PROBING THE FAULT COMPLEXITY OF THE 2017 MS 7.0 JIUZHAIGOU EARTHQUAKE BASED ON THE INSAR DATA - MDPI
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
remote sensing Article Probing the Fault Complexity of the 2017 Ms 7.0 Jiuzhaigou Earthquake Based on the InSAR Data Xiongwei Tang 1,2 , Rumeng Guo 1,2,3, *, Jianqiao Xu 1 , Heping Sun 1,2 , Xiaodong Chen 1 and Jiangcun Zhou 1 1 State Key Laboratory of Geodesy and Earth’s Dynamic, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences, Wuhan 430077, China; tangxiongwei18@mails.ucas.edu.cn (X.T.); xujq@asch.whigg.ac.cn (J.X.); heping@asch.whigg.ac.cn (H.S.); chenxd@whigg.ac.cn (X.C.); zjc@asch.whigg.ac.cn (J.Z.) 2 College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China 3 Earth System Science Programme, The Chinese University of Hong Kong, Shatin, Hong Kong, China * Correspondence: rumengguo@cuhk.edu.hk Abstract: On 8 August 2017, a surface wave magnitude (Ms) 7.0 earthquake occurred at the buried faults extending to the north of the Huya fault. Based on the coseismic deformation field obtained from interferometric synthetic aperture radar (InSAR) data and a series of finite fault model tests, we propose a brand-new two-fault model composed of a main fault and a secondary fault as the optimal model for the Jiuzhaigou earthquake, in which the secondary fault is at a wide obtuse angle to the northern end of the main fault plane. Results show that the dislocation distribution is dominated by sinistral slip, with a significant shallow slip deficit. The main fault consists of two asperities bounded by an aftershock gap, which may represent a barrier. In addition, most aftershocks are located in stress shadows and appear a complementary pattern with the coseismic high-slip regions. We propose that the aftershocks are attributable to the background tectonic stress, which may be related to the velocity-strengthening zones. Citation: Tang, X.; Guo, R.; Xu, J.; Sun, H.; Chen, X.; Zhou, J. Probing the Fault Complexity of the 2017 Ms Keywords: Jiuzhaigou earthquake; fault geometry parameters; coseismic slip distribution; InSAR 7.0 Jiuzhaigou Earthquake Based on the InSAR Data. Remote Sens. 2021, 13, 1573. https://doi.org/10.3390/ rs13081573 1. Introduction On 8 August 2017, a surface wave magnitude (Ms) 7.0 earthquake occurred in Ji- Academic Editor: Zhong Lu uzhaigou County, Aba Prefecture, Sichuan Province, China, with its epicenter at 103.82◦ E, 33.20◦ N, and a focal depth of about 20 km (Institute of Geophysics, China Earthquake Received: 12 April 2021 Administration, CEA-IGP). As of 13 August 2017, the earthquake had caused 25 deaths Accepted: 15 April 2021 and severe damage to both the Jiuzhaigou scenic area and more than 70,000 structures [1]. Published: 19 April 2021 The Jiuzhaigou earthquake occurred near the northeastern boundary of the Bayan Har block, indicating that the Bayan Har block is still active [2]. Because the southeasterly Publisher’s Note: MDPI stays neutral movement of the Bayan Har block is blocked by the North China block and the South China with regard to jurisdictional claims in block, there are many broom-like branches on the eastern end of the East Kunlun fault, published maps and institutional affil- which intersect with the Minjiang fault and the Huya fault, resulting in a complex seismic- iations. tectonic environment in this region [3]. The NWW-trending Tazang fault, the northern segment of the NNW-trending Huya fault, and other nearby branches accommodate most of the sinistral strike-slip activity of the East Kunlun fault. The nearly NS-trending Min- jiang fault and the southern segment of the Huya fault that are transverse compressional Copyright: © 2021 by the authors. structures absorb the sinistral strike-slip motion of the remaining part of the East Kunlun Licensee MDPI, Basel, Switzerland. fault, which causes the Minshan uplift on the eastern margin of the Tibetan Plateau [4,5]. This article is an open access article The complicated and intense tectonic activities of these faults make the eastern margin distributed under the terms and of the Bayan Har block one of the most seismically active regions in China, with many conditions of the Creative Commons strong earthquakes having occurred throughout its history. Examples include the 1654 Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ Tianshui earthquake (Ms = 8), the 1879 Wudu earthquake (Ms = 7), the 1933 Diexi earth- 4.0/). quake (Ms = 7.5), the 1976 Songpan earthquakes (Ms = 7.2, 6.7, and 7.2), the 2008 Wenchuan Remote Sens. 2021, 13, 1573. https://doi.org/10.3390/rs13081573 https://www.mdpi.com/journal/remotesensing
Remote Sens. 2021, 13, x FOR PEER REVIEW 2 of 14 Remote Sens. 2021, 13, 1573 strong earthquakes having occurred throughout its history. Examples include the2 1654 of 12 Tianshui earthquake (Ms = 8), the 1879 Wudu earthquake (Ms = 7), the 1933 Diexi earth- quake (Ms = 7.5), the 1976 Songpan earthquakes (Ms = 7.2, 6.7, and 7.2), the 2008 Wen- chuan earthquake earthquake (Ms = 8),(Msand= 8), theand 2013the 2013 Lushan Lushan earthquakeearthquake (Ms = 7)(Ms = 7) 1). (Figure (Figure 1). The The shorter shorter duration, continuous earthquakes, and smaller strain of release duration, continuous earthquakes, and smaller strain of release compared to previous compared to pre- phases of earthquake clustering show that the Bayan Har block is experiencing a peak ina vious phases of earthquake clustering show that the Bayan Har block is experiencing peak in activity seismic seismic [4]. activity [4]. Therefore, Therefore, further further study of study of its seismogenic its seismogenic environment environment is urgently is urgentlyThe needed. needed. 2017 The 2017 Jiuzhaigou Jiuzhaigou earthquake earthquake provides provides a great opportunity a great opportunity to under- to understand the stand theloading tectonic tectonicmechanisms loading mechanisms in thisDifferent in this region. region. Different researchresearch institutions institutions have have issued aissued seriesaofseries of reports reports on themechanisms on the focal focal mechanisms of the Jiuzhaigou of the Jiuzhaigou earthquake. earthquake. All of All them of them shown have have shown thatearthquake that this this earthquake was a was a sinistral sinistral strike-slip strike-slip eventawith event with highadip high dip angle angle and and a moment a moment magnitude magnitude (Mw) (Mw) of 6.5,of 6.5, but but are there there are significant significant discrepancies discrepancies in thein the fault fault geometry geometry and depths and focal focal depths (Table(Table 1). 