40Ar/39Ar geochronology and the paleoposition of Christmas Island (Australia), Northeast Indian Ocean

Abstract

The Christmas Island Seamount Province is an extensive zone of volcanism in the Northeast Indian Ocean, consisting of numerous submerged seamounts and flat-topped guyots. Within this region lies two subaerial island groups, Christmas Island, and the Cocos Keeling archipelago. Christmas Island has experienced multiple episodes of volcanism that are exposed sporadically along its coastline. Here, we dated these volcanics using 40Ar/39Ar geochronology and analysed them for paleomagnetism. The oldest exposed volcanism occurred in the Eocene between 43 and 37 Ma. This is followed by a time gap of ~ 33 million years, before the eruption of a younger episode of Pliocene age (4.32 ± 0.17 Ma). It has, however, been suggested by previous workers that there is a much older Late Cretaceous event beneath the limestone which is unexposed. In addition, this study conducted the first paleomagnetic analysis of samples from Christmas Island to determine its paleoposition and the paleomagnetic polarity of the sampled sites. Two normal and two reversal magnetic events have been recorded, that agree with the geomagnetic reversal timescales. Late Eocene (38-39 Ma) palaeomagnetic data suggest a palaeolatitude of − 43.5°− 11.2°+ 9.0°, which is further south than palaeolatitudes (around 30° S) derived from existing plate reconstruction models for the Australian plate. However, the Late Eocene palaeomagnetic data are limited (only two sites) and secular variation may not have been averaged out. During the Pliocene (ca. 4 Ma) we estimate a palaeolatitude of approximately 13° S. The presence of the Late Eocene ages at Christmas Island correlates well with the cessation of spreading of the Wharton Ridge (~ 43 Ma), the initiation of spreading along the South East Indian Ridge, and the transit of Christmas Island over a broad low velocity zone in the upper mantle. This suggests that changes in stress regimes following the tectonic reorganisation of the region (prior to~ 43 Ma) may have allowed deeper-origin mantle melts to rise. Similarly, changes in the plate's stress regime at the flexural bulge of the Sunda-Java subduction zone may be implicated in renewed melting at ~ 4 Ma, suggesting that tectonic stresses have exerted a first-order effect on the timing and emplacement of volcanism at Christmas Island.

Full text

40
Ar/
39
Ar geochronology and the paleoposition of Christmas Island
(Australia), Northeast Indian Ocean
Rajat Taneja
a,
⁎, Craig O'Neill
a
,MarkLackie
a
,TracyRushmer
a
, Phil Schmidt
b
, Fred Jourdan
c
a
Department of Earth and Planetary Sciences, Australian Research Council Centre of Excellence for Core to Crust Fluid Systems/GEMOC, Macquarie University, Sydney, Australia
b
CSIRO Earth Science and Resource Engineering, North Ryde, NSW, Australia
c
Western Australian Argon Isotope Facility, JdL Centre & Dept. of Applied Geology, Curtin University, Perth, Australia
abstractarticle info
Article history:
Received 14 October 2013
Received in revised form 3 April 2014
Accepted 3 April 2014
Available online 27 April 2014
Handling Editor: J.G. Meert
Keywords:
Paleomagnetism
Seamounts
Seismic tomography
Plate reconstruction
Indian Ocean
Capricorn plate
The Christmas Island Seamount Province is an extensive zone of volcanism in the Northeast Indian Ocean,
consisting of numerous submerged seamounts and flat-topped guyots. Within this region lies two subaerial
island groups, Christmas Island, and the Cocos Keeling archipelago. Christmas Island has experienced multiple
episodes of volcanism that are exposed sporadically along its coastline. Here, we dated these volcanics using
40
Ar/
39
Ar geochronology and analysed them for paleomagnetism. The oldest exposed volcanism occurred in
the Eocene between 43 and 37 Ma. This is followed by a time gap of ~33 million years, before the eruption of a
younger episode of Pliocene age (4.32 ± 0.17 Ma). It has, however, been suggested by previous workers that
there is a much older Late Cretaceous event beneath the limestone which is unexposed. In addition, this study
conducted the first paleomagnetic analysis of samples from Christmas Island to determine its paleoposition
and the paleomagnetic polarity of the sampled sites. Two normal and two reversal magnetic events have been
recorded, that agree with the geomagnetic reversal timescales. Late Eocene (38–39 Ma) palaeomagnetic data
suggest a palaeolatitude of −43.5°
−11.2°
+9.0°
, which is further south than palaeolatitudes (around 30° S) derived
from existing plate reconstruction models for the Australian plate. However, the Late Eocene palaeomagnetic
data are limited (only two sites) and secular variation may not have been averaged out. During the Pliocene
(ca. 4 Ma) weestimate a palaeolatitude ofapproximately 13°S. The presence of theLate Eocene agesat Christmas
Island correlates wellwith the cessation of spreading of the Wharton Ridge (~43 Ma), the initiation of spreading
along the South East Indian Ridge, and the transitof Christmas Island overa broad low velocity zone in the upper
mantle. This suggests that changes in stress regimes following the tectonic reorganisation of the region
(prior to~43 Ma) may have allowed deeper-origin mantlemelts to rise. Similarly, changes in the plate's stress re-
gime at the flexural bulge of the Sunda–Java subduction zone may be implicated in renewed melting at ~4 Ma,
suggesting that tectonic stresses have exerted a first-order effect on the timing and emplacement of volcanism
at Christmas Island.
© 2014 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
1. Introduction
1.1. Christmas Island Seamount Province (CHRISP)
The Northeast Indian Ocean is an understudied area, but the
occurrence of large intraplate and subduction-zone earthquakes in the
last ten years within the intraplate and subduction region of the Indo-
Australian plate and a recent cruise has led to a re-emergence of geolog-
ical interest (Abercrombie et al., 2003; DeMets and Royer, 2003;
Hoernle et al., 2011; Hall, 2012; Yue et al., 2012). The Northeast Indian
Ocean region is dominated by numerous submerged seamounts and
two ocean islands whose origin has not received much attention until
recently (Hoernle et al., 2011). The distribution of volcanism does not
follow a linear hotspot model (Morgan, 1971). The region is dominated
by numerous north–south trending fracture zones, but again, these do
not seem to be exerting a fundamental control on the distribution of
seamount volcanism, which predominantly trends east–west.
The Northeast Indian Ocean is divided into various physiographic di-
visions as shown in Fig. 1 comprising; The North West Shelf, Exmouth
Plateau, Wombat Plateau, Argo Abyssal Plain, Cuvier Abyssal Plain,
Gascoyne Abyssal Plain, and Platypus Spur. The North West Shelf is a
continental margin extending from northern Australia to the Exmouth
Plateau off the west coast that was formed due to multiple episodes of
rifting and removal of continental fragments following the disintegra-
tion of Gondwana in the Late Cretaceous (Stagg et al., 1999; Heine and
Müller, 2005). The Argo Abyssal Plain is located west of the North
West Shelf in the easternmost corner of the Indian Ocean. It is
Gondwana Research 28 (2015) 391–406
⁎Corresponding author.
E-mail addresses: taneja.rajat3@gmail.com,rajat.taneja@mq.edu.au (R. Taneja).
http://dx.doi.org/10.1016/j.gr.2014.04.004
1342-937X/© 2014 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Gondwana Research
journal homepage: www.elsevier.com/locate/gr

surrounded by the Java Trench in the north, and the submerged conti-
nental crust of the Scott Plateau in the northeast, the Rowley Terrace
in the east and the Wombat Plateausin the south. The Exmouth Plateau
lies further south of Wombat Plateau. The western edges of the Argo
Abyssal Plain are separated by the Joey and Roo Rises north of Platypus
Spur, and by the Gascoyne Abyssal Plain in the west. The Gascoyne
Abyssal Plain is bounded by the Exmouth Plateau in the west, Cuvier
Abyssal Plain in the south and the Java Trench in the north, with Argo
Abyssal Plain in the east and Wharton Basin and Christmas Islands in
the west. Investigator Ridge, a roughly 1800 km long ridge trending
north south along 98° E longitude is a distinctive feature of the Indo-
Australian plate. It can be seen prominently in bathymetric maps as a
continuous linear ridge that extends from 7°–18° S, beyond which it
becomes patchy and broken before subducting underneath Sumatra as
evident from Fig. 1.
The most important characteristic features of the Northeast Indian
Ocean are the numerous submerged seamounts, volcanic guyots
and two sub-aerial islands. The Christmas Island Seamount Province
(CHRISP) lies within the Northeast Indian Ocean, south of the Java
Trench and between the Ninetyeast Ridge and the North West Shelf
off the west coast of Australia (Fig. 1). These volcanic edifices rise up
to 3000 m below sea-level in a region where the abyssal plain depth
ranges between 5500 and 6000 m. Most of these volcanic features
occur within the latitudes 10° S and 15° S as a wide zone of seamounts
rather than a linear volcanic chain. The two sub aerially exposed island
groups are the Cocos (Keeling) Archipelago, and Christmas Island. Cocos
Fig. 1. Regional bathymetry map of Northeast Indian Ocean showing various tectonic features and the two islands (red circle) within the region. AAP, Argo Abyssal Plain; GAP, Gascoyne
Abyssal Plain; CAP, CuvierAbyssal Plain; Perth AP, Perth Abyssal Plain; RR, Roo Rise; JR, Joey Rise; ExP,Exmouth Plateau ScP, ScottPlateau; WoP, Wombat Plateau; PS, Platypus Spur; RT,
Rowley Terrace; WP, Wallaby Plateau;ZP, Zenith Plateau; CRFZ, Cape Range Fracture Zone,WZFZ, Wallaby–Zenith Fracture Zone; WB, Wharton Basin; BR, Broken Ridge; Ch Is, Christmas
Islandand Go, GoldenBo'Sunbird(85 Ma) seamount arerepresentedin red and blackrectangle. Otherseamountsare FF, FlyingFish (70 Ma);VM, VeningMeinesz (72 Ma); Sch, Shcherbakov
(~81 Ma); and Umb, Umbgrove (no age available).
392 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

