Based on hourly-resolution zenith tropospheric delay(ZTD) data from 35 GNSS stations in the Greenland region spanning the period from 2010 to 2021, combined with ERA5 reanalysis datasets and T_m model, precipitable water vapor(PWV) over the region was retrieved. Subsequently, the principal seasonal and diurnal variation characteristics, including annual, semi-annual, diurnal, and semi-diurnal cycles, were investigated. The results indicate that the GNSS stations in Greenland exhibit a certain degree of autocorrelation, with the summer season showing weaker autocorrelation compared to other seasons. The amplitude of the annual component ranges from 2 to 7 mm, while that of the semi-annual component ranges from 0 to 3 mm, with relatively minor interannual variations in their phases. The maximum values occur in summer, and the minimum values in winter. The diurnal component is only pronounced at a few stations, with amplitudes around 0.08 mm. After removing the annual, semi-annual, and diurnal components from the time series to obtain the residual time series, the maximum RMSE ranges from 1.30 to 3.65 mm, with an average RMSE of 2.08 mm. Notably, the RMSE values are relatively higher in the northern region and lower in the southern region.

A multi-scale prediction model(referred to as C-TCN-A) based on complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN) and temporal convolutional network-attention mechanism(TCN-Attention) algorithms is proposed, which can be effectively applied to missing data imputation and future trend prediction of GNSS elevation time series. The model first employs CEEMDAN for multi-scale decomposition of time series, then utilizes TCN-Attention for prediction and reconstruction of different scale components to obtain final results. To validate the model's performance, 12 observation stations were selected for 1-day and 5-day predictions, with comparisons made against multiple other models. Results demonstrate that in 1-day prediction, C-TCN-A reduces RMSE and MAE by 35%-40% and 36%-41% respectively while improving correlation coefficient R by 25%-29%. For 5-day prediction, it achieves reductions of 20%-26% in RMSE and 20%-28% in MAE, with R increasing by 26%-33%. To verify the model's universality, C-TCN-A was applied to 99 stations from the crustal movement observation network of China(CMONOC) for 1-day and 5-day predictions. Results indicate generally favorable RMSE and MAE metrics with concentrated error distribution, where most errors remain below 4 mm. Spatial analysis reveals regional performance differences, with optimal results achieved in the northwestern region.

Previous ZWD models over China only consider the effects of elevation, latitude and meteorological parameters, without taking into account the impact of monsoon climate and ocean water vapor transport on ZWD estimation. To solve this problem, we take bias correction as the breakthrough point, analyze the spatial distribution of RMSE and bias of ERA5-ZWD estimation value obtained by the ERA5 data integration method and inversion method over China from 2016 to 2019. Then, the regional bias correction model over China based on RF_Adaboost neural network is established using the ERA5 and radiosonde data from 2016 to 2018. The results show that the bias of ERA5-ZWD estimation value in 2019 is significantly reduced after correction, and the RMSE is significantly reduced in the eastern coastal areas and southern regions of China.

By establishing a BDS transmission reference control surveying model, the paper compares and analyzes the high-precision control surveying positioning performance differences between BDS-3/BDS-2 and BDS-3+BDS-2. Nine first-order control points of a project in Beijing and three IGS tracking stations in China and its surrounding areas were selected to establish a high-precision control surveying positioning network that supports both GPS and BDS-3/BDS-2 signals. Continuous observation data from three time periods(days 153-155 of 2023) were selected, and four schemes were designed for data processing. The positioning performance was compared and analyzed from two aspects: the transmission results of the initial reference and the adjustment results of control surveying. The results show that compared with GPS, BDS-3 and BDS-3+BDS-2 have only millimeter-level differences in N, E, U, horizontal, and point directions, while BDS-2 has differences ranging from millimeters to centimeters in the U and point directions, and only millimeter-level differences in the N, E, and horizontal directions. The measurement accuracy of BDS-3 is basically equivalent to that of BDS-3+BDS-2, while BDS-2 has lower accuracy than the other two, especially in the U direction.

