Continental-scale-shear-wave-splitting-analysis--Invest 2005 Earth-and-Plane

Earth and Planetary Science Letters 236 (2005) 106 – 119
www.elsevier.com/locate/epsl
Continental scale shear wave splitting analysis: Investigation
of seismic anisotropy underneath the Australian continent
Maggy Heintz*, Brian L.N. Kennett
Research School of Earth Sciences, Australian National University, Canberra, Act 0200, Australia
Received 27 September 2004; received in revised form 2 May 2005; accepted 2 May 2005
Available online 21 June 2005
Editor: S. King
Abstract
The structure of the upper mantle beneath the Australian continent is investigated using teleseismic shear wave splitting to
extract seismic anisotropy. Measurements have been performed on data recorded at 190 sites with portable broadband seismic
recorders, spanning almost the entire surface of the continent since 1992. The average time span of the various deployments,
primarily designed for surface wave tomography, is 6 months, which is rather limited for shear wave splitting analysis. However,
the data set provides a full continental scale survey using the reasonably favourable distribution of seismicity to Australia. Seismic
anisotropy has the potential to provide insights into the lithospheric structure and the possible mechanical coupling between the
crust and the upper mantle, but prior results for Australia have indicated relatively small levels of splitting and a complex pattern.
These results are confirmed with our new and far more extensive measurements across the whole continent. The pattern of seismic
anisotropy from shear wave splitting beneath Australia is rather complex and is not correlated with the almost north–south absolute
plate motion (APM) from recent models. Deviation of the asthenospheric mantle flow around the lithospheric roots associated with
the extensive Archaean and Proterozoic zones of central and western Australia could be occurring, and so mantle flow-related
anisotropy cannot be completely ruled out. Despite the limited geological outcrop, especially in Phanerozoic eastern Australia, that
is almost entirely covered by sedimentary basins, some relationships can be highlighted between the orientation of the polarization
plane of the fast S-waves and structural trends along, for instance, the Halls Creek orogen bordering the eastern edge of the
Kimberley basin or along the New England and Lachlan fold belts in the southeastern part of the continent. Such relationships
might account for anisotropy frozen in the lithosphere during post-tectonic thermal relaxation. Along the poorly constrained and
controversial Tasman Line (TL), the geological boundary between Precambrian and Phanerozoic Australia, directions of anisotropy
measured at some stations located in the close vicinity of the dlineT appear to exhibit a curvilinear trend somewhat similar to that
of the TL, suggesting that fossil deformation associated with the TL might be recorded in the pattern of seismic anisotropy.
D 2005 Elsevier B.V. All rights reserved.
Keywords: Seismic anisotropy; Lithosphere; Australian continent; Shear wave splitting; Mantle flow; Anisotropy frozen in the lithosphere;
Crust/mantle coupling
* Corresponding author. Tel.: +61 2 6125 0339; fax: +61 2 6125 0738.
E-mail addresses: [email protected] (M. Heintz), [email protected] (B.L.N. Kennett).
0012-821X/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.epsl.2005.05.003
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
1. Introduction
The analysis of seismic anisotropy has been extensively developed over the last decade to emerge as a
powerful tool for studying structure in the upper
mantle (see [1] and [2] for review). The anisotropic
parameters are assumed to be closely related to the
crystallographic fabrics developed in the upper mantle
due to past and present deformation. Comparison of
shear wave polarization anisotropy from core phases
and surface geology can provide insights into the
mechanical coupling between the crust and the
upper mantle.
Seismic anisotropy is a phenomenon analogous to
the crystallographic birefringence of calcite: when a
polarized S-wave propagates across an anisotropic
medium, it is split into two quasi S-waves with
perpendicular polarization that propagate with different velocities. Measuring seismic anisotropy at a
seismological station yields two parameters: /, the
orientation of the polarization plane of the faster Swave and dt, the delay between the arrival times of
the fast and slow split waves. The splitting measurements are commonly performed on core-refracted
shear waves such as SK(K)S or PK(K)S. These
phases are generated from a P-to-S conversion at
the core–mantle boundary (CMB) and are thus polarized along the radial direction as they enter the
mantle. The phases arrive at the station with a nearly
vertical incidence, and anisotropy measured at the
Earth’s surface thus represents a vertically integrated
effect of anisotropy from the CMB to the surface,
with no indication of the depth location of the anisotropy source. Source-side contamination, however,
is avoided. Petrophysical studies suggest that anisotropy is mostly located in the upper mantle, between
the 410 km olivine–spinel phase transition and the
Moho [1–3]. Some small contributions from the DW
layer [4], the lower mantle [5] and the crust [6–8] can
nevertheless not be ruled out.
Olivine is the most abundant and deformable mineral in the upper mantle; it develops a lattice preferred
orientation (LPO) that might result either from the
active deformation of the asthenospheric mantle that
accommodates or causes the absolute plate motion
(mantle flow related anisotropy) [9,10], or was imposed during past deformation, and then was bfrozenQ
in the lithosphere during post-tectonic thermal relax-
107
ation [11–15]. The orientation of the polarization
plane of the fast S-wave, /, is assumed to be a
proxy for the orientation of the [100] axis of olivine
in the upper mantle [16] and therefore provides a
means of investigating upper mantle structure and its
possible mechanical coupling with the crust, by studying the extent to which the measured orientations are
correlated with the superficial geological structures.
The delay time dt is a function of the intrinsic
anisotropy, the thickness of the anisotropic layer, the
orientation of the ray path with respect to the elastic
tensor of the anisotropic medium, and the vertical
coherence of the mantle fabric.
