Tectonophysics 751 (2019) 180–191

Contents lists available at ScienceDirect

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

Earthquake Geology of the Lembang Fault, West Java, Indonesia T

a,b,⁎ a b c
Mudrik R. Daryono , Danny H. Natawidjaja , Benjamin Sapiie , Phil Cummins
a
RC for Geotechnology, Indonesian Institutes of Sciences (LIPI), Indonesia
b
Institute of Technology Bandung (ITB), Indonesia
c
Australia National University (ANU), Indonesia

ARTICLE INFO ABSTRACT

Keywords: The Lembang Fault is a major fault in western Java that skirts the northern edge of Bandung, one of Indonesia's
Lembang Fault largest cities, just south of the active Tangkuban Perahu volcano. Although it has no recorded or historical large
Active faults earthquakes, the Lembang Fault shows obvious geomorphic evidence of recent activity and has long been
Slip rate thought to be active. In this study, we use geomorphic analysis to unequivocally establish that the fault has a
Paleoseismological trenching
dominantly sinistral sense of movement with a slip rate of 1.95–3.45 mm/yr. This proves that the fault is ac-
LIDAR
IFSAR
commodating trench parallel slip resulted from a slight obliquity in plate convergence at the JavaTrench. With a
length of 29 km, this suggests that the Lembang Fault could produce a Mw 6.5–7.0 earthquake with a recurrence
time of 170–670 years. We also conducted paleoseismological trenching of the Lembang Fault and found evi-
dence for at least 3 earthquakes in the 15th century, 2300–60 BCE and 19,620–19,140 BP. The 2300–60 BCE
earthquake had a measurable vertical displacement of 40 cm, which is consistence with a Mw 6.5 earthquake.
This is the first mapping of a source of crustal earthquakes in Java, Indonesia, the world's most densely popu-
lated island in one of its most tectonically active areas. The Lembang Fault and other faults in Java are likely to
pose substantial risk to not only Bandung but many of Java's major urban agglomerations.

1. Introduction Although there have been several large damaging earthquakes in

the historical record of western Java (Nguyen et al., 2015), none of
The island of Java, Indonesia, lies along the northern edge of the them have been conclusively associated with an inland active fault. The
Java subduction zone, where the Indo-Australian Plate has been sub- only major damaging earthquake that was instrumentally recorded and
ducting beneath Sundaland, at the Eurasian Plate's southern margin, was clearly associated with an inland fault was the 2006 Mw 6.4 2006
since 45 Ma (Hall et al., 2011) (Fig. 1). In Sumatra, the oceanic plate Yogyakarta earthquake, which killed over 5000 people. Java includes
obliquely converges along the Sunda trench, resulting in the formation many urban population centers whose populations grew rapidly in the
of a major strike-slip fault system known as the Great Sumatran Fault 20th century and are now much larger than Yogyakarta, including Ja-
that accommodates trench parallel slip (Bradley et al., 2017; McCaffrey, karta, Surabaya, Bandung and Semarang. Understanding the sources of
1991; Sieh and Natawidjaja, 2000; Yeats et al., 1997). Along the Java earthquakes that may threaten these major population centers is critical
trench, the oceanic plate converges almost perpendicular to the trench, to reduce future earthquake fatalities through development and appli-
with relative motion of about 67 mm/year (Simons et al., 2007). This cation of appropriate building codes.
normal convergence is thought to have resulted in a roughly north- The Lembang Fault is one potential earthquake source that is lo-
south directed maximum horizontal compressive stress that controls cated in the middle of the province of West Java, only about 10 km
active faulting in Java (Simandjuntak and Barber, 1996). However, north of West Java's capital Bandung, a city inhibited by 8.6 million
both McCaffrey's (1991) analysis of earthquake slip vectors as well as people (Barat, 2011). This fault zone is visible as a prominent landmark
the more recent analysis of GPS measurements of crustal strain by of slope breaks between a series of east-west-trending linear ridges that
Koulali et al. (2017) show that there is enough obliquity in convergence separate the north Bandung highland from a wide and flat Lembang
at the Java trench to require some accommodation of left-lateral motion basin farther to the north. The Lembang basin is filled with young al-
by crustal faults. Candidates for this accommodation are the Cimandiri luvial and Quaternary volcanic products along the front of the southern
and the Lembang faults in West Java (Abidin et al., 2009; Van flank of the active Tangkuban Perahu volcano (Marjiyono et al., 2008;
Bemmelen, 1949). This paper will focus on the Lembang Fault. Silitonga, 1973; Tjia, 1968; Van Bemmelen, 1949)(Fig. 2). The linear

⁎
Corresponding author at: RC for Geotechnology, Indonesian Institutes of Sciences (LIPI), Indonesia.
E-mail address: mudr001@lipi.go.id (M.R. Daryono).

https://doi.org/10.1016/j.tecto.2018.12.014
Received 4 July 2018; Received in revised form 1 December 2018; Accepted 13 December 2018
Available online 17 December 2018
0040-1951/ © 2018 Elsevier B.V. All rights reserved.
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 1. The location of the Lembang Fault is in the middle of west Java accommodating a left-lateral component of the slightly oblique plate convergence with
relative motion of about 67 mm/yr (Simons et al., 2007). The Cimandiri fault in the southwest and the Baribis fault in the north are predominantly reverse fault
systems.

Fig. 2. Recent seismicity of the Lembang Fault. July 22 and August 28, 2011 events (white stars) and relocated small earthquake events (magnitude less than 3 – red
stars) (Afnimar et al., 2015; Sulaeman and Hidayati, 2011; Sulaeman, 2011). According to Dam (1994), the Lembang Fault is divided into Eastern (red thick line) and
Western (red thin line) fault segments. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

fault ridges consist of various Quaternary volcanic rocks (Silitonga, Meteorologi, Klimatologi dan Geofisika) and Indonesia's Geological
1973); hence, the Lembang Fault is clearly a Quaternary fault that is Agency recorded small earthquake events on July 22, 2011 and August
potentially active. 28, 2011 (Fig. 2). Afnimar et al. (2015) have re-analyzed the hypo-
The Lembang Fault has never been properly mapped and studied in centers and focal mechanisms of these small earthquakes, showing that
detail, thus its kinematics and activity are still poorly understood. This they are aligned with the Lembang Fault.
study aims to map the Lembang Fault in detail to elucidate its geometry The July 22, 2011 earthquake at 05:46 had a magnitude of 3.4 and a
and style of faulting, measure offsets, estimate its slip rate, and find depth of 6 km, while the August 28, 2011 earthquake had a magnitude
evidence of prehistoric earthquake activity from trenching surveys. 3.4 with 1.45 km depth (Sulaeman and Hidayati, 2011). The latter
earthquake caused damage to surrounding buildings. Intensity survey
2. Seismic activity results indicated that the damaged area was located in Jambudipa and
Muril Villages, Ngamprah District, West Bandung Regency (Sulaeman,
Seismicity around the Lembang fault has not been documented in 2011). This coincides with relocated earthquake epicenters of the 2011
global earthquake catalogs, such as the EHDL catalog (Engdahl et al., event (Madrinovella et al., 2012).
2007) and the United States Geological Survey (USGS) catalog. How-
ever, local seismic networks owned and operated by BMKG (Badan

