Cross-fault deformation measurements have made an important contribution to the search for earthquake precursor information and strong earthquake prediction and forecasting efforts. Before the 2022 Qinghai Menyuan M6.9 earthquake, the cross-fault data in the Qilianshan-Haiyuan fault zone occured significant precursory anomalies. Based on the cross-fault data before earthquake, this paper uses time-series analysis, deformation anomaly intensity analysis, vertical deformation accumulation rate analysis and other methods to extract the temporal and spatial evolution characteristics of pre-earthquake anomalies, and analyse the short-term and medium-term fault activity and deformation mechanism before the earthquake. The results show that under the dynamic background of the combined action of NE-trending extrusion and NW-trending horizontal shear, the accumulation of deformation trend in the near-seismic region and its vicinity is at a high value, with a certain degree of strain accumulation background. Five years before the earthquake, the movement of the faults in the near-seismic area deviates from linearity and slows down. During the two years before the earthquake, the abnormal spatial distribution has the characteristics of migration and concentration to the epicentral area, which indicates that the crustal stress is accumuleting rapidly and locally enhanced in the near-epicentre area. Within half a year before the earthquake, multiple sites in the region of 60~120 km from the epicenter exhibited consistent tensional anomalies that were opposite to the compressional background, implying that micro-dynamic adjustment occurred after the stress in the near-seismic area accumulated to a certain extent, and pre-slip behaviors occurred.

The earthquake sequence’s spatial-temporal evolution can help to reveal the earthquake nucleation process and the interaction between earthquakes. In this paper, we relocated earthquakes in the catalog since 2009 around the 2022 Menyuan M_S6.9 focal area with the HypoDD and used them as templates to match-filter re-scan continuous waveforms, building a more complete seismic activity map before and after the mainshock. The results show there was not obvious foreshock pefore the 2022 Menyuan mainshock and it was a typical mainshock-aftershock sequence. Most of aftershocks mainly occurred in a strike-slip fault on the south side of Tuoleshan-north fault, dipping to the south at a high angle. The early aftershocks expanded along strike to the east and west with a logarithm migration relationship between distance and the elapsed time, meeting the rule of the afterslip expansion. The Lenglongling fault oblique intersects with the mainshock’s fault, and its seismicity is different from the aftershocks. The events in the Lenglongling fault occurred before the aftershock expansion front and migrated southeastward along the strike, but soon blocked. The migration continued again only after the M_S5.2 earthquake occurred 5 days later, and finally stopped near the end of the 2016 Menyuan aftershock zone. Therefore, the spatial-temporal evolution pattern of the earthquakes indicates the two faults interaction and the seismicity in the Lenglongling fault is probably triggered by the Coulomb stress change from the 2022 Menyuan mainshock.

On September 5, 2022, an M_S6.8 earthquake occurred in Luding County, Ganzi Tibetan Autonomous Prefecture, Sichuan Province. In this study, we analyzed regional seismic data and inverted the focal rupture process of the main shock. The results show that the seismic fault of this earthquake is the Xianshuihe fault of generally left-lateral strike-slip. The rupture propagated primarily in the SSE direction along the fault strike, expanding from the initial rupture point to the shallow part. The duration of the entire process was approximately 13 s, releasing a scalar seismic moment of 1.2×10¹⁹ N·m, equivalent to M_W6.6. The rupture had a length of about 18 km and a width of 14 km, with a maximum coseismic slip of 2.7 m and an estimated centroid depth of 6.2 km. Comparing our inversion results with those obtained from teleseismic waveform data, we conclude that the main rupture area of this earthquake is located on the SSE side of the epicenter, while no obvious rupture area is observed on the NNW side of the epicenter.

The gravity value of the Earth's surface will change accordingly in the process of earthquake preparation. We draw the contour maps of gravity field by using the mobile gravity observation data in Sichuan area from March 2018 to April 2022 and we study the variation characteristics of the gravity field before Lusan M_S6.1 earthquake and Maerkang M_S6.0 earthquake in 2022 by the contour maps. The results show: ① It was a period of intensive changes in gravity that about one and a half years before the Lusan M_S6.1 earthquake and Maerkang M_S6.0 earthquake. ② The Lusan M_S6.1 earthquake advanceed the occurrence of the Maerkang M_S6.0 earthquake; ③ The four quadrant area of the gravity variation contour before Lusan M_S6.1 earthquake and Maerkang M_S6.0 earthquake indicate a high concentration of stress in the area; ④ The epicenters of the Lusan M_S6.1 earthquake and Maerkang M_S6.0 earthquake were near the zero value line and has a large turning arc and are possible or high risk area for a large earthquake in the tectonic belt.

