T22D-01
Crust And Upper Mantle Structure Of The Bengal Basin And Bay Of Bengal From Surface Wave Group Velocity Dispersion Studies
Fundamental mode Rayleigh and Love wave group velocity dispersion for paths crossing the Bay of Bengal have been calculated for earthquakes in the Indo-Burman arc and the Andaman-Sumatra subduction zone recorded at seismographs in the eastern part of Peninsula India and Sri Lanka. The ray-path coverage in this study provides a better spatial sampling than any previous studies of the region. The individual dispersion curves range from 12 to 70~s and have been clustered in four spatial groups to form average dispersion curves representative of the Bengal basin, northern, central and southern Bay of Bengal. These average dispersion curves for Rayleigh and Love waves are jointly inverted to obtain shear wave velocity structure of the lithosphere. The higher frequencies/shorter periods (12--30~s) used in the inversion constrains the sediment shear wave speed and thickness while the longer periods provide information of the upper mantle structure. The results show a remarkable increase in the sediments thickness along the Bengal Fan from south to north ranging from 6 km, around the southern tip of India, to 23 km beneath the Bengal basin. The shear wave velocity models reveal a sediment saturation beyond 7-10 km of burial leading to metamorphism and eventual increase in velocity to continent like material with depth. The average crustal thickness (loose sediments overlying consolidated sediments followed by metasediments and oceanic crust) is anomalously continental (~20-36 km) rather than being simply oceanic crust overlain by sediments. The average shear wave velocity is about 3.5-3.8 km/s which is more representative of continental crusts. Finally the low velocity zone in the uppermost mantle is possibly an effect of the expected increase in temperature due to blanketing of the fan sediments over the Bay of Bengal crust. The misfits to parts of the dispersion data using a 1D isotropic model provides an indication of the presence of polarization anisotropy in the lithosphere and sets a good starting point for modeling the anisotropic structure.
T22D-02
Strong crust-mantle coupling at Ordos plateau in North China inferred from shear-wave splitting observations of three seismic arrays at its boundary
Since September 2005, three temporary seismic linear arrays consisting of total 49 portable seismic stations were deployed across the boundary of the Ordos plateau at three locations with different time periods to image the crustal and upper mantle structure of the seismically active fault zones that bound the Ordos plateau, which is a relatively stable block in North China. In this study we focus on the shear-wave splitting data of these temporary seismic arrays to obtain information of the horizontal mantle anisotropy directions at the boundaries of the Ordos plateau. About two-years data of these temporary stations plus several nearby permanent stations were analyzed for shear-wave splitting of both the SKS and the SKKS phases. Because of the relatively dense station spacing (~10 km) of these linear arrays, the data offer good lateral resolution of mantle anisotropy across the plateau boundary. Preliminary results reveal that the fast polarization direction and delay time are not significantly affected by the plateau boundary at the Weihe graben (south boundary) and Sanxi graben (east boundary). In particular fast directions are almost NW-SE trending for all the stations. The uniform fast direction of the mantle anisotropy agrees well with the surface deformation directions inferred from both GPS data and observations of Quaternary fault slip rate, which implies strong crust-mantle coupling at Ordos plateau and its adjacent regions.
T22D-03 INVITED
The upper mantle Sv wave speed structure beneath Tibet
Although it has long been accepted that the uplift of the Himalaya and Tibet result from the collision of India with Eurasia, there is still no agreement on the details concerning the geodynamical processes involved. This lack of consensus arises primarily due to uncertainties in the Tibetan upper mantle structure. To improve understanding of this structure we determine a 3D Sv wave speed and azimuthal anisotropy model for the region constrained by the analysis of more than 22000 vertical component, short propagation path, multi-mode Rayleigh wave seismograms. The dense path coverage, wide azimuthal distribution and rich higher mode content of the data set allow us to build an upper mantle model for Tibet and the surrounding region with a horizontal resolution of a few hundred kilometers and a vertical resolution of a few tens of kilometers extending to about 400 km depth. The surface wave analysis shows that sub-Moho upper mantle shear wave structure is slow down to about 130 km, but at deeper depths, down to about 250 km, the long wavelength pattern of Sv velocity of the mantle beneath the whole of Tibet is fast with respect to PREM. The high wave speed mantle at these depths beneath Tibet merges to the south with high wave speeds beneath northern India and to the east with high wave speeds beneath southeastern China. The most commonly invoked model for the uplift of Tibet involves lithospheric delamination but our results indicate that the Tibetan lithosphere is still largely, if not completely intact. The seismic model suggest that most of the plateau has been underthrust by high wave speed Indian mantle from the south and possibly to a lesser extent, high wave speed Asian mantle from the north. The surface wave results are supported by observations of frequency-dependent Sn propagation in Tibet. High frequency Sn does not propagate beneath northern Tibet but low frequency Sn is observed to propagate across all of the plateau. Low velocities in the upper mantle beneath northern Tibet have previously been noted and these, along with the recent volcanism in northern Tibet, have been cited as evidence for lithospheric delamination beneath the plateau. However, both the surface wave tomography and Sn propagation efficiency studies reported here show that the low wave speeds in the upper mantle below Tibet are only a shallow feature and not the result of lithospheric delamination. Therefore, the elevation of Tibet is not the result of uplift caused by hot, buoyant astenospheric material replacing the Tibetan lithosphere.
