| GPS technology is gaining more and more attention by the majority of scholars, becauseit has some advantages such as high precision, large scale, all-weather, real-time and so on.Especially, with the establishment of a large number of GPS continuous observation stationsof various countries, as well as the improvement of the data processing accuracy, GPS hasbeen widely applied in the fields of earth dynamics, seismology, geophysics, geology, and soon.After years of GPS continuous observation, it has accumulated abundant observationaldata, conducived to crustal deformation analysis. Usually, the method of extracting crustaldeformation information by using GPS continuous observation data is as follows: firstly,process and analyze the GPS continuous observation data precisely, and obtain the coordinatetime series of the stations, and then get the velocity fields of the study area with high accuracy.On the basis of velocity fields got above, further obtain and analyze the dynamic features suchas earth crust displacement, stress and strain fields to explore the dynamic evolution modeland its tectonic movement.Therefore, how to recognize and separate a variety of information in the GPS time series(such as the trend, periodic, error and noise), improve the accuracy and reliability of themodel parameters, so as to improve the precision of GPS time series are very important to thedeformation analysis. It is the basis for recognition and understanding of regional motion anddeformation. But beyond that, selecting the appropriate method to calculate the regionalcrustal strain characteristic parameters is also the key to explain regional crustal deformationcharacteristics reasonably.Based on the above ideas, this paper is divided into two parts: one is how to obtain thehigh precision GPS velocity field. In this part, we discuss the model and method of GPS dataprocessing with high precision, and the influence of non tectonic factors in the GPScoordinatetime series. In the second part, we discuss the method for obtaining the straincharacteristics of the study area, based on the obtained regional crustal horizontal velocity.The main researches and results are as follows: (1) Firstly, we discuss the methods and strategies of processing the GPS data with highprecision, and then analyze the influence of non tectonic factors in the GPS time series, whichinvolves principal component analysis (PCA), data interpolation, spectrum analysis and thequality load correction method based on the geophysical model.(2) Characteristics of daily position time series which dated from2006to2011in SouthCentral of America GPS fiducial network are researched in this paper. A spatial filteringalgorithm based on principal component analysis is employed to extract and remove thecommon mode errors from the coordinate time series. This method promotes the accuracy andreliability of the sites coordinate. Further, we discuss the differences influence in velocityfield of GPS stations to wipe off the common mode errors. The result shows that the commonmode errors cannot be neglected in time series analysis. Particularly when we deal with themicro deformation, the PCA method can get rid of the common mode errors effectively andimprove the reliability of result.(3) For the GPS stations coordinate time series in South Central of America area, wecorrect the quality of load (air, snow, soil water) based on the geophysical model. Theexperiments show that it works better than harmonic fitting correction, and it can effectivelyimprove the GPS positioning accuracy of the vertical. In addition, through the superimposedtime series of mass load by sequence spectrum analysis, the elevation (U) direction is largerthan the plane (N, E) direction. And it has an impact on the annual cycle.(4) For the typical regions-North America plate boundary (PBO) region, we acquire andanalyze the deformation information. On the basis of high precision velocity fields, use theDelaunay triangle method and spherical wavelet method respectively to get the straincharacteristic parameters of the study area. Finally, we compare the research findings of thispaper with the related research abroad, to verify the reliability of the spherical wavelet method.Because the wavelet provides amulti-scale estimation, it can select the corresponding framefunction according to the density of the stations to estimate speed. Therefore, it can be used inthe estimation of velocity in the area which is large and the station distribution is uneven, sothat reflect the strain situation of different regions more truly.(5) For the latest data of the Crustal Movement Observation Network of China from2011to2013, we obtain the velocity field under the framework of ITRF and relative to the Eurasian plate with fine processing. We analyze the current characteristics of crustalhorizontal movement of China.(6) The spherical wavelet method is used to calculate the parameters of regional straincharacteristics of China for the first time. The results show that the spherical wavelet methodcan adapt to the change of sites densities, and multi-scale estimation velocity field. Except asmall number of sites, the estimation error is less than2mm/a between the measured velocityfield and the estimated velocity field. So the estimated velocity field by the spherical waveletand the measured velocity field are in good consistency. From the expansion rate and strainrate of the Crustal Movement Observation Network of China that we got by using thespherical wavelet method, the structures are strongly deformed and the strain rate is high inthe western China. It is divided by the tectonically active zones in the northern and southernsides, the intensity of east strain rate is much weaker than the west, the western high value isthe most significant in the Qinghai-tibet block and its edge, Sichuan-Yunnan region andwestern Xinjiang area. |
PDF Full Text Request