2 Xiamen Center for Seismic Survey, Earthquake Administration of Fujian Province, Xiamen 361021, China;
3 University of Chinese Academy of Sciences, Beijing 100049, China
Xisha block is one of the micro-continental massifs formed in the process of the formation and evolution of the South China Sea (SCS). It is located among the Indosinia Plate, the South China Plate and the SCS Plate (Fig. 1), with Xisha trough to the north, Northwestern Sub-basin to the east, Zhongsha trough and Zhongsha block to the southeast, Southwestern Sub-basin to the south, and Guangle uplift to the west. A lot of oil- and gas-bearing Cenozoic sedimentary basins, e.g. the Qiongdongnan basin, Pearl River Mouth basin, Zhongjiannan basin, Beibuwan basin and Yinggehai basin, were distributed around Xisha block (Fig. 1). These basins were formed by the action of large-scale extensional thinning in the northern margin of the SCS (Taylor and Hayes, 1983). The thick Cenozoic sediment stratum in these basins has favorable conditions for the formation and accumulation of petroleum (Wei, 2006). Thus, the deep structure information of Xisha block is critical to provide a significant guidance for the economic exploration and development for these hydrocarbon basins. At the same time, the deep crustal structure of Xisha block contains important information about the history of formation and evolution for the SCS. More importantly, it is the foundation to study the deep dynamic mechanism and the formation and evolution of marginal seas.
Since the 1980s, many deep seismic profiles have been investigated in the northern margin of the SCS (Fig. 2)(Yao et al., 1994; Nissen et al., 1995; Qiu et al., 2001; Wu et al., 2008; Huang et al., 2011; Ao et al., 2012). The seismic surveys have greatly enriched the understandings for structural features of Xisha block, but controversies still exist concerning the tectonic properties of Xisha block. The drilling results from Yongxing Island and Chenhang Island (Fig. 2) show that the oldest layer is granitic gneiss formed at 1465 Ma (Zhang, 1991). It was deduced that Xisha block is an old continental block (Proterozoic) based on petrology and chronology from drillings. Yao (1994) concluded that the Xisha trough acted as the paleo geo-suture zone during In- dosinian movement, and speculated that Xisha block de-parted from Indochina block as a micro-continent massif based on the seismic results of ESP-W completed in 1985(Fig. 2). Qiu et al.(2000) considered that Xisha trough is a Cenozoic rift based on the results of OBH1996-4 seismic profile (Fig. 2); the two sides of Xisha trough should belong to the same block before the rifting. It suggests that Xisha block and South China margin have same origin. What are the features for the deep crustal structure of Xisha block? Where did Xisha block de- part from? What is the role played by the Xisha block in the process of formation and evolution of the SCS? The deep seismic survey and research are very necessary to answer the above scientific questions.
Previous studies mainly focused on the deep structure around Xisha block or across Xisha block in N-S direction (Fig. 2), but there are no deep seismic survey lines which cross the Xisha block in NEE-SWW direction and parallel to the tectonic trend of the northern margin of the SCS. Therefore, this paper is focused on the OBS2013-3 survey line of 2013. The P-wave velocity structure beneath the OBS2013-3 profile across Xisha block will provide new geological and geophysical evidences for Xisha block tectonic attributes in the northern SCS continental margin.
2 DATA ACQUISITION AND PRELIMINARY PROCESSINGIn April-May 2013, a deep seismic survey line OBS2013-3 in Xisha area was completed which runs along the NEE-SWW direction from the east of Guangle uplift, throughout the Xisha area, almost connects to the OBS2006-2 line (Fig. 2). The total length of the line is 238 km. 15 Chinese broadband Ocean Bottom Seis- mometers (OBS) with three-component geophones and one hydrophone were deployed using the R/V Shiyan2 from the South China Sea Institute of Oceanology, Chines Academy of Science. The average interval between the OBS stations is 12 km. The OBS pool includes 2 types of big-ball and small-ball. The sampling rates for OBSs were set up at 125 Hz for the big-ball type and 250 Hz for the small-ball type during the survey respectively. The seismic source consisted of an array of four air-guns with a total volume of 6000 in3 towed at 10 m depth below the sea surface, whose working pressure was set up at 120 kPa. The ship speed varied from 4.5 to 5.5 knots, and the initial shooting time interval was kept at 120 s. Some shots were fired at 60 s interval and 200 m distant interval near OBS13 station because of air-gun trouble.
