Based on 26 496 BDS monitoring sequences with various observation durations in 2023, we establish quantitative function expressions relating the availability rate, monitoring precision of BDS results to observation duration. The results show that there is a significant positive correlation between the availability rate of monitoring results and observation duration, with an average correlation coefficient of 0.73 for six monitoring stations. The monitoring precision in N, E, U directions exhibit significant negative correlations with observation duration, with correlation coefficients of -0.83, -0.90, and -0.77, respectively. To ensure high timeliness, high availability rate (>95%), and high precision (horizontal precision better than 3 mm, vertical precision better than 5 mm), an optimal observation duration of 3 hours is recommended.

Three missing data processing methods, namely zero padding, linear model fitting, and resampling, were proposed to evaluate the frequency stability of atomic clocks in presence of data interruption and date gap. A direct calculation method for calculating overlapping Hadamard variance was also proposed. Based on clock error simulation data, the influence of different processing methods on frequency stability calculation was compared and analyzed. The results show that when the data missing rates are below 1%, 10%, and 20%, the stability calculated by zero padding, linear model fitting, and resampling methods is relatively close to the true stability. Zero padding and linear model fitting methods are only suitable for situations with less missing data, while resampling methods have better applicability than the first two methods. Furthermore, the direct calculation method can accurately evaluate stability even when the data missing rate reaches 50%. In cases where there is a large amount of missing data, the original data can be directly used for stability evaluation. Finally, the frequency stability of the BDS-3 on-board clock with missing clock bias data was evaluated. It is shown that the stability of the rubidium clock and hydrogen clock were better than 2.12×10^-14 and 2.15×10^-15 at the sampling time of 10 000 seconds and one day, respectively, and was generally at the same level as GPS Ⅲ A rubidium clocks and Galileo hydrogen clocks.

This paper proposes a method that can calculate the integrity risk of RAIM under multiple fault conditions. This method is based on constructing a chi-square test statistic using least squares residuals for fault detection, and further calculates the integrity risk by searching for the worst case fault vector under multiple fault modes based on the probability distribution of the test statistic and positioning error. The actual experimental results show that the proposed method can effectively monitor the integrity risk of multiple faults, achieve risk alarms during periods of poor observation geometry, and ensure reliable user positioning. Under the integrity requirements of the aviation LPV-200 phase, the availability of user positioning results reached 99.47%.

Based on the observation data of ground-based GNSS network in Hubei province during 25 June to 9 July 2020 and in Henan province during 13 to 26 July 2021, the precipitable water vapor(PWV) are derived. Combined with hourly precipitation data of ground automatic weather stations, we analyze the spatio-temporal characteristics of PWV and its influence on precipitation in "Violent Meiyu" in 2020 and Henan "7·20" extreme rainstorm in 2021. The results show that the regions with high PWV value above 65 mm have a good indication of the areas where rainstorms happen. PWV increases in a fluctuating tendency before the arrival of rainstorm, the precipitation intensity and PWV reach the maximum value almost at the same time. During the process of "7·20" extreme rainstorm in 2021, the water vapor is sufficient throughout Henan province, the indication of PWV high value for rainstorm is relatively weak. In general, PWV shows a trend of slow increase and fast decrease in "7·20" extreme rainstorm. Before the occurrence of precipitation, PWV fluctuates at a high level and grows slowly. When PWV rapidly decreases, precipitation stops. Furthermore, it is also revealed that PWV usually increases 1 to 2 hours before the occurrence of rainstorm both in 2020 and 2021, which is significant for rainstorm monitoring and weather forecasting.

This study analyzed data from vertical pendulum tiltmeters, water-tube tiltmeters, extensometers, borehole strainmeters, and gravimeters at three stations within 100 km of the epicenter prior to the M_S6.8 Luding earthquake in Sichuan province. The results revealed that pre-earthquake observations from some instruments exceeded two times the mean square deviation, and the distance from the epicenter influenced the duration of the anomalies.

