Options
Deep structure beneath the Central-South Tibet crustal density modelling and azimuthal anisotropy variation inferred from Quasi-Love wases
Zhang, Sufang
2010-03-29
Contributor(s)
Braitenberg, Carla
Abstract
The area of the present study is the central part of southern Tibet. It consists of
two accreted terranes, Lhasa and Himalaya terranes, which today record the
deformation history that originated from the processes of collision between the
Eurasia and India plates. Our study of the crust/mantle structure in terms of seismic
velocity, density, anisotropy and petrologic composition are undoubtedly significant
to deepen the understanding of the continent-continent collision and its dynamics.
This PhD thesis can be briefly summarized into four parts that are listed in the
following. 1) In order to reveal the characteristics of the crust/mantle deformation that
has been generated by the Indian/Eurasia collision in the southern Tibet plateau, we
study the propagation of Quasi-Love (QL) waves. Our study is based on the results
from numerical modeling, which proved that QL is sensitive to lateral variation of
seismic anisotropy, rather than heterogeneity and other factors. The results we obtain
from processing locally observed seismograms, reveal a West-East variation of
crust/mantle deformation in each terrane of the plateau. 2) A 3D density model of
central-south Tibet is produced by modeling the Bouguer gravity field using all
existing constraints. 3) Integrating seismic velocity and density models of the crust in
the Lhasa and Himalaya terranes, we infer crustal composition models in central and
southern Tibet. 4) Combining crustal density, velocity and mineralogical composition
models, some important issues, such as the Indian slab subduction angle, and the
relationship between crustal density and earthquake occurrences are discussed.
Some results based on the gravity modeling are summarized as follows: 1) under
the constraint of the geometrical structure defined by seismic data, a 3-D density
model and Moho interface are proposed for central-south Tibet; 2) the lower crustal
density, smaller than 3.2 g/cm3, suggests the absence of eclogite or partial
eclogitization due to delamination under the central-south Tibet; 3) seismicity is
strong or weak in correspondence of the most negative Bouguer gravity anomaly, so
there is not a relationship between them; 4) the composition of the lower crust,
determined after the temperature-pressure calibration of seismic P wave velocity,
might be one or a mixture of: 1. amphibolite and greenschist facies basalt beneath the
Qiangtang terrane; 2. gabbro-norite-troctolite and mafic granulite beneath the Lhasa
terrane. When using the data set published by Rudnick & Fountain (1995), the
composition of the middle crust turns out to be granulite facies and might be pelitic
gneisses. Granulite facies used to be interpreted as residues of partial melting, which
coincides with the previous study by Yang et al. (2002) on partial melting in the
middle crust. Amphibolite facies are thought to be produced after delamination, when
underplating works in the rebound of the lower crust and lithospheric mantle.
From the seismology study, I have made the following conclusions: 1) through
numerical simulation of surface wave propagation in heterogeneous media, we find
that amplitude and polarization of surface wave only change a little when considering
heterogeneity and QL waves, generated by surface wave scattering, are caused by
lateral variation of anisotropy. 2) QL waves have been identified from the
seismograms of selected paths recorded by the Tibetan station CAD, and are utilized
to determine the variation of the uppermost mantle anisotropy of the Tibetan plateau.
The location of the azimuthal anisotropy gradient is estimated from the group
velocities of Rayleigh wave, Love wave and QL wave. We find that a predominant
south-north lateral variation of azimuthal anisotropy is located in correspondence of
the Tanggula mountain, and a predominant east-west lateral variation of azimuthal
anisotropy is found to the north of the Gandese mountain (near 85°E longitude and
30°N latitude) and near the Jinsha river fault (near 85°E longitude and 35°N
latitude).
Insegnamento
Publisher
Università degli studi di Trieste
Languages
en
File(s)