The process of earthquake incubation is accompanied by tectonic movement,material migration and density changes, which will cause changes in the Earth’s gravity field. The mobile gravity observation may capture precursor information related to earthquake incubation, thereby providing important basis for mid- and short-term earthquake prediction. Based on the mobile gravity observation data in Yunnan, the dynamic variation and evolution of regional gravity field before the Jinggu earthquake of M_S6.6 on October 7, 2014 are studied. The Bouguer gravity anomaly and the dynamic variation of regional gravity field are decomposed by using the wavelet multi-scale decomposition technology, and the crustal medium with different depth field source characteristics is separated by selecting the appropriate wavelet basis. The results show that the wavelet multi-scale results of EGM2008 Bouguer gravity anomaly data are relatively consistent with the measured gravity data. Through comparative analysis, the Jinggu earthquake occurred at the edge of the negative anomaly of the regional gravity field, near the high gradient belt that gravity changes from negative to positive. The gravity field image based on wavelet multi-scale decomposition can clearly reflect the anomaly caused by the change of deep material density in the epicenter area during the preparation and occurrence of Jinggu M6.6 earthquake, which has certain reference research value for the study of tectonic movement and seismogenic mechanism.

Field survey showed and steitistics that the macroscopic epicenter of November 30, 2021 M_S5.8 Shuanghu earthquake is located near Shenyazadi Village. The maximum intensity of the Shuanghu earthquake could reach Ⅶ degree. Its macroscopic epicenter location, maximum intensity range, major aftershock distribution, etc. are controlled by the fault activity. The surface rupture and seismic mechanism are consistent, which consistently indicates that Wuru cuo fault is the seismic structure.This earthquake produced some clear surface ruptures, which is of great significance for further study of the relationship between magnitude and rupture length.

Southwest Yunnan is an earthquake active area. The deep and shallow deformation characteristics of the lithosphere in this area is of great significance to understand its seismogenic environment and mechanism. This study is based on independent observation and collected regional GNSS data, the GNSS velocity field in Southwest Yunnan is studied. Then the multi-scale spherical wavelet method is used to calculate the strain rate tensor, and the spatial distribution characteristics of principal strain rate, surface strain rate and maximum shear strain rate are analyzed. The results show that the southwest of Yunnan is compressed by nearly NNE—SSW, and the surface compressibility is ~10 nstrain/a. The near east-west fault is characterized by left-hand strike slip, and the near north-south fault is characterized by right-hand strike slip. There is an obvious transformation between surface compression and surface expansion in the area between Heihe fault and Wuliangshan fault. Wanding fault, east of Longling-Ruili fault, east of Nantinghefault, Heihe fault, Lancang fault and Xiaojiang fault extend across Honghe fault to Menglian-Mengxing fault area has high shear strain rate, up to 50 nstrain/a. The focal mechanism solutions of the study area from 1976 to 2020 are collected, the regional damping stress inversion method is used to invert the stress tensor, and the R-value distribution reflecting the relative size of the principal stress is attained. The inversion results show that the southwest of Yunnan is affected by the principal compressive stress in the NNE—SSW direction, and the area is in the state of strike slip stress as a whole, The North-south fault near Tengchong volcano in the east of Dayingjiang fault has the characteristics of normal fault and oblique slip. The magnitude of principal stress in Southwest Yunnan is obviously different in space. The south of Nantinghe fault is the area with the largest principal stress, and its strike slip characteristics are more obvious. In addition, this study also collects seismic wave anisotropy data in the study area, compares the angle between the direction of the principal stress axis and the principal strain rate axis, the angle between the direction of the principal stress axis and the polarization direction of Pms fast wave, the angle between the direction of the principal stress axis and the polarization direction of XKS fast wave, the angle between Pms fast wave direction and XKS fast wave polarization direction, we obtains the structural deformation differences between the surface and middle upper crust, middle upper crust and lower crust, lower crust and upper mantle respectively, It is found that the lower crust and upper mantle are decoupled in Southwest Yunnan. At the same time, it is found that the difference in deformation between the lower crust and the middle upper crust near Longling-Ruili fault, Nantinghe fault and Menglian fault, indicating that the lower crust and the middle upper crust are decoupled in this area, and there may be a hot material flow channel in the lower crust, with localized lower crustal flow in partial zones of southwestern Yunnan.

