Earth science has entered a new era. A new branch of science-earthquake geodesy, has been preliminarily formed after fifty years of exploration, with the development of modern geodesy such as space technology integrated with the theory of seismology, geology, rock mechanics and complex dynamical system. Basic progress and questions are discussed from the perspective of scientific thought and practical effect. The precise multi-scale time-spacial  monitoring system is set up to explore the new field of continental dynamics, and to get the unknown natural phenomenon. It is revealed that  crustal deformation system in China mainland  is a self-organizing dynamical system, earthquake is a transient behavior of self-adjusting for keeping long-term stability in the process of evolution.Earthquake has a predictable nature,but it is difficult to achieve fully deterministic prediction. It is  possible to predict some of the earthquakes  at present, although the long-term exploration is required.Some new ways, new models, new measures and new methods, advancing the innovation and getting out of difficulties for eurthquake prediction,  are suggested on the basis of innovative scientific thinking.

Using multi period fault-crossing leveling data from Minle in Gansu province to Menyuan in Qinhai province, crust vertical deformation is analyzed comparatively before the 1986 and 2016 Menyuan MS6.4 earthquakes. The results show that: 1） The seismogenic background before the two earthquakes differs. Before the 1986 Menyuan earthquake, Qilian Mountains rise relatively sharply in Minle basin and the regional compressive stress state is enhanced; however, before 2016 earthquake, Qilian Mountains rise relatively insignificantly and the regional compressive stress is relaxed. 2) The vertical deformation before the two Menyuan earthquakes also differs. Before the 1986 Menyuan earthquake, there is a rapid rise in the vicinity of the Lenglongling fault; while before 2016 earthquake, there is an unstable change between the rise and fall in the vicinity of the Lenglongling fault. 3) Surface uplift in the vicinity of seismic faults is observed before the two Menyuan MS6.4 earthquakes, which is consistent with the expansion model of seismic development.

Using observation data of GPS, gravity and across fault in divergence area, we analyze crustal deformation anomaly distribution and dynamic image characteristics in Sichuan province before and after the Lushan 7.0, Ludian 6.5 and Kangding 6.3 earthquakes. The results show: 1) before the earthquakes during 2013 and 2014 in divergence area, the crustal deformation data in Sichuan province presents group mid-term and short-term precursory anomalies in different degrees, including cross fault deformation that appears as a short-term anomaly in the form of direct precursors, and that earthquakes occurred within one month from the end of the anomaly, which has instructive significance in strong earthquake occurrence time forecasting; 2) a prominent short-term anomaly point located about 200 kilometers from the epicenter, earthquake location and anomaly belong to the same tectonic system; 3) strong earthquakes occurred in the relative gravity anomaly zone and in faults which are parallel to or intersect the fault,which has cross fault secondary anomaly amplitude, which has instructive significance in strong earthquake location forecasting. These several moderate-strong earthquakes show that, during 2013 and 2014 the south-eastern drive function of the Bayan Har Block acceleratedsensitive fault activity in the divergence area; in the process of accelerating, the stress accumulation area ruptured because of the instability, which further resulted in brittle rock burst, thus releasing long-term accumulated energy in strong earthquakes.

Using the rigid-linear elastic model to fit 2001-2004 GPS regional velocity field and strain rate field, we obtain 1-3 years precursory anomalies before the July 4th, 2006 Wen’an MS5.1  earthquake. The results show that the earthquake occurred in the anomaly area and the edge gradient transition zone of the high abnormal body.