181
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

3. Previous studies of the Lembang Fault gradually disappears eastward. Thus, the fault terminations to the west
and east are characterized by the gradual disappearance of fault-topo-
The Lembang Fault appears in previously published literature graphic expressions. The Lembang fault is not a simple, single fault line
(Marjiyono et al., 2008; Silitonga, 1973; Sutoyo and Hadisantono, but is broken into 6 distinct sections (Fig. 3), namely: 1) the Cimeta
1992) only in small-scale maps, so its kinematics and activity are poorly strike-slip fault section from km 0 to km 6 with strike N80°E, 2) the
constrained. Tjia (1968) mapped the fault to have a total length of Cipogor Section, which has a visible fault scarp (FS) from km 5 to km 11
about 22 km with left-lateral strike-slip movement, based on his inter- with change in strike from N80°E to N100°E, 3) the Cihideung Section,
pretation of major stream offsets. Other studies, however, suggested with an oblique fault expression from km 10 to km 16.5 with strike
that the Lembang fault is a normal fault (Hidayat et al., 2008; Van N100°E, 4) the Gunung Batu Section, with a right step-over and ex-
Bemmelen, 1949). Dam (1994) divided the Lembang fault into eastern pression of a monoclinal structure from km 16.5 to km 21.5, 5) the
and western sections, the 10-km-long Eastern Lembang Fault is inter- Cikapundung Section, with a strike-slip fault expression from km 21.5
preted as a normal fault, formed by gravitational collapse that is related to km 29 and change in strike from N100°E to N90°E, and 6) the Batu
to volcanic eruptions during 126 and 135 ka (Dam, 1994; Van Lonceng Section, with a strike-slip fault expression of strike N120°E (km
Bemmelen, 1949) (see Fig. 2), while the Western Lembang Fault has 25–km 29). Detailed descriptions of these sections are given in Sup-
been thought to have an indication of movement since around 24 ka plementary Information S1 to S6. These fault sections, however, do not
(Dam et al., 1996). Djedi S. Widarto (Personal communication) sug- necessarily limit rupture propagation along the fault, since their se-
gested that the Lembang Fault is a reverse fault, based on his 2D gravity parations are only on the order of a few hundred meters, much less than
survey and modeling. More recent analysis of centroid-moment-tensors the minimum of 4-km separation that inhibits rupture propagation
of recent seismicity related to the magnitude 3 earthquake events in (Wesnousky, 2006).
2011 reveal predominantly left-lateral movement (Afnimar et al., 2015; In the Cipogor Section the fault zone is rather complicated (sup-
Madrinovella et al., 2013), thus supporting the interpretation of Tjia plementary information S2). It shows a broad zone of deformation
(1968). consisting of a compressional bulging structure and shows vertical
movement with the southern block up thrown. The amount of elevation
4. Methods difference between the peaks of the ridge and the average top of the
down-side block is up to 300 m (Supplementary information S7). This
The key to our work is the availability of IFSAR (Interferometric section also features a broad anticline on its north side, and another
Resolution Synthetic Aperture Radar) elevation data with 5-m grid re- left-lateral strike-slip fault strand that deviates from the main fault zone
solution, and LIDAR (Light Detection and Ranging) elevation data with near the tip where the Cimahi River channel runs westward and parallel
0.9 m grid resolution in the form of a DSM (Digital Surface Models) and with the Cimeta section's fault trace (i.e. Section 1). Minor structures
DTM (Digital Terrain Models). These data sets have been acquired also exist aligned in a direction perpendicular to the main fault and
during the Indonesian Earthquake Hazard Project, a collaboration of involving compressional folded structures.
Indonesian and Australian earth scientists supported by the Indonesian The Cihideung Section (Supplementary information S3), in the
and Australian governments from 2010 to 2015 (Horspool et al., 2011). central part of the fault, shows one major fault strand with many
The high resolution digital topography datasets made available by this streams that are left-laterally offset along the fault. This section is also
project enabled us to map the Lembang fault strands in great detail, characterized by three depressions: the Ciwaruga, Cibeureum, and
including their associated offsets. Cihideung basins, which are all associated with zones of left-lateral
We have conducted surface fault mapping based on inspection of stream displacements. The eastern termination of the Cihideung Section
landscape expressions of fault movements using conventional methods is marked by a 150 m wide right stepover to the western end of the
in earthquake geology (McCalpin, 1996; Yeats et al., 1997), similar to Gunung Batu Section (supplementary information S4).
the methods used to map the Great Sumatran Fault (Sieh and Along the western part of the Gunung Batu Section, the fault trace
Natawidjaja, 2000), the Anatolian Fault in Turkey (Saroglu et al., seems to dissolve into a broad monocline that shallowly dips to north.
1992), the Chelungpu Fault in Taiwan (Shyu et al., 2005), and the We interpret this bulge as the expression of a positive flower structure
Sagaing Fault in Myanmar (Wang et al., 2014). We also use various GIS that is associated with a blind thrust underlying the monocline. Here,
(Geographic Information System) techniques, such as multi-shade- the elevation difference between the top of the ridge and the flat
analysis of digital elevation relief imagery (ESRI-Mapping-Center- morphology to the north increases from 30 to 60 m in the west up to
Team, 2010) to illuminate any morphological features related to fault 200 m in the east (Supplementary information S7). The significant
movements. change in the vertical offsets coincides with the change from the
We also conducted surface geological mapping to study stratigraphy smooth-undulating relief of the Tuffaceous Sand formation in the west
and find key marker beds related to fault movements. We have sur- to the rough-high relief of the Volcanic Old-Breccia formation in the
veyed geological transects along river valleys to collect samples from east (Fig. 8).
rock outcrops, and logged their rock type and position. The Cikapundung and Batu Lonceng Sections (Supplementary in-
formation S5 and S6) form a continuous major fault line but their strike
5. Result varies significantly from west to east along the Cikapundung Section
then shifts southeastward toward Gunung Palasari. The most striking
5.1. Geometry of the Lembang Fault feature here is that the active fault strand does not run along the slope
breaks as previously thought assuming the apparent vertical movement
Inspection of IFSAR, LiDAR and observations in the course of with north block down and the arcuate ridge are related to the fault
fieldwork show that the Lembang Fault is characterized by active-fault movement with significant normal slip component. Instead, the fault
landscape features such as linear valleys (LV), fault scarps (FS), shutter cuts through the middle of the steep north slope. Inspections of detailed
ridges (SR), river offsets (RO), wind gaps (WG) and beheaded rivers morphology along the fault trace also shows no vertical movement but
(BR), which extend for 29 km in length (Fig. 3). Previously, the Lem- pure strike-slip motion. The fault scarp on the Batu Lonceng Section
bang Fault has been mapped along its 22-km length from an arcuate even shows a contrary apparent dip-slip but with the north block up;
ridge in the Cipogor river area at its western end to the arcuate–ridge However, it is probably formed due to displacements of undulated to-
hill of Palasari at its eastern end. Our mapping shows that the fault has pography parallel to the strike-slip fault line. Thus, the apparent ver-
an additional westward extension of about 6 km until it gradually dis- tical breaks and the arcuate ridge along the Cikapundung and Batu
appears in the Padalarang area. On the eastern end, the fault expression Lonceng sections are not related to active fault movement with

