CHINESE JOURNAL OF GEOPHYSICS  2013, Vol. 56 Issue (4): 400-410   PDF    
THREE-DIMENSIONAL P-WAVE VELOCITY STRUCTURE AROUND THE XIAOJIANG FAULT ZONE AND ITS TECTONIC IMPLICATIONS
WU Jian-Ping, YANG Ting, WANG Wei-Lai, MING Yue-Hong, ZHANG Tian-Zhong    
Institute of Geophysics, China Earthquake Administration, Beijing 100081, China
Abstract: As the southeastern margin of the Tibetan Plateau, the Xiaojiang fault zone plays an important role in the lateral escape of Tibetan Plateau materials. In this work, the 3-D P-wave velocity structure of the Xiaojiang fault zone and its surrounding areas was imaged by using travel time data from permanent seismic stations and a dense temporal seismic array. The results show that in the upper and middle crust, the Xiaojiang fault zone is mainly of low velocity, while high velocity anomalies exist to its east. In the middle and lower crust, a low velocity anomaly is shown in the central portion of the Xiaojiang fault zone, and high velocity anomalies exist both in the northern and southern areas. The northern high velocity anomaly near Panzhihua extends from lower crust to the surface, and the southern one in the lower crust extends to the upper mantle. We consider that the low velocity anomaly in the central portion of the Xiaojiang fault zone is caused by high temperature derived from upper mantle. The high anomalies in the northern area are most likely related to the late Paleozoic mantle plume activity which caused a large number of basic and ultrabasic mantle materials intruding into the crust. These high-velocity anomalies, with greater mechanical strength, play a certain impediment to southward escaping of Tibetan Plateau materials, and lead to the rapid uplift of the northern sub-block of the Sichuan-Yunnan active block. The north-dipping high velocity anomaly in the southern area is a further barrier for southward extrusion of the Sichuan-Yunnan active block, leading to strong deformation and intense seismicity within the low-velocity area in the overlaying upper and middle crust. Meantime, this high-velocity block contributes to stabilization of the upper mantle, which makes mantle materials of its southern region less affected by the southward escape of Tibetan Plateau materials.
Key words: Seismic tomography    P-wave velocity    Xiaojiang fault zone    Southeastern margin of Tibetan Plateau    
1 INTRODUCTION

The India-Eurasia collision is the most spectacular geologic event in the Cenozoic era on the Earth, resulting in rapid uplift of the Tibetan Plateau and surrounding orogenic belts, large-scale strike-slip faulting, and lateral escape of plateau materials. The Xianshuihe-Xiaojiang fault zone is a newly formed intra-continental strike-slip fault system, later than the southeastward extrusion of the Indochina block and the formation of the Red River fault[1, 2, 3] after the collision. The Xiaojiang fault zone is a famous seismic belt in China mainland, which consists of four N-S trending faults with width about 200 km. The four faults from west to east are the Lüzhijiang, Yimen, Puduhe, west and east Xiaojiang faults. In the past 300 years, five earthquakes of M7 or greater occurred within the belt, including one event of M8. As the southeastern margin of the Tibetan Plateau, the Xiaojiang fault zone plays an important role in the lateral escape of Tibetan Plateau materials[1, 2, 3, 4, 5]. Researches on the deep structure of the Tibetan Plateau provide insight into the evolution of the continental collision zone.

Geophysical investigations on the Xiaojiang fault zone have been highly concerned by researchers. Since the 1980s, several seismic sounding profiles were carried out through or partly through the Xiaojiang fault zone[6, 7, 8], which provide the important constraints on the crust structure in this region. However, it is still difficult to obtain three-dimensional crustal structure in this region due to insufficient number of the seismic sounding profiles. Since the 1990s, various geophysical methods such as the travel time tomography, surface wave tomography, and the receiver function inversion have been used in the Yunnan-Sichuan region[9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21], and achieved many meaningful results. However, because of the sparse seismic stations used in their study, the resolution of the images in crust and upper mantle within and around the Xiaojiang fault zone is still low.

Recently, 50 broadband seismic stations have been deployed around the Xiaojiang fault zone by our research group. Combined with permanent stations, the average station interval is about 30 km. In this work, we adopt travel time tomography technique to reveal 3-D P-wave velocity structure around the Xiaojang fault zone by using the regional seismic data recorded by the dense broadband seismic array in two years.

