Original scientific paper Seismic imaging beneath Sumatra Island and its surroundings, Indonesia, from local-regional P-wave earthquake tomography

Abstract

Sumatra Island and its surroundings, Indonesia, are one of the most active tectonics in the world. The Aceh-Andaman earthquake, one of the most destructive earthquakes in the world, occurred there. It has attracted many earth scientists to apply various methods, including seismic tomography, to understand the island's subsurface structure and tectonic system. This study is the first to delineate subsurface imaging beneath the island and its surroundings using a local-regional earthquake catalogue from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) seismic network. The tomographic imaging of P-wave (Vp) conducted in this study has successfully delineated subduction slabs (high Vp), partial melting zones (low Vp), volcanic arcs (low Vp), and Sumatran Fault zones (low Vp). The relationship between the subduction zone and the volcanic arc on the island can be seen on several vertical sections where a partial melting zone occurs at a depth of about 100 km, which functions as magma feeding for some volcanoes on the island. The oceanic slab model also exhibits a more pronounced and steeper slope towards the southern regions of Sumatra Island, possibly attributed to the slab's aging process in that direction. The results highlight the importance of the BMKG seismic network in imaging local-regional subsurface structures beneath Indonesia's archipelago, especially for the main islands such as Sumatra.

Full text

119-132
119
Rudarsko-geološko-naftni zbornik
(The Mining-Geology-Petroleum Engineering Bulletin)
UDC: 551.4:551.3
DOI: 10.17794/rgn.2023.3.10
Original scientic paper
Corresponding author: Mohamad Ramdhan
e-mail address: mohamad.ramdhan@brin.go.id
Seismic imaging beneath Sumatra Island
and its surroundings, Indonesia,
from local-regional P-wave earthquake
tomography
Bayu Pranata1, Mohamad Ramdhan2, Muhammad Hanif2, Muhammad Iqbal Sulaiman3,
Mufti Putra Maulana4, Wandono4, Sri Widiyantoro5, Sandy Kurniawan Suhardja6,
Edi Hidayat2, Pepen Supendi1, Ridwan Kusnandar1, Wiko Setyonegoro2
1 Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG), Jalan Angkasa I, No. 2, Kemayoran, Jakarta 10720, Indonesia
2 Research Center for Geological Disaster, National Research and Innovation Agency (BRIN), KST Samaun Samadikun,
Jl. Cisitu Sangkuriang, Bandung 40135, Indonesia
3 Luwu Disaster Management Agency (BPBD Kab. Luwu), Jl. Andi Djemma, Senga, Belopa, Kab. Luwu 91994, Indonesia
4 School of Meteorology, Climatology, and Geophysics of Indonesia (STMKG), Jl. Perhubungan I No. 5, Pondok Betung, Tangerang Selatan
15221, Indonesia
5
Global Geophysics Research Group, Faculty of Mining and Petroleum Engineering, Institut Teknologi Bandung, Bandung 40132, Indonesia;
Faculty of Engineering, Maranatha Christian University, Bandung 40164, Indonesia
6 University of Pertamina, Jl. Teuku Nyak Arief, Simprug, Kebayoran Lama, Jakarta 12220, Indonesia
Abstract
Sumatra Island and its surroundings, Indonesia, are one of the most active tectonics in the world. The Aceh-Andaman
earthquake, one of the most destructive earthquakes in the world, occurred there. It has attracted many earth scientists
to apply various methods, including seismic tomography, to understand the island’s subsurface structure and tectonic
system. This study is the rst to delineate subsurface imaging beneath the island and its surroundings using a local-re-
gional earthquake catalogue from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) seismic
network. The tomographic imaging of P-wave (Vp) conducted in this study has successfully delineated subduction slabs
(high Vp), partial melting zones (low Vp), volcanic arcs (low Vp), and Sumatran Fault zones (low Vp). The relationship
between the subduction zone and the volcanic arc on the island can be seen on several vertical sections where a partial
melting zone occurs at a depth of about 100 km, which functions as magma feeding for some volcanoes on the island. The
oceanic slab model also exhibits a more pronounced and steeper slope towards the southern regions of Sumatra Island,
possibly attributed to the slab’s aging process in that direction. The results highlight the importance of the BMKG seismic
network in imaging local-regional subsurface structures beneath Indonesia’s archipelago, especially for the main islands
such as Sumatra.
Keywords:
Sumatra; BMKG, P– wave; tomography; slab; fault
1. Introduction
Sumatra Island and its surroundings are among the
areas in Indonesia with very high seismic activity. Seis-
mic activity in this region is caused by the oblique sub-
duction of the Australian Plate to the Southeast Asian
Plate. The offshore plate boundary of Sumatra Island,
Indonesia, is formed due to the oblique subduction of the
Indian and Australian sub-plates beneath the Sunda Plate
at variable rates of 5-6 cm/year (DeMets et al., 2010).
Subduction resulted in two central tectonic systems: the
subduction zone fault system along the Sumatra Sea
trench and the Sumatran Fault System. The second fault
system, known as the Sumatran Fault, is a strike-slip
fault with a dextral direction that spans the 1900 km Su-
matran backbone, consisting of 20 main segments and is
located close to the volcanic arc of the island of Sumatra
(Irsyam et al., 2017; Natawidjaja, 2018; Sieh & Na-
tawidjaja, 2000). The Sumatran Fault Zone (SFZ) spans
from the Sunda Strait in the south with a tectonic strain
rate of ±6 mm/year to the Andaman Sea in the north with
a rifting rate of 37 mm/year, while exhibiting varying
slip rates, such as ~9 mm/year in the Sunda Strait and on
the Kumering Fault in south Sumatra, 14-15 mm/year
near Lake Maninjau in central Sumatra, and 16±6 mm/
year across the Batee and Takengon segments in the
Aceh region, increasing to 20±6 mm/year across the
Aceh segment (Natawidjaja, 2018). Oblique subduc-
tion plays a crucial role in forming the ±600 km Menta-

