Focal Mechanism and Seismogenic Structure of the MS 5.1 Qingbaijiang Earthquake on February 3, 2020, Southwestern China

The 3 February 2020 MS 5.1 Qingbaijiang earthquake, southwestern China, is the closest recorded MS ≥ 5.0 event to downtown Chengdu City to date, with an epicentral distance of only 38 km. Here we analyze seismic data from the Sichuan and Chengdu regional seismic networks, and employ a multi-stage location method to relocate the earthquakes that have occurred along the central and northern segments of the Longquanshan fault zone since 2009, including the MS 5.1 Qingbaijiang earthquake sequence, to investigate the seismogenic structure of the region. The relocation results indicate that the seismicity along the central and northern segments of the Longquanshan fault zone has occurred mainly along the eastern branch since 2009, with the hypocentral distribution along a vertical cross-section illustrating a steep, NW-dipping parallel imbricate structure. The terminating depth of the eastern branch is about 12 km. The distribution of the MS 5.1 Qingbaijiang earthquake sequence is along the NE–SW-striking Longquanshan fault zone. The aftershock focal depths are in the 3–6 km range, with the mainshock located at 104.475°E, 30.73°N. Its initial rupture depth of 5.2 km indicates that the earthquake occurred above the shallow decollement layer of the upper crust in this region. The hypocentral distribution along the long axis of the aftershock area highlights that this earthquake sequence occurred along a fault dipping at 56° to the NW. Our surface projection of the inferred fault plane places it near the eastern branch of the Longquanshan fault zone. We infer the MS 5.1 mainshock to be a thrust faulting event based on the focal mechanism solution via the cut-and-paste waveform inversion method, with strike/dip/rake parameters of 22°/36°/91° and 200°/54°/89° obtained for nodal planes I and II, respectively. We identify that the seismogenic fault of the MS 5.1 Qingbaijiang earthquake lies along the eastern branch of the Longquanshan fault zone, and nodal plane II represents the coseismic rupture plane, based on a joint analysis of the event relocation results, mainshock focal mechanism, and regional geological information. Our study provides vital information for assessing the seismic hazard of the Longquanshan fault zone near Chengdu City.


INTRODUCTION
The ongoing Cenozoic collision between the Indian and Eurasian plates has contributed to the uplift of the Tibetan Plateau (Molnar and Tapponnier, 1975;Houseman and England, 1986;Yin and Harrison, 2000;Tapponnier et al., 2001). However, eastward growth of the Tibetan Plateau has been impeded by the tectonically stable Yangtze Block, with the Longmenshan fold-and-thrust belt (LFTB) and the associated uplift of the Sichuan foreland basin and Longquanshan anticline forming in the eastern part of the Tibetan Plateau (Tapponnier and Molnar, 1977;Tapponnier et al., 1982;Chen et al., 1994;Burchfiel et al., 1995;Chen and Wilson, 1996;Kirby et al., 2002). The LFTB is a geologically active area of the eastern Tibetan Plateau, as indicated by the 2008 M S 8.0 Wenchuan and 2013 M S 7.0 Lushan earthquakes (Burchfiel et al., 2008;Hubbard and Shaw, 2009;Xu et al., 2009Xu et al., , 2013Yin, 2010). Numerous studies have focused on the Longmenshan fault zone following the 2008 Wenchuan earthquake Liu-Zeng et al., 2009;Li et al., 2010;Wang et al., 2010Wang et al., , 2011Wang et al., , 2012Zhang et al., 2010;Zhang, 2013), and also the structural characteristics and seismic hazard along the range front of the LFTB (Parsons et al., 2008;Toda et al., 2008;Hubbard and Shaw, 2009;Liu-Zeng et al., 2009;Hubbard et al., 2010). The Longquanshan fault zone, which forms the southeastern boundary of the Sichuan basin , has been inferred to possess a low level of seismicity that is characterized by smaller-magnitude earthquakes (Li et al., 2015;Wang and Lin, 2017); consequently, its geometric structure and seismic hazard have not been well studied. The occurrence of the 3 February 2020 M S 5.1 Qingbaijiang earthquake therefore provides an opportunity to study the structural characteristics and potential seismic hazard of the Longquanshan fault zone, since its epicenter and associated aftershocks have been located along the central and northern segments of the fault zone.
