T11E-01
Shear Velocity Structure of the Indian Shield From Surface Wave Tomography and Receiver Function Inversion
We present an improved shear velocity model for the crust and upper mantle of the Indian shield, an assembly of Precambrian cratons bounded by mobile belts and paleo-rifts, obtained by inverting receiver function data and a set of recently constructed Rayleigh wave group velocity maps of India and the surrounding regions. The latter provide a higher resolution dataset than previous global and regional studies and the inclusion of shorter paths (<500km) increases the high frequency content, providing better constraints on upper crustal structure. Dispersion curves were extracted from the group velocity maps for points on a 0.5° by 0.5° grid and inverted for 1D shear velocity structure to 120km depth using a linearised least squares inversion method. Group velocity dispersion of surface waves is sensitive to shear velocity structure but does not contain information on discrete velocity boundaries; synthetic tests show that such boundaries are blurred to the order of 20km. The availability of receiver functions which are more sensitive to impedance contrasts, allows a more accurate determination of crustal structure below seismic stations and data for more than 40 stations across India have been re-examined and jointly inverted with the dispersion data from this study. The resulting shear wave velocity structure is not only better constrained for the areas where both data sets are available, but equally importantly, provides a means of using the more extensive surface wave data to calibrate crustal thickness elsewhere in the region. Our results show that thicker crust in the region of 40-45km is present beneath the Delhi-Aravalli fold belt and an area of central India straddling the Satpura Mobile Belt, with crust as thick as 54km observed beneath the Western Dharwar Craton. We also observe the Indian crust flexed beneath the Himalayan foreland basin and high velocity anomalies in the upper mantle beneath the Dharwar Craton in the south and the Singbhum and Bundelkhand Cratons in the north.
T11E-02 INVITED
Variation in lithospheric thickness and azimuthal anisotropy in Eastern Asia
We map the thick, cold continental lithosphere of eastern Asia using a surface wave data set consisting of ~40,000 vertical component multi-mode waveforms. Our path coverage allows us to resolve lid thickness and anisotropic variations with horizontal wavelengths matching the scale of the lithospheric blocks that have coalesced to form the continental landmass or eastern Asia. We assume great circle propagation in our tomography but test the validity of this assumption for the heterogeneous structure of eastern Asia with surface wavefront tracking. An extensive region of thick lithosphere exists beneath the Siberian Platform and the West Siberian Basin and extends to the European Platform, forming the stable Eurasian craton or core. Thick lithosphere exists beneath Tibet but the lithosphere is thin beneath the Archean crust of southern India and northeastern China. The eastern portion of the Eurasian craton has controlled the geometry of continental deformation in Eastern Asia. Except for the Indian subcontinent, the amplitude of sublithospheric azimuthal anisotropy is small over most of eastern Asia.
T11E-03 INVITED
P and S Velocity Structure in China and Surrounding Areas from Seismic Travel-time Tomography
Three-dimensional P- and S-wave velocity structures are obtained for the crust and upper mantle under China and surrounding areas using arrival times from local and regional networks and from global (e.g. EHB) data sources. For the crust and uppermost mantle tomography, 500,000 P-wave arrivals and 280,000 S- wave arrivals are used. The spatial grid used in the inversion is 1 degree by 1 degree in the horizontal direction and 10 km in depth. Sharp velocity contrasts at crustal interfaces and at Moho are incorporated into the model. Corrections for crustal travel time variations are used for P-wave mantle tomography based on regional and global arrival times. The crustal and uppermost mantle velocity models correlate well with geological and tectonic features. Major features such as the Tibet, Sichuan, Tarim, and Songliao Basins, and the Ordos Plateau are well- delineated in the velocity models. Under Tibet there is a low velocity zone in the mid-crust but the Pn and Sn velocities are high. Basins are characterized by high velocities of the lower crust and upper mantle. A surprising result of the studies is that high velocity "roots" of the Sichuan and Tarim Basins extend well into the mantle. Upper mantle images show the actively subducting slabs as well as the remnants of fossil slabs.
