1
2
India-Madagascar paleo-fit based on flexural isostasy of their rifted
margins
3
4
R.T. Ratheesh-Kumara,*, C. Ishwar-Kumara, B. F. Windleyb, T. Razakamananac, R. R.
5
Naird, K. Sajeeva
6
a
7
b Department
8
c Département
9
d
10
Centre for Earth Sciences, Indian Institute of Science, Bangalore -560012
of Geology, The University of Leicester, Leicester, LE1 7RH, UK.
de Sciences Naturelles, Université de Toliara, Toliara, Madagascar
Department of Ocean Engineering, Indian Institute of Technology Madras, Chennai-600
036, India.
11
12
Email: ratheesh.geo@gmail.com
13
14
15
16
17
1
18
Abstract
19
The present study contributes new constraints on, and definitions of, the reconstructed plate
20
margins of India and Madagascar based on flexural isostasy along the western continental
21
margin of India (WCMI) and the eastern continental margin of Madagascar (ECMM). We have
22
estimated the nature of isostasy and crustal geometry along the two margins, and have examined
23
their possible conjugate structure. Here we utilize elastic thickness (Te) and Moho depth data as
24
the primary basis for the correlation of these passive margins. We employ the flexure inversion
25
technique that operates in spatial domain in order to estimate the spatial variation of effective
26
elastic thickness. Gravity inversion and flexure inversion techniques are used to estimate the
27
configuration of the Moho/Crust-Mantle Interface that reveals regional correlations with the Te
28
variations. These results correlate well with the continental and oceanic segments of the Indian
29
and African plates. The present study has found a linear zone of anomalously low-Te (1-5 km)
30
along the WCMI (c. 1680 km), which correlates well with the low-Te patterns obtained all along
31
the ECMM. We suggest that the low-Te zones along the WCMI and ECMM represent paleo-rift
32
inception points of lithosphere thermally and mechanically weakened by the combined effects of
33
the Marion hotspot and lithospheric extension due to rifting. We have produced India-
34
Madagascar assembly during the initial phase of their separation, based on the Te estimates of
35
the rifted conjugate margins, which is confirmed by Moho geometry and bathymetry of the shelf
36
margins, and by the matching of tectonic lineaments, lithologies and geochronological belts
37
between India and Madagascar.
38
Keywords: Continental margin; isostasy; effective elastic thickness; Moho; lithosphere.
39
40
2
41
1.
Introduction
42
The temporal evolution and spatial configuration of continents can be analyzed through their
43
response to long-term forces, as a function of elastic property of the lithosphere, which is
44
parameterized as effective elastic thickness (Te). The Te method has been widely used as a key
45
proxy to examine the long-term strength/ rigidity structure of the lithosphere. It can be
46
parameterized through flexural rigidity, D≡E.Te3/12(1−ν2), which is a measure of the resistance
47
of the lithosphere to flexure in response to loading (Watts, 2001), where Young’s modulus, E
48
(1011 Pa), and Poisson’s Ratio, ν (0.25), are the material properties. The elastic thickness in
49
oceanic regions has values between 0 and 65 km that approximately correspond to the depth of
50
the 450oC isotherm (Watts, 1992). In contrast, the continents exhibit a Te range as high as 80+
51
km in stable regions (Watts and Burov, 2003), and as low as ~5 km in young and tectonically
52
rejuvenated regions (Watts, 2001). Although a possible correlation between Te and the age of the
53
lithosphere was studied in Europe (Perez-Gussinye and Watts, 2005) and Australia (Simons and
54
van der Hilst, 2002), most studies demonstrated that mechanical strength is not always the first
55
order to control by the age of the lithosphere. Te may be influenced by many factors including
56
localized brittle failure of crustal rocks under deviatoric stress (Lowry and Smith, 1995),
57
“Sandwich” deformation (decoupling) when a weak ductile layer in the lower crust does not
58
allow bending stresses to be transferred between the strong brittle layers (Burov and Diamant,
59
1995), “frozen” deformation by lattice preferred orientation of olivine as result of increased melt
60
production within the upper mantle (Simons et al., 2003), and large-scale tectonic features and
61
faults (Audet and Mareschal, 2004). Te has been broadly used as a key proxy for the rigidity of
62
the lithosphere; for example, it is correlated with shear wave velocity (Perez-Gussinye et al.,
63
2007), surface heat flow (Lowry and Smith, 1995), seismogenic thickness (McKenzie and
64
Fairhead, 1997; Watts and Burov, 2003) and seismic anisotropy (Audet and Mareschal, 2004).
3
65
In the present study, we aim to appraise spatial variations of elastic thickness in the so-called
66
conjugate passive margins of India and Madagascar (Fig. 1) such as the Western Continental
67
Margin of India (WCMI) and Eastern Continental Margin of Madagascar (ECMM), in order to
68
understand how the nature of isostasy varies in these margins, and to find any possible
69
correlation/ conjugate nature between them. In contrast to other geophysical investigations in the
70
WCMI and ECMM that used seismic, gravity and bathymetry data base to constrain the
71
geometry/ structure of these passive margins, the present study using Te variations effectively
72
maps the tectonic deformations within the lithosphere that can be a proxy to understand the
73
evolution of these passive margins.
74
Previous studies of passive margins in the world have shown variable results for Te. Stern
75
and Brink (1989) estimated a Te of ~19 km in the Ross Sea where rifting occurred at about 60
76
Ma, whereas in the Valencia trough where there is a comparatively young rifting age of 20 Ma,
77
elastic thickness estimates are ~5 km (Watts and Torne, 1992). Daly et al. (2004) computed the
78
elastic thickness of the Irish Atlantic margin using a multitaper coherence method between
79
scaled bathymetry and Bouguer gravity and obtained Te values of ~6-18 km. Wyer and Watts
80
(2006) applied flexural backstripping and gravity modeling techniques to calculate the gravity
81
anomaly associated with rifting and sedimentation anomaly along the east coast, USA
82
continental margin. They iteratively compared the calculated gravity anomalies to the observed
83
free-air gravity anomaly to derive a best-fit Te structure that show significant variation of
84
0<Te<40 km, which they attributed to the strength variation in the rifted lithosphere. Several
85
studies revealed crustal thinning and depth of necking as the parameters to predict the flexural
86
response of lithospheric stretching (Fourno and Roussel, 1994; Braun and Beaumont, 1989,
87
Keen and Dehler, 1997; Ratheesh Kumar et al., 2011). Ratheesh Kumar et al. (2011) used
88
orthonormalized Hermite multitaper method to estimate Te along the northeast passive margin of
4
89
North America, and suggested that the low-Te values indicate the passive nature of the margin
90
when the loads were emplaced during the continental break-up process at high temperature
91
gradients. Chand et al. (2001) examined the cross-spectral correlation between gravity and
92
bathymetry along 1D profiles across the eastern continental margin of India (ECMI) and its
93
conjugate East Antarctica margin. They obtained Te~10-25 km and Te<5 km over the northern
94
and southern segments of the ECMI, and suggested their possible match with the Te data of the
95
corresponding congruent segments of the East Antarctica margin. Subrahmanyam and Chand
96
(2006) re-examined gravity and topography/bathymetry data over India and the adjoining
97
oceans, and suggested that ECMI evolved in a shear tectonic setting, and bears similarities with
98
its conjugate half in East Antarctica.
99
There are a few studies based on the Te estimates in the WCMI and ECMM using different
100
techniques operates in spectral domain. Chand and Subrahmanyam (2003) estimated Te of the
101
western margin of India and eastern margin of Madagascar through cross-spectral analysis of
102
gravity and bathymetry data. Based on the comparable Te results of 8-15 km for the WCMI and
103
10-13 km for ECMM, they interpreted the conjugate nature of these margins. Sheena et al.
104
(2007) employed rectangular blocks for the coherence analysis along the Konkan and Kerala
105
basin of the WCMI, which reveal a variation of lithospheric strength from 5-10 km. Choubey et
106
al. (2008) derived Te using admittance (cross-spectral relation) between 12 gravity and
107
bathymetry profiles across the Laccadive Ridge, and obtained low-Te (2-3 km) values, which
108
they attributed to the local compensation of stretched continental lithosphere. Ratheesh Kumar et
109
al. (2014) derived the spatial variation of the elastic thickness structure of the Indian Shield and
110
adjoining regions using a fan wavelet-based Bouguer coherence technique. Their Te map reveals
111
zones of significantly low Te along the western margin of the Indian Shield, which they
112
attributed to the rifting processes.
5
113
In the present study, we use a thin plate flexure model (Braitenberg et al., 2002, 2006),
114
which is an alternative to the widely used calculation of admittance/coherence of topography
115
and gravity. This analysis is based on the convolution method that models surface and
116
subsurface loads with the point load response function of the elastic plate in spatial domain.
117
The present study use Bouguer gravity and bathymetry/topography to estimate the spatial
118
variation of effective elastic thickness and the Moho configuration in the western continental
119
margin of India and the eastern continental margin of Madagascar.
120
2. Study Region: Tectonic setting
121
The position and morphological relationship of India relative to Madagascar in the past is
122
one of the most debated and current problems in understanding the tectonics of eastern
123
Gondwana (e.g., Katz and Premoli, 1979; Collins and Windley, 2002; Braun and Kriegsman,
124
2003; Ghosh et al., 2004; Collins and Pisarevsky, 2005; Collins, 2006; Ashwal et al., 2013;
125
Gibbons et al., 2013; Ishwar-Kumar et al., 2013; Rekha et al., 2013). The breakup of Gondwana
126
started with the roughly simultaneous rifting of Madagascar, Seychelles, India, Antarctica and
127
Australia from Africa at around ca. 150 Ma. Then, Madagascar, Seychelles and India separated
128
together from Antarctica and Australia at about 128-130 Ma (Biswas, 1999). At ca. 90 Ma India
129
and Seychelles further rifted from Madagascar, and at about 65 Ma India separated from
130
Seychelles (Pande et al., 2001). The record of the position and movement of India after rifting
131
from Madagascar was long and eventful. Geophysical data suggest that the Indian plate
132
continually drifted northwards after its separation from Africa in the Late Jurassic to finally
133
collide with Eurasia at about 55 Ma (Yin and Harrison, 2000). The oldest seafloor anomaly
134
recognized is c34 (83 Ma), such that the onset of seafloor spreading occurred sometime during
135
the Cretaceous Quiet Zone (120-83 Ma). The fragmentation took place in three stages: doming,
136
rifting and drifting. According to Storey et al. (1995) the Marion hotspot initiated the breakup of
6
137
India and Madagascar at about 88 Ma, resulted in the formation of the Western Continental
138
Margin of India and the Eastern Continental Margin of Madagascar (Fig. 1).