We collect, sort, process and analyze velocity field data, obtained before and after April 25, 2015 MS8.1 Nepal earthquake, from the earthquake zone and the Qinghai-Tibet plateau areas of China. According to the pre-seismic velocity field, the Himalayan tectonic belt presents compression characteristics of about 16 mm/a. Furthermore, according to the deformation characteristics, the pre-seismic Himalayan tectonic belt can be divided into eastern, central and western sections. The earthquake occurred in the middle and the main squeeze is northerly. The eastern and western sections have clockwise and counterclockwise characteristics, respectively. After the earthquake, the velocity field shows that the tectonic deformation of the impact is greater in Nepal and Southern Tibet, mainly because the pushing speed of the Himalayan tectonic belt decreases, the Southern Tibet north-south movement decreases, and the east-west speed increases. This phenomenon may have great impact on strike-slip faults in the Southern Tibet area.

In this paper,we use the displacement time series for a period between 2 and 6 years after the 2011 Kunlun mountain MS8.1 earthquake.The deformation field is simulated by post-seismic viscoelastic relaxation,then we use the semi analytic plane delamination model to analyze the lithospheric rheological structure of northern Kunlun.The lower crust and upper mantle's viscoelastic flow mainly causes post-seismic relaxation.Models of viscoelastic stress relaxation in a Maxwell half-space gives a best fitting result.The best thickness of elastic upper crust is 15 km-20 km,the Moho depth is 70 km, the valid viscosity of lower crust ranges from 1×1019 Pas to 2.5×1019 Pas and the viscosity of the upper mantle ranges from 1×1019 Pas to 6.3×1019 Pas.The result is consistent with the inversion of post-seismic viscoelastic relaxation caused by Manyi earthquake,thus it can provide significant basis for understanding tectonic movement in the Qinghai-Tibet plateau.

In this study, a pair of SAR imagestaken before and after the Pishan earthquake acquired from the Sentinel-1A satellite are used to extract the coseismic deformation field. Based on the above coseismic deformation field and the Okada model, considering the varying incidence angle and unwrapping datum error, we inverse the fault slip distribution of this earthquake. The inversion results indicate that the Pishan earthquake is a thrust earthquake with smaller dip, the fault ruptures 5 km under the surface, and the maximum slip is about 0.76 m, and the main slip area is at the depth of 9-14 km. The estimated seismic moment is 5.14×10¹⁸, approximately equivalent to the magnitude of M_W6.47.

The Yadong-Gulu rift in the Tibet plateau plays a very important role intectonic evolution in the Cenozoic Era. Its examination helps to understand the role central and southern Tibet rifts played in the evolution of the Tibet Plateau in the Quaternary. Based on 26 ENVISAT synthetic aperture radar(SAR) scenes during the period from 2003 to 2008, we take the SBAS-InSAR technology to calculate the time-dependent LOS velocity field of Yadong-Gulu rift and the surrounding region, in which the errors induced by terrain, orbit data and the atmosphere model have been detected. The results of the slip rate, which is inverted by the LOS direction deformation, show that the southern and northern portionsof the rift have a similar dip slip rate of about 1 mm/a. The locking depth is 31km in the southern section and 22 km in the northern section. The inversion of fault dip angle in the southern and northern sections is 52° and 40° respectively.

To overcome the deficiency of the single forecasting model, an EEMD-PSOGSVM prediction model of foundation pit displacement is proposed, based on chaotic time series. The EEMD is adapted to decompose the time series, then phase space reconstruction technique is used to reconstruct the sample. The PSOGSVM model is then applied to predict. A comparative study of some deep foundation pit displacement is made by using the GM (1, 1), SVM and wavelet neural network optimized by genetic algorithm models, respectively. The results show that the predictive accuracy of this method is better and more stable, and that it can be effectively applied into the prediction of foundation pit displacement.

Gross error cannot be avoided as observations will be affected by environmental and some uncertain factors. In this paper, we introduce two adaptive factors based on a robust Kalman filter to adjust imprecise dynamic models and observation models which interfuse gross error. According to the low efficiency of the robust adaptive Kalman filter, we construct achi-squared distribution based on a robust Kalman filter. We test gross error by chi-squared distribution, then use a robust adaptive Kalman filter to process data in these algorithms.Experimental results show that the algorithm of dual adaptive factors filtering can resist gross error efficiently and weaken adverse effects due to the imprecise dynamic model. A robust Kalman filter based on chi-squared distribution can resist the effects of gross error and the convergence rate will also be improved.