182
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 3. Map showing active fault traces and general morphology of the Lembang Fault, its sections, and trenching sites. Detailed figures are shown in Supplementary
Information S1–S6.

significant or predominat dip-slip component but akin to pre-existing S11).

morphology. (3) Tuff. This geological unit is colored pink in Fig. 4, and is only ex-
posed in deep river valleys such as the Cimahi, Cimeta, Cihideung
5.2. Geology and stratigraphy and Cigulung river valleys beneath the Tuffaceous sand unit (Sup-
plementary information S11). The unit consists of a white tuff layer
Geology around the Lembang Fault has been mapped but only on a (Supplementary information S12) and a pumice-rich layer that has
regional scale (Van Bemmelen, 1949) Silitonga (1973), Sutoyo (Dam, large pumice fragments with sizes 0.5–20 cm. This unit is hard and
1994; Dam et al., 1996). Using the LiDAR and IFSAR DEMs, we revised very resistant to erosion. An estimated total thickness of this unit is
the available geological mapping to add details around the Lembang about 40 m. In the Cimahi river valley, this unit is generally ex-
Fault. There are five lithological units in this area (Fig. 4), which are posed as water falls. This rock formation is thought to be related to
correlated with geological events in the Bandung area (Table 1): shallowing of the Bandung Lake water level at 25–20 ka (Dam,
1994), and major eruptions of the Sunda and Tangkuban Perahu
(1) Lake Sediment. This unit is colored green in Fig. 4. It is char- volcanoes during 35–20 ka (Dam, 1994; Dam et al., 1996;
acterized by flat-smooth topography filling in isolated depressions Kartadinata et al., 2002).
along the north side of the Lembang fault. There are five locations (4) Young (Volcanic) Breccia. This unit is colored orange in Fig. 4, and
with areas from 0.02 to 0.6 km2, which are Cihideung, Situ Umar, is found at the bottom of Cimahi, Cihideung, Cimeta river valleys
Ciwaruga, Danau Kering, and Cibeureum basins or ponds. The first (Supplementary information S11, 13). It has an open fabric, brown
three still have water impoundment at present, but the other two to reddish colour, is hard and has a matrix of coarse sands and tuff.
are dry lakes. Carbon dating analysis on samples of the sediments of The matrix includes angular fragments of andesitic rock having
the lake yields 20.19 ± 0.23 ka (Dam, 1994). dimensions of 10–100 cm.
(2) Tuffaceous Sand. This unit is colored yellow in Fig. 4, and covers (5) Old Breccia. This unit is mapped in brown colour in Fig. 4. The
50% of the study area. Its morphology is characterized by un- geomorphology is expressed as a high-relief mature hill landscape
dulated-smooth topography but it has been deeply incised by major with deep river channels at the eastern part of the Lembang Fault.
river systems. The lithology is dominated by coarse tuffaceous sand This unit consists of very indurated volcanic breccia (Supplemen-
that consists of fragments of igneous rock and tuff with size tary information S16). The matrix is sand and tuff, with angular
0.5–1 cm in open fabric. This rock is easily broken with a rock igneous rock inclusions with open fabric.
hammer. The unit is estimated to be about 45 m thick. It is com-
monly quarried in open pits (Supplementary information S10 and Based on the results of multiple dating analyses and compilation of

Fig. 4. Map of lithological units along the Lembang Fault. Table shows stratigraphic correlations with dated major geological events: (1) Lake Sediment filling
isolated depressions along the north side of the Lembang fault is characterized by flat-smooth topography. Carbon-dating analysis yields 20.19 ± 0.23 ka (Dam,
1994). (2) Tuffaceous Sand is characterized by undulated-smooth topography but deeply incised by major river systems. (4) 35–20 ka Tuff (Dam, 1994; Dam et al.,
1996; Kartadinata et al., 2002) is only exposed in the bottom of deep river valleys (5) Strongly consolidated Breccia with a matrix of coarse sands and tuff. (6) Old
Breccia consists of very indurated volcanic breccia with matrix of tuff and sandstone. The age of Old Breccia is about 205–185 ka (Nasution et al., 2004).