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 120
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
wai fault system, which operates parallel to the Suma-
tran Fault System. This process contributes to the uplift
of the accretionary wedge and the deposition of sedi-
ments in the forearc basin, resulting in the development
of elongated Neogene basin depocenters aligned parallel
to the trench in the Sumatran forearc region (Mukti et
al., 2021). The Mentawai fault system is located in the
fore-arc basin from the southernmost Sumatra to the
north of Siberut Island, the Mentawai Islands. Figure 1
shows the tectonic system of Sumatra Island and its sur-
roundings, along with several signicant earthquakes
between 2004 and 2016, as shown in Table 1. Addition-
ally, Figure S1 displays the most recent segments of the
Sumatran Fault System.
The heightened seismic activity in this region has
sparked considerable interest among earth scientists
aiming to comprehend the earthquake source character-
istics and their potential impact. Both aspects hold great
value in mitigating earthquake disasters in the area. Var-
ious geophysical methods have been devised and em-
ployed to investigate these aspects. As these characteris-
tics are inuenced by processes occurring beneath the
earth’s surface, a comprehensive understanding of the
subsurface structure becomes paramount. Seismic to-
mography serves as a primary method for comprehend-
ing the earth’s subsurface structure on a local to a global
scale, and it has been effectively utilized in the study of
Sumatra Island and its surrounding areas. Local seismic
tomographic studies in the Toba Caldera area have deter-
mined a magma reservoir 5 km below sea level charac-
terized by a high Vp/Vs value of 1.9 (Koulakov et al.,
2009). The study also described the partial melting zone
that serves as the source of the caldera magma and the
mantle wedge structure indicated by low Vp, and high
Vp/Vs values. The subsurface structure was delineated
from earthquakes recorded by the PASSCAL seismic
network consisting of 40 seismic stations (Fauzi et al.,
1996; Masturyono et al., 2001). The results of this study
were an update on previous studies that described sub-
surface structures using only P– waves (Masturyono et
al., 2001). Global seismic tomographic studies under the
Burma, Andaman, and Sumatra arcs have detected the
presence of slabs to a depth of 975 km beneath the island
of Sumatra to the Andaman (Pesicek et al., 2008). This
study has updated the previous global seismic tomogra-
phy studies by adding new data and using different
methods (Widiyantoro & van der Hilst, 1996, 1997).
Teleseismic tomography has succeeded in depicting the
subducted slab under the northern part of Sumatra Island
to a depth of 400 km and the subduction of the slab in the
southern part to 800 km (Liu et al., 2018). The study
utilized the BMKG seismic network. Joint inversion us-
ing regional earthquake and teleseismic data with the
same seismic networks has successfully described the
morphology of the slab under northern Sumatra (Liu et
al., 2019). An exciting feature of the study was the abil-
ity to explain the slab tear caused by the subduction of
the IFZ under the Toba Caldera at a depth of 120 km to
more than 400 km. The latest regional seismic tomogra-
phy study using data from the International Seismologi-
cal Centre (ISC) has determined the seismic velocity
structure of P– waves to a depth of 90 km (Osagie and
Abir, 2021).
This research focuses on analysing earthquakes in
and around Sumatra, by utilizing data recorded on the
local-regional seismic network operated by BMKG. The
study presents updated tomographic ndings of the P-
wave seismic velocity (Vp) beneath Sumatra and its sur-
rounding areas, using a more recent earthquake cata-
logue compared to previous studies. It is worth noting
that the previous BMKG earthquake catalogue has yet to
be employed for tomographic investigations in Sumatra.
However, the study did not incorporate tomographic
models of the S-wave seismic velocity (Vs). This deci-
Figure 1: The main tectonic settings in the study area consist
of the subduction zone, the Sumatran Fault Zone (SFZ), and
the Mentawai Fault Zone (MFZ) (modied from Irsyam et
al., 2017). In addition, Wharton Fossil Ridge (WFR) and
Investigator Fracture Zone (IFZ) are two tectonic features
found in the research area. The study area is in the red box in
the inset gure. Earthquake events numbering follows the
sequential chronological order. Based on the location, events
1 to 7 occurred on the subduction zone of Sumatra Island,
events 11 to 14 occurred on the mainland of Sumatra Island or
in the Sumatran Fault Zone (SFZ), and events 8 to 10
occurred in the WFR area. The plate movement rate is
modied from Natawidjaja, 2018. The red triangle shows the
volcano’s position on Sumatra Island and its surroundings
(modied from Malawani et al., 2021). The cross-sectional
lines A-A’ to I-I’ are the vertical section of the tomogram
directions that pass through Sumatra Island from north to
south. The source mechanism shown on the map is modied
from the CMT global catalogue (Ekström et al., 2012). The
topographic map in the image above uses GEBCO data
(modied from Weatherall et al., 2015).

121 Seismic imaging beneath Sumatra Island and its surroundings, Indonesia…
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
sion was attributed to the scarcity of S-wave arrival
times in the earthquake catalogue compared to the abun-
dance of P-wave arrivals. Consequently, developing S-
wave tomograms based on the existing earthquake cata-
logue would result in unreliable interpretations owing to
inadequate resolution. To overcome this challenge, fu-
ture seismic tomography studies that incorporate S-
waves should prioritize identifying and re-picking wave
phases. This process ensures that the time intervals be-
tween the two wave phases are not excessively distant,
as observed in the existing earthquake catalogue.
2. Data and Methods
This study uses the BMKG earthquake catalogue for
April 2009–December 2019 at coordinates 91.9º–108.2º
east longitude and 7.5º south–6.9º north latitude or in
Sumatra Island and its surroundings. The data used for
tomographic inversion is a catalogue that has relocated
its hypocentre parameters (Ramdhan et al., 2021). Eve-
ry earthquake event was recorded by at least six stations
so that seismic stations would accurately constrain the
event’s epicentre. The data used in this study is from
9,152 events recorded by 122 seismic stations on the is-
land of Sumatra and its surroundings. Figure 2 repre-
Figure 2: Grid conguration and hypocentre distribution
used for seismic tomographic inversion. The inverted yellow
triangle shows the seismic network that records the
earthquakes.
Table 1: Earthquake events and corresponding information analysed in this study
Event No. Magnitude Date Remarks Reference
1 Mw ~ 9.2 December 2004 Aceh-Andaman megathrust earthquake
in December 2004 (Meltzner et al., 2006)
2 Mw 8.6 March 2005 March 2005 Nias earthquake (Fujii et al., 2020)
3 Mw 8.4 September 2007 Bengkulu earthquake in September 17,
2007 (Ekström et al., 2012)
4 Mw 7.9 September 2007 Aftershock of Bengkulu earthquake (Ekström et al., 2012)
5 Mw 7.6 September 2009 Intraslab earthquake on the west coast
of Sumatra
(Earthquake Engineering
Research Institute (EERI), 2009)
6 Mw 7.8 April 2010 Earthquake near the Banyak islands
on April 7, 2010 (Haridhi et al., 2018)
7 Mw 7.8 October 2010 Megathrust earthquake around Sumatra
Island on October 25, 2010 (Ekström et al., 2012)
8 Mw 8.6 April 2012 Intraplate event near Wharton Fossil
Ridge (WFR) zone on April 11, 2012 (Ekström et al., 2012)
9 Mw 8.2 April 2012 Intraplate event near Wharton Fossil
Ridge (WFR) zone on April 11, 2012 (Ekström et al., 2012)
10 Mw 7.8 March 2016 Intraplate event near Wharton Fossil
Ridge (WFR) zone on March 2, 2016 (Ekström et al., 2012)
11 Mw 6.4 March 2007 Doublet earthquake on the Sumatran
Fault on March 6, 2007 (Nakano et al., 2010)
12 Mw 6.3 March 2007 Doublet earthquake on the Sumatran
Fault on March 6, 2007 (Nakano et al., 2010)
13 Mw 6.6 October 2009 Dikit segment earthquake in Bengkulu
Province on October 1, 2009 (Ekström et al., 2012)
14 Mw 6.5 December 2016 Pidie Jaya earthquake on December 6,
2016 (Muzli et al., 2018)
sents earthquakes the BMKG seismic network recorded,
and the grid distribution used for seismic tomographic
inversion.