The preliminary processing of OBS data includes: shooting time and position correction, OBS recorder time drift correction, OBS position correction, etc. Firstly, we use the accurate shooting time recorded by the calculagraph to correct shooting time determined by Hypack navigation; and using Monte Carlo method to correct coordinates of shots and OBSs (Zhang et al., 2013) based on the shooting positions by using the Global Positioning System (GPS) of the ship; then calculate the offsets between shot and OBS, the coordinates of OBSs are acquired along OBS2013-3 profile by using of GPS data to determine the depth of sea bottom. The drift increment was calculated according to the beginning record time, ending record time and total drift of each OBS, then the time increment will be added to each trace of the reduced time profile with a linear method. Finally, the seismic data were formatted in the standard format of the Society of Exploration Geophysicist (SEGY) for each OBS. Most of OBS data are of high quality except OBS1 station after band-pass filtering. The seismic record sections for 14 OBSs stations (Figs. 3a-7a) are displayed with a reduction velocity of 6.0 km·s-1 which recorded the seismic signals down to the Moho interface. These data provide a strong foundation for the research of deep velocity structure.
Many kinds of seismic phases were recorded by OBSs along the OBS2013-3 profile (as seen in Table 1). Here, the seismic phase (Pdw) was defined as direct water waves; Ps is a refraction from the sedimentary layer; Pg is a refractive phase in the crust; PcP and Pc are reflective and refractive waves from the Conrad interface, respectively; and PmP is the reflection phase from the Moho interface. The seismic phases Pg and PmP are identified for most OBS record sections, and Pg can be traced as far as 150 km offset at the most (Fig. 4a). However, the seismic phases (Ps) are recognized only in a few OBS record sections since the sedimentary layer in Xisha Islands is relatively thin in thickness as it is a reef-island area (Zhang, 1991). No effective seismic phases are identified in OBS01 record section, which is located at the Guangle uplift, due to the poor quality of data. OBS02-OBS05 stations are located at the transition region between the Guangle uplift and the Xisha block (Fig. 2), where the water depth changes greatly from a relatively shallow depth in the Guangle uplift to a water depth of about 1200 m in the middle of Xisha block crossing the Qiongdongnan basin and Zhongjiannan basin. The main seismic phases identified in these four record sections included Pdw, Ps, Pg and PmP; Pg and PmP appeared clearly. Taking OBS02 station as an example (Fig. 3a); Pg was observed at the offset of 5∼30 km on both branches, which undulated greatly as was influenced by sedimentary basement; and PmP was traced as far as 110 km offset in the OBS02 record section.
OBS06-OBS08 stations were located at Xisha block with a water depth of 300∼700 m (Fig. 2). What partic- ularly worth mentioning is that PcP and Pc, reflection and refraction from the Conrad interface, could be rec- ognized at OBS06, 07 and 08 stations (Figs. 4, 5, 6, 8). Taking OBS06 station as an example (Fig. 4a), the phase of PcP and Pc could be clearly recognized besides Pdw, Ps, Pg and PmP; PcP with the reduction time ca. 2.6 s were traced at -50∼-20 km offset in the left branch and at 12∼22 km offset in the right branch, respectively. PcP was used to well constrain the Conrad interface shown in the velocity model and ray tracing results beneath OBS06 station (Fig. 4c).
In addition, OBS09-OBS15 stations were located in the eastern uplift region of Xisha block where reefs were distributed and water depth changed greatly ranging from 400 to 1100 m. Seismic phases Pdw, Pg and PmP were recorded by above OBSs. Taking OBS11 station as an example (Fig. 7a), Pg and PmP could be clearly traced to 110 km offset at most; the seismic phase Pg was used to constrain the crustal velocity structure; while PmP well controlled the depth and shape of the Moho interface.