2 DATA AND METHOD

We deployed 50 broadband seismic stations around the Xiaojiang fault zone from Dec. 2008 to Dec. 2010. In this study the travel time data is mainly collected from these temporary seismic stations. Meanwhile, the travel time data from 73 permanent seismic stations in the same time period around the Xiaojiang fault zone is also used. Fig.1 shows the distribution of the temporary and permanent seismic stations. We carefully analyzed the waveform data from the temporary seismic stations, picking the arrival times and locating each earthquake if the event can be recorded by several stations. Afterwards, the observed travel time data from the permanent and temporary seismic stations are merged into one uniformed data for the same earthquake. The number of seismic phases used here is over 8 for each event, and the maximum travel time residual is 3.0 s. Finally, we picked data of 1425 earthquakes, including 11366 P-wave and 7235 S-wave data.

Fig.1 Distribution of seismic stations and epicenters used in this study  Red dots represent epicenters. Permanent and temporary seismic stations are denoted by light blue and dark blue triangles respectively. F1-Red river fault, F2-Xiaojiang fault, F3-Xiaojinhe fault. I-North sub-block of Sichuan-Yunnan active block, II-South sub-block of Sichuan-Yunnan active block.

We used the tomography method of Zhao et al.[22, 23] to determine the 3-D P-wave velocity structure around the Xiaojiang fault zone and surrounding areas. This method is capable of considering complex velocity discontinuities in crust and upper mantle and can computer theoretical travel times in a 3-D medium fast and accurately. A 3-D grid net is set up in the model to express the 3-D structure. One merit of this method is that the grid space can be adjusted according to the density of epicenters and stations and thus can be set to non-equidistant. This guarantees every node having sufficient ray-path number.

The study region is 23° -27.5°N, 101.5° -104.5°E. The lateral distance between grid nodes is 0.25°. The initial velocity model contains 8 layers, of which the first layer is at 1 km depth, the second layer is at 10 km depth and other layers increase with a 10 km step. The Moho and Conrad discontinuities are obtained from the receiver function and deep seismic sounding studies (Fig.2). The initial 1-D velocity model is based on the deep seismic sounding profiles and primary tomographic inversion results, and adjusted according to the average velocity at different depths. P- and S-wave velocities are 5.90 km/s and 3.41 km/s above the Conrad discontinuity, 6.55 km/s and 3.78 km/s between the Conrad and Moho discontinuities, respectively. Velocities below the Moho discontinuity are referred to that of the PREM.

Fig.2 Distribution of crustal thickness used for tomography inversion

We relocate the earthquakes in the inversion of velocity structure. In order to improve the accuracy of the earthquake location, inversions of P- and S-wave velocity structure are simultaneously conducted, and the theoretical travel times are calculated based on the 3-D P- and S-wave velocity models. Earthquake location and the velocity structure inversion are conducted alternately. We first relocate earthquakes according to the initial 3-D velocity model, and then invert the velocity structure based on the ray tracing according to new earthquake location results. The iteration is repeated until the stop condition is reached. The RMs P-wave travel time residual is 0.29 s after 10 iterations.

3 RESULTS

Figure 3(a, b) shows the P-wave velocity structure at depths of 1 km and 10 km. Velocity anomalies are mainly presented as strip shapes, and the long axes of the anomalies are in the N-S direction. At 1 km depth, a low P-wave velocity zone exists between the Puduhe and the east Xiaojiang faults. In the east to the Xiaojiang fault, except vicinity of the Dongchuan area, high- velocity anomalies are dominant, especially in the central and southern sections of the Xiaojiang fault. The Xiaojiang fault becomes the dividing line between high and low velocity areas from east to west, indicating that the Xiaojiang fault is obviously controlled by the shallow crustal structure. In the region from the Zemuhe fault to the Miyi-Huili-Huize area mainly shows weak high velocity. The Panzhihua and its vicinity exhibit obviously high velocity anomalies. Velocity structure at 10 km depth is similar to that at 1 km depth, but the small-scale high velocity anomalies at 1 km depth near the Mile and Wuding are obviously weaker at 10 km depth, indicating local shallow anomalies. A distinct low velocity anomaly is present beneath the Wuding-Kunming area. Strong earthquakes with Ms > 6 mainly occurred on the boundary between the high and low velocity anomalies. At depth of 20 km (Fig.3c), the long axes of the high- and low-velocity anomaly bodies are still mainly in the N-S direction. The low velocity in the Wuding-Kunming area at 10 km depth extends to the Lüzhijiang. The eastern edge of the Xiaojiang fault zone is located on the boundary between several high- and low-velocity anomalies. At depth of 30 km (Fig.3d), the shape of some anomalies changes obviously. The long axes of the velocity anomaly bodies south of 26°N change into the E-W direction. The high-velocity anomaly near Panzhihua becomes more obvious and extends further to the south. A distinct high velocity anomaly appears in the southwest of the Zemuhe and Xiaojiang faults near Qiaojia.

