Thrusting of a volcanic arc: a new structural model for Java

Benjamin Clements1,*, Robert Hall1, Helen R. Smyth2 and Michael A. Cottam1

1

SE Asia Research Group, Department of Earth Sciences, Royal Holloway University of London, Egham,
Surrey TW20 0EX, UK
2
Present address: CASP, University of Cambridge, West Building, 181a Huntingdon Road, Cambridge CB3 0DH,
UK
*
Corresponding author (e-mail: [email protected])
ABSTRACT: Java is part of a volcanic island arc situated in the Indonesian

archipelago at the southern margin of the Eurasian Plate. Sundaland continental

crust, accreted to Eurasia by the Early Mesozoic, now underlies the shallow seas to
the north of Java where there has been considerable petroleum exploration. Java has
an apparently simple structure in which the eastwest physiographic zones identified
by van Bemmelen broadly correspond to structural zones. In the north there is the
margin of the Sunda Shelf and, in southern Java, there are Cenozoic volcanic arc
rocks produced by spatially and temporally discrete episodes of subduction-related
volcanism. Between the Sunda Shelf and the volcanic rocks are Cenozoic depocentres of different ages containing sedimentary and volcanic material derived from
north and south. This simplicity is complicated by structures inherited from the
oldest period of subduction identified beneath Java, in the Cretaceous, by extension
related to development of the volcanic arcs, by extension related to development of
the Makassar Straits, by late Cenozoic contraction, and by cross-arc extensional
faults which are active today. Based on field observations in different parts of Java,
we suggest that major thrusting in southern Java has been overlooked. The thrusting
has displaced some of the Early Cenozoic volcanic arc rocks northwards by 50 km
or more. We suggest Java can be separated into three distinct structural sectors that
broadly correspond to the regions of West, Central and East Java. Central Java
displays the deepest structural levels of a series of north-directed thrusts, and
Cretaceous basement is exposed; the overthrust volcanic arc has been largely
removed by erosion. In West and East Java the overthrust volcanic arc is still
preserved. In West Java the arc is now thrust onto the shelf sequences that formed
on the Sundaland continental margin. In East Java the volcanic arc is thrust onto a
thick volcanic/sedimentary sequence formed north of the arc in a flexural basin due
largely to volcanic arc loading. All the components required for a petroleum system
are present. This hypothesis is yet to be tested by seismic studies and drilling, but, if
correct, there may be unexplored petroleum systems in south Java that are worth
investigating.
KEYWORDS: Sundaland, Java structure, thrusting, volcanic arc

INTRODUCTION
Java (Fig. 1) is one of the many islands at the Eurasian margin
in Indonesia in which subduction has been an important
influence on geological history and consequently the development of petroleum systems. Subduction is active today
all around the SE Asian archipelago, from Sumatra to the
Philippines, and has been almost continuous for much of
the Cenozoic, particularly in Indonesia. Sumatra and the
Philippines, where subduction is oblique, are characterized by
features such as major strike-slip faults, displaced and fragmented terranes, and bends in faults that may be associated
with sedimentary basins or folding and thrusting, depending on
Petroleum Geoscience, Vol. 15 2009, pp. 159174
DOI 10.1144/1354-079309-831

the geometry of the bend. Volcanic activity is also often

localized at fault bends or tips, and complexities in volcanic
activity with time have been interpreted to result from oblique
subduction of major features on the downgoing slab, such as
fractures or ridges, or tearing of the subducting slab, opening
slab windows, causing seismic gaps or anomalous heat flows.
Overall, it is reasonable to expect oblique subduction to
produce a complex geology as features migrate along subduction margins which, in turn, can be expected to influence
petroleum systems and their prospectivity.
In contrast, Java is often cited as an example of the product
of orthogonal subduction and the volcanic island appears to
have a relatively simple structure. The physiographic zones
1354-0793/09/$15.00  2009 EAGE/Geological Society of London

160

B. Clements et al.

Fig. 1. Major features of the Java

region between Borneo and the Indian
Ocean. Topography from SRTM and
GTOPO30 and bathymetry from
Sandwell & Smith (1997). Boxes are
numbered with the figures to which
they refer in this paper. Towns and
location names discussed are labelled.
Arrow shows the current Indian
OceanEurasia convergence vector at
the Java Trench from McCaffrey
(1996).

