Research On Key Issues Of Long-Term Ground Subsidence Monitoring And Forecasting Using InSAR Technology

The monitoring of ground subsidence using InSAR technology has been developed for more than 30 years,starting from the application of D-InSAR technology to the later time series InSAR technology,which is more mature than the monitoring of other geological hazards like landslides and mining.However,with the continuous launch of SAR satellites,how to integrate multi-platform data for long-term deformation monitoring,mechanism inversion and accurate prediction of ground subsidence is still one of the urgent problems,which is crucial to the monitoring and management and prevention of geological hazards directly related to surface deformation.In the existing studies,most scholars have only used a certain satellite to study a short period of deformation in a certain area,and few have studied how to obtain long-term deformation time series methods from multiple sources of data.Therefore,there is an urgent need to develop new multi-source time-series InSAR techniques to monitor long-term surface deformation.Meanwhile,with the advent of SAR big data era,how to dynamically process high-frequency InSAR data and at the same time use the dynamically updated monitoring values to predict the future development trend of ground subsidence is also the second problem that needs to be urgently addressed.With the continuous development of large-scale human engineering activities,such as land reclamation and mountain excavation,how to use InSAR technology to dynamically monitor the changes of surface deformation and surface elevation is also the third problem that needs to be solved.Based on these three aspects,this paper focuses on multi-source long-term series InSAR ground subsidence monitoring,prediction and joint ground elevation change and deformation monitoring methods,and the main content and innovation points of the paper are summarized as follows:(1)For the study of multi-source long time series InSAR ground subsidence monitoring,we proposed a long time series InSAR data processing technique fusing segmented linear interpolation and quantile-quantile adjustment(QQA),which can be applied to the fusion multiplatform data with data missing and data overlapping.Xi’an city was selected as the monitoring test area,and its archived SAR datasets from 1992 to 2022,including ERS-1/2,Envisat/ASAR,Terra SAR-X,ALSO/PALSAR-2,and Sentinel-1,were collected.The results show that the proposed method is better than the traditional SVD connection technique,and it is found that the surface deformation in Xi’an over the past 30 years has undergone a process of accelerated subsidence-decelerated subsidence-accelerated subsidence-decelerated subsidencestabilization-uplift,and the maximum ground subsidence that can be monitored exceeds 2 m.Finally,with the combination of groundwater data,it is found that the ground subsidence in Xi’an is an inelastic deformation process,and land subsidence has a strong relationship with groundwater changes and land use changes.(2)For the problem of dynamic processing of SAR massive data and dynamic prediction of land subsidence area of reclaimed,a set of dynamic prediction techniques for the ground subsidence area of reclaimed land is proposed.Shuangyu Island,Zhangzhou City,was selected as the experimental area,and 125 views of Sentinel-1 data were collected,and the ground subsidence in the study area was processed and predicted dynamically by combining sequential estimation and geotechnical models.The results show that the average standard deviation of the predicted and measured values is less than 5 mm,and the cumulative maximum land subsidence of the study area will exceed 1.3 m by 2025.(3)A joint surface deformation and surface change InSAR solution method is proposed for the spatial and temporal changes of surface elevation and multi-dimensional surface deformation caused by the excavation and urbanization.The Chengguan District of Lanzhou City was selected as the experimental area,and a total of 513 images from 2014 to 2022 of SAR data from one ascending orbit and two descending orbits of the Sentinel-1 satellite were collected,and the proposed method was used to calculate the spatial and temporal evolution of ground elevation for three orbits.The results show that the cumulative change in ground elevation in Chengguan District,Lanzhou City is between 60 and-80 m,and the correlation between two is greater than 0.89.Meanwhile,the cumulative two-dimensional deformation decomposition results calculated by the LOS deformation of three orbits show that there is not only a large vertical deformation but also a large east-west deformation in the excavation of the mountain.