Fig.3 Map showing P-wave velocity deviation at different depth  Solid circle represents epicenter of historic earthquakes large than Ms 6.0. F1-Red river fault, F2-Shiping-Jianshui fault, F3-Chuxiong-Qujiang fault, F4-Xiaojiang fault, F5-Puduhe fault, F6-Yimen fault, F7-Luzhijiang fault.

Below depth of 40 km (Fig.3e, 3f, 3g), the long axis of velocity anomalies is obviously along the E-W direction near 25.5°N. Meanwhile, an obvious high velocity anomaly appears in the north of the Red river fault with its edge extending to the north as the depth increasing. At the depth of 60 km, low-velocity anomalies mainly appear to the north of 25.5°N, and high velocity anomalies to its south. Due to the restriction of seismic stations, the ray coverage is fairly sparse in the south of the Shiping-Jianshui fault below 70 km depth, and the reliable results can not be obtained there.

We once used the checkerboard test to evaluate the resolution of the 3-D P-wave velocity model, and found that in most areas the lateral resolution is higher than 0.25°, but in some regions the resolution is relatively low. In order to keep the high resolution in most areas and at the boundary of large anomaly bodies, we still use the grid of 0.25° to divide the model space, and use the recovery test to evaluate the model resolution. The basic principle of this method is as follows. Firstly, we calculate the theoretic travel time based on the inverted 3-D velocity model. Secondly, the theoretic travel time is taken as the observed data which is applied to the inversion. Finally, we compare the new inversion model with the original model to evaluate the reliability of the imaging results. The damping and smoothing parameters used in the model recovery test are the same with those in the inversion of the real observational data.

Figure4 shows the results of the model recovery test. Fig.4(a, b) is comparison of the velocity perturbations at 10 km depth, and Fig.4(c, d) shows a comparison of velocity profiles along the N-S direction. Fig.4(a, c) is derived from the inversion of the observed data, while Fig.4(b, d) is from the recovery test. The results show that most anomalies are recovered well with slightly decreased amplitudes and smoother shapes, which are caused by the damping and smoothing in the inversion. The model recovery test shows that the inversion results in this study are quite reliable.

Fig.4 Map showing recovery test results of three-dimensional velocity model  (a) and (b) are velocity structures at depth of 10 km obtained from observed data and recovery test respectively. (c) and (d) show velocity profiles along 102.75°E inverted from observed data and recovery test respectively. The dash line is the contour of the ray density of 50. In the region above the contour, the ray density is larger than 50.
4 DISCUSSION

Near Panzhihua, there exist high-velocity anomalies with different sizes at different depths in the crust. The N-S striking high velocity anomalies at the depth of 1km (Fig.3a) and 10 km (Fig.3b) are mainly distributed in the west of the Lüzhijiang fault. According to the P-wave tomographic results[14] of Sichuan and Yunnan, these high-velocity anomalies can extend further to the north to the Xiaojinhe fault. In the depth range 20~40 km (Fig.3(d-e)), the positive anomaly in the Miyi-Huidong-Qiaojia area merges with the anomaly near Panzhihua with its edge extending to the south as the depth increases. Geological investigations show that large-scale flood basalt eruptions occurred in the late Paleozoic (256~259Ma) in Yunnan, Sichuan and Guizhou provinces[24]. Recent studies suggest the eruptions may be related to mantle plume activity[25, 26, 27, 28, 29, 30, 31]. The Panzhihua region is located in the core of the crust-scale dome caused by mantle plume before the eruption of the Emeishan basalt[29]. The plume activity caused the intrusion of basic and ultrabasic mantle materials in the crust. There is a series of the mafic and ultramafic intrusions near the Mopanshan-Lüzhijiang, Panzhihua-Chuxiong, and the Anninghe-Yimen faults[24, 32]. Zhang et al.[24] considered that the mantle detachment, basement and the deeper structure stretching are the main causes for the formation of such kind of rock mass, but it is impossible to cause large-scale upwelling of the magma at the early stage of the extensional tectonic environment, and suggested there should exist larger rock mass in the deep crust. The seismic sounding profile along the E-W direction in southern Panzhihua reveals a relatively higher velocity anomaly in the upper crust and a locally high-velocity anomaly in the lower crust beneath the Panzhihua area and its vicinity[7]. However, seismic sounding only reveals the velocity structure along profile, and is unable to determine three dimensional features. Tomographic results in this study suggest that the size of this high-velocity anomaly in the middle to lower crust is larger than that reported in previous investigations. Studies on the seismic velocity of the crust rocks from the laboratory and seismic soundings indicate that the mafic and ultramafic rocks usually show high P-wave velocity[33, 34]. We infer that the high-velocity anomaly in the Panzhihua region is caused by the basic and ultrabasic rock mass, which was produced by the intrusion of mantle material in the process of crust doming at the early stage of the mantle plume activity.

