High-precision relocation of the aftershock sequence of the January 8, 2022, MS6

Liping Fan, Boren Li, Shirong Liao, Ce Jiang,2 and Lihua Fang,4,✉

1 Institute of Geophysics, China Earthquake Administration, Beijing 100081, China

2 Guangdong Earthquake Agency, Guangzhou 510405, China

3 Fujian Earthquake Agency, Fuzhou 350003, China

4 Key Laboratory of Earthquake Source Physics, China Earthquake Administration, Beijing 100081, China

ABSTRACT The 2022 Menyuan MS6.9 earthquake, which occurred on January 8, is the most destructive earthquake to occur near the Lenglongling (LLL) fault since the 2016 Menyuan MS6.4 earthquake. We relocated the mainshock and aftershocks with phase arrival time observations for three days after the mainshock from the Qinghai Seismic Network using the double-difference method. The total length and width of the aftershock sequence are approximately 32 km and 5 km, respectively, and the aftershocks are mainly concentrated at a depth of 7-12 km. The relocated sequence can be divided into 18 km west and 13 km east segments with a boundary approximately 5 km east of the mainshock, where aftershocks are sparse. The east and west fault structures revealed by aftershock locations differ significantly. The west fault strikes EW and inclines to the south at a 71°-90°angle, whereas the east fault strikes 133° and has a smaller dip angle. Elastic strain accumulates at conjunctions of faults with different slip rates where it is prone to large earthquakes. Based on surface traces of faults, the distribution of relocated earthquake sequence and surface ruptures, the mainshock was determined to have occurred at the conjunction of the Tuolaishan(TLS) fault and LLL fault, and the west and east segments of the aftershock sequence were on the TLS fault and LLL fault,respectively. Aftershocks migrate in the early and late stages of the earthquake sequence. In the first 1.5 h after the mainshock,aftershocks expand westward from the mainshock. In the late stage, seismicity on the northeast side of the east fault is higher than that in other regions. The migration rate of the west segment of the aftershock sequence is approximately 4.5 km/decade and the afterslip may exist in the source region.

Keywords: Menyuan earthquake; aftershock sequence; double-difference relocation; Lenglongling fault; Tuolaishan fault

1. Introduction

On January 8, 2022, at 1:45 am (Beijing time), aMS6.9 earthquake struck near Menyuan county, Haibei prefecture, Qinghai province (hereinafter referred to as the“Menyuan earthquake”). The China Earthquake Networks Center reported the earthquake’s focal depth to be at 10 km (37.77°N, 101.26°E). The Qinghai Seismic Network recorded 1,434 aftershocks until 0:00 on January 11,including one earthquake of a magnitude ＞MS5.0, two of≥MS4.0, and 13 ofMS3.0. The largest aftershock (MS5.1)was recorded on January 8 at 2:09 am, which caused a 22-km surface rupture. The highest intensity was estimated at IX degrees, and the area with an intensity of VI or above was measured to be approximately 23,417 km2(http://www.cea.gov.cn/cea/xwzx/fzjzyw/5646200/index.html,last access: 13 January, 2022).

The Menyuan earthquake occurred at the intersection of the Lenglongling (LLL) fault and the Tuolaishan (TLS)fault (Figure 1). Strike-slip, thrust, primary, and secondary faults form a complex tectonic system close to the earthquake’s epicenter (Guo P et al., 2019a). It is an important part of an almost 1,000-km-long left-lateral strike-slip fault zone at the northeastern margin of the Tibetan Plateau (NETP), which controls the northeastward compression movement of the plateau relative to the Alashan block. The Menyuan earthquake was the largest destructive earthquake to occur near the LLL fault since the 2016 MenyuanMS6.4 earthquake. At the LLL fault,seismicity is strong and at least six strong earthquakes have occurred along this fault, the largest being the 1927MS8.0 Gulang earthquake (Guo P et al., 2019b). Shortly after the Menyuan earthquake, the focal mechanism,rupture process, and high-precision aftershock relocation were studied by domestic and foreign research institutions;however, the seismogenic fault geometry and spatial and temporal evolution of the aftershock sequence remain unclear. In this study, we investigated the precise locations of the aftershock sequence based on seismic data from the Qinghai Seismic Network. Our results provide a reference for determining the seismogenic fault of the Menyuan earthquake, analyzing the characteristics of the earthquake sequence, and determining the earthquake trend.