recognized by van Bemmelen (1949) were identified by him as

structural units, and they are broadly parallel to the eastwest
elongation of the island and the strike of the subduction trench
south of Java. He recognized some deformation on Java and,
although it is sometimes difficult to interpret his ideas in terms
of modern tectonic processes, van Bemmelen evidently saw
deformation as predominantly vertical responses to magmatic
growth of geanticlinal features, which could result in thrusting
and folding due to gravity-driven movements. He considered
that the Sunda mountain system was typified by outward
migration of orogenic deformation with time, said by him to be
seen clearly in Sumatra. However, according to van Bemmelen,
Java showed a contrary inward migration of deformation from
the Indian Ocean, which he appears to attribute (e.g. van
Bemmelen 1949, p. 658) to deep crustal or mantle processes
associated with volcanic activity. The main episode of deformation he identified, which did produce northward folding and
thrusts in north Java had the character of backward- or
hinterland folding.
The idea that Java lacks significant deformation has
been passed down from van Bemmelen, and major thrusts are
not shown on most geological maps, except in north Java.
Simandjuntak & Barber (1996) interpret a major backthrust, the
BarabisKendeng thrust, to run the length of the island and
report that some segments are currently active. They interpret
there to have been a change from an extensional to a
compressional regime across the subduction system and suggest this could be a consequence of subduction of features such
as the Roo Rise on the Indian Plate (Fig. 1), or that localization
of compression may be due to magmatic intrusion and consequent uplift of the arc. This suggestion implies that the
northward thrusting in north Java is part of contractional

deformation of a large wedge north of the trench, with

compression between the trench and a backstop. The reason
for a backstop in north Java is not clear; usually in arcs, the
interpreted backstop is between the arc and the trench, not
behind the arc. The concept of such a large wedge is also
unusual, and suggests that the entire block is being pushed
northward, with deformation localized at its northern margin.
Oil company geologists in Indonesia are very familiar with
the significant thrusting in north Java and, with a few exceptions, this represents the southern limit of extensive subsurface
exploration which has been mainly offshore north of the island.
The subsurface structure of the island, and certainly the crust
south of the island, is less well known, because of problems
with exposure, difficulties in mapping structures at the surface,
and the paucity of seismic data. However, as a result of our
fieldwork on Java in the last few years, we suggest that there is
more thrusting than previously recognized, which is of great
importance in understanding the structure of the arc and may
be of interest to exploration geologists investigating frontier
plays. The onshore geology of West Java is very different to that
further north. The petroleum systems of the shallow seas to the
north of Java are covered elsewhere (e.g. Longley 1997; Doust
& Sumner 2007) but there is much less information on land.
What little subsurface data do exist in SW Java (e.g. Yulianto
et al. 2007) are further to the west in the Malingping Block
(Fig. 1), where the geology is different to that of the region
onshore.
In this paper we identify important features that we suggest
are better explained by a thrust model for Java. We propose
that Palaeogene arc rocks have been thrust northwards towards
the Sunda Shelf and that a modern analogue of the Eocene to
Early Miocene sedimentary and tectonic setting is found at the

