With the continuous increasing density of the seismic network and the improvement of the seismograph observation capability, the number of observed seismic events has increased dramatically and the location accuracy has been continuously improved. Therefore, obtaining fault geometry and its parameters from massive seismic data has become an essential method for seismogenic structure research. At present, in the research of obtaining faults and their parameters based on seismic data, there are two main methods of selecting data: One is to select seismic data empirically based on the understanding of fault structures and the spatial distribution of seismic data, and then fit the fault plane from these data. However, it depends on prior information, i.e. the knowledge of existing fault structures and the linear distribution of earthquakes, and it is difficult to process relatively poor linear trends. The other is based on the spatial clustering of seismic data, which adopts unsupervised clustering technology in machine learning to select data. This method avoids the dependence on experience and is more suitable for fault segment data obtained from massive seismic data. Fault parameters can be inversed by fault segment data to determine the fault structure and give its quantitative parameters. However, the current clustering technique for obtaining fault parameters has some limitations, such as the selection of the optimal parameters being difficult, data with different densities being dealt with by the same parameters, and poor method generality. In order to automatically identify faults and obtain fault parameters based on the spatial distribution of earthquakes, and avoid the aforementioned limitations, a new method based on the improved DBSCAN algorithm is presented in this study.
The method proposed in this study uses the k-average nearest neighbor method(K-ANN)and the mathematical expectation method to generate the candidate sets of eps and minPts threshold parameters, which are selected as optimal parameters based on the density hierarchy stability. Considering the spatial density differences of seismic events on different faults and the same fault, this study performs layer-by-layer density clustering from high density to low density. First, the above steps achieve the automatic selection of optimal parameters for clustering and identifying fault segments. Secondly, the fault parameters of the identified fault segments are calculated by the combination of the simulated annealing(SA)global search method and the local search method of Gaussian Newton(GN). Then, the adjacent similar fault segments are merged. Finally, the faults and their parameters are obtained.
The reliability of the automatic fault identification method was verified by synthetic data and the double-difference location catalog of Tangshan area, China. The following results were obtained: Ⅰ. The improved DBSCAN algorithm can automatically identify the fault segments, which is verified by the application of synthetic data and the double-difference location data of the Tangshan area. Ⅱ. Based on the double-difference location data of the Tangshan area, eight fault segments were identified using the improved DBSCAN algorithm. The specific names of the 8 faults are as follows: Douhe fault segment, Weishan-Fengnan fault segment, Luanxian-Laoting fault segment, Lulong fault segment, Xujialou-Wangxizhuang fault segment, Luanxian fault north segment, Leizhuang fault segment, and Chenguantun fault segment, and their strike and dip angle are 229.1°, 230.4°, 132.2°, 31.7°, 191.3°, 31°, 229.5°, 84.9°, and 51.6°, 88.4°, 89.3°, 88.6°, 88.4°, 88.2°, 73.8° and 85.4°, respectively. The parameters of the first five faults are mostly consistent with those of previous research results. The last three faults are the newly identified faults in this study based on the seismic catalog, and the parameters of two of them have been confirmed by previous research results or focal mechanism parameters on the faults.
In a word, the improved DBSCAN algorithm in this study can realize fault segment automatic identification, but there are still some problems that need to be improved urgently. In the follow-up research, we will continue to improve the automatic fault identification method and increase its ability of automatic fault identification so as to provide more accurate fault data for related research.

The South China block, located in the east of the Eurasian plate, mainly consists of the Yangtze block and the Cathaysia block. The South China block is bounded by the eastern margin of the Qinghai-Tibet Plateau in the west, the Qinling-Dabie orogenic belt in the north, and its eastern boundary extends from the southeast coast to the north, through the Taiwan Strait, and then along the Ryukyu Island arc to the west direction. The neotectonic movement of the South China block is intense. It is not only the continental margin with the most active crustal growth and continental accretion, but also the tectonic belt with the most intense core-mantle mass transfer and the coupling zone of the inner layers of the Earth. Therefore, the crust-mantle velocity structure of the South China block and its formation and evolution have always been a hot topic in earth science research.