This study reports shear wave splitting observations across the entire Australian continent. Data have
been recorded at 190 sites temporarily equipped with
portable broadband seismic recorders (Fig. 1). For
completeness sake, the permanent stations maintained
by GEOSCOPE and IRIS are also shown on the inset
in Fig. 1. Data recorded at those stations have already
been processed and have been the object of several
papers. The purpose of the present study was therefore
not to have a look at those data again, but rather to
consider data recorded by temporarily deployed stations, covering the entirety of the continent, to see
whether the huge amount of data available would help
solve for the source location of the anisotropy measured at the surface. Except for the stations deployed
within the framework of the Tasman Line project
(white squares, Fig. 1) that will run for 2 yr (2003–
2005), most of the temporary deployments had a time
span between 4 and 8 months, since they were mainly
aimed at recording events to be used for surface and
body waves tomography as well as receiver function
studies. The average time span of 6 months is frequently not long enough to allow the recording of a
number of useful events in the distance range for
teleseismic core-refracted shear waves (i.e. between
858 and 1508) with sufficiently large body wave magnitude (N5.5) and good signal-to-noise ratio. This
study has however been motivated by the large potential of the data (121 useful events have been
recorded since 1993, see Fig. 2), the unusually
dense coverage of a continent in terms of seismological stations, and the lack of any prior continental
scale study of seismic anisotropy in Australia using
body waves. The purpose was to find as much information about the Australian lithosphere as possible,
108
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
110º
115º
120º
125º
130º
135º
140º
145º
150º
155º
-10º
Region #1
PC
SC01
COEN
KA03
KB
SC04
o
SC03
TL01
HC
KL
o
-20º
KB14
TC
SC06
TL03
WRAB
MBWA
PB
CTAO
MIB
AB
TL04
SC08
SA04
AmB
EIDS
SA07
Co
WARB
MB
SB01
SB02
ZB11
YCB
GC
SD09
Ko
WP15-16
Region #2
ARMA
SB05
Cu
AFo
STKA
SB09
YE02
BBOO
NWAO
ZB12
YNG
NEF
B
-30º
CAN
B
LF
FP
YB02
TOO
Tasman Lines
-40º
KIMBA
GA permanent stations
QUOLL
TASMAN LINE
SKIPPY
TIGGER
Permanent stations
WACRATON
Region #3
Hill [1951]
Gunn et al [1997]
TFS
Shaw et al [1996]
Scheibner [1974]
TAU
S&V [1984]
Fig. 1. Location of the 190 instrumented sites. Symbols represent the deployments they are related to. The names of the stations for which
results are discussed in the text are indicated. The dashed line represents the Tasman Line as suggested by [22]. The main geological provinces
mentioned in the text are located (AB=Arunta block; AFo=Albany Fraser orogen; AmB=Amadeus Basin; Co=Capricorn orogen; Cu=Curnamona block; FP=Fleurieu Peninsula; GC=Gawler craton; HCo=Halls Creek orogen; KB=Kimberley basin; KLo=King Leopold orogen;
Ko=Kimban orogen; LFB=Lachlan fold belt; MB=Musgrave block; MIB=Mount Isa block; NEFB=New England fold belt; PB=Pilbara
block; PC=Pine Creek; TC=Tennant Creek; TFS=Tamar Fracture System; YCB=Yilgarn cratonic block).
and to complement the results previously obtained by
[17], which mainly deal with the eastern part of the
continent.
2. Region under study
The Australian continent has experienced a complex
tectonic evolution over the past 4.6 Gy that may be
expected to have left some imprints on the lithospheric
mantle (see [18,19] for review). The tectonic evolution
of the various Precambrian cratonic blocks has been
influenced by a combination of lateral lithospheric
deformation, and vertical lithospheric accretion, perhaps by mantle plumes. The three main cratonic blocks
(North, West and South Australian cratons) formed at
ca. 1830 My and resulted from the collision between
individual blocks. The Pilbara and Yilgarn cratons
joined along the Capricorn orogen during the early
Proterozoic to form the West Australian craton. The
assemblage of the proto-Gawler and proto-Curnamona
blocks along the Kimban orogen shaped the South
Australian craton. Various small Precambrian blocks
such as the Kimberley, Pine Creek and Tennant Creek
inliers collided along sutures like the King Leopold and
Halls Creek orogens, and resulted in the formation of
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
Fig. 2. Location of the 121 events used in this study; the distance
range is 85–1508 and the magnitude is greater than 5.5.
the North Australian craton (see Fig. 1 for the location
of the various geological units). Active tectonics between 1300 and 1100 My led to the amalgamation of
Proterozoic Australia. The North, South and West Australian cratons, previously part of large continental
landmasses that may have respectively included
North America, East Antarctica and Greater India,
ended up shaping an early component of the Rodinia
supercontinent. The West and North cratonic blocks
accreted and then collided with the South Australian
craton along the Albany Fraser orogen. The Centralian
Superbasin formed between 830 and 750 My over the
junction of the three cratons, and was followed by the
breakup of Rodinia at ca. 750 My.
The Palaeozoic belts located in the eastern part of
the continent (e.g. the New England and Lachlan fold
belts, see Fig. 1) developed mainly in a backarc
environment with transient episodes of accretionary
tectonism related to west-dipping subduction. Passive
margins preserved on the east, west and south coasts
of the Australian continent formed as a result of
Gondwana breakup.
The boundary between Precambrian and Phanerozoic Australia has been interpreted to follow the
bTasman LineQ (TL), whose definition has been primarily based on the separation of geological outcrops
of Precambrian and younger basement [20], and, to a
109
lesser extent, on gravity and magnetic anomalies. This
concept has been extensively discussed and argued,
with various authors suggesting different positions
for the TL (see [21] for review and Fig. 1 for location).
The recent work of [21] however concluded that even if
based on geophysical data and geological lineaments
believed to define a given age and tectonic origin, the
TL, as defined by many authors (e.g. [22–25]), may in
fact result from a number of sources that vary in age,
protolith, and degree of deformation/metamorphism. In
this view there is, therefore, no basis for the interpretation of a coherent supracrustal Tasman dLineT.
The contrast between eastern and western Australia
is however the most striking result in surface wave
tomographic studies (see [26, 27] for the latest results,
but also [28, 29]) and is present from at least 75 to 200
km depth: a very strong contrast in seismic shear
wavespeed is indeed imaged beneath the eastern and
western parts of the continent. Shear wavespeeds faster
than the continental average extend to at least 200 km in
the cratonic zone to the west of 1408 E. There is no clear
relationship between the eastern extension of this
anomaly and any of the various locations of the Tasman
Line (e.g. [22–25]): it is in close agreement with the
original TL defined by [20] in northern Queensland
down to 100 km depth (see Fig. 1 for location), and all
over Queensland from 100 to 150 km depth. At 150 km
depth however, the NE–SW trend along the edge of the
Curnamona province is in closer agreement with more
recent suggestions [22], [24], the southernmost connection of the TL to the Tamar Fracture system in
Tasmania is markedly different between different
authors. Below 150 km depth, the boundary is mainly
oriented NS along 1408 E, and does not correlate with
any definition of the TL. Around 258 S however, a
small indention within the shape of the anomaly might
be related to the lines suggested by [22] or [24].
The results of surface wave tomography for mantle
structure are not simply related to the suggested location of the TL from crustal information; the contrast
between western Precambrian and eastern Phanerozoic Australia is nonetheless undisputed.
