Shear wave anisotropy in upper mantle beneath East Antarctica
from broadband seismic data;
- contribution to Antarctic Arrays Deployments / POLENET at
Yusuke USUI1)*, Masaki KANAO2), Atsuki KUBO3)
1) Transdisciplinary Research Integration Center, Research Organization of
Information and Systems, 1-9-10 Kaga, Itabashi-ku, Tokyo, 173-8515, Japan
* Corresponding author: firstname.lastname@example.org
2) National Institute of Polar Research, Tokyo, Japan
3) Kochi University, Kochi, Japan
Shear wave splitting of SKS waves are recorded at 8 stations around Lützow-Holm Bay,
East Antarctica. We calculate apparent splitting parameters φ (fast polarization direction)
and δt (delay time between split waves) assuming anisotropic layer of hexagonal &
horizontal symmetry. As compared with typical delay times of SKS waves having 1.2s, the
result shows generally the same degree. Two layer modeling reveals anisotropy in the
Pan-African terrains in the upper layer corresponds well to polarization of NE-SW paleo
compressional stress in the region. Lower layer anisotropy is supposed to be created by
recent asthenospheric flow, as compared with the absolute plate motion by HS2-NUVEL1
model. These kinds of field broadband stations deployed at several outcrops & continental
ice sheet around the region has significance on coming Antarctic Arrays program, as one
of the contributions in the marginal part of East Antarctica at IPY.
1. Introduction We investigate upper mantle anisotropy using
The analysis of seismic anisotropy has teleseismic SKS wave beneath LHB, and discuss
developed into a powerful tool to know the origin of the anisotropy associated with the
deformations in the Earth's interior. It has been history of the Antarctic plate motion and the effects
known to the existence of seismic anisotropy at of continental collision and/or break-up.
crust, upper mantle, D" layer and inner core. Upper
mantle anisotropy is mainly produced by lattice
2. Data set and measurements
We use broadband seismographs recorded at 8
preferred orientation of highly anisotropic mantle seismic stations from 1996 to 2004 in LHB (Figure
peridotites during deformation processes [Nicolas 1). We selected the events located within the
and Christensen, 1987]. Seismological and epicentral distances of 85° ~ 130° from each
petrophysical studies suggest that most of station. For this distance range, SKS phases are
anisotropy is restricted to the olivine stability field, clearly separated from direct S, Sdiff, and ScS. We
i.e., above 410km depth, in the lithosphere and/or chose the data of the good signal to noise ratio, and
asthenosphere [Mainprice and Silver, 1993]. simple and impulsive source time function which
The studies of the upper mantle anisotropy we could easily found SKS arrivals. In this study,
have been performed using the broadband data we use the data as follows: (1) magnitudes greater
recorded at major permanent seismic stations in than 5.6 (2) the number of events is 115 (3) the
Antarctica. However the results in East Antarctica analyzed waveforms are 205. The events and
are poor. In this study, we focus on around Syowa stations location maps are shown by Figure 1.
station in the Lützow-Holm Bay (LHB), East We calculate the splitting parameter (φ, δt) for
Antarctica. The LHB has been considered the part teleseismic SKS waves using Silver and Chan
of Pan-African orogen during assembly of . φ is fast direction of split shear wave and δt
Gondwana (500Ma). LHB is one of the most is the delay time of two split waves. The splitting
important regions to understand the history of the parameters are determined by minimizing the
continental collision and break-up. energy of the transverse component by net grid
search technique with intervals of 1° and 0.1s,
respectively (Figure 2). The error estimate of each consider that the anisotropy of upper layers is
combination of splitting parameters can be given caused by lithospheric deformation during
by 95% confidence level of F test [Silver and Chan, Pan-African orogen event (~500Ma).
Additionally, we assume more complex models 4. Conclusions
as a two-layer model with four independent We observed upper mantle shear wave splitting
parameters [Silver and Savage, 1994]. The using teleseismic SKS waves for 8 stations in LHB
apparent splitting parameters are fitted by the four East Antarctica. We propose that the anisotropy is
parameters (φ1, δt1, φ2, δt2), with index 1 caused by lattice preferred orientation of mantle
corresponding to the lower layer and index 2 to the minerals along the NE-SW paleo compressional
upper layer. stress for upper layers and asthenospheric mantle
flow for lower layers. The mapping of anisotropic
3. Results and Discussions structures allows the continuation of geological
The weighted average of the apparent splitting structures into the Earth's upper mantle. The
parameter assuming single anisotropy layer are anisotropy is explained as originating from the
shown by Figure 3, where the weights are influence of tectonic events in Pre-Cambrian ages.