Thrusting of a volcanic arc: Java

present day in the East JavaBali region. We first discuss West
Java, and then discuss thrusting and related deformation from
other parts of Java and, finally, consider briefly the implications
for hydrocarbon exploration.
WEST JAVA
The CiletuhBayah area (Fig. 2) contains key features that we
consider make the case for thrusting in southern Java most
clearly. We therefore first summarize the geology of that area
and then discuss previous and new interpretations, followed by
briefer descriptions of examples of thrusting and deformation
in other parts of West Java.
CiletuhBayah
The Ciletuh area (Fig. 2) is well known for its exposures of the
oldest rocks in West Java, which include rocks interpreted to be
Mesozoic. The Pre-Tertiary of Ciletuh Bay was reported by
van Bemmelen (1949) to consist of metamorphic basic and
ultra-basic rocks (gabbro, peridotite, serpentine) with chloritic
schists and phyllites. Schiller et al. (1991) described an ophiolitic assemblage of peridotites, gabbros, pillow lava basalts and
serpentinites associated with greenschists, mica-schists, amphibolite schists and quartzites, which has often been interpreted
as a tectonic melange (e.g. Endang Thayyib 1977; Martodjojo
et al. 1978; Parkinson et al. 1998). Little is known about the ages
of these rocks. Schiller et al. (1991) reported that radiometric
dating using the KAr method yielded very few conclusive
results due to alteration. A pebble of basalt from a conglomerate of the Ciletuh Formation yielded a Late Cretaceous age,
whereas plagioclase and whole-rock ages from a gabbro gave
Paleocene to Middle Eocene ages. By comparison with the
Lok Ulo complex of Central Java (Parkinson et al. 1998), a
Cretaceous age for the ophiolitic rocks is likely.
The Middle and possibly Upper Eocene sediments of the
Ciletuh area have been assigned to the Ciletuh Formation,
between 400 m and 1500 m thick, and are generally regarded
as having been deposited on Cretaceous basement (van
Bemmelen 1949; Marks 1957; Sukamto 1975; Endang Thayyib
1977; Martodjojo et al. 1978; Schiller et al. 1991). Most authors
have interpreted the contact to be unconformable, although
Martodjojo et al. (1978) suggested that the Ciletuh Formation rested conformably on the melange complex and van
Bemmelen (1949) reported local thrust contacts between the
pre-Tertiary and the Ciletuh Formation. The Ciletuh Formation
has been separated into two main lithofacies (van Bemmelen
1949; Marks 1957; Sukamto 1975; Schiller et al. 1991). The
pre-Tertiary rocks are unconformably and transgressively covered by the Eocene: sandstones and conglomerates with boulders of the Pre-Tertiary, mudstones, breccias and graywackes
(van Bemmelen 1949). Schiller et al. (1991) described a
quartzose lithofacies composed of mostly quartz and a wide
variety of lithic rock fragments; and a less pervasive volcanic
lithofacies composed entirely of volcaniclastic sediments.
The rocks of the Ciletuh Formation were suggested by van
Bemmelen (1949) to be shallow-marine deposits, but Schiller
et al. (1991) reinterpreted them as a sand-dominated submarine
fan deposited in deep water in a forearc or interarc position.
About 20 km northwest of Ciletuh in the Bayah area there
are more than 1000 m of Eocene quartzose conglomerates, sandstones and mudstones of the Bayah Formation
(Koolhoven 1933; Sukarna et al. 1993) divided into a southern
and northern facies. The southern facies is predominantly
conglomerates and sandstones, and van Bemmelen (1949)
remarked on its similarity to the Ciletuh quartzose lithofacies.