3. Method
Shear wave splitting measurements have been performed using the Silver and Chan algorithm [30],
110
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
which consists of minimizing the energy on the transverse component by rotating and time shifting the
traces. The energy on the transverse component is
automatically evaluated for many candidate values
of / and dt (with respective increments of 18 and
0.05 s) to retrieve the pair of values that best removes
influence of anisotropy. This method assumes that
anisotropy is located in a single horizontal layer.
Individual results have been sorted in 4 categories
(good, fair, poor and null) following [14] and are
based on: (1) the quality of the initial signal (signalto-noise ratio and interferences with other phases), (2)
the ellipticity of the particle motion in the horizontal
plane when anisotropy is present, (3) the linearization
of the particle motion by anisotropy removal, (4) and
the waveform coherence between the fast and slow
split shear waves. Measurements satisfying the four
criteria, producing similar pulse shapes and linear
particle motion after correction along with fairly
small error ellipses were rated as bgoodQ, while
those meeting only three criteria were rated as bfairQ.
A poor measurement only fulfills two criteria and a
null measurement does not show any energy on the
transverse component associated with the arrival of
the core phase of interest on the radial component; this
may be due either to an absence of anisotropy or to an
initial polarization of the incoming shear wave parallel
or orthogonal to the fast anisotropic direction.
The quality of the data has been visually inspected
and only traces showing sharp arrivals of the core
phases, very distinct from the surrounding noise,
have been kept. The influence of time windowing
and filtering has systematically been checked. When
used, the filter was a Butterworth pass-band with
lower limit equal to 0.03 Hz and upper limit varying
from 0.25 to 0.75 Hz.
To increase the robustness of the results compared to
the analysis of single events, we have used the method
of [31] (hereafter referred to as WS) to compute global
solutions. The WS method consists of simultaneously
processing multiple events, at a single station, by normalizing and summing individual error surfaces. For
those stations showing either no correlation between
backazimuthal variations or splitting parameters that
are more complex than the splitting predictions for a
single anisotropic layer with a vertical symmetry axis,
global solutions have been computed. In order to assess
the validity of these results, we also used the same data
to compute global solutions using the cross-convolution method of [32] (ML in the following). Unlike the
WS method which defines individual errors on both
anisotropic parameters, the estimate of the errors in the
ML method is performed through the use of a misfit
parameter R based on the match between the transverse
and radial components after correction (see [32] for
details). Smaller R indicates better solutions. Only
those results that are coherent between the two methods
have been plotted in Fig. 3b as grey bold lines and listed
in Table 1.
4. Observations and discussion
Surface wave tomographic studies including azimuthal anisotropy suggest a two-layer system of anisotropy beneath Australia [33–36]: in the upper layer,
directions of anisotropy are approximately oriented
east–west in [33,34], and more or less randomly in
[35], whereas in the bottom layer, directions of anisotropy appear to be north–south, parallel to the absolute
plate motion (APM) in each model.
The main challenge in studying core-refracted
shear waves is the lack of vertical resolution. The
anisotropy measured at the surface has been acquired
on the way from the CMB to the surface; the splitting
parameters therefore represent an integrated measurement and we need to consider whether a lithospheric
or an asthenospheric source, or a combination of both,
may account for the observed anisotropy.
4.1. Asthenospheric source of anisotropy
In the study case of a decoupling between a bflatQ
lithospheric plate and the underlying upper mantle,
simple asthenospheric flow would lead to splitting
parameters very coherent over large geological
domains that differ in both age and structure: the
orientation of the polarization plane of the fast Swave would be parallel to the APM, and the amplitude of the delay time would be rather constant in
each domain. At the Australian continental scale
however, considering the results as a whole, the
directions of anisotropy obtained from shear wave
splitting appear to be very scattered. Previous evidence accounting for an asthenospheric contribution
comes from observation of shear wave splitting for
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
WRAB
111
CTAO
-10∞
Region#1
a)
NWAO
CAN
Ozalaybey and Chen [1999]
Vinnik et al. [1992]
Barruol and Hoffmann [1999]
TAU
-30º
Region#2
Tasman line (Scheibner et al., 1974)
Absolute plate motion (HS2-Nuvel1, Gripp and Gordon, 1990)
Absolute plate motion (HS3-Nuvel1A, Gripp and Gordon, 2002)
Region#3
Absolute plate motion (T22A, Wand and Wand, 2001)
-40º
Faults
Good measurement
1s
Fair measurement
110º
120º
130º
140º
150º
-10º
Region#1
b)
-20º
-30º
Region#2
Region#3
Null measurement
-40º
1s
110º
Global solution
120º
130º
140º
150º
Fig. 3. Measured directions of the orientation plane of the fast shear-wave. (a) Good and fair measurements. The length of each line is
proportional to the delay time. (b) Null measurements associated with the results of the multiple-events method (grey bold lines, with the length
of the lines proportional to the delay time. When a slight difference exists between the global solutions computed either with the WS or the ML
method, we represent the results issued from the WS method). Crosses denote the absence of splitting: each branch is either parallel or
perpendicular to the backazimuth of the incoming waves and represents a possible direction of anisotropy. The dashed line represents the
Tasman Line as defined by [22]. Light grey arrows represent the APM defined by the HS2-Nuvel1 model [37], black arrows represent the APM
calculated from the HS3-Nuvel1A model [37] and dark grey arrows the APM calculated from the T22A model [38]. The inset displays
measurements obtained at the permanent stations from various studies: [46] (black lines show backazimuths of null shear wave splitting), [9]
(grey lines), [42] (inset for Geoscope station CAN: left diagram represents the azimuth of each fast split shear wave for one good (black) and one
fair (grey) measurement, while the right diagram represents the null directions).
112
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
Table 1
Multiple event parameters calculated for those stations showing no important variations in the splitting parameters with respect to the
backazimuth
Station
ARMA
KA03
KB14
SB09
STK
TL01
TL03
TL04
WP15–16
ZB11
Lat (8)
30.42
15.93
18.19
32.28
31.88
18.00
20.15
21.68
31.21
28.85
Lon (8)
151.63
127.25
127.82
146.85
141.60
141.64
139.41
140.97
121.09
152.25
/ (8)
WS method
Error (F8)
dt (s)
Error (s)
ML method
/ (8)
dt (s)
R
Nb of
phases
162.00
64.00
48.00
8.00
53.00
24.00
117.00
7.00
64.00
182.00
12.50
8.50
4.00
3.00
7.50
4.00
22.50
6.50
10.50
4.00
0.65
0.84
1.60
1.12
0.45
0.64
0.16
0.52
0.60
1.04
0.18
0.18
0.26
0.14
0.10
0.14
0.18
0.08
0.12
0.12
147.00
64.00
42.00
7.00
52.00
27.00
101.00
10.00
63.00
179.00
0.80
0.88
1.24
1.16
0.45
0.72
0.16
0.52
0.68
1.04
0.24
0.11
0.47
0.25
0.12
0.27
0.15
1.02
0.25
0.55
3
2
4
3
3
3
4
2
4
5
Multiple event parameters have been calculated using both the Wolfe and Silver [31] and Menke and Levin [32] methods. R is a misfit
parameter issued from the ML method and based on the match between the transverse and radial components after correction: the smaller the R,
the better the solution.
