S-wave splitting, also called S-wave birefringence, is a phenomenon that occurs when a transversely polarized shear wave enters some form of effective elastic anisotropic medium, and the incident shear wave splits into two almost orthogonally polarized seismic phases. These two split seismic phases have different propagation velocities and vibration directions. Theoretical studies show that the particle motion trail on a polarization diagram is no longer linear but elliptical. Proper rotation of the three-component seismic map can separate the split fast and slow transverse waves (Crampin S., 1978; Crampin S., 1981). Research also shows that under simple shear mechanism, peridot lattice preferred orientation (LPO) only depends on finite strain of the mantle, has nothing to do with the deformation process that produces strain, and the LPO is parallel to the principal tension axis of the strain ellipse, which usually represents the current direction of mantle deformation and flow (Ribe N.M., 1989; Ribe N.M., 1992). Therefore, the study of seismic anisotropy can impose constraints on mantle deformation and flow.
Hohhot, belong to the North China craton, a first-level tectonic unit, is located in the Daqingshan-Liangcheng epicontinental basin (Pan Guitang et al., 2009) and adjacent to active faults such as the Daqingshan piedmont fault, the northern Daihai fault and the Horinger fault. The analysis of anisotropy of the underground medium in Hohhot is of positive significance to understanding the nature of the underground medium, the change of stress field and seismic activity characteristics in this area. Seismic observation at Hohhot fiducial seismic station (HHC) and Horinger seismic station (HLG) has undergone the transformation of the ninth"Five-year Plan"and the tenth"Five-year Plan", and has accumulated many years' worth of digital observation data. Meanwhile, multiple mobile seismometers set up in Horinger in 2015-2016 following the"one institute and one school in three provinces"project have recorded multiple near earthquakes, which has also enriched the data for the study of near-shock shear wave splitting.1 DATA AND METHODS
In this paper, the study on the upper mantle anisotropy is carried out by mainly using the S-wave phases of teleseism recorded at Hohhot fiducial seismic station, and at the same time, the near-shock S-wave phases recorded at fixed and mobile seismic stations in Horinger, in the southern region of Hohhot, are used for the study of crustal anisotropy.1.1 Data and Methods Used for the Study of Teleseismic S-wave Splitting at Hohhot Seismic Station
MS≥6.0 distant earthquakes with epicentral distances in the scope of 90.3°-143.3° recorded in Hohhot station during 2008-2015 are selected, and SKS phases of these earthquake events are used in the study of S-wave splitting.
The Splitlab software (Wustefeld A. et al., 2008) is adopted to calculate S-wave splitting, using the rotation correlation method (hereafter referred to as the RC method) (Bowman J.R. et al., 1987) and minimum energy method (hereafter referred to as SC method) (Silver P.G. et al., 1991) at the same time for the calculation of S-wave splitting. The polarization analysis of SKS phase of a single teleseismic event is carried out using these two methods to acquire the anisotropy parameter pair (φ, Δt), of which, φ indicates fast wave polarization direction and Δt is the arrival time difference between fast and slow waves. Using two methods at the same time to calculate S-wave splitting can effectively determine weak anisotropy or invalid events (Null) (Wustefeld A. et al., 2007). The so-called invalid event refers to seismic records that do not produce splits, and the emergency of an invalid event may be associated with the isotropic medium under the seismic stations, and may also be due to the orientation of the event being parallel or perpendicular to the polarization direction of the fast wave.
According to the discriminant methods of splitting measurement proposed by Barruol et al. (1997) and Wustefeld et al. (2007), based on S-wave splitting parameters obtained from the SNR of seismic phase in the original seismic records, waveform before and after the correction of fast and slow wave, movement trail of particles and different measurement methods (RC method and SC method), we divide the splitting results into five categories, These are, a high-quality effective splitting result (good Non-Null), fair (fair Non-Null), unreliable (poor), high-quality invalid splitting result (good Null) and fair-quality invalid splitting result (fair Null). In this paper, we judge the results of S-wave splitting obtained from graphic features and different methods, so as to find effective high-quality splitting results (good Non-Null).
At present, it is generally believed that the results obtained using the minimum energy method (SC method) are more reliable and stable (Vecsey L. et al., 2008; Gao S.S. et al., 2009), therefore, although a combination of RC and SC methods is used, the measurement results obtained by the minimum energy method (SC method) will be used in the discussion and analysis of effective splitting results. Fig. 1 is an example of a high-quality effective splitting judged from calculation results of SKS phase by the use of teleseismic events recorded by station HHC.
