Breakup and early seafloor spreading between India and Antarctica

Abstract

We present a tectonic interpretation of the breakup and early seafloor spreading between India and Antarctica based on improved coverage of potential field and seismic data off the east Antarctic margin between the Gunnerus Ridge and the Bruce Rise. We have identified a series of ENE trending Mesozoic magnetic anomalies from chron M9o (~130.2 Ma) to M2o (~124.1 Ma) in the Enderby Basin, and M9o to M4o (~126.7 Ma) in the Princess Elizabeth Trough and Davis Sea Basin, indicating that India-Antarctica and India-Australia breakups were roughly contemporaneous. We present evidence for an abandoned spreading centre south of the Elan Bank microcontinent; the estimated timing of its extinction corresponds to the early surface expression of the Kerguelen Plume at the Southern Kerguelen Plateau around 120 Ma. We observe an increase in spreading rate from west to east, between chron M9 and M4 (38-54 mm yr-1), along the Antarctic margin and suggest the tectono-magmatic segmentation of oceanic crust has been influenced by inherited crustal structure, the kinematics of Gondwanaland breakup and the proximity to the Kerguelen hotspot. A high-amplitude, E-W oriented magnetic lineation named the Mac Robertson Coast Anomaly (MCA), coinciding with a landwards step-down in basement observed in seismic reflection data, is tentatively interpreted as the boundary between continental/transitional zone and oceanic crust. The exposure of lower crustal rocks along the coast suggests that this margin formed in a metamorphic core complex extension mode with a high strength ratio between upper and lower crust, which typically occurs above anomalously hot mantle. Together with the existence of the MCA zone this observation suggests that a mantle temperature anomaly predated the early surface outpouring/steady state magmatic production of the Kerguelen LIP. An alternative model suggests that the northward ridge jump was limited to the Elan Bank region, whereas seafloor spreading continued in the West Enderby Basin and its Sri Lankan conjugate margin. In this case, the MCA magnetic anomaly could be interpreted as the southern arm of a ridge propagator that stopped around 120 Ma.

Full text

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Geophys. J. Int. (2007) doi: 10.1111/j.1365-246X.2007.03450.x
GJI Marine geoscience
Breakup and early seafloor spreading between India and Antarctica
Carmen Gaina,1R. Dietmar M¨uller,2Belinda Brown,2Takemi Ishihara3
and Sergey Ivanov4
1Center for Geodynamics, Geological Survey of Norway, Trondheim, Norway
2Earth Byte Group, School of Geosciences, The University of Sydney, Australia
3Institute of Geology and Geoinformation, National Institute of Advanced Industrial Science and Technology, AIST Central 7, Tsukuba, Japan
4Polar Marine Geophysical Research Expedition, St Petersburg
Accepted 2007 March 21. Received 2007 March 21; in original form 2006 May 26
SUMMARY
We present a tectonic interpretation of the breakup and early seafloor spreading between
India and Antarctica based on improved coverage of potential field and seismic data off the
east Antarctic margin between the Gunnerus Ridge and the Bruce Rise. We have identified a
series of ENE trending Mesozoic magnetic anomalies from chron M9o (∼130.2 Ma) to M2o
(∼124.1 Ma) in the Enderby Basin, and M9o to M4o (∼126.7 Ma) in the Princess Elizabeth
Trough and Davis Sea Basin, indicating that India–Antarctica and India–Australia breakups
were roughly contemporaneous. We present evidence for an abandoned spreading centre south
of the Elan Bank microcontinent; the estimated timing of its extinction corresponds to the
early surface expression of the Kerguelen Plume at the Southern Kerguelen Plateau around
120 Ma. We observe an increase in spreading rate from west to east, between chron M9
and M4 (38–54 mm yr–1), along the Antarctic margin and suggest the tectono-magmatic
segmentation of oceanic crust has been influenced by inherited crustal structure, the kinematics
of Gondwanaland breakup and the proximity to the Kerguelen hotspot. A high-amplitude, E–W
oriented magnetic lineation named the Mac Robertson Coast Anomaly (MCA), coinciding with
a landwards step-down in basement observed in seismic reflection data, is tentatively interpreted
as the boundary between continental/transitional zone and oceanic crust. The exposure of lower
crustal rocks along the coast suggests that this margin formed in a metamorphic core complex
extension mode with a high strength ratio between upper and lower crust, which typically occurs
above anomalously hot mantle. Together with the existence of the MCA zone this observation
suggests that a mantle temperature anomaly predated the early surface outpouring/steady state
magmatic production of the Kerguelen LIP. An alternative model suggests that the northward
ridge jump was limited to the Elan Bank region, whereas seafloor spreading continued in the
West Enderby Basin and its Sri Lankan conjugate margin. In this case, the MCA magnetic
anomaly could be interpreted as the southern arm of a ridge propagator that stopped around
120 Ma.
Key words: Antarctica, Enderby Basin, plate tectonics, sea floor spreading, Kerguelen.
1 INTRODUCTION
The Enderby Basin, Princess Elizabeth Trough and Davis Sea Basin
(Fig. 1) are remote regions off the east Antarctic continental margin
where the history of breakup and earliest seafloor spreading has been
poorly constrained due to limited data coverage. They constitute a
key area to understanding the timing and orientation of breakup and
seafloor spreading between Antarctica and India.
The role of India in the dispersal of Gondwanaland has remained
one of the largest uncertainties in the Mesozoic global plate circuit
due to a lack of palaeomagnetic and marine geophysical data, as
well as a complex breakup history (e.g. Powell et al. 1988; Coffin
1992; Grunow 1999). Early African–Antarctic spreading to the
west of the Gunnerus Ridge (∼30◦E) has been dated from Early
Cretaceous with a reasonably well-defined sequence from M24
(∼153 Ma) (Roeser et al. 1996; Jokat et al. 2003). West and north
of Conrad Rise, magnetic anomalies from chron 34 to 28 (∼83.5
to ∼64 Ma) have been identified (Goslin & Schlich 1976; Royer &
Coffin 1992). Early Australian–Antarctic spreading to the east of
the Bruce Rise and Vincennes Fracture Zone (∼105◦E) has been
identified with a Late Cretaceous spreading system between chron
34 (∼83.5 Ma) and 31 (∼71 Ma) (Tikku & Cande 1999).
For the early Indian–Antarctic spreading history different scenar-
ios have been proposed. Previous work on the conjugate Indian and
Antarctic margins (Ramana et al. 2001a) is based on scarce data
with wide spacing and random orientations of ship tracks, Magnetic
C
2007 The Authors 1
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
2C. Gaina et al.
Figure 1. The satellite derived marine free-air gravity field along the East Antarctic margin, between Gunnerus Ridge and the Bruce Rise (outlined), illuminated
with an azimuth of 330◦S (Sandwell & Smith 2005). Inset figure shows the location of our study area. Lines annotated (a–g) correspond to magnetic anomaly
profiles in Fig. 4a. Isochron C34 (∼83.5 Ma) from M¨uller et al. (1997) indicates the northern boundary of Cretaceous Quiet Zone crust along the SWIR, and
lineament labelled 60.1 Ma indicates approximate timing and zone of breakup between Broken Ridge and Kerguelen Plateau (outlined). Abbreviations are as
follows: 1137, ODP Leg 183-site 1137; BR, Bruce Rise; EB, Elan Bank; EL, Enderby Land; GR, Gunnerus Ridge; KFZ, Kerguelen Fracture Zone; KL, Kemp
Land; KPO Coast, Kron Prinz Olav Kyst; MRL, Mac.Robertson Land; PB, Prydz Bay; PEL, Princess Elizabeth Land; PET, Princess Elizabeth Trough; LHB,
Lutzow-Holm Bay; QML, Queen Mary Land, RLS, Riiser-Larsen Land, SKP, Southern Kerguelen Plateau; SWIR, Southwest Indian Ridge; VFZ, Vincennes
Fracture Zone and WL, Willem II Land.
anomaly identification off the conjugate Indian margin (e.g. Ramana
et al. 1994a,b; Banergee et al. 1995) has been inhibited by masking
and interference from thick Bengal Fan sediments (>5 km) and ig-
neous structures such as the 85◦E and 90◦E ridges (Curray 1991).
This has resulted in two, equivocal models with the identification of
magnetic anomalies M11 (∼133 Ma) to M0 (∼120 Ma) by Ramana
et al. (1994a,b) and a model by Banergee et al. (1995) that proposes
that the seafloor created in the Bay of Bengal is younger than 116 Ma
within a portion of a long normal polarity interval, the Cretaceous
Normal Superchron (CNS) (∼118–83.5 Ma). More recently, Desa
et al. (2006) have identified magnetic anomalies M11 (134 Ma) to
M0 (120 Ma) south of Sri Lanka as a conjugate to the West Enderby
oceanic floor basin (Ramana et al. 2001b).
There have been some experimental three-component magne-
tometer surveys conducted in the Enderby Basin region by Japanese
Antarctic Research Expeditions (JARE) during survey work on the
icebreaker Shirase (e.g. Nogi et al. 1996). These data are thought
to be helpful in detecting trends and lineations in areas with little
data coverage. Nogi et al. (1991, 1996) have presented a model for
a Mesozoic vector sequence from M9 (∼130 Ma) with a general
NE–SW trend in the central Enderby Basin.
The uncertainty in magnetic anomaly identification in the Bay of
Bengal and sparse data coverage off the Enderby Basin, Princess
Elizabeth Trough, and in the Davis Sea has led to two alternative
models for Cretaceous plate reconstructions for the Indian Ocean.
One hypothesis is that the age of early ocean crust formed during
breakup is largely Cretaceous Normal Superchron (CNS) (∼118–
83.5 Ma) crust with little or no Mesozoic sequence (e.g. Royer &
Coffin 1992; M¨uller et al. 2000). An alternative hypothesis sug-
gests that older Mesozoic crust (∼120+Ma) does exist and that
spreading was roughly contemporaneous with the well-documented
M-sequence (M10–M0) off the Perth Abyssal Plain (M¨uller et al.
1993; Powell et al. 1988).
During the mid-Cretaceous period there was also increasing mag-
matic activity in the region, related to the development of the Ker-
gulen Large Igneous Province (LIP) from about 120 to 110 Ma (Frey
et al. 2000; Nicolaysen et al. 2001). It is likely that the history of
early seafloor spreading is complicated by ridge–hotspot interaction
during the growth of the Kerguelen Plume. The Elan Bank micro-
continent was most probably isolated from the Indian continent by
one or several ridge jumps associated with the Kerguelen Plume
(M¨uller et al. 2001; Gaina et al. 2003).
Our recent compilation of shiptrack potential field data in the En-
derby Basin, Princess Elizabeth Trough and Davis Sea offers an op-
portunity to address questions about the early breakup and spreading
history between India and Antarctica. We show the improved survey
coverage of potential field and seismic data from both new and ex-
isting sources and present an interpretation of a Mesozoic seafloor
spreading sequence based on marine magnetic anomaly data. We
examine the tectono-magmatic variation along-axis of the zone of
breakup, and spreading segmentation off the Antarctic margin, using
potential field data and seismic refraction and reflection data. We
also show supporting evidence for an abandoned ‘fossil’spread-
ing centre, in the Enderby Basin, west of Kerguelen Plateau and
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 3
discuss the development of the early spreading system in the con-
text of mantle–lithosphere interaction with the growth and influence
of the Kerguelen Plume.