(Keeling) Islands are a coral atoll archipelago consisting of 27 islands.
There are no volcanic exposures on the islands (Woodroffe et al.,
1991; Woodroffe and Falkland, 1997). Hoernle et al. (2011) dredged
volcanic rocks at depths of 2400–3100 m that yielded Late Palaeocene
ages of 55.6 ± 0.2 Ma and 56.0 ± 0.2 Ma.
Christmas Island, on the other hand, is the sole island to have record-
ed sub-aerial intraplate volcanism, in the form of basaltic rocks that are
exposed in some places on the island. Little published literature exists
concerning Christmas Island's geology. Prominent amongst these are
the works doneby An drews (1900) and Trueman (1965) on the geology
of the island. Barrie (1967) discussed the phosphates of the island with
minor volcanic examination based essentially on visual and field obser-
vations. Earlier works on Christmas Island by Rivereau (1965),Polak
(1976),Pettifer and Polak (1979),andBarrie (1967) were primarily
geomorphological and geophysical surveys. Barrie (1967),byvisual
examination of volcanic rock and its breakdown into soil, and faunal
assemblages within interbedded limestones suggested, that the main
phase of volcanism ceased in the Early Tertiary and the island became
volcanically extinct in the Pliocene. The presence of these volcanic
outcrops, therefore, makes this island significant in constraining the
tectonic and volcanic evolution of the region.
There is a report of paleomagnetic analysis carried out at Christmas
Island by the Bureau of Mineral Resources. This preliminary study
was based on one sample per site to understand rock properties and
mentioned some locations for detailed paleomagnetic investigations
(Polak, 1976). We conduct a detailed high resolution paleomagnetic
analyses using larger sample size to determine the paleomagnetic
record of Christmas Island, the only volcanically exposed representative
of submarine volcanism in the CHRISP region.
Hoernle et al. (2011) has put forward a mechanism describing the
origin of the CHRIS Province. They proposed shallow recycling of
continental lithosphere as a mechanism for volcanic generation of the
CHRISP. They argue that decompressional melting of volatile rich Arche-
an SCLM and MORB sourcemantle leads to the formation of CHRISP near
the MOR during the dispersal of Gondwana. In addition, Hoernle et al.
(2011) suggested that the volcanic province was formed near the
mid-oceanic ridge. They carried out
40
Ar/
39
Ar age dating on dredged
samples within the intraplate region, and found that they get younger
towards the west. Ages between 136.2 ± 0.17 Ma in the Argo Abyssal
Plain, and 47.0 ± 0.2 Ma in the east, south of Cocos (Keeling) Island,
have been described by Hoernle et al. (2011).
Although there are numerous tectonic models for the evolution
of the Northeast Indian Ocean and the western coast of Australia
(Veevers, 1971; Heirtzler et al., 1978; Müller et al., 1998; Heine and
Müller, 2005; Müller et al., 2008; Hall, 2012), none have integrated
age constraints from volcanism in the area, with paleomagnetic con-
straints on the volcanic suite, to examine the causative mechanism of
volcanism. In this paper, we present high resolution
40
Ar/
39
Ar radioiso-
topic dating results to determine the ages and duration of volcanism on
the island. In addition, this study presents a paleomagnetic analysis on
the Christmas Island volcanics to determine the magnetic polarity of
the lava flows, and compare these magnetic polarities with the geomag-
netic timescale to constrain the reconstructed position of Christmas
Island.
1.2. Geology of Christmas Island
Christmas Island is a roughly “T”shaped (Fig. 2) limestone capped is-
land that lies 400 km south of the Java Trench and roughly 3000 km
northwest of Perth. Its present position is 105° E, 10.5° S and is 360 m
above sea-level, rising 5700–5900 m from the bottom of sea floor.
Christmas Island has been previously surveyed essentially for exploita-
tion of its groundwater aquifers (Polak, 1976; Pettifer and Polak, 1979)
and phosphate mining operations (Doak et al., 1965; Trueman, 1965).
Large parts of the island are now a world heritage site due to its unique
fauna including the annual red crab migration (O'Dowd and Lake, 1989;
O'Dowd and Lake, 1991).
The island is believed to have been formed initially in the Late Creta-
ceous as an undersea volcano (Grimes, 2001). It now sits on the bulge of
the Australian plate subducting beneath the Sunda and Java Trenches
which has led to its re-emergence above the sea-level (Grimes, 2001).
Field studies of volcanic rocks at Sydney's Dale and Waterfall Spring
suggest that the island was volcanically active in the Eocene but this
had ceased by the Miocene (Barrie, 1967). Pettifer and Polak (1979)
by visual examination of volcanic rocks and soils suggested, in their un-
published report, that the main phase of volcanism ceased in the Early
Tertiary and the island became volcanically extinct in the Pliocene.
There is no evidence of any volcanic activity in the Oligocene on the is-
land and it has been suggested as a period of erosion (Barrie, 1967). The
rise of the island due to the island riding the flexural bulge on the
subducting plate is estimated to be around 140 mm per thousand
years (Woodroffe, 1988) and has been subjected to sea-level fluctuation
in the recent past (Grimes, 2001).
Barrie (1967) has reported the occurrence of a phosphatised volca-
nic material near the highlands on the western part of the island,
more specifically, close to the Murray Hill region. In addition, it has
been suggested that the volcanic outcrops at Flying Fish Cove and
Dolly Beach are a result of faultingor wave cutting and are pre Miocene
to early Miocene in age (Barrie, 1967). Baxter and Weeks (1984) have
reported the volcanism in Pliocene was accompanied by faulting near
the Murray Hill region. Some of the sub-aerial deposits fill limestone de-
pressions, which suggests that the island was partly emergent in Early
Pliocene (Grimes, 2001). Using geophysical investigations, Polak
(1976) suggests the presence of rift zones to be responsible for a three
pronged shape of the island. These rift zones partly cut deep erosion
channels before the island emerged (Polak, 1976).
Works by Andrews (1900) and Trueman (1965) have documented
three stages of volcanic eruptions withinterbedded sequences of Tertia-
ry limestone and phosphate rich soils. These interbeds have more spe-
cifically been exposed in the north near Flying Fish Cove and in the
east close to Dolly Beach. The Late Cretaceous volcanic core of the island
is buried, though earlier work has cited some exposures in the bottom
of some limestone caves that are now being used for groundwater ex-
ploitation purposes, and are now inaccessible (Grimes, 2001). A Late
Cretaceous phase of volcanism has been inferred from stratigraphic
and faunal correlations (Grimes, 2001), and from flexural plate model-
ling (Taneja and O'Neill, in-review). It is overlain by basalts, basanites
and trachytes of Middle to Late Eocene with limestones and dolomites
(Table 1). The Pliocene phase of volcanism at the island occurs as
minor volcanic dykes and tuffs that are overlain by phosphate rich soils.
2. Methodology and analytical techniques
2.1. Sample collection
Samples were collected during the field trip to the island in Novem-
ber, 2010. The rock cores were obtained using a water-cooled two-
stroke powered motor drill. Samples were drilled in the field and orient-
ed using a magnetic and a sun compass as described by Butler (1992)
and McElhinny and McFadden (2000). Each individual core measured
3 to 5 cm in length with a diameter of 2.5 cm. Of the total sites sampled,
three were on the east coast (Waterfall Spring, Ethel Beach and the
Dolly Beach), two on the west coast (Winifred Beach and Sydney's
Dale) and one site in north (Flying Fish Cove) as shown in Figs. 2 and
3. The three east coast sites and one western site (Winifred Beach)
were used for paleomagnetism. All of the above mentioned sites were
drilled except the Sydney's Dale due to its protected status under the
Ramsay Convention on Wetlands. These four sites and additional hand
samples from Sydney's Dale and Flying Fish Cover have been dated
using
40
Ar/
39
Ar geochronology. Dolly Beach on the east coast contains
a vertically dipping basaltic dyke protruding through the shore, with a
393R. Taneja et al. / Gondwana Research 28 (2015) 391–406

strike of ~45°, containing phenocrysts of olivine. The sub-horizontal sills
at the base of the cliffs at Ethel Beach and Waterfall Spring are vesicular
basalts. Winifred Beach and Sydney's Dale, the sampling sites on the
west coast of the Island, outcrop inland and occur under a thick canopy
of forest. Waterfall Spring and Ethel Beach are small outcrops spread
across 3–5m
2
. Furthermore, the sampling sites on the east coast are lo-
cated at the shore as comparedto those on the west coast, which are at a
higher elevation.Since most of the island is a dense tropical forest under
the management of Parks Australia and a phosphate mining hub, access
to certain parts of the island was limited or not possible, affecting sam-
pling in some areas.
2.2. Geochronological techniques
Six samples were selected from Christmas Island for
40
Ar/
39
Ar dating
and were separated into unaltered, 600–800 μm-size groundmass.
Groundmass particles were carefullyhand-picked under a binocular mi-
croscope. Selected groundmass material (consisting of homogenous
unoxidised pyroxene and plagioclase) was further leached in diluted
HF for 1 min and then thoroughly rinsed with distilled water in an
ultrasonic cleaner. Separates of plagioclase crystals have been shown
to provide more accurate results than groundmass, as it facilitates the
optical selection of the freshest grains (e.g., Hofmann et al., 2000;
Jourdan et al., 2007; Verati and Jourdan, 2013), but their small size in
our samples prevented us from using this approach.
Samples were loaded into six large wells of 1.9 cm diameter and
0.3 cm depth aluminium disc. These wells were bracketed by small
wells that included Fish Canyon sanidine (FCs) used as a neutron
fluence monitor for which an age of 28.305 ± 0.036 Ma (1σ)was
adopted (Renne et al., 2010). The discs were Cd-shielded (to mini-
mize undesirable nuclear interference reactions) and irradiated for
2 h in the Hamilton McMaster University nuclear reactor (Canada)
Fig. 2. A map of Christmas Islandshowing the location of volcanic sites sampled for geochronologyand paleomagnetism. Volcanic exposures androad network re-drawn from maps pro-
vided by Christmas Island National Park. Settlement area modified from Christmas Island tourism maps.
Structural details are redrawn from Barrie (1967)
Table 1
Main geological units at the Christmas Islands, Northeast Indian Ocean, modified from
Grimes, 2001.
Time period Geological description
Quaternary Phosphate deposits.
Pliocene (3–5 Ma) Upper volcanic series, the youngest volcanism.
Late Oligocene–Mid Miocene Upper carbonate series.
Eocene (35–40 Ma) The lower volcanic series, second episode of
volcanism.
Eocene Lower carbonate series.
Late Cretaceous–Early Tertiary
volcanic
The core v olcanic p art, mostly submerge d and
the oldest.
394 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