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 122
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
Figure 3: The general inversion algorithm for simultaneous
inversion determination conducted in this study (modied
from Grandis, 2009).
Table 2: The three-dimensional grid spacing in the horizontal (X and Y) and vertical (Z) directions, along with their
corresponding distances from the center of grid utilized to delineate the seismic velocity structure beneath Sumatra Island
and its surroundings.
Grid
direction
Number
of grid Grid distances from the center of grid (km); Center of grid is at 100.5°E and 0.0°S
X 20 -1000 -900 -800 -700 -600 -500 -400 -300 -200 -100
0 100 200 300 400 500 600 700 800 900
Y 20 -1000 -900 -800 -700 -600 -500 -400 -300 -200 -100
0 100 200 300 400 500 600 700 800 900
Z 18 -10 0 10 20 30 40 50 60 80 100
120 150 180 200 250 300 350 400
As depicted in Figure 3, the general process of tomo-
graphic inversion involves the utilization of an initial
model m0 consisting of Vp, Vp/Vs ratios, and hypocentre
parameters. This study incorporated a pre-existing 1-D
velocity model with a Vp/Vs ratio of 1.73 (Kennett et
al., 1995; Wadati, 1933). The hypocentre parameters
were derived from the BMKG earthquake catalogue.
The Δd matrix represents the discrepancy between ob-
served travel times (tobs) and calculated travel times (tcal)
of seismic waves at individual seismic stations. The cal-
culation of tcal is accomplished through forward model-
ling with the aid of g(m). The matrix J, known as the
Jacobian matrix, encompasses the rst partial deriva-
tives of the calculated travel times with respect to veloc-
ity parameters (Vp and Vp/Vs) and hypocentre parame-
ters. The Δm matrix denotes the model perturbation ma-
trix for velocity and hypocentre parameters, facilitating
the iterative renement of previous model parameters
until the minimum error criterion is met, thus signifying
the attainment of the nal model parameters (m).
The SIMULPS12 code was employed for inversion
tomography, applying simultaneous inversion of the ve-
locity model parameters (Vp) and the Vp/Vs ratio, along
with the determination of hypocentres (Evans et al.,
1994; Thurber, 1993). The pseudo-bending method was
determined by ray tracing in the code (Um & Thurber,
1987). The algorithm has been successfully applied in
various parts of Indonesia to determine subsurface struc-
tures and physical properties at different scales (Af et
al., 2021; Ramdhan et al., 2019; Supendi et al., 2020).
The initial velocity model used for the inversion tomo-
graphic input applied the 1D AK135 model (Kennett
et al., 1995). A uniform grid size of 100 km was used
for the horizontal grid model. Table 2 reveals a varying
grid size ranging from 10 km to 50 km for the vertical
grid distance. Damping determination is crucial for de-
termining the optimal value during the tomographic
inversion process. This value was derived from the
trade-off curve depicted in Figure 4, which compared
the data with the variance model. A damping value 70
was obtained from the rst iteration of the tomographic
inversion.
Prior to conducting tomographic inversion, it was
crucial to perform resolution tests to assess areas or fea-
tures that can be effectively resolved using seismic data.
This study implemented a checkerboard resolution test
(CRT), incorporating positive and negative perturba-
tions of ±5% relative to the 1-D reference velocity mod-
el, which served as input for the tests. If the inversion
results of the synthetic model exhibit similar patterns of
negative or positive anomalies resembling the input per-
turbations, it could be inferred that those areas were suc-
cessfully resolved by the seismic data, even with some
magnitude reductions due to damping, which is a limita-
tion of the resolution test (Lévěque et al., 1993; Ramd-
han et al., 2019; Rawlinson & Spakman, 2016). This
study only used a P– wave tomogram, so it was not too
deep to discuss the physical properties of the rock, which
required a tomogram of the Vp/Vs ratio, which was lin-
ear with Poisson’s ratio. These parameters were suscep-
tible to changes in uid and temperature.

123 Seismic imaging beneath Sumatra Island and its surroundings, Indonesia…
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
3. Results and Discussion
As mentioned above, this study used 9,152 events to
image the subsurface structure beneath the research area.
The total phases consist of 89,292 P-waves and 28,509
S-waves. The analysis of the tomographic inversion re-
vealed promising results, as evidenced by the concentra-
tion of residual time ranging predominantly between -1
and 1 s, with a tendency towards approaching zero val-
ues (see Figure 5). However, despite focusing on P-
wave analysis, this study incorporated S-phases in the
simultaneous inversion process to provide additional
constraints on focal depth parameters, as suggested by
previous studies (Gomberg et al., 1990; Husen &
Hardebeck, 2010). Given the close interrelation be-
tween the resolution test and seismic tomography inver-
sion, the subsequent subchapter presents the seismic to-
mography results obtained from the horizontal and verti-
cal cross-sectional views. They provided a comprehensive
interpretation of the tectonic conditions in the research
area.
3.1. Resolution Tests
The effectiveness of synthetic model inversion, vali-
dated by the input tests shown in Figure 6a and Figure
6d, is prominently depicted in Figure 6b, Figure 6c,
Figure 6e, and Figure 6f, specically for depths within
the range of 20 to 50 km in the vicinity of Sumatra Is-
land. This notable performance enabled the reliable in-
terpretation of inversion outcomes at these particular
depth intervals. The results of the CRT performance at a
depth of 80-150 km and other depths shallower than 80
km can still be interpreted within the context of the geo-
logical structure within that depth range, as illustrated in
Supplementary Material Figure S2. The performance
was closely related to the different stations and earth-
quake distributions in each area of Sumatra Island. The
oblique cross-section from A-A’ to I-I’ (which did not
align with the west-east or north-south directions) did
not result in a clear positive-negative anomaly pattern
between the grids in the horizontal axis of the vertical
sections. Instead, the positive-negative anomaly pattern
was only observable in the vertical axis below the grids
(see Figure 7). Besides conducting CRT tests, synthetic
tests with a single negative anomaly block were also per-
formed in this study, measuring 400 x 400 km2. The test
results also demonstrate satisfactory results up to a depth
of 150 km, as shown in Supplementary Figure S3.
3.2. Vp Models
The number of tomograms expressed in terms of the
P– wave velocity perturbation value or relative to the
initial model was used as the inversion tomography in-
put. This value shows the geological structure or tecton-
ic conditions in the research area, represented by a nega-
tive anomaly (red) or a positive anomaly (blue). The
positive anomaly observed in the seismic tomography
results can be attributed to the subsurface structure,
which exhibits a higher velocity compared to the initial
model. This anomaly is associated with a high-density
structure, resulting in faster wave propagation within the
medium. Subduction slabs are notable examples of posi-
tive anomalies in seismic tomography studies (Liu et
al., 2019; Supendi et al., 2020). In contrast to the posi-
tive anomaly, a negative anomaly was observed, indicat-
ing its association with prominent geological features,
including faults, partial melting zones, magma reser-
voirs, uid-rich zones, and elevated temperatures.
Figure 4: Trade-o curve to get optimum damping. The
optimum damping value used is indicated by the red circle
of 70. After one iteration of inversion, the values for various
damping levels are represented by blue circles.
Figure 5: Residual time histogram of simultaneous inversion
results from total 117,801 of P and S-P phase

