Global GPS Data Processing And Applications In Crustal Movement

GPS are widely used for its advantages of all-weather,high-precision and real-time in the fields of crustal deformation,geodynamics,seismology and so on.Especially with the intensive construction of the permanent global continuous GPS observation stations and improvement of accuracy in GPS data processing,GPS can monitor more precisely on global plate motions,regional crustal deformation of active blocks and fracture state of motion.It can provide a high-precision regularity of the displacement field of the crust surface over time.In order to study the global high-precision data processing method and study the global plate movement law,this paper has done a lot of research based on the data of 6 years from 2011 to 2016 of more than 700 GPS stations around the world.The work is as follows:(1)Researched and designed a solution to the subnet resolution of the global network.Base on the relative accuracy of the baseline,the baseline repetition rate,the Nrms value of the baseline,the network adjustment accuracy and the accuracy of the coordinates of the external accuracy of the comparison and analysis of the two programs.It is found that the method of zoning by latitude and longitude is slightly better than the method of zoning by plate in accuracy and solving efficiency.Therefore,the data processing in this paper adopts a scheme for subnetting by latitude and longitude.(2)The global observation data were solved by single day solution and reached millimeter accuracy.Based on the study of sub-network partitioning schemes,each sub-network is baseline-resolved using GAMIT/GLOBK 10.6,and the entire network is subjected to joint network adjustment.The relative accuracy of the baseline reached 10-9 m;the Nrms value of the baseline solution was around 0.18;the average repetition rate of the baseline was 0.0028 m;the average a value of the network adjustment was below 6 mm,most of which was around 4 mm.Therefore,this solution can meet the research needs of this paper.(3)A high-precision 6-year GPS coordinate time series was constructed.Coarse-difference detection and elimination of the coordinate time series revealed that 160 stations did not meet the requirements and were eliminated.The excluded stations accounted for 22.9%of all stations and the remaining stations were 536.And multi-channel singular spectrum is used to iteratively interpolate the coordinate time series to obtain high-precision continuous GPS coordinate time series.Identify common-mode errors.This paper presents a method for identifying and extracting common-mode errors using Multi-channel Singular Spectrum Analysis(MSSA),and proposes a new identification criterion for common-mode errors.Through the study of the extracted common-mode errors,and compared with the recognition criteria of common-mode errors proposed by Dong(2006),it is found that the common-mode errors extracted under the two standards are relatively close.(4)A method for identifying common-mode errors using Multi-channel Singular Spectrum Analysis(MSSA)was proposed.At the same time,a new identification criterion for common-mode errors was proposed.The average value of the sum of the eigenvalues of the reconstructed signals that are identified as common-mode errors is 85.5%.(5)Studyed global crustal movements that take into account common-mode errors.Using maximum likelihood estimation,least squares spectrum analysis and least squares fitting method,the station noise,coordinate time series parameters and station speed were estimated.The effects of common-mode errors on station noise,parameter estimation and station velocity estimation were studied.The study found that,after removing the common-mode error,the noise characteristics of a few stations changed,and the noise level of most stations decreased significantly.The average white noise and colored noise levels in the N direction decreased by 31.61%and 42.28%,respectively.The white and colored noise levels in the E direction decreased by an average of 28.74%and 38.68%,respectively,and the white and colored noise levels in the E direction decreased by an average of 38.53%and 52.15%,respectively.At the same time,the uncertainty of the parameter estimation and the uncertainty of the station's velocity estimation have also been significantly reduced after rejecting the common-mode error.It can be concluded that after the common-mode errors are eliminated,the accuracy of the coordinate time series has been significantly improved.The coordinate time sequence after rejecting the common mode error was used to estimate the speed of the station,and the global earth's horizontal and vertical movements were analyzed.