3.2 Velocity Structure ModelThe initial crustal structure model beneath OBS2013-3 profile was firstly established according to the regional geological data and previous research results (Yao et al., 1994; Qiu et al., 2000; Yan and Liu, 2002; Ao et al., 2012; Huang et al., 2011). The characteristics of seismic phases in 14 OBSs’ seismic record sections were analyzed in detail along OBS2013-3 profile. Subsequently, two-dimensional ray tracing and theoretical calculations were carried out employing the Rayinvr software (Zelt and Smith, 1992). The trial-and-error method was used to repeatedly modify the velocity model, and theoretical travel-time curve of each seismic phase was calculated (Figs. 3-9). When theoretical travel-times calculated were gradually closing to all the observed travel-times, and the root mean square (RMS) and travel-time residuals of all seismic phases were reaching minimum (Table 1), the final velocity structure model (Fig. 10a) was obtained at last. The forward simulation calculation was conducted following the procedure from single OBS station to multiple stations, from the upper to the lower of the model, from simplicity to complexity (Qiu et al., 2011) during the forward and inversion analyses. The final velocity model (Fig. 10a) could be divided into five layers: the top layer is seawater with a velocity of 1.5 km·s-1; the second layer is sedimentary with a velocity of 2.2∼3.2 km·s-1 from top to bottom. The thickness for the sedimentary layer varies largely from 0.8 to 3 km in the Xisha block. The sedimentary layer fluctuates greatly in thickness in the transition between Guangle uplift and Xisha block, where extrusive basalts may exist beneath OBS03 and OBS05 stations according to the geological data in surrounding areas. In the Guangle uplift region, the thickness of the sedimentary layer is relatively thin, about 0.3∼2.0 km.
The third and fourth layers are upper crust with a velocity of 5.0∼6.4 km·s-1 and lower crust with a velocity of 6.5∼6.9 km·s-1, respectively. The crustal velocities beneath OBS03, 04, 05 and 06 stations are relatively higher than that in surrounding area, with velocities of 5.5∼6.5 km·s-1 in the upper crust and 6.6∼6.9 km·s-1 in the lower crust. The fifth layer is the upper mantle with a velocity of 8.0 km·s-1. The structure of the upper mantle was not constrained effectively due to lack of Pn seismic phase recorded by OBS stations along the OBS2013-3 profile (Fig. 3-9). The sea bottom interface of this model was set mainly by using gravity data and water depth, and coincided with the direct water-waves with near offset in each OBS station (Fig. 9). Therefore, the sea bottom interface was treated as a known condition and remained unchanged during the simulation. Moreover, the sedimentary basement, Conrad interface and Moho interface were set up dominantly referring to the geological and geophysical data in this region, and their depths and shapes were adjusted largely during the simulation primarily by seismic phases. As seen in the modeling and ray-tracing map of all seismic phases of all stations (Fig. 9), theoretical travel-times were fit well with observed travel-times. The residual mean square (RMS) reached a minimum (79 ms), and a Chi-square value of 1.489(Table 2) in the final model. These parameters suggest that the velocity structure is highly reliable.
Logging data of Xiyong 1# well demonstrate that the crystalline basement of Xisha block is a type of ancient metamorphic rock formed at pre-Cambrian age (Zhang, 1991), which suggests that Xisha block is a continental block with continental crust as same as South China block. The deep seismic surveys on land (Yin et al., 1999) and onshore-offshore seismic results (Huang et al., 2011) present that the crustal thickness of South China block is about 32 km (Table 2). Assuming that the crust of the continental margin of South China is a normal continental crust, Xisha block is featured by a thinned continental crust to different degrees under the extension and rifting of the northern SCS (Ruan et al., 2006; Qiu et al., 2006). There are only two deep seismic survey lines, OBS2013-3 and 0BS2006-2(Fig. 2), which are parallel to the tectonic trend in the northern margin of the SCS. A joint crustal section from Guangle uplift to Northwest Sub-basin is obtained by combining these two survey lines (Fig. 11); and the crustal structures in the overlapping part are relatively consistent with each other in crustal thickness, velocity, and the depths of each interface. According to the characteristics of the velocity structure and regional gravity and magnetic data, this crustal structure was divided into four parts corresponding to four geological tectonic units: Guangle uplift, Xisha block, the transitional belt, and the Northwest Sub-basin (Ao et al., 2012). Xisha block in this joint section extends to 230 km length along the NE-SW direction, where the crustal thickness is 23 km (Fig. 11). Additionally, a series of normal faults developed well in the transitional zone, which indicates obviously extensional tectonic environment in this region (Ding et al., 2009), and where the crustal thickness is continuously thinning to 12 km (Fig. 11) but still belongs to a type of continental crust (Ao et al., 2012).
