GPS?InSAR Data Combined Methods For Three-Dimensional Surface Deformation Field

GPS,InSAR as two representative modern space geodetic technology play an increasingly important role in exploring the crustal movement and explicating the geophysical phenomenon etc.However,in terms of spatial resolution,GPS is limited by the number of receiver's antenna and thus is usually low,while InSAR can achieve surface deformation facets observation in large area.In time resolution respect,GPS can arbitrarily set receiving frequency according to the actual demand,whereas InSAR is limited by the round trip cycle of the SAR satellite,the time domain resolution is low.In the aspect of accuracy,CGPS can gain high precision result of horizontal deformation,InSAR is more sensitive to the vertical component due to the incidence angle of the SAR satellite.So how to exploit the complementary nature of two geodetic techniques to obtain high temporal-spatial resolution and precision three-dimensional deformation field of research area has a strong practical significance and research value.Modern surveying adjustment is a geodesy data processing technology based on the error theory and classical surveying adjustment,and introduces a series of new theories and methods of mathematical,such as mathematical statistics,linear algebra and matrix theory.It has been widely used in multi-means,multi-resolution and multi-temporal measurement data integration research.To use for reference it's successful application in "3S" integrated data processing,to be guided by the existing problems in GPS-InSAR data fusion,and the function model,stochastic model as core research contents,the additional system parameters adjustment,variance component estimation as the research breakthrough,to achieve the optimal integration of GPS and InSAR data sets.The main research innovations are as follows:1.Equivalence Relation between Analytical Optimization Method and Least Squares MethodThe relationship between the analytical optimization method based on Gibbs energy equation and the least squares method with considering data feature and imaging geometry is studied.Assuming the same weight matrix,the law equation of least squares method is exactly the solve linear equations of analytical optimization method,accordingly the two methods have the equivalence relation.Therefore,it is not necessary to design a complex numerical iterative model,just simply introduce the theory of modern surveying adjustment and apply in the basic least squares model to make it better.2.Ascending and Descending Orbits InSAR-GPS Data Fusion Model with Additional System Parameters(Function Model)Because of the difference imaging geometry,ascending and descending orbits InSAR theoretically can improve the accuracy of three-dimensional deformation solution.However,DEM error,the selection of phase unwrapping reference points,atmospheric error will lead different orbits InSAR data and InSAR with GPS data exist a systematic error due to the difference starting benchmark.Therefore,method of adding system parameters adjustment is put forward to weaken or eliminate this influence.But the error sources are different and intertwined with each other,it is difficult to adopt the additional parameter model which takes into account the characteristics and origins of the system error.Considering the atmospheric noise,DEM error typically in InSAR deformation monitoring error source are largely relative with the terrain,hence under the constraint of high accurate GPS observations,the system error comprehensive impact as a whole correction member is respectively added in ascending and descending orbits In SAR observation equation,in the form of polynomial function that related to the position of the solve points,thus establish the ascending and descending orbits InSAR–GPS data fusion method with additional systematic parameters.Finally,simulated and practical examples show that the proposed method is effective and feasible to obtain the high precision three-dimensional deformation field.The accuracy is improved in the ENU three directions,especially in the U direction.3.GPS-In SAR Data Combined Method Based on Vertical Variance Component Estimation(Stochastic Model)In order to solve the known problem that the given function model but the stochastic model is difficult to determine,in three-dimensional deformation field study by using GPS and InSAR data sets.Thinking of the InSAR is not sensitive to the horizontal deformation,and it's mainly depends on the GPS-Kriging interpolation,hence a method of vertical variance component estimation is proposed to combined the GPS-InSAR data,through analysis the variance component estimation stochastic model posterior estimation method.Firstly,the performance and scope of the method are analyzed by simulation experiment.It be proved can increase the efficiency and gain higher precision when the GPS density is sparse.Then it is successfully applied to extract the three-dimensional surface deformation rate in Xi'an area,the least squares method has been significantly improved.4.Wenchuan Mw7.9 Co-seismic Slip Distribution Inversion based on Automatic Fault Discretization Method(Practical Application)The above model is applied to the study of practical geology problems.Based on the GPS and InSAR co-seismic observations in Wenchuan Mw7.9 earthquake,we establish a high-precision three-dimensional deformation field by using GPS-InSAR integrated deformation analysis model with additional system parameters.And as a surface deformation constraint,merging the data from inversion of fault geometric parameters,using of triangular dislocation fault automatic discretization technology inverse its co-seismic slip distribution.The smooth transition between dislocations,amount of slip,sliding forms and the magnitude all show that the three-dimensional deformation field obtained by the function model with additional system parameters can get a more reliable sliding distribution,compared with the GPS three-dimensional deformation field result.