LITHOSPHERIC STRUCTURE AND CONTINENTAL GEODYNAMICS
XU Zhonghuai1) and SHI Yaolin2)
1) Institute of Geophysics, China Seismological Bureau, Beijing 100081, China
2) Center for Earth System Science, The Graduate School of The Chinese Academy of Sciences,
Beijing 100039, China
During the last 4 years since 1999 22nd IUGG general assembly, lithospheric structure and continental geodynamics of China continent is an active research field for Chinese geophysicists. Research interests are mainly in seismic wave velocity structure of the lithosphere and its present-day deformation.
I. REGIONAL LITHOSPHERIC STRUCTURE
Moho depth map for East Asia Since 70 s of 20th century to 2000, about 5000 km DSS profiles, including more than 2000 km reflection profiles, have been carried out in China. Based on the data of these profiles and the published DSS study results for the neighboring countries and sea regions, a map of Moho depth distribution has been compiled (Teng et al., 2002, Fig.1). In the compilation the Moho depth data are sampled every 20 to 50 km space for China mainland region, producing about 3000 datum samples, and every 50 to 100 km for surrounding regions, giving additional 1000 datum samples. Background map scale is 1:25,000,000. Depth resolution of the map reaches to ±2 km. This is the most detailed and updated Moho depth map for East Asia
Pn velocity lateral variation Wang et al. (2003) tomographically inverted nearly 140,000 Pn arrival times, and obtained lateral variation and anisotropy of Pn velocity in top mantle beneath China continent. The result indicates that high velocity is seen in the region of stable basins around Qingzang (Qinghai-Xizang) Plateau, such as Tarim, Junggar, Turpan-Hami, Qaidam and Sichuan basin, while low velocity is found in tectonically active region, such as western Sichuan and Yunnan, middle part of Qingzang Plateau, Shanxi graben and North China basin. Beneath compressive basins the velocity tends to be high, while beneath extensional basins or grabens it is usually low.
Fig.2. Pn velocity variation with respect to the average 8.05 km/s in and around China. Lines demarcate tectonic blocks.
Combilation of heat flow data Hu et al.(2001) published the third compilation (3rd edition) of heat flow data in China mainland. Their paper presents 450 new data since the 2nd edition. Up to now in China mainland there are 862 heat flow data, among which 816 have been published.
Surface wave study Many researchers investigated wave velocity structure in crust and upper mantle beneath China continent by means of Rayleigh wave dispersion analysis. They have obtained the following main results: ① The upper mantle low velocity zone beneath North China plain is shallow and thick (He et al., 2002), with its top boundary being possibly at 80-90 km (Xu et al., 2000). This supports the idea that this region is under a state of extension. ② Velocity in upper mantle beneath Qingzang Plateau is remarkably low. ③ Upper mantle velocity beneath stable Tarim basin and Yangtze platform is relatively high (He et al.,2002; Zhu et al., 2002). ④ In the boundary region between Burma and Yunnan of China there is a clear low velocity zone in upper mantle, as revealed in the wave period range of 12s to 120s (Zhu et al., 2002). ⑤ In southern China the Moho depth is 30-40 km and the crust becomes gradually thin from west to east, and low velocity zone exits in the depth range of 60-150 km, with the depth range being laterally variable (Teng et al., 2001).
Using 12000 waveform records of long period Rayleigh wave Zhu et al. (2002) calculated 4100 dispersion curves along great circle path with a period range of 8s to 250s, and then both the dispersion and waveform data were inverted to obtain a 3-D S wave velocity image for the region of East Asia and marginal seas of western Pacific Ocean (WPO). Their result indicates that, in the depth range from upper crust to 70 km deep, S-wave velocity in western part of the region is low, with the Qingzang Plateau in central part of the low velocity zone, while the velocity in eastern part of East Asia and in the region of marginal seas of WPO is high, and in the Iran-Himalaya-Burma-Indonesia plate collision zone low velocity prevails. In the upper mantle of 85 km to 250 km deep, there is a large low velocity zone in the eastern part of East Asia and in the region of marginal seas of WPO. Zheng et al.(2000) studied the lateral variation of Rayleigh wave group velocity in the sea region to the east of China, and also found that the upper mantle beneath Japan Sea and Okinawa trough shows low velocity.
Q distribution Qβ structure of crust and upper mantle in eastern and western China is obtained by inverting Rayleigh wave attenuation factors varying with periods (Li et al., 2000). Their result indicates that in large part of the Yangtze quasi-platform at the depth of 88 km there is a significant high Qβ zone, while to the east of Yangtze quasi-platform and in middle and eastern part of South China fold system the upper mantle low Qβ zone appears at a depth of 85 km. In west Yunnan folding and faulting system in top mantle there is a low Qβ layer of about 40 km thick, and down to the depth of 95 km an even lower Qβ layer reveals. In western China the upper mantle low Qβ layer exits in many regions, for example, on the path from Gaotai to Lhasa across the Qingzang plateau, at about 84 km deep, there appears low velocity and low Qβ layer.
