Study On Fault Activity In Western Sichuan Based On InSAR Data

Conventional fault activity monitoring methods(such as leveling and global positioning systems)can only monitor limited discrete points,and Interferometric Synthetic Aperture Radar is a surface observation based on the deformation monitoring means with high spatial resolution and all day All-weather observation.It makes up the shortcomings of low spatial resolution deformation monitoring based on point observation which provides favorable conditions for studying seismic activity and also provides fine observation data for fault sliding inversion based on geodetic technique.In the study of fault activity,the dislocation theory model is used for most time,but it can not be explained for some of the observed displacement field rotation.For the specific fault activity and the ground deformation field problem,the disclination theory describes the theoretical relationship between the mutual rotation in the fault plane of the elastic medium and the ground surface deformation.In this paper,a new Disclination-dislocation Model was adopted which is based on dislocation theory and disclination theory with taking the sliding and rotation of faults into account.In the typical area of the crustal deformation-the western Sichuan area,taking the Longmenshan fault and the Xinshuihe fault as examples,the Wenchuan coseismic displacement and the time series and the annual average deformation rate of the northwest part of the Xianshuihe fault were obtained by using the InSAR data and the inversion and rotation of the two faults were calculated based on the Disclination-dislocation Model.The main contents and contributions of this paper are as follows:(1)For the problem that disclocation theory cannot solve the the rotation of the fault,a new Disclination-dislocation Model was proposed to simulate the slip and rotation of the fault.And the relationship between the slip and rotation of the fault and the ground deformation was established.The model combines the advantages of the dislocation model of fault flapping and disclination model of fault rotating,which can effectively simulate the movement of the fault.The model successfully simulated the rotation characteristics of a single fault.(2)For the problem that C-band satellite signals obtained from the radar satellite in the hanging wall of of Longmenshan are largely unavailable,in this paper,ALOS PALSAR L data with wavelengths up to 23.6 cm were selected and the image of the six orbits before and after the Wenchuan earthquake were used.The two-pass differential D-InSAR technology was adopted for differential interference measurement;In order to ensure the continuity of the data between and within the tracks,the methods of weighted mosaic feathering,bottoming mosaic feathering and geometrical mosaic feathering were used in this paper.And finally the splice was deformed continuously by the weighted mosaic feathering of mosaic the coseismic displacement of Wenchuan earthquake,the associative perception of the image was enhanced.(3)In order to reduce the influence of InSAR data volume,data noise and redundant data on inversion calculation efficiency,an improved cubic volume re-sampling method was proposed in this paper.By this method,adopt the dense sampling site in main deformation area and adopt sparse sampling site in non-main area.This method solves the problem of taking samples on average and great number of data in traditional cubic convolution method.It better preserve the detailed information of InSAR data.the coseismic displacementrate of the Wenchuan earthquake and the annual average deformation rate of the northwest section of the Xianshuihe fault were achieved by this method.(4)For the characteristics of rotation during the coseismic process in the Longmenshan fault zone,based on the Disclination-dislocation Model and particle swarm optimization,GPS and InSAR data are combined to inverse the slide and rotation of Longmenshan for the first time,and compared with the sliding and rotation of the Longmenshan fault zone inversion with GPS data.The results show that compared with the results of dislocation model,the active feature of Longmenshan fault zone based on Disclination-dislocation Model is more in line with the feature of shovel surface of the Longmenshan fault zone;compared with the GPS inversion,the inversion combination of GPS and InSAR has lower strike-slip rate and dip slip rate with higher rotational angle and obvious rotational feature of fault zone.(5)In order to study the effect of Wenchuan earthquake in the fault activity of Xianshuihe and aiming at the characteristics of slowing fault shape changes of Xianshuihe,the Envisat ASAR C-band image of the 8 area in the northwest of the Xianshuihe fault was adopted in this paper about one year after the Wenchuan earthquake.The time series and the annual average deformation rate of the northwest part of the Xianshuihe fault were successfully obtained by using the SBAS_InSAR method.(6)For the existing rotation problem of the Xianshuihe,based on the Disclination-dislocation Model,particle swarm optimization was used and the InSAR data was firstly used to realize the slide and rotational inversion of the northwest part of the Xianshuihe fault.The results show that after Wenchuan earthquake,the crack rate and strike-slip rate of the Xianshuihe fault were slightly higher than those of the existing results.Segmentation feature of fault rotation was more obvious than slide.