183
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Table 1
Ages and geological event on the Bandung area.
Unit Event Event Age Range (ka) Dating Methode/Refference Offset range (m) Slip rate (mm/yr)

Lake sediment Lacustrin Sedimentation along Lembang Fault 19.04–20.42 Radiocarbon (Dam, 1994) Unknown
Tuffaceous sands Tangkubang Perahu Phase 5 Eruption 40.48–41.02 Radiocarbon (Kartadinata et al., 2002) 100–140 2.43–3.45
Tuff
Young Breccia Sunda Caldera collapse 205–180 K-Ar (Nasution et al., 2004) 400–520 1.95–2.8
Old Breccia Early Pleistocene Volcanics 1100–570 K-Ar (Sunardi and Kimura, 1998) Pre Lembang Fault or Unknown

chronological events, the Old and Young Breccia units are interpreted 5.3.2.1. Large Offsets (~460 m). The Cimahi River consists of two
to be correlated with early to middle Pleistocene Volcanism at about parallel channels on the north side of the Lembang fault that
1100–570 ka (Sunardi and Kimura, 1998) and the Sunda Caldera col- currently merge to flow into a single channel on the south side of the
lapse about 205–180 ka (Nasution et al., 2004), respectively. fault. At first glance it may appear that the Cimahi River channel has a
From the geologic history and geomorphology of the Lembang Fault spectacular right-lateral offset of about 350 m that sharply displaces
region, there are two major volcanic events that have reset the land- similar-size deep channels on the north and south sides of the fault.
scape and reorganized river drainage systems. These are: (1) The par- However, our fieldwork found that this is only an apparent offset due to
oxysmal eruption of the Sunda Volcano that led to the major caldera beheading and pirating of the upstream (northern) channel, facilitated
collapse around 205–185 ka (Nasution et al., 2004). This marked the by the presence of a thick, erosion-resistant lava unit that made the
gradual end of the Sunda-Volcano period and the beginning of the Cimahi River prone to lateral erosion.
Tangkuban Perahu-Volcano period. Note that some studies, however, Our alternative interpretation of sinistral motion is based on two
were still using the term Sunda Volcano for volcanic activity after this currently dry channels on the southern block, just east of the present
event even up to 20 ka (e.g. Dam, 1994; Dam et al., 1996), (2) There are course of the Cimahi River, that we interpret as the original continua-
apparently two major catastrophic volcanic eruptions during the tions of the two channels of the Cimahi River on the northern block. We
“Tangkuban Perahu” period, which are in 105 and 50–35 ka (Dam, surmise that sinistral motion on the Lembang Fault eventually resulted in
et.al, 1996), but for this study, however, we focus on the 50–35 ka capture of both channels on the northern block by the single channel on
event, which correlates well with the age of the old crater (40 ka) ad- the southern block that lies just westward of the two dry channels and is
jacent to the present Tangkuban Perahu crater. presently the course of the Cimahi River (see Fig. 6). Offset of the
northern block channels from their now-dry original continuations on the
5.3. Geomorphological Offsets southern block implies a large 460 m sinistral offset of the Cimahi River.
To the east of the Cimahi River, in the Maribaya area of the
5.3.1. Vertical Offsets Cikapundung Section, the current course of the Cikapundung River is
Vertical offset is measured as the distance between slope lineaments WSW north of the Lembang Fault, but turns sharply southward where it
of the northern and southern parts of the Lembang Fault (see crosses the fault. Farther to the west on the northern block, the
Supplementary information S7). These vertical offsets are indexed ac- southward flow of the Cigulung and Cikawari rivers appears to be
cording to the kilometre notation along the fault used in e.g. Fig. 3. blocked by the Lembang Fault, at which point they flow eastward and
Some cross sections have no measurements because no lineament trend into the Cikapungung River. We surmise that prior to displacement by
was discernable. Km0 to Km6 has a small offset of about 0.8 ± 3 m the Lembang Fault, all three rivers on the northern block - the Cigulung,
(Km5) and 7.4 ± 3 m (Km6) with the southern side down. Then from Cikawari and Cikapundung Rivers - joined and flowed southward. The
km 7 to km 14, the vertical offset changes abruptly to 91 ± 3 m with implied offset of the Cikapundung River is consistent with a large 460 m
the northern side down. From km 15 to km 19 the vertical offset de- offset of the Cimahi River, and this offset also aligns the current course
creases to 28.3 ± 2 m with the northern side down. Finally, km 20 to of the Cihideung River north of the Lembang Fault with a paleochannel
km 29 has a small vertical offset of about 13.1 ± 0.2 m with the on the southern block, as indicated in Fig. 6.
southern side down.
5.3.2.2. Intermediate Offsets (~120 m) related to Tangkuban Perahu
5.3.2. River Offsets Major Eruption ~40 ka. The Group 5 offsets of roughly 120 m are all
Drainage patterns, straightness trends and changes in lithology that on smaller, younger rivers and streams which incise only the Tuffaceous
were connected across the fault before being displaced by fault move- Sand unit. Hence, these offsets measure fault displacement since the
ment can be estimated by finding and matching appropriate landform deposition of this unit, which is a product of a major eruption of
markers (Burbank and Anderson, 2001; McCalpin and Nelson, 2009). Tangkuban Perahu at 40.75 ± 027 ka, obtained by Kartadinata et al.
For example, the measurement of river offsets accounts for the river (2002).
drainage flow direction and offset-followed rivers (Arrowsmith and In the westernmost section, Cimeta, the fault strand is associated
Zielke, 2009; Lienkaemper, 2001). The extent of the river on the up- with a clear deflection of streams that incise the Tuffaceous Sand and
stream side of fault line is called the head and the extent on the Tuff units, indicating a sinistral offset of approximately 120 m (Figs. 7
downstream side is called the tail. The distance between projections of and S1). No vertical offset is observed here, hence suggesting a pure
the river lineament onto the fault trace from the head and the tail is strike-slip movement.
taken as the length of the river offset. Detailed geological investigation showed that the Cimeta River in-
River offsets measured along the Lembang Fault fall into 6 groups cises through the Tuffaceous Sand, Tuff and Young Breccia units
(Fig. 5 and supplementary information Table S1). These groups have (Fig. 8). This area is mostly covered by the Tuffaceous Sand unit. Fig. 9
mean offsets and uncertainties of: 7 ± 2 m, 26 ± 12 m, 65 ± 10 m, shows outcrops of the Young Breccia unit with a fault plane striking
84 ± 9 m, 120 ± 20 m, and 460 ± 60 m, respectively. The Group 5 N230°E and 75° dip N, cutting across a lava sub-unit. However, the
offsets are quite similar to those of Tjia (1968), who measured river Cimeta River clearly has the same offset as its branch river to the west,
offsets of about 140 m. The largest river offsets are in Group 6, which which only incises the Tuffaceous Sand. We therefore infer that the
includes evidence of sinistral offsets of the Cimahi, Cihideung, Situ Cimeta River was subject to the same topographic reset at as its branch
Umar and Cikapundung rivers (Fig. 6 and Supplementary information river, and that its offset has similarly accumulated since the
S8). 40.75 ± 027 ka eruption of Tangkuban Perahu.

