Earthquake Reaearch in China  2018, Vol. 32 Issue (4): 482-490
Relocation of the Jinghe MS6.6 Earthquake Sequence on August 9, 2017 and Analysis of the Seismogenic Structure
Liu Jianming, Gao Rong, Wang Qiong, Nie Xiaohong
Earthquake Agency of Xinjiang Uygur Autonomous Region, Urumqi 830011, China
Abstract: Based on the digital waveforms of the Xinjiang Digital Seismic Network, the Jinghe MS6.6 earthquake sequence (ML ≥ 1.0) were relocated by HypoDD, The characteristics of the spatial distribution and the seismogenic structure of this earthquake sequence were analyzed. The results show that the main shock is relocated at 44.2639°N, 82.8294°E, and the initial rupture depth is 17.6km. The earthquake sequence clearly demonstrates a unilateral extension of about 20km in the EW direction, and is mainly located at a depth of 7km-17km. The depth profile along the aftershock direction shows that the focal depth of aftershocks tend to be shallower within 10km to the west of the main shock, the focal depth of the aftershock sequence with the tail direction deflecting SW is deeper. The depth profile perpendicular to the earthquake sequence shows a gradual deepening of the seismic sequence from north to south, which indicates that the fault plane is dipping south. According to the focal mechanism solution, given by the Institute of Geophysics, China Earthquake Administration, and the geological structure of the seismic source region, it is inferred that the seismogenic structure of the Jinghe MS6.6 earthquake may be the eastern segment of the Kusongmuxieke fault.
Key words: Jinghe MS6.6 earthquake     HypoDD location     Seismogenic structure     Kusongmuxieke fault

INTRODUCTION

According to the China Earthquake Networks Center, an MS6.6 earthquake, occured at 44.27°N, 82.89°E with a focal depth of 11km, struck Jinghe County of Bortala Mongolian Autonomous Prefecture, Xinjiang Uygur Autonomous Region, at 07:27 a.m. on August 9, 2017 (Beijing Time) (hereinafter referred to as the Jinghe MS6.6 earthquake). The earthquake was strongly felt in Urumqi, Bortala Mongolian Autonomous Prefecture, Changji, Ili and Karamay. According to the Xinjiang Regional Digital Seismic Network, up to September 30, 2017, a total of 396 aftershocks had been recorded in the focal area of the Jinghe MS6.6 earthquake, including 18 aftershocks with MS3.0-3.9 and 6 aftershocks with MS4.0-4.9, with the largest magnitude of MS4.7. Since 1900, 11 earthquakes measuring 6.0 or above occurred within 200km from the epicenter, including 1 earthquake measuring magnitude 7.0, which is the Xinyuan MS7.2 earthquake in 1944, whose epicenter is 94km away from this earthquake. The nearest earthquake is the 1955 Wusu MS6.5 earthquake, 59km from this earthquake, and the most recent earthquake is the 2012 Xinyuan-Hejing MS6.6 earthquake, 179km away from this earthquake. Based on field study and strong ground motion data, the Earthquake Agency of Xinjiang Uygur Autonomous Region provides distribution of the isoseismal lines of the Jinghe MS6.6 earthquake, which is nearly elliptic with a long axis in a NWW direction and an intensity of Ⅷ degree in the meizoseismal area which is about 44km long2.

The regional seismotectonic map of the Jinghe MS6.6 earthquake, published on the website of the Institute of Geology, China Earthquake Administration3, shows that the quake occurred near the Kusongmuxieike piedmont fault. This fault is a boundary fault located at the northern margin of the western section of Tianshan Mountains, and is also a regional active fault, which can be divided into three sections, namely the east, middle and west sections, from west to east according to its activity characteristics. Among them, the east section, striking 300°-310°, is about 50km long, and stretches from Jipuke in the east, traveling west to the west bank of the Ashale River in western Longkou, passing though Wulasitai, Wulanteergan and Saozimutu, which is a combination of 4 diagonal faults striking 280°-290°. The length of a single fault is 9km-13km, and the fault plane dips to the south, which presents mainly characteristics of a reverse fault (Chen Jianbo et al., 2007).