161

The northern facies, or Cipageur Member (Sukarna et al. 1993),

consists of clays and quartz sandstones with intercalated limestones of Late Eocene age (Koolhoven 1933). The Bayah
Formation is interpreted as fluvio-deltaic and shallow-marine
deposits; van Bemmelen (1949) also noted the presence of
some tuff intercalations. Schiller et al. (1991) suggested the
Bayah Formation is a shallow-water equivalent of the Ciletuh
Formation. Koolhoven (1933) noted that a few observations
strongly favour equal age, although van Bemmelen (1949)
observed that the southern facies of the Bayah Formation is
probably younger. This is reportedly the case in the Ujung
Kulon-1 well (offshore SW Java), where the Upper Eocene
Bayah Formation lies immediately above a heavily tectonized
and intensely fractured basement (Sukarno et al. 1999), indicating that sedimentation probably commenced later in this part of
Java than in the Ciletuh area.
Just west of Bayah, between the Cimadur and Cidikit Rivers,
the Upper Eocene Cicarucup Formation is exposed in the cores
of anticlines (Sukarna et al. 1993). Its thickness was estimated
by Koolhoven (1933) to be at least 400 m, although this is
uncertain because of faulting and folding (Sukarna et al. 1993).
The formation includes andesitic conglomerates, quartz sandstones, clays, limestones and thick tuff units deposited in a
littoral to shallow-marine environment (Koolhoven 1933;
Sukarna et al. 1993). Koolhoven (1933) suggested that the
formation indicated that West Java was largely emergent in the
Late Eocene, in contrast to the Middle Eocene, and van
Bemmelen (1949) observed that andesitic detritus indicated
that volcanic activity had begun in this part of West Java earlier
than elsewhere in Java.
In the Ciletuh area, the Ciletuh Formation is overlain by
volcanic rocks of the Jampang Formation which are part of the
Old Andesites of Dutch workers. The formation is considered
to be Lower Miocene based on biostratigraphic analysis of
limestone lenses in the formation (Sukamto 1975; M.
Boudagher-Fadel pers. comm. 2007) and was estimated by
Dutch workers (e.g. Verbeek & Fennema 1896; Koolhoven
1933; van Bemmelen 1949) to be up to 5 km thick. There must,
therefore, be a major break as the Upper Eocene and Oligocene
is missing. The formation consists largely of basaltic to dacitic
volcanic breccias, tuffs and lavas. In the Ciletuh area the
Jampang Formation forms a dramatic amphitheatre, up to
500 m high, with steep-sided cliffs, particularly along the
northern boundary, surrounding a window of the older Ciletuh
and Ciemas formations (Clements & Hall 2007). Close to the
coast at Ciletuh Bay are well-bedded tuffs and breccioconglomerates several metres thick, which are predominantly
basaltic andesites and andesites. These types of rock are
common in the Southern Mountains of West Java and represent the most proximal deposits of the old volcanic arc and
must have been deposited relatively close to the volcanic
centres. Thus, for the Early Miocene, the position of the
volcanic arc is known; for earlier times it is less clear.
Interpretation of the CiletuhBayah region
All previous explanations of the geological relationships in the
CiletuhBayah area of West Java have been essentially stratigraphic and autochthonous. Schiller et al. (1991) give a clearly
illustrated explanation of the Eocene, shown in Figure 3. They
suggested the quartz-rich material of the Bayah and Ciletuh
formations was derived from a Sundaland continental source to
the north. They interpreted the Bayah Formation as shelf-edge
deposits and their Ciletuh Formation as a sand-dominated
submarine fan deposited in deeper water to the south in a
forearc or interarc position. Possible equivalents of these

162

B. Clements et al.

Fig. 2. West Java, Ciletuh region: SRTM image, key map and cross-sections showing the structural interpretation discussed in this paper.

Thrusting of a volcanic arc: Java

sequences in the offshore region north of West Java are
(?)Upper EoceneLower Oligocene proximal alluvial deposits
and associated lacustrine shales (Pertamina 1996a). These are
the lowermost parts of the Talang Akar Formation and rest
directly on basement rocks. Equivalents of the thick Middle and
Upper Eocene sequences exposed on land in West Java are
probably only very thin or not present in the offshore to the
north. It appears that sedimentation started earlier at the
margins of Sundaland in SW Java than further north (Clements
& Hall 2008).
The Ciletuh amphitheatre has been interpreted as an inlier
with Eocene and older rocks in the centre, overlain unconformably by the Miocene Jampang Formation. Our observations
suggest a number of problems with the autochthonous interpretation. The character of the two Middle Eocene lithofacies
in the Ciletuh Formation is quite different. One is quartzdominated and the other is composed of volcaniclastic debris.
There is no transition observed between the two lithologies
and, in the Ciletuh area, where the two lithofacies are found
close to one another in outcrop, there is no sign of mixing of
quartz-rich and volcaniclastic material. For this reason we
assign the two lithofacies to different formations (Clements &
Hall 2007): a volcaniclastic Ciletuh Formation and a quartz-rich
Ciemas Formation. The material in the Ciletuh Formation is
almost all derived from ophiolitic basement, or volcaniclastic
sources, with rare epidote amphibolites and shallow-marine
limestones; quartzose material is typically chert fragments and
no material resembling the quartz of the Ciemas Formation is
seen. Blocks of Nummulites limestones, and some volcaniclastic
sandstones, were incorporated when incompletely lithified and
show soft sediment deformation typical of rapidly slumped
material. In the interpreted submarine fan of Schiller et al.
(1991) not only is the most distal material completely different
from the supposed source and more proximal material, it is also
the coarsest (blocks up to 10 m across) and is highly angular.
We suggest these two formations are better interpreted as
contemporary but deposited in separate settings (Fig. 3). The
Ciemas Formation, as suggested by Schiller et al. (1991), does
represent a transition from shelf-slope to deeper-water fan.
Quartz-dominated material was derived from the Sundaland
continent to the north and deposited on the shelf, some of
which was transported into deeper water. Much further south
in the forearc was the site of formation and deposition of the
volcaniclastic Ciletuh Formation. Many features of the formation indicate active tectonism. The size and angularity of blocks
require steep submarine fault scarps, exposing basement, down
which was carried partly lithified volcaniclastic material and rare
shallow-marine limestones as debris flows, into a rift in which
basaltic lavas were erupting. The association of deep-marine
deposits, active faulting of basement and steep slopes,
accompanied by volcanic activity, slumping and synsedimentary deformation is interpreted here as due to extension
in a deep-marine forearc setting accompanying the onset of
subduction and the initiation of the volcanic arc. Such features
can be observed today in places such as Tonga, or the
IzuBoninMarianas arcs. The original large distance between
the Ciemas and Ciletuh formations is now reduced to zero. We
interpret the juxtaposition to be the result of major northward
thrusting of the Southern Mountains arc. We suggest the thrust
front follows the northern edge of the Jampang Formation
along the Cimandiri Valley (Fig. 2). In the Ciletuh area, the
contact between the Jampang Formation and the underlying
Ciletuh Formation is a thrust and the amphitheatre is therefore
not an inlier but a thrust window.
We suggest that the thrust hypothesis offers a simpler
explanation of apparent stratigraphic complexities requiring