The systematic analysis method of shear wave splitting, SAM method (Gao Yuan et al., 2004), is used in this study. SAM is a shear wave splitting analysis method proposed on the basis of correlation function analysis, which mainly includes three parts, namely calculation of the correlation function, time delay calibration and polarization analysis and testing, with the characteristics of self-examination. For ease of use, SAM method software system has been recently up-graded and updated (Gao Yuan et al., 2008).
The main task of shear wave splitting analysis is to isolate fast and slow shear waves from the ground surface records. Since the fast and slow shear waves come from the same source, after time delay correction of the fast and slow shear waves, these two waves are generally correlated. The calculation of correlation function is to first rotate the waves of two horizontal components of the record and obtain two new waves. Rotation angle A starts in the north, changing within the scope of 0°-180° in a clockwise direction, with a step length of 1°. The change range of time delay is from -0.1s to 0.1s, with a step length of 0.02s. Then, the correlation coefficient of two waves is calculated, and the rotation angle and time delay corresponding to the maximum correlation coefficient are the polarization orientation of the fast shear wave and time delay of the slow shear wave. Time delay correction is performed according to the calculation results of relevant functions. Finally, the polarization analysis and test of two waves before and after time delay correction is performed. According to the results of Gao Yuan et al. (1995), if the polarization graph after time delay correction is more linear, the reliability of the calculation results is higher. Otherwise, the shear wave splitting parameters need to be recalculated and the calculation results need to be retested.2 SHEAR WAVE SPLITTING RESULTS AND DISCUSSION 2.1 Discussion on the Results of Teleseismic S-wave Splitting at Hohhot Seismic Station
In this paper, SKS phase is used in the study of S-wave splitting for 118 teleseismic events with epicenter distances of 90.3°-143.3° recorded at Hohhot fiducial station (HHC) during 2008-2015. According to the above analysis and determination method of the results, 18 high-quality effective splitting results are obtained (good Non-Null), as shown in Table 1.
By analyzing the results, it is found that the polarization orientation of fast waves at Hohhot seismic station is mainly concentrated in -12°-6°, namely in the NW direction, as shown in Fig. 2(a).
The delay time according to the calculated results is mainly about 0.9s, which is consistent with the results obtained by previous research (Chang Lijun et al., 2011, 2012). Further research finds that the anisotropy parameters beneath Hohhot seismic station, especially the fast wave polarization orientation, are not randomly distributed, but concentrated mainly within the scope of two back azimuth areas, which are 26°-37° and 114°-126° respectively, which may be related to the distribution orientation of seismic events involved in the calculation, and in the meantime also reflects the complex anisotropic structure beneath the seismic station.
Because the SKS seismic phase is mainly used in this study, the anisotropic information obtained is mainly about anisotropy of the upper mantle and above. The anisotropy of the upper mantle is caused by the lattice-preferred orientation in mantle peridotite resulting from deformation, therefore shear wave splitting measurement results directly reflect the characteristics of deformation of the upper mantle and mantle flow field (Long M.D., 2008). Some scholars believe that anisotropy is caused mainly by current mantle flow (Silver P.G., 1996, Vinnik L.P. et al., 1992). In areas of stable tectonics, anisotropy is considered to be the"fossil"anisotropy left in the lithosphere by the last large-scale tectonic movements in these areas. In tectonically active areas, anisotropy reflects the ongoing tectonic movement (Silver P.G. et al., 1991). The principal axis of anisotropy is parallel to the boundary of plates in the vicinity of the aggregation of modern plates, parallel to the stretching direction of rifts in rife zones, parallel to the orientation of mountain ranges in orogenic belts, and is parallel to the fault zone strike in the vicinity of the large-scale strike-slip fault zone (Vinnik L.P. et al., 1992). In addition, the direction of absolute plate motion (APM) is consistent with the orientation of fast wave of anisotropy, indicating that the mantle flow under lithosphere plays an important role in anisotropy (Vinnik L.P. et al., 1992).