in position 5C. The mean J-values computed from standard grains
within the small pits range from 0.00068800 ± 0.00000220 to
0.00071000 ± 0.00000099 determined as the average and standard
deviation of J-values of the small wells for each irradiation disc. Mass
discrimination was monitored using an automated air pipette and
provided a range from 1.004492 ± 0.29 to 1.006127 ± 0.38 (%1σ)per
dalton (atomic mass unit) relative to an air ratio of 298.56 ± 0.31
(Lee et al., 2006). The correction factors for interfering isotopes were
(
39
Ar/
37
Ar)
Ca
=7.30×10
−4
(±11%), (
36
Ar/
37
Ar)
Ca
= 2.82 × 10
−4
(±1%) and (
40
Ar/
39
Ar)
K
=6.76×10
−4
(±32%) (e.g., Jourdan and
Renne, 2007).
The
40
Ar/
39
Ar analyses were performed at the Western Australian
Argon Isotope Facility at Curtin University. The samples were wrapped
in degassed niobium foil packages and step-heated using a 110 W
Spectron Laser Systems, with a continuous Nd-YAG (IR; 1064 nm)
laser rastered over the sample for 1 min to ensure a homogenously dis-
tributed temperature. The gas was purified in a stainless steel extraction
line using two SAES AP10 and one GP50 getter and a liquid nitrogen
condensation trap. Argon isotopes were measured in static mode
using a MAP 215-50 mass spectrometer (resolution of ~450; sensitivity
of 4 × 10
−14
mol/V) with a Balzers SEV 217 electron multiplier mostly
using 9 to 10 cycles of peak-hopping. The data acquisition was per-
formed with the Argus programme written by M.O. McWilliams and
ran under a LabView environment. The raw data was processed using
the ArArCALC software (Koppers, 2002) and the ages have been calcu-
lated using the decay constants recommended by Renne et al. (2010).
Blanks were monitored every 3 to 4 steps and typical
40
Ar blanks
range from 1 × 10
−16
to 2 × 10
−16
mol. Ar isotopic data corrected for
blank, mass discrimination and radioactive decay are given as Supple-
mentary material. Individual errors (given in the Supplementary
material) are given at the 1σand 2σlevels, while Age and K/Ca
(Fig. 5) are given at 2σlevel. Our criteria for the determination of pla-
teau are as follows: plateaus must include at least 70% of
39
Ar. The pla-
teau should be distributed over a minimum of 3 consecutive steps
agreeing at 95% confidence level and satisfying a probability of fit (P)
of at least 0.05. Plateau ages (Fig. 5) are given at the 2σlevel and are cal-
culated using the mean of all the plateau steps, each weighted by the in-
verse variance of their individual analytical error. Mini-plateaus are
defined similarly except that they include between 50% and 70% of
39
Ar. Inverse isochrons include the maximum number of steps with a
probability of fit≥0.05. All sources of uncertainties are included in the
calculation.
2.3. Paleomagnetic laboratory techniques
The drilled cores were cut into 2.2 cm individual core samples
and used for paleomagnetic analysis. A select set of samples were
demagnetised using an alternating-field progressive demagnetiser
(Model 2G600) and another separate set of samples were thermally
demagnetised using MMTD80 programmable thermal demagnetiser at
CSIRO, North Ryde. The oven within the demagnetiser is surrounded
by 4 layers of mu-metal shielding with a residual field of less than
10 nT. It is capable of heating up to 80 one inch samples automatically.
Remanent intensities were measured using 2G Enterprise 3-axis cryo-
genic magnetometer (Model 755). Alternating field demagnetisation
was carried out in fourteen demagnetisation steps mostly with
increments of 5–10 mT up to maximum of 70 mT. Selected samples
underwent thermal demagnetisation from 200 °C in increments of
50 °C up to 500 °C, thereafter in increments of 5–20 °C up to 570 °C.
Fig. 3. Outcrop photographs during samplecollection at Christmas Island, A, Waterfall Spring (XM-19C and XM-20); B, Dolly Beach ( XM-29 and XM-30B); C, EthelBeach(XM-34
and XM-35); D, Sydney's Dale (XM-40).
395R. Taneja et al. / Gondwana Research 28 (2015) 391–406

Magnetic susceptibilities were measured using a Sapphire Instruments
SI2B susceptibility instrument.
The demagnetisation steps with their inclinations and declinations
were plotted on orthogonal vector diagrams. Magnetic remanence
locked within the sample consists of a primary magnetic component,
and often associated with it is later secondary magnetisation. This aux-
iliary magnetic character is acquired, in geological time, due to second-
ary processes like viscous acquisition or effects of stray fields from
lightning. Metamorphism of the rock body, or alteration due to chemical
reactions, can both leave secondary or later stage magnetic imprints. In-
terpretation of demagnetisation was done by preparing visually identi-
fiable univectorial decay vectors on orthogonal vector plots, called
orthogonal diagrams (Zijderveld, 1967), to determine the polarity and
progressive demagnetisation. Thishelps in identifying the primary mag-
netic component, also called Characteristic Remanent Magnetisation
(ChRM), and differentiating it from secondary components. Principal
component analysis (PCA) was used in calculating mean directions and
mean intensity of ChRM (Kirschvink, 1980). Fisherian analysis is used
for calculation of mean inclination (I
m
) and declination (D
m
) for each
site as shown in Tables 4 and 5, using the software Paleomagnetic
Tools (PMAGTOOL) V 4.2a, and an in-house software developed by Phil
Schmidt, CSIRO. Mean inclination, of a sample can be used to calculate
the paleolatitude by a simple relation,
tanIm¼2tanλð1Þ
where, λ, is the latitude of the paleosite.
3. Results
3.1. Petrology
Only fresh samples from Christmas Island were selected forgeochro-
nology, and in-situ sites were drilled for paleomagnetic analysis. The
Dolly Beach site on the east cost of the island consists of a vertical basal-
tic dyke emerging from the coast. Petrographically (Fig. 4), it contains a
fine-grained groundmass (~70 vol.%) consisting ofmicroliths of plagio-
clase (50 μm to 200 μm), glass, and opaque minerals. Phenocrysts of
olivine (25 vol.%) and clinopyroxene (5 vol.%) form the remaining 30%
of the sample. Phenocrysts of olivine vary from b50 μm to several cm
and are euhedral to subhedral. Magnetite inclusions (~10 μm) occur in
pyroxenes within the groundmass, and occasionally surrounding grain
boundaries of olivine. Waterfall Spring and Ethel Beach are very similar
petrographically, with a fine-grained groundmass (plagioclase, ground-
mass and opaque oxides) that comprises ~ 80 vol.% of the sample.
Plagioclase occurs as elongate euhedral to subhedral needles that are
randomly oriented and range from 50 μmtoover100μm. Phenocrysts
of olivine (15 vol.%) are fractured (affected by serpentinisation),
and are euhedral to subhedral (100 μmto400μm), while the remaining
5 vol.% comprises clinopyroxenes (50 μm–200 μm). Needle shaped
magnetite (50 μm) are present within chromite. Flying Fish Cove, on
the north side of the island, consists of phenocrysts of olivine
(150 μm to several cm), and minor clinopyroxene comprising total
30 vol.% of the sample, while the remaining part is fine-grained
groundmass. Plagioclase averages 100 μminsize,andforms
Fig. 4. Petrological examination ofvolcanic rocks from ChristmasIsland. Primary mineralogy consists of phenocrysts of olivine and clinopyroxene,in a fine grained matrix comprising of
plagioclase, glass,and opaque oxides.Modal abundanceof groundmass involcanic rocks fromChristmas Islandvaries between 70 vol.% and 80 vol.%. Phenocrystsof olivine are euhedral to
subhedral (~ 50 μm to several mm), while plagioclase occurs as needle like ranging from 50 μmto300μm. All photos taken in crossed polars.
396 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