In this paper, we collected continuous vertical component broadband seismic data between January 1, 2010 and December 31, 2012 from the regional networks of 609 stations and used ambient noise tomography method to inverse the three-dimensional S-wave velocity structure of South China block and its adjacent area. Firstly, the seismograms are cut into daily segments and decimated at a sampling rate of 1Hz. After the removal of the mean, trend, and instrument response, a 3~150s band-pass filter is applied. In order to reduce the effect of earthquakes and instrumental irregularities on cross-correlations, we normalized the seismograms with a time-frequency normalization method. Then, we computed daily cross-correlations for each station pairs and stacked all of them by using normalized linear stacking method to obtain cross-correlation functions. Next, the phase velocity dispersion curves of Rayleigh surface wave were extracted by frequency-time analysis method. Finally, the three-dimensional S-wave velocity structure of the study area was obtained by using nonlinear Bayesian Monte Carlo inversion method.

The results show that the S-wave velocity distribution has a good correlation with surface geological and tectonic features, and could clearly reveal the lateral velocity variation in the crustal. The shallow S-wave velocity in basin and graben area presents low velocity anomaly due to the influence of sedimentary layer. The high velocity anomaly exists in the middle and lower crust of Jianghan Basin and Sichuan Basin, indicating that the middle and lower crust of these basins are cold and hard. Due to the phenomenon of arching existing in the upper mantle of Sichuan Basin, the S-wave velocity of the crust and mantle is relatively high in the upper mantle, meanwhile, the S-wave velocity in the center of the basin is higher than that in the edge. Although both the Yangtze block and Cathaysia block are located in the South China block, their upper mantle S-wave velocity structures are quite different due to their different evolutionary processes. The high S-wave velocity of the Yangtze block indicates the internal structure of the block is relatively stable, while the low S-wave velocity of the Cathaysia block indicates the strong magmatic activity during its evolution. The crust-mantle S-wave velocities in the west of the southwest boundary of the South China block show low velocity anomalies, which may indicate the existence of asthenosphere in the middle and lower crust of the eastern margin of the Qinghai-Tibet Plateau. The S-wave velocity structures of the eastern and western parts of the Qinling-Dabie orogenic belt are quite different, and the crustal thickness transition zone is the boundary of the S-wave velocity structure, which is high in the east and low in the west. The crust-mantle S-wave velocity of Ordos block is relatively high, indicating that the inner structure of ordos block is relatively stable. However, the S-wave low velocity anomaly in the upper mantle at the southwest corner of the Ordos Basin may indicate that the heat flow of the upper mantle of the North China Craton has begun to “invade” the Ordos lithosphere.

A M_S6.0 earthquake with shallow focal depth of 16km struck Changning County, Yibin City, Sichuan Province at 22:55: 43(Beijing Time)on 17 June 2019. Although the magnitude of the earthquake is moderate, it caused heavy casualties and property losses to Changning County and its surrounding areas. In the following week, a series of aftershocks with M_S≥4.0 occurred in the epicentral area successively. In order to better understand and analyze the seismotectonic structure and generation mechanism of these earthquakes, in this paper, absolute earthquake location by HYPOINVERSE 2000 method is conducted to relocate the main shock of M_S6.0 in Changning using the seismic phase observation data provided by Sichuan Earthquake Administration, and focal mechanism solutions for Changning M_S6.0 main shock and M_S≥4.0 aftershocks are inferred using the gCAP method with the local and regional broadband station waveforms recorded by the regional seismic networks of Sichuan Province, Yunnan Province, Chongqing Municipality, and Guizhou Province. The absolute relocation results show that the epicenter of the main shock is located at 28.35°N, 104.88°E, and it occurred at an unusual shallow depth about only 6.98km, which could be one of the most significant reasons for the heavier damage in the Changning and adjoining areas. The focal plane solution of the Changning M_S6.0 earthquake indicates that the main shock occurred at a thrust fault with a left-lateral strike-slip component. The full moment tensor solution provided by gCAP shows that it contains a certain percentage of non-double couple components. After the occurrence of the main shock, a series of medium and strong aftershocks with M_S≥4.0 occurred continuously along the northwestern direction, the fault plane solutions for those aftershocks show mostly strike-slip and thrust fault-type. It is found that the mode of focal mechanism has an obvious characteristic of segmentation in space, which reflects the complexity of the dislocation process of the seismogenic fault. It also shows that the Changning earthquake sequences occurred in the shallow part of the upper crust. Combining with the results from the seismic sounding profile in Changning anticline, which is the main structure in the focal area, this study finds that the existence of several steep secondary faults in the core of Changning anticline is an important reason for the diversity of focal mechanism of aftershock sequences. The characteristics of regional stress field is estimated using the STRESSINVERSE method by the information of focal mechanism solutions from our study, and the results show that the Changning area is subject to a NEE oriented maximum principal stress field with a very shallow dipping and near-vertical minimum principal stress, which is not associated with the results derived from other stress indicators. Compared with the direction of the maximum principal compressive stress axis in the whole region, the direction of the stress field in the focal area rotates from the NWW direction to the NEE direction. The Changning M_S6.0 earthquake locates in the area with complex geological structure, where there are a large number of small staggered fault zones with unstable geological structure. Combining with the direction of aftershocks distribution in Changning area, we infer that the Changning M_S6.0 earthquake is generated by rupturing of the pre-existing fault in the Changning anticline under the action of the overall large stress field, and the seismogenic fault is a high dip-angle thrust fault with left-lateral strike-slip component, trending NW.

