Page 5 - gt1502

116°E 118°E 120°E 122°E 124°E significant proportion of its deep mantle keel (e.g., Griffin et al.,
38°N
36°N
34°N
32°N
30°N
Figure 1. Topography of eastern China and the distribution of seismic stations
(red points). Gray lines represent the boundaries between tectonic units or
major faults; the large white arrow indicates the absolute plate motion (APM)
of the Eurasian plate (Gripp and Gordon, 2002). Blue dots in inset show the
spatial distribution of earthquakes used in this study. EU—Eurasian plate;
JXF—Jiashan-Xianshui fault; NA—North American plate; NCB—North
China block; PA—Pacific plate; PS—Philippine Sea plate; QLDB—Qinling-
Dabie orogenic belts; SCB—South China block; SULU—Sulu orogenic belt;
TLF—Tan-Lu fault; XGF—Xiangfan-Guangji fault; XSF—Xinyang-Shucheng
fault; YQWF—Yintai-Qingdao-Wulian fault. All other abbreviations are
station names.
pressure (UHP) metamorphic rocks (Guo et al., 2012; Yin and
Nie, 1993). The eastward extension of the orogenic belt is thought
to have been offset sinistrally several hundred kilometers by the
Tan-Lu fault following the collision (e.g., Li, 1994). Some workers
believe the collision first occurred in the east during the Early
Triassic and propagated westward (e.g., Guo et al., 2012).
However, the direction of convergence and the location of the
suture remain ambiguous (e.g., Faure et al., 2001) because of the
irregular shape of the northern edge of the SCB and the offset
along the Tan-Lu fault (Yin and Nie, 1993). In the east, a north-
south direction of convergence has been suggested based on linear
aeromagnetic anomalies (Li, 1994). Whether this direction of
convergence occurred on a lithospheric scale or not needs to be
evaluated geophysically.
Following convergence between the SCB and NCB, extensive
cratonic reactivation affected the eastern NCB during the late
Mesozoic, and the thick cratonic lithosphere in this region lost a
1GSA supplemental data item 2015006, detailed methods, data tables, and supplementary figures, is online at www.geosociety.org/pubs/ft2015.htm. You can also
request a copy from GSA Today, P.O. Box 9140, Boulder, CO 80301-9140, USA; gsatoday@geosociety.org.
1998). From the Late Mesozoic through the Cenozoic, deforma-
tion of the eastern Asian continent was dominated by extensional
tectonics leading to the formation of several rift systems (Yin,
2010). The Pacific plate began to subduct along the eastern margin
of the Asian continent at 180 Ma (Maruyama et al., 1997). The
lithospheric deformation in eastern China caused by the conver-
gence between the SCB and NCB was probably modified during
these subsequent tectonic events.
GSA TODAY | www.geosociety.org/gsatoday/
In this study, we present shear wave splitting observations using
data sets from eastern China and explore the “frozen” LPO that may
be associated with the convergence between the SCB and NCB.
DATA, METHODS, AND RESULTS
We use teleseismic shear wave (including SKS, SKKS, and PKS)
splitting measurements to evaluate the lithosphere and upper
mantle deformation beneath this region. The broadband seismic
data used in this study were recorded during August 2007 to April
2013 employing 75 permanent stations in the eastern China.
These stations are widely distributed in the southeastern part of
the NCB and the northeastern part of the SCB. Station locations
are shown in Figure 1 and are listed in Table S1 (see the GSA
Supplemental Data Repository1). In order to observe distinct, high
signal-to-noise ratio shear wave phases, we systematically selected
seismic events with magnitudes (MW) larger than 5.3 occurring at
epicentral distances of 85°–120° and 120°–150° for SKS and PKS
phases, respectively. In these epicentral distances, the SKS or PKS
phase are both well isolated from other shear-waves and are suffi-
ciently energetic. We obtained 215 events that fit these criteria. All
the events are shown in Figure 1 and listed in Table S2 (see foot-
note 1).
We used the measurement method of Tian et al. (2011) to
extract (1) the difference in arrival times (or delay time, �t)
between the fast and slow shear waves, which is a function of the
thickness and intrinsic anisotropy of the anisotropic medium;
and (2) the orientation of the polarization planes of the fast shear
wave (or the fast polarization direction, �), which reflects the
orientation of the structure. The details of splitting measurement
are provided in the supplemental appendix (see footnote 1).
We obtained a total of 1326 pairs of good splitting parameters �
(fast polarization direction) and �t (delay time). Figure 2A presents
the whole set of individual splitting measurements plotted at each
respective station. The station averages computed from the indi-
vidual measurements are plotted in Figure 2B and listed in Table S1
(see footnote 1). At stations CSH, ANQsd and NLA, we obtained a
large number of measurements (56, 44, and 37, respectively) and
large back-azimuth variations. These back-azimuth variations of the
splitting parameters are clearly not random but rather are well orga-
nized. Such back-azimuth variations have been suggested to result
from the presence of two anisotropic layers (Silver and Savage, 1994).
Following the scheme proposed by these authors, we tried to
constrain the possible geometries of these anisotropic layers beneath
the three stations. The best fitting models are plotted in Figure 3 and
listed in Table S3 (see footnote 1).
5