80 vol.% of the groundmass volume, while 20 vol.% of the groundmass
is glass andopaques, both occurring in equal proportions. Somesections
of the sample have evidence of alteration that appears to have affected
grain boundaries of olivine. Winifred Beach samples exhibit a fine-
grained groundmass of clinopyroxene (25 vol.%, 25 μmto200μm)
that are euhedral to subhedral in shape. Plagioclase laths (60 vol.%,
25 μmto150μm) form the majority of the groundmass, together with
minor opaques and glass (15 vol.%). Olivine is the primary phenocryst
observed (~70 vol.% of the total phenocrysts), with minor
clinopyroxene (~30 vol.%). Randomly oriented needle-like magnetites
appear within the chromite. Sydney's Dale exhibits a characteristic tra-
chyte texture of flow-oriented mineral grains. The samples are veryfine
grained, making it difficultto identify minerals, but thesample predom-
inantly consists of oriented plagioclase (50 μmto200μm), with opaque
and glass and rare clinopyroxene. Plagioclase appears to have under-
gone sericitic alteration, and glass and earlier mafic phases to chlorite.
3.2.
40
Ar/
39
Ar dating
The step wise laser heating experimentfor sample from Dolly Beach
(XM-30B) gave a plateau age of 43.55 ± 0.44 Ma (MSWD = 1.4; P =
0.14; Fig. 5) including 99% of
39
Ar released. The K/Ca (Fig. 5) (derived
from
39
Ar
K
and
37
Ar
Ca
) spectra show a slightvariation typically observed
for basalt samples (McDougall and Harrison, 1999). The inverse
isochron (Fig. 5) gives an age of 43.46 ± 0.66 Ma (MSWD = 1.4; P =
0.14) and with a
40
Ar/
36
Ar intercept of 300.8 ± 2.5, indistinguishable
from the atmospheric ratio of 298.6 ± 0.3 (Lee et al., 2006)adoptedin
this study. This indicates that no excess argon is present in this sample,
and hence we take the plateau age as indicating the time since the
eruption of the magma.
Ethel Beach (XM-34) shows a two-level age spectrum suggesting
that alteration perturbed this sample to some extent. A mini-plateau
(minimum) age of ≥37.10 ± 0.66 Ma (MSWD = 0.86; P = 0.56) in-
cluding 63% of the
39
Ar released, suggests that the effects of alteration
were not extensive (Fig. 5). The shape of the K/Ca spectrum matches
the shape of the age spectrum and shows a decrease of the K/Ca ratio
towards high temperature steps (Fig. 5). The inverse isochron yielded
an age of 36.3 ± 2.1 Ma (MSWD = 0.9; P = 0.53) and a
40
Ar/
36
Ar
trapped ratio of 316 ± 20, indistinguishable from the atmospheric
ratio. We conservatively interpret the plateau age as being a minimum
age, albeit probably close to the eruption age of this sample.
Sample XM-41 at Sydney's Dale on the west coast did notyield a pla-
teau age as the age spectrum shows a barely visible, but nevertheless in-
dividual, step age variation. However, it did yield a well-behaved
inverse isochron age of 41.75 ± 0.30 Ma (MSWD = 0.85; P = 0.63)
including 92% of cumulative
39
Ar and an intercept
40
Ar/
36
Ar trapped
value of 306.1 ± 3.4 (Fig. 5). The
40
Ar/
36
Ar value suggests that some
excess
40
Ar* is present in the sample, but is fully accounted for in the
inverse isochron age calculation (e.g. Sharp and Renne, 2005; Jourdan
et al., 2012). As an exercise, we calculated a plateau age of 41.75 ±
0.18 Ma (MSWD = 0.75; P = 0.75) using a
40
Ar/
36
Ar value of 306.1
instead of 298.6 (Fig. 5) but we retain the inverse isochron age for the
discussion.
The results for the sample from Flying Fish Cove (XM-43) near the
north shore of the island, however, shows a more disturbed age spec-
trum (Fig. 5). Such a spectrum with decreasing apparent ages at high
temperature is characteristic of an alteration overprint, where the ef-
fects of alteration (usually neo-crystalized K-rich phases) on apparent
step ages are more apparent on K-poor domains compare to K-rich do-
mains. A contribution from alteration cannot be excluded from the low
temperature steps and probably lowers the age of the mini-plateau. It
only provides qualitative constraints for the age of this sample, with a
minimum age of ≥38.7 ± 0.5 Ma (MSWD = 0.48; P = 0.87) as given
by a mini-plateau including 56% of the
39
Ar released. The K/Ca spectrum
show little compositional zoning in the sample (Fig. 5). The inverse
isochron shows only little spread, but the trapped ratio (319 ± 71),
although imprecise, seems to suggest that no excess
40
Ar* is present.
Waterfall Spring (XM-19) yielded a plateau over 100% of the
39
Ar re-
leased and gave an age of 37.75 ± 0.77 Ma (MSWD = 0.65; P = 0.75,
Table 2). The K/Ca spectrum suggests little zoning in this sample
(Fig. 5). The inverse isochron has a small spreading factor preventing
an estimate of the proper age and intercept value (Fig. 5).
Winifred Beach (XM-33) on the other hand is the youngest
sample, and gave a plateau age of 4.32 ± 0.17 Ma (MSWD = 1.56;
P = 0.15) including 79% of cumulative
39
Ar (Fig. 5). The age spec-
trum shows that the step ages get slightly younger over the last
21% of the age spectrum, possibly due to minor alteration. The
breadth of the plateau suggest that it is not affected by any secondary
process, and we interpret the
40
Ar/
39
Ar plateau age as indicating the
eruption age of this sample. The inverse isochron yielded an impre-
cise intercept
40
Ar/
36
Ar ratio of 309 ± 49, indistinguishable of atmo-
spheric composition. The K/Ca ratio suggests that very little Ca was
present in this sample and there is more glass and groundmass.
3.3. Paleomagnetic analysis
The Koenigsberger ratio (Q) is the ratio of the remanent magnetiza-
tion to the induced magnetization (product of susceptibility and the
Earth's magnetic field strength). It is an effective parameter to provide
an indication of the significance of remanence magnetism. A large Q im-
plies that the magnetic material will tend to retain significant remanent
magnetization and hence is a strong magnetic recorder. Nakanishi
and Gee (1995) calculated the Q values for north-western Pacific
guyots and concluded that these are reliable enough for paleolatitude
determination. The Koenigsberger ratio (Q ratio) mathematically, is
defined as:
Q¼NRM
kHð2Þ
where, NRM, is the remanent magnetism measured before AF
demagnetisation; k is the susceptibility; H is the total magnetic field
at Christmas Island. Both remanent and field intensities are measured
in units of A/m.
Relatively high Q values for most of the samples as shown in Table 3
make them suitable for paleomagnetic analyses. The orthogonal projec-
tions displaying magnetic vector components are the most common
tool to determine and analyse paleomagnetism properties by selecting
vector end points that decay univectorially towards the origin as
shown in Figs. 6 and 7. The Ethel Beach (XM-35) and Waterfall Spring
(XM-20) sampling sites are located on the east coast, and are ~300 m
apart. The assumed primary ChRM for both the sites is resolvable after
10–15 mT demagnetisation steps. Both sites have a mean declination
that is northwards with a steep inclination of −57.2° and −65.5°
(directed upwards) respectively. The two sites being in close proximity
of each other and their similar geochronology age imply that both of
the sites have recorded the same event of magnetism and magma
solidification.
Dolly Beach (XM-29) on the east coast has a steeper inclination of
77.6° that points down, opposite to Waterfall Spring and Ethel Beach.
Stereographic projection (Fig. 6) shows a declination centred near the
south. Winifred (XM-33), on the west coast is the youngest at ~4.3 Ma
and has a primary magnetism pointing east-southeast in the down
direction. The inclination points in the same direction as the oldest
site (XM-29); it is, however, opposite to Waterfall Spring and Ethel
Beach (Fig. 6). Samples from Winifred Beach were demagnetised
thermally in addition to AF-demagnetisation (Fig. 7). This proved to
be more effective in resolving the characteristic component of up to
500 °C. Beyond this temperature erratic behaviour was observed for
thermally demagnetised specimens indicating that the samples had
been demagnetised.
397R. Taneja et al. / Gondwana Research 28 (2015) 391–406

398 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

Furthermore, two sites (Dolly Beach and Winifred) have positive
inclinations. Assuming a southern hemisphere origin of these volcanics,
this denotes that the Earth's magnetic polarity reversed while magne-
tism was locked within these sites.
These are very large errors associated with some of the site
VGP's (Table 4) in the paleomagnetic analysis and hence appear
doubtful if these can be used effectively to find paleoposition.
Therefore, a combined position was obtained for the Eocene volca-
nism at the Christmas Island (Table 5) by combining all the Eocene
dates. In addition, the Fisher mean of the individual fisher means
(means of the three Eocene VGP sites) was calculated to constrain
the errors (Table 5).
4. Discussion
4.1. Duration of volcanism
40
Ar/
39
Ar geochronology shows multiple stages of volcanism at
Christmas Island. Two suites of volcanism at the island have been
dated and identified, an older Lower Volcanic Suite (LVS), and a younger
Upper Volcanic Suite (UVS) (Fig. 8). The oldest ages within the LVS are
found at Dolly Beach on the east coast and Sydney's Dale on the west
coast. We believe that this volcanism initiated in the Middle Eocenebe-
tween 42 and 43 Ma. Further on, Late Eocene (~ 38–39 Ma) volcanism is
evident at Waterfall Spring and Ethel Beach on the east coast, and Flying
Fish Cove in the north. Hoernle et al. (2011) conducted
40
Ar/
39
Ar geo-
chronology analyses that yielded a lowest age of 37.0 ± 0.6 Ma, and
an oldest age of 43.6 ± 0.4 Ma. Hoernle et al. (2011) also obtained
intermediate ages of 39.6 ± 0.4 Ma, 40.2 ± 0.2 Ma, 41.1 ± 0.1 Ma and
42.6 ± 0.4 Ma. Combining ages from this study and those obtained by
Hoernle et al. (2011), it appears that volcanism on the island was con-
tinuous from 37 Ma to 43 Ma.
The last and final episode of volcanism at the island started after a
time gap of ~33 Myr in the Early Pliocene at 4.32 ± 0.17 Ma. Hoernle
et al. (2011) yielded similar young ages in the range of 4.31 ± 0.14–
4.52 ± 0.24 Ma. New high resolution radioisotopic dating of the basaltic
rocks obtained from this study and those dated by Hoernle et al. (2011)
confirms two dated episodes of volcanism at the island, and it strongly
contradicts a middle phase in the Miocene as argued by Barrie (1967).
A Late Cretaceous volcanic core of the island is, however, not exposed
hence no dates are available for it. However, stratigraphic studies on in-
terbedded limestones on the Island (Grimes, 2001), and recent plate
flexural modelling (Taneja and O'Neill, in-review) have suggested a
Late Cretaceous volcanic episode, which led to elastic loading of the
young oceanic lithosphere around 75–88 Ma.
4.2. Petit spot low volume volcanism
Low volume “petit spot”volcanism formed away from the mid-
oceanic ridge and subduction zones has been recently described off
the north-western Pacificplate(Hirano et al., 2001, 2006). Hirano
et al. (2008) calculated the volume of these petit spot monogenetic
volcanoes and suggested that these are usually b1km
3
.Theseare
further differentiated on the basis of acoustic reflectivities with the
petit spot features having a higher reflectance than those compared
to the older and much larger submarine volcanoes (Fujiwara et al.,
2007).
Oceanic lithosphere behaves as an elastic sheet when loaded, for
stresses below its intrinsic yield stress (Bodine et al., 1981; Watts,
2001). Hirano et al. (2001, 2006) while discussing the petit spot
intraplate volcanism at Hokkaido Rise, in the northwest Pacific Ocean
suggested that these were formed due to fracturing within the flexed
outer rise of the lithospheric bulge. The flexure induced by bending of
the plate as it approaches a subduction zone causes redistribution of in-
ternal stresses leading to fracturing of the lithosphere that may cause
decompression melting in subjacent asthenospheric mantle, potentially
allowing the ascent of small volumes of melt in nascent cracks (Hirano
et al., 2001). The Indo-Australian plate in the region below the Java
and Sunda Trenches is undergoing subduction and demonstrates
flexure on the outer rise of the subducting plate (Levitt and Sandwell,
1995; Smith and Sandwell, 1997). A cross section of such a flexure
produced along a line perpendicular to the trench passing through the
island is shown in Fig. 9.Theflexure produced by the load of Christmas
Island is overprinted on the crest of the flexed lithosphere as it subducts
at the trench. A similar local flexure is produced by Golden Bo'Sunbird, a
seamount, south west of Christmas Island as shown in Fig. 9.
Magma flow rising through these cracks reaches the surface in the
form of petit spot volcanism. The younger episode of volcanism on
Christmas Island may share some characteristics with these low-
volume eruptions. Even though there are no constraints on the volume
of volcanism at Christmas Island, as it is covered by thick sequences of
limestone, the exposure of the Pliocene phase of volcanism at Winifred
Beach on the west coast of the island, emplaced through the older
volcanic core and limestone cliffs and terraces, bears similarity to this
petit spot model due to the following reasons. Baxter and Weeks
(1984) suggested that the volcanism in the region around the Murray
Hill was accompanied by faulting that led to volcanic vents, fractures
and dykes. Hoernle et al. (2011) dredged Pliocene volcanism around
the flanks of Christmas Island and this study has sampled one location
on the island. The limited occurrences of the younger phase of volca-
nism, far less than the older dated samples, and its association with
faulting and emplacement as dykes suggest that this is a small phase.
Table 2
40
Ar/
39
Ar analysisfor six samples fromChristmas Island;p, probability of fit(P) of at least 0.05; MSWD, meansquare weighteddeviation; Age in boldare the age adopted in thisstudy. Ages
underlined and accompanied with the sign ≥are minimum ages.
Sample no. Location Plateau age
39
Ar(k) released (%) Inverse isochron
40
Ar/
36
Ar
Age 2σMSWD p Age 2σMSWD Intercept 2σ
XM-34 Ethel Beach ≥37.1 0.66 0.86 0.56 63 36.29 2.08 0.88 316 20
XM-41 Sydney's Dale 41.75 0.18
a
0.75 0.75 92 41.75 0.30 0.85 306 3.0
XM-30B Dolly Beach 43.55 0.44 1.37 0.14 99 43.46 0.66 1.38 301 2.5
XM-19C Waterfall Spring 37.75 0.77 0.65 0.75 100 39.97 1.11 0.3 255 21
XM-43 Flying Fish Cove ≥38.7 0.5 0.48 0.87 56 38.20 1.83 0.54 319 71
XM-33D Winifred Beach 4.32 0.17 1.56 0.15 79 4.27 0.46 1.85 309 49
a
Calculated using the measured
40
Ar/
36
Ar intercept ratio.
Fig. 5. Age plateau (left), inverse isochron (centre), and K/Ca (right) plot for the six sampling sites at Christmas Island. Uncorrected sample from Sydney's Dale did not produce an age
plateau, but an age was obtained from Inverse Isochron (highlighted in bold), and a corrected sample yielded an agewhich is shown above. Squares in green are those used for age cal-
culations. Ages in bold are the ages adopted in this study. Ages in italics are minimum ages.
399R. Taneja et al. / Gondwana Research 28 (2015) 391–406