T22D-04 INVITED
Integrated Seismic Arrays for Imaging the North China Craton: the ¡°Destruction of the North China Craton" Project
It has been known, mostly according to the petrological and geochemical studies particularly the xenolith's data, that the North China craton (NCC), which is part of the Archaean Sino-Korean craton, had been reactivated since Mesozoic, and experienced widespread extension and volcanism through much of the Cenozoic. Currently the NCC is characterized by a thin lithosphere (as thin as 80 km according to published studies) and strong internal deformation, where over half of the eastern China's earthquakes occurred while the two major plate boundaries about China are thousands kilometers away. Although it is seismically quite active this region encompasses the China's capital and several mega cities which together hosts a large population and are very important to China's growing economy. Supported by the Chinese earth science community, the Chinese NSF recently started a major research program, the ¡°Destruction of the North China craton" (DNCC). About 150 million RMB (~ 20 million US dollars) will be allocated for this 5-year multi-disciplinary research program which is open for competition for all the earth scientists in China. Here we report one major seismic observation project of ¡°Integrated Seismic Arrays of DNCC" just funded during the first phase funding of DNCC. This observation-driven project integrates two groups at the Institute of Geology and Geophysics, Chinese Academy of Sciences and Peking University, and both institutions have their own broadband seismometers and have recently conducted pilot portable seismic array studies in North China. Up to seven linear broadband seismic arrays, each consists of 60-100 stations, are planned within the NCC. The principle objectives are to quantify the range and degree of the craton destruction in spatial domain with major focus on the east-west variation from the previously proposed intact craton in the west to the rejuvenated region in the east and the transition zone in the middle. With the expected imaging results of the crustal and upper mantle structures this integrated seismic arrays will provide the fundamental observational information to the overarching goal of understanding the ¡°Destruction of the North China cration", which is one of the important tectonic processes in the evolution of the continents on the earth.
T22D-05
Seismic array observations at the boundary of Ordos plateau, North China
Since September 2005, three temporary seismic linear arrays consisting of total 49 portable seismic stations were deployed across the boundary of the Ordos plateau at three locations with different time periods to image the crustal and upper mantle structure of the seismically active fault zones that bound the Ordos plateau, which is a relatively stable block in North China. These arrays were conducted as a pilot study by Peking University for the implementation of the ¡°Integrated Seismic Arrays", which is a major observational program of a new research initiative, ¡°Destruction of North China craton", recently funded by the Chinese NSF. Preliminary results from over one-year observations at these arrays show that the faults, which form the Weihe graben and Sanxi graben at the boundary of the Ordos plateau, cut through the Moho indicating that these are lithospheric faults. Deformation at the upper mantle correlates to the deformation at the surface through these deep-cutting faults. Relatively uniform fast directions across these boundary faults from shear-wave splitting analyses and their agreement with the surface deformation directions inferred from both GPS data and observations of Quaternary fault slip rate imply strong crust-mantle coupling at Ordos plateau and its adjacent regions.
