We analyze and process the whole time observation data, from January 1, 2015 to May 20, 2021 of the water tube instrument and the extensometer within 200 km of the epicenter of the May 21, 2021 M6.4 Yangbi, Yunnan earthquake. We find that from 2019, the observation curves of the two sets of instruments at each station show trend change or the annual change amplitude decreases at the same time. There is a certain relationship between the time of anomaly occurrence and epicentral distance. The time of anomaly occurrence at stations near the epicenter is earlier than that at stations far away from the epicenter.

To study the distribution characteristics of the strain field before the earthquake, we conduct deep processing of the station displacement time series of 43 continuous GNSS stations in Yunnan from January 2012 to December 2018, and take 7 earthquakes with M≥5.7 as samples to obtain the surface strain and maximum shear strain distribution before earthquakes. The results show that: 1) Before the 7 earthquakes with M≥5.7, there is a significant surface strain change trend in Yunnan. In most cases, there is a pattern of strong compression and strong tension coexisting, and the cumulative surface strain generally exceeds ±4.0×10^-8. The earthquakes mostly occur in high-gradient belts where surface strain (especially the surface squeeze) changes. 2) Before the 7 earthquakes with M≥5.7, there is a significant trend in the maximum shear strain change in Yunnan, and the cumulative maximum shear strain generally exceeds 5.0×10^-8. The earthquakes mostly occur in the high gradient zone with the maximum shear strain change. 3) The surface strain (especially the surface squeeze) and the high gradient zone of the maximum shear strain change can be used as important areas for the research and judgment of the earthquake location of M≥5.7 earthquakes in the future.

To study the influence of the surrounding surface dynamic displacements caused by Madoi earthquake, using ultra long baseline modewe process 11 high rate GPS observations within 400 km of the epicenter and then analyze the time series of the horizontal displacements. The results indicate that the Madoi earthquake caused a displacement of about 25 cm at QHMD station on the eastward and did not recover to the preseismic status. On the northward, it caused about 27 cm at QHMQ and GSMA stations but recovered to preseismic status quickly. The influence of the earthquake on the dynamic displacements of the surrounding stations, especially on the northward displacement, has a high correlation with the relative azimuth of the stations to the epicenter.

We review some examples of the application of multi-means of crust deformation monitoring to medium and long term earthquake prediction and short and medium term earthquake prediction, and summarize some examples of outburst deformation anomalies and large deformation anomalies not corresponding to the earthquake. We demonstrate the validity and limitation of crust deformation method for strong earthquake prediction; we then express views and suggestions based on the discussion about some problems found in practice.

Using ALOS-2 and Sentinel-1 satellites SAR images based on the InSAR technology, we extract the coseismic surface deformation fields of the four M_W>6.0 earthquakes that occurred between October to December of 2019 in Mindanao Island, Philippines. Then, we use the InSAR data to estimate the coseismic faulting model of these four earthquakes. The inversion result shows that the rupture of three individual faults are responsible for these four earthquakes. The dextral strike rupture of two adjacent segments of the same southeast dipping fault caused the two earthquakes occurred on 2019-10-16 and 2019-10-31, respectively. The maximum slip magnitude of the 2019-10-31 earthquake is about 1.1 m, which is two times of that of the 2019-10-16 earthquake. Another fault for the earthquake occurred on 2019-10-29, and the maximum slip magnitude is about 2.0 m. An additional fault characterized by predominant sinistral strike slip is the seismogenic fault of the 2019-12-15 earthquake, with a maximum slip magnitude of about 3.0 m. The calculation of the Coulomb failure stress (CFS) change indicates that the 2019-10-16 earthquake has caused significant positive CFS change in the high-slip area of the 2019-10-29 earthquake. Moreover, the 2019-10-29 earthquake also increased the CFS in the hypocenter area of the 2019-10-31 earthquake. Finally, the cumulated CFS change on the seismogenic fault of the 2019-12-15 earthquake due to the previous three earthquakes is less than zero, which suggests that the static CFS changes may be not the main triggering factor of the 2019-12-15 earthquake.

