Time-Series InSAR Technology For Monitoring And Analysis Of Landslide Deformation In Guang’an Village

China is one of the countries with frequent geological disasters in the world,among which landslides account for a large proportion of the total number of geological disasters in China,and due to the uncertainty of landslide occurrence,it causes great loss of human,material,and financial resources every year,which always threatens the safety of people’s lives and properties.In the early morning of October 21,2017,a high-level landslide occurred in Guang’an Village,Dahe Township,Wuxi County,Chongqing City,which caused severe deformation of the local slope and great potential safety hazards.Therefore,it is of great practical significance to carry out the analysis and research of landslide deformation monitoring in Guang’an village.With the continuous development of science and technology,InSAR technology provides a new method for landslide deformation monitoring,which has the advantages of all-day,wide range,high accuracy and less restricted by climatic conditions,and has been widely used in the field of landslide identification and monitoring.However,due to the large magnitude and rapid deformation of high-level landslides,traditional DInSAR technology is susceptible to spatial and temporal decoherence and atmospheric delay,which makes it difficult to monitor single landslides comprehensively and obtain reliable deformation results,so it is important to combine effective technology for landslide deformation monitoring and prediction analysis research.To address the above problems,this paper adopts the time-series InSAR technique to analyze and predict the deformation characteristics before and after the landslide in Guang’an Village landslide.main research contents and results are as follows:(1)Monitoring and analysis of time series deformation before and after Guang’an village landslide.SBAS-InSAR and IPTA technologies were used to process 84 scenes of Sentinel-1A data covering the landslide area,of which 18 scenes were from March 17,2017,to October 19,2017,before the landslide,and October 2017 after the landslide,from the 31 st to December 26th,2019,a total of 66 scene data were collected.The data processing process was optimized,and the annual average deformation rate and time-series cumulative deformation variable before and after the landslide were obtained.By selecting typical feature points,the deformation rates obtained by the two methods are verified and analyzed,and the correlation coefficient was 0.86 before the landslide and 0.85 after the landslide,indicating that the two InSAR results are in good agreement in terms of location and magnitude.There are obvious deformation areas on the slope before and after the landslide occurs.(2)Analysis of the deformation characteristics of Guang’an Village landslide.Firstly,Offset Tracking technique was used to process the ALOS2 data before and after the landslide to obtain two-dimensional offset of the landslide,and the range and azimuthal deformations were consistent with the distribution position of the landslide area.Then,combining the time-series InSAR results with the landslide investigation data of Guang’an Village,the landslide deformation characteristics were comprehensively analyzed from three perspectives:the twodimensional offset of the landslide,the spatial distribution of the deformation before and after the landslide,and the factors influencing of landslide.The results show that the spatial distribution of time-series InSAR deformation before and after the occurrence of the landslide in Guang’an Village is consistent with the distribution of the landslide instability rupture location,deformation zoning,and the changes in optical images,which verifies the rationality of the timeseries InSAR results;and the local slope continues to deform after the landslide,under the action of triggering factors such as abnormal rainfall,chert dissolution,gully cutting,terrain fluctuation and human activities,there is still has the possibility of sliding again,which can be combined with time-series InSAR technology to enhance the detection and analysis of landslide stability.(3)The post-landslide deformation prediction model is established based on the time-series InSAR technique and PSO-SVR algorithm.Combining the spatial distribution characteristics of post-landslide deformation,feature points with significant deformation signals are selected in different deformation sub-areas,the corresponding time-series cumulative deformation variables are extracted,and the data are divided into training and test sets in the ratio of 9:1 to establish the post-landslide deformation prediction model,and the accuracy of the prediction model is evaluated by using model evaluation indexes.The results show that PSO-SVR algorithm can predict the time-series InSAR deformation results well,and the coefficient of determination of the test set data is greater than 0.5.The model fits well and can meet the requirements of engineering applications,which provides an estimation method for landslide stability analysis.