1). Figure 1. Tectonic setting and seismicity around the 2017 Jiuzhaigou earthquake. (a) Tectonic setting of the eastern margin Figure 1. Tectonic setting and seismicity around the 2017 Jiuzhaigou earthquake. (a) Tectonic setting of the eastern margin of the Tibetan Plateau. The solid blue and purple lines indicate the primary and secondary block boundaries, respectively. of the Tibetan Plateau. The solid blue and purple lines indicate the primary and secondary block boundaries, respectively. The color-coded circles represent historical earthquakes (Ms ≥ 5) from the Data Sharing Infrastructure of National Earth- The quakecolor-coded Data Centercircles (NEDC)represent historical earthquakes (http://data.earthquake.cn, (Ms ≥on5)14 accessed from the2021). April DataThe Sharing black Infrastructure beach balls show of National the focal Earthquake mechanisms of the historical earthquakes (Ms ≥ 6.5) in this region from the Global Centroid Moment Tensor,show Data Center (NEDC) (http://data.earthquake.cn, accessed on 14 April 2021). The black beach balls the Harvard focal mechanisms of the historical earthquakes (Ms ≥ 6.5) in this region from the Global Centroid Moment University Global Moment Tensor Solution (GCMT). The red beach ball represents the focal mechanism of the Jiuzhaigou Tensor, Harvard University earthquakeGlobal Moment from the Tensor Institute Solution (GCMT). of Geophysics, The red beach China Earthquake ball represents Administration the focal(http://www.cea-igp.ac.cn/en/, (CEA-IGP) mechanism of the Jiuzhaigou accessed on 14 April 2021). The inset shows the large-scale tectonic environment. The red rectangle is the region shown in earthquake from the Institute of Geophysics, China Earthquake Administration (CEA-IGP) (http://www.cea-igp.ac.cn/en/, (b). Note: QT—Qiangtang block; BH—Bayan Har block; QL—Qilian block; SC—South China; NC—North accessed on 14 April 2021). The inset shows the large-scale tectonic environment. The red rectangle is the region China; shownTB— in Tibet; TR—Tarim Basin. (b) Fault geometry and aftershocks of the Jiuzhaigou earthquake. The (b). Note: QT—Qiangtang block; BH—Bayan Har block; QL—Qilian block; SC—South China; NC—North China; TB—Tibet;blue dots are the relocated aftershocks within a month from Fang et al. [6], and the red rectangle is the projection of our optimal fault model on the TR—Tarim Basin. (b) Fault geometry and aftershocks of the Jiuzhaigou earthquake. The blue dots are the relocated surface. The gray, black, red, and purple beach balls show the focal mechanisms determined by the GCMT, United States aftershocks within a month from Fang et al. [6], and the red rectangle is the projection of our optimal fault model on the Geological Survey (USGS) (https://www.usgs.gov/, accessed on 14 April 2021), CEA-IGP, and Institute of Earthquake surface. The gray, Forecasting, Chinablack, red, and Earthquake purple beach balls Administration show (http://www.ief.ac.cn/, (CEA-IEF) the focal mechanisms determined accessed onby14the GCMT, April 2021),United States respectively. Geological Survey (USGS) (https://www.usgs.gov/, accessed on 14 April 2021), CEA-IGP, and Institute of Earthquake Forecasting, China Earthquake Administration No significant (CEA-IEF) (http://www.ief.ac.cn/, coseismic surface rupture was accessed on 14 April observed 2021), for the respectively.earth- Jiuzhaigou quake, making it challenging to determine the seismogenic fault [3,5]. Therefore, surface Table 1. Focal mechanisms of the 2017 Jiuzhaigou earthquake determined by different institutions. deformation as a direct response to the earthquake provides critical information for stud- ying the seismogenic Nodalfault Plane and I rupture model. Nodal PlaneAs II one of the important means for obtain- ing surface deformation, interferometric synthetic aperture Magnitude (Mw) radar (InSAR)Depth (km)ad- has the Strike, Dip, Rake Strike, Dip, Rake vantages of a wide◦geographical span, high precision, and high spatial resolution [7]. It is GCMT 242 , 77◦ , −168◦ 150◦ , 78◦ , −13◦ widely used to decipher earthquake cycle processes, including6.5 14.9 the interseismic locking [8], USGS 246◦ , 57◦ , −173◦ 153◦ , 84◦ , −33◦ 6.5 13.5 coseismic CEA-IGP rupture [9,10], and 328◦ , 48◦ , −11◦ postseismic afterslip 65◦ , 82◦ , 137◦ and viscoelastic 6.5 relaxation [11]. 20.0 Re- CEA-IEF 59◦ , 77◦ , 164◦ 152◦ , 75◦ , 13◦ 6.5 5.0
Remote Sens. 2021, 13, 1573 3 of 12 No significant coseismic surface rupture was observed for the Jiuzhaigou earthquake, making it challenging to determine the seismogenic fault [3,5]. Therefore, surface defor- mation as a direct response to the earthquake provides critical information for studying the seismogenic fault and rupture model. As one of the important means for obtaining surface deformation, interferometric synthetic aperture radar (InSAR) has the advantages of a wide geographical span, high precision, and high spatial resolution [7]. It is widely used to decipher earthquake cycle processes, including the interseismic locking [8], co- seismic rupture [9,10], and postseismic afterslip and viscoelastic relaxation [11]. Recently, InSAR has been applied to monitor induced earthquakes, which is helpful to scientifically guide industrial production [12,13]. After the Jiuzhaigou earthquake, many studies on surface deformation and seismotectonics have been carried out using InSAR technology. Shan et al. [14] used Sentinel-1A data to acquire the coseismic surface deformation fields and inverted the rupture model. Hong et al. [15] derived the dislocation distribution from InSAR and Global Positioning System (GPS) data. Zhao et al. [16] retrieved the coseis- mic deformation field from Sentinel-1A InSAR data. Although a lot of research has been performed on the seismogenic fault of the Jiuzhaigou earthquake, a unified knowledge has not been achieved. From previous studies, we know that different one-fault models have a wide range of dips (50◦ –84◦ ) and a small range of strikes (152◦ –155◦ ) [1,5,9,14–20]. The two-fault models using the aftershock gap as the boundary have a small difference in fault geometry on the southern fault (strike: 145◦ –148◦ , dip: 84◦ –88◦ ), but a significant difference on the northern fault (strike: 151◦ –171◦ , dip: 77◦ –88◦ ) [2,21,22]. Sun et al. [10] proposed a more complex three-fault model to resolve the crustal deformation, including a fault branch forming an obtuse angle with the main subvertical fault at the northern end [23,24]. Thus, it is necessary to further discuss the seismogenic fault and rupture model of the Jiuzhaigou earthquake. Accurate coseismic slip distribution could provide a reasonable driving source for the postseismic viscoelastic relaxation analysis, Coulomb stress change assessment, and ground motion simulation. Moreover, it is of great practical significance for improving our knowledge of the background tectonics of the source region and scientifically guiding earthquake relief efforts. In this study, four finite fault models are tested, and a brand-new two-fault model composed of a main fault and a secondary fault is finally determined as the optimal model of the 2017 Jiuzhaigou earthquake, elucidating a complex fault system. In addition, we further discuss the possible reasons for the aftershock distribution and the occurrence of the aftershock gap. 2. Data and Methods 2.1. Data We derived the coseismic deformation field in the line-of-sight direction based on the Sentinel-1 synthetic aperture radar (SAR) images following the strategies of Ji et al. [3]. Interferometric processing was carried out using the GAMMA software. Two images, including a master image (20170730) and a slave image (20170811), from the ascending Path128 were used to obtain the interferogram based on the two-pass InSAR method [25]. The shuttle radar topography mission (SRTM) digital elevation model (DEM) with 30 m resolution was used to remove the topographic phase. In order to suppress the noise, the interferogram was multi-looked by a factor of 8 in range and 2 in azimuth and filtered twice using the weighted power spectrum algorithm with windows of 128 × 128 and then 32 × 32 pixels [26]. The minimum cost flow algorithm was used to phase unwrapping. The residual orbital phase in the interferogram was removed by a nonlinear least-squares adjustment [27]. To minimize the atmospheric signal caused by the layered atmosphere, the atmospheric delay model was established based on the existing digital elevation model and removed from the interferogram [27].
