To investigate the seismogenic structure of the 2025 Maduo M_S5.5 earthquake in Qinghai province, we use the double-difference relocation method to precisely determine the hypocentral distribution of Maduo earthquake sequence, and calculate the tectonic stress field in the epicentral region. The fault plane parameters are derived using the Gauss-Newton and simulated annealing algorithms. Based on these results, we characterize the kinematic properties of seismogenic fault and compute the coseismic displacement field using an infinite half-space elastic model. The results show that the Maduo earthquake sequence is dominated by a near-horizontal SW-NE-oriented principal compressive stress axis and a NW-SE-oriented principal extensive stress axis, with R-value of 0.42. The fault plane parameters fitted by small earthquakes of Maduo earthquake sequence are strike 43.39° and dip 81.64°, suggesting that the rupture surface is a NE-striking nodal plane, which is a right-lateral strike-slip hidden fault. The coseismic displacement field reveals a maximum horizontal displacement of 12 to 13 mm and a maximum vertical displacement of 3 to 5 mm.

Using GNSS observation data of Chinese mainland from 1996 to 2016, we invert the current slip rates of active faults around Longxi block based on elastic block model. Combined with the historical earthquake catalog, we calculate the seismic moment deficit of each fault and analyze the spatial distribution of seismic b-value in Longxi block and its surrounding regions to assess seismic risk. The results show that the Haiyuan fault zone exhibits high-rate left-lateral strike-slip motion with a compression component, while the Liupanshan fault zone is dominated by crustal shortening and compression with a strike-slip component in both its northern and southern segments. The northern margin of western Qinling fault zone is primarily characterized by left-lateral strike-slip motion, with an extensional component in its western segment and a compression component in its eastern segment. In terms of seismic risk, significant seismic moment deficits have accumulated in the arcuate tectonic region where the Hasishan-Machangshan segment of Haiyuan fault zone meets the northern segment of Liupanshan fault zone. Meanwhile, the Jinqianghe, Maomaoshan, and Laohushan faults within the Tianzhu seismic gap show rapid stress accumulation and high seismic moment deficits. The Maxianshan fault zone exhibits high compression slip rate and a low b-value, indicating high stress accumulation. Additionally, the Zhuanglanghe fault zone has a substantial seismic moment deficit. These fault zones all possess the potential to generate moderate-to-strong earthquakes.

The laser interferometry absolute gravimeter uses the stable frequency laser wavelength as the precise length reference, and the selection of reference source directly affects the accuracy and precision of the measurement results. We focus on the differences in engineering applications between iodine saturated absorption stable frequency lasers and thermally stable frequency lasers. The wavelength accuracy and repeatability of thermally stable frequency lasers are analyzed in detail. We construct a beat frequency measurement system based on iodine stable frequency laser reference, which achieves the measurement of wavelength accuracy and repeatability of thermal stable frequency laser. A statistical processing model with the same final result of absolute gravity measurement is spmpled to quantify the repeatability and accuracy of output wavelength of the tested laser. The results show that the output wavelength accuracy of the tested thermally stable frequency laser is better than 4×10^-9 nm, and the repeatability is better than 1.86×10^-6 nm. That is, during the locking process of the thermal stable frequency laser, the gravity measurement error caused by wavelength accuracy is much smaller than 1 μGal, and the absolute gravity measurement error caused by wavelength repeatability is 2.79 μGal, which is acceptable for the absolute gravity measurement requirement of 10 μGal observation accuracy. This provides a complete metrological basis for the engineering application of absolute gravity instruments based on thermal stable frequency lasers in field environment.

Aiming at the problems of low accuracy of empirical models and meteorological parameter model accuracy limited by meteorological parameters in commonly used ZTD prediction models, we propose a regional ZTD prediction model based on improved multi-head attention mechanism combined with long short-term memory network (LSTM-MHA). The model is constructed using data from the third-generation global pressure and temperature model (GPT3), along with station latitude, longitude, elevation, day of year, and hour of day. Using ZTD data from 26 continuously operating reference stations (CORS) in Europe in 2023 as reference values, comparative experiments were conducted with the GPT3 model, LSTM model, and CNN model. The results show that the LSTM-MHA model achieves a mean bias close to 0, with root mean square error and mean absolute error of 13.43 mm and 12.4 mm, respectively. The accuracy of LSTM-MHA model has been improved by 60.8%, 55.2%, and 53.3% compared to the GPT3, CNN, and LSTM models. Evaluation of the LSTM-MHA model performance under different distances and prediction lengths reveals that the proposed method performs stably in both sparse station networks and long-term forecasting scenarios, demonstrating its suitability for regional multi-station prediction tasks.

To analyze the differences of different singular integral methods in geoid calculation using Stokes integral, based on the traditional methods using circular approximation and biquadratic polynomial interpolation, we derive a de-singularized method for Stokes integral within the integration radius. The results show that when the innermost area is selected as the single grid where the calculation point is located, the accuracy of traditional method and biquadratic polynomial interpolation method are similar, and the de-singularized method has been improved using the point kernel numerical integration. The mean kernel numerical integration can improve the calculation accuracy of traditional method and biquadratic polynomial interpolation method, while the de-singularized method is not significant.