184
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 5. River offsets along the Lembang Fault: (a) graph showing the length of river offsets ordered by decreasing magnitude, and grouped into classes from 1 to 6
(youngest to oldest), (b) Sections of the Lembang Fault, (c) distribution of river offsets highlighting the largest offsets of the oldest class-6 rivers, (d) Graph showing
intermediate offsets of 70 to 210 m (e) Graph showing distributions of smaller offsets, 1 to 100 m along km 21 to km 29, (f) Graph showing vertical offsets along the
fault.

185
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 6. Mismatch of the largest geomorphic offsets (a) Current morphology and (b) Shifted back to the position before a 460-m sinistral offset, observed at Cimahi,
Cihideung and Cikapundung rivers.

5.4. Slip rate Based on the Group 5 offsets of roughly 140 m, are all made on
smaller younger rivers and streams which incise only (with the exception
The rivers on which the Group 6 offsets are measured are all major, of the Cimeta River mentioned above) the Tuffaceous sand unit. Hence
well-established rivers, and while some of them currently traverse the these offsets measure fault displacement since the deposition of this unit,
Tuffaceous Sand unit, all have incised the underlying Breccia, and their for which we use the carbon dating results for this late Quaternary vol-
original channels are evident in the southern block even where they have canism of 40.75 ± 027 ka, obtained by Kartadinata et al. (2002). This
been covered by the Tuffaceous sand. Hence these offsets measure the total leads to a range of estimated slip rates for the Group 5 offsets of
dislacement of the Lembang Fault since its formation with the collapse of 2.32–3.58 mm/yr (Table 1), compatible with the Group 6 results.
the Sunda Caldera, for which we adopt the KeAr dating results of Sunardi The Group 1–4 offsets are more difficult to interpret. While the ridge
and Kimura (1998) of 185–205 ka. This leads to an estimated range of slip where these 9–84 m offsets were measured is very old (~0.5 Ma), the
rates based on the Group 6 offsets of 1.95–3.06 mm/yr (see Table 1). steep northern slopes are covered by young colluvial breccias. We

Fig. 7. The clearest evidence of 120-m sinistral offset at Pasirlangu (a) Current morphology, (b) After removing the offset along the fault line.

186
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 8. (a) Geological map at km 4 to 6 on the Cimeta River and (b) cross section along the marked-AB line showing, the deep-stream incision exposes the older layers
beneath the Tuffaceous Sands that blanketed the surface. The fault zone is exposed within the Young Breccia Unit at the bottom of Cimeta River as shown in Fig. 9.

believe that these young breccias are products of frequent gravitation higher than the estimates we obtain above. Tjia (1968) gives no indication of
slumps or downhill creep that has eroded and redeposited the old what information the anthropological age is based on, but it probably refers
volcanic formation, thus resetting the stream topography. Hence, we to Van Bemmelen's (1949) erroneous assertion that the entire eruptive his-
can only say that these offsets measure recent fault displacement, but tory of the Tangkuban Perahu volcano complex occurred within the 3000-
cannot establish a precise age and are therefore unable to use them to year history of the local Sundanese population. In any case, the anthro-
estimate slip rates. pological age estimate is apparently not the age of the offset river channel.
Tjia (1968), whose 140 m sinistal offset of a river channel is compatible In summary, the river offsets which can be reliably dated are those
to the Group 5 offsets measured here, appears to have used an “anthro- from Groups 5 and 6, and these result in sinistral fault slip rates of
pological age” to infer a slip rate for the Lembang Fault of 30 mm/yr, much 2.43–3.45 mm/yr.and 1.95–2.8 mm/yr, respectively.

Fig. 9. Outcrop at Cimeta River showing the fault with strike/dip of N230°E/75° cut the Young Breccia Unit. Behind the man is a Lava sub-unit truncated by the fault.

187
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 10. Trench logs of TR1 and TR2, parallel trenches that are 60-m apart along the Lembang Fault at Batu Lonceng. (a) TR1 eastern wall, (b) TR2 eastern wall (b)
and (c) TR2 western wall. Stratigraphy is divided into soil units: 100 (top soil, silty sands), 200 (gravelly coarse sands), 300 (silty sands), 400 (igneous fragment with
sandy silt matrix), and 500 (silty sands). AMS radiocarbon analysis of selected samples from the bottom and top of Unit 300 yields the ages from 500 to 440 BP
(1450–1510 CE) to 535–390 BP (1415–1560 CE), which can be interpreted as bracketing the occurrence time of Ruptures A and B.