Remote Sens. 2021, 13, 1573 4 of 12 2.2. Methods According to the Okada elastic dislocation theory [28], there is a linear relationship between the coseismic surface displacement and the fault dislocation distribution. Ideally, we would directly use the least-squares method to invert the fault dislocation distribution. However, due to practical limitations, to describe the fault more objectively, we need to discretize it into many meshes. Moreover, the distribution of the surface observation is usually heterogeneous, and the Green function matrix is often rank-deficient or ill- conditional. Generally, additional constraints are needed to ensure the stability of the inversion, and a dislocation smoothing constraint is usually used: F (s) = |Gs − d| 2 + α2 |Lτ | 2 → min, where G, s, and d represent the Green function, the slip vector of each sub-fault, and the displacement vector, respectively, and ||Gs − d||2 is the variance between the observations and predictions. The second term is the smoothing constraint, where L represents the finite difference approximation of the Laplacian operator multiplied by a weighting factor proportional to the slip amplitude, and τ is the shear stress drop that is linearly related to the slip distribution. The parameter α is the smoothing factor, which controls the trade-off between the model roughness and misfit. Generally, the “inflection point” is taken as the optimal value. We used the steepest descent method (SDM) [29–32] to estimate the slip model of the Jiuzhaigou earthquake. Compared with other inversion methods, the SDM has the advantage of providing stable inversion results with high efficiency. The Green function used for the inversion was based on the Crust1.0 layered model [33]. 3. Model Tests and Results We first constructed a one-fault model (Figure 2a) to retrieve the surface deformation. To obtain a high-resolution finite fault slip model, we divided the entire fault into 247 rect- angular sub-faults with a grid of 2 km × 2 km. As shown in Figure S1, we choose an optimal smoothing factor of 0.1 based on the trade-off curve. The optimal strike and dip are 154◦ and 51◦ , respectively, which is similar to the results of Shan et al. [14] (strike = 153◦ , dip = 50◦ ). The one-fault model reproduces most of the observations, but it cannot decipher the surface deformation on the southeastern end of the fault satisfactorily (Table 2, Figure 3c). The slip distribution is mainly dominated by sinistral strike slips, with some thrust components on the southern segment and small normal components on the northern segment (Figure 4a). Unfortunately, this model has a significant shallow slip on the southern end of the fault, which is not consistent with the fact that there was no surface rupture. Table 2. Parameters of finite fault models. Average Error Maximum Moment Magnitude Faults Strike Dip (m) Residual (m) (Mw) One-fault Main fault 154◦ 51◦ 0.0211 −0.0900 6.4 Northern fault 154◦ 66◦ Two-fault 0.0203 −0.0684 6.4 Southern fault 147◦ 74◦ Northern fault 156◦ 77◦ Three-fault Southern fault 147◦ 82◦ 0.0198 −0.0602 6.4 Secondary fault 204◦ 80◦ Main fault 151◦ 77◦ New two-fault 0.0188 −0.0487 6.5 Secondary fault 196◦ 77◦
Remote Sens. 2021, 13, 1573 5 of 12 Remote Sens. 2021, 13, x FOR PEER REVIEW 5 of 14 Remote Sens. 2021, 13, x FOR PEER REVIEW 6 of 14 80°) (Figure S3), improving the data fit in the southwestern region compared to the two- fault model (Table 2, Figure 3h). Figure 4c shows the slip distribution, whose high-slip zone is located at a depth of 2–10 km on the main fault, with a peak slip of 1.30 m. The slip of the northern fault is composed of sinistral strike-slip and normal motion, while the southern fault is almost exclusively composed of sinistral strike-slip motion. We conclude that the average slip of the northern fault is larger than that of the southern fault, and that the secondary fault extends to the northeast of the main fault, which is consistent with the large number of aftershocks that occurred in this region. Table 2. Parameters of finite fault models. Average Error Maximum Resid- Moment Magni- Faults Strike Dip (m) ual (m) tude ( ) One-fault Main fault 154° 51° 0.0211 −0.0900 6.4 Northern fault 154° 66° Two-fault 0.0203 −0.0684 6.4 Southern fault 147° 74° Northern fault 156° 77° Three-fault Figure 2. Projections Southernoffault the four finite147° fault models82° on the surface, 0.0198including the−0.0602 (a) one-fault 6.4 Figure (b)Projections model, 2. of the two-fault model four finite (a northern fault fault and modelsfault), a southern on the (c) surface, three-faultincluding model, and the (d) (a) one-fault model, Secondary fault 204° 80° brand-new two-fault (b) two-faultMain model model (a main (a northern151°fault and a secondary fault and 77° fault). Other symbols are the same as a southern fault), (c) three-fault model, and (d) brand-new Figure 1b. fault New two-fault two-fault Secondary model (a main fault and a secondary 0.0188 −0.0487 6.5 fault 196° 77° fault). Other symbols are the same as Figure 1b. In addition, given that the depth of the aftershocks is different in the north and the south and that there is a seismic gap [6], we also constructed a two-fault model similar to the previous studies (Figure 2b, [2,3,19,21,22,34]). The trade-off curve is shown in Figure S2a, and we set the smoothing factor as 0.15. Results demonstrate that the optimal model is composed of a northern fault (strike = 154°, dip = 66°) and a southern fault (strike = 147°, dip = 74°) (Figure S2). Our preferred strikes are similar to the model of Zheng et al. [2], and the optimal dip in the north is slightly smaller than that in the south [3,6,21,34]. Figure 3f shows that the two-fault model further improves the fit of the data (Table 2). It has a similar slip pattern to the results of joint inversion by Zheng et al. [2], whose average slip on the northern fault is larger than that on the southern fault, with a peak slip of around 0.9 m. The differences are that our results have a partial normal slip component on the northern fault, a smaller slip on the aftershock gap, and no significant slip near the surface (Figure 4b). This is consistent with the fact that the Jiuzhaigou earthquake did not result in surface rupture. Based on the coseismic deformation field and the aftershock distribution, Sun et al. [10] proposed a fault branch at the northern end of the main fault plane at an obtuse angle. The geological survey also illustrates that there is indeed a secondary fault in this area [35]. Therefore, we introduced a similar secondary fault to construct a three-fault model (Figure 2c). During the inversion, we applied the same smoothing factor as the two-fault model. The optimal three-fault model is composed of a northern fault (strike = 156°, dip = 77°), a southern fault (strike = 145°, dip = 82°), and a secondary fault (strike = 204°, dip = Figure 3. Comparison between interferometric synthetic aperture radar (InSAR) observations and predictions from the Figure three test models. 3. Comparison (a,d,g) between Coseismic line-of-sight interferometric (LOS) synthetic deformation. Simulated aperture LOS radar deformation from(InSAR) observations and the (b) one-fault model, (e) two-fault model, and (h) three-fault model. (c,f,i) Residuals. Red star represents the epicenter predictions from the three test models. (a,d,g) Coseismic line-of-sight (LOS) deformation. of the 2017 Jiu- Simulated zhaigou earthquake from the CEA-IGP. LOS deformation from the (b) one-fault model, (e) two-fault model, and (h) three-fault model. (c,f,i) Residuals. Red star represents the epicenter of the 2017 Jiuzhaigou earthquake from the CEA-IGP.
