Tectonophysics 838 (2022) 229472

Contents lists available at ScienceDirect

Tectonophysics
journal homepage: www.elsevier.com/locate/tecto

Evolution of the Western Ghats: Constraints from receiver function imaging

and harmonic decomposition
Padma Rao B. a, *, Ravi Kumar M. b
a
National Centre for Earth Science Studies, Ministry of Earth Sciences, Govt. of India, Thiruvananthapuram, India
b
CSIR-National Geophysical Research Institute, Hyderabad, India

A R T I C L E I N F O A B S T R A C T

Keywords: The Western Ghats (WG), with an average elevation of ~1.2 km, is one of the greatest escarpments that spans
Receiver functions ~1500 km parallel to the west coast of India. It is a mosaic of disparate geological formations having distinct
Harmonic decomposition structural and physical characteristics. In this study, the lithospheric structure in terms of crustal thickness, sub-
Moho
Moho layer and deformation along and across the WG are investigated using receiver functions (RFs) at 30
LVL
Rifting
broadband seismological stations. Techniques like slant stacking, common conversion point imaging and har­
Western Ghats monic decomposition are applied to the RFs. Results reveal a large crustal thickness of ~45 km in the central part
of the WG, which decreases to ~39 km in the southern and ~ 37 km in the northern parts, in conformity with the
geology. The variation in crustal thickness beneath the WG could be due to the effect of hotspot impingement and
subsequent rifting mechanism. Further, the RFs reveal a strong sub-Moho low velocity layer (LVL) along the WG.
This observation gains support from a previous inference of a low-density zone in the uppermost mantle beneath
the west coast of India. Harmonic decomposition of the RFs reveals that this LVL is anisotropic, while the Moho is
dipping/ anisotropic in nature. The trend of the symmetry axis of anisotropy (of the LVL) varies from 58.2∘ to
134.4∘, almost perpendicular to the west coast of India. We postulate that the origin of this LVL can be linked to
different rifting episodes along the west coast of India, since LVLs are manifestations of thermal/chemical
anomalies. These results support the rifting model proposed for the evolution of WG.

1. Introduction Gunnell (2001), the causes for plateau uplift can be basically divided
into three categories: (i) density reduction due to an isostatic response,
The Western Ghats (WG), known as the Sahyadri mountain range, is mechanically or thermally, (ii) crustal buoyancy owing to an increase in
a coast parallel upliftment along the passive western continental margin lithospheric thickness and (iii) plastic necking because of stretching and
of India. It has an average elevation of ~1.2 km and extends ~1500 km asymmetrical denudation of the lithosphere on both sides of the scarp,
in the NNW-SSE direction from 8.1∘N to 21.2∘N. The WG is a mosaic of known as lithospheric flexure. In the Indian context, several studies
diverse geological formations having different physical and structural attempted to understand the evolution of WG. Based on geomorpho­
characteristics, namely the Southern Granulite Terrain (SGT) in the logical analysis, Radhakrishna (1993) suggested that the WG is formed
south, Western Dharwar Craton (WDC) in the middle and the Deccan due to Cenozoic rifting and down faulting. Later, a model of rift flank
Volcanic Province (DVP) in the north. Globally, studies aimed at deci­ upliftment proposed by Matmon et al. (2002), corroborated these find­
phering the evolution of passive continental margins resulted in many ings. Recently, a multistage model was invoked to explain the evolution
different mechanisms or models. The most popular among these are, of WG based on the geophysical and geological characteristics (Sinha-
crustal thinning due to rifting (McKenzie, 1978), magmatic underplating Roy, 2018). This model linked its evolution to two rifting episodes, i.e.,
(Cox, 1980; Devey and Lightfoot, 1986), transient thermal effects rifting of greater India from Madagascar at ~91 Ma due to the Marion
(Cochran, 1983), extensional and flexural unloading (Buck, 1986; hotspot and from Seychelles and Mascarene plateau at ~65 Ma due to
Weissel and Karner, 1989), denudational isostasy (Gilchrist and Sum­ the Reunion hotspot. Further, from palaeomagnetic analysis of deep drill
merfield, 1994; Widdowson and Cox, 1996; Gunnell, 2001) and flexural samples from a part of the Koyna region within the WG, Radhakrishna
response to denudation (Gunnell and Fleitout, 2000). According to et al. (2019) inferred that igneous underplating, extension/rifting and

* Corresponding author.
E-mail address: [email protected] (P.R. B.).

https://doi.org/10.1016/j.tecto.2022.229472
Received 12 July 2021; Received in revised form 9 June 2022; Accepted 29 June 2022
Available online 9 July 2022
0040-1951/© 2022 Elsevier B.V. All rights reserved.
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

thermal collapse and subsidence of the lithosphere are the key mecha­
nisms for the evolution of WG. Apart from these studies, different
geological and geophysical studies have been undertaken to decipher
the evolution of WG (e.g., Kaila et al., 1981; Tiwari and Mishra, 1999;
Gunnell and Fleitout, 2000; Tiwari et al., 2006; Campanile et al., 2008;
Arora et al., 2012; Rai et al., 2013; Gupta et al., 2018). However, a
conclusive explanation still remains ambiguous.
Geological processes like erosion, sedimentation and uplift affect the
Earth at different scales, from the surface to deep interior. Since these
alter the lithospheric properties, it is important to decipher the sub-
surface structure to understand the geodynamics and evolution of a re­
gion. Receiver function (RF) analysis is one of the best tools that is
extensively used to investigate the seismic structure of the crust and
upper mantle (Kind and Yuan, 2011). Thus, in the present study, we
decipher the crustal, sub-crustal discontinuities and anisotropy beneath
the WG through harmonic decomposition of RFs and interpret the results
in terms of possible causes for its evolution. Most of the receiver function
studies extract the isotropic, 1D structure utilizing the radial compo­
nents of the RFs obtained by deconvolution of the radial components
with the vertical components of teleseismic P- waveforms. However,
important, second order effects like heterogeneity and anisotropy of the
sub-surface layers, can be extracted from the analysis of radial RFs
together with the transverse RFs. In this study, we use both these com­
ponents to parameterize the crustal and uppermost mantle layering.

