Research On Ground Deformation Detection Based On SBAS-InSAR Technology

Ground subsidence,also known as ground collapse or depression,is a main engineering geological problem in urban construction all around the world.The geological hazard caused by land surface deformation exert great harm and risk to the lives of the masses.Hence,searching for reliable surface deformation detection measures to achieve effective supervision and prevention is of great significance.Compared with conventional leveling technique,Differential Interferometric Synthetic Aperture Radar(D-InSAR)and Small Baseline Subset Interferometric Synthetic Aperture Radar(SBAS-InSAR)technologies show great advantages of all-weather,all-time,large scale and high precision,which have been extensively applied in the wide area supervision of ground deformation.As the matter stands,it can provide an efficient and reliable way for monitoring and detecting geological hazards.Aiming at the issues in the data process and deformation detection performance through SBAS-InSAR technology,this thesis conduct research from the following aspects:1.Based on the composition model of InSAR phase,we explored the influence of each phase component to D-InSAR deformation detection.Then,simulation experiments were conducted,validating that the effective methods can improve the performance of processing.Next,we successfully inverse the deformation in Kumamoto earthquake with the SAR data from Sentinel-1 satellite.In general,this section laid the foundation for the following SBAS-InSAR data process.2.With SBAS-InSAR technology,the linear deformation and elevation measurement estimation accuracy can reach several millimeters a year and a few meters respectively.However,the theoretical limit of their estimation accuracy is not given clearly.By theoretical modeling,we derived the Cramer-Rao bound of linear deformation and elevation measurement error estimation,and quantitatively analyzed the impact of various non-ideal factors on SBAS-InSAR deformation detection performance with Monte Carlo simulations.3.In the section,we designed a scheme for SBAS-InSAR high coherent targets and long-time series deformation data simulation,and implement the proposed scheme into an independent module for the D-InSAR simulation software of our team.Through the in-depth analysis on the 19 sets of SAR data simulated by our software,we verified the correctness of the designed scheme and the effectiveness of the software module.Afterwards,based on the Sentinel-1 SAR data in Xi'an city from December 2015 to February 2017,we retrieved the surface deformation,and then made analysis on the results of Xianyang International Airport,County Road 106(near Mazhuang)and Xianyang railway station,which are basically consistent with the current available result,proving the validity of process scheme proposed in this thesis.D-InSAR and SBAS-InSAR show great prospect both in the extraction of violent deformation in earthquake region and the supervision of moderate ground deformation,which provide vital technical support for the disaster prevention,mitigation and evaluation,whose potential applications are extremely broad and bright.