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 124
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
Figure 6: Checkerboard resolution test (CRT) Vp results for horizontal sections at a depth of 20, 30, 40, 50 km.
The CRT velocity model input for the top (a) corresponds to tomograms at depths of 20 km (b) and 40 km (c),
while the CRT velocity model input for the bottom (d) corresponds to tomograms at depths of 30 km (e) and 50 km (f).
Positive and negative perturbations of ±5% for input velocity model of CRT relative to the 1-D reference velocity model.
The velocity shown is relative to the AK135 1D model (Kennett et al., 1995).
The horizontal section shown in Figure 8 and Sup-
plementary Figure S4 shows a positive anomaly extend-
ing along Sumatra from 0 to 150 km depth, which was
indicated as a subduction slab of the Indo-Australian
Plate subducting under the Eurasian Plate. The positive
anomaly also moved northeast with increasing depth. A
dominant negative anomaly is observed in the horizontal
section at depths shallower than 30 km. This anomaly
was most likely associated with a series of volcanoes on
the island of Sumatra, known as the volcanic front line.
Apart from being caused by a volcanic arc, this anomaly
may also be related to the Sumatran Fault because the
two positions are close. In order to achieve a more com-
prehensive understanding of the tectonic framework sur-
rounding Sumatra Island, this study provides a series of
nine vertical sections spanning from the northern to the
southern regions of Sumatra. These sections are visual-
ized in Figure 9, Figure 10, and Figure 11, enhancing
the clarity of the tectonic picture.
The vertical tomogram from A–A’ to I–I’ shows that
the depth of the subduction slab becomes deeper towards
the south. This feature was related to the age of the slab,
which increased in this direction (Scotese et al., 1988).
This age relationship can be seen from the distribution of
earthquakes that occurred north of Sumatra at a depth of
less than 180 km (see Figure 2), where the subduction
slab plunged more and more to the south.
3.3. Northern Sumatra Tomographic Proles
(Section A to Section C)
The transverse lines A–A’ in Figure 9a are cross-sec-
tions that intersect the Sunda Trench, Kota Aceh, and
Sumatran Faults in the Seulimeum and Aceh segments,
as shown in Figure 1. Positive anomalies represented in
blue begin to appear from a depth of 30 km in the forearc
area to less than 150 km beneath the Sumatra Island arc,
considered a part of the Indo-Australian subduction slab
in the province of Aceh. This positive anomaly was in
line with the distribution of the earthquake hypocentre at
medium depth, which resembles the Benioff Zone. A
sizeable negative anomaly with dense earthquake distri-
bution at 0–20 km depth on the right was most likely
associated with the Sumatran Fault Zone in the Seulime-
um and Aceh segments and the 2018 Pidie Jaya earth-
quake zone. Owing to the proximity of the volcanic arc
and the Sumatran Fault, this strong negative anomaly
might also have been caused by the magma system un-
der the Seulawah Agam Volcano. The limited resolution
of seismic data causes two anomalies generated by the
Sumatran Fault, and the volcanic arc could not be sepa-

125 Seismic imaging beneath Sumatra Island and its surroundings, Indonesia…
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
Figure 7: Checkerboard resolution test (CRT) Vp results for vertical sections in the cross-section direction A–A’ to I–I’ (a-i).
The direction of the oblique/diagonal cross-section intersects the grid, displaying a CRT pattern that is not as distinct
as the CRT results from a horizontal cross-section. To visualize anomaly patterns more clearly, the color perturbation scale
is set to ± 1.5%.
Figure 8: Vp tomogram for a horizontal section at 10 km (a), 20 km (b), 30 km (c), 40 km (d), 50 km (e), and 60 km (f) depth.

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 126
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
rated too clearly. In this cross-section, a partial melting
zone was identied at a depth exceeding 100 km, exhib-
iting a negative anomaly that extended vertically to the
base of the Seulawah Agam Volcano. This zone served
as the magma source for the underlying magma reser-
voir. The 2018 Pidie Jaya earthquake (Event 14) oc-
curred between moderate and robust negative anomalies
in the contact zone. Negative anomalies are commonly
observed beneath fault zones and are associated with
stress concentrations, making these areas prone to earth-
quakes (Nugraha et al., 2013). The negative anomaly
above the subduction slab in the southern part of the
Seulawah Agam Volcano was likely linked to a mantle
wedge structure. Similar negative anomalies were also
identied above the subduction zone, specically south-
west of Toba Volcano, indicating the inuence of the IFZ
subduction (Koulakov et al., 2016).
Not much different from the A–A’ cross-section, posi-
tive anomaly imaged in the B–B’ and C–C’ cross-sec-
tions (see Figure 9b and Figure 9c) and the earthquake
hypocentre distribution in the Benioff Zone conrmed
the presence of the Indo-Australian Plate subduction
slab below the Eurasian Plate. The negative anomaly
with a dense earthquake distribution at 0–20 km depth
on the left side of these images was probably associated
with the West Andaman Fault Zone through which the
cross-section passes (Martin et al., 2014).
In the B-B’ section, the 2010 Aceh earthquake (event 6)
with a magnitude of 7.8 (Ekström et al., 2012) occurred,
featuring a hypocentre at a lesser depth than the 2004
great earthquake (event 1). However, when examining the
2004 Aceh-Andaman earthquake segment, it becomes ap-
parent that the 2010 Aceh earthquake occurred within that
very same segment (Ammon et al., 2005). Although it
caused strong level-V shocks on the MMI scale, the 2010
Aceh earthquake did not produce a tsunami (Reliefweb,
2010). The most plausible explanation is that the segment
had already entered an inter-seismic state, where the stress
accumulation had been released when the Aceh-Andaman
2004 earthquake.. The event emerged from shallower
crust destabilization in response to a previously large
earthquake. These factors make the submarine deforma-
tion energy insignicant to generate a tsunami. This event
also indicated that the segment was resetting its recur-
rence at large earthquake intervals.
The 2004 Aceh-Andaman earthquake (event 1) and
the 2005 Nias earthquake (event 2) occurred in the Indo-
Australian Plate interface zone that subducts to the Eura-
sian Plate, as shown in Figure 9c. The rupture zone or
segment of the two earthquakes was side-by-side (Am-
mon et al., 2005). Events 1 and 2 have the same pertur-
bation characteristics, as indicated by the high Vp pertur-
bation (see Figure 9c). This slight difference means the
event 1 Aceh-Andaman earthquake rupture did not prop-
agate towards Nias Island because it was below the
southern boundary of the rupture zone where there were
fold slabs in the upper and lower mantle transition zones,
indicating the megathrust segmentation in the island
(Pesicek et al., 2008). Based on Figure 9c, this section
also passed through the northern part of the Toba Cal-
dera, which had a negative perturbation anomaly at 25-
50 km depth associated with the basic magma reservoir
(Koulakov et al., 2016).
3.4. Central to Southern Sumatra Tomography
Proles (Section D to Section I)
The vertical section D–D’ in Figure 10a is a cross-sec-
tion that crosses Nias Island and the Sumatran Fault in the
Toru, Angkola, and Barumun segments. The vertical sec-
tion shows the accretionary prism complex on Nias Island
and the Sumatran Fault Zone at a depth of less than 20 km.
Figure 9: Vp tomographic inversion results for vertical
sections A–A’ (a), B–B’ (b), and C–C’ (c). The positions of the
earthquake hypocentre (small black circle), the volcano (red
triangle), and signicant earthquakes (yellow star) are all
located within a distance of 50 km from the vertical cross-
section line. The black line on the vertical tomogram
cross-section indicates the Slab 2.0 model
(Hayes et al., 2018).