The M S 5.1 Qingbaijiang earthquake occurred adjacent to the highly populated and economically developed Chengdu Plain. It is the closest M S ≥ 5.0 event to downtown Chengdu City to date, with an epicentral distance of only 38 km. Previous studies have considered the Sichuan Basin to be a stable terrane that poses no risk of strong earthquakes (Xu et al., 2006;Hubbard et al., 2010). However, recent studies have revealed that moderate-to-strong earthquakes may also occur along the Longquanshan fault zone (Li et al., 2012Wang and Lin, 2017). Numerous concerns regarding the future seismic hazard of the Longquanshan fault zone have been raised following this event. However, the limited data means that more studies are needed to reassess the seismic hazard in the Longquanshan area.
Constraints on seismogenic structures and their associated earthquake mechanisms are helpful for seismic risk assessments. The characteristics of various geological structures can generally be recognized by the spatial distribution of relocated earthquakes (Michael, 1988;Shearer, 1997;Presti et al., 2008;Yi et al., 2015Yi et al., , 2017Yi et al., , 2019, and the geometric and kinematic characteristics of these seismogenic structures can be inferred from their associated earthquake focal mechanisms (Zheng et al., 2009;Lv et al., 2013;Yi et al., 2015Yi et al., , 2017Yi et al., , 2019Yi et al., , 2020. Here we analyze seismic phase and waveform data from the Sichuan and Chengdu seismic networks, and relocate the earthquakes that have occurred along the central and northern segments of the Longquanshan fault zone since 2009, which include the M S 5.1 Qingbaijiang earthquake sequence. We then analyze the earthquake distribution along the central and northern segments of the Longquanshan fault zone and the seismogenic structure associated with the M S 5.1 Qingbaijiang earthquake sequence. We confirm the structure via a joint analysis of the earthquake relocation results, mainshock focal mechanism solution, and existing regional geological data, and analyze the seismogenic mechanism. Our results are a first step in better understanding the seismogenesis of the Longquanshan fault zone, and provide essential data for future studies on the formation mechanism and seismic hazard of the fault zone.

GEOLOGICAL SETTING AND HISTORICAL EARTHQUAKES
The Longquanshan fault zone forms the eastern boundary of the western Sichuan foreland basin, extending along the eastern side of the Longmenshan thrust-nappe structure and the western front of the Sichuan Basin. This 200-km-long NE-SW-trending fault zone starts in Zhongjiang County at its northern end, transects Jintang and Renshou counties, and terminates near Xinqiao Town, Leshan County, at its southern end (Li and Zeng, 1994). The Longquanshan fault zone consists of eastern and western branches that are distributed along the eastern and western limbs of the Longquanshan anticline, respectively ( Figure 1); a conceptual model of the fault zone is shown in Figure 2. The western branch generally strikes NE20 • -NE30 • , and consists of faults that dip to the SE at 35 • -70 • and form an imbricate nappe structure, with these faults converging along the same main reverse fault at depth, as inferred from geological mapping (Figure 2; Deng et al., 1994) and seismic profiling (Wang et al., 2008;Li et al., 2013). The eastern branch is nearly parallel to the western branch, with a strike of NE10 • -NE30 • , and it consists of faults that dip to the NW at 28 • -82 • (Huang and Tang, 1995;Huang and Jiang, 2012;Li et al., 2013). The primary strata of the Longquanshan anticline are Jurassic and Cretaceous in age , the Longquanshan fault zone extends through these Jurassic and Cretaceous strata and terminates in the Triassic strata (Huang and Tang, 1995). The Longquanshan fault zone is divided into northern, central, and southern segments, and previous studies have generally defined the western branch as the main fault of the fault zone (Figure 1; Deng et al., 1994;Huang and Tang, 1995;Wang et al., 2008). The Longquanshan fault zone was originally activated during the formation of the Tibetan Plateau at the end of the Eocene , as inferred from seismological and geological investigations, geodetic deformation surveys (Huang and Tang, 1995;Zhang et al., 2003;Burchfiel, 2004), and focal mechanism solutions (Cheng, 1981). The secondary western Sichuan Block within the Tibetan Plateau continued to move to the SE and collided with the South China Block, resulting in the formation of small-scale folds in the interior of the Sichuan Basin .   Deng et al., 1994). The western branch consists of SE-dipping faults that form an imbricate nappe structure, with these faults converging along the same main reverse fault at depth. The eastern branch consists of NW-dipping faults with a nearly parallel arrangement.