Remote Sens. 2021, 13, 1573 6 of 12 Remote Sens. 2021, 13, x FOR PEER REVIEW 7 of 14 Figure 4. Slip distribution with arrows delineating the average rake of each sub-fault for the three test models, whose Figure 4. Slip distribution with arrows delineating the average rake of each sub-fault for the three test models, whose magnitude is color-coded. (a) Slip distribution for the optimal one-fault model. (b) Slip distribution for the optimal two- magnitude is color-coded. (a) Slip distribution for the optimal one-fault model. (b) Slip distribution for the optimal two-fault fault model, including the (1) northern fault and (2) southern fault. (c) Slip distribution for the optimal three-fault model, model, including including the (1) northern the (1) secondary fault, (2)fault and (2) northern southern fault, and (3)fault. (c) Slip southern fault.distribution for the optimal three-fault model, including the (1) secondary fault, (2) northern fault, and (3) southern fault. Given that the fault geometry of the southern fault is similar to that of the northern In addition, given that the depth of the aftershocks is different in the north and the fault in the optimal three-fault model, we proposed a simpler two-fault model including south and that there is a seismic gap [6], we also constructed a two-fault model similar to the a main fault and a secondary fault that is at an obtuse angle to the northern end of the previous studies (Figure 2b, [2,3,19,21,22,34]). The trade-off curve is shown in Figure S2a, main fault (Figure 2d). The trade-off curve shows that the optimal smoothing factor is and we set the smoothing factor as 0.15. Results demonstrate that the optimal model is consistent with the three-fault model. Figure 5 illustrates that the optimal strikes of the composed of a northern fault (strike = 154◦ , dip = 66◦ ) and a southern fault (strike = 147◦ , main fault (Seg1) and the secondary fault (Seg2) are 151° and 196°, respectively, and that dip = 74◦ ) (Figure S2). Our preferred strikes are similar to the model of Zheng et al. [2], and their dips are the optimal dipboth 77°. in the Interestingly, north this brand-new is slightly smaller than that two-fault in the southmodel has theFigure [3,6,21,34]. best fit 3f among the four models (Table 2). The southeastern source region, which shows that the two-fault model further improves the fit of the data (Table 2). It has a similar is poorly recov- ered by the other slip pattern to themodels, resultsisofwelljointconstrained inversion by by Zheng this model et al.(Figure 6). Results [2], whose average showslipthat on 18 the released seismic moment is around 6.3 × 10 N∙m, equivalent the northern fault is larger than that on the southern fault, with a peak slip of around to an Mw 6.5 event, consistent 0.9 m. Thewith the results differences areof several that different our results institutions have a partial (Table normal1).slip Most of the coseismic component on the slip on the northern main fault, fault occurred a smaller slip on thearound two shallow aftershock gap, and asperities, whichslip no significant arenear separated by a the surface seismic gap with few aftershocks. The rupture is dominated by sinistral (Figure 4b). This is consistent with the fact that the Jiuzhaigou earthquake did not result in slip, accompanied by a small surface number of normal components. The high-slip regions are distributed between rupture. 2–8 kmBased ona the with peak slip of 1.51 coseismic m, consistent deformation field with no aftershock and the surface rupture (FigureSun distribution, 7c).etInal. addi- [10] tion, most aftershocks are concentrated under the high-slip patches proposed a fault branch at the northern end of the main fault plane at an obtuse an- and appear a comple- mentary gle. The pattern geologicalwith the coseismic survey slip. Interestingly, also illustrates that there isthe slip on indeed the secondary a secondary faultfault is in this also areamainly concentrated [35]. Therefore, we in two patches, introduced but it is a similar separatedfault secondary by antoaftershock construct swarm. They a three-fault extend up to 112c). model (Figure km, with athe During peak slip of 0.85 inversion, m, coinciding we applied the same with the depth smoothing of aftershocks factor as the two- (Figures 2d and 7b). Therefore, we propose that this two-fault fault model. The optimal three-fault model is composed of a northern fault (strike model is our = 156◦ , preferred dip = 77 ), a southern fault (strike = 145 , dip = 82 ), and a secondary fault (strike = 204◦ , model for◦ the Jiuzhaigou earthquake. ◦ ◦ dip = 80◦ ) (Figure S3), improving the data fit in the southwestern region compared to the two-fault model (Table 2, Figure 3h). Figure 4c shows the slip distribution, whose high-slip zone is located at a depth of 2–10 km on the main fault, with a peak slip of 1.30 m. The slip of the northern fault is composed of sinistral strike-slip and normal motion, while the southern fault is almost exclusively composed of sinistral strike-slip motion. We conclude that the average slip of the northern fault is larger than that of the southern fault, and that the secondary fault extends to the northeast of the main fault, which is consistent with the large number of aftershocks that occurred in this region.
Remote Sens. 2021, 13, 1573 7 of 12 Given that the fault geometry of the southern fault is similar to that of the northern fault in the optimal three-fault model, we proposed a simpler two-fault model including a main fault and a secondary fault that is at an obtuse angle to the northern end of the main fault (Figure 2d). The trade-off curve shows that the optimal smoothing factor is consistent with the three-fault model. Figure 5 illustrates that the optimal strikes of the main fault (Seg1) and the secondary fault (Seg2) are 151◦ and 196◦ , respectively, and that their dips are both 77◦ . Interestingly, this brand-new two-fault model has the best fit among the four models (Table 2). The southeastern source region, which is poorly recovered by the other models, is well constrained by this model (Figure 6). Results show that the released seismic moment is around 6.3 × 1018 N·m, equivalent to an Mw 6.5 event, consistent with the results of several different institutions (Table 1). Most of the coseismic slip on the main fault occurred around two shallow asperities, which are separated by a seismic gap with few aftershocks. The rupture is dominated by sinistral slip, accompanied by a small number of normal components. The high-slip regions are distributed between 2–8 km with a peak slip of 1.51 m, consistent with no surface rupture (Figure 7c). In addition, most aftershocks are concentrated under the high-slip patches and appear a complementary pattern with the coseismic slip. Interestingly, the slip on the secondary fault is also mainly concentrated in two patches, but it is separated by an aftershock swarm. They extend up to 11 km, with a Remote Sens. 2021, 13, x FOR PEER REVIEW 8 of 14 peak Remote Sens. 2021, 13, x FOR PEER REVIEW slip of 0.85 m, coinciding with the depth of aftershocks (Figures 2d and 7b). Therefore, 8 of 14 we propose that this two-fault model is our preferred model for the Jiuzhaigou earthquake. Figure 5. Search results of our preferred model. (a) Trade-off curve between the model roughness and misfit. Dips of the Figure Figure5. Search results of our preferred model. (a)(a) Trade-off curve between thethemodel roughness and misfit. Dips of the (b) (b) main5.fault Search results (Seg1) and of(c)our preferred secondary model. fault (Seg2). Trade-off Strikes curve of the (d) between model main fault (Seg1) roughness and and (e) secondary misfit. Dips fault (Seg2).of the main fault (Seg1) and (c) secondary fault (Seg2). Strikes of the (d) main fault (Seg1) and (e) secondary fault (b) main fault (Seg1) and (c) secondary fault (Seg2). Strikes of the (d) main fault (Seg1) and (e) secondary fault (Seg2).(Seg2). Figure 6. Comparison between InSAR observations and predictions from our preferred model. (a) Coseismic LOS defor- Figure 6. mation. Comparison between InSAR observations and predictions symbolsfrom oursame preferred model. (a) Coseismic LOS defor- Figure 6.(b)Comparison Simulated LOS deformation. between InSAR (c) Residuals. observations Other arefrom and predictions the our as Figure preferred 3.model. (a) Coseismic LOS mation. (b) Simulated LOS deformation. (c) Residuals. Other symbols are the same as Figure 3. deformation. (b) Simulated LOS deformation. (c) Residuals. Other symbols are the same as Figure 3.