2. Data and methodology

2.1. Waveform data

In the present study, three component teleseismic waveform data

recorded at 30 broadband seismological stations sampling the WG along
and across the strike (Fig. 1), are used for receiver function analysis. The
data sets have been collated from various organizations like the National
Centre for Earth Science Studies (NCESS), National Centre for Seis­ Fig. 1. Location of broadband seismological stations (inverted triangles) whose
mology (NCS), Institute of Land and Disaster Management (ILDM) and data are used in the present study. Tectonic units: SGT – Southern Granulite
National Geophysical Research Institute (NGRI). The codes, geograph­ Terrain (Brown), WDC – Western Dharwar Craton (Yellowgreen), DVP – Deccan
ical locations and time period of operation of these stations are given in Volcanic Province (Light‑gold), CG – Closepet Granite (Slategray), EDC –
Table 1. Further, the waveforms of earthquakes having magnitude >5.0, Eastern Dharwar Craton (Gold), CB – Cuddapah Basin (Pink), EGMB – Eastern
in the epicentral distance range of 30∘ to 95∘, and signal to noise ratio Ghats Mobile Belt (Light-green) and WG – Western Ghats (Cross-pattern). Inset
at the bottom indicates an enlarged portion of the region containing the ILDM
(SNR) ≥ 2.5 are only selected for analysis. These criteria resulted in
stations. (For interpretation of the references to colour in this figure legend, the
38,889 three component waveforms from 4041 earthquakes recorded at
reader is referred to the web version of this article.)
30 broadband seismological stations.

slowness, the amount being proportional to depth. This moveout

2.2. Receiver functions
correction has been applied by considering a reference slowness of 6.4
s/∘, corresponding to an epicentral distance of 67∘ (Kind and Vinnik,
The lithospheric structure beneath a seismological station can be
1988), using the IASP91 model. Since the phases reflected from the free
determined using receiver functions (RFs). When the primary (P) wave
surface have opposite moveouts compared to the direct conversions, it is
is incident upon a seismic discontinuity in the crust or mantle, it gets
necessary to apply the corrections individually for each type of wave.
converted into an S wave (i.e., Ps). In order to isolate the converted
Also, the difference in moveout slopes provides a simple criterion to
energy, the ZNE components are rotated into a ray coordinate system
differentiate the converted and multiple phases. Later, the bad quality
(LQT) where, the L component is in the direction of the primary wave (i.
PRFs are discarded by visual quality checking, leading to 9252 good
e., P wave), the Q component is in the direction of the SV wave and the T
quality PRFs from 3020 earthquakes, at 30 broadband seismological
component is in the direction of the SH wave (Vinnik, 1977). Further,
stations (Fig. S1). The moveout corrected PRFs at the individual stations
the source equalized SV and SH receiver functions are obtained by
clearly reveal the Moho conversions (Ps) and the corresponding P and S
deconvolving the L component from the Q and T components respec­
multiples (i.e., PpPs and PpSs+PsPs). Further, the PRFs in the slowness
tively (Langston, 1979; Owens et al., 1984). This can simply be done by
range of 4.44 s/∘ to 8.84 s/∘ are sorted in 25 narrow slowness bins and
spectral division in the frequency domain. Though, the RFs are not
stacked in each bin, to enhance the coherence of converted phases as
sensitive to the absolute velocities within the layers, these can strongly
well as multiples. Examples of Q-component PRF traces from 2 stations,
constrain the velocity contrast across the discontinuities and their
sampling different parts of the WG are shown in Fig. 2. The Q-compo­
depths. In the present study, an extended-time multitaper frequency
nent PRF traces from other stations are shown in Fig. S2.
domain cross-correlation receiver function (ET MTRF) technique
(Helffrich, 2006) is adopted to calculate the PRFs. The converted phases
(e.g., Ps phases) originating at different depths arrive at different times 2.3. Estimation of Moho depth and Poisson’s ratio (H–K analysis)
with respect to the direct P phase; the shallower conversions have
smaller and the deeper ones have larger differential times. However, the For the estimation of Poisson’s ratio (σ ) and the Moho depth (ZM)
arrival times of the Ps conversions from a given interface depend on the beneath each station, the grid search approach of Zhu and Kanamori