This paper examines bridge deflection measurement. Against the characteristics of the point cloud at the bottom surface of the bridge, two methods: fitting calculation and center of gravity calculation,are used to obtain the deflection change of the bridge under different loadings. The results are compared and analyzed and the preliminary measuring precision of this technology is confirmed. Theexperiment also verifies the feasibility of this technology in bridge deflection measurement.

Launched on March 30,2016,the IGSO-6 satellite of the Beidou navigation system has new attitude control modes compared to in-orbit operation ofother IGSO satellites of the Beidou navigation system. In particular, during the earth’s eclipse, it keeps yaw-steering mode. In this paper, multiple observation data are used to analyze the orbit accuracy of the IGSO-6 satellite.The experimental results show that: during the earth’s eclipse, the IGSO-3 satellite overlap accuracy reach 0.31 m for average and 0.67 m for the largest, while the IGSO-6 satellite gets 0.18 m for average and 0.28 m for the largest; during the changing of control mode, SLR validation reaches 0.2 m for the IGSO-3 satellite and less than 0.1 m for the IGSO-6 satellite. Compared to the IGSO-3 satellite, which experiences the earth’s eclipse at about the same time period, the IGSO-6 satellite with new attitude control mode can effectively avoid the decrease in precision of orbit determination.

A combined formula of international terrestrial reference frame (ITRF) is deduced when new local tie vectors are introduced. This paper analyses the effects of new local tie vectors on coordinates and their velocities of ITRF2008. Solved results are compared with ITRF2008 when introducing Changchun, Kunming, and Urumqi local tie vectors and their variance-covariance matrices from the 2011 ITRF2008 crustal movement observation network of China (CMONOC). The results show that coordinate differences for the most of sites are within 10 mm, and velocity differences are less than 1 mm/a. Moreover, the horizontal velocity and vertical velocity fields of the most sites are consistent with ITRF2008 velocity fields.

In order to eliminate noise in the deformation sequence, the signal is decomposed into different scales using the CEEMD method. Aiming at the problem that the signal and noise distinguish criteria are not unique, a de-noising algorithm, based on the combination of CEEMD and auto correlation analysis, is proposed to separate the signals and random signals. The algorithm is applied to a simulation experiment and to GNSS deformation monitoring data, and compared with traditional wavelet de-noising methods. Compared with the wavelet method, the algorithm has a better effect.

In this paper, RTKLIB, gLAB and G-NUT open source precise point positioning software programs are used to process the 76 IGS stations by the use of static simulation of dynamic method in PPP models. The tropospheric delay and the position calculation results are compared with the reference coordinates and tropospheric product from IGS. The results show that standard deviation of troposphere calculated using different software can achieve cm-level accuracy, while RTKLIB can achieve STD better than 1 cm. In terms of coordinate accuracy, G-NUT needs longer convergence time than RTKLIB and gLAB to reach an absolute positioning accuracy. However, all three open-source software programs can achieve cm-level accuracy using kinematic positioning.Although the incomplete data preprocessing strategy in RTKLIB will cause a greater deviation in some stations, it can still get the highest accuracy; G-NUT has the same accuracy with RTKLIB horizontally, but the vertical resultsare poor. gLAB gets the largest deviations horizontally, but vertical accuracy is better than G-NUT.