The fast wave polarization orientation of this study is consistent with the results obtained by previous researchers (Chang Lijun et al., 2011, Chang Lijun et al., 2012). By analyzing the results of fast wave polarization orientation in this study, it is found that the direction is not parallel to the direction of APM, nor the NE or NEE-striking of the regional surface structure. The anisotropy reflected in this study should be"fossil"anisotropy, which mainly exists in stable units on continental tectonics, and retains the historical information of mantle deformation.2.2 Discussion on the Results of Near-shock S-wave Splitting at Fixed and Mobile Seismic Stations in Horinger
A ML3.4 earthquake took place in Horinger of Inner Mongolia on October 29, 2015. Seismic activities in the area increased significantly after that. Seismic data for the S-wave window smaller than or equal to 45° is calculated in this study using SAM, and a total of 11 S waves are finally obtained at 1 fixed station, Horinger (HLG), and 2 mobile stations, Dahongcheng (L1503) and Heilaoyao (L1505). Categorized results are shown in the table below.
The calculation results of delay time show that the main variation range is 0.01-0.05s. The maximum value appears at HLG station, which is 0.05s, and the delay time at most of the stations is mainly concentrated in 0.01-0.04s. The change of delay time mainly reflects the intensity of anisotropy, so it can be seen from the results that the intensity of anisotropy in Hohhot is relatively stable, with high value only in HLG station area.
Calculation results show that the variation of polarization direction mainly ranges from 40°-50°, except for S-wave polarization orientation at Horinger station, which shows a change of 10°. Polarization orientation of fast waves at Dahongcheng and Heilaoyao seismic station is parallel to the orientation of the main fault zone in the region, namely the Horinger, Jiucaizhuang-Qianyaozi, northern Daihai and southern Daihai faults, basically consistent with the direction of the regional principal compressive stress, both in the NE direction, and in line with the background characteristics of tectonic stress field in the eastern margin of Ordos block, which indicates that the change of the state of stress field is not significant in Horinger in the south of Hohhot.
According to the results of this study, it is found that the SKS splitting in Hohhot area mainly reflects the anisotropy of the upper mantle, of which the polarization orientation of fast waves is mainly concentrated in the range of -12°-6°, that is, in the NW direction, and seismic events used for the achievement of fast wave polarization direction are not randomly distributed, concentrating mainly in the scope of back azimuth area within 26°-37° and 114°-126°, which may be related to the orientation distribution of seismic events involved in the calculation, and also reflect the complex anisotropic structure beneath the stations. The delay time is mainly about 0.9s. The results obtained are basically consistent with the results obtained by previous researchers, namely, that fast wave orientation is neither consistent with the direction of absolute plate motion, nor consistent with the strike of structures. The anisotropy reflected in the results is likely to be the manifestation of "fossil" anisotropy left in the thick lithosphere of the old craton.
The results of near-shock shear wave splitting in Hohhot area show that the variation of fast wave polarization direction mainly ranges from 40°-50°, except for S-wave polarization orientation at Horinger station, which shows a change of 10°. Polarization orientation of fast waves at Dahongcheng and Heilaoyao seismic stations is parallel to the orientation of the main fault zone in the region, namely the Horinger, Jiucaizhuang-Qianyaozi, northern Daihai and southern Daihai faults, basically consistent with the direction of regional principal compressive stress, both in the NE direction, which is in line with the background characteristics of the tectonic stress field in the eastern margin of the Ordos block, and also indicates that the tectonic stress field in the Hohhot region is mainly influenced by the anisotropy of the crust. The effect of the anisotropy of the upper mantle is not obvious on the stress field in the Hohhot area, and the stress field shows no evident changes, which is rather stable.This is also consistent with the relatively stable geological structure background in the northeast corner of Ordos block where Hohhot lies.
The study of teleseismic SKS splitting and near-shock S-wave splitting using data information from fixed and mobile seismic stations in Hohhot leads to a better understanding of research techniques and scope of application of the two methods, and in the process of S-wave splitting, especially in the use of SKS seismic phase for splitting study, we have gained a clear understanding of the data selection, which will be meaningful to the study of S-wave splitting using data from more seismic stations in a larger scope and areas.
