Options
The Structural Model of the Lithosphere-asthenosphere System in the Qinghai-Tibet and its adjacent Areas from Surface Wave Tomography
Zhang, Xuemei
2010-03-29
Contributor(s)
Romanelli, Fabio
Abstract
The Qinghai-Tibet Plateau lies at the continent-continent collision between the
Indian and Eurasian plates. Because of their interaction the shallow and deep
structures are very complicated with different tectonic units. The force system
forming the tectonic patterns and driving tectonic movements is exerted together by
the deep part of the lithosphere and the asthenosphere. In the recent decades, many
deep explorations have been performed and a series of important results about the
collision models of Indian and Eurasian plates and their deep structures have been
gained, but the studies on the fine structure of the lithosphere-asthenosphere system
are still a few. In order to get knowledge about their formation and evolution,
dynamic process, layers coupling and exchange of material and energy, it is important
to study the 3-D velocity structures, the material properties and physical state of the
lithosphere-asthenosphere system.
Based on the Rayleigh wave dispersion theory, we study the 3-D velocity
structures, including the crust, of the lithosphere-asthenosphere system in the
Qinghai-Tibet Plateau and its adjacent areas.
In the study area (20ºN - 50ºN, 70ºE - 110ºE) we collect long period and
broad-band seismic data from the global and regional seismic networks surrounding
the area: G (Geoscope), IC (NCDSN) , II/IU (GSN), KZ (Kazakhstan), XA (Bhutan),
XR (INDEPTH II&III), YA (2003MIT-China), and YL (Himalayan Nepal Tibet
Experiment) during the period of 1966-2007. After making instrumental correction
and proper band-pass filtering, group velocities dispersion of fundamental mode of
Rayleigh waves are measured using the frequency-time analysis (FTAN). Cluster
averaging is applied to similar ray paths and, in such a way, a set of dispersion curves,
in the period range from 8 s to 150 s is obtained along 791 paths.
A 2-Dsurface-wave tomography method capable to evaluate the average lateral
resolution, proposed by Yanovskaya, is applied to calculate the lateral variations in the
group velocity distribution at the different periods. The lateral heterogeneity
resolution has been estimated to be about 200 km in most of the study area. To be
consistent with the resolution level, the group velocity maps, at different periods, are
discretized in cells of 2o×2o. The most conspicuous low group velocity anomaly, in
the period range from 25 s to 40 s appears in whole Qinghai-Tibet Plateau, while the
Indian Plate and the Yangtze craton are characterized by high group velocity
anomalies. At the intermediate periods (50 - 80 s) the most dominant feature is the
NW-SE directed low velocity anomaly in the Qinghai-Tibet Plateau. At the long
ii
periods the velocity anomaly is comparable with the anomaly at the lower periods.
The determination of the shear-wave velocity distribution versus depth from a
surface wave dispersion curve is a severely non-linear problem. The non-linear
Hedgehog inversion method (Panza, 1981) is applied to the surface wave tomography
cellular dispersion curves to obtain shear wave velocity-depth models of the crust and
upper mantle. The non-linear inversion does not depend upon the initial model.
Since the Hedgehog is a non-linear procedure, the inversion is multi-valued, i.e.,
a set of equally probable model is accepted, which is consistent, within the chosen
parametrization, with the experimental errors. An ensemble of acceptable models is
found and in order to summarize and define the geological meaning of the results, it is
often necessary to identify a representative model. Physical and mathematical reasons
can be used to define the criteria that allow us to select a unique solutions. The Local
Smoothness Optimization (LSO) (Panza et al., 2007；Boyadzhiev et al., 2008) fixes
the solution as the one which minimizes the norm between neighbouring cells.
Applying the first iteration of LSO, the Starting Cell (SC) is chosen such that satisfies
the analogous objective criterion: the cell with the minimal divergence between the
accepted solutions. Staring from the SC, the LSO takes, as representative model, the
solution that has the smallest distance (difference in the velocity) in l2 with respect to
the models of the neighbouring cells. The LSO method is a smoothness criteria, which
can avoid the introduction of heterogeneities that can arise from a subjective choice.
However, in the Qinghai-Tibet Plateau and its vicinity, the heterogeneity is too severe
to apply LSO to the whole study area, because its deep structure is very complicated.
It is appropriate to obtain representative models by LSO method in local regions, each
with different starting cell (SC).
Since the Hedgehog non-linear inversion and the local smoothing algorithm
provide us only a mathematical solution, the representative models are chosen,
considering a priori geophysical and geological information.
The top and bottom of the lithosphere and asthenosphere are recognized from the
velocity values and velocity contrast between the layers. These thicknesses are helpful
to study the structural differences between the Qinghai-Tibet Plateau and its adjacent
areas and among different geologic units of the plateau.
Taking into account also previous investigations, the following conclusions are
reached from the distributions of the S-wave velocities in the crust and the upper
mantle and thicknesses of the crust, lithosphere and asthenosphere.
(1) The crust is very thick in the Qinghai-Tibet Plateau, and varies from 60 km to
iii
80 km. The lithospheric thickness in the Qinghai-Tibet Plateau is smaller (125-160
km) than in the adjacent areas. The asthenosphere is relatively thick, varies from 100
km to 200 km, and the thickest area lies in the western Qiangtang block (QT). India,
located to the south of the Main Boundary thrust, has a thinner crust (32-42 km), a
thicker lithosphere of 190 km and a rather thin asthenosphere of only about 80 km.
Sichuan and Tarim basins have the crust thickness less than 50 km. Their lithospheres
are thicker than the Qinghai-Tibet Plateau, and their asthenospheres are thinner.
(2) The uppermost mantle of the Indian Plate is subducted almost horizontally
beneath the Himalaya block (HM) and the Lhasa block (LS), and the subduction is
delimited by the Bangong-Nujiang suture belt (BNS). The Indian lithospheric lid is
also subducted with a large-angle beneath the Eurasian Plate before the
Yalung-Zangbo suture belt (YZS). The low velocity lower crust and asthenosphere,
detected in central Qinghai-Tibet Plateau, show that in the Qiangtang block (QT) the
temperature is high, well in agreement with the active Cenozoic volcanism in the area.
We also think that the underplating of the asthenosphere may thin the lithosphere and
that the buoyancy might be the main mechanism of deep dynamics of the uplift of the
Qinghai-Tibet hinterland.
(3) Inside the plateau two blocks can be recognized, divided by an NNE striking
boundary running between 90ºE ~ 92ºE. The shear-wave velocities of the crust and
the thicknesses of the lithosphere and asthenosphere in the eastern Qinghai-Tibet
Plateau are different from those in the western one. The width of the boundary
between the eastern Qinghai-Tibet Plateau and the western one may be 2° ~ 3°.
(4)The continental surface loss by the kinematic shortening is not compensated
by the increment of the crust thickness due to the collision of Indian Plate and
Eurasian Plate. Therefore we may deduce that the crustal material is laterally extruded
along a channel between the Jinshajiang suture belt (JSJS) and Banggong-Nujiang suture
belt (BNS), and rotated around the eastern Himalayan Syntaxes because of the
obstacle of the Yangze block. The source of the lateral extrusion may be in the
Qiangtang block (QT).
Insegnamento
Publisher
Università degli studi di Trieste
Languages
en
File(s)