Remote Sens. 2021, 13, 1573 8 of 12 Remote Sens. 2021, 13, x FOR PEER REVIEW 9o Figure 7. Slip distribution fordistribution Figure 7. Slip our preferred formodel. (a) Projection our preferred model.of(a) theProjection slip modelofon thethe surface. slip modelSlip distribution on the surface. of the (b) secondary fault, and (c) main fault. The white dots represent the relocated aftershocks. The slip distribution with ar- Slip distribution of the (b) secondary fault, and (c) main fault. The white dots represent the relocated rows delineating the average rake of each sub-fault is color-coded. Other symbols are the same as Figure 1b. aftershocks. The slip distribution with arrows delineating the average rake of each sub-fault is color-coded. Other symbols are the same as Figure 1b. 4. Discussion 4. Discussion 4.1. Uncertainty Test 4.1. Uncertainty Test In order to verify the effectiveness of our preferred model, we applied the bootstr In order to verify method the effectiveness to estimate of our preferred the uncertainty. model, is Bootstrapping wea applied statistical theestimation bootstrap method method to randomly estimate the uncertainty. resamples Bootstrapping from the original dataistoacreate statistical estimation multiple samples. method. That is, m grou It randomlyofresamples data are from the original randomly data selected to create from multiple n groups samples.data of original That foris,inversion, m groups of and B resu data are randomly could be selected obtainedfrombynrepeating groups ofthe original data forB inversion, experiment times, whichandallows B results thecould standard dev be obtainedtions by repeating the experiment to be inferred B times, [36]. Therefore, which the allows standard the standard deviation deviations obtained by the bootstr to be inferred [36]. Therefore, the standard deviation obtained by the bootstrap method not only contains the influence of the original observation error, but also refle method not only contains the errorthe influence caused of the data by different original observation distribution. error, but It is widely used also reflects in the the of mo estimation error caused by different data uncertainty [36,37]. distribution. It is widely used in the estimation of model uncertainty [36,37]. To ensure reliable results, the value of B was set to 100 [38], and correspondi To ensure reliable results,were resampling datasets the value createdofinBthe was setThese case. to 100datasets [38], andwere corresponding used for inversions w resampling the datasets were created in the case. These datasets were used same model space. Figure 8 shows the estimated uncertainties based onfor inversions with the bootstr the same model space. Figure 8 shows the estimated uncertainties based on method, and the maximum standard deviation is less than 0.22 m. Although the standathe bootstrap method, and the maximum deviation mainlystandard deviation concentrates in theissouthern less thanpart 0.22ofm.the Although main fault, the standard it is small relative deviation mainly concentrates in the southern part of the main fault, it the peak slip of asperity. Besides, the standard deviation in the aftershock is small relative gap is very lo to the peakindicating slip of asperity. Besides, the standard deviation in the aftershock gap is very there is indeed a slip deficit in this region. Overall, the uncertainty of our op low, indicating there is indeed mal two-fault model ais slip verydeficit in this region. low, revealing Overall,distribution the dislocation the uncertainty of is reliable and s our optimal two-fault model is very low, revealing the dislocation distribution is reliable ble. and stable.
Remote Sens. 2021, 13, 1573 9 of 12 Remote Sens. 2021, 13, x FOR PEER REVIEW 10 of 14 Figure Figure 8. 8. The The standard standard deviations deviations of of the the (a) (a) main main fault, fault, and and (b) (b) secondary secondary fault, based on fault, based on the the bootstrap bootstrap method. method. 4.2. 4.2. Fault Fault Geometry Geometry and and Properties Properties In In our ouroptimal optimalmodel, model,the thesecondary secondary fault is atisaatlarge fault obtuse a large obtuseangleangle to thetomain the fault, main termed backward branching. Similar phenomena were also observed fault, termed backward branching. Similar phenomena were also observed in the 1992 in the 1992 Landers earthquake [39] and the Landers earthquake [39]1999 and Hector the 1999 Mine earthquake Hector [40]. The mechanism Mine earthquake of this back- [40]. The mechanism of ward branching this backward is the rupture branching is the stopping or slowing rupture stopping down on or slowing the main down on the fault plane, main faultwhich plane, causes which the rupture causes to jump to the rupture to jump a nearbyto afault [39].fault nearby This[39]. phenomenon can also becan This phenomenon accommo- also be dated by a significant accommodated shallow slip by a significant deficit shallow [40]. slip Apparently, deficit no significant [40]. Apparently, surface rupture no significant surface was observed rupture in the Jiuzhaigou was observed earthquake. in the Jiuzhaigou There isThere earthquake. a small is aslip small zoneslipatzone the south at the of the south junction between of the junction the backward between the backward branching branchingandandthethe main main fault, fault,which whichmay mayhave have been caused by a rigid caused rigid block blockstopping stoppingor orslowing slowingdowndownthe therupture rupture onon thethemain main fault. Moreover, fault. Moreo- the 1992 ver, Landers the 1992 earthquake Landers and the earthquake and1999 theHector Mine earthquake 1999 Hector Mine earthquakeboth occurred within a both occurred complex within a fault complexsystem. faultThe Jiuzhaigou system. earthquakeearthquake The Jiuzhaigou appears to be a similar appears case, to be whose main a similar case, seismogenic whose fault is the hidden main seismogenic fault isfaults with nofaults the hidden obvious withsurface trace surface no obvious [10] or afterslip trace [10] [41], or indicating a young fault [10,41]. This less mature fault system is characterized afterslip [41], indicating a young fault [10,41]. This less mature fault system is character- by weak faultby ized planes weakand more fault complex planes and morerupture processes. complex rupture processes. 4.3. Aftershock 4.3. Aftershock Distribution Distribution Based on Based onthethedistribution distribution of of relocated aftershocks, relocated therethere aftershocks, is an aftershock gap of around is an aftershock gap of 5 km, which may be caused by one or multiple of the following around 5 km, which may be caused by one or multiple of the following three factors: three factors: (1) The(1) coseismic slip was large and the stress release was sufficient; (2) The strikes of the faults The coseismic slip was large and the stress release was sufficient; (2) The strikes of the changed at this turning point; (3) There were unbroken asperities [6]. From the InSAR faults changed at this turning point; (3) There were unbroken asperities [6]. From the In- surface deformation (Figure 3), we find that there are few surface deformations in the SAR surface deformation (Figure 3), we find that there are few surface deformations in aftershock gap with the indication of no significant coseismic slip below, which is also the aftershock gap with the indication of no significant coseismic slip below, which is also verified by our preferred slip model. We thus rule out the first possibility. During our finite verified by our preferred slip model. We thus rule out the first possibility. During our fault tests, the three-fault model suggests that the geometry parameters of the northern finite fault tests, the three-fault model suggests that the geometry parameters of the north- fault are similar to those of the southern fault, precluding the second possibility. In addition, ern fault are similar to those of the southern fault, precluding the second possibility. In Li et al. [41] stated that no early afterslip is observed following the Jiuzhaigou earthquake, addition, Li et al. [41] stated that no early afterslip is observed following the Jiuzhaigou indicating the accumulated strain is unlikely to be released by postseismic deformation in earthquake, indicating the accumulated strain is unlikely to be released by postseismic the gap. Therefore, we argue that unbroken asperities hindered the coseismic rupturing, deformation in the gap. Therefore, we argue that