Additionally, if thePliocene phase of volcanism is indeed produced due
to flexure, then the onset of volcanism should correlate with the onset
of flexure, which can be determined from present observations of the
forebulge morphology, using plate reconstructions and past plateveloc-
ities. To determine the paleoposition of the erupted younger volcanics,
we backtracked the island from its present position at 10.5° S using a
plate motion rate of 7 cm/yr (Tregoning et al., 1994). Tracing the island
back to its position at 4.32 ± 0.17 Ma, we calculate its reconstructed po-
sition 302 ± 11.9 km (12.70°–12.85° S) southwestof its present day lo-
cation as shown in Fig. 9. Two additional small volcanic features are
present between the point of lithospheric cracking, and Golden
Bo'Sunbird seamount. These two features might represent petit spot
volcanic features documented by Hirano et al. (2001) for which addi-
tional dredging expeditions should be carried out. This, however, is be-
yond the scope of the current study.
Hirano et al. (2006) report petit spot volcanism could represent
a new intraplate type volcanic setting. They, along with Buchs et al.
(2013), are of the opinion that petit spot volcanism has been inadequate-
ly sampled, compared to other known volcanic islands in the Pacificand
elsewhere, and that this volcanic setting could be more common than
previously thought, and it also presents a tremendous scope for further
geophysical, geochemical and geodynamic investigations.
4.3. Magnetic polarity and paleoposition
The Eocene epoch has been marked by magnetic reversals and
paleomagnetism analysis can uncover changes in magnetic polarity
locked within a rock. This study matched magnetic polarity and geo-
chronological ages with Candeand Kent's (1995) and Wei's (1995) Geo-
magnetic Polarity Time Scale (GPTS) in a comparison with Earth's
paleomagnetic field record. Waterfall Spring and Ethel Beach, with
ages of 37.8 ± 0.8 Ma and 37.1 ± 0.7 Ma respectively, have normal po-
larity (present day). Dolly Beach, with an age of 43.4 ± 0.5 Ma, has a
positive inclination (Table 5), implying a reverse paleomagnetic field.
Dolly Beach falls within polarity chron C20n–C20r as summarised in
Table 6. Winifred Beach, with an age of 4.3 ± 0.2 Ma, has a reverse po-
larity as evident from positive inclination in stereographic projection in
Fig. 7 and Table 6.
Paleomagnetic investigations for Dolly Beach and Waterfall Spring
have very steep inclinations, and this indicates a high paleolatitude for
these sites. Polak (1976) found similar high inclinations for two of the
three sites where measurements were taken. An inclination of −65.5°
has been reported for Waterfall Beach and 67.9° for Dolly Beach —
similar and comparable to our analysis. The third site on the west
coast near Winifred Beach has a low inclination of 16.1°. The magnetic
polarity of both data is similar. Polak (1976) has, however, not given
details regarding the paleomagnetic analysis of these samples. The re-
sults are based on just one sample per site and it is not clear if drilled,
or in-situ samples were used for these calculations —lumps and chips
have been mentioned while describing some rock properties.
Paleomagnetic analyses of the Ethel Beach, Waterfall Spring and
Dolly Beach have given a VGP latitudes of −37.8°
−34.1°
+19.4°
;−47.6°
−8.3°
+7.1°
and −66.2°
−22.3°
+19.9°
respectively. There are large error associated with
individual site VGP's, hence the average of the Eocene volcanism was
calculated as shown in Table 5. By doing this, multiple flows of lava
are generally included within the combined data and are more effective
in averaging any secular variation experienced while the basaltic body
cooled. The pole calculated from combining the results can only be
considered to be a virtual geomagnetic pole (VGP). The paleolatitude
is therefore not tightly constrained and can only be taken as indicative.
While we tried to maximise the sampling of volcanic flows a lack of
suitable outcrops limits the statistical averaging of the VGPs. Same can
be said for the Winifred Beach where a single flow sampled may
record a transient field during a reversal. The combination of Fisher
means of the 3 VGP sites (Table 5) gives a combined paleolatitude
of −49.7°
−28.7°
+19.0°
(with a full range of paleolatitude between −30.47°
and −78.4°) This position is further south than the reconstructed posi-
tion of the Island as using a moving hotspot model (O'Neill et al., 2005),
which positions the island at 30° S at 45 Ma, and 28° S at37 Ma. The po-
sition of the Island during most recent phase of volcanism during the
Pliocene, at ~5 Ma, would be around 11.5 –12° S.
The drilling operations were carried out using a portable hand drill
and it is probable that an insufficient number of flows were sampled
within the sites to reduce the errors. Similar issues were encountered
by Nakanishi and Gee (1995) in their magnetic studies in the North-
western Pacific Ocean Guyots, where they suffered from secular varia-
tions and subsequent differences in paleolatitudes. Dupont-Nivet et al.
(2010), while calculating paleolatitude and the age of Indo-Asia colli-
sion suggested k N50 and n ≥5 as a sufficient dataset to represent
time averaged geomagnetic variations. Montes-Lauar et al. (1995)
adopted a methodology of using two samples per site to determine
the mean, they admit poor statistical quality in the data but consider
worth including these due to consistent directions.
Hence, instead of using the three Eocene sites, like above, the study
combines the two Late Eocene (Waterfall Beach and Ethel Beach)
sites, to get a slightly shallower paleolatitude of −43.5°
−11.2°
+9.0°
(with a
full range of paleolatitude between −34.4° and −54.7°). This is still
quite different from the reconstructed position, but the errors induced
are significantly smaller. It seems quite reasonable to combine the two
sites because, firstly, they have similar ages (Table 2) and secondly,
they are separated by 300 m and therefore, represent different lava
flows of the same sequence, whereas Dolly Beach is 7–9 km south of
the two sites. By combining these two events,we hope to have sampled
at least two flows of the same event. And with an increased number of
samples we hope to have constrained secular variations slightly better
as compared to individual sites. Given the relationship between the
two sites, it is possible that the recalculated paleolatitude might be a
spot reading of the geomagnetic field, and should be considered indica-
tive. Moreover, the reconstructed position is now closer to the position
ascribed by reconstruction model of O'Neill et al. (2005) of ~30 ° S, when
the error range is considered.
Seismic tomography models of Grand (2002) and Montelli et al.
(2004) document the presence of a low velocity zone beneath the
Eocene position of the Christmas Island as shown in Fig. 10.Grand's
(2002) seismic tomography model has very clearly documented a
seismic low velocity zone underneath the reconstructed position of
Table 3
Koenigsberger ratio (Q) for paleom agnetic samples f rom Christmas Isla nd. The total
magnetic field at Christmas Island is 46800 nT (H = 37.2 A/m). The high intensity for
Winifred Beach represents minimum value measured and could be stronger, thus their
Q values could be higher, while the Q values for other sites are well within the ranges
measured by 2G and are precise.
Location Sample Susceptibility, SI NRM (mA/m) Q
Winifred Beach XM-33A2 0.01145 17790 42.43
XM-33B2 0.01042 18810 49.27
XM-33B3 0.01058 17930 46.28
XM-33C2 0.01142 16880 40.35
XM-33E2 0.01107 17650 43.54
XM-33A1 0.01173 17790 41.40
XM-33B1 0.01050 16860 43.83
XM-33C1 0.01147 17080 40.64
XM-33D1 0.01058 18870 48.72
XM-33E1 0.01108 18220 44.89
Waterfall Spring XM-20C2 0.00130 1070 22.55
XM-20A1 0.00131 1340 27.95
XM-20C1 0.00119 1090 25.22
XM-20D1 0.00227 2930 35.22
Dolly Beach XM-29B2 0.00528 8280 42.85
XM-29C2 0.00529 4280 22.09
XM-29B1 0.00579 7050 33.28
XM-29C1 0.00569 8210 39.40
XM-29D1 0.01229 9690 21.54
Ethel Beach XM-35A1 0.00316 8210 71.04
XM-35B1 0.00256 4150 44.30
400 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

Christmas Island in the Eocene down to at least 900 km, and potentially
deeper down to 1600 km, beyond which the resolution is poor. Montelli
et al. (2004) documented a plume in the East Indian Ocean using both P
and S waves centred around 35° S and 100° E down to a depth of
1900 km below which they argue the resolution is poor. They also
discussed a deep plume underneath the Cocos (Keeling) Island and a
starting plume South of Java. Therefore the association of deep seated
low velocity seismic anomalies around the reconstructed position of
Christmas Island is striking (Montelli et al., 2004), and may be implicat-
ed in the re-emergence of volcanism at Christmas Island.
Oceanic islands are susceptible to tilting as they begin to ride the
bulge produced due to subduction at the trench. Tilt associated with
Christmas Island is between 0.23° and 1.27° and the Island presently
sitsrightatthecrestofthebulge(Fig. 9). The effect of tilt is more clearly
seen at the Golden Bo'Sunbird seamount, which is riding the bulge,
south-west of Christmas Island. After correcting the inclinations for
Fig. 6. Representativeorthogonal vectorplots (left)and site stereonetplots (right) illustratingdemagnetisation. Blue(solid) data pointsare projected ontothe horizontalplane, red (open)
data pointsare projected ontovertical plane. Firsttick on orthogonalvector plot represents 1000 mA/m. Black star (withinstereonet) is thepresentmagneticfield,D = 1.5°, I = −39° and
the Black square is the dipole field at Christmas Island.
401R. Taneja et al. / Gondwana Research 28 (2015) 391–406

