Based on the 21st century earth science development strategy, through the perspective of the scientific methodology and scientific culture, several issues that may exist in the disciplinary development are rethought: an “engine” role of the observation theories and techniques; mining and understanding of the observation data; multiscale simulation of dynamic system; predicted promotion; disciplinary framework and scientific cognitive approaches.

The formula to express relationship between the surface horizontal, vertical deformation and fault rotation  is derived from a disclination model. In order to verify the validity of this new derived formulas, the surface deformation  in Longmenshan fault zone is computed by using the new derived formula. The computed results are compared with those computed by dislocation model which have same fault parameters as disclination model except slip components and rotations in the same region. The results manifest that：1） the surface deformation induced by fault rotation is as same order as that induced by fault slip in magnitude;2）The surface horizontal displacements due to fault rotation has similar space distribution characteristics with those due to fault slip, but the direction and magnitude of  the horizontal displacement induced by fault rotation  have a little difference  with those induced by fault slip in space contribution；3）The vertical displacements caused by fault rotation have smaller space area than those induced by fault slip, and the vertical displacement lie near two ends of the fault.

With the homogeneous elastic strain block model and the horizontal crustal movement velocity field obtained by GPS measurements, the movement parameters and strain parameters of the northeastern margin of the Qinghai-Tibet block are calculated. The uniform deformation of the blocks, and the relative boundary motion between the adjacent blocks including strikeslip and extrusion rate are calculated from the each block movement parameters and strain parameters. Combining with the tectonic characteristics and GPS velocity profile locating at Haiyuan-Liupan Arc, the present day tectonic activity are discussed.

On the basis of the GPS horizontal velocity field data in the period of 1999-2007, using the block model and GPS velocity profiles across faults and combining with the distribution of the strain rate field near it.we analyzed the present-day tectonic deformation characteristics of the Haiyuanliupanshan fault zone from its segments and the entirety.The results show that the characteristics of the Tuolaishan fault and the Lenglongling fault is strike-slip with compression; The deformation of the Jinqianghe fault is the most distinct, and its shear strain accumulation is the highest. The strain accumulation on the Maomaoshan segment is larger than that of the Laohushan segment.The left slip rate of the special Haiyuan fault is 5 mm/a, and this fault is still creeping and adjusting since the Haiyuan earthquake occurrence. The compressive velocity across northern segment of the liupanshan fault is 2.4 mm/a, whose compressive strain accumulation is obvious. However the deformation of middle and south segment of the liupanshan fault is not distinct. The strike slip with compressionstrike slipcompression transformation pattern presentes themselevs in the entire fault system from west to east.

By using 38 ENVISAT ASAR images between 2003-10—2009-06 of TianJin area, we conduct the differential interferometry and PS pixel selection experiment, and then extract the subsidence of experimental area. According to the time and space characteristics, using space and time filter to separate the atmospheric delay, orbit error,DEM residual phase,we obtained average settlement rate of test site, and finally selected PS pixels in four typical regional mapped the time sequence settlement diagram.

Statistical analysis on the earthquakes with magnitude greater than 2.0 in the Three Gorges area shows that the tectonic earthquake and collapsed earthquake have different corner frequency. The majority of the former is above 5.0Hz, and that of the latter is comparatively lower, around 2.5～4.5Hz. It can be taken as a identification basis, whereas, in real application, we should make comprehensive analysis of many other parameters. The collapsed earthquakes mainly occurred in the region with many mines and Karst, such as Xietan, Zigui county, Huofeng, Badong and so on. And the tectonic ones are prone to occur along the faults.

The largest magnitude is M_L3.8 small earthquake swarm which occurred on October 30, 2012, Quyuan town of Zigui County in the Three Gorges reservoir region. The studies show that the small earthquake swarm are directly related to the Three Gorges Reservoir impounding; the earthquake swarm gathered in the northern end of Xiannvshan fault zone, earthquakes most located within 10 km near the crystalline basement; linear scale of hypocenter body is smaller, mainly limited radius of 5 km of range; mainshock and larger seismic performance typical stereotype earthquake characteristics; earthquake swarm triggered the fault structure Xiannvshan stress to adjust in the high water level of the reservoir impounding, small earthquake swarm are reservoir triggered tectonic earthquakes.