The Indian-Eurasia collision caused rapid uplift of the Tibetan Plateau and eastward escape of plateau material. The Sichuan-Yunnan active block is one of the fastest escaping blocks in the Tibetan Plateau. This block is rhombic, confined by the Xianshuihe-Anninghe- Zemuhe-Xiaojiang fault and the Jinshajiang-Honghe fault. According to the fault slip rate and paleomagnetic investigation results[35], the Sichuan-Yunnan block can be further divided into the northwestern Sichuan sub-block in the north and the central Yunnan sub-block in the south, which is separated by the Lijiang-Xiaojinhe fault. The elevation changes dramatically from north to south across the fault. To the north of the fault, there are many mountains with elevation of 4500~6500 m, and the average elevation is over 3500 m. To the south of the fault, the average elevation drops to 2000 m. Geological investigations reveal obvious differential movement on two sides of the fault. The horizontal slip rate of the northwestern Sichuan sub-block is 2 mm/a higher than that of the central Yunnan sub-block, and the mean differential vertical movement reaches 1.0~1.3 mm/a[35, 36]. However, what causes the movement differences between these two sub-blocks? Few studies on this subject have been done so far. Tomographic results in this study suggest that there is a high-velocity anomaly centered on the Panzhihua vicinity in the crust, which extends to the Lijiang-Xiaojinhe fault in the north, reaches about the latitude of 26°N in the south, and partly expands to the Nanhua-Chuxiong fault in the west of the study region. We speculate that this high-velocity anomaly is related to the basic and ultrabasic rock mass produced by the intrusion of the mantle materials at the early stage of the Emeishan basalt eruption. The intrusion enhanced the mechanical strength of the crust, which forMs a stronger block centered on the Panzhihua region. The southward extrusion of the plateau materials is obstructed by this block, leading to rapid uplift in the north.

The intersection region of the Xiaojiang and Red River faults is the southern boundary of the Sichuan-Yunnan active blocks. The P-wave velocity in this region mainly shows low-velocity anomalies in the upper and middle crust. Below 30 km depth, there exists an obvious high-velocity anomaly in the crust and the uppermost mantle, and its front edge extends to the north with increasing depth (Fig.3 and Fig.5). The horizontal displacement along the Xiaojiang fault zone since the Neogene changes significantly, about 50 km in northern, the 42 km in middle and 12 km in southern segments[36]. GPS observations reveal that the left-lateral strike-slip rate in the middle and the northern segment of the Xiaojiang fault zone is about 8~10 mm/a, while in the southern segment[37, 38] that reduces to 4 mm/a. To the north of the Red River fault, the Qüjiang-Shiping fault zone, which is composed of the Qüjiang fault and Shiping fault[39], is dominantly of dextral strike slip with a compressive thrust component. The modern right lateral slip/shear deformation rate in the Qüjiang-Shiping fault zone is about 4.5 mm/a[37]. On this fault zone, 5 earthquakes with M > 7 and 11 earthquakes with M > 6 occurred in the recent 500 years. Based on the P wave velocity structure, we consider that the formation of the southern boundary of the Sichuan-Yunnan block is not only related to the Red River fault, but also to the existence of the high-velocity anomaly body in the middle-lower crust and the upper most mantle in this region (Fig.3 and Fig.5). The northward dipping high-velocity and the high-strength anomaly body plays a certain impediment to the southward escaping of the Sichuan-Yunnan block and causes the large deformation of the low-velocity medium in the upper and middle crust, leading to the dominantly dextral strike-slip Qüjiang and the Shiping fault shows a compressive thrust component. The impedance of the high velocity anomaly to the southward escape of the Sichuan-Yunnan block can also be indirectly confirmed by the feature of velocity anomalies. In the 24°N-26°N, the long axis of the low-velocity anomaly body in the upper and middle crust is nearly in the N-S direction, while that in the lower crust and upper mantle is altered to the E-W direction. The E-W striking low-velocity anomaly is probably caused by the two-side compression deformation from the high-velocity anomaly blocks beneath the Panzhihua region in the north and the Red River fault in the south. The difference of structure features between the upper and the lower crust indicates distinctly different motions between the upper and the lower crust in this region.