139
2.1 The Western Continental Margin of India
140
The western continental margin of India (WCMI) is characterized by a wide continental
141
shelf (>300 km) and thick shelf sediments (7-8 km) of Indus fan origin (Zutshi et al., 1995).
142
South of the Vengurla arch (15°46'13"N, 73°40'48"E) (Fig. 1) the shelf is narrow (<100 km) and
143
characterized by 3-4 km-thick sediments that are mainly derived from denudation of the Western
144
Ghats and concentrated in small, localized depressions (Zutshi et al., 1995). The offshore shelf
145
basins can be regionally classified into three: northern Kutch and Sourashtra basins, central
146
Bombay basin, and southern Konkan and Kerala basins (Fig. 1). To the west of the shelf margin
147
the transitional crust is restricted by the Kori-Comorin ridge, a typical longitudinal ridge
148
identified close to the foot of the continental slope along the western continental margin, which
149
could possibly be the ocean-continent boundary (Biswas, 1987, 1988). The western margin of
150
India is geomorphologically similar to other rifted continental margins like Parana of Brazil,
151
Karoo in southeast Africa, and Etendeka in southwest Africa (Widdowson, 1997).
152
The northern segment of the WCMI is occupied by plume-generated flood basalts of the
153
Deccan Traps (Beane et al., 1986) that have a maximum thickness of >3 km, and which migrated
154
southwards during the plume activity (Jay and Widdowson, 2008). Trace element geochemical
155
data indicate increasing degrees of partial melting from north to south (Peng and Mahoney,
156
1995); the shallower and higher degrees of melting in the south were explained by Kumar (2003)
157
as the result of lithospheric thinning, which would be consistent with the progressive southward
158
opening of the India-Madagascar rift. The Deccan Traps started to erupt 65 Ma ago from the
159
Réunion hotspot (Courtillot et al., 2003).
7
160
2.2 The Eastern Continental Margin of Madagascar
161
The eastern continental margin of Madagascar (ECMM) has a narrow coastal plain marked
162
by NNE/SSW-striking Cenozoic normal faults that impart a remarkable, strong linearity to the
163
coastline. Cretaceous basalts and minor rhyolites (ca. 88 Ma) are prominent all along the ECMM
164
(Storey et al., 1997); these coastal rift volcanic rocks and central flood basalts formed within a
165
short period between 92 and 84 Ma (Melluso et al., 2001). Swarms of coast-parallel dolerite
166
dykes that have K-Ar ages ranging from 97 to 89 Ma (Storetvedt et al., 1992) intruded during
167
Early Cretaceous rifting along the NE coast of Madagascar (Bauer et al., 2011). Apatite fission
168
track data suggest that some escarpments on the eastern coast date from the time of rifting with
169
India (Seward et al., 1999).
170
The geochemical signatures of the eastern coast basalts reflect their different mantle source
171
regions as the rift with India opened from north to south (Storey et al., 1997; Melluso et al.,
172
2001, 2002). On the northeastern coast low-Ti basalts are similar to the low-Ti flood basalts of
173
the Deccan Traps on the opposite northwestern coast of India. To the south along the central
174
coast of Madagascar basalt geochemistry is dominated by an Indian Ocean-type MOR-source
175
mixed with a component of old continental mantle lithosphere (Mahoney et al., 2008). On the
176
southeast coast, high Fe-Ti basalts are similar to those on the East Greenland volcanic rift
177
margin, and Nd, Pb and Sr isotopic data indicate a significant Marion hotspot plume component
178
(de Wit, 2003). Paleomagnetic data from the basalts combined with magnetic anomalies and
179
fracture zones of the Indian Ocean provide strong evidence that the Marion hotspot was situated
180
within 100 km of southern Madagascar when it separated from the Seychelles-India continent at
181
about 90-88 Ma (Storey et al., 1997; Torsvik et al., 1998, 2000; Reeves and de Wit, 2000).
182
3. Data and Method
8
183
Our study areas cover most of the western continental margin of India and the eastern
184
continental margin of Madagascar. The database used for this study comprises gravity,
185
bathymetry/topography and sediment thickness. The bathymetry data (Figs. 2a and 3a) were
186
obtained from GEBCO Digital 1-minute bathymetry data (National Oceanic and Atmospheric
187
Administration, 2003). We merged the gravity data for land and ocean by using the land ocean
188
deconvolution technique (Kirby and Sawain, 2008). The free-air gravity data were derived from
189
the global marine gravity field from ERS-1 and GEOSAT geodetic mission altimetry of
190
Anderson and Knudsen (1998) and Anderson et al. (2008). The free-air gravity anomaly data
191
(Gf) was converted to Bouguer gravity anomalies (Gb) (Figs. 2b and 3b) using the slab
192
formula of Parker (1972):
193
Gb = Gf + 2GH
(1)
194
where =1670 kg.m-3 is the density contrast between surface rock and water, H is the
195
bathymetry (in meters) and G is the gravitational constant.
196
We use the sediment thickness Model (Figs. 2c and 3c) of Divins (2003), which was compiled
197
by the National Geophysical Data Centre (NGDC) of NOAA (National Oceanic and
198
Atmospheric Administration), and has a resolution of 5 × 5 arc minute.
199
3.1 Flexure modeling in spatial domain (convolution method):
200
We adopt a methodology that operates in the spatial domain introduced by Braitenberg
201
et al. (2006), which they successfully used in their analysis of the South China Sea Ridge. In
202
this method, Moho depths are first estimated from forward modeling of gravity anomalies;
203
then, the lithosphere rigidity is inverted in order to retrieve isostatic Moho depth undulations
204
compatible with those previously obtained.
9
205
In this method, we first model the Crust Mantle Interface/ Moho depth undulations, which
206
contribute to the long wavelength part of the observed gravity field, whereas the short
207
wavelength part is generated by superficial masses viz., sediment layers or intra-crustal density
208
inhomogeneities. However, sometimes sedimentary basins can also produce long-wavelength
209
signals. Hence, it is essential to estimate the gravity effect of sediments in isostatic flexure
210
modeling. Furthermore, on a passive continental margin, large amounts of sediments will simply
211
erase any signal of load-induced topography (i.e. flat topography is unrelated to flexure). For
212
these reasons we isolated the effect of sediments from the observed gravity and bathymetry to
213
recover the basement structure. As an initial step, the base of the sediments was generated by
214
subtracting the sediment thickness from the bathymetry. The obtained sediment corrected
215
basement (Figs. 2e and 3e) will now represent the actual bathymetric features that are previously
216
masked by the thick sediment cover. A linear density variation with depth, ρ(z) is calculated
217
from the following expression.
218
ρ(z) = ρtop + (ρlow - ρtop) hsed / (hlow - htop)
(2)
219
where ρtop and ρlow represents the density values corresponding to the top (2.25 Mg/m3) and
220
bottom (2.7 mg/m3) layers, hsed is the sediment thickness, hlow and htop represents the depth to the
221
top and bottom layers respectively. The gravity effect is then calculated by subtracting the
222
density of the reference model from ρ(z). The obtained gravity effect of sediments (Figs. 2d and
223
3d) is then subtracted from the observed gravity to obtain the sediment-corrected gravity (Figs.
224
2f and 3f), which is used for the flexural modeling. In order to filter the input gravity field, we
225
defined a cut-off wavelength that suppresses all wavelengths smaller than 100 km. This
226
sediment-corrected Bouguer gravity field is then inverted by applying an iterative algorithm that
227
alternates downward continuation with direct forward modeling (Braitenberg et al., 1997). Thus,
228
we obtained Moho undulations inverted from the Bouguer gravity data.
10
229
The next modeling step is the flexure inversion, an independent means to determine the
230
physical flexural model of Moho undulations, and it allows the gravity-deduced Moho
231
undulations to be checked for compatibility. The flexure is calculated by the convolution of the
232
crustal load (i.e., topographic and subsurface loads) with the point-load flexure response curves
233
(Braitenberg et al., 2002, 2003). In order to avoid separate analyses and inversions on land and
234
ocean areas, we scaled the sediment-corrected ocean bathymetry (h) to equivalent topography
235
(h’) using the equation, h’ = (ρc−ρw)h/ρc, where ρc and ρw are the densities of crust and water,
236
respectively. The equivalent topography represents the bathymetry that one would assume if
237
there were no water present under isostatic conditions (Daly et al., 2004; Kirby and Swain,
238
2008). Accordingly the derived equivalent topography for the WCMI (Fig. 4a) and ECMM (Fig.
239
4b) are used in the present convolution scheme. A series of flexural response functions are used
240
in the convolution to model the crust-mantle interface undulations, each corresponding to a Te
241
value between 0 and 20 km. The spatial variation of Te is calculated on sliding square windows
242
of side length 100 sq. km that shifted every 20 km. The obtained Te value for a specific window
243
is the one that minimizes the root mean square (rms) difference between the flexure Moho and
244
the observed Moho derived from the gravity inversion. The elastic model parameters used in the
245
flexure analysis are given in Table 2.
246
3.2 Advantages and limitations
247
We now discuss possible concerns regarding the validity of the present method and
248
sensitivity of input parameters for inferring Te. We assumed a continuous-plate rather than
249
broken-plate model for the present analyses. Braitenberg et al. (2002) tested the convolution
250
method in a synthetic model situation, and successfully used the continuous plate model to
251
recover the spatial variation of elastic thickness over the Eastern Alps. However, they observed
252
some discrepant decrease in Te values in the main Alpine range, and explained it as the possible
11
253
result of recent tectonic forces acting at the border of two merging plates. Braitenberg et al.