2 PHYSIOGRAPHIC SETTING
The region off the East Antarctic margin once conjugate to Southern
Greater India (e.g., Powell et al. 1988; Harley & Henson 1990;
Ramana et al. 1994a) extends over a large part of the Enderby Basin,
the Princess Elizabeth Trough and the Davis Sea Basin (Fig. 1). The
Enderby Basin is a wide area located between the Kerguelen Plateau
and the Antarctic margin, bounded to the northwest by the Kerguelen
Fracture Zone and by the Crozet Basin. It includes the area conjugate
to the eastern Indian continental margin, and extends across Enderby
Land, Kemp Land, Mac Robertson Land and Princess Elizabeth
Land. The Princess Elizabeth Trough, to the east of the Enderby
Basin, is a narrow zone that separates the southern extension of the
Kerguelen Plateau from the Antarctic margin. The Davis Sea, further
to the east, is the area where Southern Greater India and Australia
were once joined to Antarctica, in the region offshore Wilhelm II
Land and Queen Mary Land. The Russian and Japanese Antarctic
programs named the western Enderby Basin the Cosmonaut Sea and
the eastern Enderby Basin the Co-operation Sea (e.g. Joshima et al.
2001).
Our study area extends from approximately 30◦E to 105◦E be-
tween the Gunnerus Ridge and Bruce Rise (Fig. 1). Gunnerus Ridge
is a narrow, submarine ridge that lies perpendicular to the Antarc-
tic margin; it consists mostly of continental crust with an igneous
structure at its northern tip (Bergh 1987; Roeser et al. 1996). To
the east of Gunnerus Ridge, Sri Lanka has been correlated with the
onshore Pre-Cambrian terranes at Lutzow-Holm Bay (e.g. Fedorov
et al. 1982; Yoshida et al. 1992; Buchel 1994, Shiraishi et al. 1994;
Kriegsman 1995; Yoshida et al. 1996; Lawver et al. 1998). In be-
tween Gunnerus Ridge and Bruce Rise lies the Prydz Bay-Lambert
Graben structure; geophysical and geological correlation suggests
that it is part of a pre-existing N–S Palaeozoic intracontinental rift,
conjugate with the Mahanadi Graben in eastern India (Fedorov et al.
1982; Stagg 1985; Lisker et al. 2003). The graben has undergone
an active tectonic history with repeated intrusive and extrusive ig-
neous activity, and inferred uplift around Miocene to Recent times
(Wellman & Tingey 1982). The Bruce Rise is a crystalline conti-
nental basement plateau that has been correlated to the Naturaliste
Plateau off the southwest Australian margin (Murakami et al. 2000;
Stagg et al. 2004). The eastern flank of the Bruce Rise is adjacent to
the Vincennes Fracture Zone, which is conjugate to the Perth Frac-
ture Zone and related to Australian–Antarctic spreading along the
Southeast Indian Ridge (Tikku & Cande 1999).
The physiographic setting of the region is dominated by igneous
structures related to the Kerguelen Large Igneous Province (LIP),
as well as large areas of relatively thick sediment. It is possible
that a large area of seafloor and potentially a number of continen-
tal fragments are now overprinted by voluminous igneous activ-
ity associated with the Kerguelen Plume, which began forming at
the Southern Kerguelen Plateau at about 118 ±2Ma(Freyet al.
2000; Nicolaysen et al. 2001). ODP Legs 119 (sites 738–746), 120
(sites 747–751) and 183 (sites 1135–1142) were drilled with the
object of investigating the Kerguelen LIP. In particular, the recov-
ery of core material including garnet-biotite gneiss on Elan Bank
at ODP 183 (site 1137) indicated a continental origin (Nicolaysen
et al. 2001). Several wide-angle and reflection seismic profiles ac-
quired by French and Australian surveys describe the crustal prop-
erties and structure of the microcontinent (Charvis & Operto 1999;
Gladczenko & Coffin 2001; Borissova et al. 2003) and the basins
directly off the Kerguelen Plateau, including the Labuan Basin
(Borissova et al. 2002).
The marine free-air gravity field anomalies derived from satellite
altimetry (Smith & Sandwell 1997) do not resolve many shorter
wavelength free-air gravity features off the Antarctic margin due
to areas of thick ice and/or sediment cover (e.g. McAdoo & Laxon
1997; Rotstein et al. 2001). Sediment thickness is commonly 1–
2 km in the abyssal plain and increases up to 6–8 km in areas toward
the continental margin (Mizukoshi et al. 1986; Murakami et al.
2000; Stagg et al. 2004). As a consequence, the identification of
basement fabric, such as the gravimetric expressions of fracture
zones (e.g. Goslin & Schlich 1982) is difficult. Further to the north of
the study area a series of relatively short NNE-trending fracture zone
lineations can be observed and the might indicate the development of
oceanic crust between East Antarctica and India/Sri Lanka younger
than 120 Myr. The parallel series of NNE fracture zones located
north of Gunnerus Ridge also extend into the Riiser-Larsen Sea;
they are more likely to belong to the Africa–Antarctica (AFR–ANT)
spreading system described by Nogi et al. (1996) and Jokat et al.
(2003). To the north they are truncated by NNW fracture zones in
the Crozet Basin (Fig. 1), which includes CNS crust, south of the
C34 isochron (∼83.5 Ma).
3DATA
In addition to existing open-file geophysical data from the
National Geophysical Data Center (NGDC 1998) and older French
surveys (R/V Marion Dufresne surveys 1, 5, 11, 38 and 47) (e.g.
Goslin & Schlich 1982), there are several more recent offshore
survey programs that have vastly improved shiptrack data cover-
age in the Enderby Basin region. These surveys include: Japan
National Oil Company (JNOC) TH83, TH84, TH85, TH98 and
TH99 (e.g. Mizukoshi et al. 1986; Murakami et al. 2000; Joshima
et al. 2001), Soviet Antarctic Expeditions/Russian Antarctic Expe-
ditions surveys 31–47 (e.g. Gandyukin et al. 2002; Golynsky et al.
2002) and Geoscience Australia surveys 228 and 229 (e.g. Stagg &
Colwell 2003). There has been a unique opportunity through vari-
ous data agreements to compile virtually all the available shiptrack
magnetic anomaly, free-air gravity and bathymetry data in the En-
derby Basin, Princess Elizabeth Trough and Davis Sea, between ap-
proximately 30◦E and 110◦E. This has provided the most complete
magnetic anomaly data set for the region to date; Fig. 2 provides
a summary of the coverage of both the older and new data sets.
The marine free-air gravity field derived from satellite altimetry
(Sandwell & Smith 2005) and GEBCO bathymetry (GEBCO 2004)
were used in this study, in conjunction with magnetic anomaly data
to identify the age of seafloor spreading and tectonic structure of the
Enderby Basin. Additional access to some seismic reflection profiles
(Australian and Japanese survey data) and sonobuoy velocity–depth
data (Japanese, Russian and Australian surveys) has allowed us to
build upon our interpretation of potential field data. A detailed de-
scription of this integrated seismic data set is included in Stagg et al.
(2004).
3.1 Magnetic anomalies
The newly collated magnetic anomaly data (Fig. 2) were reformatted
and assembled into one database. Magnetic data collected at these
latitudes are often more affected by electromagnetic disturbances
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
4C. Gaina et al.
Figure 2. Trackline coverage of magnetic anomaly data in the Enderby Basin region. Tracklines in pale grey represent older data sets from NGDC (1998)
and R/V Marion Dufresne. The colored tracklines represent recent data compilation from: Russian data (blue); Japanese data (green); and new Australian data
(red). Annotations and abbreviations as in Fig. 1, SEIR, Southeast Indian Ridge; the Kerguelen Plateau is outlined and shaded grey.
(e.g. Kleimenova et al. 2003), in addition to wind and instrument
noise. The magnetic anomaly data were filtered with a high-pass
filter of 300 km and smoothed with a low-pass filter of 10 km. The
filtering process cleaned up the magnetic anomaly data sufficiently
to apply a gridding technique, following the method of the North
American Magnetic Anomaly Group (NAMAG 2002). The filtered
data were gridded with the GMT (Wessel & Smith 1991) spline grid-
ding tool ‘surface’and a near neighbour weighting algorithm based
on an elliptical allocated search area, with its long axis oriented
along the strike of the magnetic lineations, and an 0.5◦grid interval.
The gridding method is dependent on line spacing, sample density
and interpolation errors, so data gaps greater than a 0.25 degrees
cell radius were excluded in order to reduce sampling function bias,
using the GMT ‘near-neighbour’routine. The improved coverage of
magnetic anomaly data has allowed gridding techniques to augment
the identification of lineations, trends and sequences. The colour-
shaded magnetic anomaly grid highlights trends that are not seen
as easily with a conventional along-track anomaly plot, and allows
interpretation of seafloor fabric character (Fig. 3).
From along-track and gridded magnetic anomaly data, a series of
ENE trending magnetic anomalies can be observed off the Antarctic
margin, between the Gunnerus Ridge and Bruce Rise, in particular
east of 55◦longitude. In the Enderby Basin to the west of the Kergue-
len Plateau and to the south of Elan Bank (central Enderby basin), a
symmetric magnetic anomaly sequence from M9o (∼130.2 Ma) to
M2o (∼124.1 Ma) has been observed on either side of a central axis,
as suggested by Gaina et al. (2003). The character and distance of the
magnetic anomaly profiles between both M2 lineations suggests the
existence of a palaeospreading axis, which trends ENE. Spreading is
likely to have slowed prior to extinction and the width of the central
axis anomaly indicates it ceased after M2, possibly around M0 time
(∼120 Ma). A new active ridge north of Elan Bank replaced the
extinct ridge due to the proximity of the Kerguelen Plume (Gaina
et al. 2003), the timing of this jump roughly coincides with the first
recorded surface expression of the plume at the Southern Kergue-
len Plateau dated at about 118 ±2 Ma (e.g. Coffinet al. 2002).
North of Elan Bank the magnetic anomaly pattern is very chaotic
resembling CNS crust (∼118–83.5 Ma). This also implies most of
the crust subsequently accreted to the Indian Plate is younger than
M2, and is largely Cretaceous Quiet Zone crust. This would explain
the difficulty in discerning a magnetic anomaly sequence in the Bay
of Bengal (Curray 1991; Ramana et al. 1994a,b; cf. Banergee et al.
1995)
Fig. 4 presents a series of north–south oriented magnetic anomaly
stacked-profiles from the central, western and eastern Enderby basin
and the correlation with the interpreted seafloor spreading sequence.