Figure 1. Faults, historical earthquakes, and seismic stations around the Menyuan earthquake. Blue triangles denote seismic stations; black circles denote historical earthquakes ≥ MS5.0 (data source: China Earthquake Data Center,http://data.earthquake.cn, last access: 13 January, 2022); purple circle denotes 2016 MS6.4 Menyuan earthquake (Liu M et al.,2019; focal mechanism data source: https://www.globalcmt.org/CMTsearch.html, last access: 13 January, 2022); blue circles denote 2022 Menyuan earthquake sequence; red star denotes mainshock of the 2022 Menyuan earthquake (focal mechanism data source: https://www.cea-igp.ac.cn/kydt/278812.html, last access: 13 January, 2022).

2. Data and method

In this study, we collected seismic observation reports from 01:45 on January 8 to 0:00 on January 11, 2022,analyzed by the Qinghai Seismic Network, which contained data on 1,434 earthquakes of which 478 were recorded by more than three stations. Magnitudes of 477 aftershocks range fromML0.5 toMS5.1, with the minimum magnitude of completeness beingML0.8. A total of 86 stations were used for the preliminary location, with a good azimuth coverage of an average opening angle 194°.There are 13 stations within 200 km of the epicentral distance from the mainshock, the nearest being approximately 35 km. A total of 6,615 phase arrivals were used to locate 478 earthquakes, including 3,488 Pg, 2,782 Sg, and 335 Pn phases, with an average of approximately 14 phase arrivals for each earthquake. Travel-time curves for all phases are plotted in Figure 2, which show good linear and concentrated trends, indicating that the arrival time picking accuracy is high and initial earthquake locations are reliable.

Figure 2. Travel time curves of P-wave and S-wave phases.

In this study, the double-difference location method was used to relocate the Menyuan earthquake sequence(Waldhauser and Ellsworth, 2000). It is based on the assumption of a constant seismic velocity in a small region, and fits observational travel-time difference of an event pair within a certain distance recorded by the same station to the theoretical difference with adjusting relative locations of earthquakes in the pair, which can considerably improve relative location accuracy. This algorithm has been widely used for obtaining highprecision seismic catalogs and determining fault structures(Chen JH et al., 2009; Fang LH et al., 2013; Zhang GW and Lei JS, 2013; Guo H and Zhang HJ, 2017; Li YQ et al., 2019; Song Q et al., 2020; Wang WL et al., 2021). The 1-D velocity model (Table 1) used in the relocation refers to the P-wave velocity obtained by Jia SX and Zhang XK(2008) using deep seismic sounding, and the crustal thickness (57 km) and P/S ratio (1.73) from a receiver function study by Wang WL et al. (2017).

Table 1. Velocity model used for aftershock relocation.

To guarantee reliable results, data used for relocation were selected according to the following criteria: (1) the distance between hypocenters in an event pair is ＜ 12 km;(2) the hypocentral distance is ＜ 200 km; and (3) the number of phase travel-time pairs in an event pair is ≥ 6.After selection, 393 earthquakes were relocated with 12,204 P-wave and 11,210 S-wave travel-time pairs.

3. Results and discussion

In this study, we obtained relocated hypocenters and origin times of the mainshock and 374 aftershocks. The relocation errors in north-south, east-west, and depth are 0.38 km, 0.35 km, and 0.57 km, respectively; the mean values of travel-time residuals before and after relocation are 0.35 s and 0.12 s (Figure 3). Reduction of travel-time residuals after relocation demonstrates that relative locations between earthquakes are more accurate.

Epicenters and the space-time distribution of the mainshock and aftershocks after relocation are shown in Figure 4. The mainshock was located at 37.769°N,101.274°E, and the depth was 12.6 km. The aftershock sequence is distributed linearly along the east-west direction with a total length of approximately 32 km, and mainly concentrates at a depth of 7-12 km. Aftershocks exist on the east and west of the mainshock with the majority above it, indicating that the mainshock ruptured upward. In the aftershock sequence of this study, early and large earthquakes were primarily distributed on the west side of the mainshock. Aftershocks were sparse and the fault strike changes at approximately 5 km (37.76°N,103.50°E) to the east side of the mainshock. Consequently,the aftershock sequence was divided into two sections by this boundary: west and east. The west segment was 18 km long, with a width gradually decreasing from 4 km in the west to approximately 2 km in the east; the east segment was approximately 13 km long and 5 km wide.

Figure 3. Histogram of errors in relocation (a-c) and histogram of travel-time residuals in initial location and relocation (d).