127 Seismic imaging beneath Sumatra Island and its surroundings, Indonesia…
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
This region was associated with negative anomalies and
high seismicity. Barber et al., (2005) described a diagram
of the Sumatran subduction system showing Nias Island as
part of an accretionary prism zone. The accretionary prism
zone is characterized by the accumulation of uplifted sedi-
ments to produce new small islands. This vertical section
also showed a partial melting zone, indicated by a negative
anomaly as magma feeding for the Sorikmarapi Volcano.
The positive anomaly was associated with the slab widen-
ing to the east (deected) at a depth of 150 km. The anom-
aly extends away from the Benioff Zone curve and was
represented by the earthquake hypocentre. The same phe-
nomenon was also observed in the cross-section of E–E’,
as shown in Figure 10b. As indicated by the CRT test, the
CRT resulted in the slab area were valid, as shown in Sup-
plementary Figure S2e. However, further S–wave tomog-
raphy studies and other geophysical methods were re-
quired to conrm this hypothesis. The 2009 Padang earth-
quake (event 5) occurred in the intraslab zone and was
associated with positive anomalies, as seen in sections
E–E’. The earthquake was one of the largest intraslab
earthquakes ever recorded on the island of Sumatra after
2000 (Ekström et al., 2012). Doublet earthquakes (events
11 and 12) were observed on the cross-section associated
with a negative anomaly.
The vertical cross-section F–F’ shown in Figure 10c
was a cross-section that cut through the Sunda Trench,
Figure 10: Vp tomographic inversion results for vertical
sections D–D’ (a), E–E’ (b), and F–F’(c). The positions of the
earthquake hypocentre (small black circle), the volcano
(red triangle), and signicant earthquakes (yellow star) are
all located within a distance of 50 km from the vertical
cross-section line. The black line on the vertical tomogram
cross-section indicates the Slab 2.0 model
(Hayes et al., 2018).
Figure 11: Vp tomographic inversion results for vertical
sections G–G’ (a), H–H’ (b), and I–I’(c). The positions of the
earthquake hypocentre (small black circle), the volcano (red
triangle), and signicant earthquakes (yellow star) are all
located within a distance of 50 km from the vertical cross-
section line. The black line on the vertical tomogram
cross-section indicates the Slab 2.0 model
(Hayes et al., 2018).

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 128
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
Mentawai Fault, and Sumatran Fault in the Dikit seg-
ment, in which one of the signicant earthquakes in the
Sumatran Fault occurred (event 13). The cross-sectional
tomogram in this section showed anomalous variations
at a depth of 0–40 km, a dense collection of earthquakes
occupying the negative anomaly area, which was an ac-
cretionary prism complex and weak zone in the Menta-
wai and Sumatran Faults in the Dikit segment. Two sig-
nicant earthquake events, namely, the 2007 Bengkulu
aftershock (event 4) and the 2010 Mentawai earthquake
(event 7), occurred within the zone associated with the
positive anomaly. Earthquakes within the oceanic slab
commonly manifest as positive anomalies or regions
with high Vp values attributed to the high density of the
slab. This phenomenon was evident in the 2018 Lombok
earthquake series, consisting of an event of Mw 6.4
event on July 29, 2018, and an Mw 7.0 event on August
5, 2018. Both earthquakes occurred within the Flores
Oceanic Crust, which exhibited a notably high Vp anom-
aly (Af et al., 2021).
The vertical cross-sections G–G’ and H–H’ in Figure
11a and Figure 11b intersected the Sunda Trench, Men-
tawai Fault, and Sumatran Fault in the Ketaun and Musi
segments. The slab slope of the two vertical sections was
sharper than that of the northern section owing to the age
of the slab, which was aging in the south direction (Sco-
tese et al., 1988). The 2007 Bengkulu earthquake (event
3) occurred in the interface zone, represented in vertical
section G–G’. Rupture modelling based on GPS data
showed a large slip under the archipelagic belt and shal-
low waters; therefore, the earthquake did not cause a sig-
nicant tsunami (Ambikapathy et al., 2010). The cross-
sectional tomogram in this section showed a positive
anomaly aligned with the subduction slab, represented
by an earthquake in the Benioff Zone. The slab model on
the G–G’ transverse line appeared to be a negative
anomaly under the forearc. This negative anomaly was
thought to be related to slab dehydration. Subduction
slabs and oceanic crust carried away large amounts of
seawater in the pores and hydro minerals. As a result of
the increase in temperature and pressure caused dehy-
dration (release of water content) towards the crust layer
above it. At shallow depths, water was expelled by sub-
ducting sediment compaction and loss of porosity in the
upper oceanic crust. At a depth of approximately 100
km, this uid could cause a partial melting phenomenon
that causes the formation of a volcanic arc (Tatsumi,
1989). Most of the uid is released under the forearc,
and some is released under the back-arc. The presence of
these hydro minerals caused a decrease in seismic veloc-
ity such that the anomaly value becomes negative
(Hyndman & Peacock, 2003). The transverse line I–I’
in Figure 11c is a vertical section that intersects the Sun-
da Trench, the Mentawai Fault, and the Mount Krakatau
complex. In this section, the negative anomaly in the
forearc section was related to slab dehydration and be-
neath the Mount Krakatau complex, which was related
to the magma reservoir beneath the mountain complex.
This negative anomaly could also be related to the Su-
matran Fault in the Sunda segment through which this
section passes.
The study ndings allowed for the depiction of a tec-
tonic model showing the geological features of Sumatra
Island and its surrounding region, as depicted in Figure
12. The chromatic gradient from orange to red corre-
sponds to low-velocity anomalies, whereas the gradient
from green to blue signies high-velocity anomalies.
Low-velocity anomalies are associated with the volcanic
arc and Sumatran Fault Zone, while high-velocity anom-
alies characterize the subduction zone, with deeper re-
gions displaying increasing blue hues indicative of ele-
vated velocity and density values within the slab.
Figure 12: The tectonic
framework of Sumatra
Island and its adjacent
areas discerned from the
analysis of nine vertical
cross-sections of P-wave
tomography