5.5. Paleoseismological trenching to trench at this site was based on a resistivity survey Gr8ADD
(Supplementary information S24). The Gr8ADD resistivity survey was
We dug three trenches manually, TR1 and TR2 at km 26.3 (Batu made using a SYSCAL Pro48 resistivity meter. The results from this
Lonceng Section), and TR3 at km 11.4 near Panyairan Danau Village survey showed an obvious fault unconformity at the TR3 location. TR3
(Cihideung Section, see Fig. 3). TR1 and TR2 are parallel trenches about is 20 m long, 1.5 m wide, and 2.75 m deep (Supplementary information
60 m apart crossing the Lembang Fault. TR1 is in a dry, intermittent S25). TR3 could not be dug deeper because of extensive groundwater
river valley, and TR2 is on the hill between two river valleys. Trench seepage. TR3 has seven units. Unit 100 is a Rubbish Clay unit formed
TR3 is located on an existing recharge farming drainage. due to farming activity. Unit 200 is grey sandy silt. Unit 300 is light
TR1 has dimensions 20 m length × 1 m width × 2.25 m depth grey sandy silt. Unit 400 is grey sandy silt. Unit 500 is brown sandy silt.
(Supplementary information S17). Only its eastern wall is available to Unit 600 is black (some parts are brown) sandy silt. Unit 700 is orange
describe because the other side was used as an access stair, so it was too to brown sandy silt. An obvious rupture cuts units 500, 600 and 700
deformed to interpret. The eastern wall has five layers (Fig. 10), unit (Figs. 11 and 12). The minimum rupture age for this earthquake is
100 (top soil, silty sands), unit 200 (gravelly coarse sands), unit 300 2115–2010 BP or about 60 BCE (Unit 400, sample PD8), and the
(silty sands), unit 400 (igneous fragment with sandy silt matrix), and maximum age is 4465–4450 BP (Unit 500, sample PD4) or about
unit 500 (silty sands, supplementary information S18 and S19). In this 2300 BCE. Unit 700 has age about 13,435–13,300 BP (sample
section, it is possible to infer two ruptures (Ruptures A and B) that cut PD13ADD) (supplementary information S26). The upward displace-
into the base of unit 300 but do not emerge at the surface. AMS ment of the base of unit 700 indicated in Fig. 11 is at least 40 cm.
radiocarbon dating shows unit 300 has a bottom age of 500–440 BP Thus, our paleoseismology work on the Lembang Fault shows very
(dates 1450–1510 CE, sample P1-2) and at its top the age is 535–390 BP clear evidence of one earthquake in 2300–60 BCE, and a 15th century
(dates 1415–1560 CE, sample P1-3). These ages can be interpreted as earthquake. The 2300–60 BCE rupture shows about 40 cm vertical
bracketing the ages of Ruptures A and B, which therefor occurred movement. Considering the ratio of the maximum geomorphic left-
sometime in 1450–1560 CE. lateral offsets of about 500 m and the vertical offsets of no more than
TR2 also has five units (supplementary information S20, S21, and 100 m (Fig. 5), we may roughly estimate that ratio of left-lateral to
S22). Unit 100 is grey, top soil and sandy silt. Unit 200 is brown, sandy vertical slip is approximately 5 to 1. Hence the amount of the lateral slip
silt. Unit 300 is brown, sandy silt, with black colored-soil at the top. of the 2300–60 BCE might be as much as ~200 cm, which is, according
Unit 400 is gravelly coarse sand, and unit 500 is brown, sandy silt. From to the scaling relations for displacement of Wells and Coppersmith
this section can be inferred evidence of Ruptures C and D. Rupture C (1994), corresponds to a Magnitude 7 earthquake.
has a maximum earthquake age of 5705–5620 BP (sample TR2-5, top of
unit 300). Rupture D has a minimum earthquake age of 19,620–19,140 6. Conclusions
BP (sample TR2-9, bottom of unit 300). Unit 500 has age
20,525–20,290 BP (sample TR2-12). This paper has described the detailed location, length, movement
The TR3 site is in a closed morphology basin which often floods mechanism, and geological slip rate on the Lembang Fault. Although it
during the rainy season (Supplementary information S23). The decision has no recorded history of large earthquakes, the Lembang Fault shows

188
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Fig. 11. Exposures of the Lembang Fault (red lines)

cutting young sediments at trench TR3 along the
eastern wall (top) and western wall (bottom). The
vertical spacing between the strings tacked to each
wall is 50 cm. The soil stratigraphy is divided into
seven units (detail explanation on Fig. 12). On these
figures are shown Unit 400 (grey sandy silt), Unit
500 (brown sandy silt), Unit 600 (black - some part is
brown, sandy silt), Unit 700 (orange to brown sandy
silt). An obvious rupture cuts units 500, 600 and 700.
The vertical offset of the base of unit 700 is ap-
proximately 40–60 cm. The minimum rupture age
for this earthquake is 2115–2010 BP or about 60 BCE
(Unit 400, sample PD8), and the maximum age is
4465–4450 BP (Unit 500, sample PD4) or about
2300 BCE. (For interpretation of the references to
colour in this figure legend, the reader is referred to
the web version of this article.)

obvious geomorphic evidence of recent activity and has long been have clearly established the sense of motion or quantitatively de-
thought to be active. Previous studies of the Lembang Fault (Van termined the geologic slip rate, nor has any study ever established that
Bemmelen, 1949; Tjia, 1968; Kaars and Dam, 1997, among others) have the fault has hosted large earthquakes. In this study we have used
strongly suggested that it is active, but other than Tjia (1968), none geomorphic evidence to unequivocally establish that the Lembang fault

Fig. 12. Log of trench TR3 in Panyairan Danau Village. (top) eastern wall and (bottom) western wall. TR3 could not be dug deeper because of extensive groundwater
seepage. TR3 has seven units. Unit 100 is a Rubbish Clay unit formed due to farming activity. Unit 200 is grey sandy silt. Unit 300 is light grey sandy silt. Unit 400 is
grey sandy silt. Unit 500 is brown sandy silt. Unit 600 is black (some part is brown) sandy silt. Unit 700 is orange to brown sandy silt. An obvious rupture cuts units
500, 600 and 700. The minimum rupture age for this earthquake is 2115–2010 BP or about 60 BCE (Unit 400, sample PD8), and the maximum age is 4465–4450 BP
(Unit 500, sample PD4) or about 2300 BCE. Unit 700 has age about 13,435–13,300 BP (sample PD13ADD). (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)