Dense cross sections perpendicular to the fault strike can better reveal 3D fault structures (Figure 5). We fit fault shapes with aftershock distribution. The fault strikes 90° in sectionAA’ and 133° in sectionBB’. InCC’ andDD’,aftershocks are clustered at a depth of approximately 10 km;thereby making determination of dip angles of the fault impossible. InEE’,FF’ andGG’, geometric parameters of the fault deduced from the distribution of aftershocks, the fault is inclined to the south with a dip angle of 71° to 90°.InII’ andJJ’, aftershocks are far too dispersed to enable identification of the fault shape. The fault strike and dip angle change gradually from west to east; however, this change is particularly significant in the east segment. The west fault strikes E-W with a 71°-90° dip angle toward the south, and the east fault strikes approximately 133° with a smaller dip angle. Although only data collected in three days was utilized in this study, the fault structure we obtained was identical to that obtained by Yang HF et al.(2022) using data collected over two weeks. However, our result differs from those of Xu Y et al. (2022), wherein an aseismic layer was found at a depth of 10-15 km.

Multiple researchers (Jiang LZ et al., 2007; Chen WB,2003; Guo P et al., 2017) suggest that the eastern part of the TLS fault is on the north side of TLS, ends north of Liuhuanggou at the bottom of LLL, and the western end of the LLL fault is south of Liuhuanggou. Fault traces in this study were obtained from the Seismic Active Fault Survey Data Center of China Earthquake Administration(https://www.activefault-datacenter.cn/, last access: 13 January, 2022). These traces are derived from satellite data with an error of 100 m to several kilometers. In Figures 4 and 5, the east segment of the aftershock sequence coincides with the surface trace of the LLL fault. The west segment is at a depth of 7-12 km and inclines to the south with a 71°-90° angle, whose surface projection is approximately 1.5 km to the south of the surface trace of the TLS fault. Based on the high-precision locations of aftershocks and surface traces of the LLL fault and TLS fault, we inferred that the Menyuan earthquake occurred at the junction of the LLL fault and TLS fault; the aftershock activity of the west segment is related to the TLS fault and that of the east segment is related to the LLL fault. At present, no studies investigating the TLS fault have been conducted. If the fault’s dip direction is not south, the seismogenic fault of the Menyuan earthquake may be an unknown fault located to the south side of the TLS fault.

Figure 4. Spatial and temporal distribution of the aftershock sequence after relocation. Blue star denotes the mainshock;circles denote aftershocks; grey solid lines denote faults. The color bar indicates origin times of earthquakes relative to the mainshock. (a) Aftershock locations and the location of cross-section AA’. (b) Depth of aftershocks along AA’.

A study on the focal mechanism of the Menyuan earthquake indicates that this event occurred on a strikeslip fault (https://www.cea-igp.ac.cn/kydt/278809.html,last access: 13 January, 2022). The cross-section of the aftershock sequence near the mainshock shows that the fault is nearly vertical, which is consistent with the study of the focal mechanism. The temporal and spatial evolution of the aftershock sequence shows that it propagates westward (Figure 6) and upward (FF’ section in Figure 5) in the early stage, which agrees with the rupture direction of the mainshock (Yang HF et al., 2022).Aftershocks start bursting to the east side of the mainshock approximately 1.5 h after the mainshock, and the sequence expands on both sides. To the west of the mainshock,aftershocks expand from 6 km at 1 h after the mainshock to approximately 12 km in less than 1 day, with an expansion rate of approximately 4.5 km/decade. This phenomenon is similar to that observed by Peng ZG and Zhao P (2009) in the 2004 Parkfield earthquake. We inferred from the space-time evolution of aftershocks that the afterslip may have occurred in the epicenter region(Perfettini et al., 2019). Research on the source rupture process shows that the rupture scale was greater on the east side of the mainshock than on the west side (https://www.cea-igp.ac.cn/kydt/278809.html, last access: 13 January,2022). This asymmetry of the coseismic rupture results in the shape of the aftershock sequence. Additionally, the aftershock distribution indicates that the coseismic rupture may have bilateral characteristics. Joint inversion of strong motion and geodetic records around the epicenter could provide more information regarding the coseismic process.In the last stage of the aftershock sequence, the seismicity in the northeastern part of the east fault is stronger than that in other regions, indicating that aftershocks migrate to the northeast.

Figure 5. Cross-sections of the aftershock sequence perpendicular to the fault’s strike. The star denotes the mainshock.Circles are aftershocks and sizes scale with magnitudes. (a) Locations of cross-sections. Grey lines denote fault traces. Black lines denote fitted faults. Solid pink lines denote cross-section locations and dash pink lines on both sides of per solid pink line limit the range of the projection. Red squares are locations at which parameters of fitted faults are measured, and numbers around squares stand for fault-strike/fault-dip with fault dip direction. (b) Cross-sections.