The available studies have shown that the three factors:reservoir water level elevation, water level change rate and specific water level elevation can affect the reservoir seismic activity. In order to comprehensively discuss the three factors, we introduce the fuzzy mathematical methods, define the membership function of the reservoir water level and consider the  fluid diffusion mechanism hypothesis of reservoir earthquake, and then discuss how to select a subordinate function threshold, and how long that the water will affect the seismic activity. On this basis, a quantitative study method for  the relation between the  water level changes and seismic activity is given. With the method we studied the relation between earthquakes and the water level of three reservoirs in  Shanxi Zhejiang, Shuikou Fujian, Xinfengjiang Guangdong.The results show that the initial water level of
reservoir  is the main external factor affecting seismic activity in the reservoir area, but the impact  time of three reservoirs are different that in Xinfengjiang Reservoir is about six years, the Shuikou Reservoir about five years, Shanxi Reservoir about 4 years.

Three classical algorithms for parameter adjustment： Lemke’s method, potential function method and iterative multiplier method,based on K-T conditions are discussed, and some useful conclusions from the results of the case study which applied the three algorithms are given.

The deformation monitoring network adjustment method with considering the prior information are the paper researched. The functional model is given and the optimum solution formulas are deduced. Whether the prior information is considered or not, the parameter estimation formulas are the same. The estimation accuracy and applicable condition analysis were also carried out. The study reveals the relationships among prior accuracy, the quality of virtual observations and the adjustment accuracy. The calculation example of deformation monitoring network adjustment shows that considering the prior information can improve the accuracy of the parameter estimation

Taking elevation component time series of GNSS permanent stations as examples, effects of data defect and noise level on results of GNSS time series analyzing   are discussed based on both simulation and observation data. The results show that the higher the noise level is, the greater the effect will be. If the days of data defect are the same, the effect of continuous data defect is much greater than the effect of discrete data defect. The interpolation can’t remedy the data defect effectively in a high noise level if days of data defect are more than a certain number of days. On the basis, the request of observation data in different reliability requirements is discussed. It is pointed out  when annual and shorter periodic variations are considered and to ensure deviation of linear velocity be less than 10%, for example, when the noise lever is 9mm, then the defect days for continuous data losing should be less than 11 days and for discrete data losing, the defect days should be less than 12 days. Last but not least, an advice is given that in the study of vertical crustal movement using elevation component time series of GNSS CORS, the operation status of GNSS permanent station should be strictly screened.

A combined GPS/GLONASS precise point positioning (PPP) approach is proposed to estimate zenith tropospheric delay (ZTD),and the results are compared with those from GPSonly PPP. The
new ZTD products of IGS are used as references to assess the estimated ZTD. The numerical computations are conducted using IGS data collected at three Chinese stations within one GPS week. The results show that the different settings of elevation angle cutoff and spectral density value of tropospheric delay parameter have a large impact on the ZTD estimates. Compared with GPS-only PPP, the ZTD estimates obtained from the combined GPS/GLONASS PPP are more consistent with IGS products.

The relations between the accuracy of COMPASS pseudorange observations and elevation angles of satellite are studied, and the pseudorange point positioning accuracy of the constellation which contains 4 Geostationary Orbit（GEO）satellites and 5 Inclined Geo-synchronization Orbit（IGSO）satellites is analyzed. The results demonstrate multi-path effects of COMPASS pseudorange and observation noise are about 0-3 m, and their accuracies are about 0.3-0.7 m. At the higher elevation angles, multi-path effects and observation noise
are the smaller, and both of the multipath effects and observation noise on all GEO satellites are smaller than those on IGSO satellites. The COMPASS pseudorange point positioning accuracy（RMS）in the N, E and U directions are about 3.33, 3.45 and 8.84 m, respectively.