Yongxing Island is the biggest island in Xisha block. Shidao mobile seismic station was deployed in Yongxing Island, the positions of OBS14 station along the line OBS2013-3 and Shidao seismic station are almost the same. The crust thickness beneath OBS14 station along the OBS2013-3 model is about 20 km and Moho interface lies at the depth of 22 km (Fig. 10). Compared with the crust thicknesses (26.5 km and 28 km, respectively) obtained by Ruan (2006) and Qiu (2006) using the receiver function method respectively, the crust thickness beneath OBS14 station is thinner by 4∼6 km. This difference in crustal thicknesses was caused by different methods. The velocity structures acquired by receiver function method are in fact the shear-wave (S-wave) velocity structures (Ruan et al., 2006; Qiu et al., 2006). And S-wave is more sensitive to partial melting of mantle than P-wave. That is to say, Moho depth detected by P-wave is shallower than that detected by S-wave, which means that partial melting may exist at the lower crust or the top of upper mantle in Xisha block.
The pseudo three-dimensional structure model across Xisha block (Fig. 12) is acquired by combining the crustal structure along the onshore-offshore seismic survey line OBS2011, which crosses Xisha block in NW-SE direction, with the crustal structure (Fig. 11) from Guangle uplift to Northwest Sub-basin in NE-SW direction. Chenhang mobile seismic station located at the intersection of two combined profiles (OBS2011, OBS2013-3 and OBS2006-2) is mostly equivalent to the location of OBS8 along the line OBS2013-3. The crustal thickness is 24 km and Moho interface lies at the depth of 26 km beneath OBS08(Fig. 10a) which is 80 km away in SW direction from Shidao seismic station (Fig. 2). On the other hand, the crust thickness beneath Chenhang seismic station is 26∼28 km obtained by the combination of earthquake receiver function method and the OBS wide-angle reflection/refraction method (Huang et al., 2011). The results of crustal structures are similar to each other between Shidao and Chenhang seismic stations. The Moho depth constrained by P-wave is shallower than that constrained by S-wave. It shows again that partial melting may exist in the lower crust or at the top of upper mantle in Xisha block. The length of Xisha block in NE-SW direction is 230 km as calculated by the crustal section from Guangle uplift to Northwest Sub-basin (Fig. 11); while the extent of Xisha block in NW-SE direction is 175 km measured by the crustal section of OBS2011 line (Fig. 12). Thus, the volume of the Xisha block is estimated roughly to be 9.2×105 km3 assuming an average crustal thickness of 23 km. The crust of the Xisha block is a thinned continental crust on the whole.
The velocity model of OBS2013-3 survey line (Fig. 10a) did not only provide the detailed velocity informa-tion of Xisha block, but also determine accurately the boundary between the upper and lower crust and their thickness ratio by picking up reflective phase (PcP) and refractive phase (Pc) from the Conrad discontinuity (Fig. 8). The average crustal thickness of Xisha block (23 km) is thinner than the normal crustal thickness of the South China block (Table 2). Combined with the drilling results of Xiyong No.1# well (Zhang, 1991), it was concluded that Xisha block is a paleo-continental block with a thinned crust. However, whether Xisha block was dismembered from the Indochina block or from the South China block is still an unsolved scientific question and one of the focuses of this survey and research.
The basement in Xisha block was pre-Cambrian metamorphic rocks based on petrology and chronology from drilling of Xiyong No.1#, and Kunsong block in the middle of Vietnam had similarly the metamorphic basement. Thus, Yao et al.(1994) deduced that Xisha block was disintegrated from the Indochina block in Proterozoic era. Moreover, the latest research shows that Indochina-Sunda sub-plate at the west of SCS, which extends southeastwards from the Indochina Peninsula to western Kalimantan Island, is a pre-Cambrian continental block (ca. 2500 Ma). That is to say, the Kunsong block in the central South Vietnam is cut by the N-Shuge fault belt at the western margin of the SCS (namely the East Vietnam fault); the Paleozoic and Mesozoic strata of the Kunsong block were distributed to the eastern margin around itself and not in the SCS area (Xia et al., 2014). In addition, the depths of Moho interfaces change quite dramatically from Indochina block to Xisha block based on the Moho depth map (Xia et al., 2014). It is inconsistent to the characteristics of the crustal structure formed by extension and rifting. Therefore, our research does not support this view.