2
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Table 1 wave velocity extracted from CRUST1.0 (Laske et al., 2013) based on the
Details of broadband seismological stations whose data are utilized in the pre­ location of the station (Table S1) is used to calculate the Moho depth or
sent study. crustal thickness (ZM) beneath each station. The obtained crustal
S. Station Lat Long Elevation Data Period thickness and Poisson’s ratio values vary from 34.8 to 50.0 km and 0.20
No. Code (∘N) (∘E) (m) to 0.29 (Table S1).
02–11–2018 to 09-
1 AGMB 13.507 75.095 659 08-2020 2.4. Common conversion point (CCP) images
04-10-2013 to 05-08-
2 ALDI 09.700 77.031 767 2016
In order to capture the crustal thickness variations along and across
04-10-2013 to 05-08-
3 CTRA 09.815 77.217 878 2016 the WG, we generated common conversion point (CCP) images along 4
05-10-2010 to 28-07- profiles, ab and cd (across WG), ef and gh (along WG) (Fig. 3) using good
4 ERD 11.351 77.634 143 2014 quality PRFs. The profiles have been selected based on the piercing
02–01–2005 to points of the Ps phase, at all the stations, at a depth of 50 km (Fig. 3). To
5 GOA 15.491 73.824 58 30–01–2017
02–11–2018 to 08-
construct the velocity model required for this exercise, the velocities
6 JODA 15.176 74.490 588 03-2021 below the station locations are extracted from CRUST1.0 (Laske et al.,
05-04-2013 to 28-07- 2013) down to a depth of Moho and from AK135 model below it
7 KAL 11.071 77.446 316 2014 (Kennett et al., 1995). This station specific model is utilized to back-
29-09-2010 to 28-07-
project the PRF amplitudes to depth along the ray path, by assuming
8 KAR 10.914 78.044 49 2014
04-10-2013 to 04-07- that all the energy arises from a Ps conversion. Later, the back-projected
9 KLMV 09.794 76.944 761 2016 amplitudes are binned in 2 × 2 km grids at various offsets and depths, to
08–11–2018 to 04- enhance the spatial coherence of the converted phases. The amplitudes
10 KNUR 11.872 75.588 44 03-2021 in the corresponding bins are stacked and projected onto a 2D reference
02–01–2005 to 03-
11 KOD 10.230 77.470 2345 02-2008
plane whose strike coincides with the chosen profile. The summed am­
01-02-2013 to 31-12- plitudes in the corresponding bins are plotted with a colour scheme
12 KOLH 16.834 74.281 593 2014 where blue indicates positive polarity and red represents negative po­
16-09-2011 to 29-07- larity. The CCP images along the 4 profiles are shown in Fig. 3. Plotting
13 MET 11.279 76.973 343 2014
of the Moho depth values obtained from H–K analysis on the corre­
02–11–2018 to 08-
14 MGLI 16.172 74.126 900 03-2021 sponding CCP images reveals a good correlation. The images along (ef
02–01–2005 to 30- and gh) and across the WG (ab and cd) clearly indicate variations in
15 MNGR 12.940 74.820 31 07-2014 crustal thickness.
26-09-2010 to 28-07-
16 NAM 11.185 78.187 93 2014
29-09-2010 to 13-03-
2.5. Harmonic decomposition
17 NAN 11.986 76.873 681 2014
10-10-2010 to 30-07- In general, for horizontal interfaces associated with a purely
18 OTY 11.489 76.649 1837 2014 isotropic structure, the converted energy is present only on the P and SV
11-10-2010 to 29-07-
components of the receiver functions. However, in case of dipping and/
19 PAL 10.530 77.527 227 2014
07-04-2000 to 02-03- or anisotropic interfaces, some of the Ps converted energy appears on the
20 PCH 10.530 76.340 60 2017 T-RFs, in view of the P to SH conversions (Levin and Park, 1997, 1998).
05-04-2013 to 30-07- In case of dipping interfaces, the Ps converted energy on the T-RFs
21 PKD 10.877 76.443 92 2014 changes polarity twice in a full back azimuthal span (i.e., a two-lobed
27-04-2014 to 05-02-
22 PMBA 09.388 77.156 994 2015
polarity flip). However, in the presence of anisotropy with a hexago­
24–01–2009 to nal symmetry and a horizontal symmetry axis, the Ps converted energy
23 POO 18.529 73.849 560 30–01–2017 on the T-RFs changes polarity four times in a full back azimuthal span (i.
09–11–2018 to 06- e., a four-lobed polarity flip). However, a combination of two- and four-
24 SBMN 12.660 75.618 134 03-2021
lobed polarity occurs in case of anisotropy with a plunging axis of
10–11–2018 to 06-
25 SDPR 14.333 74.760 535 03-2021 symmetry and/or a dipping isotropic interface. One way to distinguish
26-03-2011 to 28-07- between the two scenarios is by examination of energy on the T-RFs at
26 TRI 10.666 78.741 114 2014 the time of 0 s. The dipping isotropic interface shows a change in po­
27–01–2009 to larity associated with the Ps phase on the T-RFs accompanied by an
27 TRVM 08.508 76.958 64 30–01–2017
12-05-2013 to 31-12-
arrival with opposite polarity at time 0 s. This energy is not present in
28 VAL 20.217 73.447 277 2014 the case of anisotropy with a plunging axis of symmetry. However, this
05-03-2014 to 31-07- analysis becomes ambiguous in the presence of multiple layers, showing
29 VLKV 09.528 77.112 903 2016 dip and anisotropic effects, since multiple arrivals may interfere at zero-
09-04-2013 to 30-07-
time. A technique to decompose the receiver functions into back
30 VYT 11.663 76.111 738 2014
azimuthal harmonics i.e., the harmonic decomposition technique
developed by Bianchi et al. (2010) can overcome the limitations of vi­
(2000) has been applied to the PRFs. In this approach, the pair of sual analysis of amplitude and polarity changes in a given time window
Poisson’s ratio (σ) and Moho depth (ZM) that maximizes the summation and enable characterization of the crustal and uppermost mantle
of the amplitude of Ps + PpPs-(PpSs+PsPs) at the corresponding pre­ structure in terms of anisotropy and dipping nature of the interfaces.
dicted times, is chosen as the optimal one for that station. Clear Moho This method involves extraction of the back azimuthal harmonics from
multiples (PpPs and PpSs+PsPs) at the stations yielded well-constrained RF data as a function of the incoming P-wavefield direction. RFs can be
estimates of σ and ZM (as demonstrated from the low standard errors of represented as a scaled sum of coskϕ and sinkϕ at every time step, where
estimations). In this procedure, the estimation of the depth to the Moho k is the harmonic degree and ϕ is the back azimuth. This technique
beneath the station majorly depends on the mean crustal P wave velocity performs stacking of Q- and T-RFs with a phase shift of +Π/2k in back
(Chevrot and van der Hilst, 2000). In the present study, a mean crustal P azimuth to extract the signal related to anisotropy and/or dipping in­
terfaces (modelled part). The stacking of Q- and T-RFs with a phase shift

3
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Fig. 2. Example Q(SV) component P-receiver functions stacked in narrow slowness bins and the corresponding differential slowness stacks. Summation traces
moveout corrected for converted (bottom) and multiple (top) phases are shown in the top panel. Red and blue shaded regions indicate the Moho and LVL and the
corresponding multiples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

of -Π/2k represents the unmodeled part, that supresses the effects of the rose diagrams of T-RF energy in selected time windows are shown in
anisotropy and/or dipping structures. In the present analysis, we limit to Figs. 4, 5 and the corresponding harmonic decomposition along with the
the first three harmonic degrees i.e., k = 0, 1 and 2. Therefore, PRFs are particle motion and least square fits are shown in Fig. 6.
decomposed into five gathers i.e., k = 0 term (constant term), k = 1 term
(cosϕ and sinϕ) and k = 2 terms (cos2ϕ and sin2ϕ). Here, the constant 3. Results and discussion
term represents the bulk isotropic seismic velocity variation with depth.
In general, since the flat interfaces in an isotropic medium do not The PRFs at 30 broadband seismological stations indicate a variable
contribute to the T-RFs, the signals from them are visible only in the crustal thickness along and across the WG (Fig. 3). The times of the Ps
constant term i.e., a simple stack of Q-RFs over all the back azimuths. converted phases from the Moho range from ~3.5 s to 5.8 s among the
The first harmonic terms show a back azimuthal variation with a 2Π stations. The crustal thickness values determined from H–K analysis vary
periodicity and the second harmonic terms show a back azimuthal from 34.8 km to 50 km (Fig. S4 and Table S1). Images of the crust
variation with a Π periodicity. The first order terms are primarily sen­ revealed by the CCP stacks along profiles ef and gh, clearly indicate a
sitive to dipping interfaces and/or anisotropy with a plunging axis of significant variation in the crustal thickness, akin to the results from H–K
symmetry and expected to produce a two-lobed periodicity (Maupin and analysis (Fig. 3). Interestingly, the crust is shallow (~39 km and ~37
Park, 2007; Piana Agostinetti et al., 2011; Audet, 2015). However, the km) in the southern and northern parts of the WG, compared to that in
second order terms are sensitive to anisotropic layers with a horizontal the central part (~45 km). The crustal thickness is in conformity with
symmetry axis and produce a four-lobed periodicity (Maupin and Park, the geological units along the WG, being less beneath the southern SGT
2007; Shiomi and Park, 2008; Bianchi et al., 2010; Audet, 2015; Cossette and DVP, compared to that in the WDC. These results agree with those
et al., 2016). Moreover, anisotropy with a vertical axis of hexagonal from previous studies (e.g., Singh et al., 2017). Profile ab across the WG
symmetry cannot affect the first and second order terms (Tarayoun et al., reveals a variation in the Moho (Fig. 3), with the crustal thickness
2017). In the present study, we initially analysed the rose plots of T-RF increasing from ~39.8 km to ~43 km. Further, for a better under­
energy in specified time windows and checked the back azimuthal standing of the variation in crustal thickness across the WG, we refer to
coverage and polarity flipping. Further, those stations having a good the CCP image along the cd profile, traversing across it. Results from this
back azimuthal coverage are only considered for harmonic decomposi­ profile also reveal a variation in the Moho i.e., from ~48.6 km to ~43
tion analysis. These criteria resulted in a choice of 14 out of the 30 km (Fig. 3). These results clearly indicate that the crustal thickness
stations. Examples of Q- and T-RFs at stations TRVM, POO along with beneath the WG is distinct from the surrounding SGT region. Our results