129 Seismic imaging beneath Sumatra Island and its surroundings, Indonesia…
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
4. Conclusions
The successful implementation of the simultaneous in-
version method in this study has led to the development of
a novel local-regional seismic tomography model for P-
waves utilizing the BMKG seismic network. Comprehen-
sive mapping of the slab geometry from northern to south-
ern Sumatra revealed distinct patterns down to a depth of
150 km, with a notable increase in a slope towards the
southern regions attributed to the inuence of the older
slab. Furthermore, the analysis has effectively captured
key geological features, including the partial melting
zone, mantle wedge structures, volcanic arcs, and Suma-
tran Fault Zone. The spatial correlation observed between
signicant earthquakes and their respective sources high-
lights the positive anomalies associated with subduction
zone events and negative anomalies linked to earthquakes
originating from the Sumatra Fault Zone. These ndings
emphasize the signicant role of the BMKG seismic net-
work in facilitating comprehensive local-regional seismic
tomography studies and enabling the identication of pre-
viously undetected features. To further enhance the under-
standing, future studies should prioritize the repicking of
the S-wave phase to obtain Vp/Vs tomograms, enabling a
more accurate determination of the physical properties of
the rocks.
Acknowledgement
The National Research and Innovation Agency (BRIN)
- Research Center for Geological Disaster is sincerely ac-
knowledged for their valuable support, while the Center
for Earthquakes and Tsunami (PGT) of the Indonesian
Agency for Meteorology, Climatology, and Geophysics
(BMKG) are gratefully acknowledged for providing the
arrival time of the earthquake catalogue, greatly contrib-
uting to this research. The study received support from
Komite Kajian Gempabumi dan Tsunami 2021. Most g-
ures in this study were plotted using the Generic Mapping
Tools (GMT) program (Wessel and Smith, 1998).
Data availability
Raw data were generated at the Center for Earth-
quakes and Tsunami (PGT), BMKG. Derived data sup-
porting the ndings of this study are available from the
corresponding author MR on request.
5. References
Papers:
Af, H., Nugraha, A. D., Muzli, M., Widiyantoro, S., Zulfakri-
za, Z., Wei, S., Sahara, D. P., Riyanto, A., Greeneld, T.,
Puspito, N. T., Priyono, A., Sasmi, A. T., Supendi, P., Ardi-
anto, A., Syahbana, D. K., Rosalia, S., Cipta, A., & Husni,
Y. M. (2021). Local earthquake tomography of the source
region of the 2018 Lombok earthquake sequence, Indone-
sia. Geophysical Journal International, 226(3), 1814–
1823. https://doi.org/10.1093/gji/ggab189
Ambikapathy, A., Catherine, J., Gahalaut, V., Narsaiah, M.,
Bansal, A., & Mahesh, P. (2010). The 2007 Bengkulu
earthquake, its rupture model and implications for seismic
hazard. Journal of Earth System Science, 119(4), 553–560.
https://doi.org/10.1007/s12040-010-0037-2
Ammon, C. J., Ji, C., Thio, H.-K., Robinson, D., Ni, S., Hjor-
leifsdottir, V., Kanamori, H., Lay, T., Das, S., Helmberger,
D., & others. (2005). Rupture process of the 2004 Suma-
tra-Andaman earthquake. Science, 308(5725), 1133–1139.
https://doi.org/10.1186/BF03353379
Barber, A. J., Crow, M. J., & Milsom, J. S. (2005). Sumatra:
Geology, Resources and Tectonic Evolution. Geologi-
cal Society of London. https://doi.org/10.1144/GSL.
MEM.2005.031
DeMets, C., Gordon, R. G., & Argus, D. F. (2010). Geologi-
cally current plate motions. Geophysical Journal Interna-
tional, 181(1), 1–80. https://doi.org/10.1111/j.1365-246X.
2009.04491.x
Earthquake Engineering Research Institute (EERI). (2009).
Learning from earthquakes the Mw 7.6 Western Sumatra
earthquake of September 30, 2009. EERI Special Earth-
quake Report, 2009, 1–12.
Ekström, G., Nettles, M., & Dziewoński, A. (2012). The global
CMT project 2004–2010: Centroid-moment tensors for
13,017 earthquakes. Physics of the Earth and Planetary In-
teriors, 200, 1–9. https://doi.org/10.1016/j.pepi.2012.04.002
Fauzi, McCaffrey, R., Wark, D., Sunaryo, & Haryadi, P. Y. P.
(1996). Lateral variation in slab orientation beneath Toba
Caldera, northern Sumatra. Geophysical Research Letters,
23(5), 443–446. https://doi.org/10.1029/96GL00381
Fujii, Y., Satake, K., Watada, S., & Ho, T.-C. (2020). Slip dis-
tribution of the 2005 Nias earthquake (M w 8.6) inferred
from geodetic and far-eld tsunami data. Geophysical
Journal International, 223(2), 1162–1171. https://doi.
org/10.1093/gji/ggaa384
Gomberg, J. S., Shedlock, K. M., & Roecker, S. W. (1990).
The effect of S-wave arrival times on the accuracy of hy-
pocenter estimation. Bulletin of the Seismological Society
of America, 80(6A), 1605–1628.
Haridhi, H. A., Huang, B.-S., Kuo-Liang, W., Denzema, D.,
Prasetyo, R. A., & Chao-Shing, L. (2018). A study of large
earthquake sequences in the Sumatra subduction zone and
its possible implications. TAO: Terrestrial, Atmospheric
and Oceanic Sciences, 29(6), 6. https://doi.org/ 10.3319/
TAO.2018.08.22.01
Hayes, G. P., Moore, G. L., Portner, D. E., Hearne, M.,
Flamme, H., Furtney, M., & Smoczyk, G. M. (2018).
Slab2, a comprehensive subduction zone geometry model.
Science, 362(6410), 58–61. https://doi.org/10.1126/sci-
ence.aat4723
Husen, S., & Hardebeck, J. (2010). Earthquake location accu-
racy. CORSSA. https://doi.org/10.5078/corssa-55815573
Hyndman, R. D., & Peacock, S. M. (2003). Serpentinization of
the forearc mantle. Earth and Planetary Science Letters,
212(3–4), 417–432. https://doi.org/10.1016/S0012-821X
(03)00263-2