By utilizing the method of dominant azimuth of the geoelectric field, an analysis of the dominant azimuth anomaly changes and characteristics was conducted on the geoelectric field data within a 400 km range of the epicenter of the 2023 Jishishan M_S6.2 earthquake. The results show that the earliest anomaly before the earthquake occurred at the Guyuan station, which is located at the intersection of the Haiyuan fault and the Liupanshan fault. During the period of the anomaly, the Jingtai station, located in the middle section of the Haiyuan fault, also showed a rapid and significant deflection of the azimuth. The max azimuth α changes at both stations deviated from the normal background value by about 45°, which is considered a medium to short-term anomaly. The last anomaly to appear was at the Dawu station, located in the Bayan Har block, where the azimuth angle α showed a rapid deflection and jump, deviating from the normal background value by about 90°, which is considered a short-term anomaly. The number of three abnormal stations meets the prediction index system of the geoelectric field dominant azimuth. The evolution of the anomaly shows that the anomalies have quasi-synchronous characteristics in terms of appearance, duration, and end time. In terms of the distance from the epicenter, Guyuan station is the farthest, while Jingtai and Dawu stations are closer to the earthquake epicenter. Before the earthquake, due to the internal compression of the plate, the areas where the three geoelectric field stations are located and the surrounding stress compression changes were intense. When rocks were subjected to pressure and fracture, the fracture structure of the rock mass led to significant changes in the dominant azimuth of the geoelectric field in succession. The subsequent end of the three geoelectric fields anomaly after the earthquake also indirectly reflected the process of the gradual recovery of the rock mass fracture structure after stress release, indicating that the data changes of the three stations can truly reflect the changes in the underground medium.

By selecting time and spatial scales suitable for regional characteristics and using accelerating moment release to predict the location of earthquakes, retrospective testing was conducted on 8 earthquakes with M_S4.0 or higher in Heilongjiang province and adjacent areas since 2009. The results showed that 5 earthquakes observed accelerating moment release phenomenon before the earthquake, while in the other 3 were not observed. At the same time, before some earthquakes, it can be observed that there was a phenomenon of accelerated areas migrating towards the epicenter. This indicates that accelerating moment release has certain indicative significance for predicting the occurrence location of moderate earthquakes in Heilongjiang province and adjacent areas.

This paper attempts to link the strength of regional crustal stress with its earthquake occurrence probability, and proposes a method to calculate regional earthquake occurrence probability based on apparent stress data, which mainly solves the following problems: 1) The singular value is calculated using multifractal theory, and the stress rich region, poor region and background region are revealed. 2) A new method of fitting singular value and b value is derived, and b value under the background of singular value are obtained, so that the strength of crustal stress and b value are unified. 3) A suitable method is designed to calculate the shear stress loading rate of the reference area, and then other shear stress loading rate value are calculated by using the specified formula. 4) When the lognormal model is used to calculate the probability of earthquake occurrence, the fixed integral is used to process the model parameters, and the error of shear stress loading rate is the main source of the uncertainty of the result. 5) Compared with the normal distribution model and the BPT model, it is found that the lognormal distribution model agrees well with the normal distribution model, but there is a big difference between the lognormal distribution model and the BPT model, which may be due to the different algorithms of each model. 6) According to the calculation and analysis, the areas with a higher probability of earthquake occurrence of magnitude 5 or above in the study region are located within the enclosed areas of Yiyuan-Linqu-Anqiu-Zhucheng-Yishui, Yishui-Wulian-Juxian-Yinan and Linqu-Changle in the north of the study area. The conditional probability of earthquake occurrence in 10 years is up to 0.034, and the cumulative probability is up to 0.861. The conditional probability of earthquake occurrence in 30 years is up to 0.097, and the cumulative probability is up to 0.87.The conditional probability of earthquake occurrence in 50 years is up to 0.15, and the cumulative probability is up to 0.878.

In order to more clearly depict the fault distribution in the seismic profile of urban active fault detection, seismic attributes are extracted from the seismic data of urban active fault detection based on the theory and implement methods of variance body, coherence body, instantaneous phase and wave impedance inversion, and comprehensively analyze the application effects of various attributes, and use the attribute fusion method to superimpose display the seismic attributes. The results show that using seismic attribute analysis and attribute fusion can reduce the difficulty of fault identification, improve the accuracy of fault identification, and the fault distribution is more regular.

This article adopts the double-difference tomography method to obtain the three-dimensional P-wave velocity structure and seismic relocating results of the Wushan section in the Three Gorges reservoir, based on the travel time data of 2 319 earthquakes recorded by 49 stations from January 2008 to June 2020. The results show that earthquakes are mainly concentrated on the north bank of the Yangtze river, with a nearly EW oriented strip distribution, and are not more than 10 km away from the Yangtze river system. There are two obvious P-wave high-speed bodies concentrated mainly at the intersection of the Yangtze river and the faults in the research area. One is distributed in the area trapped by the Gaoqiao fault and the Zhoujiashan-Niukou fault, and the other is distributed in the area where the Wushan fault intersects with the Yangtze river. The distribution of high-speed bodies is consistent with the dip of nearby faults. The velocity cross-section shows that the high-speed zone extends to a depth of 12 km. The Wushan fault, Gaoqiao fault, and Zhoujiashan-Niukou fault serve as the main fracture channels for reservoir water infiltration.