4
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Fig. 3. Common conversion point (CCP) images along the four profiles. Red crosses in ab, cd, ef and gh profiles indicate the Moho depth values obtained from the
Zhu and Kanamori (2000) technique. Black dashed lines indicate the LVL, marked based on CCP images. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)

5
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Fig. 4. Examples of Q(SV) (top left panel) and T(SH) component (top right panel) P-receiver functions at TRVM station without moveout correction, stacked in
narrow back azimuthal bins. Summation trace is shown in the top panel. Solid and dotted lines indicate Moho and LVL respectively. Rose diagrams of transverse
component P-receiver functions as a function of back azimuth, corresponding to the Moho (bottom left panel) and LVL (bottom right panel). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)

gain support from those of the previous studies (Chand and Sub­ WG indicate the influence of hotspots on the Moho and the crust. Results
rahmanyam, 2003; Tiwari et al., 2006; Singh et al., 2017). A thinner from this study corroborate an alteration of the crustal thickness due to
crust beneath the southern and northern parts of the WG compared to hotspots and subsequent rifting (Fig. 8).
the central part, may be due to the impingement of the head of the Further, at most of the stations, we observed a strong negative phase
Marion hotspot beneath the southern part of WG and the Reunion hot­ just after the Ps conversion from the Moho (Figs. 2 and S2). The time of
spot beneath the northern part. Moreover, there is no major influence of this negative phase varies from 4.5 s to 7.9 s. This signature is also
the hotspot heads beneath the central part of WG where we observed a observed in the CCP stacks especially beneath the WG (Fig. 3). To
thicker crust compared to the southern and northern parts. Further, a confirm whether this phase is from a low velocity layer (LVL) beneath
plot of the plume affected regions in the western part of India, extracted the Moho, we computed the differential slowness stacks (Vinnik et al.,
from Raval and Veeraswamy (2003) (Fig. 8) supports the results from 2005), which distinguish converted phases from the multiples and other
the present study. Also, the dykes along the south-western margin of WG artefacts. This is due to the fact that the differential slowness is negative
have been dated at 90 Ma (Radhakrishna et al., 1994, 1999) and related for converted phases and positive for multiples. These results clearly
to the Marion hotspot, while the Deccan flood basalts are linked to the validate that the negative phase just after the Moho is a conversion from
thermo-magmatic Reunion plume interaction at 65 Ma (Courtillot et al., an LVL(Figs. 2 and 3). Generally, LVLs are thermal or chemical anom­
2000). These surface signatures at the southern and northern parts of alies originating from regional or local sources. The major cause

6
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Fig. 5. Examples of Q(SV) (top left panel) and T(SH) component (top right panel) P-receiver functions at POO station, without moveout correction, stacked in narrow
back azimuthal bins. Summation trace is shown in the top panel. Solid and dotted lines indicate Moho and LVL respectively. Rose diagrams of transverse component
P-receiver functions as a function of back azimuth, corresponding to the Moho (bottom left panel) and LVL (bottom right panel). (For interpretation of the references
to colour in this figure legend, the reader is referred to the web version of this article.)

responsible for an LVL beneath the Moho is linked to thermal effects like To understand the nature of Moho and the sub-Moho LVL along the
partial melting (Kumar et al., 2002; Benz and McCarthy, 1994). It is WG, we applied the harmonic decomposition method to RFs at stations
postulated that even small quantities (<1–2%) of melts may cause a with a good back azimuthal coverage. The back azimuthal stack sections
substantial decrease in seismic velocities (Mavko, 1980). Also, carbon­ of the PRFs and rose diagrams of the T-RF amplitudes in the selected
ates can generate melt (expressed as an LVL) by decreasing the solidus time windows (i.e., at the Moho and LVL) are used to check the suit­
temperature (Presnall and Gudfinnson, 2005). An LVL beneath the WG ability. The time windows are selected based on the Ps conversion time
may not be linked only to partial melting beneath the DVP since this LVL of the Moho and the LVL just after the Moho, with an offset of ±0.2 s i.e.,
extends to the south of the DVP region too, i.e., we observed this LVL the time window starts from 0.2 s before the peak amplitude of Ps and
even beneath the southern parts of DVP region (WDC and SGT). More­ ends at 0.2 s after the peak amplitude of Ps in case of the Moho and the
over, this signature is not clear and not resolved properly beneath re­ peak amplitude of negative phase just after the Moho is considered as
gions away from the west coast (WG) i.e., after 300 km in profiles ab and the reference in case of the LVL. The PRFs at 14 out of the 30 stations
cd. These results indicate the existence of an LVL all along the WG. This satisfied this criterion. Later, we analysed the Q-, T-RFs and rose dia­
LVL could be associated with rifting of the Indian sub-continent, since grams of the T-RFs in the selected time windows, in detail, to decipher
this feature is majorly along the continental margin. the nature of the discontinuity based on the polarity flip. Results from

7
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Fig. 6. Examples of computed harmonic decomposition

components (constant, K = 1: cosϕ, sinϕ and K = 2: cos2ϕ,
sin2ϕ) for stations TRVM (first row panels) and POO (third
row panels). Solid and dotted lines in top panels indicate the
Moho and LVL respectively. Particle motion defined using
the K = 1 harmonic components for the Moho and LVL, for
stations TRVM (second row panels) and POO (fourth row
panels). Red lines in the bottom panels indicate the least
square fit line of the particle motion. (For interpretation of
the references to colour in this figure legend, the reader is
referred to the web version of this article.)