To address the challenges of excessive noise points and weak effective signals in satellite laser ranging(SLR) echo data, which hinder accurate and rapid signal identification, this study proposes a composite real-time signal identification method. The method integrates an improved traditional histogram-based approach with the Graz fast filtering algorithm, optimizing both identification accuracy and processing efficiency. A simulated platform was established to replicate signal generation processes, preliminarily validating the algorithm's advantages. Experimental validation using measured data(1 ms per data point) demonstrated that the proposed algorithm achieved a false detection rate of 8.71%, significantly lower than the 32.71% and 43.52% rates of two conventional secondary filtering algorithms, while maintaining comparable missed detection rates(about 10%). Post-processing analysis of 1 million echo datasets(100 000 echoes per satellite for 20 satellites) revealed an average computation time of 4.31 s, outperforming the conventional algorithms(7.28 s and 12.45 s).

In order to simultaneously improve the efficiency and reliability of point cloud plane segmentation, based on the multi-scale supervoxel and region growing algorithm, supervoxels are used as the growing units, and the voxel region growing algorithm is used to achieve fine point cloud plane segmentation. The results show that the proposed method has good segmentation effect on both indoor and outdoor point cloud data, which is superior to RANSAC, Khaloo and global energy optimization methods. Moreover, the Precision, Recall and F₁-score of the method in this paper are all higher than 0.90.

In this paper, the stochastic finite-fault method is used to simulate the ground motion of the Wulong M_S5.0 earthquake. The simulation results for four stations within 100 km of the epicenter showed that the simulated acceleration time histories, peak ground acceleration, and response spectra are generally consistent with the actual records. Some discrepancies were observed at individual stations due to the influence of model parameters and environmental noise. Simulations were performed for 323 grid points within 100 km of the epicenter, and the spatial distribution characteristics of the ground motion obtained were consistent with the intensity field survey data. This study provides data support for seismic fortification and risk assessment in the Chongqing region and has both theoretical and practical significance for mitigating earthquake disasters.

Using the arrival time data of natural earthquakes from January 2009 to June 2023 in Shanxi and adjacent areas, the double-difference tomography method was employed to obtain the relocated catalog of the region, as well as high-resolution three-dimensional crustal velocity structure and velocity ratio structure. The tectonic environment for historical earthquakes of magnitude 6 or above in this area was discussed. The two-dimensional profiles of velocity and velocity ratio versus depth for historical earthquakes of magnitude 6 or above that passed through the southern part of Shanxi show that the 1303 Hongtong earthquake of magnitude 8 occurred on a gradient belt of high and low velocity anomalies, with a low-velocity anomaly beneath the earthquake, indicating that the ductile creep of the lower-middle crust can lead to localized stress and strain concentration in the brittle upper crust, triggering earthquakes. The earthquake is located in an area of low velocity ratio, and the Linfen basin has a low velocity ratio beneath it, with slightly higher velocity ratios on both sides of the mountainous areas, which is related to the process of crustal extension and thinning beneath the basin. Both the velocity and velocity ratio structures show that the structural characteristics beneath Luoyunshan piedmont fault are steep in the shallow part and gentle in the deep part, with the fault cutting extending to a depth of about 30 km, which is similar to the results of existing seismic reflection profiles and may be the reason for multiple earthquakes of magnitude 6 or above in this area. Most historical earthquakes in the southern part of Shanxi occurred near the gradient belt of high and low velocity ratios, which may be related to these areas being located at the boundary of the basin and mountain, where the crustal material composition changes more dramatically and is more likely to accumulate stress and strain, thus generating earthquakes.

Utilizing the Yunnan small earthquake monthly catalog and seismic wave data recorded by the seismic network, we calculated the correlation coefficients of spectral amplitudes for Yunnan M≥5.0 earthquake sequences since 2008. A comprehensive analysis of traditional seismological parameters, including the h and b values, for different types of earthquake sequences revealed that the correlation coefficients of spectral amplitudes for foreshock type seismic sequences are significantly higher than those for earthquake swarm and mainshock-aftershock type sequences. Moreover, the range of variation is relatively small, with values exhibiting a concentrated distribution. In contrast, traditional seismological parameters show a more concentrated b value and a wider distribution range for the h value. Analysis of the correlation coefficients of spectral amplitudes for 8 typical foreshock type earthquake sequences indicates a good correspondence between changes in the correlation coefficients of spectral amplitude and the magnitude intensity of foreshock type earthquake sequences. The analysis suggests that a high correlation coefficient of spectral amplitude is indicative of high regional stress levels, and fluctuations in the correlation coefficient of spectral amplitude may be influenced by the loading and unloading of seismic stress.