The Longquanshan fault zone is a seismic zone that has generally hosted moderate-to-weak earthquakes, with segmented seismicity across the fault zone. Only two recorded M S ≥ 5.0 earthquakes have occurred along the southern segment of the fault zone (Figure 1), including the destructive 24 January 1967 M S 5.5 Renshou earthquake, which is the largest event along the fault zone to date; this event had a 4 km focal depth and epicenter located on the southeastern side of the western branch (Cheng, 1981;Xu et al., 2006). Fault gouge dating has revealed that the Longquanshan fault zone was active during the early to late Pleistocene, with the mid-Pleistocene being the most active period (Huang and Tang, 1995). Late Pleistocene activity was limited to the southern and central segments of the western branch, which might explain why moderate earthquakes have frequently occurred along these segments (Huang and Tang, 1995).

RELOCATION OF THE M S 5.1 QINGBAIJIANG EARTHQUAKE SEQUENCE Seismic Data and Relocation Method
The seismic phase data used for the earthquake relocation are from the Sichuan (black triangles in Figure 3) and Chengdu (green triangles in Figure 3) regional seismic networks. The M S 5.1 Qingbaijiang earthquake sequence occurred in an area with dense, distributed seismic station coverage, thereby providing reliable data for our study. There were 55 broadband seismic stations within a 300 km epicentral distance of the mainshock, with an additional two portable short-period (SP) seismometers deployed near the epicenter immediately after the mainshock (red triangles in Figure 3). The minimum magnitude of completeness for the M S 5.1 earthquake sequence was estimated to be 1.4 by the goodness-of-fit test (Wiemer and Wyss, 2000).
The events in the study area since the M S 5.1 mainshock have been highly concentrated near the mainshock. We selected the events with epicentral distances within 15 km of the mainshock as aftershocks. The aftershocks of the M S 5.1 event were sparse, with a scattered distribution of only 75 M L ≥ 0.1 aftershocks recorded by the end of October 2020 (Figure 4). The magnitudetime (M-t) plot ( Figure 4A) shows that the aftershocks of the M S 5.1 event occurred mainly during the first 2 months, whereas the aftershocks after 1 April 2020 were scattered. The frequencytime (N-t) plot ( Figure 4B) shows that the monthly aftershock frequency of the sequence has been very low since April 2020, which indicates a rapid decay in the aftershock frequency.
We collected the phase data for 281 earthquakes since 2009 along the central and northern segments of the Longquanshan fault zone to obtain the necessary number of earthquakes for relative locations, and analyzed both the temporal and spatial characteristics of the seismicity along these segments. We relocated these earthquakes using the multi-stage method developed by Long et al. (2015). The details of our procedure are as follows.
We first located the initial earthquake locations using HYPOINVERSE2000 (Klein, 1989) based on the eastern Sichuan velocity model (Zhao and Zhang, 1987). We then selected the events that were recorded at six or more seismic stations and possessed a station distribution with a maximum azimuth gap of less than 180 • . The lower limit number of observed stations in our study region was set to six, based on a series of test results, whereby the location accuracy was heavily reduced for fewer stations and the amount of relocated events decreased significantly with higher station limits, both of which inhibited the ability to accurately determine the geometry of the seismogenic fault. We calculated a new local one-dimensional (1-D) velocity model ( Table 1) and station corrections via the VELEST program (Kissling et al., 1994(Kissling et al., , 1995 prior to the event relocations. We then employed hypoDD (Waldhauser and Ellsworth, 2000) to relocate the events using this new velocity model and applying the station corrections. We set a 3 km search radius to identify any earthquake clusters, since there were no tight earthquake distributions in the study area.
We relocated 199 earthquakes, including 60 events that were associated with the M S 5.1 Qingbaijiang earthquake sequence. The average errors in the horizontal and vertical directions were 0.6 and 0.8 km, respectively, and the travel time residual was 0.2 s. The mainshock was located at 104.475 • E, 30.731 • N, with a focal depth of 5.2 km.

Analysis of the Relocation Results
The dashed rectangle in Figure 5A  zone is reflected in these observations. The aftershocks of the M S 5.1 Qingbaijiang earthquake were ∼10 km southwest of the Xinglong earthquake cluster, and the epicenter of the M S 5.1 mainshock was located between the eastern and western branches of the Longquanshan fault zone, ∼5 and 6 km from the eastern and western branches, respectively. The aftershock distribution strikes NE-SW, parallel to the Longquanshan fault zone, and is ∼6.0 km long and ∼2.5 km wide. The M S 5.1 mainshock was located just north of the central part of the aftershock area ( Figure 5B). Here we define 1 March 2020 as the boundary between the earlier and later aftershocks of the M S 5.1 mainshock. The early aftershocks in February were generally smaller-magnitude events that were distributed to the southwest of the M S 5.1 mainshock (red circles in Figure 5B), whereas the later aftershocks were relatively larger-magnitude events that were distributed to the east of the mainshock and included all of the relocated M L ≥ 2.0 earthquakes (green circles in Figure 5B).