254
(2006) assumed a continuous plate model and demonstrated the use of the convolution method
255
for the estimated spatial variation of Te in a mixed land-ocean setting in and around the South
256
China Sea. They derived the spatial variation of Te for the oceanic lithosphere, and the included
257
terrestrial parts are blanked in their results. We follow the approach of Braitenberg et al. (2002,
258
2006) by assuming that the present study regions are paleo-rift margins where the continental
259
and oceanic plates are expected to be very well coupled and hence can be considered as a
260
continuous plate, in which case the formalism assumes that we have only vertical loads with no
261
horizontal stresses. In other words, the continuous-plate model assumes there is no edge of chaos
262
in the passive margin setting in contrast to an active rift or subduction zone setting, where two
263
different plates will be pushed/pulled from the side and a broken-plate model is likely more
264
applicable. Furthermore, we used equivalent topography (Figs. 4a and b) rather than simple
265
bathymetry, and that allows the land-loading equations to be applied for a whole land-ocean
266
setting (Pérez-Gussinyé et al., 2004). Kirby and Swain (2008) used scaling in their synthetic
267
modelling and demonstrated its use in recovering Te in mixed a land-sea setting with a
268
negligibly small bias. Recently, Jiménez-Díaz et al. (2014) by using both multitaper and wavelet
269
(Bouguer coherence) methods in and around Central America successfully demonstrated that Te
270
can be better recovered in a mixed land-ocean setting when using the scaling (equivalent
271
topography) and land-loading equations.
272
The sensitivity of the model to input parameters (e.g. density contrast within a plate) would
273
be another complicating factor in the flexure modeling of a mixed land-ocean setting. In the
274
absence of constraining data, we set a constant density contrast (Δρ) across the crust-mantle
275
interface (CMI). However, we tested the model sensitivity with different combinations of density
276
contrast (Δρ~350-600 kg/m3) and reference depth (d~20-35 km) standard ranges of the CMI, and
12
277
a best-fit Te is deduced from the minimum of root mean square error (rms) between the observed
278
and computed CMI. The best results (i.e., minimum rms) were obtained for the set of parameters
279
Δρ~450 kg/m3 and d~30 km.
280
Earlier Te estimates of comparable passive margins, such as the flexural analyses of Chand
281
and Subrahmanyam (2003), Sheena et al. (2007, 2012), Tiwari et al. (2007), and Chaubey et al.
282
(2008) were carried out in spectral domain along an 1D profile or using some discrete windows
283
of variable size, but they could not produce the spatial variations necessary for the effective
284
elastic thickness (Te), which we consider a serious shortcoming. In contrast to the earlier studies,
285
the thin plate flexure model applied in the present study operates in the spatial domain
286
(convolution method), which has such a significant advantage over the spectral methods that it
287
overcomes the numerical instabilities in the admittance/coherence calculations. The spatial
288
variation is achieved by dividing the analysis area into overlapping windows of size 100 sq. Km,
289
where Te is calculated and inverted for each window, and then moving the centre of each
290
window by 20 km in order to cover the entire investigated area for each new estimate. This
291
provides spatial variations of the flexural properties with higher resolution than any of the
292
spectral approaches. Another significant advantage is that this analysis can be made over an area
293
that is not necessarily rectangular, as required for the spectral analysis. Recently, Ratheesh
294
Kumar and Windley (2013) used flexure inversion technique in combination with a spectral
295
technique (Morelet wavelet based Bouguer coherence) to derive the Te structure of the
296
Ninetyeast Ridge in the Indian Ocean, and demonstrated that both the spatial and spectral
297
estimates provide spatial variations that are mutually complementary.
298
4. Results
13
299
The effective elastic thickness (Te) maps estimated for the data windows (a and b in Fig. 1)
300
over the WCMI and ECMM are presented in Figs. 5 and 6 respectively. We also present the
301
Moho models derived from gravity inversion and flexure inversion analysis of the WCMI (Figs.
302
7a and b) and ECMM (Figs. 7c and d). The gravity inversion and flexure inversion results are in
303
good agreement, because the residual Moho (Figs. 7e and f) (mismatch between the gravity
304
inversion-derived Moho and flexure inversion-derived Moho) has a very low range (average of
305
±3 km). In Figs. 7e and f, it is clear that the marginal segments of India and Madagascar show an
306
rms range of ±1 km. The model of Moho undulation on each window is determined for a specific
307
Te, and hence the low range of the root mean square error is yet another quality check of the
308
present Te results.
309
Tables 3 display a comparison of Te values obtained in the present study and t the
310
estimates from the previous studies over various tectonic provinces in and around the WCMI and
311
ECMM.. The obtained Te maps (Figs. 5 and 6) correlate well with the morphological features in
312
the study regions, and resolve regional-scale structures. A narrow linear patch of anomalously
313
low-Te is immediately observed on the western Indian shelf region (Fig. 5) that receives our
314
main attention in this study. Away from the shelf margin, the Laccadive Ridge exhibits a
315
significantly low-Te signature, whereas the adjacent terrains to its east and west are
316
distinguished by higher Te values that separate the ridge from the shelf margin and Arabian
317
basin, respectively. To the north, the Laxmi ridge (Fig. 5) exhibits a similar low-Te range with
318
higher values on its sides. Over most of the Arabian basin the elastic thickness is significantly
319
low (Te<3 km). Different tectonic provinces included in the study area within the southern
320
Indian shield exhibits significant Te variations. Within the continental regime of India included
321
in the data window, significant Te variations are observed over the Deccan Volcanic Province
322
(DVP), Dharwar craton, and southern granulite terrain (SGT) (Fig. 5).
14
323
In the case of Madagascar (Fig. 6), its Archaean cratonic interiors exhibit higher Te values (~20
324
km), whereas, the marginal zones are characterized by a significantly low-Te range (1-10 km).
325
The entire stretch of the ECMM including the narrow shelf zone and the adjacent ocean basin
326
exhibits uniformly low-Te values, similar to those obtained from the WCMI. Towards the
327
southern end of the eastern margin, the low-Te estimates in the shelf adjacent to the Madagascar
328
basin, and the fossil ridge segment together contribute the features of an extensively weak
329
lithosphere. Farther away from this margin, the Reunion (also called La Reunion) and Mauritius
330
chains exhibit higher Te (~18-20 km), whereas in its northward continuity, the Nazareth Bank
331
region shows low-Te values (2-8 km) (Fig. 6).
332
The Moho models clearly depict the transition from thick continental to thin oceanic
333
crusts, and exhibit significant undulations that correlate with regional-scale features. The
334
continental interiors of India and Madagascar show a high crustal thickness (>35 km), which
335
decreases towards the margins. The Moho undulations beneath the continental shelves of India
336
and Madagascar correlate well with each other, and both are in the range of 25 to 30 km. In the
337
WCMI, the Laccadive ridge is underlain by a 20-25 km thick crust, whereas in the Laxmi ridge
338
and its surroundings, it is in the range of 15-20 km. In the Arabian basin the crustal thickness
339
decreases progressively from north (<15 km) to south (~8 km). In the ECMM, the ocean basin
340
adjacent to the narrow shelf exhibits a uniformly low crustal thickness (average ~10 km) from
341
north to south, with a significant and extensive thin crust observed in the southernmost regimes.
342
The Reunion-Mauritius-Nazareth Bank chain in the Moho map is distinguished by a higher
343
crustal thickness than its surrounding oceanic lithosphere. A progressively increased crustal
344
thickness is particularly evident from the Reunion (~20 km), Mauritius (~25 km), and Nazareth
345
Bank (~30 km) regimes. The present Moho depth values are in good agreement with the
346
published seismic and gravity constrained estimates (Table 4).
15
347
5. Discussion
348
The spatial variations of elastic thickness and Moho depth in the continental-oceanic
349
realms of India and Madagascar reveal some important insights into the evolution and
350
deformation of the different lithologic units. An interesting observation in the present study is
351
the NW-SE trending linear zone of significantly low-Te (0-5 km) obtained along the WCMI
352
(particularly in the shelf region) and its remarkable resemblance to the similar value zone and
353
pattern of Te obtained along the ECMM. The present low Te result in the WCMI is consistent
354
with the spectral estimates of Sheena et al. (2012), who inferred low Te variations over the
355
Konkan Basin (Te~5 km) and the Kerala Basin (Te~10 km) by considering the lithospheric
356
necking model. Several studies supported the rift-related lithospheric deformations along the
357
WCMI and its congruent ECMM. Chand and Subrahmanyam (2003) obtained an elastic
358
thickness of 8-15 km for WCMI and 10-13 km for ECMM using a one-dimensional free air
359
admittance function, and suggested that these low-Te values represent the signatures of rifting
360
between India and Madagascar. The Moho topography derived from Bouguer gravity inversion
361
by Fourno and Roussel (1994) revealed a NE-trending zone of substantially thinned crust in the
362
Precambrian basement of eastern and central Madagascar, which they attributed to the separation
363
of India during the Cretaceous. Windley and Razakamanana (1996) suggested that the Moho
364
topography and zone of thinned crust of Fourno and Roussel (1994) concided with a zone of
365
extensional structures in the basement related to extensional collapse of the Neoproterozoic
366
orogen, and Kusky et al. (2010) pointed out that the thinned crustal zone is expressed by a post-
367
Miocene graben system along the centre of Madagascar, whch may be an incipient expression of
368
the East African Rift System along an extension of a diffusive plate boundary.
369
370
16
371
5.1. The anomalously low Te Zones along the passive margins—Rift related?
372
Although we obtained comparable Te results along the conjugate margins of India and
373
Madagascar, there is a need to clarify why one would expect the Te to be similar on both
374
margins. By considering the well-documented tectonic history of the WCMI, two major episodes
375
of lithospheric deformation can be taken into account for the anomalously low Te signature: 1.
376
The deformation as a result of the rifting processes including lithospheric stretching, crustal
377
necking and emplacement, and volcanic loading during the early phases of India-Madagascar
378
separation date back from ca. 90 Ma BP; 2. The lithospheric deformation caused by the Reunion
379
hotspot during the northward drift of India at about 65 Ma ago. Thus, there is a point of potential
380
confusion regarding which parameter (rift or plume) played the predominant role in formulating
381
the elastic thickness along these passive margins. The later possibility was ruled out by Chand
382
and Subrahmanyam (2003) and Choubey et al. (2008). According to their idea, the Indian Plate
383
moved at such a fast rate (13.5 cm/year) over the Reunion hotspot between 66 and 48 Ma that
384
the thermal rejuvenation may have been insufficient to change the plate strength.