Fig. 7. Representative orthogonal vector plot (top left) and site stereonet plot (top right) and a closeup of orthogonal vector diagram(middle) illustrating thermaldemagnetisationlevel
up to 570 °C for Winifred Beach.Representative orthogonal vector plot (bottom left) and site stereonetplot (bottom right) for Winifred Beach obtained using AF-demagnetisation. First
tick on orthogonal vectorplot represents 1000 mA/m. Blue (solid) data points are projected onto the horizontal plane, red (open)data points are projected onto ver tical plane. Black star
(within stereonet) is the present magnetic field, D = 1.5°, I = −39° and the Black square is the dipole field at Christmas Island.
Table 4
Summary of paleomagnetic results of foursites from Christmas Island. D
m
°andI
m
° represent mean declination and mean inclination respectively; α
95
°, semi-angle of cone of 95% confi-
dence about mean direction; k, Fishers precision parameter, n, number of specimen in a site; λ° represents the paleo-latitude; dp and dm, are the errorsassociated with the pole.
Site Demag Age (Ma) D
m
°I
m
°α
95
° k n Paleolatitude (λ° S) Paleolat. error VGP Pole Error
+−Lat° Long° dp dm
Waterfall AF 37.75 ± 0.77 8.8 −65.5 5.8 255.7 4 47.6 7.1 −8.3 52.1 275.8 7.6 9.4
Ethel AF 37.10 ± 0.66 354.7 −57.2 23.5 28.5 3 37.8 19.4 −34.1 62.2 294.5 25.0 34.3
Dolly AF 43.55 ± 0.45 162.2 77.6 13.1 35.0 5 66.2 22.3 −19.9 −32.9 113.9 23.0 24.5
Winifred TH 4.32 ± 0.17 137.3 24.7 19.2 23.6 4 12.9 12.8 −10.2 −48.1 187.7 11.0 20.6
402 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

the observed tilt, paleolatitude of the Pliocene phase of volcanism is be-
tween 12.03°and 12.88° S, and Waterfall and Ethel Beach combined is in
the range of 41°–43° S which is still significantly larger than existing re-
construction models.
It is possible that the volcanic sites sampled were influenced by
structural and tectonic forces that were not possible to determine in
the field and might be responsible for such high inclination. Barrie
(1967) and Polak (1976) have documented that the island has been af-
fected by a network of faults. Trueman (1965) mentions faulting has
played a major role in shaping the coastline of the island with normal
faults of high angle and tens of metres of displacement being observed
at many locations. Faults around the Murray Hill region have already
been mentioned earlier (Polak, 1976; Baxter and Weeks, 1984).
Rivereau (1965) mentions that the geomorphology of the island is
influenced by terraces, faults and scarps. It is further discussed that
the coastlines are affected by faulting, especially, the south-west and
east-south-east, whereas the north coast is little affected by it. Two
major networks of fault-lines have been reported east–west and
north-north-west (Rivereau, 1965). It is conceivable that these volcanic
sampling sites were influenced by faulting and associated block move-
ments (Fig. 2), but our field observations at our sample sites suggest
minimal interference.
The Indo-Australian plate has undergone major transformation in
terms of plate motion since its breakup from Gondwana and its subse-
quent northward journey. Following the cessation of spreading in the
Wharton Basin around ~43 Ma, and the initiation of spreading along
the South East Indian Ridge (Liu et al., 1983 and Müller et al., 2000),
the once separate Indian and Australian plates became one rigid plate.
The northern boundary of the plate changes from acontinental collision
in the west, to a subduction zone in the east. This change in tectonic set-
ting along the northern margin, and the relative motion of India and
Australia, is causing torsional stresses within the centre of the plate
(Cloetingh and Wortel, 1986). Deplus et al. (1998) discussed the distinct
deformation on either side of the Ninetyeast Ridge. Towards the west,
folds and east–west oriented reverse faults are common, whereas, on
the east of the ridge, left lateral strike slip faulting is prevalent along a
north–south trending fracture zone (Deplus et al., 1998). The deforma-
tion zone has been called as the Capricorn Plate (Royer and Gordon,
1997). Such a distinct deformation pattern within the diffuse plate
boundary could have altered the tectonic fabric of the seamount prov-
ince, which might influence the reconstructed paleoposition of Christ-
mas Island. The discrepancies in the paleolatitude of the Eocene phase
of Christmas Island with those from global reconstructions models
(O'Neill et al., 2005; Seton et al., 2012) could possibly be related to
Fig. 9. Backtracking the position of ChristmasIsland and associated Pliocenevolcanism at the seaward side of the outerrise of the subductingtrench. Black line represents theprofile of the
oceanic lithosphere in a NE-SW section perpendicular to the trench, red line showing the bending and flexing of the lithosphere. Golden Bo'Sunbird and two small seamounts (SW of
Christmas Island) represent petit spot features produceddue to flexure.
Fig. 8. A proposeddiagrammatic vertically exaggerated west to east cross section of Christmas Island showing the main submerged Late Cretaceous volcanic and LVS and UVS. There is a
lower volcanic sequence outcropping on the west coast at Sydney's Dale but has not been shown in this section.
Modified from Grimes (2001)
Table 5
Summary of paleomagnetic results by combining the Eocene sites from Christmas Island to determine a single pole position. Abbreviations are same as Table 4.
Site D
m
°I
m
°α
95
°k n Paleolatitude(λ°) Paleolat. error Pole Lat° Pole Long° dp dm
+−
Waterfall + Ethel Beach 1.7 −62.2 8.3 54.5 7 −43.5 9.0 −11.2 56.9 283.2 10.00 12.9
Eocene phase 11.4 −78.2 12.9 12.2 12 −67.3 20.0 −20.3 −32.6 280.3 22.9 24.4
Mean of 3 VGP's(W + E + D) 357.4 −67.0 17.1 52.9 3 −49.7 19.0 −28.7 −50.7 288.1 23.4 28.3
403R. Taneja et al. / Gondwana Research 28 (2015) 391–406

this deformation, and provide a further scope to constrain the paleo-
magnetic history of this intraplate region.
Additionally, dynamic tomography produced due to mantle flow
variations and density changes can induce elevation or subsidence
in continental and oceanic areas (Nyblade and Robinson, 1994;
Lithgow-Bertelloni and Silver, 1998; Moucha et al., 2008). Dynamic
subsidence of the order of 300 m has been reported in the Southeast
Asia by Wheeler and White (2002). Variations of these natures have
been observed in Southern and Eastern Africa, producing regions of
elevation of the order of 1 km, that are underlain by large scale low
velocity zones (Lithgow-Bertelloni and Silver, 1998). The East Indian
Ocean contains three significant low velocity zones, described by
Montelli et al. (2004), and suggested to be deep mantle plumes,
which can cause local uplift (Hager et al., 1985; Braun, 2010). Global dy-
namic tomography models have been presented by Heine et al. (2008)
and Steinberger et al. (2001).Heine et al. (2008) constrained present
day dynamic topographic in the CHRISP between −500 to 500 m,
and −100 to 500 m around the reconstructed position of Christmas
Island at 70 Ma. Moucha et al. (2008) prepared global ocean dynamic
topography, and constrained itwithin −500 to 0 m. Thus, the Northeast
Indian Ocean and CHRISP have been affected by the variation in sea-
floor depths in response to dynamic tomography, and this could disturb
the preserved tilts within the sampled units, during the progressive
cooling of magmatic material, though the magnitude of this effect is
likely to be similar, or less, than that ascribed to tilting due to subduc-
tion flexure.
The re-emergence of volcanism at Christmas Island in the Late
Eocene is synchronous with tectonic reorganisation in the East Indian
Ocean around 43 Ma (Liu et al., 1983). Plate reorganisation within the
Pacific Plate is exemplified by the change in the Hawaii Emperor Bend
(HEB). Dalrymple and Clague (1976) first suggested an age of 43 Ma,
but Sharp and Clague (2006), using new radioisotopic
40
Ar/
39
Ar data
from additional islands revised it to 50 Ma. Tarduno (2007), however,
is critical of such a value, and rather argues for an age of 47 Ma.
Whittaker et al. (2007) suggested major reorganisation events correlat-
ed with HEB, such as subduction of Pacific-Izanagi spreading ridge, and
initiation of Marianas and Tonga–Kermadec subduction. Such events
also caused plate motion changes in the Australian plate. Slab-pull
north of Australia, due to westerly subducting Wharton Ridge, led to a
change in Australian plate motion, from northwest, to north, around
50 Ma (Whittaker et al., 2007). Shortly after, spreading in the
Wharton Ridge ceased at 43 Ma (Liu et al., 1983, Müller et al.,
Fig. 10. Paleo-position of Christmas Island overlayed on Grand (2002) seismic tomography model at 350 and 900 km depth slices. Blue circle represents an approximate position of
Christmas Island as per reconstruction model, black represents reconstructed paleo-position according to this study, and purple circle represents present position of the island. The
paleo-position of the Eocene age is shown by vertical bar at 40 Ma where the circle represents the paleolatitude. This position is further south of that suggested by moving hotspot
model of O'Neill et al. (2005).
Table 6
Magnetic polarity of the four sampling sites from Christmas Islandand their comparison with the Geomagnetic Polarity Time Scale (Cande and Kent, 1995).
Location Sample no. Age (Ma) Polarity Magnetic chron (Cande and Kent, 1995) Magnetic chron (Wei, 1995)
Waterfall XM-20 37.8 ± 0.8 Normal C17n.1n–C17n.2n–C18n.1n C17n.1n–C17n.2n–C18n.1n
Ethel XM-35 37.1 ± 0.7 (min age) Normal C16n.2r–C17n.1n–C17n.2n C16n.2n–C17n.1n–C17n.2n
Dolly XM-29 43.4 ± 0.5 Reverse C20n–C20r C20n–C20r
Winifred XM-33 4.3 ± 0.2 Reverse C2An.3r–C3n.2n (Gilbert) C2An.3r–C3n.2n (Gilbert)
404 R. Taneja et al. / Gondwana Research 28 (2015) 391–406