refracted S-waves beneath northern Australia [37]:
the authors explained the observed discrepancy between measurements resulting from waves travelling
through the low velocity zone, the transition zone,
and through the top of the lower mantle in terms of
transverse isotropy located within the low velocity
zone under the unusually thick mantle lid beneath
north Australia. In the present study, we cannot
highlight any simple relationship between the orientations of the polarization plane of the fast S-wave
and the APM of Australia determined either from the
HS2-Nuvel1 [38], HS3-Nuvel1A [39] or T22A [40]
models (arrows on Fig. 3). Following one school of
thought [9,10], if APM is nevertheless envisioned as
the main source of anisotropy, we have to consider
rather complex deviations of the mantle flow around
the cratonic keels of the various continental domains
to reconcile the scattered directions of anisotropy
measured at the surface and the APM. Results
from surface wave tomography of the Australian
continent suggest that the topography of the lithosphere–asthenosphere boundary may well be rather
complex [28,29]. This could account for such deviations of the mantle flow, and an asthenospheric
contribution is therefore not ruled out.
4.2. Lithospheric source of anisotropy
Due to the limited amount of reliable measurements performed at each station, related to the short
time span of recording, together with the sedimentary
cover overlying the Palaeozoic basement almost in the
entire eastern part of the continent, no direct correlation is observed between the measured orientation of
the polarization plane of the fast S-waves and the
mapped superficial structures. It is therefore impossible to assert that anisotropy frozen in the lithosphere is
the main source of anisotropy.
With a closer look at the data, some correlations
can however be seen, especially (1) in the Kimberley
region, (2) in the southern part of the Yilgarn cratonic
block, and (3) along the southeastern margin of the
continent (see Fig. 1 for location).
In the Kimberley block (KB, Fig. 1), the directions
of anisotropy measured within the framework of the
KIMBA experiment (July–Oct. 1997 then May–Oct.
1998, white diamonds on Fig. 1) show various orientations. We can divide the results into three different
groups with respect to the surface geology, from east
to west: the Halls Creek orogen (HCo, Fig. 1), the
Kimberley basin, and the King Leopold orogen (KLo,
Fig. 1).
In the eastern part, directions of anisotropy are
mainly oriented N 50–608 E and are sub-parallel to
the structural trend of the Halls Creek orogen. In the
western part however, the measured directions of
anisotropy are ~N 908 E and not obviously correlated with the NW–SE structural trend of the King
Leopold orogen (Fig. 4). Within the Kimberley
basin and along the boundaries between both the
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
113
-11º
1s
a)
Good measurement
-12º
Fair measurement
-13º
SC01
-14º
-15º
KA01
KB08
KB04
KA03
-16º
KB03
KA02
KB07
KA05
SC03
KA10
-17º
KB06
KB09
KA09
KB11
KB12
-18º
KB14
KA07
KA08
-19º
SC06
-20º
120º
121º
122º
123º
124º
125º
126º
127º
128º
129º
130º
131º
132º
-11º
b)
-12º
-13º
SC01
-14º
-15º
KA01
KB08
KB04
KA03
-16º
KB03
KA02
KB07
KA05
SC03
KA10
KA09
-17º
KB06
KB09
KB11
KB12
-18º
KB14
KA07
KA08
-19º
SC06
-20º
120º
121º
122º
123º
124º
125º
126º
127º
128º
129º
130º
131º
132º
Fig. 4. Zoom on region 1 (see Fig. 3 for location) superimposed on the geological map of the Kimberley region (obtained from the Geological
Survey of Western Australia website). Null and fair measurements are respectively represented as black crosses and grey lines. Stations whose
results are discussed in the text are indicated. (a) Good and fair measurements. (b) Null measurements.
114
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
fold belts and the basin, measurements performed
highlight null results. Results are coherent between
the stations in this region and if anisotropy is
detected, the fast anisotropic direction is either oriented N 308 E or N 1208 E, approximately parallel
to the network of faults mapped within the basin
(Fig. 4). At those stations located within the basin
and along the King Leopold orogen however, only
one or two measurements have been considered as
reliable per station; it may therefore have little physical meaning and does not allow any straightforward
interpretation. Along the Halls Creek fold belt,
results obtained at station KB14 are coherent: 4
measurements have been performed on 3 events
with different backazimuths (120.9–194.38), leading
to a well-constrained global solution (/ = 48 F 48,
dt = 1.6 F 0.26 s with the WS method, and / = 428,
dt = 1.24 s, R = 0.47 with the ML method). Measurements performed at station SC06 confirm those
obtained by [17] and highlight a direction of anisotropy (208 b / b 388) that coincides with the NNE
trend of the major shear zones that separate the
Kimberley from the central Arunta block [17]. The
results obtained at SC01 are also in good agreement
with those performed by [17]: one good measurement led to / = 18 F 14.58 and dt = 0.28 F 0.08 s.
Unlike [17] however, no reliable measurements
have been kept at station SC03, except three very
coherent null measurements showing either the absence of anisotropy along the ray path or two possible orientations of the polarization plane of the fast
S-wave: N 648 E or N 1548 E.
In the southern part of the Yilgarn cratonic block,
two main directions of anisotropy are observed on
both sides of 308 S latitude (Fig. 5). The N 40/508 W
directions of anisotropy observed north of this line
are parallel to some regional faults (Fig. 5a). South of
308 S latitude, a set of 10 stations recorded the same
event (day 00341, latitude 39.578, longitude 54.808,
depth 30 km, magnitude 7.5). The coherence between
the individual results per station is particularly striking (Fig. 5b) because it extends over 600 km, whereas the pattern of anisotropy from shear wave splitting
is usually expected to vary in continental areas that
experienced long and complex history such as Precambrian platforms. All stations indeed exhibit a null
result which corresponds either to an absence of
anisotropy beneath the stations or to the polarization
plane of the fast S-wave oriented N 45/508 E or N 40/
458 W (Fig. 5c). A N 40/458 W orientation of the
polarization plane of the fast S-wave would approximately match the global NW–SE trend of the faults
in the southwestern part of the Yilgarn craton; however, despite the good consistency of the results at the
deployment scale, no firm interpretation can be envisaged on the basis of a single measurement per
station.