The  Te map of Anhui and surrounding regions is calculated from EIGEN-6C4 gravity anomalies and ETOPO1 data using the isotropic fan wavelet method. The tectonic structure and geodynamic features of this area are discussed according to the  Te map. The results indicate that the Te of Anhui and its surrounding regions varies from 5 to 75 km. The  Te values of the Dabie orogenic belt and lower Yangtze region are low, as are the Te values along Tan-Lu fault. High  Te values appear at Hehuai basin and Subei basin. The  Te distribution mainly reflects the particular characteristics of tectonic structure and geodynamic features of corresponding areas.Most of the seismic activities located in regions with  Te values vary from 10 to 50 km; the spatial distribution of earthquakes indicates that earthquakes tend to occur in regions with low Te or steep Te gradients,further indicating that lithospheres of regions with low Te or steep Te gradients show higher seismicity rateand are not strong enough to resist deformation.

Based on the analysis of 998 groups of leveling data in 38 survey fields, distributed in 13 administrative divisions throughout Xinjiang, we come to the conclusion that the overall precision EGM2008 in Xinjiang is 0.146 m. We summarize the distribution regularities of height anomaly precision and changes varying with distance of EGM 2008, and propose that, within a certain scope, the EGM 2008 transformation results of GPS ellipsoidal height can replace fourth and lower grade leveling.

New Baytap-G tide analysis accessional software is designed based on C#.NET programming language. The software not only provides a simple and easy way to access through the visual user interface, but also implements the automatic format transformation and filling for the parameter, water level, and pressure data.

The paper compares four reference map building algorithms. The methods, evaluated using the parameters, integral mean square error, root mean squared error and the leave-one-out cross validation method, are verified in two scenarios. The results show that the co-Kriging interpolation algorithm has highest precision but is slowest, while standard Gaussian process regression has slightly lower precision but faster construction speed. Hence, co-Kriging algorithm is suitable for the high-precision system; while for the high real-time system, standard Gaussian process regression method suits best.

With seismic waveform data from 2001 to 2015 recorded by the Three Gorges digital earthquake network, the inelastic attenuation coefficients in different water impoundment phases are calculated. The well-fitting results suggest that the linear relationship between the Q value and the frequency is significant. In general, the Q0 value and η value at different impoundment stages are similar, and relatively not so large. This indicates that the tectonic structures in the study area are very complex and the frequency of the mirco-small earthquakes is high. In addition, the medium in this region is very heterogeneous. The Q value near the Xiannvshan-Jiuwanxi fault is large, which shows that the region is relatively tectonically stable. While the Q value in Badong region is small, it might denote that tectonic activity is stronger. For the Xietan region, the Q value becomes larger with the water level increasing, showing that the medium becomes more and more homogeneous.

We select concentrations of earthquakes of varying times, spaces and magnitudes, and calculate station magnitude bias and network-average values, based on digital waveform data recorded by regional seismic networks and relevant seismic phase data as collected by the China Earthquake Networks Center. We perform quantitative statistical analysis of the magnitude deviation of each station after eliminating effects due to source, path and direction. Considering different tectonic structures in Sichuan Basin, the western Sichuan plateau and the Panzhihua-Xichang area, the problem of local magnitude bias is discussed from the point of view of regional seismic wave attenuation characteristics, in order to render the magnitude correction value of station more reasonable. In our study, for stations in Sichuan basin and the south section of Longmenshan fault, greater magnitudes are functions of low attenuation in structure and amplification of both basin effects and seismic stations;these require negative correction. For stations in the northern and middle sections of Longmenshan fault, their lower magnitudes require positive correction. For stations in western Sichuan plateau, the possible causes of the lower magnitudes are severe dependence upon source region due to low Q value, path effect attenuation, and small magnitude estimated by weak S-wave. For stations in the Panzhihua-Xichang area, magnitude deviation varies with the location of epicenter;the greater magnitudevaluesoccurwhen earthquakes occur in Sichuan basin appear to be well correlated with edge effect of sedimentary basin on strong ground motion; these require negative correction. Furthermore, the ray path passes through the Sichuan plateau and Panzhihua-Xichang area, so that the two high attenuation areas may be the main cause of low magnitude values for plateau earthquakes, thus requiring positive correction.