We provide three depth profiles (A-A' , B-B' , and C-C' in Figure 5A) to further understand the seismogenic structure associated with the 2020 Qingbaijiang earthquake. These profiles focus on the M S 5.1 Qingbaijiang earthquake sequence (B-B' and C-C' in Figure 6) and the earthquake clusters in the Xinglong-Sanhe-Shisun area (A-A' in Figure 5C).
Profile A-A' is perpendicular to the strike of the Longquanshan fault zone and intersects the main earthquake cluster in the Xinglong-Sanhe-Shisun area ( Figure 5A). All of the relocated events within 5 km of the profile are plotted to discern the spatial distribution of the earthquake clusters. Three clusters are identified, which appear to form a parallel imbricate feature consisting of three steep NW-dipping faults FIGURE 3 | Distribution of seismic stations (triangles) and known faults (red lines) around the epicenter of the M S 5.1 Qingbaijiang earthquake (star). There were 55 broadband seismic stations (black and green triangles) and two portable SP seismometers (red triangles) within a 300 km epicentral distance of the mainshock.   with focal depths concentrated in the upper 12 km of the crust. The two earthquake clusters along the eastern part of the profile are relatively small, distributed below Sanhe Town and Shisun Town, respectively, and not associated with surface ruptures. These two clusters occurred along planes dipping at ∼60 • to the NW. The largest earthquake cluster near Xinglong Town has a "shovel-shaped" distribution that is nearly vertical in the uppermost crust and NW-dipping below 8 km depth. We suggest that the distribution of these earthquake clusters, which are located just below the surface expression of the eastern branch, reflects the geometry of this branch, which is consistent with regional geological and geophysical survey results (Huang and Tang, 1995). Profiles B-B' and C-C' capture the short-and long-axis distributions of the M S 5.1 Qingbaijiang earthquake aftershock area, respectively, showing all of the aftershocks that fall within 5 km of the profiles (Figure 6). The B-B' profile ( Figure 6A) highlights a NW-dipping feature, with the aftershock distribution spanning depths of 3-6 km and the mainshock and most of the aftershocks being located above 5 km depth. The larger aftershocks (M L ≥ 2.0) were deeper than the smaller aftershocks. Beneath the Longquanshan anticline there is a decollement layer in the upper crust at 6-9 km depth Jia et al., 2010;Li et al., 2015). It can therefore be inferred that the M S 5.1 Qingbaijiang earthquake and its aftershocks occurred above this shallow detachment surface in the upper crust. The lack of aftershocks near the surface may indicate that the rupture did not extend to the surface. The distribution of the entire sequence in this section indicates that the M S 5.1 Qingbaijiang earthquake occurred along a seismogenic fault plane dipping 56 • to the NW, which is consistent with the dip direction of the eastern branch. If we extend this fault plane to the surface, it appears to meet the eastern branch fault. Therefore, we can infer that the M S 5.1 Qingbaijiang earthquake ruptured the eastern branch of the Longquanshan fault zone at 3-6 km depth. The C-C' profile ( Figure 6B) shows a northward deepening of the earthquake sequence, with smaller aftershocks in the southern section that are concentrated at ∼4 km depth and larger aftershocks in the northern section that are all below 4 km depth, indicating segmentation along the eastern branch of the Longquanshan fault zone.

FOCAL MECHANISM INVERSION Seismic Data and Focal Mechanism Solution Method
The focal mechanism solution, centroid depth, and moment magnitude of the M S 5.1 Qingbaijiang earthquake were determined via the cut-and-paste (CAP) waveform inversion method (Zhao and Helmberger, 1994;Zhu and Helmberger, 1996). We chose the CAP method over other inversion methods since it needs fewer stations, and the inversion results are insensitive to lateral variations in the crust and less dependent on the velocity model (Tan et al., 2006;Zheng et al., 2009;Long et al., 2010;Yi et al., 2012;Luo et al., 2015). The CAP method can also constrain the focal depth using the surface wave amplitudes in the waveforms, thereby yielding a reliable focal centroid depth (Luo et al., 2015).