8
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

the rose diagrams of the T-RF amplitudes at the Moho reveal a two lobed show that the Moho along the WG is either dipping and/or anisotropic in
polarity flip at 6 stations namely AGMB, ALDI, CTRA, MNGR, NAN, nature. Further, to calculate the dip direction and/or trend of the sym­
TRVM and a mix of two and four lobed polarity flips at the remaining 8 metry axis of anisotropy, we estimate the linear trend of the particle
stations namely ERD, GOA, JODA, KAR, NAM, PAL, POO, TRI. The two motion of the K = 1 terms within the selected time window by a least
lobed polarity flip at 6 stations may indicate a dipping isotropic Moho square fit. These results indicate that the dip direction of the Moho (from
and the combination of two and four lobed flip at 8 stations indicates north) and/or trend of anisotropy all along the WG varies from 58.0∘ to
anisotropy with a plunging axis of symmetry or a dipping isotropic 106.8∘ in the southern part, 38.6∘ to 122.7∘ in the central part and 148.2∘
interface. We analysed the energy at zero time for the stations where we in the northern part.
observed such two and four lobed polarity flip, to distinguish between Further, results from rose diagrams corresponding to the LVL clearly
anisotropy with a plunging axis of symmetry or a dipping isotropic indicate a four lobed polarity flip at 3 stations namely AGMB, GOA,
interface. Results reveal energy with an opposite polarity at 0 s, with TRVM, which suggests anisotropy with a horizontal symmetry axis.
respect to the Ps energy (Fig. S3), which may indicate a dipping Moho. Further, a combination of two and four lobed polarity flip observed at
However, as discussed in the harmonic decomposition section, this the remaining 11 stations namely ALDI, CTRA, ERD, JODA, KAR,
analysis becomes ambiguous when there are multiple interfaces at MNGR, NAM, NAN, PAL, POO, TRI, reveals anisotropy with a plunging
depth, since multiple zero-time arrivals may interfere with each other. axis of symmetry and/or a dipping isotropic interface. Further, we
Further, for a detailed investigation of the nature of the identified analysed the energies on the K = 1 and 2 terms corresponding to the LVL
discontinuity, five components of the hormonic decomposition at each and observed a dominance of energy on the K = 1 term and no signifi­
station i.e., k = 0 (constant term), k = 1 (cosϕ and sinϕ terms) and k = 2 cant energy on the unmodelled components. These results along with
(cos2ϕ and sin2ϕ terms) are measured. Later, the amplitudes on the 5 the energies on the T-RFs within the selected time window of LVL
components within the selected time windows along with the number of indicate anisotropy with a plunging axis of symmetry all along the WG.
loops within it (i.e., Moho) are analysed. This analysis indicates domi­ The calculated trend of the apparent symmetry axes of anisotropy from
nance of energy on the k = 1 & 2 terms (cosϕ; sinϕ and cos2ϕ; sin2ϕ the least square fit of the particle motion within the selected time win­
components) in the time window corresponding to the Moho. Moreover, dow, varies from 58.2∘ to 134.4∘. Specifically, these values vary from
there is no significant energy in the unmodelled harmonic stacks, which 58.2∘ to 134.4∘ in the southern part, 60.2∘ to 122.7∘ in the central part
argues that the two lobed and/or mix of two and four lobed signals at the and 107.7∘ in the northern part of WG. The strong negative phase just
Moho may not be caused by a dip in the Moho (Liu and Park, 2017), but after the Ps converted phase and the harmonic decomposition results of
due to anisotropy. These results along with the energy analysis of T-RFs this phase at most of the stations (Figs. 4, 7) suggest an anisotropic LVL

Fig. 7. Map showing the least fit line of the particle motion calculated using the K = 1 terms of the harmonic analysis for 14 stations, for the Moho (left panel) and
LVL (right panel). Length of the bars is proportional to the maximum elongation of particle motion and angle represents the dip direction or trend of the symmetry
axes of the Moho (left panel) and trend of the symmetry axes of anisotropy in the LVL (right panel). Blue and brown colour lines in the left panel indicate a two lobed
polarity flip of amplitude on T-RFs with the dominance of amplitude on the k = 1 terms and combination of two and four lobed polarity flip with amplitude at 0 s,
dominance of amplitude on the k = 1 term respectively. Blue and brown colour lines in the right panel indicate four lobed polarity flip of T-RFs amplitude with the
dominance of amplitude on the k = 1 terms and combination of two & four lobed polarity flip with the dominance of amplitude on the k = 1 term respectively. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