Under the background of thrusting stress regime, a large number of strike-slip earthquakes occurred on the Miyaluo Fault during the Wenchuan earthquake sequence process, which is in the southern part of the Longmenshan Fault. In order to find the cause of their occurrence, stress tensors in subregions near the Miyaluo Fault are estimated. The result shows that in both north and south side of the Miyaluo Fault, the direction of principal compressive stress is nearly perpendicular to the Longmenshan Fault, and its dip is nearly horizontal, and the direction of tensile stress is nearly vertical. While in the Miyaluo fault zone, the direction of principal compressive stress is SWW-NEE, and its dip is nearly horizontal, the direction of principal tensile stress is NNW-SSE, also its dip is nearly horizontal. It is consistent with sinistral shear stress state in the Miyaluo fault zone. It was referred that the behavior of Miyaluo Fault during the Wenchuan earthquake sequence process was caused by tearing effect generated from unbalanced forces of two sides of the fault. To understand the rupture mode of the aftershocks in subregions as described above, the total seismic moment tensors are estimated by adding the corresponding component separately of the seismic moment tensor of aftershocks in each region. The result shows the similar trend of total seismic moment tensor components in the north and south side of the Miyaluo Fault(indicating the consistency of rupture mode in the north and south side of the Miyaluo Fault), and most seismic moment tensor components in the south side is higher than that in the north side, especially the compression component perpendicular to Longmenshan Fault and expansion component in the vertical direction. It indicates that thrusting component in the southeast direction in the south side is greater than that in the north side, and the thrusting difference causes the sinistral tearing effect of the Miyaluo Fault. We also find that the sinistral tearing component of the Miyaluo Fault is the same order of magnitude with the thrusting difference of its two sides, which indicates that the tearing effect of Miyaluo Fault can be completely explained by thrusting difference of its two sides. According to the analysis, we put forward the dynamic model of the Miyaluo Fault, which can explain the above phenomenon.

A collapse happened in Pingyi County, Shandong Province, on December 25, 2015. The displacement field, stress field and Coulomb failure stress change on the Mengshan frontal fault generated by the collapse are calculated by using point collapse model in isotropic medium. The result shows that: (1) The maximum horizontal displacement is located at the center of the collapse with value of~18mm. The horizontal displacements are greater than 1mm within~5km of the collapse with its direction pointing to the collapse center. The maximum subsidence is located at the center of the collapse with the value of 4mm. The subsidence is greater than 1mm within ~3km of the collapse. The displacement field decays so rapidly that can be ignored at far away from the collapse for the shallow source, which caused local displacement field. (2) Influenced by the free surface, the contraction area stress within ~5km of the collapse with the order of 1000Pa and expansion area stress in farther away areas at depth of 2km are estimated. the expansion area stress of 1000Pa is estimated at the~5km from the collapse center. Then the expansion area stress decays to 100Pa at the distance of ~10km from the collapse. The maximum compressive and extensional principal stresses are estimated as 10000Pa at the depth of 2km. The compressive stress axes present radical direction pointing to the collapse within ~5km of the center. In farther away from the collapse, The extensional principal stress axes present radical direction pointing to the center of the collapse. With farther distance to the collapse, the compressive and extensional stress decay rapidly to the order of 100Pa. (3) The Coulomb failure stress on the northwestern part of the Mengshan frontal fault, which is known as active segment of the Mengshan frontal fault, is decreased by the collapse with maximum value of 2500Pa. Whereas, the Coulomb failure stress on the southeastern part of the Mengshan frontal fault, which is known as left-lateral normal slip fault segment in Quaternary period, is increased by the collapse with maximum of 2400Pa, to which attention would be paid in seismic hazard analysis.

