| InSAR (Interferometric Synthetic Aperture Radar), which was first proposed in the 1970s, developed rapidly in the 1990s, is booming to the present as an Earth observation technology. It has produced great impact on many research fields of earthsciences. One of the most successful applications of InSAR is to survey the deformation caused by earthquakes. However, one-dimensional measurements that is the projection of the surface three-dimensional deformation on the line of sight (LOS) direction, has greatly limited the capability of InSAR technique in characterizing the true surface displacements. People often investigate strategies for mapping surface deformation in three dimensions by using multiple interferograms with different imaging geometries. Since the side-looking geometry of the SAR sensor and its near-polar orbiting satellite platform cause the InSAR measurement extremely insensitive to the north component, error amplification phenomenon can be found in the combination of multiple InSAR measurements from different viewing geometries, and the most exaggerated errors are presented in the north component. It is known that the crux of retrieving the true three-dimensional deformation is to get the north displacement (or the along track displacement) exactly, and a new technology called MAI which is the further development of InSAR in inferring the azimuthal displacement, is able to provide effective constraints on the horizontal coseismic deformation field. This thesis selects the 2008 Gaize, Tibet earthquake with wealthy SAR data and tectonic significance as the studying object. It utilizes a least square iterative approximation solution with a priori knowledge and the fusion of InSAR and MAI measurements to suppress the error amplification. This work concentrates on deriving the complete 3D co-seismic surface deformation fields caused by the 2008 Gaize earthquake at first, and then inverts the fault slip distribution based on the Okada dislocation model in an elastic half space. The results and conclusions in this thesis are summarized below.1. Acquisition of the one-dimensional co-seismic deformation fields by InSAR and MAI. Firstly this work obtained four LOS co-seismic measurements by carrying out the differential interferometry on three tracks (2 ascending and 1 descending) of Envisat ASAR images and one track (ascending) of ALOS PALSAR images. The ENVISAT ASAR images acquired from descending were selected based on data coherence and optimal baseline and then processed with the method of MAI to generate one azimuthal displacement field. Two remarkable subsidence centers and complex features of inconsecutive and obscured fringes are observed in the all four LOS co-seismic measurements. Azimuthal measurement shows that the northwest part (the hanging wall side of the source fault) of the whole deformation area has more obvious southward motion relative to the others (the footwall side).2. This work resolved the co-seismic 3D displacements of the 2008 Gaize earthquakes by combination of multi-pass LOS measurements, a least square iterative approximation solution with a priori knowledge (also called indirect solution) and the fusion of D-lnSAR and MAI measurements, respectively and it is done within the common area covered by the four one-dimensional co-seismic interferograms. Then all 3D displacements were analyzed and evaluated based on the error estimate formulae and the simulated deformation field. The results show that the error amplification phenomenon in direct solution can be significantly decreased by using the indirect solution and the fusion of D-lnSAR and MAI measurements, and the feature of the co-seismic surface deformation caused by the Gaize earthquakes can be better characterized by the latter two. This study chose the results induced by the indirect solution as the best complete 3-D displacement maps by considering the principles of best combination of estimation precisions and relation between deformation characteristics and the seismogenic fault. The optimal results show that in addition to northward movement with amplitude less than 5cm around the aftershock fault, the north-south deformation field on the hanging wall of the main-shock fault suggests the overall southward movement with the maximum of 10cm. The boundary dividing the two sides of the main-shock fault is very clear in the vertical movements, and two remarkable subsidence centers are observed on the hanging wall, while amplitude of the west one (48.9cm) is larger than the east (41.4cm). The horizontal displacement field exhibits the eastward rotation characteristic with magnitude increasing from west to east, and obviously separating movements around both the main-shock fault and aftershock fault, and convergence features at the two subsidence centers on the hanging wall. Three-dimensional displacements indicate that the induced slip is predominantly vertical and rotates to the east in the horizontal plane.3. The thesis selected some intererograms with high correlation and large spatial coverage for the inversion. Combining with tectonic background and seismic characteristics, this work obtained fault parameters of the 2008 Gaize Mw6.4 main-shock and Mw5.9 aftershock based on the Okada’s dislocation model in an elastic half space. The main-shock occurred on a west-dipping fault with the strike of 219° and dip of 50°, with maximum slip of 1.39m and the average rake of -65.66°, and the focal depth is about 6.51km, so it is dominated by normal faulting with small left-lateral strike slip. The Mw5.9 aftershock is a nearly pure normal faulting located northwest to the main-shock fault, striking 212° and dipping to 50°. The maximum slip is about 0.87m that occurred at the depth of 4.21km along the fault plane, and the average rake is about -75.01°. The moments released by the two earthquakes are 4.29×1018N·m and 1.32×1018N·m, respectively.The inversion results above show that the 2008 Gaize earthquakes all occurred on high-angle normal faults. And three-dimensional displacements indicate that the induced slip is predominantly vertical. The tectonic background indicates that the basin where the epicenter is located at its northern edge, is situated in the intersection of the striking EW, dextral strike-slip Dongco fault, Gase fault and the striking NE Riganpeicuo fault with left-lateral slipping. The tension-torsional energy in the Dongcuo basin, the eastward extension induced by the Dongco fault, Gase fault, and the southwestward contraction by the Riganpeicuo fault may be jointly responsible for occurrence of the 2008 Gaize earthquakes... |
PDF Full Text Request