385
The Reunion hotspot traces can be observed in both ECMM and WCMI. The major
386
bathymetric features in the ocean to the east of Madagascar such as Nazareth Bank, Mauritius
387
and Reunion Island, which are considered to be the remnants of the Reunion hotspot, exhibit
388
anomalously thick crust and significant Te variations. The crustal thicknesses beneath Reunion
389
(~20 km), Mauritius (~25 km), and the Nazareth Bank (~30 km) obtained in the present study are
390
consistent with the estimates of Torsvik et al. (2013). Tiwari et al. (2007), using a free-air
391
admittance technique to estimate Te along the Deccan- Reunion hotspot track, obtained a
392
decrease in Te from 30 km over Reunion and Mauritius to 13 km over the Nazareth Bank. They
393
suggested that the higher Te regions resulted from intraplate emplacement on old lithosphere,
394
whereas the lower Te estimates in the Nazareth Bank were due to emplacement on the flank of
17
395
the Central Indian Ridge, where both plume and mid-ocean ridge basalts were emplaced on
396
young lithosphere. Our present Te results in Reunion and Mauritius with their high values (Te
397
~18-20 km), and Nazareth Bank with its low value (Te~5 km) support the concept of different
398
emplacement mechanism, as proposed by Tiwari et al. (2007).
399
The Laccadive-Chagos ridge in the WCMI is considered to be the track of the Indian plate
400
over the Reunion hotspot (See Fig. 5 for the hotspot track). Chaubey et al. (2008) analyzed the
401
isostatic compensation mechanism beneath the Laccadive Ridge using free air admittance, and
402
they obtained a significantly low-Te value of ~2.5 km. Their results favor a model of Airy
403
isostatic compensation beneath the Laccadive ridge that resulted when stretched continental
404
lithosphere was loaded during an initial stage of rifting. The present study obtained a
405
significantly low-Te (1-3 km) over the Laccadive Ridge (Fig. 5) with a crustal thickness estimate
406
of ~20-25 km, which may support the idea that the crustal loads in this ridge segment were
407
isostatically compensated as a result of thermal rejuvenation of the lithosphere and subsequent
408
volcanic loading by hotspot magmatism in the Late Cretaceous-Early Tertiary. The effect of the
409
Reunion volcanism is also apparent in the continental part of the Indian plate. Recently a
410
published Te map of the Indian Shield by Ratheesh-Kumar et al. (2014) clearly demarcated the
411
Deccan volcanic province affected by the Reunion hotspot volcanism. In the present study, an
412
anomalous low-Te (1-5 km) zone to the north of the Dharwar Craton, supports the idea of
413
Ratheesh-Kumar et al. (2014) that this part of the lithosphere had a long thermal interaction with
414
the Reunion plume centre.
415
From Fig. 5 the linear low Te pattern of the Deccan Volcanic Province coincides with the
416
Reunion hotspot track. In contrast, the anomalously low Te zone observed parallel to the shelf
417
region shows a markedly different trend. These two contrasting Te patterns may imply two
418
different possible tectonic events that resulted in lithospheric deformation. The Te map of
18
419
Ratheesh-Kumar et al. (2014) shows zones of anomalously low Te in the western margin of the
420
Western Dharwar Province and in the adjacent shelf region, which they inferred as thermally and
421
mechanically weakened lithosphere caused by the combined action of the Marion hotspot and
422
rift-related lithospheric extensional processes. Most importantly, the present study finds a
423
profound correlation between the NNW/SSE-trending zone of anomalously low-Te (Fig. 5) and
424
the prominent linear bathymetric features including the mid-shelf basement ridge, inner-shelf
425
graben, shelf margin basin and the Prathap Ridge complex that has a similar trend transect along
426
the shelf basement. Subrahmanyam et al. (1995) suggested that the mid-shelf basement ridge and
427
the Prathap Ridge complex are rift-related ridges formed during the separation of India from
428
Madagascar around 84 Ma, and that they followed the pre-existing trends of the Precambrian
429
basement fabric. According to Chaubey et al. (2002), the presence of rotated fault blocks at the
430
shelf margin basin, and the emplacement of the volcanic Prathap Ridge complex indicate a failed
431
rift and volcanism of the stretched continental regime of the basin. We now suggest that the
432
anomalously low-Te zone is the sum effect of the flexural response of the rift-related
433
surface/subsurface structural features and the volcanic emplacements along the WCMI. Support
434
for our interpretation comes from available geochronological data related to both passive
435
margins. The rifting of India from Madagascar was accompanied by the formation of
436
voluminous flood basalt flows and dolerite dykes with subordinate rhyolite flows along the rifted
437
margin of India (Pande et al., 2001). The rhyolites and rhyodacites from St. Mary’s island off the
438
western coast of southern India have K-Ar ages in the range of 97-80 Ma (Valsangkar et al.,
439
1981), and a
440
Madagascar. Torsvik et al. (2000) obtained U-Pb zircon age of ca. 91 Ma from the St. Mary’s
441
island, which they linked with the late Cretaceous magmatic province in Madagascar (≈ 84-92
442
Ma) and Analalava gabbro pluton (~91 Ma). Also related to the rifting are ENE/WSW-striking
443
dykes in Karnataka, western India that have a
40
Ar-39Ar age of ca. 86 Ma (Pande et al., 2001) related to rifting of India from
40
19
Ar-39Ar age of about 88-90 Ma (Anil Kumar et
444
al., 2001), and leucogabbro and felsite dykes from southwestern India that have a K-Ar age of
445
ca. 85 Ma (Pande et al., 2001). The eastern coast of Madagascar contains several mafic-
446
ultramafic complexes, which are remnant signatures of rifting that is dated at 92-84 Ma (Storey
447
et al., 1995; Melluso et al., 1997, 2001, 2002, 2005; Torsvik et al., 1998; Mahoney et al., 2008;
448
Cucciniello et al., 2010, 2011). The Antampombato–Ambatovy complex in the east-central part
449
of the Cretaceous flood basalt province of Madagascar has an 40Ar/39Ar incremental heating age
450
of ca. 90 Ma and U–Pb age of ca. 90 ± 2 Ma (Melluso et al., 2005). Mahoney et al. (2008)
451
suggested high-level, pre-breakup lithospheric extension between India and Madagascar, inferred
452
from the great concentration of rhyolite dykes and significant crustal contamination of basalt on
453
the central eastern coast of Madagascar. These lines of evidence clearly suggest that the Marion
454
hotspot and associated rifting processes contributed to the weak strength of both the WCMI and
455
ECMM. Thus the present study conclude that the anomalously low Te zones along the
456
continental shelf and adjacent oceanic regimes indicate the deformations within the passive
457
margin lithosphere, and apparently these deformations can be best explained by the rift related
458
processes including lithospheric thinning/ necking and hotspot interactions. This idea now
459
clearly justifies why similar values of Te along the conjugate margins of India and Madagascar
460
can be correlated to examine their possible conjugate nature.
461
5.2 Fit of Conjugate Margins Reconstructed from Te Correlation:
462
In Fig. 5 we find that the zone of lithospheric deformation along the WCMI, indicated by
463
anomalously low Te pattern, is extending in a NW-SE trend for a total length of ~1680 km. This
464
length value is strikingly coinciding with the full stretch of the ECMM characterized with similar
465
and uniformly low Te zone (Fig. 6). We matched these characteristic linear low Te zones
466
between the two conjugate margins and obtained a fit position of Madagascar against India,
467
which is presented in Fig. 10. The Moho models derived from flexure inversion analysis have
20
468
been examined to find any possible match of the conjugate margins on the fit position deduced
469
from Te model. In the present Moho models, a crustal thinning towards the continental margins
470
of India (Fig. 8) and Madagascar (Fig. 9) is evident, which define the actual extend of
471
continental margin. Here, the ocean ward extent of continental margin/ shelf can be defined by a
472
rectilinear zone of Moho configuration (~25 km deep). It is obvious that the Madagascar is
473
characterized by a narrow shelf zone, while the shelf of India is comparatively broader and its
474
width increases towards north. We matched the two shelf zones of similar Moho configuration
475
on a fit position derived from Te match, and obtained a close geometrical fit of Moho between
476
the continental margins of India and Madagascar (Fig. 11). We then superimposed the
477
bathymetry contours on this fit position (Fig. 11), and find that 1000 m isobath that represents
478
both the shelf margins are also show a reasonably good close-fit. Thus, the ‘fit position’ of the
479
continental margins deduced from the Te correlation is well justified by the Moho and
480
bathymetry configurations, and ultimately produce a unique paleo-continental configuration of
481
India and Madagascar.
482
483
5.3 Previous Perspectives
484
The exact position of India against Madagascar still remains a matter of debate, and clearly
485
there is a primacy to re-examine the previous data in the light of the present study to better
486
understand, or even resolve, this problem. Many published plate reconstructions used the
487
matching of structural lineaments to find the original form and coherence of the WCMI and the
488
ECMM (Katz and Premoli, 1979; Storey et al., 1995; Braun and Kriegsman, 2003; Ghosh et al.,
489
2004; Ishwar-Kumar et al., 2013). The time of breakup of the matching margins was defined by
490
several geochronological methods such as K-Ar dating (Valsangkar et al., 1981) and apatite
491
fission track analysis (Chand and Subrahmanyam, 2003, Emmel et al., 2006). Also, Marks and
21
492
Tikku (2001) used free-air gravity and topography in combination with magnetic anomaly data
493
to reconstruct the gravity and topography fields in the Cretaceous period in order to determine
494
the correct fit of Africa, Antarctica and Madagascar. Katz and Premoli (1979) defined two
495
possible positions of Madagascar relative to India based on the matching of tectonic lineaments,
496
namely the Bhavani lineament of southern India and the Itremo and Ranotsara lineaments in
497
Madagascar, respectively. According to Powell et al. (1997), the southern tip of India was over
498
1000 km south of Madagascar during the initial stages of its northward drift (Brian, see the
499
reviewers Comment No.8 given in Revision Notes, and please include the references Reeves
500
and de Wit (2000), Schettino and Scotese (2005), Gaina et al (2007) as suggested by the
501
reviewer). However, according to Gibbons et al. (2013), the southern tip of India lies ~250 km
502
north of the southern edge of Madagascar, and they proposed a dextral-transtensional motion
503
between these two continents that culminated in a diachronous rifting. Torsvik et al. (2000)
504
postulated fit of India and Madagascar prior to and during the early phase of Madagascar-India
505
separation in the Late Cretaceous by correlating the breakup related paleomagnetic anomaly runs
506
sub-parallel with the Southwest India. Mishra et al. (2014) schematically showed the position of
507
India-Seychelles bank and Madagascar in a 65-70 Ma plate reconstruction, primarily based on
508
the paleostress trends deduced from field as well as remotesensing analyses in the western part
509
of Deccan large igneous province. They inferred predominantly N-S trending zone of
510
extensional deformation in the western Deccan region which they matched with faults
511
interpreted from seismic data to postulate strike-slip rifting between India and Seychelles.