9
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

beneath the Moho along the Western Ghats. The trend of the apparent Therefore, the results from the present study in terms of variable crustal
symmetry axes of anisotropy within this layer is almost perpendicular to structure and anisotropic LVL vindicate rifting as a plausible mechanism
the strike of the WG. A sub-Moho LVL beneath the WG is also reported in for the evolution of WG.
previous studies (Tiwari et al., 2006). Also, such a layer in the shallow The west coast of Indian sub-continent is a passive margin that is
upper mantle beneath the DVP has been documented (Kumar and exposed to multi stage rifting episodes: (i) due to the Marion hotspot
Mohan, 2005). The observed coast perpendicular trend of the apparent activity, the greater India (the Indian sub-continent with Seychelles and
symmetry axis of anisotropy within the LVL also supports its rifting the Mascarene plateau) got separated from Madagascar between 91 and
origin. Lithospheric extension or stretching associated with the rifting 83 Ma ago (Molnar et al., 1988; Storey et al., 1995; Torsvik et al., 2000),
process aligns the anisotropic minerals and/or flow in the direction of (ii) due to the Reunion hotspot activity, the Indian sub-continent broke
extension and this process produces anisotropy parallel to the exten­ away from Seychelles and the Mascarene plateau between 66 and 65 Ma
sion/stretching direction (Vinnik et al., 1992). This result also correlates (White and McKenzie, 1989; Vandamme et al., 1991; Ramana et al.,
with the observations from other continental rifting models (Werner and 2015). The significant variation in the crustal thickness near or at the
Kahle, 1980). coast i.e., beneath the WG, altered by the rifting mechanism engenders
The two major hypotheses proposed to explain the evolution of WG the possibility of block faulting and vertical tectonics in the upliftment
are directly linked to the lithospheric structure. The first one is the of WG. Results from geomorphological studies (e.g., Radhakrishna,
buoyancy model of the crust which postulates existence of extra mass 1993) also indicate that the WG is formed due to Cenozoic uplifts, rifting
either at the surface or at the base of the crust in terms of underplating. and down faulting. Lithospheric stretching in rift zones reduces the
The second hypothesis is the isostatic model, which requires a decrease thickness of the crust, causes the surface to subside and enables for­
in density, primarily due to mass reduction either by local heat sources mation of extensional basins (McKenzie, 1978). Faults generated due to
or the effect of erosion. However, the crustal thickness values and dip­ extension may create space for sediment deposition in the brittle upper
ping/anisotropic nature of the Moho derived from this study do not crust and the lower crust could accommodate the necking by a ductile
support these hypotheses because the crustal thickness is not uniform mechanism. In such regions, the crustal material is replaced with the
along the WG and there is no significant correlation between the high density upper mantle, which creates subsidence by negative
elevation of WG and crustal thickness. In contrast, we observe a strong buoyancy. Moreover, the rift flanks far from the tectonically extended
lateral variation along the WG. As explained in the above paragraph, this areas may encounter subsidiary uplift due to the long wavelength lith­
variation could be due to alteration owing to interaction with the hot­ ospheric necking (Fig. 8). Further, the study of Osmundsen and Redfield
spots and subsequent rifting. Further, the anisotropic LVL beneath the (2011) suggests that the crustal taper linked to rifting and its strength
Moho with the coast perpendicular trend of the symmetry axes all along gradient has significant controls on the topography of the escarpment.
the WG also supports the rifting mechanism. These results corroborate The crustal taper could be due to thinning of the crust by the effect of
with the results of Tiwari et al. (2006). Their observations of varied detachment related extension, whose hinterland is the continental part
crustal thickness along the WG, negative isostatic anomalies varying with a great upliftment at the edges (Osmundsen and Redfield, 2011). In
from − 40 to − 70 mGals, along with a negative Ps phase just after the this case, the position of crustal thinning will depend on the distance to
Moho conversion can be attributed to a sub-Moho LVL. Further, they the continent-ocean boundary that leads to a taper-break where the
interpreted that these could be related to a rifting mechanism. Presence thickness of the crust is less than ~10 km, which is inversely propor­
of low-density material in the upper mantle level (e.g., Pandey et al., tional to the elevation of the escarpment (Davis and Kusznir, 2002;
1996) is well explained by the negative isostatic anomalies over the WG. Osmundsen and Redfield, 2011). In the case of WG, if the break in the

Fig. 8. Right Panel: The red shaded areas indicate the probable zones of influence of plume heads. Tectonic units: SGT – Southern Granulite Terrain (Brown), WDC –
Western Dharwar Craton (Yellow green), DVP – Deccan Volcanic Province (Light gold), CG – Closepet Granite (Slate gray), EDC – Eastern Dharwar Craton (Gold), CB
– Cuddapah Basin (Pink), EGMB – Eastern Ghats Mobile Belt (Light green) and WG – Western Ghats (Cross-pattern). Left Panel: Cartoon diagram showing the possible
alterations in the Moho, rifting process and the corresponding upliftment due to the influence of plumes/hotspots. The direction (red line) in the circle indicates the
trend of the symmetric axes of anisotropy in the LVL, extracted from the harmonic decomposition results. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)