T11E-04
Colliding Lithosphere beneath Tibet: A Preliminary Synthesis of Results from Project Hi- CLIMB
The mainstay of Hi-CLIMB (An Integrated Study of the Himalayan-Tibetan Continental Lithosphere during
Mountain Building) is a hybrid linear/regional seismic array that has an aperture of over 800 km by 800 km.
With over 200 stations, this is by far the most extensive deployment of broadband, transportable seismic
arrays to date, with a dense spacing of only 3-8 km along the linear array. The resulting dataset, when
combined with those from other large-scale field experiments in central Asia, provides unprecedented
resolution throughout the thickened crust, into the lithospheric mantle, and reaching the bottom of the mantle
transition zone near a depth of about 700 km.
A preliminary synthesis of our latest results leads to the following new findings. Migration of receiver-functions
over a range of frequencies reveals a curious, wide zone of Moho disruption that extends over a distance of
about 250 km beneath the Bangong-Nujian suture (BNS), a near-surface feature associated with the collision
between Lhasa and Qiangtang terranes in the Mesozoic in central Tibet. In spite of this disruption in Moho,
average thickness of thickened Tibetan crust thins northward, from about 75 km beneath the southern Lhasa
terrane to about 63 km under the northern Qiangtang terrane. The variation in crustal thickness under near-
constant elevation indicates a large deviation from Airy isostasy in northern Tibet which must be
compensated by processes in the mantle. Multi-scale, finite frequency tomography of teleseismic P-wave
travel times show regions of low P-wave speeds within thickened crust that correspond well with zone of low
electric resistivity deduced from two north-south trending magnetotelluric profiles (MT). However, regional
coverage from seismic tomography clearly indicates that zones of low P-wave speed are not inter-connected
along the trend of the collision zone, so wide-spread, southward crustal channel flow, if present, is currently
inactive. Two step-wise increases in the magnitude of shear-wave birefringence are interpreted as the
southern termination of the Asian lithospheric mantle north of the Indus-Yalung suture and the leading,
northern edge of the Indian lithospheric mantle north of the BNS [Chen and Ozalaybey, GJI, 1998], the latter
also being evident from travel-time tomography.
The current configuration of mantle lithosphere, when combined with the onset of the latest episode of intra-
continental magmatism, current rate of ground motion, and a remnant of mantle lithosphere currently resting
just above the 660-km discontinuity under the Qiangtang terrane, facilitates a partial palinspastic
reconstruction of the colliding lithosphere back to about 15 Ma ago.
http://www.Illinois.edu/goto/pubs
T11E-05
Three-Dimensional Seismic Velocity Structure of the Crust and Upper Mantle beneath Western Tibet from Multiscale, Finite-Frequency Travel-Time Tomography
Three-dimensional seismic velocity structure of western Tibet has been resolved down to depths about 400 km from tomographic inversion of relative travel-time residuals of teleseismic P- and S-waves recorded during the Hi-CLIMB Experiment. A broad swath of high wave-speeds extends northward under much of the Lhasa terrane, passing beyond the Bangong-Nujang Suture and abruptly turns to low wave-speeds beneath the Qiangtang terrane in northern Tibet. The strong velocity contrast manifestly marks the leading edge of underthrusting Indian lithosphere. Two disjointed low-velocity anomalies straddling the Indus-Yarlung Suture in the uppermost 100 km or less are collocated with regions of high conductivity revealed from magnetotelluric imaging profiles, interpreted by many researchers as evidence for partial melt or aqueous fluids that weaken the lower crust and induce ductile flow. However, regional coverage of our data clearly shows that the two seismically slow regions throughout the thickened crust are not connected, suggesting that the popular geodynamic model of pervasive, south-directed channel flow in the mid-to-lower crust beneath southern Tibet is either presently inactive or in need of modifications.