After the Jinghe MS6.6 earthquake, many domestic and foreign research institutions have carried out a series of studies on relocating the earthquake and its focal mechanism solution, which provides important references for satisfying the need of earthquake relief work and conducting follow-up scientific research. The website of the United States Geological Survey (USGS) provides a focal depth of about 20km for the Jinghe MS6.6 earthquake and thrust-type dislocation4. Early precise location results of the earthquake sequence given by the Institute of Geophysics, China Earthquake Administration5, and show that the earthquake sequence is distributed in the EW direction, and distribution of deep aftershocks is S-dipping, with a focal depth of about 15km. The focal mechanism solution inverted by the CAP method shows the nature of a thrust earthquake, and focal centroid depth is 20km. Relocation results given by the China Earthquake Networks Center6 show that the earthquake sequence is distributed in the EW direction, and deep aftershocks reveal that the seismogenic fault plane is nearly vertical. The focal mechanism solution inverted by the CAP method displays the nature of a thrust earthquake, and the hypocenter and centroid depths are 23km. The results given by different research institutions indicate that there are some differences in focal depth. Meanwhile, a different trend of distribution of deep aftershocks makes it difficult to understand the seismogenic structure of this earthquake.

Existing research results indicated (Waldhauser F. et al., 2000; Yi Guixi et al., 2015, 2017) that the spatial distribution and focal mechanism solution of an accurately positioned main shock and its aftershocks are of great significance for rapid determination of seismogenic structure after the earthquake, determination of future trends and post-disaster reconstruction planning. There are few stations near the epicenter of the Jinghe MS6.6 earthquake, and there are only 2 fixed seismic stations within a 100km radius from the epicenter. After the earthquake, the Earthquake Agency of Xinjiang Uygur Autonomous Region started a level Ⅱ emergency response and set up 2 mobile stations in the earthquake area. The first mobile station started to operate at 06:00 p.m. on August 9, and data was transmitted to the Xinjiang Earthquake Networks Center in real time, abundant sequence data have been accumulated, which provides an important basis for follow-up earthquake tracking and research work.

In this paper, based on seismic phase data recorded by fixed stations and mobile stations in the study area, the Jinghe MS6.6 earthquake sequence is relocated by using the HypoDD method, and more accurate spatial distribution images are obtained. On this basis, the spatial distribution characteristics and possible seismogenic structure of the Jinghe MS6.6 earthquake are discussed in combination with other related research achievements of the earthquake and study of geological structure in the earthquake zone.

1 METHOD AND DATA 1.1 The Double Difference Location Method

At present, the double difference (HypoDD) method (Waldhauser F. et al., 2000) is widely used by domestic researchers and significant research results have been obtained (Lyu Jian et al., 2008a, 2008b; Fang Lihua et al., 2011, 2013, 2014; Wang Weilai et al., 2014; Long Feng et al., 2015). This method is a relative earthquake locating method, which was developed on the basis of the main shock seismic locating method, it does not only reduce the influence of inaccuracy of the velocity structure model, but can also be used for the accurate location of an earthquake in a large area (Yang Zhixian et al., 2003). The premise of using the double difference algorithm is: the distance between two earthquakes relative to the distance between earthquakes and the seismic station and medium velocity change are small enough, thus two ray paths can be approximately viewed as the same, and the time difference of two events recorded at the same station can be attributed to the spatial location difference of the two events.

The difference between the observed travel time difference of two earthquakes, i and j, to station k and the theoretical travel time difference is defined as double-difference residual and is expressed as drkij

 ${\rm{d}}\mathit{r}_k^{ij} = r_k^i - r_k^j = {(t_k^i - t_k^j)^{{\rm{obs}}}} - {(t_k^i - t_k^j)^{{\rm{cal}}}}$ (1)

where, (tki-tkj)obs is the observed travel time difference, and (tki-tkj)cal is the theoretical travel time difference. Equations such as formula (1) obtained from all earthquakes (i, j=1, 2, …, N) and all stations (k=1, 2, …) are expressed in matrix form as below

 $\mathit{\boldsymbol{WGm}} = \mathit{\boldsymbol{Wd}}$ (2)

where, G is the matrix containing M×4N -order partial derivative (M is the number of observed double difference, and N is the number of earthquakes), d is the data vector containing the double difference equation (1), m represents the vector with a length of 4N, which contains variation of underdetermined source parameters, and W denotes a diagonal matrix weighted to each equation. In the calculation, the following formula is used as a constraint where the average displacement of all earthquakes after relocation is zero, that is, the centroid is not moving. By further iteration, the positioning residuals are gradually reduced, and final positioning results are obtained.