In this paper,we present a method which allows to calculate the mean stress field according to the total seismic moment released by earthquakes.The exact method is as follows: First,we calculate the scalar seismic moment released by each earthquake according to the statistical relationship between earthquake magnitude and its seismic moment; Second,we calculate the seismic moment tensor released by each earthquake according to the relationship between focal mechanism solution and seismic moment tensor; Then,we can get the total seismic moment tensor released in a specific time period of the study area; Finally,we calculate the eigenvector and eigenvalue of the total seismic moment tensor,the obtained eigenvector is corresponding to the mean stress field direction released by the study area. We tested the method by using the synthetic focal mechanism to which random error was added and with the focal mechanism data of Tangshan aftershock zone.The testing results show that,the released stress field of the study area obtained by our method is in consistency with the regional stress field. So our method can be applied to solve regional stress field.The more focal mechanism data used,the more stable the result would be,and closer to the real regional stress field. One of the advantages of this method is that it uses magnitude as the weight of each earthquake,so the contribution difference of the earthquake size in the stress field inversion can be better reflected. Another advantage is that it does not need to know which nodal plane of the focal mechanism is the real fault plane when we calculate stress field.

In this research,we made a primary research on the Coulomb stress triggering of the March 11,2011 M_W=9.0 Tohoku earthquake sequence by using the software of Coulomb 3.2,the earthquake rupture models obtained by Hayes and Guangfu Shao et al. , and the aftershocks data from Harvard CMT catalogue(Centroid Moment Tensor)and Japan F-net catalogue. Our results suggest that: Firstly,the M_W7.2 foreshock,which occurred on March 9,has stress triggering effect on the M_W9.0 main shock; Secondly,the statistical result of the aftershocks triggered by the main shock shows that different statistical results would be obtained when using different main shock rupture model,aftershock catalogue,equivalent friction coefficient and different nodal plane of focal mechanism as the receive plane. The minimum and maximum triggering rates of main shock to aftershock are 56.8%and 75.3%,respectively; and thirdly,when calculating the Coulomb stress by using earthquake focal mechanism,the shear stress on the two nodal planes is the same theoretically. However,in actual calculation,the shear stress would be different between the two nodal planes,due to the non-orthogonality of the two nodal planes or rounding off decimal places in the focal mechanism results. But,the difference of the shear stress on the two nodal planes is relatively small. More attention should be paid on the selection of receiver fault plane from the two nodal planes,when discussing the stress triggering of one specific earthquake or making statistics of aftershock triggering rate,because the selection of receiver fault plane would have a certain effect on the shear stress,and have greater effect on the normal stress,thus the Coulomb stress would be affected.

The precisely located earthquake catalogue is important to seismicity, seismic tomography and crustal stress inversion studies. It also has great application value in rapid report of an earthquake that just occurred. By considering the arrival time uncertainty, and the constraints on station elevation and seismic depth, we propose a relatively accurate method to estimate hypocentral location and its uncertainty based on inversion theory. Our method can combine the arrival times of Pg wave, Sg wave, Pn wave and Sn wave in hypocenter location, so it increases the location accuracy by involving more data; and it can be also used in local and regional earthquake location simultaneously. In order to test our location method, we located earthquakes by using the simulated data with different uncertainty of Pg,Sg,Pn,Sn arrivals. The result shows that the location determined by using our method is more accurate than that by using other method. We apply it to earthquakes occurring in the period from 2001 to 2008 in Sichuan area, and obtained a more clustered hypocentral distribution convergent to the fault zones. The result provides a solid foundation for studies of seismicity, geometry of the active faults and seismic tomography in Sichuan region. It is also helpful to study the seismicity precursors before the Wenchuan earthquake.