512
Recent studies by Torsvik et al. (2013) and Ashwal et al. (2013) proposed the presence of
513
micro-continental fragments between the paleo-continental configuration of India-Madagascar.
514
Based on a combined geophysical-geochronological approach, Torsvik et al. (2013) concluded
515
that Mauritius and parts of Saya de Malha, Nazareth and Chagados-Carajos Banks in the
22
516
Southern Mascarene Plateau, and Laccadive and Chagos from the conjugate Indian margin, are
517
the fragments of a Paleo-Proterozoic microcontinent called ‘Mauritia’, existed in between
518
southern India and Madagascar. However, the whole interpretation was heavily depending on
519
the rare detrital U-Pb zircons separated from the basaltic beach sands of Mauritius. The results
520
of the paper argued on a Proterozoic zircon source, possibly from a micro-continent under the
521
basaltic cover, but the statistical population of datasets is considerably poor (a total of 8 analysis)
522
for provenance analysis and the whole presented result consist only one concordant age analysis
523
at ca. ~790 Ma. Hence the available results are not conclusive to establish the presence of
524
continental fragments proposed to be separated either from African plate, or from the Indian
525
Plate.
526
microcontinent or continental fragments or blocks within the plume trails between India and
527
Madagascar, but a detailed geochronology-geology based study with precise and statistically
528
convincing dataset can only solve this puzzle. The present study is highly oriented on the
529
lithospheric deformations within the rifted margins of Madagascar and southern India, which
530
however not considering the existence of microcontinent, still attained a unique fit configuration
531
for India and Madagascar. Therefore, the present plate reconstruction has not even considered
532
the Seychelles microcontinent since the Late Cretaceous plate reconstructions (Torsvik et al.,
533
2000; Gibbons et al., 2013) place this microcontinent far to the north of the India-Madagascar
534
paleo-welding zone. Furthermore, as the present plate reconstruction is based on the rift-related
535
lithospheric deformations estimated from the Te calculations along the two passive margins, the
536
fit of continents represents a period prior to and during the early phase of India-Madagascar
537
separation (i.e., during the formation of continental shelf margins) in the Late Cretaceous at ca.
538
88-90 Ma.
Conversely, we are not completely discarding the possibilities on the presence of
23
539
The originality of the present research relies on the fact that for the first time the study
540
maps the spatial variation of elastic thickness and the Moho undulation in the passive margins of
541
India and Madagascar that brings together a profound fit of continental margins. While the
542
previous geophysical approaches (Todal and Edholm, 1998; Minshull et al., 2008; Yatheesh et
543
al., 2009; Torsvik et al., 2013) used wide-angle seismic data and/or gravity anomaly data that
544
mainly explained geometry of the lithosphere beneath major structural features and its
545
correlations with rift related phenomena in the present continental margins, the present study
546
essentially estimate lithospheric deformations and evaluate its affinity to rift-related processes by
547
integrating all available geological and geophysical datasets.
548
5.4 In the Light of Present Plate Reconstruction
549
Fig. 12 present plate tectonic reconstruction of India-Madagascar paleo-fit obtained from
550
the present study, and shows how the shear/ suture zones and the various lithological units across
551
the two continents correlate in the ‘fit position’ deduced from the Te correlation (Fig. 10). It is
552
obvious that the present plate reconstruction (Fig. 12) precisely connect the key shear/suture
553
zones as well as the lithological units across India and Madagascar. In spite of alternative
554
controversial interpretations (e.g. Tucker et al. 2011; Brandt et al., 2014), the Betsimisaraka
555
suture zone of northeastern Madagascar (Kröner et al., 2000; Collins and Windley, 2002; Collins
556
et al., 2006) was correlated with the recently proposed Kumta suture zone of southern India by
557
Ishwar-Kumar et al. (2013) and Collins et al. (2007) correlated the Betsimisaraka suture with the
558
Palghat-Cauvery shear zone of southern India. Recently, based on microscopic and mesoscopic
559
structures and Th-U-Pb monazite ages, Rekha et al. (2013) proposed a correlation between
560
crustal blocks in western India and NE Madagascar. Although their correlation is in agreement
561
with lithological units and ages, the reconstructed position between India and Madagascar is
562
inconsistent with the present results. From the structural, geological and geochronological data
24
563
Ishwar-Kumar et al. (2013) pointed out that the ca. 1300 Ma Kumta suture in western India
564
separates the ca. 3200 Ma Karwar block in the west from the ca. 2570 Ma Dharwar block in the
565
east. Ishwar-Kumar et al. (2013) suggested a possible extension of the Betsimisaraka suture in
566
Madagascar, the Kumta suture, which re-enters into southern India as the ca. 900 Ma Coorg
567
suture (also named the Mercara suture by Santosh et al., (2014). The geophysically-based
568
correlations between India and Madagascar proposed in the present study are in good agreement
569
with the correlations of Ishwar-Kumar et al. (2013) (Fig. 12, inset map), even though they differ
570
in their definitions of the closeness of fit as a result of their different treatment of the bathymetric
571
data. The difference is that Ishwar-Kumar et al. (2013) revealed the fit of continents prior to the
572
India-Madagascar separation, while the present study used the rift-related constraints to correlate
573
them. However, considerating the fact that the Moho depth and bathymetry data have changed
574
the degree of closeness has not made any significant difference in the correlations of structural
575
and lithological units between India and Madagascar, which remain the same in both studies
576
(Fig. 12).
577
The Archean gneisses of the Antongil and Masora cratons (3320-3150 Ma) (Tucker et al.,
578
2011) of Madagascar correlate well with the tonalitic gneisses of the Dharwar craton (~2500-
579
3200 Ma) (Beckinsale et al., 1980; Nutman et al., 1992; Peucat et al., 1995; Windley and
580
Razakamanana, 1996; Jayananda et al., 2000; Ishwar-Kumar et al., 2013) of southern India. In
581
the Dharwar craton the age of the basement gneisses and older greenstone belts is about 3200
582
Ma, and the younger gneisses and younger greenstone belts have an age of 2500 Ma. In
583
Madagascar the Antananarivo Block contains 2700-2500 Ma gneisses intruded by younger
584
granites (820-520 Ma) (Kröner et al., 2000). The Betsimisaraka suture, which has a wide range
585
of detrital ages (2950-740 Ma) (Kröner et al., 2000; Tucker et al., 2011), is correlated with
586
Kumta suture zone (3280-2993 Ma detrital ages) that has a 1385-1326 Ma K-Ar biotite
25
587
metamorphic age (Ishwar-Kumar et al., 2013) and the Mercara suture zone (McSZ) in SW India
588
that has a K-Ar biotite age of ca. 933 Ma (Ishwar-Kumar et al., 2013). The southern part of the
589
Palghat-Cauvery shear zone in southern India mainly consists of high-grade charnockites and
590
metasedimentary belts that can be compared with granulite facies metasedimenatry rocks of
591
Madagascar south of the Ranotsara shear zone (Collins et al., 2012).
592
6. Conclusions
593
This study contributes new data on the spatial variation of effective elastic thickness over
594
the western continental margin of India and the eastern continental margin of Madagascar,
595
obtained from flexure inversion analysis (convolution method), which reveals a possible
596
correlation of their conjugate margins. A high resolution database of the undulations on the
597
Moho/Crust-Mantle Interface, derived from gravity and flexure inversion analyses, and their
598
regional correlations with the Te variations, adds a new perspective to the present interpretations.
599
We demonstrate that elastic thickness is a useful diagnostic tool, which can be corroborated and
600
integrated with crustal geometry, bathymetry, structure, lithology and geochronological datasets
601
to evaluate the evolution and deformation of the lithosphere. The following conclusions can be
602
drawn from the present study.
603
1. The Te and Moho results exhibit significant variations of elastic thickness over the
604
continental-oceanic margins of India and Madagascar, which reveal important insights into
605
the evolution and deformation of different lithological units. The Moho data demonstrate
606
geotectonic segmentation with a transition from thick crust (>35 km) beneath the continents
607
to thin crust (8-15 km) beneath the oceans with a transitional crust (thickness~25 km)
608
beneath the continental shelves. We observe that the cold and stable segments of the
609
continental lithosphere exhibit higher Te values, while thermally or mechanically
26
610
rejuvenated lithospheric segments are mechanically weak. Most of the oceanic parts of the
611
Indian and African plates included in our study generally exhibit thinned crust and low-Te
612
ranges, whereas the hotspot fossil ridges exhibit variable Te that correlate with their
613
emplacement setting.
614
2. We conclude that the significantly low-Te zones along the western continental margin of
615
India and eastern continental margin of Madagascar represent their paleo-rift inception
616
points, affected by significant lithospheric extension due to rifting combined with the effect
617
of Marion hotspot volcanism. The low-Te zone along the western margin of India is
618
attributed to the presence of a failed rift and of the volcanism in the stretched continental
619
lithosphere, which is manifested by coincident linear structural features along the shelf
620
basement. The correlation of this zone of low-Te using a 1000 m isobath enable a best
621
possible fit of India against Madagascar. This is confirmed by an excellent geometrical fit
622
between the bathymetry and the Moho configuration of both shelf margins. The derived
623
paleo-fit of the continents is consistent with and supported by geological constraints such as
624
the matching of shear zones, lithologies and geochronological belts.