189
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

has a dominantly sinistral sense of motion with a slip rate of References

1.95–3.45 mm/yr. This confirms that Lembang Fault currently accom-
modates trench parallel slip movements due to obliquity of the Java Abidin, H.Z., Andreas, H., Kato, T., Ito, T., Meilano, I., Kimata, F., Natawidjaja, D.H.,
subduction as inferred from earthquake focal mechanism analysis Harjono, H., 2009. Crustal deformation studies in Java (Indonesia) using GPS. J.
Earthq. Tsunami 3 (2), 77–88.
(McCaffrey, 1991) and geodetic data modeling (Koulali et al., 2017). Afnimar, Yulianto, E., Rasmid, 2015. Geological and tectonic implication obtained from
With a length of 29 km, this suggests that the Lembang Fault could first seismic activity investigation around Lembang fault. Geosci. Lett. 2 (4).
produce a Mw 6.5–7.0 earthquake with a recurrence time of Arrowsmith, J.R., Zielke, O., 2009. Tectonic geomorphology of the San Adreas Fault zone
from high resolution topography: an example from the Cholame segment.
170–670 years. We also conducted paleoseismological trenching of the Geomorphology 113, 70–81.
Lembang Fault and found evidence for at least 3 earthquakes: in the Barat SIAK Provinsi Jawa, 2011. Profil Kependudukan Jawa Barat Tahun 2011. 2014.
15th centruy, 2300–60 BCE and around 18,000 BCE. The 2300–60 BCE Bradley, K.E., Feng, L., Hill, E.M., Natawidjaja, D.H., Sieh, K., 2017. Implication of the
diffuse deformation of the Indian Ocean lithosphere for slip partitioning of oblique
earthquake might be a magnitude 7 earthquake, based on its amount of plate convergence in Sumatra. J. Geophys. Res. Solid Earth 122, 572–591.
the observed vertical slip and the ratio of left-lateral to vertical slip. Burbank, D., Anderson, R., 2001. Tectonic Geomorphology. Blackwell Science, Inc.
This is the first unequivocal identification of a source of crustal (274 pp.).
Dam, M.A.C., 1994. The Late Quaternary Evolution of the Bandung Basin, West Java,
earthquakes in Java, Indonesia, the world's most densely populated
Indonesia [PhD: Vrije Universitet].
island in one of its most tectonically active areas. The Lembang Fault Dam, M.A.C., Suparan, P., Nossin, J.J., Voskuil, R.P.G.A., Group, G, 1996. A chronology
and others like it in Java are likely to pose substantial risk to not only for geomorphological developments in the greater Bandung area, West-Java,
Bandung but many of Java's major urban agglomerations. This has been Indonesia. J. SE Asian Earth Sci. 12 (1/2), 101–115.
Engdahl, E.R., Villasenor, A., Deshon, H.R., Thurber, C.H., 2007. Teleseismic relocation
recognised recently, with the number of active faults in Java considered and assessment of seismicity (1918–2005) in the region of the 2004 Mw 9.0 Sumatra-
in the seismic hazard map of Indonesia increasing from four in Irsyam Andaman and the 2005 Mw 8.6 Nias Island Great earthquakes. Bull. Seismol. Soc.
et al. (2010) to 26 in Irsyam et al. (2017). Although the 26 faults in the Am. 97, S1–S19.
ESRI-Mapping-Center-Team, 2010. Hillshade Tools, the Swiss Hillshade Model and the
2017 hazard map have geomorphic indications of recent activity (e.g., MDOW (Multi-Directional Oblique Weighting) Model. http://mappingcenter.esri.
Marliyani et al. (2016)), and their potential for earthquake occurrence com/index.cfm?fa=arcgisResources.gateway.
can be estimated from geodetic slip rates and maximum length, none Hall, R., Cottam, M.A., Wilson, E.J., 2011. The SE Asia Gateway: history and tectonics of
the Australia-Asia Collision. pp. 75–109 (London Geological Society 355, Special
have reported geological evidence of actual earthquakes. Moreover, Publication).
although Java has a history of destructive earthquakes (Musson, 2012; Hart, Earl W., Bryant, William A., 1997. Alquist-Priolo Earthquake Fault Zoning Act -
Newcomb and McCann, 1987), some of which are almost certainly as- Seismic Hazards Mapping Act. California Department of Conservation, Division of
Mines and Geology.
sociated with crustal faults (Nguyen et al., 2015), none have been Hidayat, E., Brahmantyo, B., Yulianto, E., 2008. Analisis Endapan Sagpond pada Sesar
confidently ascribed to a mapped crustal fault. In this context, this Lembang. Geoaplika 3 (3), 151–161.
study's confirmation that not all strain release in the crust of Java is Horspool, N., Natawidjaja, D.H., Yulianto, E., Lawrie, S., Cummins, P., 2011. An
Assessment on the Use of High Resolution Digital Elevation Models for Mapping
aseismic is highly significant.
Active Faults in Indonesia: Natural Hazards Impact Project Risk and Impact Analysis
The results of detailed active fault mapping are important for Group Geoscience Australia.
earthquake disaster mitigation at a local scale. In the US, an area within Irsyam, M., Sengara, W., Aldiamar, F., Widiyantoro, S., Triyoso, W., Hilman, D.,
a distance of 15 m from active fault lines (30 m wide of both sides) Kertapati, E., Meilano, I., Suhardjono, Asrurifak, M., Ridwan, M., 2010. Development
of Seismic Hazard Maps of Indonesia for Revison of Seismic Hazard Map in SNI 03-
should not be inhabited (Hart and Bryant, 1997). In New Zealand, the 1726-2002: Research Report Submitted to the Ministry of Public Works by Team for
distance is 20 m (40 m wide of both sides) (McClymont, 2001). This Revision of Seismic Hazard Maps of Indonesia.
study reports the first detailed mapping of tectonic geomorphological Irsyam, M., Natawidjaja, D.H., Meilano, I., Widiyantoro, S., Rudyanto, A., Hidayati, S.,
Triyoso, W., Hanifa, N.R., Djarwadi, D.L.F., Sunarjito, 2017. Peta Sumber dan bahaya
features along the Lembang Fault in Indonesia, and could be used to gempa Indonesia tahun 2017: Pusat Studi Gempa Nasional (PUSGEN).
establish a similar exclusion zone in which residential dwellings should Kaars, S., Dam, R., 1997. Vegetation and climate change in West-Java, Indonesia during
be prohibited. Further studies are needed to identify and map other the last 135,000 years. Quat. Int. 77, 67–71.
Kartadinata, M.N., Okuno, M., Nakamura, T., Tetsuo, K., 2002. Eruptive history of
active faults in Java and elsewhere in Indonesia, so that such mitigation Tangkuban Perahu Volcano, West Java, Indonesia: a preliminary report. J. Geogr.
measures can be adopted more widely. 111, 404–409.
Koulali, A., McClusky, S., Susilo, S., Leonard, Y., Cummins, P., Tregoning, P., Meilano, I.,
Efendi, J., Wijanarko, A.B., 2017. The kinematics of crustal deformation in Java from
GPS observations: Implication for fault slip partitioning. Earth Planet. Sci. Lett. 458,
Acknowledgement 69–79.
Lienkaemper, J.J., 2001. 1857 Slip on the San Andreas fault southeat of Cholame,
California. Bull. Seismol. Soc. Am. 91 (6), 1659–1672.
We are grateful to Australian Department of Foreign Affairs and
Madrinovella, I., Widiyantoro, S., Nugraha, A.D., Triastuty, H., 2012. Studi Penentuan
Trade for the financial support for the Graduate Research for dan Relokasi Hiposenter Gempa Mikro Sekitar Cekungan Bandung. Jurnal
Earthquake and Active Tectonics (GREAT) - Institute Teknologi Geofisika 2.
Bandung (ITB), Research Centre for Geotechnology – Indonesian Madrinovella, I., Widiyantoro, S., Nugraha, A.D., Triastuty, H., 2013. Studi Mekanisme
Fokus Gempa Mikro Sekitar Cekungan Bandung. Jurnal Geofisika 14 (1).
Institutes of Sciences (LIPI). We thank Bambang S. Widoyoko, Nandang Marjiyono, Soehaimi, A., Kamawan, 2008. Identifikasi Sesar Aktif Daerah Cekungan
Supriatna, Dadan D. Wardhana, Purna S. Putra and Ilham Arisbaya for Bandung dengan Citra Landsat dan Kegempaan. JSDG XVIII (2), 81–88.
georesistivity assistance. We thank Ayu, Januar Ridwan and Eka Marliyani, G.I., Arrowsmith, J.R., Whipple, K.X., 2016. Characterization of slow slip rate
faults in humid areas: Cimandiri Fault zone, Indonesia. J. Geophys. Res. Earth Surf.
Gamara for assistance during geological mapping. Indonesian 121, 2287–2308.
Volcanologists and PVMBG colleagues Sri Hidayati, Ahmad Solikhin, McCaffrey, R., 1991. Slip vectors and stretching of the Sumatran fore arc. Geology 19,
Supartoyo, Imun Maimunah and Afif. We thank Rob Govers 881–884.
McCalpin, J., 1996. Paleoseismology. In: Dmowska, D., Holton, J. (Eds.), International
(Tectonophysics), Rishav Mallick (Earth Observatory of Singapore) and Geophysics Series. Academic Press, San Diego.
Noelynna T. Ramos (University of Phillipines) for constructive com- McCalpin, J.P., Nelson, A.R., 2009. Introduction to paleoseismology. In: McCalpin, J.P.
ment and suggestion to improve the manuscript. (Ed.), Paleoseismology - International Geophysics. Vol. 95 Elsevier.
McClymont, B., 2001. Building on the Edge - the Use and Development of Land on or
Close to Fault Lines. Parliamentary Commissioner for The Environment, Wellington.
Musson, R.M.W., 2012. A Provisional Catalogue of Historical Earthquakes in Indonesia:
Appendix A. Supplementary data Open Rep.OR/12/073,22.
Nasution, A., Kartadinata, M.N., Kobayashi, T., Siregar, D., Sutaningsih, E., Hadisantono,
R., Kadarstia, E., 2004. Geology, Age Dating, Geochemistry of the Tangkuban Perahu
Supplementary data to this article can be found online at Geothermal Area, West Java, Indonesia. J. Geotherm. Res. Soc. Jpn. 26 (3), 285–303.
doi:https://doi.org/10.1016/j.tecto.2018.12.014. These data include Newcomb, K., McCann, W., 1987. Seismic history and seismotectonics of the Sunda Arc. J.
the Google maps of the most important areas described in this article. Geophys. Res. 92 (B1), 421–439.
Nguyen, N., Griffin, J., Cipta, A., Cummins, P.R., 2015. Indonesia's Historical Earthquakes