Using the data of GPS continual operational reference system to retrieve the precipitable water vapor. According to the heavy rainfall from 24th to 26th July in 2011,the relationship between the variation characteristic of multistations’ precipitable water vapor (PWV) and heavy rainfall was analyzed. And the relationship between the variation of precipitable water vapor and moisture transportation was also analyzed. By means of analysis of time series of GPS data from multistations, the paper obtains change characteristic of three phases of PWV during heavy rainfall period. Before the heavy rainfall period coming, PWV has a rapid rise. The length, extent and magnitude of the rapid rise reflect accumulation of water vapor, it has a better corresponding relationship with precipitation process during heavy rainfall period; and the process that the rapid rise of PWV reaches the peak has concomitant obvious precipitation, after entering high fluctuation phase, the precipitation holds on, in the course of PWV descends, the precipitation steps down, up to stop. In the phase, the fluctuation of time series of PWV and spatial distribution have a better corresponding relationship with shortterm change of precipitation during heavy rainfall period; reposefully diminishing of change amplitude and average magnitude of PWV reflect that the heavy rainfall period is about to end. According to the comparison among the precipitable water vapor of five stations, the variation of GPS precipitable water vapor is corresponds to the moisture transport from the southwest to the northeast.

Taking the GPS fiducial stations, BJFS and BJSH as the examples, the time series in elevation direction from 2000 to 2006 were decomposed into a finite number of intrinsic mode function，whose frequency  was decreasing, and the residual representing the trend of the time series. The linear rates of the two stations were -0.533 mm/a, -2.083 mm/a respectively. Furthermore every component is transformed with Hilbert algorithm, though the procedure twodimensional marginal spectrum and three-dimensional Hilbert spectrum in time-frequency-energy space were obtained and analyzed. The marginal spectrum has perfect timefrequency concentration; and then the frequency of high amplitudes is no more than 3 cycle/a. The main components in time series of the both stations are a secular nonlinear trend，and semiannual signals, and annual signals which is obvious and about 351 days. Moreover, Hilbert spectrum has high resolution and the ability of outlier detection. Afterwards the time series was processed by EMD lowpass filtering and the result is compared to that of sym6, db8 wavelet. De-noising results by EMD was much superior, slick and no glitch, which indicates that this method is more efficient in eliminating instantaneous strong noise

In order to further improve the accuracy of regional tropospheric delay model without meteorological parameters, a new model is proposed and taking the GPS data of Hong Kong CORS as an example. The modeling process  has two main steps, the first is to analyze the periodical characters of wet and dry components tropospheric delay and the factors which affect  the wet and dry components respectively, the second is to model based on the correlation between tropospheric delay and spacetime. The test result shows that the new model is more precise than the UNB3.

The precise reduced dynamic and kinematic orbits for GOCE satellite have been determined with Position And Navigation Data Analyst (PANDA) software for period of DOY 306 to 332 of 2009.The orbit accuracy is assessed with the following four methods: the differences between the reduced-dynamic orbits and the Precise Science Orbit (PSO) computed by Astronomical Institute, University of Berne (AIUB), the differences between the reduced-dynamic orbits and the kinematic orbits, the 6 hours overlap orbit differences between the reduced-dynamic orbits and the external orbit validation with SLR measurements. The results indicate that the accuracy of the reduced-dynamic orbit is about 1~2 cm in three directions respectively

Usually,when we correct the InSAR atmosphere effect based on MODIS water vapor data,the data may be polluted by the cloud.In order to improve the correction,we compared three kinds of interpolation methods:inverse distance weighted interpolation, Kriging and radial basis function. It is concluded that the Kriging method is best.The Kriging method was  used in Harbin atmospheric delay correct test. The experimental results show that the RMS differences between InSAR and GPS  reduced by 31.25% after the correction.