625
3. Based on the present results, we assume a close-fit (inset map of Fig. 12) between India and
626
Madagascar before rifting (Lawver et al., 1997). The rift-related stretching and subsequent
627
thinning of this congruent lithosphere led to the formation of a continuous shelf common to,
628
and between, India and Madagascar. The increasing degrees of partial melting demonstrated
629
from north to south by Peng and Mahoney (1995) reveal that the Marion plume activity off
630
of southern Madagascar during the time of rifting (Torsvik et al., 2000) may have had an
631
integrated effect on the reduced mechanical strength of the lithosphere beneath both
632
continental margins. Ultimately, physical separation of these continents possibly resulted in
633
two individual shelves on either side of a mid-oceanic ridge (Fig. 12). The separation of
27
634
India by rifting and then drifting from the relatively stationary continental mass of
635
Madagascar might be a main reason for the persistence of their asymmetric conjugate
636
margins. The failed rift and volcanism of the stretched lithosphere possibly created the
637
regional-scale features such as ridges, grabens and faults, and sub-crustal loads such as
638
magmatic underplating along the congruent margin of India, where their flexural responses
639
are frozen into the lithosphere and resulted in a low-Te anomaly.
640
4. The present study conclude that the passive margins will retain their original structural and
641
mechanical behavior since rifting, if they were not influenced by any later major tectonic
642
processes, and hence the effective elastic thickness can be used in such tectonic setting as a
643
powerful proxy to examine the conjugate nature of passive margins.
644
Acknowledgements
645
R.T. Ratheesh Kumar gratefully acknowledges an IISc-Research Associate Fellowship. We
646
utilized the laboratory facilities developed through Ministry of Earth Sciences, Government of
647
India project MoES/ATMOS/PP-IX/09. This study is a contribution to ISRO-IISc Space
648
Technology Cell project ISTC/CEAS/SJK/291.
649
References
650
Anderson, O.B., Knudsen, P., Berry, P., Freeman, J., Pavlis, N., Kenyon, S., 2008. The DNSC08
651
Ocean wide altimetry derived gravity field, Presented EGU (2008), session G1, General
652
assembly, Vienna, Austria, April 14 -18.
653
654
Anderson, O.B., Knudsen, P.O., 1998. Global marine gravity field from ERS-1 and GEOSAT
geodetic mission altimetry. Journal of Geophysical Research 103, 8129-8137.
28
655
Anil Kumar, Pande, K., Venkatesan, T.R., Bhaskar Rao, Y.J., 2001. The Karnataka Late
656
Cretaceous dykes as products of the Marion hotspot at the Madagascar-India break-up event:
657
evidence from
658
28, 2715-2718.
40
Ar/39Ar geochronology and geochemistry. Geophysical Research Letters
659
Ashwal, L.D., Solanki, A.M., Pandit, M.K., Corfu, F., Hendriks, B.W.H.,. Burke, K., Torsvik,
660
T.H., 2013. Geochronology and geochemistry of Neoproterozoic Mt. Abu granitoids, NW
661
India: Regional correlation and implications for Rodinia paleogeography. Precambrian
662
Research 236, 265–281.
663
664
665
666
Audet, P., Mareschal, J.C., 2004. Variation in elastic thickness in the Canadian Shield. Earth and
Planetary Science Letters 226, 17-31.
Bassin, C., Laske, G., Masters, G., 2000. The current limits of resolution for surface wave
tomography in North America. EOS Trans AGU 81, F897.
667
Bauer, W., Walsh, G.J., De Waele, B., Thomas, R.J., Horstwood, M.S.A., Bracciali, L.,
668
Schofield, D.I., Wollenberg, U., Lidke, D.J., Rasaona, I.T., Rabarimanana, M.H., 2011.
669
Cover sequences at the northern margin of the Antongil Craton, NE Madagascar.
670
Precambrian Research 189, 292-312.
671
Beane, J.E., Turner C.A., Hooper, P.R., Subbarao, K.V., Walsh, J.N., 1986. Stratigraphy,
672
composition and form of the Deccan basalts, Western Ghats, India. Bulletin Volcanologique
673
48, 61-83.
674
675
Beckinsale, R.D., Drury, S.A., Holt, R.W., 1980. 3360 M.Y. old gneisses from south Indian
craton. Nature 283, 469-470.
29
676
677
678
679
Biswas, S.K. 1987. Regional tectonic framework, structure and evolution of the western
marginal basins of India. Tectonophysics 135, 307-327.
Biswas, S.K., 1988. Structure of the western continental margin of India and related igneous
activity. Geological Society of India Memoir 10, 371-390.
680
Biswas, S.K., 1999. A review on the evolution of rift basins in India during Gondwana with
681
special reference to Western India basin and their hydrocarbon prospects. PINSA 65 (3),
682
261-283.
683
Braitenberg, C., Ebbing, J., Gotze, H.J., 2002. Inverse modelling of elastic thickness by
684
convolution method: The Eastern Alps as a case example. Earth and Planetary Science
685
Letters 202, 387-404.
686
Braitenberg, C., Pettenati, F., Zadro, M., 1997. Spectral and classical methods in the evaluation
687
of Moho undulations from gravity data: The NE Italian Alps and isostasy. Journal of
688
Geodynamics 23, 5 –22.
689
690
691
692
Braitenberg, C., Wang Y., Fang, J., Hsu, H.T., 2003. Spatial variations of flexure parameters
over the Tibet-Quinghai plateau. Earth and Planetary Science Letters 205, 211–224.
Braitenberg, C., Wienecke, S., Wang, Y., 2006. Basement structures from satellite-derived
gravity field: South China Sea ridge. Journal of Geophysical Research 111, B05407.
693
Brandt, S., Raith, M.M., Schenk, V., Sengupta, P., Srikantappa, C., Gerdes, A., 2014. Crustal
694
evolution of the Southern Granulite Terrane, south India: New geochronological and
695
geochemical data for felsic orthogneissses and granites. Precambrian Research, in press.
696
Braun, I., Kriegsman, L.M., 2003. Proterozoic crustal evolution of southernmost India and Sri
697
Lanka in Proterozoic East Gondwana: Supercontinent Assembly and Breakup. In: Yoshida,
30
698
M., Windley, B.F., Dasgupta, S., (Eds.), Geological Society of London, Special
699
Publications, 206, 169-202.
700
701
702
703
704
705
Braun, J., Beaumont, C., 1989. A physical explanation of the relationship between flankuplifts
and the breakup unconformity at rifted continental margins. Geology 17, 760–764.
Burov, E.B., Diament, M., 1995. The effective elastic thickness of continental lithosphere: what
does it really mean?. Journal of Geophysical Research 100, 3905-3927.
Chand, S., Radhakrishna, M., Subrahmanyam, C., 2001. India-East Antarctica conjugate
margins: Gravity and isostasy. Earth and Planetary Science Letters 185, 225-237.
706
Chand, S., Subrahmanyam, C., 2003. Rifting between India and Madagascar mechanism and
707
isostasy. Earth and Planetary Science Letters 210, 317-332.
708
Chaubey, A.K., Gopala Rao, D., Srinivas, K., Ramprasad, T., Ramana, M.V., Subrahmanyam,
709
V., 2002. Analyses of multichannel seismic reflection, gravity and magnetic data along a
710
regional profile across the central-western continental margin of India. Marine Geology 182,
711
303–323.
712
Chaubey, A.K., Srinivas, K., Ashalatha, B., Gopala Rao, D., 2008. Isostatic response of the
713
Laccadive Ridge from admittance analysis of gravity and bathymetry data. Journal of
714
Geodynamics, 46, 10-20.
715
716
717
718
Collins, A.S., 2006. Madagascar and the amalgamation of Central Gondwana. Gondwana
Research 9, 3-16.
Collins, A.S., Windley, B.F., 2002. Tectonic evolution of central and northern Madagascar and
its place in the final assembly of Gondwana. Journal of Geology 110, 325-339.
31
719
720
Collins, A.S., Pisarevsky, S.A., 2005. Amalgamating eastern Gondwana: the evolution of the
circum-Indian orogens. Earth Science Reviews 71, 229-270.
721
Collins, A.S., Clark, C., Sajeev, K., Santosh, M., Kelsey, D.E., Hand, M., 2007. Passage through
722
India: the Mozambique ocean suture, high-pressure granulites and the Palghat-Cauvery shear
723
zone system. Terra Nova 19, 141-147.
724
Collins, A.S., Kinny, P.D., Razakamanana, T., 2012. Depositional age, provenance and
725
metamorphic age of metasedimentary rocks from southern Madagascar. Gondwana
726
Research 21, 353-361.
727
728
Courtillot, V., Davaille, A., Besse, J., Stock J., 2003. Three distinct types of hotspots in the
Earth’s mantle. Earth and Planetary Science Letters 205, 295-308.
729
Cucciniello, C., Conrad, J., Grifa, C., Melluso, L., Mercurio, M., Morra, V.,. Tucker, R.D.,
730
Vincent, M., 2011. Petrology and geochemistry of Cretaceous mafic and silicic dykes and
731
spatially associated lavas in central-eastern coastal Madagascar. In: Srivastava, R.K (ed.),
732
Dyke Swarms: Keys for Geodynamic Interpretation, 345-375.
733
Cucciniello, C., Melluso, L., Morra, V., 2011. New
40
Ar-39Ar ages and petrogenesis of the
734
Massif d'Ambre volcano, northern Madagascar. Geological Society of America Special
735
Papers 478, 257-281.
736
Daly E., Brown, C., Stark, C.P., Ebinger, C.J., 2004. Wavelet and multitaper coherence methods
737
for assessing the elastic thickness of the Irish Atlantic margin. Geophysical Journal
738
International 159, 445-459.
32
739
Daly, E., Brown, C., Stark, C.P., Ebinger, C.J., 2004. Wavelet and multitaper coherence methods
740
for assessing the elastic thickness of the Irish Atlantic margin. Geophysical Journal
741
International 159(2), 445–459.
742
743
744
de Wit, M.J., 2003. Madagascar: heads it’s a continent, tails it’s an island. Annual Reviews Earth
Planetary Sciences 31, 3-48.
Divins, D.L., 2003. Total Sediment Thickness of the World's Oceans & Marginal Seas, NOAA
745
National
746
CO, http://www.ngdc.noaa.gov/mgg/sedthick/sedthick.html
Geophysical
Data
Center,
Boulder,
747
Emmel, B., Jacobs, J., Kastowski, M., Graser, G., 2006. Phanerozoic upper crustal tectono-
748
thermal development of basement rocks from of central Madagascar: an integrated structural
749
and fission-track study. Tectonophysics 412, 61-86.
750
751
Fourno, J-P., Roussel, J., 1994. Imaging of the Moho depth in Madagascar through the inversion
of gravity data: geodynamic implications. Terra Nova 6, 512-519.