The Mesozoic global geomagnetic reversal timescale of Gradstein
et al. (1994) was used to create a synthetic seafloor spreading model
and assign ages to oceanic crust. Model 1 (Fig. 4a) suggests that the
seafloor spreading in the entire Enderby basin became extinct some
time between M2o and M0 as a result of a northward ridge jump
due to the Kerguelan plume. A prominent positivemagnetic anomaly
can be well distinguished on each side of the extinct ridge and we
have interpreted it as chron M4. The youngest magnetic anomaly in
this model is interpreted to be chron M2, is characterized by lower
amplitude magnetic anomaly peaks separated by a trough and is
visible on most of the profiles. The oldest magnetic anomaly (chron
M9) is characterised by a high amplitude, narrow peak northward
of the MCA anomaly (well defined on profiles b, e and g, southern
flank), but it is more difficult to be interpreted on the northern flank
due to the proximity of the Kergulen plateau and later volcanic in-
trusions [see profiles and e and c (northern flank)]. Half-spreading
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 5
Figure 3. Gridded magnetic anomaly data after the new data compilation. Grey areas indicate gaps in data coverage greater than 0.25◦spacing. Annotations
and abbreviations as in Fig. 1; MCA, Mac.Robertson Coast Anomaly Lineation.
rates are calculated from the central Enderby Basin, based on a
complete, symmetric M-sequence. These rates also show the slow-
ing of spreading prior to extinction of the early active ridge from
39 mm yr–1(M9y–M4o) to 22 mm yr–1(M4o-M2o) and they com-
pare well to the half-rates for the Perth Basin 36 mm yr–1(M10-M5)
to 32 mm yr–1(M5-M2), and 24 mm yr–1(M2-M0) (Mihut 1997;
Mihut & M¨uller 1998). Model 2 (Fig. 4b) analyses the magnetic
data in the western Enderby basin in concert with the conjugate
south of Sri Lanka magnetic anomalies. Half seafloor spreading
rates from 39 to 30 mm yr–1have been modelled for the western
Enderby Basin and from 36 to 22 mm yr–1for south of Sri Lanka
region.
In this interpretation chrons M0, M2 and M4 are well defined on
two profiles south of Sri Lanka (Fig 4b, profiles jr116 and c2706),
but less visible on profiles SE of Sri Lanka. The magnetic anomaly
profiles from the conjugate Enderby basin show two high amplitude,
narrow peaks (identified as M2 and M0, profiles agso229(1) and
c1704(4) on Fig. 4b) south of a zone of chaotic and more subdued
magnetic anomalies (identified as CNS), but chron M4 is less well
defined [profiles agso229(1) and c1704(3) on Fig. 4b].
The filtered and projected along track magnetic anomaly data
were plotted at a scale to show a more detailed sequence inter-
pretation for the spreading segments in the western and eastern
Enderby Basin (Fig. 5). The extent of the identified Mesozoic mag-
netic anomaly sequence and extinct ridge axis varies across the
margin due to data coverage. The magnetic anomaly data coverage
in the western Enderby Basin (∼30◦E–65◦E) remains limited north
of 62◦S. In the Princess Elizabeth Trough and Davis Sea (between
∼75◦E and 105◦E) there is a relatively narrow area of crust be-
tween the Antarctic continental margin and the Kerguelen Plateau.
In this area data further to the north are obscured by the Kerguelen
Plateau igneous structure and to the east there is a truncation of
Mesozoic crust by the sharp transition to the E–W trending anoma-
lies of the fast-spreading Southeast Indian Ridge between Australia
and Antarctica (Fig. 3).
Since the magnetic anomalies in the western Enderby Basin have
lower amplitude and are more difficult to correlate with the central
and eastern Enderby Basin, we also evaluate an alternative model
for the evolution of this area (Figs 4b and 5b). N–S oriented frac-
ture zones identified from the free-air gravity anomaly appear to
isolate a V-shaped zone of oceanic crust in the central Enderby
Basin, also the MCA magnetic anomaly has a northern counterpart
that outlines the southern end of the N–S fracture zones (Fig. 5b).
This configuration resembles a ridge propagator very similar to the
ones described NW of Australia by Robb et al. (2005) and Mihut &
M¨uller (1998), suggesting that the northward ridge jump occurred
only in the central and eastern Enderby basin (which were closer
to the Kerguelan plume), whereas seafloor spreading in the west-
ern Enderby basin continued after 124 Ma. A sequence of magnetic
anomalies and fracture zones identified south of Sri Lanka (Desa
et al. 2006) may represent the conjugate of the seafloor spreading
system NE of Gunerus Ridge. However, based on the overall ge-
ometry of seafloor spreading in the Enderby basin, the location of
the boundary between continental and oceanic crust and regional
plate kinematics, we interpret a M9o to M0 sequence in both west-
ern Enderby basin and S of Sri Lanka. This model requires an
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
6C. Gaina et al.
Figure 4. Selected, representative magnetic anomaly profiles along the East Antarctic margin, between the Gunnerus Ridge and the Bruce Rise. Location of
profiles (a–g) indicated in Figs. 1 and 5. Corresponding shiptrack bathymetric profiles also displayed where there are bathymetric highs, including Bruce Rise
and Elan Bank; also the bathymetric trough feature in the Kron Prinz Olav (KPO) coast segment. MCA, Mac.Robertson Coast Anomaly; XR, extinct ridge. A
synthetic profile is included based on the geomagnetic timescale of Gradstein et al. (1994), using a depth to the top of the magnetized layer of 5.5 km and a
seafloor deepening calculated by Parson & Sclater’s (1977) half-space cooling formula. The thickness of magnetized layer is 0.3 km. The body is assumed to
have been magnetized at 55 ◦S latitude.
asymmetric seafloor spreading and possibly a plate boundary be-
tween the western and central Enderby Basin. A drawback of this
model is that although the magnetic anomalies south of Sri Lanka
could represent Mesozoic crust older than CNS, the amplitude and
shape of magnetic anomalies in the western Enderby basin north of
interpreted M0 (Fig. 5b) do not have a clear CNS signature. Until
additional information will better constrain the age of the crust in
the western Enderby Basin, our preferred model is model 1 that
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 7
Figure 4. (Continued.)
postulates a northward propagation of seafloor spreading ridge
around 120 Ma for the entire Enderby basin.
4 BREAKUP AND EARLY SPREADING
SEGMENTATION
We have used other available geological and geophysical data in
order to develop a more integrated tectonic model for breakup and
early seafloor spreading. While we can identify a seafloor spread-
ing sequence with a general ENE trend along the Antarctic margin
(Figs 4 and 5), from west to east we can observe an increase in
spreading rate and tectono-magmatic variations across the spread-
ing segments from the potential field data (e.g. Figs 3 and 5), seismic
refraction velocity–depth data (e.g. Fig. 6) and seismic reflection
profiles (e.g. Figs 7–9). Tectono-magmatic variations are expected
along a margin of this size, not just from the magmatic processes
that create seafloor but also the conditions of breakup along differ-
ent terranes and pre-existing structures, as well as the kinematics
of opening, ridge propagation and plate separation. The variation
in spreading segment character observed along the margin suggests
that breakup segmentation is related to a combination of these fac-
tors as well as mantle–lithosphere interaction from the developing
Kerguelen Plume.
Here, we present a description of observed large-scale varia-
tion for area over 3000 km long off the Antarctic margin between
the Gunnerus Ridge and Bruce Rise. These differences are de-
scribed below, broadly divided into areas of longitude for the seg-
ments from west to east between: (1) the Kron Prinz Olav Coast
(KPO) and the Enderby Land Promontory (∼30◦E–60◦E), (2) the
Mac.Robertson Coast and Prydz Bay (∼60◦E–80◦E) and (3) the
Princess Elizabeth Trough (PET) and Davis Sea Basin (∼80◦E–
105◦E).
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
8C. Gaina et al.
Figure 5. Free air gravity anomaly (Sandwell & Smith 2005) and magnetic anomaly data in the (a) western and eastern Enderby basin (model 1) and b)
western Enderby basin and S of Sri Lanka (model 2). Magnetic anomaly identifications are shown as red circles and the interpreted isochrones as white and
grey lines. Interpreted fracture zones are drawn in orange. A V-shaped lineation identified in the gravity anomaly is interpreted as a change in the spreading
direction (probably at 99 Ma) both NW of the Enderby basin and south of Sri Lanka (b). The MCA anomaly (south of the central Enderby basin) and a
northern counterpart identified in the gravity anomaly are shown in magenta and are interpreted as the trace of a propagator (b, model 2). Tracklines are in
black with cruise names attached. For a more detailed interpretation of the central Enderby basin magnetic anomalies see Gaina et al. 2003. Annotations as in
Fig. 1.
4.1 Zone of breakup variations
Along-axis margin segmentation is identified based on observations
of the variable nature of the identified boundary between continen-
tal and oceanic crust. In the western Enderby basin, along the Kron
Prinz Olav coast segment, the COB has no continuous boundary
signal or distinct seismic character. Near the end of the continental
slope a magnetic trough is observed inboard of the southern limit
of identified seafloor spreading anomalies (Fig. 5). The magnetic
trough roughly corresponds to a trough/low in shiptrack free-air
gravity and bathymetry data. It also lies above a basement ‘hol-
low’feature in the seismic reflection profiles, near the end of the
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 9
Figure 6. Selected grids for the Enderby Land margin viewed from 220◦azimuth. (a) Bathymetry [Etopo5); (b) Gridded sonobouy velocity data at the 6km/s
isovel. Seismic crustal character and depth estimated from Crust 2.0 (Masters et al. 2000), depth to (c) to top of basement; (d) to base of upper crust and (e)
to base of lower crust (depth with respect to sea level). The grid cell is 2 minutes. Basement and upper and lower crust highs are in warm colours, basement
and upper and lower crust lows are in colder colours. Note variations from west to east, including the Prydz Bay and Lambert Graben structure between
approximately 70◦E and 80◦E.
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
10 C. Gaina et al.
Figure 7. Seismic reflection profile 229/35 reproduced from Stagg et al. (2004) to show oceanic basement morphology. Location of profile (a) indicated on
Fig. 1. Free-air satellite derived gravity anomaly and magnetic anomalies are juxtaposed here to show the correlation with the interpreted extinct ridge (XR)
and seafloor spreading sequence.
Figure 8. Seismic reflection profile 228/01 reproduced from Stagg et al. (2004) to show oceanic basement morphology. Location of profile (b) indicated on
Fig. 1. Free-air satellite derived free-air gravity anomaly and magnetic anomalies are juxtaposed here to show the correlation with the interpreted extinct ridge
(XR) and seafloor spreading sequence. As this profile extends across a maximum segment length, from the south (the continental rise) to the north (to the
Kerguelen Fracture Zone), presumably it represents the full spreading segment sequence. Here we speculate on the location of M0 and CNS crust based on this
assumption.
continental sediment wedge which is underlain bythe seafloor multi-
ple. This feature can be observed on several profiles and corresponds
to the interpreted boundary between continental and oceanic crust
from Stagg et al. (2004) (Figs 7 and 8).