We propose an optimized land subsidence prediction method combining empirical wavelet Transform(EWT) and Prophet prediction model to solve the problems of low prediction accuracy of single traditional method and unstable prediction process. Taking the Dexing mining area in Shangrao City, Jiangxi Province, as an example, using the 30-view sentinel No.1 image, we carry out the SBAS-InSAR subsidence study and obtain the time series data of the subsidence of the area in the study time span. This method first performs EWT adaptive decomposition on the original sedimentation time series data, decomposes to produce empirical scale components and a series of empirical wavelet components, performs Prophet prediction and superimposition reconstruction on each component to obtain the final predicted settlement data, and finally adopts retrospective prediction to verify the prediction accuracy and reliability of the proposed method. Experimental results show that the EWT-Prophet combined model is overall better than the single Prophet model and the traditional ARMA model. Compared with the other two methods, the EWT-Prophet model improves the root mean square error by 51.53% and 59.03% respectively and the average percentage error is increased by 57.81% and 64.85% respectively, indicating that the prediction effect of this method is better, has better applicability, and provides an effective method for the prediction of large-area mining area settlement.

Aiming at the limitations of wavelet neural network, we use particle swarm algorithm to optimize the wavelet neural network. On this basis, a fitting model of GPS elevation abnormality is established. In order to prevent the problems of the particle swarm algorithm from falling into local minima and slow convergence, the particle swarm algorithm is improved by using a strategy combining inertia weight non-linear decreasing and adaptive learning factor, so as to improve the training accuracy of the model.Taking the measured GPS data of a mining area as an example, we verify the fitting performance of the model. The results show that the improved wavelet neural network model has higher accuracy and stability in GPS height fitting.

Aiming at the problems of gross errors in the GNSS elevation coordinate time series, instability in the elevation coordinate time series, and difficulty in detecting outliers caused by nonlinearity, we establish a gross error detection method combining SSA and S_n estimator with singular spectrum analysis. Using experimental simulation data and measured data respectively, we compare the detection effect of this method with the conventional 3σ and MAD methods. The results show that the overall detection rate of the SSA-S_n gross error detection method is better and that it satisfies the demand for high-reliability GNSS coordinate time series.

The complex nonlinear characteristics contained in the length-of-day variations parameter sequence seriously affects prediction accuracy.In order to explore whether the introduction of atmospheric angular momentum sequence can help improve the prediction accuracy, this paper proposes a Prophet fitting extrapolation joint vector autoregression residual compensation combined model to predict length-of-day variations. The sequence between 2008 and 2020 is selected for predicting experiments. At the same time, two schemes of Prophet-AR and traditional LS-AR, which ignore the atmospheric angular momentum sequence, are designed for comparison. The results show that the prediction accuracy of the three schemes decreases successively, which shows that Prophet algorithm can better fit the nonlinear signal than the LS algorithm to reduce the prediction error of the combined model. It also shows that the introduction of the atmospheric angular momentum sequence can effectively improve the prediction accuracy when the prediction model is consistent. It is comprehensively shown that the Prophet-VAR combined prediction model, which takes into account the atmosphere angular momentum, can be applied to high-precision prediction of length-of-day variations.

In order to study the high-order ionospheric delay and its influence, we use actual measurement data of IGS combined with specific experiments to discuss the geomagnetic model, TEC data model, GPS frequency and station distribution and other factors. We analyze the delay of high-order terms of the ionosphere and their influence on the baseline under different TEC acquisition methods and geomagnetic model processing modes. Experimental results show that the TEC obtained by different methods is a key factor affecting high-order ionospheric delay, its influence on carrier phase observations is up to 10.0 mm, and its impact on baseline is even up to cm level. Geomagnetic field is a secondary factor of high-order ionospheric delay. The influence on carrier phase and baseline is at the mm level; high-order ionospheric delay is related to the station area, showing systematic deviation in mid-latitudes, while high-order ionospheric delay near the equator shows occasional characteristics, and its main influence on the baseline is related to the distance and azimuth of the baseline.