752
Ghosh, J.G., de Wit, M.J., Zartman, R.E., 2004. Age and tectonic evolution of Neoproterozoic
753
ductile shear zones in the Southern Granulite terrane of India, with implications for
754
Gondwana studies. Tectonics 23, TC3006, doi: 1029/2002TC001444.
755
Gibbons, A.D., Whittaker, J.M. Muller, R.D., 2013. The breakup of East Gondwana:
756
Assimilating constraints from Cretaceous ocean basins around India into a best-fit tectonic
757
model. Journal of Geophysical Research 118, 808–822.
33
758
Ishwar-Kumar, C., Windley B.F., Horie, K., Kato, T., Hokada, T., Itaya, T., Yagi, K., Gouzu, C.,
759
Sajeev, K., 2013. A Rodinian suture in western India: New insights on India-Madagascar
760
correlations. Precambrian Research 236, 227-251.
761
Jay, A.E., Widdowson, M., 2008. Stratigraphy, structure and volcanology of the SE Deccan
762
continental flood basalt province: implications for eruptive extent and volumes. Journal of
763
Geological Society of London 165, 177-188.
764
Jayananda, M., Moyen, J.F., Martin, H., Peucat, J.-J., Auvray, B., Mahabaleswar, B., 2000. Late
765
Archean (2550-2520 Ma) juvenile magmatism in the Eastern Dharwar craton, southern
766
India: constraints from geochronology, Nd-Sr isotopes and whole rock geochemistry.
767
Precambrian Research 99, 225-254.
768
Jiménez-Díaza, A., Ruiza, J., Pérez-Gussinyéc, M., Kirby, J.F., Álvarez-Gómez J.A., Tejero, R.,
769
Capote, R., 2014. Spatial variations of effective elastic thickness of the lithosphere in
770
Central America and surrounding regions. Earth and Planetary Science Letters 391, 55–66.
771
772
773
774
Karner, G.D., Watts, A.B., 1982. On isostasy at Atlantic-type continental margins. Journal of
Geophysical Research 87, 2923-2948.
Katz, M.B., Premoli, C., 1979. India and Madagascar in Gondwana land based on matching
Precambrian lineaments. Nature 279, 312-315.
775
Kirby,J.F., Swain,C.J., 2008. An accuracy assessment of the fan wavelet coherence method for
776
elastic cthickness estimation. Geochemistry Geophysics Geosystems 9(3), Q03022.
777
Correction: Geochemistry Geophysics Geosystems 9(5) (2008) Q05021.
34
778
Kröner, A., Hegner, E., Collins, A.S., Windley, B.F., Brewer, T.S., Razakamanana, T., Pidgeon,
779
R.T., 2000. Age and magmatic history of the Antananarivo block, Central Madagascar, as
780
derived from zircon geochronology and Nd isotopic systematics. American Journal of
781
Science 300, 251-288.
782
Kusky, T.M., Toraman, E., Raharimahefa, T., Rasoazanamparany, C., 2010. Active tectonics of
783
the Alaotra-Ankay graben system, Madagascar: possible extension of Somalian-African
784
diffuse plate boundary. Gondwana Research 18, 274-294.
785
Lawver, L.A., Gahagan, L.M., Dalziel, I.W.D., 1997. A tight fit-early Mesozoic Gondwana, a
786
plate reconstruction perspective. National Institute of Polar Research (Tokyo), Memoir 53,
787
214-229.
788
789
Lowry, A.R., Smith, R.B., 1995. Strength and rheology of the western US Cordillera. Journal
of Geophysical Research 100, 17947-17963.
790
Mahoney, J.J., Saunders, A.D., Storey, M., Randriamanantenasoa, A., 2008. Geochemistry of the
791
Volcan de l’Androy basalt-rhyolite complex, Madagascar Cretaceous igneous province.
792
Journal of Petrology 49, 1069-1096.
793
794
Marks, K.M., Tikku, A.A., 2001. Cretaceous reconstructions of East Antarctica, Africa and
Madagascar. Earth and Planetary Science Letters 186, 479-495.
795
McKenzie, D., Fairhead, J.D., 1997. Estimates of the effective Te of the continental
796
lithosphere from Bouguer and free air gravity anomalies. Journal of Geophysical
797
Research 102, 27523-27552.
35
798
Melluso, L., Morra, V., Brotsu, P., D’antonio, M., Bennio, L., 2002. Petrogenesis of the Late
799
Cretaceous tholeiitic magmatism in the passive margins of northeastern Madagascar.
800
Geological Society of America Special Papers 362, 81-85.
801
Melluso, L., Morra, V., Brotzu, P., Mahoney, J.J., 2001. The Cretaceous igneous Province of
802
Madagascar: geochemistry and petrogenesis of lavas and dykes from the central-western
803
sector. Journal of Petrology 42, 1249-1278.
804
Melluso, L., Morra, V., Brotzu, P., Razafiniparany, A., Ratrimo, V., Razafimahatratra D., 1997.
805
Geochemistry and Sr-isotopic composition of the late Cretaceous flood basalt sequence of
806
northern Madagascar: Petrogenetic and geodynamic implications. Journal of African Earth
807
Sciences, 34, 371-390.
808
Melluso, L., Morra, V., Brotzu, P., Tommasini, S., Renna, M.R., Duncan, R.A., Franciosi, L.,
809
d’Amelio, F., 2005. Geochronology and petrogenesis of the Cretaceous Antampombato-
810
Ambatovy complex and associated dyke swarm, Madagascar. Journal of Petrology 46, 1963-
811
1996.
812
813
National Oceanic and Atmospheric Administration, General Bathymetric Chart of Oceans, 2003.
http://www.ngdc.noaa.gov/mgg/gebco/grid.
814
Nutman, A.P., Chadwick, B., Ramakrishnan, M., Viswanatha, M.N. 1992. SHRIMP U-Pb ages
815
of detrital zircon in Sargur supracrustal rocks in western Karnataka, southern India. Journal
816
of the Geological Society of India 39, 367-374.
817
818
O’Neill, C., Müller, D., Steinberger, B., 2003. Geodynamic implications of moving Indian
Ocean hotspots. Earth and Planetary Science Letters 205, 151-168.
36
819
Pande, K., Sheth, H.C., Bhutani, R., 2001.
40
Ar-39Ar age of the St. Mary’s Islands volcanics,
820
southern India: record of India-Madagascar break-up on the Indian subcontinent. Earth and
821
Planetary Science Letters 193, 39-46.
822
Peng, Z.X., Mahoney, J.J., 1995. Drill hole lavas from the northwestern Deccan Traps, and the
823
evolution of Réunion hotspot mantle. Earth and Planetary Science Letters 134, 169-185.
824
Pérez-Gussinyé, M., Lowry, A.R., Watts, A.B., 2007. Effective elastic thickness of South
825
America and its implications for intracontinental deformation. Geochemistry Geophysics
826
Geosystems 8(5), Q05009, doi:10.1029/2006GC001511.
827
Pérez-Gussinyé, M., Lowry, A.R., Watts,A.B., Velicogna, I., 2004. On the recovery of the
828
effective elastic thickness using spectral methods: examples from synthetic data and from
829
the Fennoscandian Shield. Journal of Geophysical Research 109.
830
831
Pérez-Gussinyé, M., Watts, A.B., 2005. The long term strength of Europe and its implications
for plate-forming processes. Nature 436, 381-384.
832
Peucat, J.-J., Bouhallier, H., Fanning, C.M., Jayananda, M., 1995. Age of Holenarsipur schist
833
belt, relationships with the surrounding gneisses (Karnataka, South India). Journal of
834
Geology 103, 701-710.
835
Powell, C., McA., Li, Z.X., Muller, D., Watkeys, M.K., 1997. The India-Madagascar transform
836
boundary: implications for Jurassic and Cretaceous Gondwanaland breakup. In: Proceedings
837
of the UNESCO-IUGS-IGCP-348/368 International Field Workshop on the Proterozoic
838
Geology of Madagascar (Cox, R., Ashwal, L.D., Eds). Miscellaneous Publication,
839
Gondwana Research Group, Osaka City University 5, 73-74.
37
840
Radha Krishna, M., Verma, R.K., Arts, K.P., 2002. Lithospheric structure below the eastern
841
Arabian Sea and adjoining West Coast of India based on integrated analysis of gravity and
842
seismic data. Marine Geophysical Researches 23: 25–42.
843
Ratheesh Kumar, R.T., Tanmay, K.M., Kandpal, S.C., Sengupta, D., Nair, R.R., 2011. Elastic
844
thickness estimates at north-east passive margin of North America and its implications.
845
Journal of Earth System Science 120 (3), 1–12.
846
Ratheesh Kumar, R.T., Windley, B.F., 2013. Spatial variations of effective elastic thickness
847
over the Ninetyeast Ridge and implications for its structure and tectonic evolution.
848
Tectonophysics 608, 847-856.
849
850
851
852
Ratheesh-Kumar, R.T., Windley, B.F., Sajeev, K., 2014. Tectonic inheritance of the Indian
Shield: New insights from its elastic thickness structure. Tectonophysics 615–616, 40–52.
Reeves, C., de Wit, M., 2000. Making ends meet in Gondwana: retracing the transforms of the
Indian Ocean and reconnecting continental shear zones. Terra Nova 12, 272-280.
853
Rekha, S., Viswanath, T.A., Bhattacharya, A., Prabhakar, N., 2013. Meso/Neoarchean crustal
854
domains along the north Konkan coast, western India: The Western Dharwar Craton and the
855
Antongil-Masora Block (NE Madagascar) connection. Precambrian Research 233, 316-336.
856
Santosh, M., Yang, Q.Y., Shaji, E., Tsunogae, M., Ram Mohan, M., Satyanarayanan, M., 2014.
857
An exotic Mesoarchean microcontinent: the Coorg, Block, southern India. Gondwana
858
Research, in press.
859
Seward, D., Grujic, D., Schreurs, G., 1999. Exhumation history of the East Madagascar
860
continental margin: inferences from apatite fission track analysis. Journal of African Earth
861
Sciences 27, 75-78.
38
862
Sheena, V.D., Radhakishna, M., Subrahmanyam, C., 2007. Estimates of effective elastic
863
thickness along the southwest continental margin of India using coherence analysis of
864
gravity and bathymetry data – geodynamic implications. Journal of Geological Society of
865
India 70, 475–487.