In the central Enderby Basin between approximately 60◦E–73◦E,
the southern limit of identified seafloor spreading anomalies is
marked by a prominent series of high-amplitude magnetic anoma-
lies (over 500 nT), which lie to the north of a magnetic subdued
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 11
Figure 9. Composite seismic reflection profiles 33–63, TH99–27, 228/06 and TH99–06 re-produced from Stagg et al. (2004) to show oceanic basement
morphology and internal crustal character. Location of profiles indicated on Fig. 1. Free-air satellite derived gravity anomaly and magnetic anomalies are
juxtaposed here to show the correlation with the interpreted extinct ridge (XR) and seafloor spreading sequence. Note the northward step in basement
corresponding to the Mac.Robertson Coast Anomaly (MCA).
zone (Fig. 3). The high amplitude anomalies form a large arcuate
lineation that trends approximately E–W, oblique to the ENE ori-
entation of the seafloor spreading lineations (Fig. 3). This anomaly,
named the Mac Robertson Coast Anomaly (or MCA), coincides with
a landwards step-down in basement that is observed in the regional
Japanese and Australian seismic data (e.g. Fig. 8). It is interpreted
to mark a major crustal boundary, probably the boundary between
continental and oceanic crust (COB) or a propagator.
In the Princess Elizabeth Trough (PET) the boundary between
continental and oceanic crust appears to be closer to the continental
margin shelfbreak. Mizukoshi et al. (1986) speculate from their ob-
servations of seismic reflection data (e.g. their Line SMG7, ∼79◦E)
that a steep rise in acoustic basement and corresponding negative
free-air anomaly may indicate the COB, with the seaward bound-
ary of crystalline continental basement marked by a large fault. Off
the Bruce Rise there is the narrowest gap between the interpreted
continent–ocean boundary off the crystalline basement plateau and
the M9o chron, the southern limit of the interpreted seafloor spread-
ing sequence (Fig. 5).
4.2 Spreading segment variation
Spreading segment variation that occurs within/along the same ridge
system have been attributed to differences in mantle temperature
(e.g. along the present day Southwest Indian Ridge (SWIR), Sauter
et al. 2004). Variation between the spreading segment centres and
ends is also to be expected, especially at slow spreading ridges
(Patriat & Segoufin 1988).
Oceanic basement morphology is also reflected in variation of
magnetic anomaly character. Here we present characteristics of
spreading segments of the Enderby Basin that are most likely related
to magma supply, spreading rate and proximity to the developing
Kerguelen Plume thermal anomaly.
The magnetic anomalies in the KPO segment trend roughly ENE
but have more variable shape than those to the east of 60◦E (Fig. 3). In
general, thick sediments and water depth contribute to the variation
in magnetic anomaly character as do differences in basement topog-
raphy (Srivastava & Roest 1995). There are relatively thick sediment
sequences and drifts observed in the KPO segment, as well as the
observed rough oceanic basement topography with quite varied re-
lief (Figs 7 and 8). The variable character of magnetic anomalies
is likely due to varying depth to source with a combination of the
effects of sediment thickness and pronounced basement roughness
and relief. However, this pattern might be also due to a dense dis-
tribution of offset fracture zones that cannot be clearly mapped due
to the N–S oriented ship tracks and thick sediment cover.
In contrast, the region situated between the Enderby Land
Promontory and the Princess Elizabeth Trough includes the widest
area of extended continental crust off Prydz Bay and the widest zone
of Mesozoic seafloor crust between the margin and the Elan Bank
microcontinent (Fig. 1). This area exhibits relatively high-amplitude
magnetic anomaly lineations. It has been suggested that the increase
in amplitude of magnetic anomalies towards the Aegir Ridge in the
Norwegian-Greenland Sea could reflect an increase in thickness of
the magnetized layer as the ridge became closer to the Iceland hot
spot (Jung & Vogt 1997). The Mac Robertson Coast and Prydz Bay
segment of the margin was closer to the Kerguelen Plume at the
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
12 C. Gaina et al.
time of formation and this is reflected in the type of oceanic crust
observed. Charvis & Operto (1999) indicate areas of crustal thick-
ness up to 10–13 km from OBS data transects near the southwest of
Elan Bank.
In the central Enderby Basin, Princess Elizabeth Trough and
Davis Sea area deep-crustal Australian seismic profiles image dif-
ferent types of oceanic basement morphology and crustal structure
(e.g. Stagg et al. 2004; Figs 7–9). Seismic profiles in the western
Enderby Basin off the Kron Prinz Olav (KPO) coast segment exhibit
a pronounced oceanic basement roughness with a crustal structure
characterized by chaotic internal reflectors. The seismic profiles
intersect several fracture zones at an oblique angle, which might ex-
plain part of the basement roughness. In comparison, to the east of
this segment, in the central Enderby Basin and Davis Sea, oceanic
basement topography observed in Australian and Japanese seismic
reflection profiles is generally low relief without any pronounced
roughness. Profiles in the eastern segments often exhibit high am-
plitude reflectors characterizing internal crustal features and strong
Moho. In particular, the seismic character of many profiles in the
MCA sector and the Davis Sea, close to the Kerguelen Plateau area,
exhibit an internal reflector pattern that looks similar to an irregular
series of diagonal crosses; they are comparable to the type of crust
observed in the Cuvier Abyssal Plain off Western Australia (Colwell
et al. 1994).
The PET spreading segment lies off a narrower area of continen-
tal shelf. The area also comprises of a reasonably complex zone of
oceanic crust as it was formerly adjacent to the Cretaceous triple
junction between Antarctica, Greater India and Australia and an
area of crust left over from the division of the Broken Ridge and
Kerguelen Plateau by the SEIR at about 61 Ma. The Broken Ridge-
Kerguelen Plateau experienced some shear motion prior to their
separation, and some minor deformation that may be related to this
event is imaged in seismic reflection lines as a deformation zone
with more chaotic or disturbed sediments. Surprisingly, there is no
evidence from seismic reflection profiles for the rough oceanic base-
ment topography that is observed at the conjugate Diamantina Zone,
which extends from Broken Ridge to the Great Australian Bight.
Stagg et al. (2004) suggest the division of at least two crustal
provinces between the western and eastern Enderby Basin segments
based on the observed variation of oceanic crustal morphology and
seismic structure. This variation is reflected in differences of the
velocity–depth structure of oceanic crust between the western and
eastern spreading segments in the Enderby Basin (Fig. 6). The area
between the eastern and western Enderby Basin provinces appears
to be a highly segmented offset zone probably because it is the
location of maximum curvature in the spreading system, north of
the Enderby Land promontory. Margin segmentation is also likely
influenced by major continental crustal discontinuities. Recent on-
shore/continental seismic lithospheric thickness studies (Morelli &
Danesi 2004) indicate a discontinuity at about 60◦E and at 90◦E.
There is also a cluster of seamounts populating the region near
Elan Bank (Fig. 1) as opposed to other segments where there are
only a few scattered seamounts observed closer to the Antarctic con-
tinental margin. This seamount cluster appears to be at an oblique
orientation to the early spreading direction, and follows rather a
similar trend to the younger Kerguelen Fracture Zone. The prox-
imity to the Kerguelen Plateau leads to the suggestion that the tim-
ing and nature of their emplacement are related to the Kerguelen
plume. Volcanic elongated ridges are thought to be formed mainly
by extrusive volcanismin a melt-channel. For instance, the Musician
seamounts chain related to the Hawaiian Plume are oblique to the
hotspot trace (Kopp et al. 2003), where hotspot–ridge interaction
leads to asthenospheric channeling from the plume to the nearby
spreading centre over a maximum distance of 400 km. Often the
amount of excess hotspot volcanism is related to the spreading rate,
where intermediate-slow ridges can focus the plume to the ridge but
not be enough to entrain all the excess melt (Jellinek et al. 2003).
The extra magmatic activity is not unexpected in this region where
ridge–hotspot interaction isolated the Elan Bank microcontinent.
4.3 A ‘Fossil’spreading centre
Several lines of evidence suggest the existence of an extinct ‘fossil’
spreading centre in the central Enderby Basin. From the magnetic
data, we could not clearly identify chron M0, but the two conjugate
M2o magnetic anomalies observed in the central Enderby basin are
separated by a trough that signifies seafloor spreading continued
until the next normal polarity (approximately 120 Ma). A synthetic
model that matches the observed pattern and distances between con-
jugate M2o chrons indicates that seafloor spreading rate dropped to
around 8 mm yr–1(half spreading rate) before becoming extinct
(Fig. 4a).
In addition to the observed magnetic anomaly pattern, there are
corresponding changes in free-air gravity and oceanic basement
morphology around the central axis of the abandoned spreading
centre, which suggests that there is a remnant crustal feature in this
zone. Sedimentary sequences are up to 2 km thick and therefore
too thick to reveal a bathymetric expression of the extinct ridge fea-
ture. A basement feature that might be interpreted as an abandoned
spreading centre is visible in the seismic data from the central En-
derby basin (Fig. 9), but the rough basement structure of the western
Enderby basin makes difficult to recognize any possible fossil ridge
(Figs 7 and 8). The location of the extinct ridge identified by mag-
netic anomalies is characterized by a step in the satellite derived
free air gravity anomaly in the western Enderby basin (Figs 7 and
8), and a peak in the gravity anomaly in the central Enderby basin
(Fig. 9).
The differences in the gravity anomaly signature along the extinct
ridge might indicate differences in the last stage of seafloor spread-
ing within the Enderby basin. Many studies suggest that before active
accretion ceases, there is a protracted period whereby the spread-
ing rate slows (e.g. Osler & Louden 1995; Grevemeyer et al. 1996;
Livermore et al. 2000). The change in crustal thickness or magma
composition as spreading slows is often reflected by a change in the
gravity anomaly at extinct ridges. Some observations and models
suggest the spreading rate and crustal variation in this zone results
in more anomalous features like a deeper ridge or thinner crust, and
possibly even the remains of a palaeomagma chamber structure.
Evidence from seismic refraction studies at the extinct spreading
centre in the Labrador Sea indicates thinner crust and lower crustal
velocities (e.g. Osler & Louden 1995). These features associated
with the Labrador extinct ridge correspond to a gravity low. The
gravity high observed at the inferred extinct spreading centre in
the Enderby Basin is one of the few extinct ridges characterized by
a positive gravity anomaly.This positive gravity anomaly may be
related to a crustal structure generated by a relatively higher than
normal spreading rate prior to extinction, as explained below.
The largest variation of oceanic crustal thickness as a function
of spreading rate is observed when spreading rates are less than
15–10 mm yr–1(e.g. Reid & Jackson 1981; White 1992; Bown &
White 1995). Most half-spreading rates calculated at other extinct
ridges indicate a slowing to quite low rates prior to extinction of the
active ridge; for example, the Labrador system slowed from 10 to
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 13
3mmyr
–1(Srivastava & Keen 1995) and the Aegir system slowed
to between 8 and 5 mm yr–1(Jung & Vogt 1997). In the Enderby
Basin, the western and eastern segments slowed from 22 mm yr–1
(29 mm yr–1, respectively)to <10 mm yr–1. The rates may have been
higher in the Enderby Basin prior to extinction as most of the early
motion during early seafloor spreading was taken up by the Indian
continent moving away from a relatively stationary Antarctica; and
in the late Cretaceous there were plate boundary forces from the
Tethys subduction system to the north. This may partly account for
the relatively high spreading rates in the last stages of spreading
before the ridge jump. Also to consider is the spatial relationship
linked to melting depth and spreading rate with ridge morphology.