In this paper, we transferred the normal height of ALOS AW3D30 to GNSS ellipsoidal height based on the EGM96, and analyzed the precision of this digital elevation model in the northeastern Qinghai—Xizang Plateau (QXP) combining with the ground GNSS measurements. We also calculated the gravitational free air anomalies in the northeastern QXP based on the campaign gravity measurements, and evaluated the precision of the EIGEN-6C4 gravitational mode. The results showed that the standard deviation of the difference between ALOS AW3D30 and GNSS results is 3.4 m, which demonstrated that the precision of the ALOS AW3D30 in this region was better than 4 m. The measured gravity anomalies showed generally negative with local positive values, ranging from -177×10^-5 to 166×10^-5 m·s^-2. The modeled gravity free air anomalies of the EIGEN-6C4 changed from -163×10^-5 to 142×10^-5 m·s^-2. The difference between the EIGEN-6C4 gravity anomalies and measured results span from -119×10^-5 to 63×10^-5 m·s^-2, which indicated that the EIGEN-6C4 gravity anomalies were poorly constrained in the northeastern QXP and the measured gravity anomalies should be combined when analyzing gravity anomalies in this region. At last, we fused the EIGEN-6C4 modeled gravity anomalies with the measured ones by the “remove-restore” method, and improved the precision of gravity free air anomalies in the northeastern QXP, especially around the eastern Qilian— Haiyuan Fault and its surrounding areas. The improved regional gravity anomalies can provide a reference for the study of gravity field in the northeastern QXP.

Haiyuan fault zone is a large NWW-trending boundary fault zone located on the northeastern margin of the Qinghai-Tibet Plateau. In order to study the crustal velocity structure and deformation characteristics of this region, the Cross-Haiyuan Fault Zone temporary Seismic Array set up by the Institute of Earthquake Forecasting, China Earthquake Administration (SACHY-Array) and vertical continuous waveform data of a total of 61 stations (including 40 temporary stations and 21 permanent stations) of permanent seismic stations in the study area were used. Based on the background noise cross-correlation method, the surface wave phase velocity dispersion curve were extracted, and the Rayleigh wave phase velocity and azimuth anisotropy images with a period range of 5~30 s and a resolution of 1°×1° were obtained by inversion. The results show that, within a short period of 5~10 s, the eastern Hexi Corridor transition zone, the northwestern Ordos Block, and the southern Yinchuan graben all present low-velocity anomalies, while the eastern Qilian orogenic fold belt presents relatively high-velocity; the southwest side of the Haiyuan fault zone presents fast waves the polarization directions are NWW and NW. The polarization direction of fast waves in the western margin of Ordos Block and its adjacent areas is mainly near NS, and the anisotropy direction is basically consistent with the regional fault strike. In the period of 15~30 s, the low-velocity anomalies in the east of the Hexi Corridor transition zone and the southern part of the Yinchuan Graben gradually weakened, and the range continued to decrease. Around the 15 s period, the low-velocity bodies in the east of the Hexi Corridor transition zone were staggered under the Yantongshan fault zone. There are high-speed anomalies in the center of the Longzhong Basin and the southwestern margin of Ordos; the direction of azimuthal anisotropy is basically consistent with the short period, but the anisotropy intensity is weak. The authors believe the Haiyuan fault zone is a transition zone between high and low velocities. The Hexi Corridor transition zone on the north side of the fault zone exhibits low velocity anomalies, while the Qilian fold orogenic belt on the south side exhibits high velocity anomalies, indicating that the Haiyuan fault zone and its adjacent areas have complex crustal structures and may have local crustal deformation and destruction; the azimuthal anisotropy of the crust is an important constraint index of crustal deformation, indicating that the anisotropy of the Haiyuan fault zone and its adjacent areas may be mainly caused by the shear deformation of the crust in the northeastern margin of the Qinghai-Tibet Plateau, the north-east pushing of the Qinghai-Tibet Plateau is the main source of dynamic in this area.

In this paper, the phase velocity distribution images of Southwest Yunnan (21.0°N~29.5°N, 97.0°E~103.5°E) are obtained with continuous seismic records at 190 mobile stations in the first periods of ChinArray (2011-01—2014-06) and 35 broadband permanent stations of Yunnan regional network. The ChinArray was setup in the southeast of the Qinghai-Xizang Plateau. Inversion results reveal significant lateral heterogeneity in the velocity structure of the crust and upper mantle in the study area. Short period images (6~12 s) indicate a strong correlation between upper crust velocity and surface geological structures. The low velocity anomaly in Tengchong volcanic area and its southern side is related to the magmatic sac or partial melt in the upper crust, and the shape is different in depth. A high-velocity anomaly is observed between Chenghai fault and Puduhe fault, consistent with the body wave travel time inversion results. The inversion indicates a large high-velocity anomaly in Panxi area, potentially related to basalt uplift in this area. Phase velocity images with periods of 14~25 s reveal low-velocity anomalies in the Sichuan-Yunnan rhombic block and Tengchong volcanic area, suggesting the presence of lower crustal flow. Comparing 14~40 s images, due to variations in crustal thickness, the high-velocity anomaly gradually increases in the southern region. Furthermore, the low-velocity anomaly in the Tengchong volcanic area extends from the surface to the upper mantle, but it does not appear to be continuous with the low velocity anomaly in the Sichuan-Yunnan rhombic block. This low velocity anomaly may be a channel of mantle heat flow. The intersection of the Tengchong fault and Longling fault appears to be the boundary between high and low-velocity anomalies. Two M≥7 earthquakes in 1976 may occurred beneath this area.