OBS973-1 survey line is located in the southern continental margin of the SCS, its crustal structure across the central Nansha block (Qiu et al., 2011) provides important information for comparison of the southern and northern continental margins of the SCS. As seen in the one-dimensional models of Xisha block and Nansha block (Fig. 10b), both crustal structures are quite similar with the thinned continental crust (ca.25 km in thickness), crustal velocity of about 5∼6.9 km·s-1, and the ratio (nearly 1:2) of the upper to the lower crustal thickness. Therefore it was deduced that Xisha block and Nansha block had a conjugate relation with the fossil spreading ridge of the Southwest Sub-basin acting as the axis center (Fig. 1). Xisha block was departed from South China block. Xisha block, Zhongsha block and Nansha blocks, all of them were a part of the paleo South China continent in Early Cretaceous era. Since the extension of the continental margin and the seafloor spreading of the SCS during the Cenozoic, they gradually drifted to their own current positions.
The above contrasts of simple crustal structures (only velocity and thickness), of course, are insufficient to come to a conclusion. A joint P-wave and S-wave comprehensive inversion will be carried out for OBS2013-3 and OBS2009-1 located in Nansha block in the future research. Two-dimensional S-wave velocity structures will be acquired by identifying and picking up the converted S-wave travel times (Zhao et al., 2010); then petrological properties will be further gained to constrain their tectonic properties of Xisha block and the Nansha block.
5 CONCLUSIONSThe crustal velocity structure along OBS2013-3 line is obtained by ray-tracing and simulation of 14 OBS data. Based on comparison of crustal structures and analysis of tectonic setting around the study area, two main points are concluded as follows:
(1) The sedimentary thicknesses of Guangle uplift unit (0∼30 km in the model) and of Xisha block (70∼240 km in the model) are about 0.5∼2.7 km and 0.8∼3.0 km, respectively. While in the transition region (30∼70 km in the model), the thickness of sediments changes obviously due to the sedimentary basement variation, the velocity of sediment and upper crust is higher than other 2 parts (Fig. 10). The reflected seismic phases (PcP) from the Conrad interface were identified in three seismic record sections of OBS06, 07 and 08 stations (Fig. 8). These seismic phases provided a good constraint for the boundary between the upper and lower crust in Xisha block. The ratios of thickness of upper crust and lower crust were calculated close to 1:2 for Xisha block and 1:1 for Guangle uplift. The crust thickness of Xisha block keeps relatively stable with the average thickness of about 23 km, Moho interface at the depth of 23∼27 km and the velocity of the upper mantle is 8.0 km·s-1 (Fig. 10a).
(2) Xisha block is a paleo block with a thinned continental crust deduced by contrast and analysis on crust velocity structures of the surrounding areas. It has a size of ca. 9.2×105 km3. Xisha block, Zhongsha block and Nansha block belonged originally to the South China block in the early Cretaceous. The velocity structures of Xisha block and Nansha block are very similar. Xisha block and Nansha block are conjugate with each other by the fossil spreading ridge of the Northwest Sub-basin.
ACKNOWLEDGMENTSWe are grateful to the crew of the R/V Shiyan 2 and scientists who took part in the scientific cruise. Insightful comments by Sun J L and Xia S H are kindly acknowledged. The GMT software (Wessel and Smith, 1995) was used in this study. This work is supported by the National Natural Science Foundation of China (91428204, 41306046, 91028002, 41249908, 41176053) and Public Cruise Program of NSFC (contract NORC- 2013-08).