Huang & Solomon (1998) use earthquake observations to suggest
the maximum centroid depth at an active ridge crest increases from
2 to 3 km at faster spreading rates (20–23 mm yr–1half-rate) up
to 5–6 km at slower spreading rates (2.5–5mmyr
–1half-rate). In
the case of the extinct spreading centre in the Enderby Basin, the
spreading rates would not be slow enough to result in lower than
expected crustal thickness, the palaeoridge also would not be as
deep as those formed at lower spreading rates, and hence not be
linked to a gravity low as at many other extinct ridges.
If spreading in the Enderby Basin ceased between chron M2 and
M0, then this would be consistent with the active ridge relocating
northward, towards the Kerguelen Plume. There seems to be a gen-
eral case for noticeable changes in spreading rate in other basins
in the Indian Ocean around this time, which suggests a general re-
organization before the onset of the CNS. There is an observed
decrease in spreading rate between M4 and M0 in the Weddell Sea
(Kovacs et al. 2002). Slowing spreading rates and subsequent ridge
jumps have been observed around M2-M0 in Natal Valley and So-
mali Basin off Africa (Marks & Tikku 2001), and around 118 Ma
in the Perth Basin off Western Australia (Mihut & M¨uller 1998).
Also in the Enderby Basin the rotation poles change from SW from
M9-M2 to ENE for M2-M1 and this is perhaps indicative of a tec-
tonic event.
Extinct spreading centres have been identified in several areas
affected by mantle plumes or by the re-organization of plate bound-
aries. The slowing of spreading and extinction of the palaeospread-
ing ridge, prior to a ridge jump has been observed at the Aegir
Ridge in the Norwegian Sea (Jung & Vogt 1997), the Labrador Sea
extinct ridge (Srivastava & Roest 1999), and the Phoenix Ridge
in the Drake Passage (Livermore et al. 2000). An example similar
to the Enderby Basin scenario is that associated with the Iceland
Plume, where ridge–hotspot interaction stranded the Jan Mayen mi-
crocontinent with a major ridge jump up to 400 km from the Aegir
Ridge to the Kolbeinsey Ridge (Grevemeyer et al. 1996). How-
ever, the closest example to a propagating ridge of Enderby Basin
length (almost 2500 km) is probably the Mascarene Basin ridge
that jumped north of the Seychelles microcontinent. Schlich (1982)
and Masson (1984) proposed that the Deccan-Reunion hotspot ini-
tiated seafloor spreading between India and the Seychelles along the
northern Carlsberg Ridge at Chron 27 (61 Ma), after spreading in
the Mascarene Basin became extinct. More recently, Bernard et al.
(2005) proposed a gradual ridge jump from the Mascarene Basin
toward the hotspot, separating the Seychelles microcontinent. Due
to the timing of the interpreted ridge jump in the Enderby Basin that
coincides with CNS (and therefore the lack of magnetic reversals),
we could not identify the detailed expression of the new seafloor
spreading ridge north of the Elan Bank. The proposed ridge jump
and microcontinent formation model are more fully explained in
Gaina et al. (2003). The plate kinematic model showing how ridge–
hotpot interaction at the Kerguelen Plume caused at least one ridge
jump, transferring most early ocean crust in this area to the Antarctic
plate and isolating the Elan Bank microcontinent from the Indian
continent is presented in Fig. 10.
5 DISCUSSION
The new data and observations presented here help to offer insights
into the evolution of the East Antarctic margin and its relationship
with the Kerguelan plume during breakup and early seafloor spread-
ing. In the case of the spreading system between the Enderby Basin
and Perth Basin, the incipient development of the Kerguelen Plume
is a likely factor during breakup and early spreading with varying
degrees of lithosphere–mantle and ridge–hotspot interaction. Here,
we discuss the tectonic and magmatic along-axis margin segmenta-
tion in the context of the growth and development of the Kerguelen
Plume.
5.1 Kerguelen Plume activity
There is evidence of a mantle thermal anomaly related to the Ker-
guelen Plume prior to its formation of the Southern Kerguelen
Plateau (SKP) at about 118 ±2 Ma. Coffinet al. (2002) provide a
comprehensive overview and summary of isotopic data characteriz-
ing the timing, distribution and magma output from the Kerguelen
Plume. The first identifiable surface expression related to a Kergue-
len mantle source (Bunbury Basalts, Western Australia (WA) around
132 Ma, Coffinet al. 2002) coincides with the time just prior to
the onset of seafloor spreading in the Perth and Enderby Basins
(∼130 Ma). There has also been minor Neocomian (137–127 Ma)
magmatic activity observed in the Perth Basin (Gorter & Deighton
2002). In general however, there is a lack of voluminous rift-related
magmatism or seaward dipping reflectors (SDR’s) in the Enderby
Basin and Perth Basin. This seems to suggest that early breakup oc-
curred with minor magmatic activity, about 15 Myr prior to the first
surface expression at the SKP ∼118 Ma and the Rajmahal Traps,
India (∼117 Ma).
Recent work on estimates of palaeolatitude for the Kerguelen
Plume (Antretter et al. 2002; O’Neill et al. 2003) suggest a south-
ward movement of the plume to its present day position, which is
now thought to be active underneath the Heard Island volcano. The
maximum diameter of a plume’s near-surface influence is estimated
to be 1500–2000 km (White & McKenzie 1989). By placing the early
plume at around 35◦S with a plume diameter up to 2000 km, then
this could encompass part of the Enderby and Perth margins, and
possibly part of the Cuvier margin. As White & McKenzie (1989)
suggest, volcanism at least for the southern WA margin could be
caused by a ‘broad thermal anomaly’from a mantle plume present
in the area shortly before breakup, particularly as the long period of
rifting and basin subsidence prior to breakup had little associated
volcanism. While presumably part of the WA and Antarctic mar-
gin was underlain by the Kerguelen Plume not all margin segments
are truly volcanic. The initial output of the plume was not hot or
voluminous enough despite the thinned passive margin lithosphere
and decompression melting for wide-scale SDR’s, but plume-related
magmatism appears to manifest at the surface in several separate lo-
cations suggesting a series of aerial and subaerial alkaline basaltic
events (Coffinet al. 2002).
Ideas about plume–lithosphere interaction include a variety of
possibilities about plume conduit size and shape with their growth
and development over a lifecycle. Coffinet al. (2002) suggests a
plume may not behave with only a plume-head configuration but a
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
14 C. Gaina et al.
Figure 10. Plate reconstruction models for early breakup between India (IND) and Antarctica (ANT) based on new seafloor spreading isochrons between
the Gunnerus Ridge and Bruce Rise, and the model in Gaina et al. (2003) including the subsequent ridge jump to north of Elan Bank (EB) and growth of
the Kerguelen Plume (KP). Poles of rotation are from Gaina et al. (2003). Stars indicate hotspots (from left to right: Deccan-Seychelles, Marion, Kerguelen,
St. Paul-Amsterdam) in Indian Ocean reference frame based on O’Neill et al.. (2003). BB, Bunbury Basalts; BR, Bruce Rise; NP, Naturaliste Plateau; Md,
Madagascar; RJ, Rajmahal Traps and Sri, Sri Lanka. C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 15
series of smaller conduits may also originate from the core–mantle
boundary so it can exist both at on-ridge or off-ridge locations. For
example, the Ontong-JavaLIP is inter preted as bi-modal (Berrovicci
& Mahoney 1994). If the Bunbury Basalts are dated at about
132 Ma and Rajmahal Traps are about 117 Ma, and they are isotopi-
cally linked to a Kerguelen Plume mantle source, then this implies
that igneous activity pre-dates Kerguelen LIP formation. In compar-
ison, dating of Reunion hotspot magmatism indicates it was active
before the voluminous production at the Deccan Traps, and appears
far to the northwest of the flood basalt outcrops (Mahoney et al.
1995; O’Neill et al. 2003). In particular, Ar–Ar dating yields ages
around 73–72 Ma, while the bulk of Deccan flood basalts are around
66–65 Ma. Indeed these models suggest that accumulation of the Re-
union plume-head under the (slow moving) continental lithosphere
did not just cause the huge surface expression of the Deccan traps,
but had an earlier surface expression as intrusions 300–400 km north
from the main Deccan body.
5.2 Plume driven breakup?
The segmentation observed along the east Antarctic margin between
Gunnerus Ridge and Bruce Rise raises the question of the nature of
its formation and the magmatic conditions in the zone of breakup and
early seafloor spreading. It has been suggested that a primary cause
of along-margin segmentation is due to the variation in the amount of
underplated igneous crust and the strength of the lithosphere (e.g.
Callot et al. 2002). Work in the present-day Afar rift (Ebinger &
Casey 2001) suggests along axis magmatic segmentation, related to
strain distribution (faulting and dyking) and a Palaeogene mantle
plume (asthenosphere temperature). Callot et al. (2002) emphasize
the important role of small-scale intrusions of mantle material in
breakup from their observations and analogue models for the North
Atlantic divergent margins. For example, across all spreading seg-
ments we observe M9o (130.2 Ma) as the southern limit of identifi-
able seafloor spreading lineations. The distance between the oldest
magnetic anomaly, M9o (130.2 Ma) and the interpreted continent–
ocean boundary is between 20 and 50 km in the western, eastern
and parts of central Enderby basin (Fig. 5). However, the distance
increases between the ENE-trending M9o lineation and the MCA
anomaly (interpreted as COB) to around 100km at the eastern end
of the MCA segment ∼70◦E. Seismic reflection profiles image un-
equivocal oceanic crust in this zone (e.g. Joshima et al. 2001; Stagg
et al. 2004) and there are no coherent magnetic anomalies or observ-
able lineations past M9o before the COB. The crust betweenthe shelf
break and MCA anomaly could be a transitional type crust formed by
stretched continental crust that was subsequently modified by mag-
matic intrusions and/or mantle exhumation/initial oceanic accretion
as described by Shilington et al.. (2006) along the Newfoundland
margin. In this case, the MCA anomaly might reflect a significant
amount of melt emplacement, or as in our alternative model, a ridge
propagator. The East Coast Magnetic Anomaly(ECMA) off the east-
ern U.S. continental margin, is characterized by a prominent mag-
netic anomaly that corresponds to an abrupt change in seismic and
magnetic profiles at the edge of a continental transition zone about
70 km wide. The MCA and ECMA are both prominent anomalies
whose formation is linked to specific magmatic conditions. At the
ECMA Holbrook et al. (1994) infer the emplacement of highly mafic
material in the continent–ocean transition zone during rifting using
seismic constraints on crustal thickness and magnetic susceptibil-
ity models of the East Coast and Brunswick magnetic anomalies.