We are concerned with problems of poor positioning accuracy, reliability and integrity in a single navigation and positioning method in an urban environment; we propose a PPP/INS/ODO integrated navigation scheme without ionospheric combination of GPS+BDS-3. Using the GPS+BDS-3 dual system, we derive the non-ionization combined function model. We give the combined model of PPP/INS/ODO and analyze the processing strategy of time asynchronous and lever arm. Combining odometer information and non-integrity constraint information further improves the positioning performance of the combined system. Using static and dynamic experimental data, we analyze the positioning performance of the GPS+BDS-3 non-ionization combined PPP and the positioning performance of the PPP/INS/ODO combined system, respectively. Experimental results show that the GPS and BDS-3 combination has significantly improved positioning accuracy compared to a single system; the PPP/INS/ODO combination system has a position RMSE better than 1 m in a complex environment, which can provide technical support for car navigation in a complex environment.

Aiming at the complex characteristics of the inertial navigation system carried by the underwater vehicle in the coning motion environment, we study and analyze the classic multi-sample algorithm, optimal multi-sample algorithm, uncompressconing compensation algorithm, and iterative algorithm based on quaternion differential equations. Using simulation experiments with the cone motion as the test input, we verify the accuracy and applicability of the attitude solution. The results show that the use of more than 3 sub-sample iterative algorithms for underwater gravity-assisted inertial navigation attitude calculation can meet the accuracy requirement of angular drift less than 1″/h.

GNSS positioning algorithm based on Doppler shift reconstruction solves the problem of abnormal observation signal by reconstructing Doppler shift. Combined with the cross difference weighting method based on vertical elevation selection, we solve the integer ambiguity under the condition of poor observation signal quality, and obtain a high correct estimation rate of integer ambiguity. The results show that the algorithm can achieve 100% ambiguity fixed rate and cm accuracy in the static case of low satellite redundancy. In the dynamic case, the accuracies of E,N and U directions are 0.679 0 m, 0.052 9 m and 0.022 8 m respectively, which are the same as the traditional RTK Positioning, and the fixed ambiguity is increased by 26.38 percentage points.

We evaluate the precision of broadcast ephemeris of BDS-2 and BDS-3 using precise ephemeris products. We further compare and analyze the influence of different BDS-2 and BDS-3 constellation combinations on single point positioning(SPP). We propose a SPP weighted optimization model based on SISRE(signal in space range error). The experimental results show that the broadcast ephemeris orbit errors of MEO, IGSO and GEO satellites in BDS-2 constellation are 2.404 m, 3.030 m and 12.574 m respectively. The broadcast ephemeris orbit errors of MEO and IGSO satellites in BDS-3 are about 0.5 m and 0.8 m respectively. For SISRE, BDS-2 is on average better than 2 m and BDS-3 is on average better than 1 m. BDS-3 satellite clock error and stability with H clock and Rb clock are basically the same. At the same time, SPP results show that BDS-2/BDS-3 IGSO/MEO constellation combination has the highest positioning accuracy. The SISRE weighted model is used to solve the SPP, and the optimized positioning accuracy is 9.61%, 18.55%, and 11.19% respectively in the N,E,U directions, proving the effectiveness of the model.

The temperature of GNSS monument and the bedrock can be influenced by surface air temperature changes, which results in thermoelastic deformation and accumulates as displacement of GNSS station. The quantity analysis on thermoelastic effect is important in separating accurately the signals and noises in coordinate time series of GNSS reference stations. In this paper, based on correlation theory of elastic mechanics and thermoelastic mechanics, we give the analytical solution thermoelastic deformation of axisymmetric monument under one-dimensional temperature field. Results suggest that the vertical displacement of monument is dependent on the Poisson's ratio, thermal expansion coefficient, temperature difference and the height of the monument and bedrock. We propose a more rigorous thermoelastic effect model, with the vertical displacement of GNSS station caused by temperature changes being considered in three parts: the displacement of aboveground part of the monument, underground part of that, and the bedrock. We compare the difference between the proposed model and literature results.