The first arrival travel times of a deep reflection profile, approximately 30 km long, implemented along the northwest-southeast direction across the northern segment of the Tangshan fault in 2023 were inverted to obtain a detailed shallow velocity structure profile in the northern part of Tangshan. The velocity structure profile reveals that from 2.5 to 8.5 km along the survey line, the Cenozoic Era sediments are about 150 m thick and relatively uniform. The Fengtai-Yejituo fault is hidden at the survey line coordinate of 3.5 km, trending northwest, and F_2-1 fault is hidden at 8.5 km, trending southeast; there is a hidden bulge between these two faults that is low in the northwest and high in the southeast. From 8.5 to 15.2 km along the survey line, the Cenozoic Era sediments gradually thicken from northwest to southeast, with a maximum thickness of about 250 m, indicating that the Jinggezhuang depression is a gull-shaped Cenozoic fault depression. The Douhe fault is exposed at 15.2 km, trending northwest. From 15.2 to 24 km along the survey line, it is the Tangshan uplift, with only local areas having very thin Cenozoic Era sediments, and the P-wave velocity at the same depth is much higher than on both sides; there is a nearly vertical low-velocity zone under Weishan, extending at least 600 m downward, which is speculated to be the fault zone of the Tangshan-Weishan-Changshan south slope fault belt. From the survey line coordinate of 24 km to the end of the survey line, it is the northwestern part of the Kaiping depression, showing a regular basin edge morphology, with Cenozoic Era sediments of about 130 m. The Douhe fault and the Tangshan-Weishan-Changshan fault are active faults of the Quaternary.

Based on the Sentinel-1A radar image data of the 2023 Jishishan M_S6.2 earthquake, we use D-InSAR technology to obtain the coseismic deformation field of ascending and descending orbits, use SDM inversion program to obtain the slip distribution of seismogenic fault, and use Coulomb3.3 program to obtain Coulomb stress changes of regional fault. The results show that the coseismic deformation fields generated by this earthquake are mainly uplifted, and the maximum LOS deformation is 6.8 cm for ascending orbit and 7.6 cm for descending orbit, and the optimal solution of seismogenic fault is 303° of strike, 52° of dip, and 89° of slip. The maximum slip is 0.5 m, which occurs at a depth of 11.89 km from the surface, and the rupture does not reach the surface. The cumulative seismic moment released was 1.48×10¹⁸ Nm, with a moment magnitude of M_W6.1. The Coulomb stress results indicate that the stress in the southern margin of Lajishan fault, the northwest section of north edge of western Lajishan fault, the northwest section of Daotang-Linxia fault, and the northwest section of north edge of western Qinling fault are in the state of obvious stress loading, and the above regions are more probable to have earthquakes in the next period of time, which need to be paid attention.

In this study, we at first employ a simplified 1-D layered spherical Earth model to compute the co- and post-seismic vertical displacement and geoid change on the seafloor surface due to the 2004 Sumatra earthquake. We then use a pseudo-spectral algorithm to iteratively solve SLE and obtain RSL and geoid changes of long wavelength(>90 km). Results show that the contribution of redistributed seawater mass on RSL is secondary, which can be neglected when we focus on the long wavelength RSL. The effect of geoid change due to the redistribution of seawater mass mainly stems from the self-gravitation of seawater disturbed by earthquakes, while the contribution from elastic load response is an order of magnitude smaller. In addition, we examine two approximate methods of ocean water effect—solving SLE with a uniform ocean covering and Bouguer layer approximation. Comparison results show that both two methods can effectively estimate the co- and post-seismic ocean water effect of the 2004 Sumatra earthquake. The seawater mass effect estimated by Bouguer correction formular is in better agreement with that obtained from solving SLE with a realistic ocean.