T22D-06 INVITED
Crust and Upper Mantle Structure Beneath Tibet and SW China From Seismic Tomography and Array Analysis
We will present a summary of the results of our seismological studies of crust and upper mantle heterogeneity and anisotropy beneath Tibet and SW China with data from temporary (PASSCAL) arrays as well as other regional, national, and global networks. In 2003 and 2004 MIT and CIGMR (Chengdu Institute of Geology and Mineral Resources) operated a 25 station array (3-component, broad band seismometers) in Sichuan and Yunnan provinces, SW China; during the same period Lehigh University (also in collaboration with CIGMR) operated a 75 station array in east Tibet. Data from these arrays allow delineation of mantle structure in unprecedented detail. We focus our presentation on results of two lines of seismological study. Travel time tomography (Li et al., PEPI, 2006; EPSL, 2007) with hand-picked phase arrivals from recordings at regional arrays, and combined with data from over 1,000 stations in China and with the global data base due to Engdahl et al. (BSSA, 1998), reveals substantial the structural complexity of the upper mantle beneath SE Asia. In particular, structures associated with subduction of the Indian plate beneath the Himalayas vary significantly from west Tibet (where the plate seems to have underthrusted the entire plateau) to east Tibet (where P-wave tomography provides no evidence for the presence of fast lithosphere beneath the Plateau proper). Further east, fast structures appear in the upper mantle transition zone, presumably related to stagnation of slab fragments associated with subduction of the Pacific plate. (2) Surface wave array tomography (Yao et al., GJI, 2006, 2007), using ambient noise interferometry and traditional (inter station) dispersion analysis, is used to delineate the 3-D structure of the crust and lithospheric mantle at length scales as small as 100 km beneath the MIT and Lehigh arrays. This analysis reveals a complex spatial distribution of intra-crustal low velocity zones (which may imply that crustal-scale faults influence the pattern of middle/lower crustal flow). We will also show preliminary results of surface wave inversion for azimuthal anisotropy, which – combined with previous results from shear wave splitting (Lev et al., EPSL, 2006) – give insight into the deformation of the upper mantle beneath the area under study.
T22D-07
Shear Velocity Structure and Anisotropy in the Crust and Upper Mantle Beneath SE Tibet From Ambient Noise and Teleseismic Surface Wave Array Tomography
Southeastern Tibet plays a very important role in the dynamic evolution of the Tibetan Plateau. Models of middle or lower crustal channel flow have been proposed to explain the deformation and tectonics in Tibet. High resolution array tomography can help to determine the existence and geometry of low velocity layers (LVLs) in the crust where channel flow may occur. In this study we adopt recent tomographic method from ambient noise and teleseismic surface wave array analysis to study the shear velocity structure and azimuthal anisotropy in the crust and upper mantle beneath SE Tibet using data from temporary array stations deployed by MIT and Lehigh University in 2003-2004. We measure Rayleigh wave phase velocity dispersion curves in the period band 10-50 s from inter-station empirical Green's functions recovered by 10 months continuous recordings. We also measure phase velocity dispersion curve in the period band 20-200 s from teleseismic surface wave two-station analysis. In the period of overlap (20-50 s), ambient noise and two-station analysis yield very similar dispersion. Finally we average dispersion data from ambient noise and two-station analysis to obtain inter-station dispersion curves in the period band 10-200 s, and then produce 2-D phase velocity maps at each period. The path coverage is very dense, yielding a horizontal resolution 50-100 km in the crust and 100-200 km in the upper mantle. Neighborhood algorithm is used to invert the point-wise dispersion curve at each grid point from 2-D phase velocity maps for the 1-D shear velocity structure. Finally we combine all the point-wise 1-D models to construct 3-D shear velocity structure from the crust to 410 km depth beneath SE Tibet. Our results demonstrate that LVLs dominate at mid-crustal depth beneath SE Tibet, which is consistent with previous MT studies which mainly show low resitivity layers at similar depth range in this area. We also observe some LVLs existing in the lower crust, e.g., Yunnan province. The horizontal transition of LVLs to normal or high velocity layers in SW China seems to correlate with some major fault zones. In the upper mantle, we observe a pronounced high velocity layer (HVL) at 150-220 km depth range beneath almost the whole study region. How this HVL will be related to the Indian lithosphere subduction is an important issue to study. Beneath this HVL to 400 km, shear velocity is apparently slower than the global ak135 value, suggesting a thick and pronounced asthenosphere beneath SE Tibet. The magnitude of horizontal heterogeneity beneath the study region is very large from upper crust to 250 km depth and generally small from 250 km to 400 km depth. Beneath Tengchong Volcanic area, we observe clear low velocity anomaly in the upper mantle. Shear velocity appears fast at 150-220 depth range beneath eastern Lhasa block, but changes to be slow below 220 km depth. The region around Eastern Himalayan Syntaxis shows high velocity feature at almost all depth range. With the dense path coverage in the array area, we are going to invert for the surface wave azimuthal anisotropy in the crust and upper mantle and investigate how the surface wave anisotropy correlates with the surface deformation from GPS studies and mantle anisotropy from shear wave splitting studies. With both high-resolution structure and anisotropy results, we are expecting to understand more about the tectonics and dynamics beneath the Tibetan Plateau.