to monitor the deformation of the area for a long time and analyze its movement trend.Traditional methods of landslide surface deformation monitoring,such as precise leveling and global positioning system(GPS)measurement,have high accuracy and reliability,but they are time-consuming and laborious,and can not be used for largescale deformation monitoring.In contrast,InSAR technology has the advantages of wide coverage,high resolution,all-day and all-weather monitoring,and high monitoring accuracy.It is suitable for large-scale landslide deformation monitoring.However,the traditional DInSAR technology is affected by time-space decorrelation and atmospheric delay,which seriously limits the application of DInSAR technology in deformation monitoring,especially in landslide monitoring with long time and slow displacement.In order to solve this problem,multi-temporal radar interferometry(MTInSAR)algorithms such as Persistent Scatterer InSAR(PSI),Small Baseline Subset(SBAS)and Stacking-InSAR(Stacking-InSAR)have been proposed successively,which effectively suppress the effects of decorrelation and atmospheric delay in DInSAR technology on monitoring results.In order to obtain and analyze the long-term motion state of large optical packet,it is necessary to select appropriate data and methods.The main work of this paper is to process two images of adjacent time intervals in 84 Sentinel-1A SAR image data by DInSAR.It is found that there are obvious interference fringes in the interferogram of large optical packet area,which indicates that there is less vegetation coverage in this area and the image coherence is strong.So Stacking and SBAS are two sequential InSAR techniques to monitor the deformation of large optical packet area for a long time.Then two digital elevation models,SRTM DEM and Tan DEM,are used to process data and generate interferograms respectively.It is found that the interference fringes generated by the two DEMs are obviously different.Referring to the data,it is found that the height change of Daguangbao landslide is as high as 500 m in less than 4 km horizontal distance before and after the occurrence of Daguangbao landslide.SRTM DEM was made before the earthquake and Tan DEM was made after the earthquake.After comparing the differences between the two DEMs,it is found that the trend is consistent with the existing results.So the selected digital elevation model is Tan DEM.Based on the multi-temporal InSAR method,the deformation of Daguangbao Mountain is monitored.Based on time series InSAR method,Daguangbao Mountain is monitored for deformation monitoring.There are three active moving areas in Daguangbao Mountain.One area is the cliff formed by the isolated peaks of Daguangbao Mountain after the landslide.The remaining two areas,one area is the accumulation area of a large number of loose deposits,the other area is the exit of landslide flow.The most active movement is caused by a large number of loose deposits and unstable slope instability in the region.From 2014 to 2018,the maximum deformation of Daguang Baoshan body along the LOS direction is 20 cm.The minimum deformation is 5mm.At the same time,in order to verify the reliability of the monitoring results,the time series deformation of Daguangbao landslide was obtained from the elevated SAR image data of PALSAR2 covering the research area from 2017 to 2018 based on SBAS algorithm and compared with the time series deformation obtained from Sentinel-1A data in the same period.The results show that the time series deformation information obtained by the two data sources is basically the same,which verifies the reliability of the deformation monitoring results.In this paper,the Daguangbao landslide is monitored for a long time based on the multi-temporal InSAR algorithm,and three regions are still sliding.The three sliding areas remained in a uniform sliding state during the four years from 2014 to 2018.Without the influence of external conditions(such as heavy rainfall and earthquakes),these three areas will continue to slide in the next few years.If there are precipitation and earthquakes and other induced factors,landslides may occur.It also shows that the Daguangbaoshan massif has not yet restored to its steady state before the Wenchuan earthquake,and the deformation monitoring in this area needs to be continued.This paper demonstrates the advantage of time series InSAR technology in landslide longtime deformation monitoring.With the improvement of technology and algorithm,time series InSAR technology will be more widely used in landslide monitoring.