190
M.R. Daryono et al. Tectonophysics 751 (2019) 180–191

Modelled Examples for Improving the National Hazard Map: Geoscience Australia, dan Bencana Geologi 2 (3), 185–190.
Canberra. Sunardi, E., Kimura, J., 1998. Temporal chemical variation in Late Cenozoic volcanic
Saroglu, F., Emre, O., Kuscu, I., 1992. Active Fault Map of Turkey: General Directorate of rocks around the Bandung Basin, West Java, Indonesia. J. Mineral. Petrol. Econ. Geol.
Mineral Research and Exploration, Scale 1:1,000,000. 93, 103–128.
Shyu, B., Sieh, K., Chen, Y.-G., Liu, C.-S., 2005. Neotectonic architecture of Taiwan and its Sutoyo, Hadisantono, R., 1992. Geological Map of Tangkuban Parahu Volcano, West
implications for future large earthquakes. J. Geophys. Res. 110. Java, Bandung, Indonesia.
Sieh, K., Natawidjaja, D., 2000. Neotectonics of the Sumatran fault, Indonesia. J. Tjia, H.D., 1968. The Lembang Fault, West Java. Geologie En Mijnbouw 47 (2), 126–130.
Geophys. Res. 105 (B12), 28295–28326. Van Bemmelen, R., 1949. The Geology of Indonesia. Government Printing Office, The
Silitonga, P.H., 1973. Geologic Map of the Bandung Quadrangle, Java: Geological Survey Hague, Netherlands (732 p).
of Indonesia. Ministry of Mines. Wang, Y., Sieh, K., Tun, S.T., Lai, K.-Y., Myint, T., 2014. Active tectonics and earthquake
Simandjuntak, T.O., Barber, A.J., 1996. Constrating tectonic style in the Neogene oro- potential of the Myanmar region. J. Geophys. Res. Solid Earth 119, 3767–3822.
genic belts of Indonesia. Geol. Soc. Lond. 106, 185–201. Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships among magnitude,
Simons, W., Socquet, A., Vigny, C., Ambrosius, B., Abu, S., Promthong, C., Subarya, C., rupture length, rupture width, rupture area, and surface displacement. Bull. Seismol.
Sarsito, D., Matheussen, S., Morgan, P., Spakman, W., 2007. A decade of GPS in Soc. Am. 84 (4), 974–1002.
Southeast Asia: Resolving Sundland motion and boundaries. J. Geophys. Res. 112. Wesnousky, S.G., 2006. Predicting the endpoints of earthquake ruptures. Nature 444
https://doi.org/10.1029/2005JB003868. (7117), 358–360.
Sulaeman, C., 2011. Laporan Tanggap Darurat Gempabumi Muril M3 Tanggal 28 Agustus Yeats, R.S., Sieh, K.E., Allen, C.R., 1997. The Geology of Earthquakes. Oxford Univ. Press,
2011: PVMBG - ESDM. New York.
Sulaeman, C., Hidayati, S., 2011. Gempa Bumi Bandung 22 Juli 2011. Jurnal Lingkungan

191