Imaging Seismic Velocity Structure of Crust and Upper
Mantle At Southern Tibet by Receiver Function
G. Jin 1 , Y. J. Chen 1 , E. Sandvol 2, D.C. Wilson 3
Institute of Theoretical and A pplied Geophysics (ITAG), School of Earth and Space Sciences, Peking University, B eijing
Department of G eological Sciences, U. Missouri , Columbia , MO 65211, United States
3 G eological Sciences Dept ., U of Texas at Austin, Austin, TX 78712, United States
E xploring the velocity structure is a key to understand the procession of the uplift of Tibet plateau . Several seismical
experiments have been programmed to explore the crust and mantle structure under Tibet , including Indepth I(1992), Indepth
II(1994-1995), Indepth III(1999) et al. From June , 2004 to September 2005, as a part of HiClimb project , a 2-D survey
including 32 broadband stations are installed to record signal from nature earthquakes . (Fig 1) In this paper, we focus on the
position of discontinuities and velocity structure under southern Tibet using receiver function method .
3 P restack and P oststack MIGRATION
Since the crustal lateral heterogeneities are significant in
this area , we adopt different kinds of methods , including
both pre -stack and post -stack migrations , to explore the
crustal structure. We adopt 2.5D Kirchhoff style migration
[David Wilson and Richard Aster, 2005] as pre -stack
migration , while wave equation migration method [Ling
Chen et al. 2005] as post-stack migration , which
provides a further correction on the result of CCP.
S IY N
Fig1: Study area and distribution of stations, Moho 0
and mantle discontinuities pierce points .
W XSR E
2 DATA AND RECEIVER FUNCTION
From June 2004 to September 2005, all the
earthquakes of which magnitude over 5 and epicenter 4 10
distance between 30° to 90° are involved and finally
Figure 3: Result of west-east
nearly 1000 high qualified receiver function are selected
to process in the following methods . All the receiver crust profiles : W1 and W3. 660
functions are firstly selected automatically due to their Similar structure is moving from
signal to noise ratio , waveform and error in the iterative south -west to north -east.
Strong upper crust discontinuity Figure 4: Result of mantle
deconvolution . After that, a carefully manual selection is profiles shows 410km and
performed to insure the quality of the data . is observed .
660km discontinuities are not
affected by the collision and
subsequence subduction .
4 S lantstack and f-k filter for Moho depth
The "slant stacking" method transforms the receiver
function series from the time domain to the H-Vp/Vs
domain is developed by Zhu and K anamori . In
the stations that the data is reluctant enough, a f-k filter
is processed to improve data quality .
•Two independent middle crust
discontinuities can be observed
south and north of the Yalu-
Zangbu suture zone and there
is also a low velocity layer
north of the Yalu -Zangbu suture
•We observe the lower crust Figure 5: Moho depth
undertrusting south of the Yalu-
Zangbu suture zone , which may
Figure 2. N1 profile : ①: middle crust discontinuity in
the south dipping and disappeared. ②: lower crust
also cause the heterogeneity in
•There is strong Moho doublet in the direction of west to
the in .
east norththe southern Tibet crust, a strong discontinuity in
underthrusting south of the Yalu-Zangbu suture zone .
③: middle crust discontinuity north of the Yalu- the upper crust is observed spreading from SSW to
Zangbu suture zone and the presence of a low
velocity layer . ④: Mohodoublet . •Mantle discontinuities (410km and 660km) are quite flat,
which indicates that the collision of Indian and Eurasia
continent doesn ’t affect deep structure.
•Moho near the north -south rifts seems to be thinner .