Based on reduced-dynamic method, we utilize sliding window to achieve near-real time orbit calculation and recursive update for low Earth orbit satellites, and verify the accuracy and effectiveness of orbit determination. Taking the observation data of GRACE-FO and Swarm-B satellites from January 15 to 26, 2025 as an example, the results show that there is no significant systematic error in the tangential, normal, and radial orbit overlapping arc segments calculated using sliding window. Compared with the official released precise science orbit, the root mean square (RMS) of orbit difference in calculation part in three directions is all within 5 cm, and the residual RMS of satellite laser ranging (SLR) validations are 3.03 cm, 4.14 cm, and 3.15 cm, respectively. At a predicted arc length of 60 minutes, the average radial orbit RMS of three satellites is about 10 cm. In addition, changing the initial value of subsequent arc segment orbit from standard point positioning (SPP) to reusing the calculation and extrapolation result of previous arc segment orbit can shorten the single arc segment solution time from 10 minutes to 5 minutes.

We investigate orbit determination and prediction for low Earth orbit (LEO) satellites using high-precision ranging data from a sparse regional network of ground stations. Aiming at applications with limited observational data, three strategies for resolving the atmospheric drag coefficient (C_D) are proposed: 1) Unconstrained estimation using the long-arc C_D solution as the initial value. 2) Using the long-arc C_D solution as the initial value, and the accuracy factor of long-arc C_D solution as the accuracy constraint. 3) Directly using the long-arc C_D solution as a fixed value. The results show that the mean orbit determination residuals for all three strategies are below 0.17 m. Compared to the precise orbit, the 3D orbit determination accuracy for all three methods is better than 10 m. For the 24-hour predicted arc, the unconstrained solution achieves a RMS better than 55 m, while both the constrained and fully constrained solutions achieve an accuracy better than 15 m. This study verifies the feasibility of independent LEO satellite orbit determination using ground-based high-precision ranging data.

Aiming at the quantitative evaluation of observation environment complexity in GNSS high-precision positioning, a comprehensive subjective and objective weighted combinative distance-based assessment (CODAS) method based on data quality indicators such as data integrity rate, cycle slip ratio, pseudorange multipath error, carrier-to-noise ratio and PDOP is proposed. According to different weighting strategies, the CODAS, TOPSIS and VIKOR methods are used to evaluate the observation environment complexity of 120 monitoring stations, and the Spearman correlation coefficient is used to measure the correlation between evaluation results of each method and the accuracy of PPP kinematics solution. The results show that the comprehensive subjective and objective weighting can improve the accuracy of CODAS method compared with the single weighting. The Spearman correlation coefficients of evaluation results of CODAS method with RMSE and ME are 0.897 and 0.924, respectively, which are better than TOPSIS and VIKOR methods. Parameter sensitivity analysis shows that the mean standard deviation of evaluation results of CODAS method under different trade-off parameters is less than 1.5, indicating good stability.

A "decomposition-reconstruction-prediction" satellite clock offset prediction method based on singular spectrum analysis (SSA) is proposed. This method first decomposes the clock offset time series into trend, periodic, and random signals. Then, based on the significant difference in Hurst exponent among the three types of signals, the reconstructed clock offset signal is determined. Finally, the SSA iterative interpolation algorithm is used to extrapolate the reconstructed clock offset signal and obtain the final predicted clock offset time series. Taking GPS and BDS-3 satellite clock offsets as an example, the results show that SSA can accurately extract trend and periodic signals in clock offset time series. Compared to the quadratic polynomial model with periodic term, grey model and autoregressive integrated moving average model, the proposed method has a smaller root mean square error in prediction and more stable prediction results. For the 12-hour prediction, the average prediction accuracy of GPS satellite clock offset is improved by 38.81%, 59%, and 72.11% compared to the other three models, respectively, while the accuracy of BDS-3 satellite clock offset is improved by 71.54%, 74.82%, and 70.09%, respectively. For the 24-hour prediction, the improvements for GPS satellite clock offset are 51%, 70.66%, and 83.99%, respectively, while the improvements for BDS-3 satellite clock offset are 86.36%, 85.25%, and 81.05%, respectively. It is concluded that the proposed method is effective and practical for satellite clock offset prediction.

This study inverts a Mascon model based on inter-satellite range-rate observations from the ChiGaM satellite mission. The strategies employed during the inversion include: 1) introducing piece-wise constant acceleration (PCA) parameters to mitigate the impact of force model errors in dynamic orbit determination; 2) applying an empirical model that considers the 3-cycle-per-revolution (3-CPR) term to absorb low-frequency noise in the range-rate data; and 3) integrating multi-source data to construct the forward model and the a priori constraint matrix for parameters. Satellite data from March 2022 to December 2023 were processed to develop the Mascon model IEU-CGS-Monthly2023M. Comparative analysis with contemporaneous Mascon products from the GRACE-FO mission shows that the mass change trends derived from our model are in high agreement with the reference models at both global and regional scales. The geoid degree errors of all models closely coincide within the dominant signal degree bands. In nine selected river basins, the average correlation coefficient between our model and the reference models exceeds 0.9. Overall, the performance of the proposed model is comparable to that of the reference models. This study demonstrates high-precision Mascon inversion using ChiGaM range-rate data, validates the satellite's capability to capture time-variable gravity signals, and can serve as a reference for related data processing and applications.