The resulting model for the ECMA suggests the amount of igneous
material produced would require either a mantle plume, for which
there is no evidence, or increased small-scale convection causing
increased melt production during rifting. Its origin in relation to a
plume or non-plume setting is still debated despite numerous studies
(cf. White & McKenzie 1989; Talwani & Abreu 2000).
There are some notable differences between the ECMA and MCA
anomalies, the ECMA is twice as long and lies adjacent to a narrow
passive margin segment, whereas the MCA lies against a wide zone
of stretched continental crust off Prydz Bay. However different in
their rift and magmatic settings, a significant amount of melt em-
placement is implied to form both the ECMA and MCA. Initial melt
production at the MCA segment would be reasonably voluminous
as early half-spreading rates are estimated to be over 48 mm yr–1.
As opposed to the ECMA, seaward dipping reflectors (SDR) are
scarce (we have tentatively interpreted SDR south of the central En-
derby, Fig. 9). The Antarctic margin of the central Enderby Basin
appears to be in a zone of previously thinned lithosphere or transi-
tional crust. As an alternative, the interpreted MCA could be rather
compared with the Blake Spur Magnetic Anomaly (BSMA) situated
offshore of ECMA in presumably transitional oceanic crust. Many
studies (e.g. White & McKenzie 1995; Marks et al. 1999) describe
how large quantities of decompression melting can be generated
during rifting and breakup in either areas of previously stretched
lithosphere or zones thinner than the surrounding lithosphere. In
the case of the Enderby Basin, both cases apply, where there is
the thinned area of crust of the Lambert Graben-Prydz Bay adja-
cent to the east of an Archaean Craton (the Napier Complex, at the
Enderby Land Promontory). There is also the likely factor of the
incipient thermal anomaly related to the Kerguelen Plume preferen-
tially developing near the boundary of this zone (e.g. Courtillot et al.
1999). In general, LIPs are thought to form preferentially at cratonic
boundaries or at a contrast in lithospheric thickness as these act as
a‘focusing’mechanism (e.g. Callot et al. 2002). One perspective
is that the relief of the base of the lithosphere can act like an in-
verted drainage system that either traps plume material or channels
it to zones of higher relief. Nielsen et al. (2002) use this idea as a
basis to explain observations of nearby volcanic and non-volcanic
margin segments, off the Greenland and Labrador margins, near the
Icelandic mantle plume; whereby plume material can be channeled
into lithospheric thin spots while cratons may act as barriers as melt-
ing, cooling and dehydration impedes lateral flow due to increased
viscosity and decreased buoyancy. The initial surface expression of
Kerguelen Plume at SKP is where presumably there was the weakest
area of crust for early breakup magmatism and seafloor spreading to
nucleate. The combined effects of thinner than normal crust (more
decompression melting) and a developing mantle plume (increasing
thermal anomaly) could explain the crustal features observed in this
sector.
An additional observation is important for linking our observa-
tions to the mode of rifting along the Enderby margin, and that
is related to the Precambrian lower crustal rocks (e.g. high grade
granulites and igneous charnockites) exposed along the coastline
of Enderby Land, Kemp Land and Mac.Robertson Land (Young
et al. 1997), which constitute the most well exposed area of the East
Antarctic Precambrian Shield. The lower crust is exposed in rifts
where the strength ratio between the upper and lower crust is large,
resulting in ‘metamorphic core complex’mode of extension (Wijns
et al. 2005). This ratio is large when the lower crust is relatively
weak compared to the upper crust, as is the case when rifting oc-
curs above anomalously hot mantle. Our observations suggest that
it might be possible that this margin formed in a metamorphic core
complex (MCC) mode. MCC-mode rifting was likely triggered by
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
16 C. Gaina et al.
the thermal mantle anomaly associated with the formation of the
margin-parallel MCA magnetic anomaly, preceding the arrival of
the Kerguelen Plume head. Rifting in this mode resulted in expo-
sure of lower crustal rocks at the coastline and a relatively narrow
rift, as expressed in the small total width of the conjugate eastern
Indian-Enderby margin pair (Fig. 10). The Antarctic continental
margin is the wider area of the conjugate rift zone, and this may be a
result of more resistant ‘lock zones’at the ends of the rift located at
the Gunnerus Ridge and Bruce Rise, following the model of Dunbar
& Sawyer (1996).
Although a thermal anomaly seems to have influenced this mar-
gin after breakup (leading to northward relocation of plate boundary
and microcontinent formation), the break-up central Gondwanaland
was probably triggered by passive rifting driven by changes in plate
driving forces, not active rifting driven by mantle upwelling. Evi-
dence for a major thermal mantle anomaly, such as that associated
with the Iceland Plume during breakup of the Norwegian-Greenland
Sea, is missing along the Enderby and Perth Basin margins. The
most likely driving force for the separation of India from Antarctica
and Australia was the gradual subduction of the Neotethys spread-
ing ridge north of India (e.g. Stampfli 2000; Stampfli and Borel
2004), leading to the reduction of ridge push forces and increasing
northward-directed slab pull forces to the north of India.
6 CONCLUSIONS
The timing and direction of early seafloor spreading in the area off
the Antarctic margin, once conjugate to part of the Southern Greater
Indian margin, along the largely unknown region of the Enderby
Basin and Davis Sea, has been analysed using a new data compila-
tion. These data provide a basis to examine along-axis breakup and
early seafloor spreading across the margin. A considerableamount of
tectono-magmatic variation is observed both along and across-axis,
related to spreading rate, magma production and distance from the
developing Kerguelen Plume. For example, the prominent magnetic
anomaly boundary signal and sharp basement step correlated with
the Mac.Robertson Coast Anomaly (MCA) is not observed else-
where in the Enderby Basin, Princess Elizabeth Trough or Davis
Sea. In the central Enderby Basin we show evidence for an aban-
doned ‘fossil’spreading centre that might continue to the west of
the Kerguelen Plateau, east of Gunnerus Ridge. The estimated tim-
ing of its extinction corresponding to the early surface expression
of the Kerguelen Plume at the Southern Kergulen Plateau around
120 Ma and the subsequent formation of the Elan Bank microconti-
nent. Alternatively, the ridge jump occurred only in the central En-
derby basin, due to the proximity of the Kerguelan plateau, whereas
seafloor spreading continued in the western Enderby basin and con-
jugate south of Sri Lanka basin.
It is likely a mantle temperature anomaly predated the early sur-
face outpouring/steady state magmatic production of the Kerguelen
LIP. However the scarcity of seaward dipping reflectors in the
Enderby Basin and Perth Basin suggests that no major mantle plume
existed during breakup (∼130 Ma). Our observations suggest that
it is possible that the Antarctic-Enderby margin formed in a meta-
morphic core complex (MCC) mode that was likely triggered by a
thermal mantle anomaly in this area, preceding the arrival of the
Kerguelen Plume head.
ACKNOWLEDGMENTS
The authors wish to thank Steve Cande, Karsten Gohl and Tim Min-
shull for thorough reviews that improved the paper considerably.
BJB acknowledges the support of a Geoscience Australia post-
graduate scholarship and thanks Howard Stagg and Jim Colwell
for the discussion of their seismic reflection profile interpretation,
and Alexey Goncharov for discussion of the refraction data. We
acknowledge PMGRE and VNIIOK in St. Petersburg for making
magnetic data collected by Russian Antarctic expeditions (31–47)
available for this work. Data were made available under the pro-
visions of a collaborative research agreement between Geoscience
Australia, PMGRE and VNIIOK. The figures in this paper were cre-
ated using GMT (Wessel & Smith 1991), apart from Figs 7–9 which
include the seismic reflection profiles, re-produced with permis-
sion from Geoscience Australia and JNOC/JOGMEC. The synthetic
magnetic profiles were calculated using MODMAG by Mendel et al.
2005.
REFERENCES
Antretter, M., Steinberger, B., Heider, F., Soffel, H. & Banergee, B., 2002.
Paleolatitudes of the Kerguelen hotspot: new paleomagnetic results and
dynamic modelling, Earth Planet. Scie. Lett., 203, 635–650.
Banergee, B., Sengpta, J. & Banergee, P.K., 1995. Signals of Bar remian
(116 Ma) or younger oceanic crust beneath the Bay of Bengal along 14N
latitude between 81E and 93E, Mar. Geol., 128, 17–23.
Bercovici, D. & Mahoney, J., 1994. Double flood basalts and plume
head separation at the 660 kilometer discontinuity, Science, 266, 1367–
1369.
Bergh, H.W., 1987. Underlying fracture zone nature of Astrid Ridge off
Antarctica, Queen Maud Land, J. Geophys. Res., 92, 475–484.
Bernard, A., Munschy, M., Rotstein, Y. & Sauter, D., 2005. Refined spreading
history of the Southwest Indian Ridge for the last 96 Ma, with the aid of
satellite data, Geophys. J. Int., 16, 765–778.
Borissova, I., Moore, A., Sayers, J., Parums, R., Coffin, M.F. & Symonds,
P.A., 2002. Geological framework of the Kerguelen Plateau and adjacent
ocean basins, Geoscience Australia Record 2002/05.
Borissova, I., Coffin, M.F., Charvis, P. & Operto, S., 2003. Structure and
development of a microcontinent; Elan Bank in the southern Indian Ocean,
Geochem. Geophys. Geosys., 4.
Bown, J.W. & White, R.S., 1995. Effect of finite extension rate on
melt generation at rifted continental margins, J. Geophys. Res., 100,
18 011–18 029.
Buchel, G., 1994. Gravity, Magnetic and Structural Patterns at the Deep-
Crustal Plate Boundary Zone between West Gondwana and East Gond-
wana in Sri-Lanka, Precambrian Res., 66, 77–91.
Callot, J.P., Geoffroy, L. & Brun, J.P., 2002. Development of volcanic
passive margins: three-dimensional laboratory models, Tectonics, 21,
doi:10.1029/2001TC901019.
Charvis, P. & Operto, S., 1999. Struture of the Cretaceous Kerguelen vol-
canic province (southern Indian Ocean) from wide-angle seismic data, J.
Geodyn., 28, 51–71.
Coffin, M.F., 1992. Subsidence of the Kerguelen Plateau: the Atlantis con-
cept. In: S.W. Wise, Jr., R. Schlich and et al. (Editors), Proc. ODP, Sci.
Results, pp. 945–949.
Coffin, M., Pringle, M.S., Duncan, R.A., Gladczenko, T.P., Storey, M.,
M¨uller, R.D. & Gahagan, L.A., 2002. Kerguelen Hotspot magma output
since 130 Ma, J. Petrol., 43, 1121–1139.
Colwell, J.B., Symonds, P.A. & Crawford, A.J., 1994. The nature of the Wal-
laby (Cuvier) Plateau and other igneous provinces of the west Australian
margin, AGSO, J. Aus. Geol. Geophys., 15, 137–156.