In recent years, deep learning technology has become increasingly prevalent in seismic phase picking and earthquake location research. This paper employs the EQTransformer, a deep neural network, to pick P and S arrivals from continuous data recorded by 34 digital seismic stations in the Baihetan reservoir area from 2016 to 2018. Phase association and preliminary localization by REAL, followed by optimization of earthquake location using VELEST and hypoDD. The research reveals that seismic phase picking based on deep learning is significantly more efficient than traditional manual methods in the Baihetan Reservoir area. The accuracy of the EQTransformer-picked P and S first-arrivals is comparable to that of manually picked phases, with average time differences of 0.03 s and 0.07 s, conforming to a normal distribution. The number of events after preliminary location by REAL (13815) is nearly twice of the routine catalog (7862), and we ulfimately obtain high-precision locations of 7108 earthquakes by the hypoDD. The estimated magnitude is on average 0.27 lower than that in the routine catalog, and the magnitude difference is primavily within 0.7. The minimum magnitude of completeness is changed from M_L1.4 in the routine catalog to M_L0.6+0.27, effectively filling the magnitude gap of the routine catalog and enriching the data on moderate and small earthquakes in the Baihetan reservoir area.

The geological structure in the central part of the Songliao Basin is well-developed, with a complex system of intersecting faults. In the past, the area has experienced several strong earthquakes, with the main seismogenic structures being the Fuyu-Zhaodong Fault and the Hulanhe Fault within the Songliao Basin. Soil hydrogen gas measurements were conducted on the Fuyu-Zhaodong Fault and the secondary faults of Hulanhe Fault, namely the Suihua-Menggushan Fault, revealing significant changes in soil hydrogen gas levels near the active fault zones. The results show that ① the hydrogen gas anomaly threshold concentration ratio for each measuring line in the Fuyu-Zhaodong Fault was 1.8~2.3, with an anomaly peak concentration ratio of 3.0~6.4; ② the hydrogen gas anomaly threshold concentration ratio for each measuring line in the Suihua-Menggushan Fault was 1.8~2.5, with an anomaly peak concentration ratio of 3.5~5.7; ③ comparison of the abnormal background, threshold, and peak concentrations between the two faults indicated that the values for the Fuyu-Zhaodong Fault were significantly higher than those for the Suihua-Menggushan Fault. This may be related to the recent significant seismic activity of the Fuyu-Zhaodong Fault, possibly containing information about the enhancement of pre-earthquake tectonic activity. ④ Significant changes in soil hydrogen gas were observed near active fault zones, which may be related to the nature of the two faults, geological structure units, local stress environments, and the strength of fault activity.

Background information and precursor information are the main observation data of a station. It is a key link to improve the effectiveness of seismic activity monitoring and prediction through accurate identification and correct extraction of precursor information. In this paper, the standard-year dynamic curve morphological analysis method was used to detrend the daily water (gas) radon value of 8 underground fluid stations in Gansu Province. The smoothing of the “forward slip” of the years of daily value was carried out and the mean square error as the upper and lower limits of the smooth curve was calculated. The abnormal values exceeding the upper and lower limits were used as anomalous information for earthquake case statistics, so as to gain earthquake reflecting ability and spatial response characteristics of the precursory anomalies of water (gas) radon at 8 underground fluid stations. It is concluded from the study that the standard annual dynamic curve analysis method could be an effective method to identify the precursory short-impending anomaly information of groundwater (gas) radon, and expected to play an important technical support role in the precursory data analysis and abnormal processing of underground fluids.

With the accumulation of four-component borehole strain observation data and the increase of seismic stations, it is important to improve the efficiency of processing and analysis of these data. Due to the high resolution of borehole strain observation data in an ultra-wide frequency band, borehole strain observation data are significant for earthquake and earthquake prediction research. Using the normal stress petal diagram to display the change of in-situ stress, we can not only qualitatively analyze the relative in-situ stress change at the station, but also quantitatively read the normal stress observed in any direction at a certain time. In this paper, this method is combined with data visualization technology and chart data combined with map display as the main body of the platform builds a visualization analysis platform of four-component strain data based on B/S architecture. This platform can intuitively display the spatiotemporal variation law of in-situ stress recorded by the borehole strain gauge and realize the visual analysis and interaction of borehole strain data and shared access by multiple users.