Based on the GPS observations and sitespecific pressure observations of the Hong Kong satellite positioning reference station, the effect of pressure parameters on estimation of ZTD and ZHD were analyzed. The result shown, in the practical operation of the GPS/MET, the estimate of the ZTD parameters are less depended on the accuracy of the pressure, you can even use the GPT model to get these pressure parameters. However, when estimating the ZHD, it is required to use the pressure observations in high accuracy. For those stations that the GPS receiver and the pressure sensor are not located in the same position, the correction due to the height should be made.

To improve the prediction accuracy of satellite clock error, least squares support vector machines（LS-SVM） is employed. The impact of the kernel function type on LS-SVM is analyzed. Furthermore, the prediction accuracy is compared with that of the secondary are polynomial and grey system model. The results show that the LS-SVM method has higher accuracy than two other methods, and the linear kernel function is better than others for the method.

The basic principle about satellite orbit determination is described and the satellite orbit determination classes based on the ObjectOriented method and C++ program language is designed and programed. Aiming at the enormous calculation about orbit determination, parallel algorithm and code optimized method based on VS2010 PPL and the new standard of C++ named C++0x are proposed. The experiment indicates, compared to serial algorithm, parallel algorithm and other code optimized methods executing on the dual cores’ computer can take full advantage of the hardware and improve the computing efficiency.

In order to improve GNSS integrity, a new integrity monitoring method of GNSS aided by INS with considering the phase Doppler smoothing code is proposed. In this algorithm, by the use of phase Doppler smoothing code to reduce the random error, the parity vector can be obtained by using measurement derived from INS and GNSS for fault detection and isolation. The simulative results show that the integrity monitoring method of GNSS aided by INS can increase the fault detection rate, when considering the phase Doppler smoothing code  the performance improvement is more significant.

In order to calculate the transformation of Cartesian coordinates, we need to know the two sets of coordinates of the common points, select different common points to establish the corresponding error equation and obtain the 7parameters or 13parameters of the transformation. We made the research on comparing estimated points calculated by transformation parameters with designed points in outerspace and innerspace around these common points through the simulation of 165 survey points in 11 rows and 15 columns of two sets coordinates. Taking the Euclid distance of the two sets of points as the basic surface fitting data to establish the surface model for calculating accuracy and with modified Shepard’s method, we have abtained  some general distribution law about the accuracy of calculating coordinate points.

The accuracy of different kernels are compared with each other by using the gravity data along the offshore of American sea in DNSC08GRA model. As the results show, in our test region, the modified Stokes kernels can improve the accuracy by 3.4 cm compared with the spherical Stokes kernel, while the difference between different modified kernels is tiny. When verifying the results with GPS/leveling data, it demonstrates the existence of a 1.17 m systematic  variation between the geoid fitting before and after. Besides, all the computing accuracy can improved by 7 cm after fitting.

In the process for researching threedimensional coordination transformation adapted to arbitrary rotation angle, with the case of only three control points , an improved constrained weighted total least squares method(ICWTLS) was proposed to solve the model of control points in the two sets having random errors (EIV model).If the control points are rich (four or more),we put forward the multivariate total least squares method (observation equation has not only one row,MTLS) to solve the parameters, and the problem of the repeated random elements of  coefficient matrix’s corrections being not equal,and to improve the accuracy of the parameter values.

Using the Chapman function and the ionosphere inversion methods and processes combined with the spherical harmonic function. The ionospheric electron density is calculated by using the data measured from the GPS CORS in the Chinese region under the static subsolar. The delays through the ionosphere calculated are compared with the known data and the RMSE of residuals is 0.311 m. The ionospheric electron density variations of the vertical structure and the certain height level of ionosphere are also analyzed. The results show that from the residuals analysis and the electron density trend, the accuracy of the electron density calculated with the method is high and it can be more intuitive to reflect the consistency of the electron density changes between different layers of ionosphere.

