Study On Coseismic Deformation Of Yushu Ms 7.1 Earthquake Using InSAR And GPS Measurements

Global Positioning System(GPS) and Interferometric Synthetic Aperture Radar(InSAR) play key roles in seismic deformation mapping. GPS can reliably obtain accurate measurements, while InSAR can accurately provide continuous ground deformation field with large coverage. Based on integrated GPS and InSAR measurements, seismic slip distribution can be effectively inverted, which has a vital significance in studying earthquake mechanism.The 2010 Ms7.1 Yushu earthquake is typically a destructive event, occurring in central Tibetan Plateau, causing large casualties and massive loss, with epicenter at the Ganzi‐Yushu Fault. Research on coseismic deformation of the seismic event is crucial to understanding its mechanism, disaster cause and implication to geodynamic process of the Tibetan Plateau.The main research works in this dissertation are listed below:The Key data processing strategy of InSAR are systematically analyzed. The principle of SAR imaging, InSAR observation model, and the error source of InSAR is deeply discussed. Especially, the algorithms of phase unwrapping, current mainstream methods, and existing problems are intensively investigated. Using the open source programs, ROI_PAC and SNAPHU, the test datasets and the InSAR data of Yushu earthquake are processed, with various strategies implemented. The key tricks of InSAR data processing are analyzed, and new processing strategies are proposed for the InSAR data processing of the Yushu earthquake.The basic principle of phase unwrapping is investigated. Common phase unwrapping algorithms, including branch‐cut method, least square method and network flow based method are introduced, respectively. The InSAR data processing software ROI_PAC is discussed from different aspects, including software installation procedure, data processing strategies, software structure and data post‐processing.A new method for unwrapping the Yushu Earthquake’s InSAR result is proposed in this dissertation. This method firstly separate the interferogram into two parts based on the earthquake rupture. Next, the minimum cost flow method in SNAPHU software is applied to each part respectively. Finally, the unwrapping results from the two parts are carefully combined. The experimental results are verified by using GPS data, which demonstrates that the proposed method can effectively improve the accuracy of Yushu Earthquake’s InSAR co‐seismic deformation field.A novel approach of the InSAR deformation field down‐sampling is proposed to obtain the optimal coseismic deformation of the Yushu earthquake, combining the quart‐tree algorithm and manual pixel addition. Considering the massive pixels(maybe exceeding million) and the low smoothness in local portions, the new approach is employed for the InSAR data processing of the Yushu earthquake. This can ensure that the major feature of original deformation field is remained after downsampling. And the downsampled dataset can satisfy ensuing source model inversion, which requires limit of data quantity.Integrated InSAR and GPS deformation is utilized to invert for the source model of the Yushu earthquake. Multiple weighting and smoothing factors are tested to analyze their effect on the inversion result. The input data are divided into four groups: a) GPS data only; b) InSAR data only; c) GPS and InSAR, the weighting factor of InSAR is 1.0; d) GPS and InSAR, the weighting factor of InSAR is 0.1. respectively. The experimental results demonstrate that by integrating GPS and InSAR data, the precision of inversion can be improved and the resolution of the fault slip can be enhanced.