|Barruol G., Silver P.G., Vauchez A. Seismic anisotropy in the eastern United States: deep structure of a complex continental plate[J]. Journal of Geophysical Research, 1997, 102(B4): 8329–8348. DOI:10.1029/96JB03800.|
|Bowman J.R., Ando M. Shear-wave splitting in the upper-mantle wedge above the Tonga subduction zone[J]. Geophysical Journal International, 1987, 88(1): 25–41. DOI:10.1111/j.1365-246X.1987.tb01367.x.|
|Chang Lijun, Wang Chunyong, Ding Zhifeng. Upper mantle anisotropy in the Ordos Block and its margins[J]. Science China Earth Sciences, 2011, 54(6): 888–900. DOI:10.1007/s11430-010-4137-2.|
|Chang Lijun, Wang Chunyong, Ding Zhifeng. Upper mantle anisotropy beneath North China[J]. Chinese Journal of Geophysics, 2012, 55(3): 886–895.|
|Crampin S. Seismic-wave propagation through a cracked solid: polarization as a possible dilatancy diagnostic[J]. Geophysical Journal International, 1978, 53(3): 467–496. DOI:10.1111/gji.1978.53.issue-3.|
|Crampin S. A review of wave motion in anisotropic and cracked elastic-media[J]. Wave Motion, 1981, 3(4): 343–391. DOI:10.1016/0165-2125(81)90026-3.|
|Gao S.S., Liu K.H. Significant seismic anisotropy beneath the southern Lhasa Terrane, Tibetan Plateau[J]. Geochemistry, Geophysics, Geosystems, 2009, 10(2): Q02008. DOI:10.1029/2008GC002227.|
|Gao Yuan, Zheng Sihua, Wang Peide. Shear wave splitting study on small earthquake swarm of 1992 in Dongfang of Hainan, south of China[J]. Chinese J.Geophys., 1996, 39(2): 221–232.|
|Gao Yuan, Liu Xiqiang, Liang Wei, Hao Ping. Systematic analysis method of shear-wave splitting: SAM software system[J]. Earthquake Research in China, 2004, 20(1): 101–107.|
|Gao Yuan, Shi Yutao, Liang Wei, Liu Xiqiang, Hao Ping. Systematic analysis method of shear-wave splitting SAM (2007): software system[J]. Earthquake Research in China, 2008, 24(4): 345–353.|
|Long M.D., Silver P.G.. The subduction zone flow field from seismic anisotropy: a global view[J]. Science, 2008, 319(5861): 315–318. DOI:10.1126/science.1150809.|
|Lv Qingtian, Ma Kaiyi, Jiang Mei, Hirn A., Nercessian A. Shear-wave anisotropy in the south of Qinghai-Tibet Plateau[J]. Acta Seismologica Sinica, 1996, 18(2): 215–223.|
|Pan Guitang, Xiao Qinghui, Lu Songnian, Den Jinfu, Feng Yimin, Zhang Kexin, Zhang Zhiyong, Wang Fangguo, Xing Guangfu, Hao Guojie, Feng Yanfang. Subdivision of tectonic units in China[J]. Geology in China, 2009, 36(1): 1–28.|
|Ribe N.M. Seismic anisotropy and mantle flow[J]. Journal of Geophysical Research, 1989, 94(B4): 4213–4223. DOI:10.1029/JB094iB04p04213.|
|Ribe N.M. On the relation between seismic anisotropy and finite strain[J]. Journal of Geophysical Research, 1992, 97(B6): 8737–8747. DOI:10.1029/92JB00551.|
|Silver P.G., Chan W.W. Shear wave splitting and subcontinental mantle deformation[J]. Journal of Geophysical Research, 1991, 96(B10): 16429–16454. DOI:10.1029/91JB00899.|
|Silver P.G. Seismic anisotropy beneath the continents: probing the depths of geology[J]. Annual Review of Earth and Planetary Sciences, 1996, 24: 385–432. DOI:10.1146/annurev.earth.24.1.385.|
|Vecsey L., Plomerová J., Babuška V. Shear-wave splitting measurements-Problems and solutions[J]. Tectonophysics, 2008, 462(1-4): 178–196. DOI:10.1016/j.tecto.2008.01.021.|
|Vinnik L.P., Makeyeva L.I., Milev A., Usenko A.Y. Global patterns of azimuthal anisotropy and deformations in the continental mantle[J]. Geophysical Journal International, 1992, 111(3): 433–447. DOI:10.1111/gji.1992.111.issue-3.|
|Wüstefeld A., Bokelmann G. Null detection in shear-wave splitting measurements[J]. Bulletin of the Seismological Society of America, 2007, 97(4): 1204–1211. DOI:10.1785/0120060190.|
|Wüstefeld A., Bokelmann G., Zaroli C., Barruol G. SplitLab: a shear-wave splitting environment in Matlab[J]. Computers & Geosciences, 2008, 34(5): 515–528.|