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 130
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
Kennett, B. L. N., Engdahl, E. R., & Buland, R. (1995). Con-
straints on seismic velocities in the Earth from traveltimes.
Geophysical Journal International, 122(1), 108–124. htt-
ps://doi.org/10.1111/j.1365-246X.1995.tb03540.x
Koulakov, I., Kasatkina, E., Shapiro, N. M., Jaupart, C., Vasi-
levsky, A., El Khrepy, S., Al-Ari, N., & Smirnov, S.
(2016). The feeder system of the Toba supervolcano from
the slab to the shallow reservoir. Nature Communications,
7(1), 1–12. https://doi.org/10.1038/ncomms12228
Koulakov, I., Yudistira, T., & Luehr, B.-G. (2009). P, S veloc-
ity and VP/VS ratio beneath the Toba caldera complex
(Northern Sumatra) from local earthquake tomography.
Geophysical Journal International, 177(3), 1121–1139.
https://doi.org/10.1111/j.1365-246X.2009.04114.x
Lévěque, J.-J., Rivera, L., & Wittlinger, G. (1993). On the use
of the checker-board test to assess the resolution of to-
mographic inversions. Geophysical Journal Internatio-
nal, 115(1), 313–318. https://doi.org/10.1111/j.1365-246X.
1993.tb05605.x
Liu, S., Suardi, I., Yang, D., Wei, S., & Tong, P. (2018). Tel-
eseismic traveltime tomography of northern Sumatra. Ge-
ophysical Research Letters, 45(24), 13–231. https://doi.
org/10.1029/2018GL078610
Liu, S., Suardi, I., Zheng, M., Yang, D., Huang, X., & Tong, P.
(2019). Slab morphology beneath northern Sumatra re-
vealed by regional and teleseismic traveltime tomography.
Journal of Geophysical Research: Solid Earth, 124(10),
10544–10564. https://doi.org/10.1029/2019JB017625
Malawani, M. N., Lavigne, F., Gomez, C., Mutaqin, B. W., &
Hadmoko, D. S. (2021). Review of Local and Global Im-
pacts of Volcanic Eruptions and Disaster Management
Practices: The Indonesian Example. Geosciences, 11 (3),
Article 3. https://doi.org/10.3390/geosciences11030109
Martin, K. M., Gulick, S. P. S., Austin Jr., J. A., Berglar, K.,
Franke, D., & Udrekh. (2014). The West Andaman Fault:
A complex strain-partitioning boundary at the seaward
edge of the Aceh Basin, offshore Sumatra. Tectonics,
33(5), 786–806. https://doi.org/10.1002/2013TC003475
Masturyono, McCaffrey, R., Wark, D. A., Roecker, S. W., Fau-
zi, Ibrahim, G., & Sukhyar. (2001). Distribution of magma
beneath the Toba caldera complex, north Sumatra, Indone-
sia, constrained by three-dimensional P wave velocities,
seismicity, and gravity data. Geochemistry, Geophysics,
Geosystems, 2(4). https://doi.org/10.1029/2000GC000096
Meltzner, A. J., Sieh, K., Abrams, M., Agnew, D. C., Hudnut, K.
W., Avouac, J., & Natawidjaja, D. H. (2006). Uplift and sub-
sidence associated with the great Aceh-Andaman earth-
quake of 2004. Journal of Geophysical Research: Solid
Earth, 111(B2). https://doi.org/10.1029/2005JB003891
Mukti, M. M., Maulin, H. B., & Permana, H. (2021). Growth
of forearc highs and basins in the oblique Sumatra subduc-
tion system. Petroleum Exploration and Development,
48(3), 683–692. https://doi.org/10.1016/S1876-3804(21)
60054-X
Muzli, M., Umar, M., Nugraha, A. D., Bradley, K. E., Widi-
yantoro, S., Erbas, K., Jousset, P., Rohadi, S., Nurdin, I., &
Wei, S. (2018). The 2016 Mw 6.5 Pidie Jaya, Aceh, North
Sumatra, earthquake: Reactivation of an unidentied sinis-
tral fault in a region of distributed deformation. Seismo-
logical Research Letters, 89(5), 1761–1772. https://doi.
org/10.1785/0220180068
Nakano, M., Kumagai, H., Toda, S., Ando, R., Yamashina, T.,
Inoue, H., & Sunarjo. (2010). Source model of an earth-
quake doublet that occurred in a pull-apart basin along the
Sumatran fault, Indonesia. Geophysical Journal Interna-
tional, 181(1), 141–153. https://10.1111/j.1365-246X.2010
.04511.x
Natawidjaja, D. H. (2018). Updating active fault maps and
sliprates along the Sumatran Fault Zone, Indonesia.
118 (1), 012001. https://doi.org/10.1088/1755-1315/118/
1/012001
Nugraha, A. D., Ohmi, S., Mori, J., & Shibutani, T. (2013).
High resolution seismic velocity structure around the
Yamasaki fault zone of southwest Japan as revealed from
travel-time tomography. Earth, Planets and Space, 65(8),
871–881. https://doi.org/10.5047/eps.2012.12.004
Osagie, A. U., & Ahmad Abir, I. (2021). Seismic tomographic
imaging of P wave velocity perturbation beneath Sumatra,
Java, Malacca Strait, Peninsular Malaysia and Singapore.
Journal of Earth System Science, 130(1), 1–13. https://doi.
org/10.1007/s12040-020-01530-w
Pesicek, J. D., Thurber, C. H., Widiyantoro, S., Engdahl, E. R.,
& DeShon, H. R. (2008). Complex slab subduction be-
neath northern Sumatra. Geophysical Research Letters,
35(20). https://doi.org/10.1029/2008GL035262
Ramdhan, M., Widiyantoro, S., Nugraha, A. D., Métaxian, J.-
P., Rawlinson, N., Saepuloh, A., Kristyawan, S., Sembir-
ing, A. S., Budi-Santoso, A., Laurin, A., & Fahmi, A. A.
(2019). Detailed seismic imaging of Merapi volcano, In-
donesia, from local earthquake travel-time tomography.
Journal of Asian Earth Sciences, 177, 134–145. https://
doi.org/10.1016/j.jseaes.2019.03.018
Rawlinson, N., & Spakman, W. (2016). On the use of sensitiv-
ity tests in seismic tomography. Geophysical Journal In-
ternational, 205(2), 1221–1243. https://doi.org/10.1093/
gji/ggw084
Scotese, C. R., Gahagan, L. M., & Larson, R. L. (1988). Plate
tectonic reconstructions of the Cretaceous and Cenozoic
ocean basins. Tectonophysics, 155(1–4), 27–48. http://doi.
org/10.1016/0040-1951(88)90259-4
Sieh, K., & Natawidjaja, D. (2000). Neotectonics of the Suma-
tran fault, Indonesia. Journal of Geophysical Research:
Solid Earth, 105(B12), 28295–28326. https://doi.org/
10.1029/2000JB900120
Supendi, P., Nugraha, A. D., Widiyantoro, S., Abdullah, C. I.,
Rawlinson, N., Cummins, P. R., Harris, C. W., Roosmawa-
ti, N., & Miller, M. S. (2020). Fate of Forearc Lithosphere
at Arc-Continent Collision Zones: Evidence From Local
Earthquake Tomography of the Sunda-Banda Arc Transi-
tion, Indonesia. Geophysical Research Letters, 47(6),
e2019GL086472. https://doi.org/10.1029/2019GL086472
Tatsumi, Y. (1989). Migration of uid phases and genesis of
basalt magmas in subduction zones. Journal of Geophysi-
cal Research: Solid Earth, 94(B4), 4697–4707. https://doi.
org/10.1029/JB094iB04p04697