In the southeast of the continent, stations ZB11,
ARMA, CNB and YB02 (see Fig. 1 for location)
highlight directions of anisotropy almost parallel to
the structural trend of the New England and Lachlan
fold belts (Fig. 3); i.e. NNE–SSW within the New
England fold belt (stations ZB11 and ARMA) and
more NE–SW along the Lachlan fold belt (stations
CNB and YB02). These measurements may account
for anisotropy frozen in the lithosphere during the
post-tectonic thermal relaxation related to the formation of the Palaeozoic Lachlan and New England fold
belts. The results obtained at station ZB11 moreover
confirm those previously measured by [17], accounting for a global solution of / = 182 F 48 and
dt = 1.04 F 0.12 s with the WS method, and
/ = 1798, dt = 1.04 s and R = 0.55 with the ML method.
CNB ( 35.3158; 149.3638) is a permanent station
deployed and maintained by Geoscience Australia.
Data processed at this station however correspond to
a time span of recording of only 6 months, for the
sake of comparison with the other data presented in
this study. Good measurements performed at CNB
(Fig. 6) on 3 core-refracted phases related to 2 events
with different backazimuths (6.08 and 129.58) give
consistent directions of / (N 558 E, N 228 E, N 278
E), whereas the same events recorded at the nearby
(b 40 km) permanent GEOSCOPE station CAN
( 35.3218; 148.9998) were of rather poor quality
and did not allow us to perform any measurement.
More strikingly, studies taking into account several
years of data recorded at station CAN reveal a well
constrained apparent isotropy for SKS and similar
phases (see [41–43] and inset in Fig. 3). The lateral
resolution reached through studying core phases arriving at a station with a near vertical incidence is
around 30 km. The inconsistency in terms of splitting
parameters observed at CAN and CNB stations is
therefore very interesting. No conclusion can be highlighted in view of only three good measurements,
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
a)
b)
-27º
-27º
1s
WS01
WS01
WP04
-28º
WP02
WP04
WV07
SE04
WP03
WP01
-28º
WV04
WP06
WV05
WR07
WV02
SE04
WP03
WV01
WV06
WV02
WP08
WP11
WP18
-30º
SE06
WP13
WP12
SE05
WP18
-30º
WT10
SE06
WP13
WR08
WP25
WT09
-31º
WT08
WP15-16
WP14
WR09
WP14
SE07
WR10
WR10
-33º
118º
119º
X 10+4
X 10+4
X 10+4
X 10+4
2
Lag (s)
3
1440 1450 1460 1470 1480 1490 1500 1510 1520 1530
WR10
0
1
Seconds
SKSac
2
Lag (s)
3
4
SKKSac
ScS
sSKSac
S
pSKSac
X 10+4
-50
X 10+4
0
5
0
-5
-10
X 10+4
Azimuth (degrees)
50
5
0
-5
-10
5
0
-5
-10
5
0
-5
-10
4
50
0
SKSac
0
1
2
Lag (s)
3
1430 1440 1450 1460 1470 1480 1490 1500 1510 1520
Seconds
SKSac
0
1
2
3
4
2
3
4
Lag (s)
SKKSac ScS
pSKSac S
50
0
-50
4
0
1430 1440 1450 1460 1470 1480 1490 1500 1510 1520
Seconds
WR09
0
0
-50
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
SKKSac
ScS
sSKSac
S
pSKSac
50
sSKSac
SKKSac
pSKSac ScS pS sS
S
1420 1430 1440 1450 1460 1470 1480 1490 1500 1510 1520
Seconds sSKSac
-50
WT10
124º
50
WT08
1430 1440 1450 1460 1470 1480 1490 1500 1510 1520
Seconds
SKKSac
ScS
sSKSac
S
pSKSac
3
X 10+4
10
5
0
-5
4
X 10+4
SKSac
1
123º
1
0
SKSac
1
0
-1
-1
Lag (s)
50
-1
-1
1
SKKSac
sSKSac
ScS
S
pSKSac
Azimuth (degrees)
0
2
Lag (s)
X 10+4
10
5
0
-5
1
SKKSac S ScS pS
pSKSac
X 10+4
X 10+4
10
5
0
-5
SKSac
Azimuth (degrees)
X 10+4
10
5
0
-5
X 10+4
WT06
0
0
1420 1430 1440 1450 1460 1470 1480 1490 1500 1510 1520
Seconds sSKSac
X 10+4
4
X 10+4
3
SKSac
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
X 10+4
2
Lag (s)
-50
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
1
0
0
-50
0
1
2
Lag (s)
3
4
-50
1
0
1400 1410 1420 1430 1440 1450 1460 1470 1480 1490
Seconds
0
1
2
Lag (s)
3
4
50
Azimuth (degrees)
2
1
0
-1
-2
2
1
0
-1
-2
2
1
0
-1
-2
2
1
0
-1
-2
1
Azimuth (degrees)
0
sSKSac
ScS
SKKSac
S
pSKSac
122º
Azimuth (degrees)
X 10+4
0
-50
X 10+4
SKSac
121º
Azimuth (degrees)
X 10+4
WT04
50
10
5
0
-5
120º
SKIPPY5
sSKSac
SKKSac
pSKSac ScS pS
S
SKSac
10
5
0
-5
Azimuth (degrees)
X 10+4
X 10+4
X 10+4
X 10+4
X 10+4
X 10+4
117º
WACRATON2
Azimuth (degrees)
0
1420 1430 1440 1450 1460 1470 1480 1490 1500 1510
Seconds
X 10+4
116º
10
5
0
-5
50
WT09
X 10+4
124º
X 10+4
Azimuth (degrees)
X 10+4
X 10+4
X 10+4
X 10+4
WT03
1380 1390 1400 1410 1420 1430 1440 1450 1460 1470
Seconds
X 10+4
123º
10
5
0
-5
-50
WT05
X 10+4
122º
50
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
10
5
0
-5
-10
X 10+4
121º
sSKSac
SKKSac
pSKSac ScS pS
S
SKSac
2
1
0
X 10+4
120º
WACRATON1
c)
WR08
119º
X 10+4
118º
X 10+4
117º
X 10+4
116º
2
1
0
-1
WR09
WT03
SE07
-1
WP17
WT04
WT02
-32º
-33º
2
1
0
-1
WP25
WP15-16
WT06
WT05
WT03
WT02
-32º
WT08
WP17
WT06
WT05
WT10
WT09
-31º
WT04
WV06
WV05
WV03
WP08
WP11
WP12
WR08
WV04
WP06
-29º
SE05
WV07
WR07
WV01
WP02
WP01
WV03
-29º
2
1
0
-1
115
0
-50
1390 1400 1410 1420 1430 1440 1450 1460 1470 1480
Seconds
0
1
2
Lag (s)
3
4
Event 2000.341
Distance ~ 95º
Magnitude 7.5
Fig. 5. Zoom on region 2 (see Fig. 3 for location) superimposed on the geological map of the southern part of the Yilgarn craton (obtained from
the Geological Survey of Western Australia website). (a) Good (black lines) and fair (grey lines) measurements. (b) Null measurements. The
dashed line represents the 308 S parallel. (c) Example of the consistency between the different null measurements obtained for the same event
recorded at different stations within the framework of the WACRATON deployment.