The airborne LiDAR technology provides a new and rapid way for detecting 3D building. In order to separate building points from mass LiDAR point clouds, a progressive densifying TIN method is first used to filter terrain points, and then top surface points of building are classified with principal component and normal vector. After initial building facets are segmented with region growing method, some fence regions are removed and small holes are filled by using morphological closing operator. Ultimately, small vegetation regions are erased on the basis of area criterion of building. Three sampled complex urban areas are selected to test extraction effects of building. The results show that the efficiency and quality of building defection with this method are better than 92%.

Aimed at the shortcomings of the Poisson iteration and polar coordinate template methods, an improved method of Poisson integration for discrete grid data is presented, which is proved efficient for improving the accuracy of integration for discrete grid data.

To solve the problem for the current measurement data processing in the computation of largescale network adjustment, a parallel computing strategy for the massive geodetic network based on the model of partitioned adjustment is analyzed and  quick adjustment solution service programs which are grounded on Web Services（WS）technology are designed. Combined with efficient numerical parallel computing algorithms and data compression algorithms, the adjustment subcalculating Web services under .NET platform is set up and the validity of the adjustment parallelization is demonstrated by some examples. The results show that the WS technology can significantly improve the efficiency of adjustment calculation and the existing resource utilization.

The optimum weight ratio of different observation values should be obtained in combined GPSBD2 positioning since there is a systematic difference between GPS and BD2. The variance component estimation（VCE） formula is used to determine the varianceratios of GPS and BD2 observations in the pseudorange single point positioning and phase relative positioning. The statistics of RMS of positioning errors with different weight ratios indicate that the results of VCE are reasonable and the prior weight ratios of GPSBD2 pseudorange and phase observations can be 2∶1 and 1∶1 respectively.

In order to assess the system performance, the GPS constellation and the moving trajectory of receiver are simulated, a 11dimensional state PositionVelocityAcceleration (PVA) navigation model is used for analysis. The EKF and EKPF model are separately adopted to obtain the navigation solution, the simulated results show that both the two models are satisfactory, while the navigation performance of the proposed EKPF algorithm is superior to EKF, especially in vertical direction, and impact of multipath effect has been reduced.

Code Minus Carrier method can be used to estimate the primary multipath error in a certain accuracy level. Howerer there is still strong noise in the esitmated results of multipath. In order to solve the problem, Least Mean Squares（LMS） method is proposed for filtering the primary multipath error by setting appropriate filter length and stepsize parameter. Taking the advantage of repeating feature, LMS method can separate real multipath error from the estimated serial effectively. The numerical example shows that multipath effect can be reduced significantly and the percentage of multipath error correction can reach more than 80%, which will provide accurate model correction for navigation and positioning of user.

The wavelet transform theory and  principle of wavelet denoising are discussed.Forthermore three different denoising method  for GPS dynamic monitoring data denoising are analyzed. The experimental results show that the wavelet transform especial soft threshold denoising method is an effective method for the GPS dynamic monitoring data processing.It can improve the accuracy of monitoring data.

The distribution of GPS slant observations taken by continuously operating Yangtze River Delta GPS network this work is described . The distribution of  the grid coverage of slant data at different temporal and spatial resolutions are given. From this,the feasibility of 3D water vapor tomography in a test region is discussed. At last, a set of additional instruments in GPS network scheme for water vapor tomography is researched by using simulation method with ECMWF data. It is concluded that the bad geographic distribution of GPS network in Yangtze River Delta causes difficulties in tomography, and the overall accuracy can be improved by adding some stations, especially in lower atmosphere.

Aim to determine the mathematical model of  directional error data of maglev gyro total station. The time series processing method was used to analyze the initial current data of stator observed by the maglev total station, all data are used to calculate directional angle, which belong to unstable time series process. At different environmental conditions, the random sequence data showed a periodic, trend, conditional variance and singular value features. In this paper,the difference technique is used to extract the trend error and BoxJenkins method is used for modeling the stable stochastic process. According to the differential data characteristics, we can use the different model to predict the future monitoring values, to estimate and control the current data which used to calculate the directional angle, in order to improve the accuracy of directional angle.