Figure 6. Spatiotemporal migration of the aftershock sequence along section AA’. Red star denotes mainshock. Circles are aftershocks and sizes scale with event magnitudes. The red dash line marks the approximate slope of migration of aftershocks along AA’.

The Indian Plate has been pushing northward since the Cenozoic, resulting in the continuous shortening and thickening of the Tibetan Plateau lithosphere along with the lateral extrusion of the plateau material (Tapponnier et al., 1990). The northeastern extrusion is restricted by the Alashan block which is characterized by strong middle and upper crust (Zhong SJ et al., 2017), causing the development of large lithospheric-scale shear zones. The Qilian-Haiyuan fault zone is the most important of these,and accommodates the convergence of the Tibetan Plateau and the Alashan block. The TLS fault and the LLL fault comprise the western and central parts of the Qilian-Haiyuan shear zone, with the TLS fault located on the west and the LLL fault on the east, forming a right-step en echelon pattern (Chen WB, 2003). GPS studies have indicated that slip rates on the north and south sides of the Qilian-Haiyuan fault zone are quite different, and the differences of slip rates between blocks on two sides along the fault are inconsistent (Duvall and Clark, 2010; Zheng WJ et al., 2013). The TLS fault is mainly sinistral strikeslip, with the slip rate of the east part being approximately 3-4 mm/a. The LLL fault is sinistral strike-slip with minor normal components, with a slip rate of approximately 6 mm/a (Guo P et al., 2017). Other faults to the east of the LLL fault in the Qilian-Haiyuan fault zone slide at a rate of approximately 4 mm/a. Studies based on GPS velocities show that the maximum shear stress rate of the fault near the mainshock reaches 2,000 Pa·a-1, meaning large stress accumulation, indicating this section of the fault being locked (Shi FQ et al., 2018; Wang M and Shen ZK, 2020).Slip-rate differences of different faults lead to deformation heterogeneity and accumulation of elastic strain at their intersections, making these regions more prone to large earthquakes. The mainshock of the Menyuan earthquake was located at the junction of the west segment of the LLL fault and the east segment of the TLS fault, where the strike of the LLL fault changes resulting in a stepover zone where elastic strain accumulates, thereby making the region prone to strong earthquakes.

4. Conclusions

We relocated the mainshock and aftershocks of the 2022MS6.9 Menyuan earthquake based on the earthquake bulletin of the Qinghai Seismic Network using the doubledifference method. This allowed us to conduct a preliminary analysis of the geometry of the seismogenic fault, space-time distribution characteristics and the seismogenic mechanisms of this event.

The distribution of aftershocks indicates that the coseismic rupture may have bilateral characteristics. The focal depth distribution of the aftershocks shows that the mainshock ruptured upward. The aftershocks have obvious spatiotemporal migration. In the first 1.5 hours after the mainshock, aftershocks expand to the west and then to both the west and the east. Migration rate of the west segment aftershocks is about 4.5 km/decade and the afterslip may have occurred in the focal region.

The continuous extrusion of NETP towards the Alashan block results in different slip rates of the TLS fault and the LLL fault, which makes accumulation of elastic strain at the junction of these two faults where is prone to strong earthquakes. The Menyuan earthquake occurred at the intersection of the two faults. The total length of the aftershock sequence is approximately 32 km and can be divided into east and west segments of different structures with a boundary approximate 5 km east of the mainshock. The west segment is approximately 18 km long, strikes EW and inclines to the south with a dip angle of 71°-90°. The east segment is about approximately 13 km long with a strike of 133° and a smaller dip angle.

The result of this study is a rapid product of large earthquake emergency management, providing a valuable reference for the identification of seismogenic faults and the compilation of intensity maps. Further research focused on building a more comprehensive and highprecision aftershock catalog using deep-learning methods is required to investigate the process of the Menyuan earthquake in detail.

Acknowledgements

We thank the reviewers for their valuable feedback and suggestions that have helped to improve the manuscript.We thank Qinghai Earthquake Agency and China Earthquake Networks Center for providing the aftershock bulletin. We thank Zhimin Li of Qinghai Earthquake Agency for providing the field investigation data. The relocated aftershock sequence catalog is available at https://www.researchgate.net/publication/358199036_Prec ise_relocation_of_the_aftershock_sequences_of_the_2022_M69_Menyuan_earthquake. This study was jointly funded by the National Key Research and Development Program of China (No. 2021YFC3000702), the Special Fund of the Institute of Geophysics, China Earthquake Administration (No. DQJB21Z05) and the National Natural Science Foundation of China (No. 41804062).