Based on TEC data obtained from the CODE and electron density(N_e) data acquired by the CSES, the seismo-ionospheric perturbations prior to three M_S≥6.0 earthquakes in western China from 2020 to 2022(that is the 2020 Jiashi M_S6.4 earthquake, the 2021 Yangbi M_S6.4 earthquake and the 2022 Maerkang M_S6.0 earthquake) were analyzed. The ionospheric anomalies were found to be concentrated in 1 week before the earthquakes. The synchronization anomalies of CODE TEC and CSES N_e occurred 5 and 2 days before the Jiashi M_S6.4 earthquake, 7 days before the Yangbi M_S6.4 earthquake and 6 days before the Maerkang M_S6.0 earthquake. In addition, it is possible that these three strong earthquakes may affect the ionosphere through the geochemical and electromagnetic channels, as described by the lithosphere-atmosphere-ionosphere coupling(LAIC) model. There were earthquake precursor anomalies such as geochemical anomalies, water level and electromagnetic anomalies near the focal area in the 5 to 15 days before the earthquakes.

This paper develops an integrated InSAR atmospheric delay correction method that combines external atmospheric data with topographic data. The basic approach involves introducing external atmospheric data to perform rough atmospheric delay correction on InSAR interferograms, followed by the establishment of a global correction model that links regional elevation to the interferometric phase, thereby suppressing the topography-related atmospheric delay components with long wavelength distributions. The interferogram is then processed in blocks, and a locally adaptive topography-related atmospheric delay error model is established for each block to achieve refined atmospheric delay correction for each sub-region. This study uses the Zagunao river basin, an area where landslides frequently occur, as a case study. Different atmospheric correction methods are applied to correct atmospheric delay errors in InSAR interferograms and to identify landslides. The results show that as the atmospheric correction progresses, the correlation between InSAR phase and topography significantly decreases, validating the effectiveness of the integrated atmospheric correction method. Furthermore, regional time series deformation analysis and landslide identification demonstrate that the integrated atmospheric delay correction method can effectively restore deformation signals from medium- and small-sized landslides, and significantly improve the accuracy of landslide boundary detection. Statistical analysis reveals that, compared to a single external atmospheric correction, the combined global topography-related atmospheric correction increases the number of identified landslides by 75%, and after further introducing local adaptive atmospheric correction, the number of identified landslides increases by an additional 33%.

To address the challenges of insufficient real-time interaction capabilities and limited functionalities in wireless node seismometers during geophysical exploration and seismic monitoring, this study designs an Android-based integrated platform for node seismometer monitoring and field deployment. The platform leverages bluetooth low energy(BLE) technology and implements a custom encrypted wireless communication protocol to enable data exchange and device control between mobile terminals and node instruments. Key functionalities include real-time visualization of single/triaxial seismic velocity signals, status monitoring, and synchronized configuration of single or multiple devices. Additionally, the platform integrates ultra-high frequency(UHF) RFID technology, QR code scanning, and electronic maps to achieve rapid field deployment of seismic node arrays. SQLite databases and a local file system are employed to ensure reliable recording of survey lines, parameter configurations, and system log management. Experimental results demonstrate that the platform can enhance the efficiency of field geophysical surveys and mobile seismic monitoring.

In this study, natural and non-natural seismic events in Guangdong and surrounding areas were selected as the research objects, and the typical physical characteristics of P waves and S waves in frequency domain were extracted from seismic waveform data. We used the feature combination optimal solution algorithm to input the best feature parameters into the support vector machine model for sample training, and identified the non-natural earthquakes in Guangdong and surrounding areas. The test results showed that the accuracy of earthquake classification is 99.3%. In addition, the non-natural seismic identification module based on support vector machine was developed and connected to the daily cataloging system of Guangdong seismic network, which realized the quasi-real-time automatic identification of seismic event types. The application results showed that the recognition accuracy of 130 manually cataloged seismic events was 96.9%.