Courtillot, V., Jaupart, C., Manighetti, I., Tapponier, P. & Besse, J., 1999. On
causal links between flood basalts and continental breakup, Earth Planet.
Sci. Lett., 166, 177–195.
Curray, J.R., 1991. Geological history of the Bengal Geosyncline, J. Explor.
Geophys., 12, 209–219.
Desa, M, Ramana, M.V. & Ramprasad, T., 2006. Seafloor spreading magnetic
anomalies south of Sri Lanka, Marine Geol., 229, 227–240.
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 17
Dunbar, J.A. & Sawyer, D.S., 1996. Three-dimensional dynamical model
of continental rift propagation and margin plateau formation, J. Geophys.
Res., 101, 27 845–27 863.
Ebinger, C.J. & Casey, M., 2001. Continental breakup in magmatic
provinces: an Ethiopian example, Geology, 29, 527–530.
Fedorov, L.V., Ravich, M.G. & Hofmann, J., 1982. Geologic comparison of
southeastern peninsular India and Sri Lanka with a part of East Antarc-
tica (Enderby Land, MacRobertson Land, & Princess Elizabeth Land), in
Antarctic Geoscience, pp. 73–78, ed. C., Craddock, International Union
of Geological Sciences (IUGS), Oslo.
Frey, F.A. et al., 2000. Origin and evolution of a submarine large ig-
neous province; the KerguelenPlateau and Broken Ridge, souther n Indian
Ocean, Earth Planet. Scie. Lett., 176, 73–89.
Gaina, C., M¨uller, R.D., Brown, B. & Ishihara, T., 2003. Microcontinent
formation around Australia, in Evolution and Dynamics of the Australian
Plate, pp. 405–416, eds Hillis, R.R. & M¨uller, R.D., Geological Society
of America, Boulder, CO, USA.
Gandyukhin, V., Gouseva, Y., Kudryavtsev, G., Ivanov, S. & Leitchenkov,
G., 2002. Crustal structure, seismic stratigraphy and tectonic history of the
Cosmonaut Sea sedimentary basin (Antarctica, southern Indian Ocean),
Exploration and Protection of Mineral Resources (original in Russian),
9, 27–31.
Gladczenko, T.P. & Coffin, M.F., 2001. Kerguelen Plateau crustal structure
and basin formation from seismic and gravity data, J. Geophys. Res., 106,
16583–516 601.
Golynsky, A.V., Alyavdin, S.V., Masolov, V.N., Tscherinov, A.S. &
Volnukhin, V.S., 2002. The composite magnetic anomaly map of the East
Antarctic, Tectonophysics, 347, 109–120.
Goslin, J. & Schlich, R., 1976. Structural limits of the South Crozet Basin;
relations to Enderby Basin and the Kerguelen-Heard plateau, Int’l. Geol.
Cong., 25(3), 884–885.
Gorter, J.D. & Deighton, I., 2002. Effects of igneous activity in the offshore
northern Perth Basin—evidence from petroleum exploration wells, 2D
seismic and magnetic surveys, in Western Australian Basins Symposium
III, pp. 875–899, eds Anon, PESA, Perth, Australia.
Goslin, J. & Schlich, R., 1982. Structural limits of the southern Crozet Basin
and its relation to the Enderby Basin and the Kerguelen-Heard Plateau, in
Antarctic Geoscience, pp. 79–85, ed. Craddock, C., International Union
of Geological Sciences (IUGS), Oslo.
Gradstein, F.M., Agterberg, F.P., Ogg, J.G., Hardenbol, J., van Veen, P.,
Thierry, J. & Huang, Z., 1994. A Mesozoic timescale, J. Geophys. Res.,
99, 24 051–24 074.
Grevemeyer, I., Weigel, W., Whitmarsh, R.B., Avedik, F. & Dehghani, G.A.,
1996. The Aegir Rift—crustal structure of an extinct spreading axis, Ma-
rine Geophys. Res., 19, 1–23.
Grunow, A.M., 1999. Gondwana events and palaeogeography: a palaeomag-
netic review, J. Afric. Earth Sci., 28, 53–69.
Harley, S.L. & Hensen, B.J., 1990. Archaean and Proterozoic high-grade
terranes of East Antarctica (40–80 E); a case study of diversity in gran-
ulite facies metamorphism, in High-Temperature Metamorphism and
Crustal Anatexis, pp. 320–370, eds Ashworth, J.R.& Brown, M.C., Unwin
Hyman, London.
Holbrook, W.S., Reiter, E.C., Purdy, G.M., Sawyer, D., Stoffa, P.L., Austin,
J.A. & Makris, J., 1994. Deep structure of the U.S. Atlantic continen-
tal margin, offshore South Carolina, from coincident ocean bottom and
multichannel seismic data, J. Geophys. Res., 99, 9155–9178.
Huang, P.Y. & Solomon, S.C., 1998. Centroid depths of mid-ocean ridge
earthquakes; dependence on spreading rate, J. Geophys. Res., 93, 13 445–
13 477.
Jellinek, M.A., Helge, M., Gonnerman, M. & Richards, M.A., 2003. Plume
capture by divergent plate motions: implications for the distribution of
hotspots, geochemistry of mid-ocean basalts and estimates of the heat
flux at the core-mantle boundary, Earth Planet. Sci. Lett., 205, 361–378.
Jokat, W., Boebel, T., Koenig, M. & Meyer, U., 2003. Timing
and geometry of early Gondwana breakup, J. Geophys. Res., 108,
doi:10.1029/2002JB001802.
Joshima, M., Ishihara, T., Nakajima, T., Sugiyama, K., Tsuchida, K., Kato,
A., Murakami, F. & Brown, B., 2001. Preliminary results of the TH99
geological and geophysical survey in the Cooperation Sea and Prydz Bay
area, Polar Geosci., 14, 235–244.
Jung, W.Y. & Vogt, P.R., 1997. A gravity and magnetic anomaly study of the
extinct Aegir Ridge, Norwegian Sea, J. Geophys. Res., 102, 5065–5089.
Kleimenova, N.G., Kozyreva, O.V., Schott, J.-J., Bitteryl, M., Bitterly, J. &
Ivanova, P.,2003. Dayside geomagnetic Pc5 pulsations in the conditions of
a strongly disturbed solar wind during the magnetic storm of 21 February
1994, Int. J. Geomagnet. Aeron., 3, 229–244.
Kopp, H., Kopp, C., Phipps Morgan, J., Flueh, E.R. & Weinrebe, W., 2003.
Fossil hot spot-ridge interaction in the Musicians Seamount Province:
geophysical investigations of hot spot volcanism at volcanic elongated
ridges, J. Geophys. Res., 108, doi:10.1029/2002JB002015.
Kovacs, L.C., Morris, P., Brozena, J. & Tikku, A., 2002. Seafloor spreading
in the Weddell Sea from magnetic and gravity data, Tectonophysics, 347,
43–64.
Kriegsman, L.M., 1995. The Pan-African event in East Antarctica: a view
from Sri Lanka and the Mozambique Belt, Precambrian Res., 75, 263–
277.
Lawver, L.A., Gahagan, L.M. & Dalziel, I.W.D., 1998. A tight fit–Early
Mesozoic Gondwana, a plate reconstruction perspective, Memoirs Nat’l.
Inst. Polar Res., 53, 214–229.
Lisker, F., Brown, R. & Fabel, D., 2003. Denudational and thermal his-
tory along a transect across the Lambert Graben, northern Prince Charles
Mountains, Antarctica, derived from apatite fission track thermochronol-
ogy, Tectonics, 22, doi:10.1029/2002TC001477.
Livermore, R. et al., 2000. Autopsy on a dead spreading center: the Phoenix
Ridge, Drake Passage, Antarctica, Geology, 28, 607–610.
Mahoney, J.J., Jones, B.W., Frey, F.A., Salters, V.J.M., Pyle, D.G. & Davis,
H.L., 1995. Chem. Geol., 120, 315–345.
Marks, K.M. & Tikku, A.A., 2001. Cretaceous reconstructions of East
Antarctica, Africa, and Madagascar, Earth Planet. Scie. Lett., 186, 479–
495.
Marks, K.M., Stock, J.M. & Quinn, K.J., 1999. Evolution of the Australian-
Antarctic discordance since Miocene time, J. Geophys. Res., 104, 4967–
4981.
Masson, D.G., 1984. Evolution of the Mascarene basin, western Indian
Ocean, and the significance of the Amirante arc, Marine Geophys. Res.,
6, 365–382.
Masters, G., Laske, G., Bolton, H. & Dziewonski, A., 2000. The relative
behavior of shear velocity, bulk sound speed, and compressional velocity
in the mantle: implications for chemical and thermal structure, in Earth’s
Deep Interior, pp. 63–87, eds Karato, S., AGU Geophysical Monograph.
McAdoo, D. & Laxon, S., 1997. Antarctic tectonics: constraints from an
ERS-1 satellite marine gravity field, Science, 276, 556–560.
Mendel, V.; Munschy, M. & Sauter, D., 2005. MODMAG, a MATLAB
program to model marine magnetic anomalies, Comput. & Geosci., 589–
597.
Mihut, D., 1997. Breakup and Mesozoic seafloor spreading between the
Australian and Indian plates, Ph.D. thesis, University of Sydney, Sydney,
NSW, Australia.
Mihut, D. & M¨uller, R.D., 1998. Volcanic margin formation and Mesozoic
rift propagators in the Cuvier Abyssal Plain off Western Australia, J.
Geophys. Res., 103, 27 135–27 149.
Mizukoshi, I., Sunouchi, H., Saki, T., Sata, S. & Tanahasi, M., 1986. Prelim-
inary report of geological and geophysical surveys off Amery Ice Shelf,
East Antarctica, Memoirs Nat’l. Inst. Polar Res. Special Issue, 43, 48–61.
Morelli, A. & Danesi, S., 2004. Seismological imaging of the Antarctic
continental lithosphere: a review, Glob. Planet. Change, 42, 155–165.
Murakami, F., Ishihara, T., Odo, H., Fujimoto, M., Oda, H., Sugiyama, K.
& Joshima, M. 2000. Preliminary report on geological and geophysical
survey results in the Princess Elizabeth Trough and its vicinity by R/V
Hakurei-maru, Polar Geosci., 13, 165–186.
M¨uller, R.D., Royer, J.Y. & Lawver, L.A., 1993. Revised plate motions rel-
ative to hotspots for combined Atlantic and Indian Ocean hotspot tracks,
Geology, 16, 275–278.
M¨uller, R.D., Roest, W.R., Royer, J.Y., Gahagan, L.M. & Sclater, J.G., 1997.
Digital isochrons of the worlds ocean floor, J. Geophys. Res., 102, 3211–
3214.
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
18 C. Gaina et al.
M¨uller, R.D., Gaina, C., Tikku, A.A., Mihut, D., Cande, S.C. & Stock, J.M.,
2000. Mesozoic/Cenozoic tectonic events around Australia, in The History
and Dynamics of Global Plate Motions, pp. 161–188, eds Richards, M.A.,
Gordon, R. & van der Hilst, R., AGU Geophysical Monograph.