because splitting parameters measured for a single, or
a small number of phases, may have very little physical meaning and can be correctly interpreted only in
the context of a large number of measurements cov-
ering a long time span and/or a wide range of backazimuths. It seems however that we may have a
meaningful change of elastic properties in the vicinity
of Canberra, on a scale of a few tens of kilometres.
116
40
20
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
SKSac
a)
SKKSac
S ScS
1.0
1.0
0.5
0.5
Phi = 55 +/- 6.5º
Dt = 0.5 +/- 0.05s
50
0
0.0
0.0
20
-0.5
-0.5
840
0
-20
40
842
844
846
840
842
844
846
20
0
0
Event 2003.167
Dist. 90.9º
Az. 188.7º
Baz. 6.0º
Depth 174.0km
5
5
-20
40
Azimuth (degrees)
-20
40
-50
0
0
20
0
-20
810
820
830
840
850
860
870
880
0
5
-5
0
1
5
2
Lag (s)
3
4
Phi = 22 +/- 2.5º
Dt = 0.75 +/- 0.10s
SKKSac
b)
0.5
0.5
0
-10
20
10
0
0.0
0.0
-0.5
-0.5
-1.0
-1.0
-10
20
50
Azimuth (degrees)
10
0
-5
Seconds
20
-5
-5
800
676
678
680
676
678
680
10
0
10
10
0
0
-10
-10
0
Event 2003.171
Dist. 122.4º
Az. 219.5º
Baz. 129.5º
Depth 558.0km
-10
20
-50
10
0
-10
640
650
660
670
680
690
700
710
-20
-20
720
-10
0
10
-20
-20
0
-10
0
1
10
2
Lag (s)
3
4
Seconds
10
Phi = 27 +/- 3º
Dt = 0.95 +/- 0.15s
SKSac SKSdf
c)
0
-10
20
10
0
0.5
0.5
0.0
0.0
-0.5
-0.5
-1.0
-10
20
50
-1.0
582
584
582
586
584
10
0
5
5
586
Azimuth (degrees)
20
0
Event 2003.171
Dist. 122.4º
Az. 219.5º
Baz. 129.5º
Depth 558.0km
-10
20
-50
0
0
10
0
-5
-5
-10
0
550
560
570
580
590
600
610
620
-10
-5
0
5
-10
-5
0
5
1
2
Lag (s)
3
4
Seconds
Fig. 6. Shear wave splitting measurements performed at station CNB on two different events highlighting good (a) and very good (b and c)
results. Studies taking into account several years of data recorded at the nearby (b40 km) permanent GEOSCOPE station CAN conclude a net
isotropic behaviour.
Similar features have been observed at the portable
Kimberley array in South Africa, significant changes
in shear wave splitting being observed over scales of
b35 km [44]: such small-scale variations might therefore represent a more common situation than is currently appreciated. On the basis of this anomaly, a
third long-term station will be installed beginning of
2005 in the same region, in order to study this peculiar
local feature of anisotropy and see whether it is well
constrained or just a question of probability related to
the backazimuths of the events recorded during a 6
month period.
In their study, [17] reported directions of anisotropy in eastern Australia that exhibit a curvilinear
trend somewhat similar to the poorly constrained and
controversial bTasman LineQ (TL), delimiting the
Proterozoic shields of central Australia and the Phanerozoic regions of the Tasman fold belt in eastern
Australia. To some extent, we agree with the statement of [17], especially since we measured additional directions of anisotropy at stations SB05, SB01,
SA04 and TL04 (see Fig. 3), located in the close
vicinity of the proposed TL as defined by [22],
directions that are locally parallel to the trend of
the line.
The TL and the Trans European Suture Zone
(TESZ) in Europe both represent geological features
separating Proterozoic from Phanerozoic terranes; we
could therefore have expected to observe a behaviour
of the seismic waves along the TL similar to the one
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
reported on both sides of the TESZ. Along the Sorgenfrei–Tornquist Zone (STZ), the northwestern part
of the TESZ, results of SKS shear wave splitting have
shown that azimuthal anisotropy directions seem to
follow the trend of the STZ [45]. These observations
are consistent with alignment of mantle rocks parallel
to the trend of the STZ which may have taken place
during large-scale transcurrent motion along the STZ
and that were subsequently frozen in the lithosphere.
In our study, we however do not observe such a
parallelism between the orientation of the polarization
plane of the fast S-waves and the structural trend of
the Tasman Line.
The Tasman Line project is designed to be operational until at least April 2005, and only 1 yr of data
has yet been analyzed. We therefore expect to record
more useful data to be able to investigate whether the
Tasman Line could be analogous to the TESZ in
Europe, and to determine the role it played during
the formation of the Australian continent. The preliminary results seem encouraging. The recent review of
[21] suggests however that previously proposed Tasman blinesQ, even if based on geophysical data, might
result from the association of lineaments or structural
units of different ages that experienced different deformation and metamorphism, and therefore do not
form a dlineT sensu stricto, as is the case in northern
Europe.
Another interesting feature to report is common
discrepancies between measurements performed on
SKS and SKKS phases at stations SB01 and SB05.
The ray paths followed by SKS and SKKS phases are
different in the lower mantle and DW layer, and almost
superimposed in the upper mantle. This kind of discrepancy might therefore account for the potential
location of anisotropy within the lower mantle and/
or the DW layer.
Finally, we report stations showing clear evidence
of isotropy. Measurements performed on various
events covering a reasonable range of backazimuths
at stations SC08 (backazimuth of 54.48 and 162.88),
SD09 (101.88s b baz. b 148.58), ZB12 (89.68 b baz.
b 132.88) and YE02 (87.78 b baz. b 143.78) exhibit
null results accounting for an absence of anisotropy
underneath those stations. Looking at the geographical distribution of the stations (Fig. 1), it appears
that stations YE02 and ZB12 are located in the
vicinity of the isotropic permanent CAN station.