10
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

crustal taper is assumed to be at the Laccadive ridge i.e., at a distance of Audet, P., 2015. Layered crustal anisotropy around the San Andreas fault near Parkfield,
California. J. Geophys. Res. 120 (5), 3527–3543.
~300 km, an elevation of 1.65 km is predicted based on the relation
Benz, H.M., McCarthy, I., 1994. Evidence for an upper mantle low velocity zone beneath
derived by Osmundsen and Redfield (2011) (Sinha-Roy, 2018). This the southern Basin and Range-Colarado plateau transition zone. Geophys. Res. Lett.
corroborates well with the present day average elevation of ~1.2 km for 21, 509–512.
the WG. These results agree with the taper related escarpments of Bianchi, I., Park, J., Piana Agostinetti, N., Levin, V., 2010. Mapping seismic anisotropy
using harmonic decomposition of receiver functions: an application to Northern
similar ages, for instance, the 85–65 Ma old rifted continental margins of Apennines, Italy. J. Geophys. Res. 115, B12317.
Australia and Greenland. Thus, the rifting theory well explains the origin Buck, W.R., 1986. Small scale convection induced by passing rifting: the cause of uplift of
of WG and mostly supports a multi-stage development. rift shoulders. Earth Planet. Sci. Lett. 77, 362–372.
Campanile, D., Nambiar, C.G., Bishop, P., Widdowson, M., Brown, R., 2008.
Sedimentation record in the Konkan-Kerala Basin: implications for the evo1ution of
4. Conclusions the Western Ghats and the Western Indian passive margin. Basin Res. 20 (1), 3–22.
Chand, S., Subrahmanyam, C., 2003. Rifting between India and Madagascar- mechanism
and isostacy. Earth Planet. Sci. Lett. 210, 317–332.
(i) A clear change in the crustal thickness is observed along the WG. Chevrot, S., van der Hilst, R.D., 2000. The Poisson ratio of the Australian crust:
The crustal thickness increases from ~39 km to ~45 km from geological and geophysical implications. Earth Planet. Sci. Lett. 183, 121–132.
south to the centre of the WG and then decreases towards the Cochran, J.R., 1983. Effects of finite extension times on the development of sedimentary
basins. Earth Planet. Sci. Lett. 66, 289–302.
north, from ~45 km to ~37 km. The variations corroborate well Cossette, É., Audet, P., Schneider, D., Grasemann, B., 2016. Structure and anisotropy of
with the geological units, regions of interaction of the hotspots the crust in the Cyclades, Greece, using receiver functions constrained by in situ rock
and the subsequent rifting phenomenon. textural data. J. Geophys. Res. 121 (4), 2661–2678.
Courtillot, V., Gallet, Y., Rocchia, R., Feraud, G., Robin, E., Hofmann, C., Ghevariya, Z.G.,
(ii) A strong negative Ps conversion following the Moho phase is
2000. Cosmic markers, 40Ar/39Ar dating and paleomagnetism of the KT sections in
indicative of a sub-Moho LVL beneath the WG. the Anjar area of the Deccan large igneous province. Earth Planet. Sci. Lett. 182 (2),
(iii) Harmonic decomposition of receiver functions reveals a dipping/ 137–156.
anisotropic nature of the Moho with the dip direction or trend of Cox, K.G., 1980. A model for flood basalt volcanism. J. Petrol. 21, 629–650.
Davis, M., Kusznir, N., 2002. Are buoyancy forces important during the formation of rift
the symmetry axis varying from 38.6 ∘ to 148.2∘ all along the WG. margins? Geophys. J. Int. 149, 524–533.
Also, the LVL is found to be anisotropic with a plunging axis of Devey, C.W., Lightfoot, P.C., 1986. Volcanological and tectonic control of stratigraphy
symmetry. The trend of symmetry axes varies from 58.2∘ to and structure in the western Deccan traps. Bull. Volcanol. 48, 195–207.
Gilchrist, A.R., Summerfield, M.A., 1994. Tectonics model of passive margin evolution
134.4∘, almost perpendicular to the west coast of the Indian sub- and their implications for theories of long term landscape development. In: Kirby, M.
continent. J. (Ed.), Process Models and Theoretical Geomorphology. John Wiley and Sons Ltd,
(iv) The crustal thickness variations, with dipping/anisotropic Moho pp. 55–84.
Gunnell, Y., 2001. Dynamics and kinematics of rifting and uplift at the western
and anisotropic sub-Moho LVL shed light on the different rifting continental margin of India, Insights from geophysics and numerical models. In:
episodes along the west coast of India, which could be the key Gunnell, Y., Radhakrishna, B.P. (Eds.), Sahyadri, the Great Escarpment of the Indian
causative factor for the evolution of WG. Subcontinent. Geological Society of India, Memoir No. 47, Bangalore, India,
pp. 475–496.
Gunnell, Y., Fleitout, L., 2000. Morphotectonic evolution of the Western Ghats, India. In:
CRediT authorship contribution statement Summerfield, M. (Ed.), Geomorphology and Global Tectonics. John Wiley and Sons.,
pp. 321–338
Gupta, S., Kanna, N., Kumar, S., Sivaram, K., 2018. Crustal Thickness and Composition
Padma Rao B: Conceptualization, Data curation, Formal analysis,
Variation along the Western Ghats of India through Teleseismic Receiver Function
Investigation, Methodology, Validation, Visualization, Writing – orig­ Analysis. J. Geol. Soc. India 92 (5), 523–528.
inal draft, Writing – review & editing. Ravi Kumar M: Validation, Helffrich, G., 2006. Extended-time multitaper frequency domain cross-correlation
Writing – review & editing. receiver-function estimation. Bull. Seismol. Soc. Am. 96 (1), 344–347.
Kaila, K.L., Reddy, P.R., Dixit, M.M., Lazrenko, M.A., 1981. Deep crustal structure at
Koyna, Maharashtra indicated by deep seismic sounding. J. Geol. Soc. India 22 (1),
Declaration of Competing Interest 1–16.
Kennett, B.L.N., Engdahl, E.R., Buland, R., 1995. Constraints on seismic velocities in the
Earth from travel times. Geophys. J. Int. 122, 108–124.
The authors declare that they have no known competing financial Kind, R., Vinnik, L., 1988. The upper mantle discontinuities underneath the grf array
interests or personal relationships that could have appeared to influence from p-to-s converted phases. J. Geophys. 62 (1), 38–147.
the work reported in this paper. Kind, R., Yuan, X., 2011. Seismic, receiver function technique. In: Gupta, H.K. (Ed.),
Encyclopedia of Solid Earth Geophysics. Encyclopedia of Earth Sciences Series.
Springer, Dordrecht. https://doi.org/10.1007/978-90-481-8702-7_12.
Acknowledgements Kumar, M.R., Mohan, G., 2005. Mantle discontinuities beneath the Deccan volcanic
province. Earth Planet. Sci. Lett. 237, 252–263.
Kumar, M.R., Ramesh, D.S., Saul, J., Sarkar, D., Kind, R., 2002. Crustal structure and
We sincerely acknowledge the Ministry of Earth Sciences, Govern­ upper mantle stratigraphy of the Arabian shield. Geophys. Res. Lett. 29 (8) https://
ment of India for supporting this project under the core program of doi.org/10.1029/2001GL014530.
National Centre for Earth Science Studies (NCESS). We thank the Na­ Langston, C.A., 1979. Structure under mount rainier, Washington, inferred from
teleseismic body waves. J. Geophys. Res. 84 (B9), 4749–4762.
tional Center for Seismology, National Geophysical Research Institute
Laske, G., Masters, G., Ma, Z., Pasyanos, M., 2013. Update on crust1. 0 - a 1-degree global
and Institute of Land & Disaster Management for sharing the data. We model of earth’s crust. In. Geophys. Res. Abstr. 15, 2658.
are grateful to Dipankar Saikia for the RF and HD codes. All the figures Levin, V., Park, J., 1997. P-SH conversions in a flat-layered medium with anisotropy of
arbitrary orientation. Geophys. J. Int. 131, 253–266.
are made using the GMT software (Wessel and Smith, 1995). We
Levin, V., Park, J., 1998. P-SH conversions in layered media with hexagonally symmetry
sincerely thank the reviewers for their thoughtful comments, which anisotropy: a cookbook. Pure Appl. Geophys. 151, 669–697.
significantly improved the quality of the manuscript. Liu, Z., Park, J., 2017. Seismic receiver function interpretation: P s splitting or
anisotropic underplating? Geophys. J. Int. 208 (3), 1332–1341.
Matmon, A., Bierman, P., Enzel, Y., 2002. Pattern and tempo of great escarpment
Appendix A. Supplementary data erosion. Geology 30 (12), 1135–1138.
Maupin, V., Park, J., 2007. Theory and observations: Wave propagation in anisotropic
Supplementary data to this article can be found online at https://doi. media. In: Romanowicz, B., Dziewonski, A. (Eds.), Treatise on Geophysics,
Seismology and Structure of the Earth, 1, 289–321. Elsevier, New York.
org/10.1016/j.tecto.2022.229472. Mavko, G.M., 1980. Velocity and attenuation in partially melted rocks. J. Geophys. Res.
85, 5173–5189.
References McKenzie, D., 1978. Some remarks on the development of sedimentary basins. Earth
Planet. Sci. Lett. 40, 25–32.
Molnar, P., Pardo-Casas, F., Stock, J., 1988. The Cenozoic and late cretaceous evolution
Arora, K., Tiwari, V.M., Singh, B., Mishra, D.C., Grevemeyer, I., 2012. Three dimensional
of the Indian Ocean Basin: uncertainties in the reconstructed positions of the Indian,
lithospheric structure of the western continental margin of India constrained from
African and Antarctic plates. Basin Res. 1 (1), 23–40.
gravity modelling: implication for tectonic evolution. Geophys. J. Int. 190 (1),
131–150.

