The synchronization accuracy of the pseudo-range single-point positioning method is low, making it unable to meet the high-precision time synchronization requirements of low Earth orbit (LEO) satellites. Meanwhile, the inter-satellite link time synchronization method faces challenges in large-scale LEO constellation applications due to factors such as complex payloads, high equipment costs, and susceptibility to interference from the space environment.This study utilizes onboard GPS observation data from the GRACE-FO satellites to design and investigate an inter-satellite time synchronization method for low Earth orbit satellites based on precise point positioning (PPP). Experimental results show that the orbit determination accuracy (RMS) of the GRACE-FO satellites in all directions is approximately 7 cm, and the GNSS timing accuracy (STD) of the two satellites is 0.78 ns and 0.77 ns, respectively, with short-term stability (1 280 s) reaching 2.22×10^-13 and 2.13×10^-13, and long-term stability (10 240 s) reaching 2.69×10^-14 and 3.21×10^-14. Meanwhile, the inter-satellite time synchronization accuracy (STD) is 0.46 ns, with short-term stability (1 280 s) of 2.24×10^-13 and long-term stability (10 240 s) of 3.47×10^-14. These results validate the feasibility of the proposed algorithm and provide an effective approach for high-precision inter-satellite time synchronization in LEO satellite systems.

Aiming at the limitation of real/near-real-time GNSS PWV retrieval under missing meteorological parameters, three PWV estimation models without the need for measured meteorological parameters were established in the Guangxi region based on the XGBoost model. First, the XGBZ-PWV model was developed with inputs including station time (DOY and HOD), location (Longitude, Latitude, and Height), and GNSS ZTD, and the output feature being GNSS PWV. Then, based on the XGBZ-PWV model, two empirical PWV values were incorporated to establish the XGBZG-PWV and XGBZE-PWV models, respectively. For comparison, the GPT3 model was used to provide pressure and temperature for PWV retrieval based on GNSS ZTD (GPT3-PWV model). The accuracy of the established models was validated using GNSS PWV retrieved from GNSS ZTD, ERA5 surface pressure, and temperature in the Guangxi region in 2022 as the reference value. The results show that, compared to the GPT3-PWV model, the estimation accuracy of the XGBZ-PWV, XGBZG-PWV, and XGBZE-PWV models improved by 22.98%, 29.03%, and 31.45%, respectively, with the XGBZE-PWV model performing the best. During two extreme rainfall events in 2022, the spatiotemporal evolution characteristics of PWV and rainfall were analyzed. The results demonstrate that the XGBZE-PWV model maintains good applicability even under extreme weather conditions.

Based on over 20 years of MODIS satellite remote sensing infrared data, we extract brightness temperature low-frequency information using anomaly method and spatial anomaly superposition method, and investigate the temporal and spatial evolution and characteristics of infrared radiation anomalies before the Wenchuan M_S8.0 earthquake on May 12, 2008. The results show that there was a significant radiation enhancement anomaly before the earthquake. In terms of time, from 2006 to 2007, there was a trend of radiation enhancement in the core area located west of the epicenter. Conversely, from 2008 until the earthquake occurrence, this radiation enhancement exhibited a weakening trend. Spatially, the core area of radiation enhancement is located in the eastern section of Bayan Har and Qiangtang blocks, encompassing approximately 5.8×10⁵ km². The spatial distribution characteristics are consistent with the dynamic background of Wenchuan earthquake.

This study proposed a novel monitoring scheme that integrates GNSS interferometric reflectometry (GNSS-IR) with GNSS positioning for monitoring coastal absolute sea level changes. Using over 10 years of observational data from seven coastal stations in Hong Kong as an example, research on absolute sea level change monitoring in nearshore areas was conducted. The results indicated that after excluding the stations HKSL and KYC1, which had lower data quality, the GNSS-IR-derived relative sea level changes from the remaining stations showed good agreement with tide gauge data in their monthly averages. For most stations, the RMSE was less than 6 cm, the correlation coefficient was greater than 0.86, and the difference in estimated sea level rise trends was less than 1 mm/a compared to tide gauges (with a regional average difference of only 0.036 mm/a). In absolute sea level monitoring, after applying the dynamic atmospheric correction (DAC), the difference in regional average rates between coastal GNSS and satellite altimetry at corresponding points was reduced from -3.40 mm/a to -0.76 mm/a, indicating highly consistent trends. Compared with 50-year long-term absolute sea level data from Hong Kong tide gauges, the deviations were mostly less than 1 mm/a, and all stations fell within reasonable error margins for regional absolute sea level monitoring. The research demonstrates that the fusion of GNSS-IR and GNSS positioning can effectively address traditional monitoring gaps, providing a scalable new technical approach for monitoring coastal sea level changes and conducting risk assessments.

Aiming at the problem that it is difficult for the temporal decomposition model to accurately distinguish the effects of induced factors on different displacement components, and the prediction accuracy is insufficient under the uncertainty of meteorological data, a landslide displacement prediction network model based on optimized time series decomposition and feature selection is proposed. Firstly, the variational modal decomposition (GA-VMD) method optimized by singular spectral analysis (SSA) and genetic algorithm is combined with induced factors to decompose the landslide displacement. Subsequently, an improved Nishihara model with fusion inducible factors is constructed to predict the trend term displacement, and the combined network of convolutional neural network and gated recurrent unit (CNN-SE-GRU) combined with compression and excitation network was used to model the period term displacement, and the random term displacement is reconstructed through frequency domain analysis. Finally, the probability interval of the displacement prediction results is constructed by combining kernel density estimation (KDE) and Monte Carlo simulation. Taking the Heifangtai landslide in Gansu province as an example, the RMSE and MAPE of the prediction model are 1.52 mm and 0.38%, respectively, and the prediction accuracy of the model is significantly improved compared with the traditional prediction model, providing more reliable technical support for landslide early warning.