866
Sheena, V.D., Radhakrishna, M., Chand, S., Subrahmanyam, C., 2012. Gravity anomalies,
867
crustal structure and rift tectonics at the Konkan and Kerala basins, western continental
868
margin of India. Journal of Earth System Sciences 121(3), 813–822.
869
Simons, F.J., R. Van Der Hilst, R.D., Zuber, M.T., 2003. Spatiospectral localization of isostatic
870
coherence anisotropy in Australia and its relation to Seismic anisotropy: Implication for
871
lithospheric deformation. Journal of Geophysical Research 108, ETG 1-8.
872
Simons, F.J., Van Der Hilst, R.D., 2002. Age-dependent seismic thickness and mechanical
873
strength of the Australian lithosphere. Geophysical Research Letters 29(11), 1529, doi:
874
10.1029/2002GL014962.
875
876
Stern, T., Brink, U.S.T., 1989. Flexural uplift of the Transantarctic mountains. Journal of
Geophysical Research. 94, 10315-10330.
877
Storetvedt, K.M., Mitchell, J.G., Abranches, M.C., Maaloe, S., Robin, G., 1992. The coast-
878
parallel dolerite dykes of East Madagascar: age of intrusion of intrusion, remagnetization
879
and tectonic aspects. Journal of African Earth Sciences 15, 237-249.
880
Storey, M., Mahoney, J.J., Saunders, A.D., 1997. Cretaceous basalts in Madagascar and the
881
transition between plume and continental lithosphere mantle sources. In: Mahoney J.J.,
882
Coffin M.F. (Eds.). Large Igneous Provinces. American Geophysical Union, Monograph
883
100, 95-122.
39
884
Storey, M., Mahoney, J.J., Saunders, A.D., Duncan, R.A., Kelley, S.P., Coffin, M.F., 1995.
885
Timing of hotspot-related volcanism and the breakup of Madagascar and India. Science 267,
886
852-855.
887
888
889
890
891
892
Subrahmanya, K.R., 1998. Tectono-magmatic evolution of the west coast of India. Gondwana
Research 1, 319-327.
Subrahmanyam, C., Chand, S., 2006. Evolution of the passive continental margins of India—a
geophysical appraisal. Gondwana Research 10, 167-178.
Sutra, E., Manatschal, G., 2012. How does the continental crust thin in a hyper-extended rifted
margin? Insights from the Iberia margin. Geology 40, 139-142.
893
Tiwari, V.M., Grevemeyer, I., Singh, B., Phipps Morgan, J., 2007. Variation of effective elastic
894
thickness and melt production along the Deccan–Reunion hotspot track. Earth and Planetary
895
Science Letters 264, 9-21.
896
Tiwari, V.M., Mishra, D.C., 1999. Estimation of effective elastic thickness from gravity and
897
topography data under the Deccan Volcanic province, India. Earth and Planetary Science
898
Letters 171(2), 289–299.
899
Torsvik, T.H., Amundsen, H., Hartz, E.H., Corfu, F., Kusznir, N., Gaina, C., Doubrovine, P.V.,
900
Steinberger, B., Ashwal, L.D., Jamtveit, B., 2013. A Precambrian microcontinent in the
901
Indian Ocean. Nature Geoscience 6, 223-227.
902
Torsvik, T.H., Tucker, R.D., Ashwal, L.D., Carter, L.M., Jamtveit, B., Vidyadharan, K.T.,
903
Venkataramana, P., 2000. Late Cretaceous India-Madagascar fit and timing of breakup-
904
related magmatism. Terra Nova 12, 220-225.
40
905
Torsvik, T.H., Tucker, R.D., Ashwal, L.D., Eide, E.A., Rakotolosofo, N.A., de Wit, M.J., 1998.
906
Late Cretaceous magmatism in Madagascar. Paleomagnetic evidence for a stationary
907
Marion hotspot. Earth and Planetary Science Letters 164, 221-232.
908
Trivedi, D., Tanmay, K.M., Sengupta, D., Nair, R.R., 2012. Reappraisal of effective elastic
909
thickness in the southwest Indian Ocean, and its possible implications. Annals of
910
Geophysics 55, 2, doi: 10.4401/ag-5171.
911
Tucker, R.D., Roig, J.Y., Delor, C., Amerlin, Y., Goncalves, P., Rabarimanana, M.H., Ralison,
912
A.V., Belcvher, R.W., 2011. Neoproterozoic extension in the Greater Dharwar Craton: a
913
reevaluation of the “Betsimisaraka suture’ in Madagascar. Canadian Journal of Earth
914
Sciences 48, 389-417.
915
Valsangkar, A.B., Radhakrishnamurthy, C., Subbarao, K.V., Beckinsale, R.D., 1981.
916
Paleomagnetism and potassium-argon age studies of acid igneous rocks from St. Mary
917
Islands. Geological Society of India 3, 265-276.
918
919
920
921
922
923
Watts, A.B., 1992. The effective Te of the lithosphere and the evolution of foreland basins.
Basin Research 4,169-178.
Watts, A.B., 2001. Isostasy and Flexure of the Lithosphere. Cambridge University Press,
Cambridge, UK.
Watts, A.B., Burov, E., 2003. Lithospheric strength and its relationship to the elastic and
seismogenic layer thickness. Earth and Planetary Science Letters 213, 113–131.
924
Watts, A.B., Torné, M., 1992. Subsidence history, crustal structure, and thermal evolution of the
925
Valencia Trough: a young extensional basin in the western Mediterranean. Journal of
926
Geophysical Research B13, 20,021-20,041.
41
927
Widdowson, M., 1997. Tertiary palaeosurfaces of the SW Deccan western India: Implications
928
for passive margin uplift; In: Widdowson, M. (ed). Palaeosurfaces: Recognition,
929
reconstruction and palaeoenvironmental interpretation: Geological Society of London,
930
Special Publications 120, 221–248.
931
Windley, B.F., Razakamanana, T., 1996. The Madagascar-India connection in a Gondwana
932
framework. In: Santosh, M., Yoshida, M. (eds) The Archaean and Proterozoic Terrains in
933
Southern India within East Gondwana. Gondwana Research Group Memoir 3, 25-37.
934
935
936
937
938
939
Wyer, P., Watts, A.B., 2006. Gravity anomalies and segmentation at the East Coast, USA
continental margin. Geophysical Journal International 166, 1015–1038.
Yin, A., Harrison, T.M., 2000. Geological evolution of the Himalayan-Tibetan orogen. Annual
Review of Earth Planetary Sciences 28, 211-280.
Zutshi, P.L., Jain, M.M., Shah, T., Malhotra, S., 1995. Geology and hydrocarbon resources of
the Indian offshore region. Oil Asia 1-21.
940
941
942
Figure Captions:
943
Fig. 1: Tectonic setting of the western Indian Ocean superimposed on a GEBCO (1×1 minute
944
grid) bathymetry map. The windows (a) and (b) represent the selected areas over the western
945
continental margin of India and eastern continental margin of Madagascar respectively.
946
Acronyms: BB- Bombay Basin, KoB- Konkan Basin, KeB-Kerala Basin.
42
947
Fig. 2: (a) Bathymetry (b) Bouguer gravity anomaly (c) Sediment thickness (d) Gravity effect of
948
the sediments (e) Sediment-corrected bathymetry (basement depth) (f) Sediment-corrected
949
gravity of the western continental margin of India.
950
Fig. 3: (a) Bathymetry (b) Bouguer gravity anomaly (c) Sediment thickness (d) Gravity effect of
951
the sediments (e) Sediment-corrected bathymetry (basement depth) (f) Sediment-corrected
952
gravity of the eastern continental margin of Madagascar.
953
Fig. 4: Equivalent topography map of the conjugate continental margins and adjoining oceanic
954
terranes of India (a) and Madagascar (b) derived from sediment-corrected bathymetry.
955
Acronyms are given in Table 1.
956
Fig. 5: Effective elastic thickness of the western continental margin of India. Topography shaded
957
relief is superimposed. Red dotted line represents the Reunion hotspot track (after Torsvik et al.,
958
2013). Acronyms are given in Table 1.
959
Fig. 6: Effective elastic thickness of the eastern continental margin of Madagascar. Red dotted
960
line represents the Reunion hotspot track (after Torsvik et al., 2013). Acronyms are given in
961
Table 1.
962
Fig. 7: Moho undulations obtained from constrained gravity inversion and flexural inversion
963
techniques for the western continental margin of India (a and b) and eastern continental
964
margin of Madagascar (c and d) respectively. (e) and (f) show the Residual Moho (the
965
mismatch between gravity inversion-derived Moho and flexural inversion-derived Moho)
966
respectively for the WCMI and ECMM.
967
Fig. 8: Moho configuration of the WCMI (derived from flexure inversion method) super
968
imposed by the topography/bathymetry shaded relief. Acronyms are given in Table 1.
43
969
Fig. 9: Moho configuration of the ECMM (derived from flexure inversion method) super
970
imposed by the topography/bathymetry shaded relief. Acronyms are given in Table 1.
971
Fig. 10. A map showing the correlation between the elastic thickness maps of India and
972
Madagascar. The low elastic thickness regions (in blue, which are about 1600 km long) in the
973
eastern shelf of Madagascar and western shelf of India match very well. This shows the possible
974
region that was affected during the rifting and separation of India. The 1000 m isobath for the
975
eastern shelf of Madagascar and western shelf of India is used for defining the close fit.
976
Acronyms are given in Table 1.
977
Fig. 11: The geometrical fit between the Moho configurations of the shelf margins of India and
978
Madagascar, obtained at the same reconstructed positions of the plate margins based on the Te
979
correlations presented in Figure 10. Acronyms are given in Table 1.
980
Fig. 12: Plate tectonic reconstruction map of India-Madagascar paleo-fit deduced from the
981
elastic thickness (Figure 10), Moho and bathymetry (Figure 11) correlations, exhibit fit of
982
tectonic lineaments as well as age and lithology of tectonic provinces between the two
983
continents. Shear zones based on correlations of Collins and Windley, 2002; Ishwar-Kumar et
984
al., 2013. Paleo-coordinates after O’Neil, et al., 2003. The inset map shows a close-fit position of
985
both continental margins and the matching of the shear zones (modified after Ishwar-Kumar et
986
al., 2013). Acronyms are given in Table 1.
987
988
44