We use the harmonic amplitude ratio method to analyze the fluxgate second sampling observation data in Qinghai area. On this basis, we apply the rate accumulation method to analyze the daily change rate of continuous data. Then, according to the need to identify the abnormal changes of data, we carry out the accumulation calculation of the analysis period. We further analyze the relationship between the abnormal change and strong earthquake, and summarize the characteristics of harmonic amplitude ratio of fluxgate second value in Qinghai region so as to complete the process from qualitative to quantitative recognition of anomalies. The results shows that the curves of the geomagnetism harmonic amplitude ratio after rate accumulation processing are continuously rising to normal, while the curves are continuously decreasing and the horizontal fluctuation are abnormal. When the duration of the anomaly is more than 1 a and the number of anomalous curves is more than half, the probability of M≥5 earthquakes around the station is high, and the curve period mainly ranges from 20 to 170 s. The magnitude is proportional to the epicenter distance, that is, the higher the earthquake magnitude, the greater the epicenter distance.

We use the time-varying gravity field model, the GLDAS hydrological model and lake water volume change data to invert the changes in groundwater storage in the Three-River headwater region from 2003 to 2020. We extract the trend and periodic changes of groundwater storage through singular spectrum analysis. The result shows that from 2003 to 2012, the groundwater storage in the Three-River headwater region showed an overall increasing trend, with an annual growth rate of 0.7 cm/a; from 2013 to 2020 the groundwater storage in the Three-River headwater region showed a decreasing trend, with an annual growth rate of -0.3 cm/a. The annual variation of groundwater storage in the Three-River headwater region is mainly affected by rainfall replenishment, which lags behind the recharge of groundwater by rainfall by two months.

To investigate the reliability and stability of the RZB-type borehole strain sensor with the function of the range expansion and in-situ calibration, we design a qualitative comparison experiment of this sensor with conventional sensors under cave conditions, and quantify the daily drift of the sensor. The superiority of the RZB borehole strain gauge sensor is verified by plotting and analyzing the observed data from 2019-10 to 2020-05 and calculating statistics.

Based on the calculation of the pore pressure distribution and the measured water level fluctuation of nine fluid wells in Tangshan area before, during and after the Guye MS5.1 earthquake. The results show that: 1) Before the earthquake, the Guye area of Tangshan was in the process of stress accumulation, the compression area was nearly in the east-west direction, and the tension area was nearly in the north-south direction. The measured well water level rise and fall were basically consistent with the compression and tension. 2) During the earthquake, the compression and tension areas do not change, but the stress is released. The measured well water rise and fall are inconsistent with the compression and tension. 3) After the earthquake, the compression area evolves into the tension area, and the tension area evolves into the compression area. This paper explains the adjustment process of seismic static stress from pre earthquake to post earthquake. By calculating the distribution of well bore pressure before, during and after the earthquake, we find that the distribution of well bore pressure is in four quadrants around the fracture zone of the strike slip fault, which can be explained by the locking shear model before the earthquake.

We use natural earthquakes and blasting events in Inner Mongolia and its surrounding areas (96°-126°E, 36°-54°N) from 2016 to 2021. Firstly, we use db7, sym6 and rbio1.5 wavelet basis functions to decompose the event waveforms into discrete wavelet, static wavelet and wavelet packet, and then extract three characteristic parameters of energy ratio, Shannon entropy and energy entropy. 288 groups of experiments are conducted by random combination of different wavelet decomposition methods, kernel function, support vector machine and eigenvalue. The results show that the highest recognition rate of “DWT+υ-SVC+db7+linear kernel+energy ratio+ Shannon entropy+energy entropy” is 95%, which indicates this method is more suitable for Inner Mongolia. It can provide a reliable reference for the identification of natural earthquake and blasting events in seismic network.