To calculate the focal mechanism of the mainshock, we analyzed the waveform records of permanent broadband seismic stations of the Sichuan Regional Seismological Network within a 50-300 km epicentral distance of the mainshock. We employed the frequency-wavenumber method (Zhu and Helmberger, 1996) to calculate Green's function using both a local 1-D velocity model (Table 1) and the model of Yi et al. (2020). We set a 2 s duration for the source time, 30 and 60 s waveform window lengths, and 0.05-0.20 Hz and 0.05-0.10 Hz band pass filters for the body waves and surface waves, respectively, in the CAP inversion. The inversion was conducted in 5 • and 1 km increments to constrain the fault plane parameters and depth of the earthquake, respectively.
The obtained focal mechanism solution, centroid depth, and moment magnitude for the M S 5.1 Qingbaijiang earthquake are shown in Table 2. The results for the two different velocity models are similar ( Table 2); we selected the result based on the velocity model in Table 1 as the focal mechanism solution for the event. The best fit to the waveforms was obtained at a 4 km centroid depth (Figure 7), with > 70% of the waveform components yielding correlation coefficients of > 93% (Figure 8).
The 4 km centroid depth of the M S 5.1 mainshock is close to the relocated 5.2 km initial rupture depth, both of which indicate that the M S 5.1 event occurred above the shallow detachment layer in the upper crust. The parameters of the focal mechanism solution (strike/dip/rake) are 22 • /36 • /91 • for nodal plane I and 200 • /54 • /89 • for nodal plane II, which indicate a thrust earthquake with a NE-SW rupture direction.

ANALYSIS OF THE SEISMOGENIC FAULT OF THE M S 5.1 QINGBAIJIANG EARTHQUAKE AND IMPLICATIONS FOR SEISMIC HAZARD
Our preliminary analysis indicates that the seismogenic fault of the M S 5.1 Qingbaijiang earthquake is along the eastern branch of the Longquanshan fault zone, with the coseismic rupture plane dipping 56 • to the NW (nodal plane II). The initial rupture  Table 1   4  25  35  95  199  55  87  291  10  95  80  201  3 4.64 Yi et al. (2020) depth obtained from the relocation analysis and the centroid depth obtained from waveform inversion suggest that this M S 5.1 event was a moderate earthquake that occurred above the shallower detachment layer beneath the Longquanshan anticline. The sub-horizontal P-axis (azimuth of 291 • ) indicates that the tectonic stress field in the focal region is characterized by an subhorizontal principal compressive stress that trends WNW, which is consistent with the regional tectonic stress field in the South China Block (Heidbach et al., 2018;Wang and Shen, 2020). The earthquake depth, which is consistent with that of the destructive 1967 M S 5.5 Renshou earthquake, may indicate that M S ≥ 5.0 destructive earthquakes along the Longquanshan fault zone have mainly occurred in the shallow part of the upper crust.
Since the 2008 Wenchuan earthquake, many studies have examined the seismic hazard of the Longquanshan fault zone. For example,  reported a ∼0.1 bar increase in the Coulomb stress of the Longquanshan fault zone after the Wenchuan earthquake. Qian and Han (2011) reported a 0.4-0.6 bar increase in stress along the northern segment of the Longquanshan fault zone after the Wenchuan earthquake, with this stress increase gradually decreasing from north to south along the fault zone, which is consistent with the observation that the seismicity along the fault zone was mainly clustered along the northern segment after the Wenchuan earthquake. Parsons et al. (2008) calculated the Coulomb stress FIGURE 7 | Centroid depth variation in the residual error of the focal mechanism solution for the M S 5.1 Qingbaijiang earthquake. change after the Wenchuan earthquake and suggested that the event increased both the stress state in the Sichuan Basin and the risk of earthquake occurrence. A comparative analysis of the seismicity before and after the 2008 Wenchuan earthquake found a significant increase in both the earthquake frequency and intensity in the Sichuan Basin (Yi et al., 2019). Moderate earthquakes occurred successively in the low-seismicity area of the Sichuan Basin, such as the 19 February 2013 M S 4.7 Santai-Yanting earthquake, and the 29 July 2014 M S 4.9 Zitong and M S 4.6 Santai earthquakes, which could be related to stress loading in the Sichuan Basin due to the 2008 Wenchuan earthquake (Yi et al., 2019). However, Parsons et al. (2008) showed that the central and northern segments of the Longquanshan fault zone were in a negative zone of coseismic Coulomb stress disturbance due to the Wenchuan earthquake. Toda et al. (2008) reported a 0.1-0.3 bar stress reduction on a thrust fault to the south of Chengdu after the Wenchuan earthquake. Therefore, the influence of the 2008 Wenchuan earthquake on the seismicity and seismic hazard of the Longquanshan fault zone needs to be further investigated.