unbroken asperities hindered the coseis- resulting in this small seismic gap. mic rupturing, resulting in Figure 7 illustrates this that small most seismic gap. aftershocks occurred in the down-dip regions of the Figurehigh-slip coseismic 7 illustrates that most asperities, aftershocks presenting occurred in the a complementary down-dip pattern. regions Generally, of the it can be coseismic high-slip asperities, presenting a complementary pattern. Generally, interpreted as (1) direct triggering from coseismic stress change [42], or (2) aftershocks it can be interpreted as (1) direct triggering from coseismic stress change [42], or that are driven by afterslip in the velocity-strengthening zone [30,43]. Therefore, we first (2) aftershocks that usedarethedriven by afterslip numerical in the velocity-strengthening code PSGRN/PSCMP [44] followingzone [30,43]. Therefore, the strategies of Guo etwe al. first [45] used the numerical code PSGRN/PSCMP [44] following the strategies to estimate the static Coulomb stress changes caused by the coseismic slip. Since of Guo et al. [45]theto estimate Jiuzhaigou theearthquake static Coulomb is a stress changes strike-slip caused event, we bysetthe thecoseismic effective slip. Since friction the Jiuzhai- coefficient to gou earthquake is a strike-slip event, we set the effective friction coefficient 0.4 [24,45–47]. We used the optimal slip model as the driving source. The receiver fault to 0.4 [24,45– 47]. was We used set to ourthe optimalfault preferred slip model. model as the driving Results show thatsource. mostThe receiver fault aftershocks was in occurred settheto our preferred fault model. Results show that most aftershocks occurred in the Coulomb
Remote Sens. Remote Sens. 2021, 2021, 13, 13, 1573 x FOR PEER REVIEW 11 of 10 of 12 14 stress shadows (Figure S4). Therefore, we propose that the aftershock distribution may be Coulomb stress shadows (Figure S4). Therefore, we propose that the aftershock distribution related to aseismic afterslip, indicating the velocity-strengthening region. Further research may be related to aseismic afterslip, indicating the velocity-strengthening region. Further is needed here. Helmstetter and Shaw [48] suggested that the aftershocks may be related research is needed here. Helmstetter and Shaw [48] suggested that the aftershocks may be to the tectonic stress on the fault, which could explain the inconsistency between the related to the tectonic stress on the fault, which could explain the inconsistency between mechanisms of the aftershocks and that of the mainshock. Therefore, we calculated the the mechanisms of the aftershocks and that of the mainshock. Therefore, we calculated static Coulomb the static Coulomb stress changes stress on the changes on optimally the optimallyoriented failure oriented planes failure for the planes for 2017 Jiu- the 2017 zhaigou Jiuzhaigou earthquake earthquake under thethe under tectonic background tectonic background stress field. stress In this field. study, In this we set study, we the set maximum principal stress, middle principal stress, and minimum principal the maximum principal stress, middle principal stress, and minimum principal stress stress to −10to 4 kPa, −104−10kPa,kPa, 3 −103and kPa,0 and kPa,0respectively, basedbased kPa, respectively, on previous studies on previous [23,24]. studies Although [23,24]. Althoughthe tectonic stresses may have large uncertainties, it does not prevent us from the tectonic stresses may have large uncertainties, it does not prevent us from analyzing analyzing them rationally. FigureFigure them rationally. 9 reveals the static 9 reveals theCoulomb stress changes static Coulomb at different stress changes depths and at different the depths distribution of the aftershocks and the distribution within 5within of the aftershocks km, from 5 km,which from we whichfindwethat most find thatof the of most after- the shocks occurred in the region where the Coulomb stress increased. Therefore, aftershocks occurred in the region where the Coulomb stress increased. Therefore, the the after- shocks of theofJiuzhaigou aftershocks earthquake the Jiuzhaigou may have earthquake maybeenhavecontrolled by the background been controlled tectonic by the background stress field. tectonic stress field. Figure 9. Figure 9. Static Static Coulomb Coulomb stress changes on stress changes on the the 3D 3D optimally optimally oriented failure planes oriented failure planes created created by by the the Jiuzhaigou Jiuzhaigou earthquake earthquake at different depths and the distribution of the aftershocks within 5 km. Coulomb stress changes at a depth of (a) 5 km, (b) at different depths and the distribution of the aftershocks within 5 km. Coulomb stress changes at a depth of (a) 5 km, 10 km, and (c) 15 km. Black dots represent the relocated aftershocks. (b) 10 km, and (c) 15 km. Black dots represent the relocated aftershocks. 5. 5. Conclusions Conclusions In In this this study, we tested study, we tested four four different different finite finite fault fault models models based based on on the the InSAR InSAR data, data, and and determined determined that that the the seismogenic fault of seismogenic fault of the the 2017 2017 MwMw 6.5 6.5 Jiuzhaigou earthquake was Jiuzhaigou earthquake was a brand-new brand-new two-fault two-faultmodel. model.The Thehigh-slip high-slip regions regionsin in thethe main fault main are are fault distributed distributedbe- tween between 2–82–8 km, with km, witha peak slipslip a peak of 1.51 m. m. of 1.51 There are are There twotwo asperities thatthat asperities are are perfectly de- perfectly marcated demarcated by by thethe aftershock aftershock gap, which gap, may which mayindicate a barrier. indicate TheThe a barrier. secondary faultfault secondary is atisa large at a large obtuse angle to the main fault, whose slip extends to 11 km, with a peak 0.85 obtuse angle to the main fault, whose slip extends to 11 km, with a peak slip at slip m. The m. at 0.85 mainThefault mainoffault the Jiuzhaigou earthquake of the Jiuzhaigou is the NW-trending earthquake is the NW-trendinghiddenhidden fault offault the Huya of the fault, Huyawhich fault, is one ofisthe which onebranches of the East of the branches of Kunlun the Eastfault. Kunlun There were fault. a fewwere There after- a shocks at both ends few aftershocks of the at both main ends fault of the where main faultthe Coulomb where stress increased, the Coulomb which deserve stress increased, which deserve our our particular particular attention. attention. Supplementary Materials: Materials: The Thefollowing followingare are available available online online at www.mdpi.com/xxx/s1, Figure at https://www.mdpi.com/article/10 S1—Search results of the .3390/rs13081573/s1, one-fault Figure model; S1—Search Figure results ofS2—Search results the one-fault of Figure model; the two-fault model; S2—Search Figure results of S3—Search results the two-fault model;of Figure the three-fault S3—Search model; Figure results S4—Static of the Coulomb three-fault model;stress changes Figure caused S4—Static by the Coulomb Jiuzhaigou earthquake stress changes caused byat the different depthsearthquake Jiuzhaigou and the distribution ofdepths at different the aftershocks within 5 km.of the and the distribution aftershocks within 5 km. Author Contributions: Conceptualization, R.G. and X.T.; methodology, J.X.; software, R.G.; valida- tion, AuthorJ.X.,Contributions: R.G., H.S., X.C.Conceptualization, and J.Z.; formal analysis, R.G. andR.G.; writing—original X.T.; methodology, J.X.; draft preparation, software, R.G.;X.T.; val- writing—review and editing, X.T., J.X., R.G., H.S., X.C. and J.Z.; visualization, X.T. and idation, J.X., R.G., H.S., X.C. and J.Z.; formal analysis, R.G.; writing—original draft preparation, R.G.; super- vision, J.X., H.S., X.C. and X.T.; writing—review and J.Z.; project editing, administration, X.T., J.X., R.G., H.S.,X.C. X.C.and and J.Z.; funding J.Z.; acquisition, visualization, X.T. J.X., H.S., and R.G.; X.C. and J.Z. J.X., supervision, All authors have H.S., X.C. andread J.Z.;and agreed project to the published administration, X.C.version offunding and J.Z.; the manuscript. acquisition, J.X., H.S., X.C. This Funding: and J.Z. All authors research have read was funded andB-type by the agreed to the published Strategic version of Priority Program ofthe themanuscript. Chinese Acad- emy of Sciences, grant number XDB41000000, and the National Natural Science Foundation of China, grant numbers 41974023, 41874094, 41674083, 41874026.