Fig.5 Vertical depth sections along the latitude direction  (a) and (c) are vertical depth sections of the velocity perturbations and absolute velocities along 101.75°E, respectively; (b) and (d) are vertical depth sections of the velocity perturbations and absolute velocities along 102.72°E, respectively. The dash lines indicate faults, and arrows represent the direction of block movement. The solid circles represent historical earthquakes. The white line indicates the number of ray hit count as 50. In the region above the white line, the number of hit count is larger than 50. F1-Red River fault, F2-Shiping-Jianshui fault, F3-Chuxiong-Qujiang fault.

SKS splitting studies[40, 41] show that the polarization direction of the fast shear waves in the middle and the northern segments of the Sichuan-Yunnan block (in the north of 27°N) is coincident with the direction of block movement. However, it changes abruptly in the middle and southern regions, and alters into a nearly E-W direction in the southern area, which is almost perpendicular to the block movement, and identical to the fast polarization direction in its southern area. Therefore, we infer that the high-velocity anomaly body C in the southern part of the Sichuan-Yunnan block may extend downward into the upper mantle (Fig.5(a, b)), which plays an important role in the deformation of materials in the upper mantle. The existence of the high strength material makes itself and its southern region less affected by the southward extrusion of plateau material.

There are two local high velocity anomaly bodies beneath the Qüjiang-Shiping fault zone in the upper crust (D and E in Fig.5b), which lie in the north of Qüjiang fault and Shiping-Jianshui fault respectively. Earthquakes with M > 6 mainly occurred around the high-velocity anomalies. Unfortunately, it is hard to determine reliable focal depths from the historical earthquake data. Considering the Ms7.7 Tonghai earthquake on Jan 5th, 1970 has a focal depth of 13 km, and the focal depths of earthquakes with M > 3 in 1999-2001 determined by the waveform fitting technique are almost all among 9~12 km[42], we consider that the focal depths of the historical strong earthquakes most are likely in this depth range. In the P-wave velocity map at the 10 km, strong earthquakes are mainly located at the edge of the low-velocity anomaly or within the small high-velocity anomaly bodies between two low-velocity anomalies (Fig.3b) in this area. On the velocity map at 20 km, strong earthquakes are mainly located at the edge of the low-velocity anomalies. A local high-velocity anomaly body may play as an obstacle, which facilitates accumulation of strain energy. On the other hand, the boundary zone between high and low velocity anomalies is in a state between brittle and ductile, so if the energy is accumulating high enough, brittle rupture can occur. Based on the seismic sounding profiles and earthquake location results, Wen et al.[37] suggested that the Qüjiang-Shiping fault zone consists of a nappe structure thrusting from north to south, the major fault dips toward NE-NNE and intersects with a basement detachment zone at a depth of about 12 km. Seismic tomography shows a similar image but reveals that the depth of the detachment zone should be deeper, about 25 km depth on the upper boundary of the high-velocity anomaly distributed in the middle-lower crust and the upper mantle. In the upper and middle crust, a distinct low-velocity anomaly is imaged between the Puduhe fault and east Xiaojiang fault. The center of this anomaly is located at the Wuding-Kunming area. Below 30 km depth, the low-velocity anomalies are distributed in a small area around the Wuding city. The vertical cross section shows that the shape of the low-velocity anomaly looks like a mushroom (Fig.5). Based on the heat flow data[43, 44, 45, 46, 47, 48], the Xiaojiang fault zone has high heat flow with a mean value of 85 mW/m2, which ranks second only to the Tengchong volcanic area[45] in Yunnan. We consider that the low-velocity anomaly beneath the Xiaojiang fault zone is probably correlated with the high heat flow at depth and the center area for the heat flow exchange between the crust and upper mantle is around Wuding. Owing to the limitation of the regional tomography, the reliable depth extension of the low-velocity anomaly can not be obtained and more information is needed to constrain it in future study.