131 Seismic imaging beneath Sumatra Island and its surroundings, Indonesia…
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
Um, J., & Thurber, C. (1987). A fast algorithm for two-point
seismic ray tracing. Bulletin of the Seismological Society
of America, 77(3), 972–986. https://doi.org/10.1785/BSSA
0770030972
Wadati, K. (1933). On the travel time of earthquake waves
(Part I). Geophys. Mag., 7, 87–99. https://cir.nii.ac.jp/crid
/1571417124417426560
Weatherall, P., Marks, K. M., Jakobsson, M., Schmitt, T., Tani,
S., Arndt, J. E., Rovere, M., Chayes, D., Ferrini, V., &
Wigley, R. (2015). A new digital bathymetric model of the
world’s oceans. Earth and Space Science, 2(8), 331–345.
https://doi.org/10.1002/2015EA000107
Wessel, P., & Smith, W. H. (1998). New, improved version of
Generic Mapping Tools released. Eos, Transactions Amer-
ican Geophysical Union, 79(47), 579-579. http://doi.org
/10.1029/98EO00426
Widiyantoro, S., & van der Hilst, R. (1996). Structure and evo-
lution of lithospheric slab beneath the Sunda arc, Indone-
sia. Science, 271(5255), 1566. https://doi.org/10.1126/sci-
ence.271.5255.1566
Widiyantoro, S., & van der Hilst, R. (1997). Mantle structure
beneath Indonesia inferred from high-resolution tomo-
graphic imaging. Geophysical Journal International,
130(1), 167–182. https://doi.org/10.1111/j.1365-246X.
1997.tb00996.x
Chapter in book:
Thurber, C. H. (1993). Local earthquake tomography: Veloci-
ties and Vp/Vs-theory. In H. M. Iyer & K. Hirahara (Eds.),
Seismic Tomography: Theory and practice (pp. 563--583).
Chapman and Hall.
Books:
Evans, J. R., Eberhart-Phillips, D., & Thurber, C. H. (1994).
User’s manual for SIMULPS12 for imaging vp and vp/vs;
a derivative of the ‘Thurber’ tomographic inversion
SIMUL3 for local earthquakes and explosions (Report No.
94–431; Open-File Report). USGS Publications Ware-
house. https://doi.org/10.3133/ofr94431
Grandis, H. (2009). Pengantar pemodelan inversi geosika.
Himpunan Ahli Geosika Indonesia (HAGI) (in Indo-
nesian)
Irsyam, M., Widiyantoro, S., Natawidjaja, D., Meilano, I.,
Rudyanto, A., Hidayati, S., Triyoso, W., Hanifa, N., Djar-
wadi, D., Faizal, L., & others. (2017). Peta sumber dan
bahaya gempa Indonesia tahun 2017. Pusat Penelitian
Dan Pengembangan Perumahan Dan Permukiman, Ke-
menterian Pekerjaan Umum Dan Perumahan Rakyat (in
Indonesian).
Ramdhan, M., Priyobudi, P., Imananta, R. T., Muzli, M., Su-
pendi, P., Yusuf Hadi Perdana, Nugraha, J., Jatnika, J., Ali,
Y. H., Panjaitan, A. L., Nugraha, M. F., Kristyawan, S.,
Sembiring, A. S., Setyahagi, A. R., & Yogaswara, D. S.
(2021). Katalog Gempabumi Indonesia: Relokasi Hi-
posenter dan Implikasi Tektonik (1st ed., Vol. 1). Bidang
Informasi Gempabumi dan Peringatan Dini Tsunami
(BMKG) (in Indonesian)
Internet source:
Reliefweb. (2010, April 20). M7.8 Northern Sumatra, Indone-
sia, Earthquake of 6 April 2010—Indonesia | ReliefWeb.
Reliefweb. URL: https://reliefweb.int/map/indonesia/
m78-northern-sumatra-indonesia-earthquake-6-april-2010
(accessed: 1st April 2023)

Pranata, B.; Ramdhan M.; Hanif, M.; Sulaiman, M.I.; Maulana, M. P.; Wandono; et al. 132
Copyright held(s) by author(s), publishing rights belongs to publisher, pp. 119-132, DOI: 10.17794/rgn.2023.3.10
SAŽETAK
Seizmičko modeliranje na području otoka Sumatre i njegove okolice, Indonezija,
pomoću P-valne seizmičke tomograje lokalnih i regionalnih potresa
Otok Sumatra i njegova okolica, Indonezija, jedno su od najaktivnijih tektonskih područja na svijetu. Tamo se dogodio
potres Aceh-Andaman, jedan od najrazornijih potresa na svijetu. Privukao je mnoge znanstvenike koji su u svojim istra-
živanjima primijenili različite metode, uključujući seizmičku tomograju, kako bi razumjeli podzemnu strukturu i
tektonski sustav otoka. Ova studija prva je koja prikazuje model podzemlja ispod otoka i njegove okolice koristeći se
lokalno-regionalnim katalogom potresa iz seizmičke mreže Indonezijske agencije za meteorologiju, klimatologiju i
geoziku (BMKG). Tomografski model brzine P-valova (Vp), uspješno je razgraničio subduciranu ploču (velika brzina
P-valova), zonu djelomičnoga taljenja (mali Vp), vulkanski luk (mali Vp) i rasjedne zone Sumatre (mali Vp). Odnos
između subdukcijske zone i vulkanskoga luka na otoku može se vidjeti na nekoliko vertikalnih presjeka gdje se na dubi-
ni od oko 100 km javlja zona djelomičnoga taljenja koja služi kao izvor magme za neke vulkane na otoku. Model oceanske
subducirane ploče također pokazuje izraženiji i strmiji nagib prema južnim regijama otoka Sumatre, što se vjerojatno
može pripisati procesu starenja ploče u tome smjeru. Rezultati naglašavaju važnost BMKG seizmičke mreže u identi-
kaciji lokalno-regionalnih podzemnih struktura ispod indonezijskoga arhipelaga, posebno za glavne otoke kao što je
Sumatra.
Ključne riječi:
Sumatra, BMKG, P-val, tomograja, subducirana ploča, rasjed
Author’s contribution
Bayu Pranata (1) (Dr.), Mohamad Ramdhan (2) (Dr.), Muhammad Hanif (3) (M.Sc) carried out tomography analyses,
conceptualization, processing, and writing the manuscript. Muhammad Iqbal Sulaiman (4) (B.Sc) performed tomo-
graphic inversion. Mufti Putra Maulana (5) (B.Sc) conducted data visualization. Wandono (6) (Dr.), Sri Widiyantoro
(7) (Prof. Dr.), Sandy Kurniawan Suhardja (8) (Dr.) & Edi Hidayat (9) (Dr.) contributed to the manuscript editing and
interpretation of the results. Pepen Supendi (10) (Dr.) & Ridwan Kusnandar (11) (M.Sc) conducted data collection
(resources). Wiko Setyonegoro (12) (M.Com) examined the results and manuscript editing.