11
P.R. B. and R.K. M. Tectonophysics 838 (2022) 229472

Osmundsen, P.T., Redfield, T.F., 2011. Crustal taper and topography at passive Sinha-Roy, S., 2018. A hybrid multistage model of evolution of the Western Ghats at the
continental margins. Terra Nova. https://doi.org/10.1111/j.1365.2011.0101. passive western continental margin of India. J. Geol. Soc. India 92 (5), 533–541.
Owens, T.J., Zandt, G., Taylor, S.R., 1984. Seismic evidence for an ancient rift beneath Storey, M., Mahoney, J.J., Saunders, A.D., Duncan, R.A., Kelley, S.P., Coffin, M.F., 1995.
the Cumberland plateau, Tennessee. J. Geophys. Res. 116, 618–636. Timing of hot spot-related volcanism and the breakup of Madagascar and India.
Pandey, O.P., Agarwal, P.K., Negi, J.G., 1996. Evidence of low density sub-crustal Science 267 (5199), 852–855.
underplating beneath western continental region of India and adjacent Arabian Sea: Tarayoun, A., Audet, P., Mazzotti, S., Ashoori, A., 2017. Architecture of the crust and
geodynamical considerations. J. Geodyn. 21, 365–377. uppermost mantle in the northern Canadian Cordillera from receiver functions.
Piana Agostinetti, N., Bianchi, I., Amato, A., Chiarabba, C., 2011. Fluid migration in J. Geophys. Res. 122 (7), 5268–5287.
continental subduction: the Northern Apennines case study. Earth Planet. Sci. Lett. Tiwari, V.M., Mishra, D.C., 1999. Estimation of effective elastic thickness from gravity
302 (3–4), 267–278. and topography data under the Deccan Volcanic Province. India. Earth Planet. Sci.
Presnall, D.C., Gudfinnson, G.H., 2005. Carbonate rich melts in the oceanic low velocity Lett. 171 (2), 289–299.
zone and deep mantle, in Plates, plumes and paradigms. In: Foulger, G.R., Natland, J. Tiwari, P.K., Surve, G., Mohan, G., 2006. Crustal constraints on the uplift mechanism of
H., Presnall, D.C., Anderson, D.L. (Eds.), Geol. Soc. Am. Special paper, 388, the Western Ghats of India. Geophys. J. Int. 167, 1309–1316.
pp. 207–216. Torsvik, T.H., Tucker, R.D., Ashwal, L.D., Carter, L.M., Jamtveit, B., Vidyadharan, K.T.,
Radhakrishna, B.P., 1993. Neogene uplift and geomorphic rejuvenation of the Indian Venkataramana, P., 2000. Late cretaceous India-Madagascar fit and timing of break-
peninsula. Curr. Sci. 64, 787–793. up related magmatism. Terra Nova 12 (5), 220–224.
Radhakrishna, T., Dallmeyer, R.D., Joseph, M., 1994. Palaeomagnetism and 36Ar40Ar Vandamme, D., Courtillot, V., Besse, J., Montigny, R., 1991. Paleomagnetism and age
vs. 39Ar40Ar isotope correlation ages of dyke swarms in Central Kerala, India: determinations of the Deccan Traps (India): results of a Nagpur-Bombay Traverse
Tectonic implications. Earth Planet. Sci. Lett. 121 (1–2), 213–226. and review of earlier work. Rev. Geophys. 29 (2), 159–190.
Radhakrishna, T., Maluski, H., Mitchell, J.G., Joseph, M., 1999. 40Ar/39Ar and K/Ar Vinnik, L.P., 1977. Detection of waves converted from P to SV in the mantle. Phys. Earth
geochronology of the dykes from the south Indian granulite terrain. Tectonophysics Planet. Inter. 15, 39–45.
304 (1–2), 109–129. Vinnik, L.P., Makeyeva, L.I., Milev, A., Usenko, A.Y., 1992. Global patterns of azimuthal
Radhakrishna, T., Mohamed, A.R., Venkateshwarlu, M., Soumya, G.S., Prachiti, P.K., anisotropy and deformations in the continental mantle. Geophys. J. Int. 111 (3),
2019. Mechanism of rift flank uplift and escarpment formation evidenced by 433–447.
Western Ghats. India. Sci. Rep. 9 (1), 1–7. Vinnik, L., Kurnik, E., Farra, V., 2005. Lehmann discontinuity beneath North America: no
Rai, S.S., Borah, K., Das, R., Gupta, S., Srivastava, S., Prakasam, K.S., Meena, R., 2013. role for seismic anisotropy. Geophys. Res. Lett. 32 (9) https://doi.org/10.1029/
The south India Precambrian crust and shallow lithospheric mantle: Initial results 2004GL022333.
from the India Deep Earth Imaging Experiment (INDEX). J. Earth Syst. Sci. 122 (6), Weissel, J.K., Karner, G.D., 1989. Flexural uplift of rift flanks due to mechanical
1435–1453. unloading of the lithosphere during extension. J. Geophys. Res. 94, 13919–13950.
Ramana, M.V., Desa, M.A., Ramprasad, T., 2015. Re-examination of geophysical data off Werner, D., Kahle, H.G., 1980. A geophysical study of the Rhinegraben-I. Kinematics and
Northwest India: Implications to the late cretaceous plate tectonics between India geothermics. Geophys. J. Int. 62 (3), 617–629.
and Africa. Mar. Geol. 365, 36–51. Wessel, P., Smith, W.H.F., 1995. New version of the generic mapping tools released. Eos.
Raval, U., Veeraswamy, K., 2003. Modification of geological and geophysical regimes Trans. Am. Geophys. Union 76, 329.
due to interaction of mantle plume with Indian lithosphere. J. Virtual Explor. 12 White, R., McKenzie, D., 1989. Magmatism at rift zones: the generation of volcanic
(12), 117–143. continental margins and flood basalts. J. Geophys. Res. 94 (B6), 7685–7729.
Shiomi, K., Park, J., 2008. Structural features of the subducting slab beneath the Kii Widdowson, M., Cox, K.G., 1996. Uplift and erosional history of the Deccan traps, India:
Peninsula, Central Japan: Seismic evidence of slab segmentation, dehydration, and evidence from laterites and drainage patterns of the western ghats and konkan coast.
anisotropy. J. Geophys. Res. 113, B10318. Earth Planet. Sci. Lett. 137, 57–69.
Singh, A., Kumar, M.R., Mohanty, D.D., Singh, C., Biswas, R., Srinagesh, D., 2017. Crustal Zhu, L., Kanamori, H., 2000. Moho depth variation in southern California from
structure beneath India and Tibet: New constraints from inversion of receiver teleseismic receiver functions. J. Geophys. Res. 105 (B2), 2969–2980.
functions. J. Geophys. Res. 122, 7839–7859.

12