 $\sum\nolimits_{i = 1}^N {\mathit{\Delta }{\mathit{\boldsymbol{m}}_\mathit{i}} = 0}$ (3)
1.2 Data

In this study, 27 fixed stations and 2 mobile stations within 400km of the epicenter of Jinghe MS6.6 earthquakes are used to relocate the earthquake sequence (Fig. 1). Data used for relocating are mainly from seismic phase observation reports provided by the Regional Digital Seismic Network of Xinjiang. 370 ML≥1.0 earthquakes recorded by at least 3 stations with 6 or more seismic phases from August 9 to September 30, 2017 are selected for relocation. Among them, there were 3, 678 P-wave arrival time data, 2, 945 S-wave arrival time data, and on average, each earthquake has 18 seismic phases. In order to check the reliability of seismic data in observation reports, seismic phase travel time curves of P waves and S waves are drawn (Fig. 2). As can be seen from Fig. 2, travel time curves of Pg(Pn) and Sg(Sn) seismic phases are clearly distinguished and have low dispersion, therefore, it is believed that travel time data for these four seismic phases are of high reliability.

 Fig. 1 Distribution of seismic stations for relocation

 Fig. 2 Travel time curves of Sn(Sg) and Pn(Pg) seismic phases

In this paper, the HypoDD method is used to relocate the Jinghe MS6.6 earthquake sequence, and 212 relocation results of the earthquake sequence were obtained. In the process of relocation, the minimum number of connections and minimum number of observations are set to be 8, and the maximum distance of an event pair to be 10km. The weight of the P-wave and S-wave arrival time are set as 1.0 and 0.5, respectively. Calculations are done in 3 groups for 15 repeated iterations during earthquake relocation. Due to the lack of artificial seismic sounding results around the epicenter, several layered velocity structure models near the study area are tried in the process of relocation, and finally the results of CRUST1.0 layered velocity structure model are used (Table 1).

Table 1 Velocity structure model in relocation
2 RELOCATION RESULTS OF THE EARTHQUAKE SEQUENCE 2.1 Planar Distribution Characteristics of Relocation

Based on the HypoDD method and parameter setting described in previous section, relocation results of the Jinghe MS6.6 earthquake sequence are obtained. The Jinghe MS6.6 earthquake is relocated to be centered at 44.2639°N, 82.8294°E and the origin time is 07 : 27 : 51 a.m. on August 9, 2017, with an initial epicenter rupture depth of 17.6km. The average relative errors of the earthquake sequence relocation in EW, SN and vertical direction are 0.5km, 0.4km and 0.9km respectively and the average travel time residual is 0.08s.

Fig. 3(a) and Fig. 3(b) are the results before and after relocation of the Jinghe MS6.6 earthquake sequence. It can be seen from Fig. 3 that the earthquake sequence is dispersedly distributed before relocation with no obvious predominant distribution, while the distribution of earthquakes is more concentrated after relocation and the locations of epicenters are obviously constricted. The planar distribution map after relocation(Fig. 3(b)) shows that the aftershock-concentrated areas of the Jinghe MS6.6 earthquake are mainly distributed in the eastern section of the Kusongmuxieke piedmont fault, stretching about 20km, extending unilaterally in the EW direction (273°) on the whole, which is consistent with the intensity distribution features7 given in post-earthquake emergency response research, and is consistent with the 4 faults striking 280°-290° in the eastern section of the Kusongmuxieke piedmont fault. In addition, aftershocks of the Jinghe MS6.6 earthquake mainly took place within 20 days after the main shock, and after that, the distribution of aftershocks shows a sign of SW shifting at the end of the near-EW direction.

 Fig. 3 Distribution of epicenters of the Jinghe MS6.6 earthquake sequence before (a) and after(b) relocation

7 Earthquake Agency of Xinjiang Uygur Autonomous Region, 2017, Report on Earthquake Disaster Investigation and Assessment Results of Jinghe MS6.6 Earthquake on August 9, 2017.

2.2 Profile Distribution after Relocation

The focal depth profile shows that earthquake focal depths are mainly concentrated in 7km-18km after relocation (Fig. 4), and there are few aftershocks in the shallow crust. This is consistent with the results of the post-earthquake emergency response investigations that show no significant surface rupture zones are found near the epicenter, and is also consistent with the results obtained by Zhang Yong et al. in Institute of Geophysics, CEA4 that the source rupture process of this earthquake shows little static slip in the shallow crust (less than 0.3m).