[1] | Ao W, Zhao M H, Qiu X L, et al. 2012. Crustal structure of the Northwest Sub-basin of the South China Sea and its tectonic implication[J]. Earth Science-Journal of China University of Geosciences , 37 (4): 779–790. |
[2] | Cai Z R, Xia B, Wan Z F, et al. . Relationship on the movement characteristics of the main fault and Red river fault in western South China Sea at late Oligocene[J]. Geotectonica et Metallogenia, 32 (4): 436–440. |
[3] | Ding W W, Li M B, Zhao L H, et al. 2009. Cenozoic tectono-sedimentary characteristics and extension model of the Northwest Sub-basin, South China Sea[J]. Earth Science Frontiers , 16 (4): 147–156. |
[4] | Huang H B, Qiu X L, Xu H L, et al. 2011. Preliminary results of the earthquake observation and the onshore-offshore seismic experiments on Xisha Block[J]. Chinese J. Geophys. , 54 (12): 3161–3170. DOI: 10.3969/j.issn.0001-5733.2011.12.016 |
[5] | Nissen S S, Hayes D E, Buhl P, et al. 1995. Deep penetration seismic soundings across the northern margin of the South China Sea[J]. Journal of Geophysical Research, 100 (B11): 22407–22433. |
[6] | Qiu X L, Zhou D, Xia K Y, et al. 2000. OBH experiment and crustal structure of Xisha trough, South China Sea[J]. Tropic Oceanology , 19 (2): 9–18. |
[7] | Qiu X L, Ye S Y, Wu S M, et al. 2001. Crustal structure across the Xisha Trough, northwestern South China Sea[J]. Tectonophysics, 341 (1-4): 179–193. |
[8] | Qiu X L, Zeng G P, Xu Y, et al. 2006. The crustal structure beneath the Shidao Station on Xisha Islands of South China Sea[J]. Chinese J. Geophys. , 49 (6): 1720–1729. |
[9] | Qiu X L, Zhao M H, Ao W, et al. 2011. OBS survey and crustal structure of the SW Sub-basin and Nansha Block, South China Sea[J]. Chinese J. Geophys. , 54 (12): 3117–3128. DOI: 10.3969/j.issn.0001-5733.2011.12.012 |
[10] | Ruan A G, Li J B, Hao T Y, et al. 2006. Inversion of far-field seismic waves at Shidao Seismographic Station[J]. Acta Oceanologica Sinica , 28 (2): 85–92. |
[11] | |
[12] | |
[13] | Wessel P, Smith W H F. 1995. New version of the generic mapping tools released[J]. EOS Trans. AGU, 76 (33): 329. |
[14] | Wu Z L, Ruan A G, Li J B, et al. 2008. New progress of deep crust sounding in the mid-northern South China Sea using Ocean Bottom Seismometers[J]. South China Journal of Seismology , 28 (1): 21–28. |
[15] | Xia K Y, Xia Z W, Zhao M H, et al. 2014. The geological basis of the China historical sea lines in South China Sea[J]. Acta Oceanologica Sinica , 36 (5): 77–89. |
[16] | Yan P, Liu H L. 2002. Analysis on deep crust sounding results in northern margin of South China Sea[J]. Journal of Tropical Oceanography , 21 (2): 1–12. |
[17] | Sun Z, Liu H L. 2011. Preface:Conjugate continental margins of the South China Sea-natural laboratory of sedimentary process[J]. Earth Science-Journal of China University of Geosciences , 36 (5): 785–786. |
[18] | Yao B C, Zeng W J, Chen Y Z, et al. 1994. Xisha trough of South China SeaAn ancient suture[J]. Marine Geology & Quaternary Geology , 14 (1): 1–10. |
[19] | Yin Z X, Lai M H, Xiong S B, et al. 1999. Crustal structure and velocity distribution from deep seismic sounding along the profile of Lianxian-Boluo-Gangkou in South China[J]. Chinese J. Geophys. , 42 (3): 383–392. |
[20] | Zelt C A, Smith R B. 1992. Seismic traveltime inversion for 2-D crustal velocity structure[J]. Geophysical Journal International, 108 (1): 16–34. |
[21] | Zhang L, Zhao M H, Wang J, et al. 2013. Correction of OBS position and recent advances of 3D seismic exploration in the central Sub-basin of South China Sea[J]. Earth Science-Journal of China University of Geosciences , 38 (1): 33–42. |
[22] | |
[23] | Zhao M H, Qiu X L, Ye C M, et al. 2004. Analysis on deep crustal structure along the onshore to offshore seismic profile across the Binhai (littoral) Fault Zone in northeastern South China Sea[J]. Chinese J. Geophys. (in Chinese). 2004, 47 (5): 845–852. |
[24] | Zhao M H, Qiu X L, Sha S H, et al. 2010. Seismic structure in the northeastern South China Sea:S-wave velocity and Vp/Vs ratios derived from three-component OBS data[J]. Tectonophysics, 480 (1-4): 183–197. |