117
The clear and consistent results obtained at station
CNB and accounting for a direction of the polarization plane of the fast S-wave oriented ~308, parallel
to the structural trend of the Lachlan fold belt, are
therefore even more intriguing and definitely require
further investigation.
5. Discussion and conclusions
We retrieved shear wave splitting parameters at
118 of 190 sites temporarily equipped with broadband seismological stations deployed all over the
Australian continent since 1992. Despite the impressive amount of stations covering the entire continent
together with the large potential of seismicity to be
recorded in Australia, the short time span of recording of most of the stations (average of 6 months)
together with the sedimentary cover preventing outcrop in a large part of the Phanerozoic eastern
Australia, did not allow us to highlight any clear
and coherent pattern of anisotropy from shear wave
splitting at the continental scale. However, due to the
complex tectonic evolution of the Australian continent that begun 4.6 Gy ago, together with the presence of large cratonic blocks deeply rooted in the
upper mantle, the lithosphere–asthenosphere boundary might well be rather complex, and mantle-flow
related anisotropy can therefore not be completely
ruled out: deviation of the mantle flow around the
cratonic keels, leading to a complex pattern of anisotropy at the surface, might still be envisioned.
This idea is supported by the results of surface
wave tomography that may suggest a rather complex
lithosphere–asthenosphere boundary.
At a regional scale, some evidence of fast split Swaves being polarized in planes oriented parallel to
the local structural trends (southeastern coast and
along the Halls Creek orogen, for instance) may account for anisotropy frozen in the lithosphere during
post-tectonic thermal relaxation.
It seems therefore difficult to interpret the complex pattern of anisotropy from shear wave splitting
beneath Australia in terms of either mantle-flow
related anisotropy or anisotropy frozen in the lithosphere: a contribution from both the lithospheric and
sublithospheric mantle is likely. The duration of
most of the deployments is not sufficient for us to
118
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
be able to pin down the contribution from each
source of anisotropy by, for instance, performing
2-layer modelling of the anisotropy. The current
Tasman Line project is however designed to last
for at least 2 yr (2003–2005) and aims at recording
events simultaneously on both sides of the Tasman
Line, the controversial boundary between the Precambrian western and Phanerozoic eastern Australia.
Only 1 yr of data has been included in the present
study. The forthcoming data should allow us to
build a reliable database on both sides of the Tasman Line and to better constrain the preliminary
results, which appear to exhibit a curvilinear trend
sub-parallel to the structural trend of the Tasman
Line and therefore suggestive of fossil anisotropy
frozen in the lithosphere.
Acknowledgements
The collection of data in the field has relied on
many members of RSES. We are grateful to all the
people who took part in fieldwork over the years, and
to Armando Arcidiaco for data handling. We are
indebted to Martha Savage and two anonymous
reviewers for their helpful comments.
References
[1] P.G. Silver, Seismic anisotropy beneath the continents: probing
the depths of Geology, Annu. Rev. Earth Planet. Sci. 24 (1996)
385 – 432.
[2] M.K. Savage, Seismic anisotropy and mantle deformation:
what have we learned from shear wave splitting? Rev. Geophys. 37 (1) (1999) 65 – 106.
[3] D. Mainprice, P.G. Silver, Interpretation of SKS-waves using
samples from the subcontinental lithosphere, Phys. Earth
Planet. Inter. 78 (1993) 257 – 280.
[4] J.M. Kendall, P.G. Silver, Investigating causes of DW anisotropy, Geodyn. Ser. 28 (1998) 97 – 118.
[5] J. Wookey, J.M. Kendall, G. Barruol, Mid-mantle deformation
inferred from seismic anisotropy, Nature 415 (6873) (2002)
777 – 780.
[6] D.E. McNamara, T.J. Owens, P.G. Silver, F.T. Wu, Shear wave
anisotropy beneath the Tibetan Plateau, J. Geophys. Res. 99
(B7) (1994) 13655 – 13665.
[7] G. Herquel, G. Wittlinger, J. Guilbert, Anisotropy and
crustal thickness of Northern-Tibet. New constraints for
tectonic modelling, Geophys. Res. Lett. 22 (14) (1995)
1925 – 1928.
[8] G. Barruol, D. Mainprice, A quantitative evaluation of the
contribution of crustal rocks to the shear wave splitting of
teleseismic SKS waves, Phys. Earth Planet. Inter. 78 (3–4)
(1993) 281 – 300.
[9] L.P. Vinnik, L.I. Makeyeva, A. Milev, A.Y. Usenko,
Global patterns of azimuthal anisotropy and deformations
in the continental mantle, Geophys. J. Int. 111 (3) (1992)
433 – 447.
[10] P. Bormann, G. Gruenthal, R. Kind, H. Montag, Upper mantle
anisotropy beneath Central Europe from SKS wave splitting:
effects of absolute plate motion and lithosphere–asthenosphere
boundary topography? J. Geodyn. 22 (1–2) (1996) 11 – 32.
[11] P.G. Silver, W.W. Chan, Implications for continental structure
and evolution from seismic anisotropy, Nature 335 (1988)
34 – 39.
[12] A. Vauchez, A. Nicolas, Mountain-building: strike-parallel
motion and mantle anisotropy, Tectonophysics 185 (3–4)
(1991) 183 – 201.
[13] D.E. James, M. Assumpção, Tectonic implications of S-wave
anisotropy beneath SE Brazil, Geophys. J. Int. 126 (1996)
1 – 10.
[14] G. Barruol, P.G. Silver, A. Vauchez, Seismic anisotropy in the
eastern United States: deep structure of a complex continental
plate, J. Geophys. Res. 102 (B4) (1997) 8329 – 8348.
[15] G. Barruol, A. Souriau, A. Vauchez, J. Diaz, J. Gallart, J.
Tubia, J. Cuevas, Lithospheric anisotropy beneath the Pyrenees from shear wave splitting, J. Geophys. Res. 103 (B12)
(1998) 30039 – 30053.
[16] W. Ben-Ismail, D. Mainprice, An olivine fabric database: an
overview of upper mantle fabrics and seismic anisotropy,
Tectonophysics 296 (1998) 145 – 157.
[17] G. Clitheroe, R.D. Van der Hilst, Complex Anisotropy in the
Australian Lithosphere from Shear-Wave Splitting in BroadBand SKS Records, AGU Monograph Series, 1997.