Gravity inversion for lithosphere lower boundary Using the technique of separating wavelet transformed wave field Fang et al. (2001) found deep gravity anomaly from Bouguer anomaly observation. Under some known constraints they inverted the deep gravity anomaly in a series expression of spherical harmonics in the region of eastern China and its vicinity, and obtained an image of the lower boundary of lithosphere: its depth varies from 35 km in eastern sea region to 110 km inland, the lithosphere is thicker in South China than in Northeast and North China, and there are two strong gravity gradient zones trending NE-SW and NNE-SSW, respectively.
Yunnan By inverting teleseismic receiver functions deduced from analyzing the broadband records of 23 digital seismic stations in Yunnan Province Wu et al.(2001a) obtained S-wave velocity structure in the depth range of 0 to 100 km beneath the stations. In northwestern Yunnan the crust thickness is 62 km or so, while in southern Yunnan it is only 32 to 34 km. The thick crust
Fig.3. Two vertical profiles of S-wave velocity variation in Yunnan region obtained from receiver function
inversion Left plot shows positions of the two profiles
thins out from northwest to southeast and is limited within the region bounded by Xiaojiang fault and Yuanjiang fault, with its shape in concordance with the Sichuan-Yunnan rhomb block. In eastern and southern Yunnan, where crust is thin, the Moho discontinuity is clear. In the region with thick and variable crust the Moho usually manifests itself as a zone of high gradient of S-wave velocity. In Fig.3 are given two vertical profiles of S-wave velocity variation obtained from receiver function inversion. From the figure we see that the velocity image shown in western profile (Fig. 3b) is obviously different from that in eastern profile (Fig. 3c).
Altyn Tagh fault zone In a Sino-France cooperative research during late August 1995 to January 1996, 30 one component short-period seismographs were deployed along a 400 km long profile for doing P wave tomographic study. Shi et al. (1999a) inverted 3883 P-wave arrival times picked up from the seismic records and obtained the deep velocity structure. The result reveals that the Altyn Tagh fault is a 40 km wide low velocity belt, which extends vertically down to a depth of about 150 km. The result also shows that the Altyn Tagh fault cuts through a Tarim-like lithosphere, which previously plunged down to the Qaidam basin.
Northeastern margin of Sino-Korean platform Lu et al.(2002) tomographically inverted 38,000 P-wave arrival times and reconstructed a 3-D velocity structure of crust and upper mantle in northeastern marginal region of Sino-Korean platform. It is shown that in the crust and upper mantle there is notable lateral velocity inhomogeneity, which extends down to 1200 km depth, and several low velocity zones are seen in the upper and middle crust in the region of Haicheng, Chaoyang, Yixian, south of Dandong and Tangshan,etc..
Antarctic Based on the waveform data of the Rayleigh wave traveled along Earth's great circle across the station South Pole (SPA) and Scott (SBA), Shu and Jiao (1999) computed the phase velocity dispersion and by inversion obtained the lithosphere S-wave velocity structure in the depth range down to 200 km below the two stations. Their result shows that beneath Trans-Antarctic mountain the crust thickness is 45 km, and there is a clear low velocity layer between 55 and 75 km, indicating possible existence of magma.
Some authors made improvement in the method of investigating lithospheric structure.
Smooth constraint technique in genetic algorithm Smooth constraint is important in linear inversion, but it is difficult to apply it directly to model parameters in genetic algorithms. If the model parameters are smoothed in iteration, the diversity of models will be greatly suppressed and all the models in population will tend to be identical in a few iterations, so the optimal solution meeting requirement can not be obtained. Wu et al.(2001b) introduced an indirect smooth constraint technique in genetic inversion. In this method the new models produced in iteration are smoothed, and then used as theoretical models in calculating misfit function, but in the process of iteration only the original models are used in order to keep the diversity of models. This technique was shown to be effective in a test of surface wave and receiver function inversion.
Evaluating solution with covariance matrix After converting nonlinear equations to linear ones in seismic tomography, the inversion problem is reduced to the one of solving ill-posed equations. Following the principle of finding resolution matrix of the solution, Liu and Chang (2000) put forward an evaluation criterion by making use of the covariance matrix of the solution for LSQR algorithm. Making correlation analysis can provide a quantitative evaluation of the solution for those inversion algorithms which can not give the resolution matrix.
Use of polarization data Liu et al. (2000) used broadband three-component seismic records of the Beijing station (BJI) and calculated P-wave polarization of teleseismic events. The polarization data were then inverted to obtain the subsurface velocity structure, especially the detail of velocity discontinuities, in the region around the Beijing station. The result shows existence of an obvious low velocity zone in crust to the west of the station, in good agreement with previous studies. This verifies that polarization data could be applied to inversion for velocity structures. Travel time and polarization data can be jointly used to study velocity structure, while polarization data are more suitable for delineating the boundary of velocity anomalies.
II. DEEP SEISMIC SOUNDING SURVEY