 Fig. 4 Histogram of depth distribution

In order to display structural features of the seismogenic fault plane of this earthquake in detail, two profiles of focal depths are given (Fig. 5). Among them, one is profile A-A′ in the direction of the major axis along the distribution of epicenters, and the other is profile B-B′ nearly perpendicular to the major axis of epicenter distribution. Profile A-A′(Fig. 5(a)) shows that aftershocks of the Jinghe MS6.6 earthquake expand unilaterally in nearly an EW direction with a spreading length of about 15km. Focal depths of aftershocks tend to be shallower within 10km west of the main shock and aftershock sequence, and there are many aftershocks occurring there. Earthquakes shifting in the SW direction at the end of the aftershock sequence have deeper focal depths and are less in number (Fig. 5(a)). Profile B-B′reflects focal depth distribution characteristics along the dipping direction of fault, and clearly shows that the focal depth of the earthquake sequence presents the characteristics of gradual deepening from north to south. In addition, profile B-B′reveals that the possible seismogenic fault of this earthquake dips to the south, which is consistent with the dipping direction of the eastern section of the Kusongmuxieke piedmont fault (Chen Jianbo et al., 2007).

 Fig. 5 Focal depth distribution along profile A-A′ and B-B′ The distance from earthquake to the axis of profile is less than 10km

The optimal double-couple nodal plane solution4 for the Jinghe MS6.6 earthquake is obtained by the Institute of Geophysics, China Earthquake Administration, using the CAP inversion, and the results show that this earthquake belongs to the thrust type, which is consistent with the motion characteristics of the eastern section of the Kusongmuxieke piedmont fault near the epicenter. Among them, nodal planeⅠstrikes at 262°, with a dip angle of 45° and a rake angle of 80°, nodal plane Ⅱstrikes at 96°, with a dip angle of 46° and rake angle of 100°, and both nodal plane Ⅰand Ⅱ are in a near EW direction. Combined with spatial distribution characteristics of precise locating, it is believed that the S-dipping nodal plane Ⅱis the seismogenic fault plane for the earthquake, which is consistent with the S-dipping of the eastern section of the Kusongmuxieke piedmont fault. Therefore, based on the focal mechanism of the main shock, spatial distribution characteristics of the earthquake sequence and geological conditions in the earthquake zone, it is inferred that the eastern section of the Kusongmuxieke piedmont fault is the seismogenic structure for the Jinghe MS6.6 earthquake.

3 CONCLUSION

In this paper, based on observation reports of the regional digital seismic network of Xinjiang, the Jinghe MS6.6 earthquake sequence is relocated by using the HypoDD method, relocation results for the earthquake sequence are obtained, and in combination with the focal mechanism solution acquired by the Institute of Geophysics, China Earthquake Administration, by using the CAP method, spatial distribution of the earthquake sequence, variation of focal depth on different profiles and its possible seismogenic structure are analyzed. The following conclusions are drawn.

(1) The Jinghe MS6.6 earthquake is relocated to be centered at 44.2639°N, 82.8294°E and the origin time is 07 : 27 : 51 a.m. on August 9, 2017, with an initial epicenter rupture depth of 17.6±0.9km, shallower than the focal centroid depth of best fit, 20km, obtained by Institute of Geophysics, China Earthquake Administration using the CAP method.

(2) The double difference location results show that the earthquake sequence expands unilaterally about 20km in a near EW direction (273°) on the whole, which is roughly consistent with the 4 faults striking 280°-290° in the eastern section of the Kusongmuxieke piedmont fault. Predominant distribution of focal depths is 7km-17km.Depth profile along the aftershock distribution shows that the rupture of the Jinghe MS6.6 earthquake sequence starts in the deep and then spreads to shallow crust, and aftershock sources tend to be shallower within 10km west of the main shock, aftershocks shifting in the SW direction at the end of the aftershock sequence have deeper focal depth. The depth profile perpendicular to the earthquake sequence shows that focal depth of the earthquake sequence presents the characteristics of gradual deepening from north to south, indicating that the seismogenic fault plane is dipping to the south, which is consistent with the trend of the eastern section of the Kusongmuxieke piedmont fault.

(3) The focal mechanism solution given by the Institute of Geophysics, China Earthquake Administration, shows that the Jinghe MS6.6 earthquake belongs to the thrust type, and the S-dipping nodal plane Ⅱ is the seismogenic fault plane for the earthquake, which is consistent with the fault type and dip of the eastern section of the Kusongmuxieke piedmont fault.