T11E-06 INVITED
Imaging the crust and upper mantle in the southern and southeastern Tibetan plateau: A three-dimensional full-wavefield approach
Finite-frequency traveltime tomography based on body-wave ray theory in conjunction with the Born approximation has shown wave-speed anomalies that are indicative of the delamination of the mantle lithosphere and possibly parts of the lower crust as a likely cause for the north-south-tending rifts in the southern and southeastern Tibetan plateau. The images also show evidence that contradicts the notion of substantial under-thrusting of the Indian mantle lithosphere beneath southern and southeastern Tibet. To further improve resolution and verify the interpretations, we carry out a three-dimensional (3D) full-wavefield finite-frequency tomography of the area, incorporating regional earthquakes and ambient signals recorded by the permanent and portable broadband seismic stations. The new approach accounts for complex wave propagation, enables fuller utilization of an arrival on all three components of seismic records, and linearizes the inverse problem by iteratively updating the 3D reference model. An important benefit of this physically realistic modeling of full wavefield is the consistency of the system of equations in inversion, which is particularly important for the integration of different types of observations and physical properties. We calculate synthetic waveforms using a non-staggered grid finite-difference code in a polar-spherical coordinate system. Frequency-dependent phase and amplitude anomalies are measured by the cross- correlation between synthetic and recorded seismograms. The finite-frequency structural sensitivity kernels of these measurements are calculated by the scattering-integral method. The final joint 3D P- and S-speed structural model is obtained through a non-linear iterative process. The results will be used to address the questions about the under-thrusting of the Indian mantle lithosphere, the origin of the north-south-trending rifts, and crustal and mantle lithosphere delamination.
T11E-07
Seismic Evidence for the Moho in southern Tibet as a Phase Change Boundary
Since the Tibetan plateau has the thickest crust on the earth it challenges the traditional interpretation of the seismic Moho. At a depth of about 70-km, granulite (believed to be the major component of the lower continental crust) beneath the southern Tibet can be transformed into eclogite, which is seismically similar to mantle olivine. Here the phase change boundary between granulite and eclogite becomes the seismic ˇ°Mohoˇ±, instead of a traditional definition of the Moho being a material boundary between crust and mantle rocks. Because that the phase transition is sensitive to local P-T conditions, depth of this ˇ°Mohoˇ± varies with different P-T settings as opposed to the traditional Moho which is relatively stable to local P-T variations. This is consistent with our receiver function observations from two broadband seismic profiles in southern Tibet (International HI-CLIMB project and Peking University RISE project). The HI-CLIMB N-S profile shows that Moho becomes shallower near Indian subduction plate while RISE E-W profile shows that Moho is deeper beneath the Yadong-Gulu rift. These variations can be explained by the process of eclogitization of granulite that forms a phase transition boundary seismically observed as the ˇ°Mohoˇ± and the depth of the ˇ°Mohoˇ± varies with the local P-T conditions. Because this phase transition has a positive Clapeyron slope, that is, pressure increases with an increase in local temperature, the relatively wamer rift region has a deeper ˇ°Mohoˇ± due to a deeper transition depth while the cold subduction region has a shallower ˇ°Mohoˇ±.
T11E-08
Exploring the Rheology of Tibet from Postseismic Deformation Following Recent Large Earthquakes
The rheological structure of the crust beneath the Tibetan plateau is a big unknown in solid earth geophysics. During the last eleven years, there have been ten large earthquakes across the plateau with magnitudes between 6.0 and 7.9. The focal mechanisms are a mixture of normal, thrust and strike-slip. Together these earthquakes offer a good opportunity to explore the rheology of the crust beneath the plateau, through analysis of the postseismic transient deformation that occurs in the years following each event. We focus on three major earthquakes, all with M > 7, across the northern part of the plateau. These are the 1997 Manyi and the 2001 Kokoxili strike-slip events, and the 2008 Yutian normal faulting event. We present InSAR observations of postseismic motion for each case, using ERS, Envisat and ALOS data. The deep stress relaxation that is responsible for the observed surface transients is modeled in terms of different candidate rheologies for the mid to lower crust of Tibet, our aim being to determine the rheological model that best explains the geodetic measurements. Guided by earlier work by Ryder et al. (2007, Geophys. J. Int.) on the Manyi postseismic phase, we test linear viscoelastic models incorporating both univiscous and biviscous layers, and we also model localized afterslip. The modeling results are considered in the context of complementary geophysical data bearing on rheological properties. Additionally, we plan to look for postseismic signals following other, smaller earthquakes across the plateau, and run first order models to interpret the observations.