Remote Sens. 2021, 13, 1573 11 of 12 Funding: This research was funded by the B-type Strategic Priority Program of the Chinese Academy of Sciences, grant number XDB41000000, and the National Natural Science Foundation of China, grant numbers 41974023, 41874094, 41674083, 41874026. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Acknowledgments: We thank the National Earth System Science Data Center, National Science & Technology Infrastructure of China (http://www.geodata.cn, accessed on 14 April 2021) for the data provided. We are grateful to Lingyun Ji of the Second Monitoring Center, China Earthquake Administration for providing the InSAR data. We used the Generic Mapping Tools (GMT) open- source collection of computer software tools to create the figures, which is developed and maintained by Paul Wessel and Walter H. F. Smith. Conflicts of Interest: The authors declare no conflict of interest. References 1. Li, Q.; Tan, K.; Wang, D.Z.; Zhao, B.; Zhang, R.; Li, Y.; Qi, Y.J. Joint inversion of GNSS and teleseismic data for the rupture process of the 2017 Mw6.5 Jiuzhaigou, China, earthquake. J. Seism. 2018, 22, 805–814. [CrossRef] 2. Zheng, A.; Yu, X.; Xu, W.; Chen, X.; Zhang, W. A hybrid source mechanism of the 2017 Mw 6.5 Jiuzhaigou earthquake revealed by the joint inversion of strong-motion, teleseismic and InSAR data. Tectonophysics 2020, 789, 228538. [CrossRef] 3. Ji, L.; Liu, C.; Xu, J.; Liu, L.; Long, F.; Zhang, Z. InSAR observation and inversion of the seismogenic fault for the 2017 Jiuzhaigou MS7.0 earthquake in China. Chin. J. Geophys. 2017, 60, 4069–4082. 4. Deng, Q.; Cheng, S.; Ma, J.; Du, P. Seismic activities and earthquake potential in the Tibetan Plateau. Chin. J. Geophys. 2014, 57, 2025–2042. 5. Xu, X.; Chen, G.; Wang, Q.; Chen, L.; Ren, Z.; Xu, C.; Wei, Z.; Lu, R.; Tan, X.; Dong, S.; et al. Discussion on seismogenic structure of Jiuzhaigou earthquake and its implication for current strain state in the southeastern Qinghai-Tibet Plateau. Chin. J. Geophys. 2017, 60, 4018–4026. 6. Fang, L.; Wu, J.; Su, J.; Wang, M.; Jiang, C.; Fan, L.; Wang, W.; Wang, C.; Tan, X. Relocation of mainshock and aftershock sequence of the Ms7.0 Sichuan Jiuzhaigou earthquake. Chin. Sci. Bull. 2018, 63, 649–662. [CrossRef] 7. Zebker, H.A.; Rosen, P.A.; Goldstein, R.M.; Gabriel, A.; Werner, C.L. On the derivation of coseismic displacement fields using differential radar interferometry: The Landers earthquake. J. Geophys. Res. Space Phys. 1994, 99, 19617–19634. [CrossRef] 8. Hussain, E.; Wright, T.J.; Walters, R.J.; Bekaert, D.P.S.; Lloyd, R.; Hooper, A. Constant strain accumulation rate between major earthquakes on the North Anatolian Fault. Nat. Commun. 2018, 9, 1–9. [CrossRef] 9. Nie, Z.; Wang, D.-J.; Jia, Z.; Yu, P.; Li, L. Fault model of the 2017 Jiuzhaigou Mw 6.5 earthquake estimated from coseismic deformation observed using Global Positioning System and Interferometric Synthetic Aperture Radar data. Earthplanets Space 2018, 70, 55. [CrossRef] 10. Sun, J.; Yue, H.; Shen, Z.; Fang, L.; Zhan, Y.; Sun, X. The 2017 Jiuzhaigou Earthquake: A Complicated Event Occurred in a Young Fault System. Geophys. Res. Lett. 2018, 45, 2230–2240. [CrossRef] 11. Zhao, D.; Qu, C.; Bürgmann, R.; Gong, W.; Shan, X. Relaxation of Tibetan Lower Crust and Afterslip Driven by the 2001 Mw7.8 Kokoxili, China, Earthquake Constrained by a Decade of Geodetic Measurements. J. Geophys. Res. Solid Earth 2021, 126. [CrossRef] 12. Malinowska, A.A.; Witkowski, W.T.; Guzy, A.; Hejmanowski, R. Mapping ground movements caused by mining-induced earthquakes applying satellite radar interferometry. Eng. Geol. 2018, 246, 402–411. [CrossRef] 13. Sopata, P.; Stoch, T.; Wójcik, A.; Mrocheń, D. Land Surface Subsidence Due to Mining-Induced Tremors in the Upper Silesian Coal Basin (Poland)—Case Study. Remote Sens. 2020, 12, 3923. [CrossRef] 14. Shan, X.; Qu, C.; Gong, W.; Zhao, D.; Zhang, G. Coseismic deformation field of the Jiuzhaigou MS7.0 earthquake from Sentinel-1A InSAR data and fault slip inversion. Chin. J. Geophys. 2017, 60, 4527–4536. 15. Hong, S.; Zhou, X.; Zhang, K.; Meng, G.; Dong, Y.; Su, X.; Zhang, L.; Li, S.; Ding, K. Source Model and Stress Disturbance of the 2017 Jiuzhaigou Mw 6.5 Earthquake Constrained by InSAR and GPS Measurements. Remote Sens. 2018, 10, 1400. [CrossRef] 16. Zhao, D.; Qu, C.; Shan, X.; Gong, W.; Zhang, Y.; Zhang, G. InSAR and GPS derived coseismic deformation and fault model of the 2017 Ms7.0 Jiuzhaigou earthquake in the Northeast Bayanhar block. Tectonophyicsics 2018, 726, 86–99. [CrossRef] 17. Chen, W.; Qiao, X.J.; Liu, G.; Xiong, W.; Jia, Z.G.; Li, Y.; Wang, Y.B.; You, Z.L.; Long, F. Study on the coseismic slip model and Coulomb stress of the 2017 Jiuzhaigou MS7.0 earthquake constrained by GNSS and InSAR measurements. Chin. J. Geophys. 2018, 61, 2122–2132. 18. Shen, W.; Li, Y.S.; Jiao, Q.; Xie, Q.; Zhang, J. Joint inversion of strong motion and InSAR/GPS data for fault slip distribution of the Jiuzhaigou 7.0 earthquake and its application in seismology. Chin. J. Geophys. 2019, 61, 115–129. 19. Xie, Z.; Zheng, Y.; Yao, H.; Fang, L.; Liu, C.; Wang, M.; Shan, B.; Zhang, H.; Ren, J.; Ji, L.; et al. Preliminary analysis on the source properties and seismogenic structure of the 2017 Ms7.0 Jiuzhaigou earthquake. Sci. China Earth Sci. 2018, 61, 339–352. [CrossRef]