(4) Because the relationship between faults seen on the surface and deep seismogenic structure is very complicated, the prediction of deep seismogenic structures or source position by surface faults is not deterministic. However, there are very few studies on the profile of Kusongmuxieke piedmont fault, so the earthquake con't be projected on the geological profile. In this study, based on spatial distribution and focal mechanism solution of the Jinghe MS6.6 earthquake sequence, it is inferred that the eastern section of the Kusongmuxieke piedmont fault is the seismogenic fault for the Jinghe MS6.6 earthquake.

ACKNOWLEDGEMENT

We extend heartfelt thanks to research professor Fang Lihua from the Institute of Geophysics, China Earthquake Administration, for guidance on the use of HypoDD Software, the Monitoring Center of Eathquake Agency of Xinjiang Uygur Autoruonous Region for providing the waveform data and observation reports for this study, and to the reviewers for their thorough review of this article and constructive suggestions for revision. Some maps in this study are drawn by GMT software.

REFERENCES
 Chen Jianbo, Shen Jun, Li Jun, et al. Preliminary study on new active characteristics of Kusongmuxieke Mountain front fault in the west segment of North Tianshan[J]. Northwestern Seismological Journal, 2007, 29(4): 335–340 (in Chinese with English abstract). Fang Lihua, Wu Jianping, Wang Weilai, et al. Relocation of the mainshock and aftershock sequences of MS7.0 Sichuan Lushan earthquake[J]. Chinese Science Bulletin, 2013, 58(20): 1901–1909 (in Chinese with English abstract). Fang Lihua, Wu Jianping, Wang Weilai, et al. Relocation of the 2014 MS7.3 earthquake sequence in Yutian, Xinjiang[J]. Chinese Journal of Geophysics, 2014, 58(3): 802–808 (in Chinese with English abstract). Fang Lihua, Wu Jianping, Zhang Tianzhong, et al. Relocation of mainshock and aftershocks of the 2011 Yingjiang MS5.8 earthquake in Yunnan[J]. Acta Seismologica Sinica, 2011, 33(2): 262–267 (in Chinese with English abstract). Lv Jian, Su Jinrong, Jin Yuke, et al. Discussion on relocation and seismotectonics of the MS8.0 Wenchuan earthquake sequences[J]. Seismology and Geology, 2008a, 30(4): 917–925 (in Chinese with English abstract). Lv Jian, Zheng Yong, Ni Sidao, et al. Focal mechanisms and seismogenic structures of the MS5.7 and MS4.8 Jiujiang-Ruichang earthquakes of Nov.26, 2005[J]. Chinese Journal of Geophysics, 2008b, 51(1): 158–164 (in Chinese with English abstract). Wang Weilai, Wu Jianping, Fang Lihua, et al. Double difference location of the Ludian MS6.5 earthquake sequences in Yunnan Province in 2014[J]. Chinese Journal of Geophysics, 2014, 57(9): 3042–3051 (in Chinese with English abstract). Yang Zhixian, Chen Yuntai, Zheng Yuejun, et al. Relocation of earthquake in central-western China using the double difference earthquake location algorithm[J]. Science in China (Ser. D), 2003, 33(Suppl): 129–134 (in Chinese with English abstract). Yi Guixi, Long Feng, Liang Mingjian, et al. Focal mechanism solutions and seismogenic structure of the 8 August 2017 M7.0 Jiuzhaigou earthquake and its aftershocks, northern Sichuan[J]. Chinese Journal of Geophysics, 2017, 60(10): 4083–4097 (in Chinese with English abstract). Yi Guixi, Long Feng, Wen Xueze, et al. Seismogenic structure of the M6.3 Kangding earthquake sequence on 22 Nov. 2014, Southwestern China[J]. Chinese Journal of Geophysics, 2015, 58(4): 1205–1219 (in Chinese with English abstract). Long Feng, Wen X. A more accurate relocation of the 2013 MS7. 0 Lushan, Sichuan, China, earthquake sequence and the seismogenic structure analysis[J]. J Seismol, 2015, 19(3): 653–665. DOI:10.1007/s10950-015-9485-0. Waldhauser F., Ellsworth W.L. A double-difference earthquake location algorithm:method and application to the Northern Hayward Fault, California[J]. Bull Seism Soc Am, 2000, 90(6): 1353–1368. DOI:10.1785/0120000006.
2017年8月9日精河6.6级地震序列重定位与发震构造初步研究