5 CONCLUSIONS

A relatively large-scale high velocity anomaly is imaged in the crust in the Panzhihua and its vicinity. We suggest that it is probably related to the late Paleozoic mantle plume activity, which caused a large amount of basic and ultrabasic mantle material intruded into the crust. The intrusion of mantle material enhances the mechanical strength of the crust and forMs a stronger block centered on the Panzhihua region, which partially hinders the southward escaping of Tibetan Plateau material and leads to the rapid uplift of the northern sub-block of the Sichuan-Yunnan active block as well.

In the southern area of the Sichuan-Yunnan active block, there exist low-velocity anomalies in the upper and middle crust and an obviously high velocity anomaly in the lower crust and upper mantle. The highvelocity and mechanically stronger anomaly in the middle and lower crust obstructs the Sichuan-Yunnan block from escaping southward rapidly, and causes the larger deformation of the low-velocity area in the upper and middle crust, which leads to the Qüjiang-Shiping dextral strike-slip fault zone having a compressional thrust component. The Qüjiang-Shiping fault zone consists of a nappe structure thrusting from north to south, and the major fault dips toward the NE-NNE, and intersects with a basement detachment zone at a depth of about 25 km. Combining the SKS splitting studies, we infer that the high-velocity anomaly body in the southern area of the Sichuan-Yunnan block extends deep in the upper mantle, which plays an important role in the upper mantle deformation by reducing the effects of southward extrusion of the plateau material on its southern region.

In the middle of the Xiaojiang fault zone, there exists a widely distributed low-velocity anomaly in the upper and middle crust. The anomaly can extend into the upper mantle. Considering the high mean heat flow in the Xiaojiang fault zone reaches about 85 mW/m2, we infer that this low-velocity anomaly is related to the deep thermal process. Different from the N-S direction long axis of the low-velocity anomaly in the upper and middle crust, the low-velocity anomaly in the low and middle crust might be deformed greatly under the pressing of the high-velocity anomaly bodies on its north and south sides, making the long axis change into the E-W direction. This phenomenon also indicates decoupling of the motion between the upper and lower crust.

ACKNOWLEDGMENTS

We thank D. P. Zhao for providing the source codes of the travel time tomography. This work was supported by the National Natural Science Foundation of China (41074068) and the Special Funds for Research of Earthquake Science (200708035).

