Three-dimensional Deformation Extraction And Thickness Inversion Of Landslides In Alpine And Canyon Region Based On DS-InSAR Technology

As one of the main types of geological disasters in China,landslides are widely distributed,highly destructive,and pose a serious threat to people’s lives and property safety,as well as the safety construction and operation of infrastructure projects.Therefore,identifying and monitoring landslide hazards is of great significance for improving the ability to prevent and control geological disasters.The geological environment and landform conditions in mountainous canyon areas are complex,and their landslide hazards have characteristics such as high position and concealment,making traditional ground investigation methods ineffective.Interferometric Synthetic Aperture Radar(In SAR)technology,with advantages such as all-weather,wide coverage,and high precision,has been widely used in early identification and monitoring of landslide hazards.However,traditional time-series In SAR technology is susceptible to factors such as low coherence and atmospheric delay in the identification and monitoring of landslide hazards in mountainous canyon areas,resulting in fewer monitoring points,lower accuracy in identifying landslide hazards,and poor accuracy in extracting landslide three-dimensional deformation results.Therefore,it is urgent to enhance the application potential of In SAR technology in the identification and monitoring of landslide hazards,enhance China’s ability to prevent and reduce geological disasters,and effectively maintain and guarantee the safety of people’s lives and property.In response to the susceptibility of traditional time-series In SAR technology in mountainous canyon areas to low coherence factors,this thesis fully explores the distributed scatterers in mountainous canyon areas using the Distributed Scatterer In SAR(DS-In SAR)technology,effectively improving the number and density of monitoring points.In response to the seasonal vertical layered atmospheric delay interference phenomenon commonly found in mountainous canyon areas,this thesis proposes a mountainous canyon area tropospheric delay correction method called optimized multi-temporal move-window linear model(OMMLM),which effectively reduces the influence of terrain-related atmospheric delay interference on monitoring results.In response to the problem of poor accuracy in extracting landslide three-dimensional deformation results,this thesis proposes a landslide three-dimensional deformation extraction method that considers the sensitivity and spatial continuity of ascending and descending orbit satellite monitoring,and combines the mass conservation model and finite difference method to invert the landslide activity thickness,which can provide technical support for studying landslide movement mechanisms and evaluating landslide risk levels.The main work and research results of this thesis are as follows:(1)Summarize the basic principles and error sources of In SAR and DS-In SAR technologies,summarize the basic principles of traditional time-series In SAR and DS-In SAR technology,analyze the advantages of DS-In SAR technology in the identification and monitoring of landslide hazards in mountainous canyon areas,and summarize the basic principles and characteristics of In SAR atmospheric delay effect.(2)Analyze the characteristics of atmospheric delay dominated by seasonal vertical stratification in mountainous canyon areas,and based on the linear relationship between vertical stratification phase and elevation,propose OMMLM for mountainous canyon area tropospheric delay correction method.Compared with the results of spatial-temporal filtering,ERA5,and GACOS correction,this method can effectively reduce the interference of tropospheric delay in mountainous canyon areas.(3)Study the sensitivity differences of SAR satellites in deformation monitoring in mountainous canyon areas,propose a landslide three-dimensional deformation extraction method based on the sensitivity weighting of ascending and descending orbit monitoring and spatial smoothing constraint.(4)Taking the Gongjue section of the Jinsha River as an example,this thesis utilizes OMMLM and DS-In SAR technology to obtain the regional surface deformation distribution and carry out landslide hazard identification and monitoring.The results show that the proposed method can significantly reduce the missed detection rate and false detection rate of landslide hazard identification in high mountain and canyon areas.By analyzing precipitation,temperature,and other data,the spatiotemporal deformation characteristics of typical landslides are also studied,and the results demonstrate that precipitation is one of the main driving factors of landslide deformation in the study area.The thesis includes 47 figures,7 tables,and 113 references.