M¨uller, R.D., Gaina, C., Roest, W.R. & Lundbek Hansen, D., 2001. A recipe
for microcontinent formation, Geology, 29, 203–206.
N.A.M.A.G –North American Magnetic Anomaly Group, 2002. Digital data
grids for the magnetic anomaly map of North America, USGS Open-File
Report 02-414.
NGDC, 1998. Worldwide Marine Geopyhsical Data GEODAS CD-ROM,
version 4.0, 1998, National Oceanic and Atmospheric Administration, US
Department of Commerce, Boulder, CO.
Nicolaysen, K., Bowring, S., Frey, F., Weis, D., Ingle, S., Pringle, M.S.
& Coffin, M.F., 2001. Provenance of Proterozoic garnet-biotite gneiss
recovered from Elan Bank, Kerguelen Plateau, southern Indian Ocean,
Geology, 29, 235–238.
Nielsen, T.K., Larsen, H.C. & Hopper, J.H., 2002. Contrasting rifted margin
styles south of Greenland: implications for mantle plume dynamics, Earth
Planet. Scie. Lett., 200, 271–286.
Nogi, Y., Seama, N., Isezaki, N., Funaki, M. & Kaminuma, K., 1991. The
directions of the magnetic anomaly lineations in Enderby Basin, off
Antarctica, in Sixth International Symposium on Antarctic Earth Sciences,
pp. 446, eds Anon, Cambridge University Press, New York.
Nogi, Y., Seama, N., Isezaki, N. & Fukuda, Y., 1996. Magnetic anomaly
lineations and fracture zones deduced from vector magnetic anomalies in
the West Enderby Basin, in Weddell Sea Tectonics and Gondwana Break-
up, pp. 265–273, eds Storey, C., King, C. & Livermore, R., Geological
Society of London, London.
O’Neill, C., M¨uller, R.D. & Steinberger, B., 2003. Geodynamic implications
of moving Indian Ocean hotspots, Earth Planet. Sci. Lett., 215, 151–
168.
Osler, J.C. & Louden, K.E., 1995. Extinct spreading centre in the Labrador
Sea: crustal structure from a two dimensional seismic refraction velocity
model, J. Geophys. Res., 100, 22 661–22 278.
Parsons, B. & Sclater, J.G., 1977. An analysis of the variation of ocean floor
bathymetry and heat flow with age, J. Geophys. Res., 82(5), 803–827.
Patriat, P. & Segoufin, J., 1988. Reconstruction of the Central Indian Ocean,
Tectonophysics, 155, 211–234.
Powell, C.M., Roots, S.R. & Veevers, J.J., 1988. Pre-breakup continental
extension in East Gondwanaland and the early opening of the eastern
Indian Ocean, Tectonophysics, 155, 261–283.
Ramana, M.V., Nair, R.R. & Sar ma, K.V.L.N.S., 1994a. Mesozoic anomalies
in the Bay of Bengal, Earth Planet. Scie. Lett., 121, 469–475.
Ramana, M.V. et al., 1994b. Magnetic studies in the northern Bay of Bengal,
Marine Geophys. Res., 16, 237–242.
Ramana, M.V., Ramprasad, T. & Desa, M., 2001. Seafloor spreading mag-
netic anomalies in the Enderby Basin, East Antarctica, Earth Planet. Scie.
Lett., 191, 241–255.
Reid, I. & Jackson, H.R., 1981. Ocean spreading rate and crustal thickness,
Marine Geophys. Res., 5, 165–172.
Robb, M., Taylor, B. & Goodlife, A.M., 2005. Re-examination of the mag-
netic lineations of the Gascoyne and Cuvier abyssal plains, off NW
Australia, Geophys. J. Int., 163, 42–55.
Roeser, H.A., Fritsch, J. & Hinz, K., 1996. The development of the crust
off Donning Maud Land, East Antarctica, in Weddell Sea Tectonics and
Gondwana Break-up, eds Storey, B.C., King, E.C. & Livermore, R.A.,
Geology Society Special Publication, Bath, UK.
Rotstein, Y., Munschy, M. & Bernard, A., 2001. The Kerguelen Province
revisited: additional constraints on the early development of the Southeast
Indian Ocean, Marine Geophys. Res., 22, 81–100.
Royer, J.Y. & Coffin, M.F., 1992. Jurassic to Eocence plate tectonic recon-
structions in the Kerguelen Plateau region, Proc. ODP Sci. Rept., 120,
917–928.
Sandwell, D. & Smith, W.H.F., 2005. Retracking ERS-1 altimeter wave-
forms for optimal gravity field recovery, Geophys. J. Int., 163, 79–
89.
Sauter, D., Carton, H, Mendel, V., Munschy, M., Romevaux-Jesti, C., Schott,
J-J. & Whitechurch, H., 2004. Ridge segmentation and magnetic structure
of the Southwest Indian Ridge (at 50 degrees 30E, 55 degrees 30E and 66
degrees 20E): implications for magmatic processes at ultraslow-spreading
centers, Geochem., Geophys., Geosys., 5, doi:10.1029/2003GC000581.
Schlich, R., 1982. The Indian Ocean: aseismic ridges, spreading centers and
oceanic basins, in The Ocean Basins and Margins, pp. 47–51, eds Nairn,
A. & Stehli, F., Plenum, New York.
Shillington, D. et al., 2006. Evidence for asymmetric nonvolcanic rift-
ing and slow incipient oceanic accretion from seismic refection
data of the Newfoundland margin, J. Geophys. Res., 111, BO9402,
doi:10.1029/2005JB003981
Shiraishi, K., Ellis, D.J., Hiroi, Y., Fanning, C.M., Motoyoshi, Y. & Nakai,
Y., 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: im-
plications for Gondwana assembly, Geology, 102, 47–65.
Smith, W.H.F. & Sandwell, D.T., 1997. Global sea floor topography
from satellite altimetry and ship depth soundings, Science, 277, 1956–
1962.
Srivastava, S.P. & Keen, C.E., 1995. A deep seismic reflection reflection
profile across the extinct mid-Labrador Sea spreading centre, Tectonics,
14, 372–389.
Srivastava, S.P.& Roest, W.R.,1995. Nature of thin crust across the southwest
Greenland margin and its bearing on the location of the ocean-continent
boundary, in Proceedings of the NATO-ARW workshop on Rifted Ocean-
Continent Boundaries, pp. 95–120, eds Banda, E., Torne, M. & Talwani,
M., Kluwer Academic Publisher, Spain.
Srivastava, S.P. & Roest, W.R., 1999. Extent of the oceanic crust in the
Labrador Sea, Mar. Petrol. Geol., 16, 65–84.
Stagg, H.M.J., 1985. The structure and origin of Prydz Bay and
MacRobertson Shelf, East Antarctica, Tectonophysics, 114, 315–
340.
Stagg, H.M.J. & Colwell, J.B., 2003. The deep-water East Antarctic conti-
nental margin, from 38–152E: overview of a new integrated geophysical
data set, in Terra Nostra—Abstracts from the 9th International Sympo-
sium on Antarctic Earth Sciences, pp. 307–308, ed. D.K. F¨utterer, Alfred
Wegener Stiftung, Berlin, Potsdam.
Stagg, H.M.J., Colwell, J.B., Direen, N.G., O’Brien, P.E., Bernardel, G.,
Borissova, I., Brown, B.J., Ishihara, T., 2004. Geology of the continental
margin of Enderby and MacRobertson Lands, East Antarctica: insights
from a Regional data set, Marine Geophys. Res., doi:10.1007/s11001-
005-1316-1.
Stampfli, G.M., 2000. Tethyan Oceans, in Tectonics and Magmatism in
Turkey and Surrounding Area, pp. 1–23, eds Bozkurt, E., Winchester,
J.A. & Piper, J.D.A., Geological Society of London Special Publication,
London.
Stampfli, G.M. & Borel, G.D., 2004. The TRANSMED transects in space
and time: constraints on the paleotectonic evolution of the Mediterranean
domain, in The TRANSMED Atlas: The Mediterranean Region from Crust
to Mantle, pp. 53–80, eds Cavazza, W., Roure, F., Spakman, W., Stampfli,
G.M. & Ziegler, P., Springer Verlag, London.
Talwani, M. & Abreu, V., 2000. Inferences regarding initiation of oceanic
crust formation from the U.S. East Coast continental margin and conju-
gate South Atlantic margins, in Atlantic Rifts and Continental Margins,
pp. 211–234, eds Mohriak, W. & Talwani, M., American Geophysical
Union Geophysical Monograph 115, Washington.
Tikku, A.A. & Cande, S.C., 1999. The oldest magnetic anomalies in the
Australian-Antarctic Basin: are they isochrons?, J. Geophys. Res., 104,
661–677.
Wellman, P. & Tingey, R.J., 1982. A gravity survey of Enderby and Kemp
Lands, in Antarctic Geoscience, pp. 937–940, ed. Craddock, C., Interna-
tional Union of Geological Sciences (IUGS), Oslo.
Wessel, P. & Smith, W.H.F., 1991. Free software helps map and display data,
EOS Trans. AGU, 72, 445–446.
White, R.S. & McKenzie, D.P., 1989. Magmatism at rift zones: the generation
of volcanic continental margins and flood basalts, J. Geophys. Res., 94,
7685–7729.
White, R.S., 1992. Magmatism during and after continental breakup, in Mag-
matism and the Causes of Continental Breakup, pp. 1–16, eds Storey, B.C.,
Alabaster, T. & Pankhurst, R.J., Geological Society Special Publication,
London.
C
2007 The Authors, GJI
Journal compilation C
2007 RAS

June 3, 2007 16:26 GeophysicalJournalInternational gji3450
Breakup and early seafloor spreading between India and Antarctica 19
White, R.S. & McKenzie, D., 1995. Mantle plumes and flood basalts, J.
Geophys. Res., 100, 17 543–17 586.
Wijns, C., Weinberg, R., Gessner, K. & Moresi, L., 2005. Mode of crustal
extension determined by rheological layering, Earth Planet. Scie. Lett.,
236, 120–134.
Yoshida, M., Funaki, M. & Vitanage, P.W., 1992. Proterozoic to Mesozoic
East Gondwana; the juxtaposition of India, Sri Lanka, and Antarctica,
Tectonics, 11, 381–391.
Yoshida, M., Bindu, R.S., Kagami, H., Rajesham, T., Santosh, M. &
Shirahata, H., 1996. Geochronologic constraints of granulite terranes of
South India and their implications for the Precambrian assembly of Gond-
wana, J. Southeast Asian Earth Sci., 14, 137–147.
Young,D.N., Zhao, J., Ellis, D.J.& McCulloch, M.T., 1997. Geochemical and
Sr-Nd isotopic mapping of source provinces for the Mawson charnock-
ites, east Antarctica: implications for Proterozoic tectonics and Gondwana
reconstruction, Precambrian Res., 86, 1–19.
C
2007 The Authors, GJI
Journal compilation C
2007 RAS