2000; Singh et al., 2011). Spreading rates in the SEIR changed from
140 mm/yr to 60 mm/yr at 51.9 Ma to 45.3 Ma, respectively, when
plate reorganisation took place in the Australian plate (Cande et al.,
2010). Deplus et al. (1998) suggested that there are differential stress
patterns on either side of the Ninetyeast Ridge, with reverse faults in
the west, and left-lateral north–south strike slip faults in the east. Thus,
the combined change in plate reorganisation at 50 Ma, slowing down
of SEIR between 51 and 45 Ma, and closure of the Wharton Ridge, result-
ed in changed stress regime of now intact Indo-Australian Plate. This
could have led to re-emergence of volcanism in the seamount province
that has yet only been documented only at Christmas Island.
To summarise, our study shows that the position of Christmas Is-
land can be calculated using paleomagnetic data. Our study has
attempted to constrain such a position of the island with reasonable
errors and found it to be further south than that predicted by the
existing reconstructions model of O'Neill et al. (2005) and Seton
et al. (2012). The paleoposition for the Pliocene phase of the island
appears to constrain the island effectively, and is consistent with
global reconstructions (Seton et al., 2012). Paleomagnetic analysis
and plate reconstructions all position the island around 12–13° S at
4.3 Ma.
5. Conclusion
Volcanism in the Northeast Indian Ocean is represented in the form
of submerged flat topped and rounded peak seamounts and two sub-
aerially exposed islands, Christmas Island and Cocos (Keeling) Archipel-
ago. Of these, Christmas Island is the only island within the region to
have recorded this intraplate volcanism in the form of basaltic rocks
that are presently exposed above sea-level.
40
Ar/
39
Ar dating of the
island has revealed ongoing Cenozoic volcanism, from 43 Ma to
37 Ma. This was followed by a more recent Pliocene volcanic phase at
4.32 ± 0.17 Ma. The younger volcanic event is probably related to the
onset of lithospheric plate flexure, and associated asthenospheric de-
compression, on the seaward side of the outer rise (Fig. 9). An older
event, in the Late Cretaceous, probably formed the core of island and
is presently not exposed on the surface. We carried out a paleomagnetic
investigation by drilling basalts at Christmas Island and have attempted
to constrain island's position. The high paleolatitude of the island in the
Eocene is suggestive of a more southerly position than that predicted by
plate reconstruction models. This could either be due to inadequately
averaged secular variation, due limited outcrop, or because of intraplate
deformation within the Capricorn Plate, altering the tectonic fabric. The
youngest phase, at 4.3 Ma, resolved through thermal demagnetisation,
yielded a paleoposition of ~ 13° S which agrees well with the edge of
the lithospheric bulge at that time, and with existing plate reconstruc-
tion models. It was argued by Hirano et al. (2008) that low volume in-
traplate volcanism could be observed at similar tectonic settings to the
Japan and Tonga trench, and we put forward the Indo-Australian plate
subduction setting as a likely candidate for this phase of intraoceanic
volcanism. The Late Eocene event, however, is coincident with the ces-
sation of spreading at Wharton Ridge at 43 Ma and the subsequent
opening of the South East IndianRidge. Tectonic reorganisation changed
the stress regime of the once separate Indian and Australian plates,
allowing deeper fertile mantle, imaged in seismic tomography, to
melt, and ascend through the plate, which may have resulted in the
re-activation of volcanic activity at Christmas Island between 43 and
37 Ma.
Acknowledgement
We are grateful to Trond Torsvik and an anonymous reviewer for
their constructive comments that greatly improved this manuscript.
The authors are thankful to the staff at Western Australian Argon Iso-
tope Facility at Curtin University and at CSIRO, North Ryde, for helping
with the analytical techniques. We also thank Christmas Island National
Park staff for their time and discussions on sharing the locations of vol-
canic exposures at Christmas Island. This work was in part supported by
Australian Research Council (ARC) funding DP110104145, DP0880801,
and FT100100717. This is contribution 444 from the ARC Centre of Ex-
cellence for Core to Crust Fluid Systems (http://www.ccfs.mq.edu.au)
and 936 in the GEMOC Key Centre (http://www.gemoc.mq.edu.au).
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.gr.2014.04.004.
References
Abercrombie, R.E., Antolik, M., Ekström, G., 2003. The June 2000 Mw 7.9 earthquakes
south of Sumatra: deformation in the India–Australia Plate. Journal of Geophysical
Research 108, 2018.
Andrews, C.W., 1900. A Monograph of Christmas Island (IndianOcean): Physical Features
and Geology. British Museum (Natural History), London.
Barrie, J., 1967. The Geology of Christmas Island. Bureau of Mineral Resources, Australia,
104. (Unpublished report).
Baxter,J.L., Weeks, G.C., 1984. Phosphatised volcanic ore fromthe Christmas Island, Indian
Ocean. Western Australia Institute of Technology, 20. (Unpublished report).
Bodine, J., Steckler, M., Watts, A., 1981. Observations of flexure and the rheology of the
oceanic lithosphere. Journal of Geophysical Research 86, 3695–3707.
Braun, J., 2010. The many surface expressions of mantle dynamics. Nature Geoscience 3,
825–833.
Buchs, D.M., Pilet, S., Cosca, M., Flores, K.E., Bandini, A.N., Baumgartner, P.O., 2013. Low-
volume intraplate volcanism in the Early/Middle Jurassic Pacific basin documented
by accreted sequences in Costa Rica. Geochemistry, G eophysics, Geosystems 14,
1552–1568.
Butler, R.F., 1992. Paleomagnetism: Magnetic Domains to Geologic Terranes. Blackwell
Scientific Publications, Boston.
Cande, S.C.,Kent, D.V., 1995. Revised calibration of thegeomagnetic polarity timescale for
the Late Cretaceous and Cenozoic. Journal of Geophysical Research 100, 6093–6095.
Cande, S.C., Patriat, P., Dyment, J., 2010.Motion between the Indian, Antarctic and African
plates in the early Cenozoic. Geophysical Journal International 183, 127–149.
Cloetingh, S., Wortel, R., 1986. Stress in the Indo-Australian plate. Tectonophysics 132,
49–67.
Dalrymple, G.B., Clague, D.A., 19 76. Age of the Hawaiian–Emperor bend. Earth and
Planetary Science Letters 31, 313–329.
DeMets, C., Royer, J.,2003. A new high-resolution model for India–Capricorn motion since
20 Ma: implications for the chronology and magnitude of distributed crustal
deformation in the Central Indian Basin. Current Science 85, 339–345.
Deplus, C., Diament, M., Hébert, H., Bertrand, G., Domin guez, S., Dubois, J., Malod, J.,
Patriat, P., Pontoise, B., Sibilla, J.-J., 1998. Direct evidence of active deformati on in
the eastern Indian oceanic plate. Geology 26, 131–134.
Doak, B.,Gallagher, P., Evans,L., Muller, F., 1965. Low-temperaturecalcination of “C”-grade
phosphate from Christmas Island. New Zealand Journal of Agricultural Research 8,
15–29.
Dupont-Nivet, G., van Hinsbergen , D.J.J., Torsvik, T .H., 2010. Persistently low Asian
paleolatitudes: implications for the India-Asia collision history. Tectonics 29, TC5016.
Fujiwara, T., Hirano, N., Abe, N., Takizawa, K., 2007. Subsurface structure of the “petit-
spot”volcanoes on the northwestern Pacific Plate. Geophysical Research Letters 34,
L13305.
Grand, S.P., 2002. Mantle shear-wave tomography and the fate of subducted slabs. Philo-
sophical Transactions of the Royal Society of London. Series A: MathematicalPhysical
and Engineering Sciences 360, 2475–2491.
Grimes, K.G., 2001. Karst features of Christmas Islands. Helictite 37, 41–58.
Hager, B.H., Clayton, R.W., Richards, M.A., Comer, R.P., Dziewonski, A.M., 1985. Lower
mantle heterogeneity, dynamic topography and the geoid. Nature 313, 541–545.
Hall, R., 2012. Late Jurassic–Cenozoic reconstructions of the Indonesian region and the
Indian Ocean. Tectonophysics 570–571, 1–41.
Heine, C., Müller, R.D., 2005. Late Jurassic rifting along the Australian North West Shelf:
margin geometr y and spreading ridge configuratio n. Australian Jou rnal of Earth
Sciences: An International Geoscience Journal of the Geological Society of Australia
52, 27–39.
Heine, C., Müller, R.D., Steinberger, B., Torsvik, T.H., 2008. Subsidence in intracontinental
basins due to dynamic topography. Physics of the Earth and Planetary Interiors 171,
252–264.
Heirtzler, J.R., Cameron, P., Cook, P.J., Powell, T., Roeser, H.A., Sukardi, S.,Veevers, J.J., 1978.
Abyssal plain. Earth and Planetary Science Letters 41, 21–31.
Hirano, N., Kawamura, K., Hattori, M., Saito, K., Ogawa, Y., 2001. A new type of intra-plate
volcanism; young alkali-basalts discovered from the subducting PacificPlate,north-
ern Japan Trench. Geophysical Research Letters 28, 2719–2722.
Hirano,N., Takahashi, E., Yamamoto, J.,Abe, N., Ingle, S.P., Kaneoka, I., Hirata, T., Kimura,J.I.,
Ishii, T., Ogawa, Y., 2006. Volcanism in response to plate flexure. Science 313,
1426–1428.
Hirano, N., Koppers, A.A.P., Takahashi, A., Fujiwara, T., Nakanishi, M., 2008. Seamounts,
knolls and petit-spot monogenetic volcanoes on the subducting Pacific Plate. Basin
Research 20, 543–553.
405R. Taneja et al. / Gondwana Research 28 (2015) 391–406