[18] J.S. Myers, R.D. Shaw, I.M. Tyler, Tectonic evolution of
Proterozoic Australia, Tectonics 15 (6) (1996) 1431 – 1446.
[19] P.G. Betts, D. Giles, G.S. Lister, L.R. Frick, Evolution of the
Australian lithosphere, Aust. J. Earth Sci. 49 (2002) 661 – 695.
[20] D. Hill, Geology, in: M.G. (Ed.), Handbook of Queensland,
Australian Association for the Advancement of Science, Brisbane, 1951, pp. 13 – 24.
[21] N.G. Direen, A.J. Crawford, The Tasman Line: where is it,
what is it, and is it Australia’s Rodinian breakup boundary?
Aust. J. Earth Sci. 50 (2003) 491 – 502.
[22] E. Scheibner, Fossil fracture zones (transform faults), segmentation and correlation problems in the Tasman Fold Belt
System, in: A.K. Denmead, G.W. Tweedale, A.F. Wilson
(Eds.), The Tasman Geosyncline–A Symposium in Honour
of Professor Dorothy Hill, Queensland Division Geological
Society of Australia, Brisbane, 1974, pp. 65 – 96.
[23] R.D. Shaw, P. Wellman, P.J. Gunn, A.J. Whitaker, C. Tarlowski, M.P. Morse, Guide to using the Australian crustal
elements map, Australian Geological Survey Organisation
Record 1996/30, 1996.
[24] J.J. Veevers, C.M. Powell, Epi-Adelaidean: regional shear, in:
J.J. Veevers (Ed.), Phanerozoic Earth History of Australia,
Clarendon Press, Oxford, 1984, pp. 278 – 284.
M. Heintz, B.L.N. Kennett / Earth and Planetary Science Letters 236 (2005) 106–119
[25] P.J. Gunn, P. Milligan, T. Mackey, S. Liu, A. Murray, D.
Maidment, R. Haren, Geophysical mapping using the national
airborne and gravity datasets; an example focusing on Broken
Hill, AGSO J. Aust. Geol. Geophys. 17 (1997) 127 – 136.
[26] B.L.N. Kennett, S. Fishwick, A.M. Reading, N. Rawlinson,
Contrasts in mantle structure beneath Australia—relation to
Tasman lines? Aust. J. Earth Sci. 51 (2004) 563 – 569.
[27] S. Fishwick, B.L.N. Kennett, A.M. Reading, Contrasts in
lithospheric structure within the Australian craton—insights
from surface wave tomography, Earth Planet. Sci. Lett. 231
(2005) 163–176.
[28] F.J. Simons, A. Zielhuis, R.D. Van Der Hilst, The deep structure of the Australian continent from surface wave tomography, Lithos 48 (1999) 17 – 43.
[29] E. Debayle, B.L.N. Kennett, The Australian continental upper
mantle: structure and deformation inferred from surface waves,
J. Geophys. Res. 105 (11) (2000) 25423 – 25450.
[30] P.G. Silver, W.W. Chan, Shear wave splitting and subcontinental mantle deformation, J. Geophys. Res. 96 (B10) (1991)
16429 – 16454.
[31] C.J. Wolfe, P.G. Silver, Seismic anisotropy of oceanic upper
mantle: shear wave splitting methodologies and observations,
J. Geophys. Res. 103 (1) (1998) 749 – 771.
[32] W. Menke, V. Levin, The cross-convolution method for interpreting SKS splitting observations, with application to one and
two layer anisotropic Earth models, Geophys. J. Int. 154 (2)
(2003) 379 – 392.
[33] E. Debayle, B.L.N. Kennett, Anisotropy in the Australian
upper mantle from waveform inversion, Ann. Geophys. 16
(1998) 37.
[34] E. Debayle, SV-wave azimuthal anisotropy in the Australian upper mantle: preliminary results from automated Rayleigh waveform inversion, Geophys. J. Int. 137 (3) (1999)
747 – 754.
[35] F.J. Simons, R.D. Van Der Hilst, Seismic and mechanical
anisotropy and the past and present deformation of the
Australian lithosphere, Earth Planet. Sci. Lett. 211 (2003)
271 – 286.
119
[36] B.L.N. Kennett, S. Fishwick, M. Heintz, Lithospheric structure
in the Australian region—a synthesis of surface wave and
body wave studies, Explor. Geophys. 35 (2004) 242 – 250.
[37] C. Tong, O. Gudmundsson, B.L.N. Kennett, Shear wave splitting in refracted waves returned from the upper mantle transition zone beneath northern Australia, J. Geophys. Res. 99 (B8)
(1994) 15783 – 15797.
[38] A.E. Gripp, R.G. Gordon, Current plate velocities relative to
the hotspots incorporating the NUVEL-1 global plate motion
model, Geophys. Res. Lett. 117 (8) (1990) 1109 – 1112.
[39] A.E. Gripp, R.G. Gordon, Young tracks of hotspots and
current plate velocities, Geophys. J. Int. 150 (2002)
321 – 361.
[40] S. Wang, R. Wang, Current plate velocities relative to hotspots: implications for hotspot motion, mantle viscosity and
global reference frame, Earth Planet. Sci. Lett. 189 (2001)
133 – 140.
[41] N. Girardin, V. Farra, Azimuthal anisotropy in the upper
mantle from observations of P-to-S converted phases: application to Southeast Australia, Geophys. J. Int. 133 (3) (1998)
615 – 629.
[42] G. Barruol, R. Hoffmann, Upper mantle anisotropy beneath Geoscope stations, J. Geophys. Res. 104 (1999)
10757 – 10773.
[43] S. Ozalaybey, W.P. Chen, Frequency-dependent analysis of
SKS/ SKKS waveforms observed in Australia: evidence for
null birefringence, Phys. Earth Planet. Inter. 114 (3–4) (1999)
197 – 210.
[44] M.J. Fouch, P.G. Silver, D.R. Bell, J.N. Lee, Small-scale
variations in seismic anisotropy near Kimberley, South Africa,
Geophys. J. Int. 157 (2) (2004) 764 – 774.
[45] K. Wylegalla, G. Bock, J. Gossler, W. Hanka, T.W. Group,
Anisotropy across the Sorgenfrei–Tornquist Zone from shear
wave splitting, Tectonophysics 314 (1999) 335 – 350.
[46] S. Ozalaybey, W.-P. Chen, Frequency-dependent analysis of
SKS/SKKS waveforms observed in Australia: evidence for
null birefringence, Phys. Earth Planet. Inter. 114 (1999)
197 – 210.