Remote Sens. 2021, 13, 1573 12 of 12 20. Gui-Xi, Y.; Feng, L.; Ming-Jian, L.; Hui-Ping, Z.; Min, Z.; You-Qing, Y.; Zhi-Wei, Z.; Yu-Ping, Q.; Si-Wei, W.; Yue, G.; et al. Focal mechanism solutions and seismogenic structure of the 8 August 2017 M7.0 Jiuzhaigou earthquake and its aftershocks, northern Sichuan. Chin. J. Geophys. 2017, 60, 4083–4097. 21. Hu, X.; Sheng, S.; Wang, Y.; Liang, S. Fault Plane Parameters of 2017 Jiuzhaigou Ms7. 0 Earthquake Determined by Aftershock Distribution. J. Seismol. Res. 2019, 42, 366–371. 22. Liu, G.; Xiong, W.; Wang, Q.; Qiao, X.; Ding, K.; Li, X.; Yang, S. Source Characteristics of the 2017 Ms 7.0 Jiuzhaigou, China, Earthquake and Implications for Recent Seismicity in Eastern Tibet. J. Geophys. Res. Solid Earth 2019, 124, 4895–4915. [CrossRef] 23. Toda, S.; Stein, R.S.; Richards-Dinger, K.; Bozkurt, S.B. Forecasting the evolution of seismicity in southern California: Animations built on earthquake stress transfer. J. Geophys. Res. Space Phys. 2005, 110, 1–17. [CrossRef] 24. Wang, J.; Xu, C. Coseismic Coulomb stress changes associated with the 2017 MW6.5 Jiuzhaigou earthquake (China) and its impacts on surrounding major faults. Chin. J. Geophys. 2017, 60, 4398–4420. 25. Massonnet, D.; Feigl, K.L. Radar interferometry and its application to changes in the Earth’s surface. Rev. Geophys. 1998, 4, 441–500. [CrossRef] 26. Goldstein, R.M.; Werner, C.L. Radar interferogram filtering for geophysical applications. Geophys. Res. Lett. 1998, 25, 4035–4038. [CrossRef] 27. Rosen, P.A.; Hensley, S.; Zebker, H.A.; Webb, F.H.; Fielding, E.J. Surface deformation and coherence measurements of Kilauea Volcano, Hawaii, from SIR-C radar interferometry. J. Geophys. Res. Space Phys. 1996, 101, 23109–23125. [CrossRef] 28. Okada, Y. Surface deformation due to shear and tensile faults in a half-space. B Seismol. Soc. Am. 1985, 75, 1135–1154. 29. Guo, R.; Zheng, Y.; Diao, F.; Xu, J. Rupture model of the 2013 M W 6.6 Lushan (China) earthquake constrained by a new GPS data set and its effects on potential seismic hazard. Earthq. Sci. 2018, 31, 117–125. 30. Guo, R.; Zheng, Y.; Xu, J.; Jiang, Z. Seismic and Aseismic Fault Slip Associated with the 2017 M w 8.2 Chiapas, Mexico, Earthquake Sequence. Seismol. Res. Lett. 2019, 90, 1111–1120. [CrossRef] 31. Wang, R.; Diao, F.; Hoechner, A. SDM—A geodetic inversion code incorporating with layered crust structure and curved fault geometry. In Proceedings of the EGU General Assembly 2013, Vienna, Austria, 7–12 April 2013. 32. Wang, R.; Martín, F.L.; Roth, F. Computation of deformation induced by earthquakes in a multi-layered elastic crust—FORTRAN programs EDGRN/EDCMP. Comput. Geosci. UK 2003, 29, 195–207. [CrossRef] 33. Laske, G.; Masters, G.; Ma, Z.; Pasyanos, M.E. CRUST1.0: An updated global model of Earth’s crust. In Proceedings of the EGU General Assembly 2012, Vienna, Austria, 22–27 April 2012. 34. Wang, Y.; Zhao, T.; Li, C.X.; Liu, C. Relocations and focal mechanism solutions of the 2017 Jiuzhaigou, Sichuan MS 7.0 earthquake sequence. Process Geophys. 2019, 34, 469–478. 35. Yi, S.-J.; Wu, C.-H.; Li, Y.-S.; Huang, C. Source tectonic dynamics features of Jiuzhaigou Ms 7.0 earthquake in Sichuan Province, China. J. Mt. Sci. 2018, 15, 2266–2275. [CrossRef] 36. Schnaidt, S.; Heinson, G. Bootstrap resampling as a tool for uncertainty analysis in 2-D magnetotelluric inversion modelling. Geophys. J. Int. 2015, 203, 92–106. [CrossRef] 37. McLaughlin, K.L. Maximum-likelihood event magnitude estimation with bootstrapping for uncertainty estimation. B Seismol. Soc. Am. 1988, 78, 855–862. 38. Efron, B.; Tibshirani, R. Bootstrap Methods for Standard Errors, Confidence Intervals, and Other Measures of Statistical Accuracy. Stat. Sci. 1986, 1, 54–75. [CrossRef] 39. Fliss, S.; Bhat, H.S.; Dmowska, R.; Rice, J.R. Fault branching and rupture directivity. J. Geophys. Res. Space Phys. 2005, 110, 110. [CrossRef] 40. Oglesby, D.D.; Day, S.M.; Li, Y.; Vidale, J.E. The 1999 Hector Mine Earthquake: The Dynamics of a Branched Fault System. B. Seismol. Soc. Am. 2003, 93, 2459–2476. [CrossRef] 41. Li, Y.; Bürgmann, R.; Zhao, B. Evidence of Fault Immaturity from Shallow Slip Deficit and Lack of Postseismic Deformation of the 2017 Mw 6.5 Jiuzhaigou Earthquake. Bull. Seism. Soc. Am. 2020, 110, 154–165. [CrossRef] 42. Stein, R.S.; King, G.C.P.; Lin, J. Stress Triggering of the 1994 M = 6.7 Northridge, California, Earthquake by Its Predecessors. Science 1994, 265, 1432–1435. [CrossRef] 43. Guo, R.; Zheng, Y.; An, C.; Xu, J.; Jiang, Z.; Zhang, L.; Riaz, M.S.; Xie, J.; Dai, K.; Wen, Y. The 2018 Mw 7.9 Offshore Kodiak, Alaska, Earthquake: An Unusual Outer Rise Strike-Slip Earthquake. J. Geophys. Res. Solid Earth 2020, 125, e2019JB019267. 44. Wang, R.; Lorenzo-Martín, F.; Roth, F. PSGRN/PSCMP—a new code for calculating co- and post-seismic deformation, geoid and gravity changes based on the viscoelastic-gravitational dislocation theory. Comput. Geosci. 2006, 32, 527–541. [CrossRef] 45. Guo, R.; Zheng, Y.; Xu, J. Stress modulation of the seismic gap between the 2008 M s 8.0 Wenchuan earthquake and the 2013 M s 7.0 Lushan earthquake and implications for seismic hazard. Geophys. J. Int. 2020, 221, 2113–2125. [CrossRef] 46. Lin, X.; Chu, R.; Zeng, X. Rupture processes and Coulomb stress changes of the 2017 Mw 6.5 Jiuzhaigou and 2013 Mw 6.6 Lushan earthquakes. Earthplanets Space 2019, 71, 1–15. [CrossRef] 47. Shan, B.; Zheng, Y.; Liu, C.; Xie, Z.; Kong, J. Coseismic Coulomb failure stress changes caused by the 2017 M7.0 Jiuzhaigou earthquake, and its relationship with the 2008 Wenchuan earthquake. Sci. China Earth Sci. 2017, 60, 2181–2189. [CrossRef] 48. Helmstetter, A.; Shaw, B.E. Relation between stress heterogeneity and aftershock rate in the rate-and-state model. J. Geophys. Res. Space Phys. 2006, 111. [CrossRef]
You can also read