References
[1] Tapponnier P, Molnar P. Active faulting and tectonics in China. J. Geophys. Res., 1977, 82(20):2905-2930.
[2] Allen C R, Gillespie A R, Han Y, et al. Red River and associated faults, Yunnan Province, China:Quaternary geology, slip rates, and seismic hazard. Geological Society of America Bulletin, 1984, 95(6):686-700.
[3] Xu Z Q, Yang J S, Li H B, et al. On the tectonics of the Indian-Asia collision. Acta Geolocia Sinica (in Chinese), 2011, 85(1):1-33.
[4] King R W, Shen F, Burchfiel B C, et al. Geodetic measurement of crustal motion in southwest China. Geology, 1997, 25(2):179-182.
[5] Wang G, Wang E Q. Extensional structures within the compressional orogenic belt and its mechanism:A case study for the late Cenozoic deformation in central Yunnan. Seismology and Geology (in Chinese), 2005, 27(2):188-199.
[6] Cui Z Z, Lu D Y, Chen J P, et al. The deep structural and tectonic features of the crust in Panxi area. Chinese J. Geophys. (in Chinese), 1987, 30(6):566-580.
[7] Xiong S B, Zheng Y, Yin Z X, et al. The 2-D structure and it's tectonic implications of the crust in the Lijiang-Panzhihua-Zhehai Region. Chinese J. Geophys. (in Chinese), 36(4):434-444.
[8] Zhang Z J, Bai Z M, Wang C Y, et al. Crustal structure of Gondwana- and Yangtze-typed blocks:An example by wide-angle seismic profile from Menglian to Malong in western Yunnan. Science in China (Series D) (in Chinese), 2005, 34(5):387-392.
[9] Liu J H, Liu F T, Wu H, et al. 3-D velocity images of the crust and upper mantle beneath North-South zone in China. Chinese J. Geophys. (in Chinese), 1989, 32(2):143-151.
[10] Liu F T, Liu J H, He J K, et al. The subducted slab of Yangtz continental block beneath the Tethyan orogen in western Yunnan. Chinese Science Bulletin (in Chinese), 2000, 45(1):79-84.
[11] Huang J, Zhao D, Zheng S. Lithospheric structure and its relationship to seismic and volcanic activity in southwest China. J. Geophys. Res., 2002, 107(B10), 2255, doi:10.1029/2000JB000137.
[12] Wu J P, Ming Y H, Wang C Y. The S wave velocity structure beneath digital seismic stations of Yunnan province inferred from teleseimci receiver function modeling. Chinese J. Geophys. (in Chinese), 2001, 44(2):228-230.
[13] Wu J P, Huang Y, Zhang T Z, et al. Aftershock distribution of the Ms8.0 Wenchuan earthquake and 3-D P-wave velocity structure in and around source region. Chinese J. Geophys. (in Chinese), 2009, 52(2):320-328.
[14] Wang C Y, Mooney W D, Wang X L, et al. Study on 3-D velocity structure of crust and upper mantle in Sichuan-Yunnan region, China. Acta Seismologica Sinica (in Chinese), 2002, 24(1):1-16.
[15] Wei W, Sun R M, Shi Y L. P-wave tomographic images beneath southeastern Tibet:Investigating the mechanism of the 2008 Wenchuan earthquake. Sci. China Earth Sci., 2010, 53(9):1252-1259, doi:10.1007/sl1430-010.4037-5.
[16] Lei J, Zhao D, Su Y. Insight into the origin of the Tengchong intraplate volcano and seismotectonics in southwest China from local and teleseismic data. J. Geophys. Res., 2009, 114, B05302, doi:10.1029/2008JB005881.
[17] He Z Q, Ye T L, Su W. S-wave velocity structure of the middle and upper crust in the Yunnan region. Chinese J. Geophys. (in Chinese), 2004, 47(5):838-844.
[18] Yao H, van der Hilst R D, Montagner J P. Heterogeneity and anisotropy of the lithosphere of SE Tibet from surface wave array tomography. J. Geophys. Res., 2010, 115, B12307, doi:10.1029/2009JB007142.
[19] Hu J F, Su Y J, Zhou X G, et al. S wave velocity of the crust and the structure of Possion ratio in Yunnan and meanings. Science in China (Series D) (in Chinese), 2003, 33(8):714-722.
[20] Li Y H, Wu Q J, Tian X B, et al. Crustal structure in the Yunnan region determined by modeling receiver functions. Chinese J. Geophys. (in Chinese), 2009, 52(1):67-80.
[21] Xu Z, Xu M J, Wang L S, et al. A study on crustal anisotropy using P to S converted phase of the receiver function:Application to Ailaoshan-Red River fault. Chinese J. Geophys. (in Chinese), 2006, 49(2):438-448.
[22] Zhao D, Hasegawa A, Horiuchi S. Tomographic imaging of P and S wave velocity structure beneath northeastern Japan. J. Geophys. Res., 1992, 97(B13):19909-19928.
[23] Zhao D P, Hasegawa A, Kanamori H. Deep structure of Japan subduction zone as derived from local, regional, and teleseismic events. J. Geophys. Res., 1994, 99(B11):22313-22329.
[24] Zhang Y X, Luo Y N, Yang C X. The Panxi Rift (in Chinese). Beijing:Geological Press, 1988.