Hoernle,K.,Hauff,F.,Werner,R.,vandenBogaard,P.,Gibbons,A.D.,Conrad,S.,
Müller, R.D., 2011. Origin of Indian Ocean Seamount Province by shallow
recycling of continental lithosphere. Nature Geoscience 4, 883–887.
Hofmann, C., Féraud, G., Courtillot, V., 2000.
40
Ar/
39
Ar dating of mineral separates and
whole rocks from the Western Ghats lava pile: further constraints on duration and
age of the Deccan traps. Earth and Planetary Science Letters 180, 13–27.
Jourdan, F.,Renne, P.R., 2007. Age calibration of the FishCanyon sanidine
40
Ar/
39
Ar dating
standard using primary K–Ar standard s. Geochimica et Cosmochimica Acta 71,
387–402.
Jourdan, F., Féraud, G., Bertrand, H., Watkeys, M.K., Renne, P.R., 2007. Distinct brief major
events in the Karoo large igneous province clarified by new
40
Ar/
39
Ar ages on the
Lesotho basalts. Lithos 98, 195–209.
Jourdan, F., Reimold, W.U., Deutsc h, A., 2012. Dating te rrestrial impact structures.
Elements 8, 49–53.
Kirschvink, J., 1980. The least-squares line and plane and the analysis of paleomagnetic
data. Geophysical Journal of the Royal Astronomical Society 62, 699–718.
Koppers, A.A.P., 2002. ArArCALC-software for 40Ar/39Ar age calculations. Computers &
Geosciences 28, 605–619.
Lee, J.Y., Marti, K., Severinghaus, J.P., Kawamura, K., Yoo, H.S., Lee, J.B., Kim, J.S., 2006. A
redeterminati on of the isotopic ab undances of atmo spheric Ar. Geochimica et
Cosmochimica Acta 70, 4507–4512.
Levitt,D.A., Sandwell, D.T.,1995. Lithosphericbending at subductionzones based on depth
soundings and satellite gravity. Journal of Geophysical Research 100, 379–400.
Lithgow-Bertelloni, C., Silver,P.G., 1998. Dynamic topography, plate driving forces and the
African superswell. Nature 395, 269–272.
Liu, C.-S., Curray, J.R., McDonald, J.M., 1983. New constraints on the tectonic evolution of
the eastern Indian Ocean. Earth and Planetary Science Letters 65, 331–342.
McDougall, I., Harrison, T.M., 1999. Geochronology and Thermochronology by the
40
Ar/
39
Ar Method. Oxford University Press.
McElhinny, M.W., McFadden, P.L., 2000. Paleoma gnetism: Continents and Oceans.
Academic Press, San Diego.
Montelli, R., Nolet, G., Dahlen, F.A., Masters, G., Engdahl, E.R., Hung, S.-H., 2004. Finite-
frequency tomography reveals a var iety of plumes in the mantle. Science 303,
338–343.
Montes-Lauar, C.R., Pacca, I.G., Melfi, A.J., Kawashita, K., 1995. Late Cretaceous alkaline
complexes, southeastern Brazil: paleomagnetismand geochronology.Earth and Plan-
etary Science Letters 134, 425–440.
Morgan, W.J., 1971. Convection plumes in the lower mantle. Nature 230, 42–43.
Moucha, R., Forte, A.M., Mitrovica, J.X., Rowley, D.B., Quéré, S., Simmons, N.A., Grand,
S.P., 2008. Dynamic topography and long-term sea-level variations: there is no
such thing as a stable continental platform. Earth and Planetary Science Letters
271, 101–108.
Müller, R.D., Mihut, D., Baldwin, S., 1998. A new kinematic model for the formation and
evolution of the west and northwest Australian margin. In: Purcell, P.G., Purcell, R.
R. (Eds.), The Sedimentary Basins of Western Australia, 2. Petroleum Exploration
Society of Australia, WA Branch, Perth, WA, pp. 55–72.
Müller,R.D., Gaina, C., Tikku,A., Mihut, D., Cande,S.C., Stock, J.M.,2000. Mesozoic/Cenozoic
tectonic events around Australia. In: Richards, M.A., Gordon, R.G., van der Hilst, R.D.
(Eds.), The History and Dynamics of Global Plate Motions. AGU, Washington, DC,
pp. 161–188.
Müller, R.D., Sdrolias, M., Gaina, C., Roest, W.R., 2008. Age, spreading rates, and spreading
asymmetry of the world's ocean crust. Geochemistry, Ge ophysics, Geosystems 9,
Q04006.
Nakanishi, M., Gee, J.S., 1995. 34: Pal eomagnetic investigations of volcanic rocks:
paleolatitudes of the northwestern Pacific guyots. In: Haggerty, J.A., Premoli Silva, I.,
Rack, F., McNutt, M.K. (Eds.), Proceedings of the Ocean Drilling Program. Scientific
results. Ocean Drilling Program, pp. 585–604.
Nyblade, A.A., Robinson, S.W., 1994. The African superswell. Geophysical Research Letters
21, 765–768.
O'Dowd, D.J., Lake, P.S., 1989. Red crabs in rain forest, Christmas Island: removal and
relocation of leaf-fall. Journal of Tropical Ecology 5, 337–348.
O'Dowd, D.J., Lake, P.S., 1991. Red crabs in rain forest, Christmas Island: removal and fate
of fruits and seeds. Journal of Tropical Ecology 7, 113–122.
O'Neill, C., Müller, D., Steinberger, B., 2005. On the uncertainties in hot spot reconstruc-
tions and the significance of moving hot spot reference fr ames. Geochemistry,
Geophysics, Geosystems 6, Q04003.
Pettifer, G.R., Polak, E.J., 1979. Christmas Island (Indian Ocean) geophysical survey for
groundwater, 1976. Bureauof Mineral Resources,Australia, 75. (Unpublished report).
Polak, E.J., 1976. Christmas Island (Indian Ocean) geophysical survey for groundwater,
1973. Bureau of Mineral Resources, Australia, 66. (Unpublished report).
Renne, P.R., Mundil, R., Balco, G., Min, K., Ludwig, K.R., 2010. Joint determination of 40
K decay constants and 40Ar*/40 K for the Fish Canyon sanidine standard, and im-
proved accuracy for 40Ar/39Ar geochronology. Geochimica et Cosmochimica
Acta 74, 5349.
Rivereau, J.C., 1965. Notes on a Geomorphological Stu dy of Christmas Isl and, Indian
Ocean, Bureau of Mineral Resources, Geology and Geophysics, Canberra, 5. (Unpub-
lished report).
Royer, J.Y., Gordon, R.G., 1997. The motion and boun dary between the Capricorn and
Australian plates. Science 277, 1268–1274.
Seton, M., Müller, R.D., Zahirovic, S., Gaina, C., Torsvik, T., Shephard, G., Talsma, A., Gurnis,
M., Turner, M., Maus, S., Chandler, M., 2012. Global continental and ocean basin
reconstructions since 200 Ma. Earth-Science Reviews 113, 212–270.
Sharp, W.D., Clague, D.A., 2006. 50-Ma initi ation of Hawaiian-Emperor bend records
major change in Pacific plate motion. Science 313, 1281–1284.
Sharp, W.D., Renne, P.R., 2005. The
40
Ar/
39
Ar dating of core recovered by the Hawaii
Scientific Drilling Project (phas e 2), Hilo, Hawaii. Geochemistry, Geophysics,
Geosystems 6, Q04G17.
Singh, S.C., Carton, H., Chauhan, A.S., Androvandi, S., Davaille, A., Dyment, J., Cannat, M.,
Hananto, N.D., 2011. Extremely thin crust in the Indian Ocean possibly resulting
from Plume–Ridge Interaction. Geophysical Journal International 184, 29–42.
Smith, W.H.F., Sandwell, D.T., 1997. Global sea floor topography from satellite altimetry
and ship depth soundings. Science 277, 1956–1962.
Stagg, H.M.J., Wilcox, J.B., Symonds, P.A., O'Brien, G.W., Colwell, J.B., A. H.P.J., C-S. L., G.
M.A.M., Struckmeyer, I.M., 1999. Architecture and evolution of the Au stralian
continental margin. AGSO Journal of Australian Geology & Geophysics 17, 17–33.
Steinberger, B., Schmeling, H., Marquart, G., 2001.Large-scal e lithosphe ric stress field and
topography induced by global mantle circulation. Earth and Planetary Science Letters
186, 75–91.
Taneja, R., O'Neill, C., 2014. Geophysical Investigation of seamount province of the North-
east Indian Ocean. Marine Geophysical Research (in-review).
Tarduno, J.A., 2007. On the motion of Hawaii and other mantle plumes. Chemical Geology
241, 234–247.
Tregoning, P., Brunner, F.K., Bock, Y., Puntodewo, S.S.O., McCaffrey, R., Genrich, J.F., Calais,
E., Rais, J., Subarya, C., 1994. First geodetic measurement of convergence across the
Java Trench. Geophysical Research Letters 21, 2135–2138.
Trueman, N.A., 1965. The phosphate, volcanic and carbonate rocks of Christmas Island
(Indian Ocean). Journal of the Geological Society of Australia 12, 261–283.
Veevers, J.J., 1971. Phanerozoic history of Western Australia related to continental drift.
Journal of the Geological Society of Australia 18, 87–96.
Verati,C.l, Jourdan, F., 2013.Modelling effect of sericitizationof plagioclase on the
40
K/
40
Ar
and
40
Ar/
39
Ar chronomete rs: implication for dating basaltic rocks and min eral
deposits. Geological Society, London, Special Publications 378.
Watts, A.B., 2001. Isostasy and Flexure of the Lithosphere. Cambridge University Press,
Cambridge.
Wei, W., 1995. Revised age calibration points for the geomagnetic polarity time scale.
Geophysical Research Letters 22, 957–960.
Wheeler, P., White, N., 2002. Measuring dynamic topography: an analysis of Southeast
Asia. Tectonics 21, 1040.
Whittaker, J.M., Müller, R.D., Leitchenkov, G., Stagg, H., Sdrolias, M., Gaina, C., Goncharov,
A., 2007. Major Australian–Antarctic plate reorganization at Hawaiian–Emperor bend
time. Science 318, 83–86.
Woodroffe, C., 1988. Vertical movement of isolated oceanic islands at plate margins: evi-
dence from emergent reefs in Tonga (Pacific Ocean), Cayman Islands (Caribbean Sea)
and ChristmasIsland (Indian Ocean). Zeitschrift für GeomorphologieSupplementband
69, 17–37.
Woodroffe, C.D., Falkland, A.C., 1997. Chapter 31 Geology and hydrogeology of the Cocos
(Keeling) Islands. In: Vacher, H.L., Terrence, M.Q. (Eds.), Developments in Sedimen-
tology. Elsevier, pp. 885–908.
Woodroffe, C.D., Veeh, H.H., Falkland, A., McLean, R.F., Wallensky, E., 1991. Last intergla-
cial reef and subsidence of the Cocos (Keeling)Islands, Indian Ocean. Marine Geology
96, 137–143.
Yue, H., Lay, T., Koper, K.D., 2012. En echelon andorthogonal fault ruptures of the 11 April
2012 great intraplate earthquakes. Nature 490, 245–249.
Zijderveld, J., 1967. AC demagnetization of rocks: analysis of results. In: Collinson, D.W.,
Creer, K.M., Runcorn, S.K. (Eds.), Methods in Paleomagnetism. Elsevier, Amsterdam,
pp. 254–268.
406 R. Taneja et al. / Gondwana Research 28 (2015) 391–406