Previous studies have considered the western branch of the Longquanshan fault zone to be the main fault of the entire anticline Huang and Tang, 1995;Tang et al., 1996;Wang et al., 2008;Wang and Lin, 2017), leading to the conclusion that the western branch represents a greater seismic hazard (Huang and Tang, 1995;Zhang et al., 2020). Luo et al. (2008) believe that the Longquanshan anticline formed under E-W compression related to collision between the Qinghai-Tibet Plateau and South China Block, which formed an anti-thrust pattern along the front of the Longmenshan thrust belt. The Longquanshan anticline is a shallow structure that has formed via multi-stage subduction and subsequent emplacement of deep material from the South China Block to the deeper sections of the LFTB and the West Sichuan Plateau. The eastern branch is a back-fault associated with the entire fold system, similar to the back-fault on the main fault plane of the 2013 M S 7.0 Lushan earthquake . However, another possibility is that the Longquanshan fault zone and LFTB are components of the same structural system, such that the Longquanshan fault zone forms the southeastern boundary of the LFTB nappe foreland basin in western Sichuan Zhang et al., 2003;Liu, 2007). The eastern branch of the Longquanshan fault zone would be considered the main fault in this combined structural system, whereas the western branch would be a backfault associated with the entire fold system. Most earthquakes, including the M S 5.1 Qingbaijiang earthquake sequence, have occurred along the eastern branch since 2009, supporting this latter structural model. Further investigations of the dynamic FIGURE 8 | Comparison of synthetic (red) and observed (black) waveforms at the best fitting depth for the M S 5.1 Qingbaijiang earthquake. The relative time shift (in s; upper) and correlation coefficient (in%; lower) are provided below each trace. The station names are given to the left, with the epicentral distance (in km; upper) and azimuth (in degree; lower) provided.
Frontiers in Earth Science | www.frontiersin.org origin, tectonic setting, and seismic hazard of the Longquanshan fault zone need to be undertaken to resolve these discrepancies.

CONCLUSION
We analyzed seismic phase and waveform data recorded by the Sichuan and Chengdu regional seismic networks to relocate the earthquakes that have occurred along the central and northern segments of the Longquanshan fault zone since 2009, including the M S 5.1 Qingbaijiang earthquake sequence, using the multi-stage method. We also calculated the focal mechanism of the M S 5.1 mainshock using the CAP waveform inversion method.
The M S 5.1 Qingbaijiang earthquake occurred along a NE-SW-striking fault, consistent with the trend of the Longquanshan fault zone, at an initial rupture depth of 5.2 km. The earthquake clusters that have occurred along the Longquanshan fault zone since 2009, including the M S 5.1 Qingbaijiang earthquake sequence, were located along NW-dipping features that formed a parallel imbricate pattern, which is identical to the fault geometry along the eastern branch of the Longquanshan fault zone. The mainshock had a centroid depth of 4 km and a thrust-type focal mechanism along a NW-dipping fault plane, similar to the structure of the eastern branch. Therefore, the M S 5.1 Qingbaijiang earthquake was a pure thrusting event along the eastern branch of the Longquanshan fault zone, and nodal plane II of the focal mechanism solution represents the coseismic rupture plane. We conclude that the M S 5.1 Qingbaijiang earthquake was a moderate earthquake above the shallow detachment layer beneath the Longquanshan Anticline and was controlled by the regional tectonic stress field, as inferred from regional geological constraints.
The focal depth of the M S 5.1 Qingbaijiang earthquake, which is consistent with that of the 1967 M S 5.5 Renshou earthquake, may indicate that the destructive M S ≥ 5.0 earthquakes along the Longquanshan fault zone occur primarily in the shallow part of the upper crust. Our study provides valuable data for future studies on the seismogenesis and seismic hazard of the Longquanshan fault zone.

AUTHOR'S NOTE
MZ, Master, engineer, mainly engaged in seismic activity research.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
MZ mainly completed the writing of the manuscript. FL mainly completed the data processing and supported the completion of the manuscript. GY mainly formed the idea of the manuscript and supported the completion of the manuscript. ML, JX, and SW participated worked for the manuscript. All authors contributed to the article and approved the submitted version.

FUNDING
The research was supported by the National Science Foundation of China (41574047) and the National Key R&D Program of China (2018YFC150330501 and 2020YFA0710603-01).