[25] Chung S L, Jahn B M. Plume-lithosphere interaction in generation of the Emeishan flood basalts at the Permian-Triassic boundary. Geology, 1995, 23(10):889-892.
[26] Xu Y G, Chung S L, Jahn B M, et al. Petrologic and geochemical constraints on the petrogenesis of Permian-Triassic Emeishan flood basalts in southwestern China. Lithos, 2001, 58(3-4):145-168.
[27] Xu Y G, He B, Chung S L, et al. Geologic, geochemical, and geophysical consequences of plume involvement in the Emeishan flood-basalt province. Geology, 2004, 32(10):917-920.
[28] Lo C H, Chung S L, Lee T Y, et al. Age of the Emeishan flood magmatism and relations to Permian-Triassic boundary events. Earth Planet. Sci. Lett., 2002, 198(3-4):449-458.
[29] He B, Xu Y G, Chung S L, et al. Sedimentary evidence for a rapid Kilometer-scale crustal doming prior to the eruption of the Emeishan flood basalts. Earth and Planetary Science Letters, 2003, 213(3-4):391-405.
[30] Zhang Z C, Mahoney J J, Wang F S, et al. Geochemistry of picritic an d associated basalt flows of the western Emeishan flood basalt province, China:evidence for a plume-head origin. Acta Petrologica Sinica (in Chinese), 2006, 22(6):1538-1552.
[31] Li J, Xu J F, Suzuki K, et al. Os, Nd and Sr isotope and trace element geochemistry of the Muli picrites:Insights into the mantle source of the Emeishan Large Igneous Province. Lithos, 2010, 119(1-2):108-122.
[32] Zhou M F, Robinson P T, Michaellesher C, et al. Geochemistry, Petrogenesis and metallogenesis of the Panzhihua gabbroic layered intrusion and associated Fe-Ti-V oxide deposits, Sichuan Province, SW China. Journal of Petrology, 2005, 46(11):2253-2280.
[33] Christensen N I, Mooney W D. Seismic velocity structure and composition of the continental crust:A global view. J. Geophys. Res., 1995, 100(B6):9761-9788.
[34] Stadtlander R, Mechie J, Schulze A. Deep structure of the southern Ural mountains as derived from wide-angle seismic data. Geophys. J. Int., 1999, 137(2):501-515.
[35] Xu X W, Cheng G L, Yu G H, et al. Tectonic and paleomagnetic evidence for the clockwise rotation of the Sichuan-Yunnan rhombic block. Seismology and Geology (in Chinese), 2003, 25(1):61-70.
[36] He H L, Fang Z J, Li P. A preliminary approach to the fault activity of southern segment on Xiaojiang west branch fault. J. Seism. Res. (in Chinese), 1993, 6(3):291-298.
[37] Wen X Z, Du F, Long F, et al. Tectonic dynamics and correlation of major earthquake sequences of the Xiaojiang and Qujiang-Shiping fault systems, Yunnan, China. Sci. China Earth Sci., 2011, 54(10):1563-1575, doi:10.1007/s11430-011-4231-0.
[38] Shen Z K, Zhao C K, Yin A, et al. Contemporary crustal deformation in east Asia constrained by Global Positioning System measurements. J. Geophys. Res., 2000, 105(B3):5721-5734.
[39] Wang E, Burchfiel B C, Royden L H, et al. Late Cenozoic Xianshuihe-Xiaojiang, Red River, and Dali Fault Systems of southwestern Sichuan and central Yunnan, China. Geol. Soc. Amer., 1998, 327:1-108.
[40] Chang L J, Wang C Y, Ding Z F. A study on SKS splitting beneath the Yunnan region. Chinese J. Geophys. (in Chinese), 2006, 49(1):197-204.
[41] Wang C Y, Chang L J, Lü Z Y, et al. Seismic anisotropy of upper mantle in eastern Tibetan Plateau and related crust-mantle coupling pattern. Science in China (Series D), 2007, 50(8):1150-1160.
[42] Wu J P, Ming Y H, Wang C Y. Source mechanism of small to moderate earthquakes and tectonic stress field in Yunnan Province. Acta Seismologica Sinica (in Chinese), 2004, 26(5):457-465.
[43] Wang J Y, Huang S P. Compilation of heat flow data in the China continental area. Scientia Geologica Sinica (in Chinese), 1988, (2):196-204.
[44] Wang J Y, Huang S P. Compilation of heat flow data in the China continental area (2nd Edition). Seismology and Geology (in Chinese), 1990, 12(4):351-366.
[45] Xu Q, Wang J A, Wang J Y, et al. Terrestrial heat flow and its tectonic significance in Yunnan, China. Geotectonica et Metallogenia (in Chinese), 1992, 16(3):285-299.
[46] Zhou Z H, Han X M, Zhou R Q. Geothermal structure of lithosphere in Tengchong region, Yunnan province. Seismological and Geomagnetic Observation and Research (in Chinese), 1996, 17(4):34-41.
[47] Hu S B, He L J, Wang J Y. Compilation of heat flow data in the China continental area (3rd edition). Chinese J. Geophys. (in Chinese), 2001, 44(5):611-626.
[48] Yuan Y S, Ma Y S, Hu S B, et al. Present-day geothermal characteristics in South China. Chinese J. Geophys. (in Chinese), 2006, 49(4):1118-1126.