inversely proportional to the standard deviations In past studies, these upper mantle structures
for each solution. Although the station STR is no have been only performed at near coast in
clear evidence, we observed two null directions Antarctica. However, in near future, it is thought
associated fast polarization direction. Note that the that the Antarctic Array projects make it possible,
delay times at all stations are the same degree in and we can study the structure of mantle including
comparison to the average value of the results in the inland of Antarctica. This study result is able to
global continental studies, and the fast polarization contribute to resolve the past tectonism and recent
directions are systematically parallel to near coast and/or past mantle flows in Antarctic Plate.
line in the LHB. For 6 stations except PAD and
STR, azimuthal variations of the splitting 5. Figures and tables
parameters do exists. In this case, we modeled
two-layer model of azimuthal anisotropy. From a
geodynamic point of view, since two layers may
correspond to anisotropy in the lithosphere and
asthenosphere, such a model is reasonable.
The results of four parameters are AKR (131°,
0.7s, 43°, 1.7s), LNG (124°, 0.9s, 35°, 1.5s), SKL
(6°, 0.6s, 39°, 0.8s), SKV (49°, 0.1s, 40°, 1.1s),
SYO (145°, 0.4s, 57°, 1.3s), and TOT (133°, 0.9s,
47°, 1.9s). Investigations of seismic anisotropy
may contribute to ideas about influence of recent
or fossil mantle flows and/or the tectonic evolution
of the study regions. Fast polarizations directions
of the lower layer are generally parallel to the
directions of Absolute Plate Motion (APM) based
on the HS2-NUVEL1 [Gripp and Gordon, 1990].
The directions are about N120°E and the velocity
is about 1cm/yr in this study region. The APM
velocities are slow for East Antarctica and the
delay time is small relative to upper layer. We
consider that it is reasonable that the structures of
lower layers anisotropy might have been produced
asthenospheric mantle flow.
The upper layers don’t coincide with the APM
direction (the difference is about 70° ~ 90°). We
should consider the anisotropic structure which is
past tectonic events of East Antarctica. The
direction of Gondwana continent break up was Fig. 1. (top) The global distributions of source and stations.
NW-SE [Nogi et al., 1995]. This is perpendicular Gray circles are source and triangles are stations. Gray lines
to the observed direction. In general, the fast are ray paths. (bottom) Local map with the seismic stations
polarization directions are consistent with NE-SW (triangles) in this study. Gray lines indicate SKS wave path
paleo-compressional stress [Board et al., 2005]. We and black line passing through above 410km.
Acknowledgements: This research was partially
supported by Function and Induction Research
Project, Transdisciplinary Research Integration
Center, Research Organization of Information and
Board, W. S., H. E. Frimmel, and R. A. Armstrong
(2005): Pan-African tectonism in the Western
Maud Beld: P-T-t Path for high-grade gneisses in
the H.U. Sverdrupfjella, East Antarctica, J.
Petrology, 46, 671-699.
Gripp, A. E. and R. Gordon (1990): Current plate
Fig. 2. (a) A example of the SKS waveforms and horizontal velocities relative to the hot spots incorporating
particle motions for both original seismograms (top) and the NUVEL-1 global plate motion model,
reconstructed SKS waveforms by correction of the splitting Geophys. Res. Lett., 7, 1,109-1,112.
parameters (bottom). (b) Contour map of the calculated Mainprice, D. and P. G. Silver (1993): Interpretation
transverse energy as a function of variable splitting of SKS-waves using samples from subcontinental
parameters. lithosphere, Phys. Earth Planet. Inter., 78,
Nicolas, A. and N. I. Christensen (1987): Formation
of anisotropy in upper-mantle peridotites –a
review, in Composition, Structure and Dynamics
of the Lithosphere-Asthenosphere System, eds
Fuchs, K. and Froidevaux, C., AGU Geodyn. Ser.,
Nogi, Y., N. Seama, and N. Isezaki (1995): Magnetic
anomaly lineation and fracture zone deduced
from vector data of the geomagnetic anomaly
field in the West Enderby Basin adjacent to the
Weddell Sea, Geological Society Special
Silver P. G. and M. K. Savage, (1994): The
interpretation of shear-wave splitting in the
presence of two anisotropic layers, Geophys. J.
Int., 119, 949-963.
Silver P. G. and W. W. Chan (1988): Implications for
continental structure and evolution from
anisotropy, Nature, 335, 34-39.
Silver P. G. and W. W. Chan (1991): Shear wave
splitting and subcontinental mantle deformation, J.
Fig. 3. Rose diagrams and weighted average values of the Geophys. Res., 96, 16,429-16,454.
apparent splitting parameters on station locations (triangles).
Bold lines are average fast polarization directions and dash
lines are null directions.