163

large vertical movements and relative sea-level changes. The

field relationships can be explained assuming normal stratigraphic contacts, unconformities and special arguments about
sediment distribution but, in our view, they require some
unlikely patterns of uplift and subsidence. For example, the
absence of the Upper Eocene and Oligocene at Ciletuh can be
explained if the area was elevated in the Late Eocene and
became a structural high on which no rocks were deposited.
This would require a rapid emergence from several kilometres
of water depth and a major tectonic event. There is no evidence
on seismic lines a short distance to the west in the offshore
Malingping block (Keetley et al. 1997; Yulianto et al. 2007) for
such an event. Indeed, less than 20 km to the north of Ciletuh,
in the Bayah area, there is a thick sequence of Upper Eocene
shallow-marine quartz-rich clastic rocks, implying continued
subsidence. If the Ciletuh block was a rapidly elevated topographic high, it would have been a source of sediment, yet there
is almost no sign of Ciletuh-derived sediment in the Bayah
Formation, and relatively small amounts of volcaniclastic or
basement material are present in Oligocene formations north of
the Cimandiri River.
The character of the Jampang Formation indicates a calcalkaline arc, suggesting a prolonged period of volcanism evolving from the deep-water basalts of the Middle Eocene Ciletuh
Formation to the largely emergent andesites and pyroclastics of
the Jampang Formation. The apparent absence of Upper
Eocene and Oligocene volcanic material has led to suggestions
that volcanic activity, and possibly subduction, did not begin
until the Late Oligocene (e.g. Hamilton 1988) although, as
noted above, van Bemmelen (1949) observed that volcanic
activity had begun in the Bayah area in the lower part of the
Paleogene, that is earlier than in the rest of Java. We suggest
that the paucity of volcanic debris to the north was, in part,
because during most of the Palaeogene the Southern Mountains
arc in West Java was submerged, relatively non-explosive, but
most importantly, relatively distant from the Bayah area and
Cimandiri Valley. The Southern Mountains Arc and the shelf
sequences dominated by quartzose Sundaland-derived material
were originally separated by tens of kilometres rather than the
few kilometres of the present day, and have been brought
closer together or juxtaposed by northward thrusting.
DEFORMATION ELSEWHERE IN SOUTH JAVA:
WEST JAVA
Northern Jampang Plateau
The plateau that forms the amphitheatre around Ciletuh Bay
extends towards the southeast, east and northeast forming the
Jampang Plateau (Figs 2A and 4). The Jampang Formation
comprises volcanic breccias, andesite and basalt flows, tuffs,
lapilli tuffs, pumiceous tuffs, volcaniclastic sandstones and
mudstones and intercalated limestones. Dips of bedding in this
central part, and in many other areas of the Southern Mountains, are broadly toward the south at low angles (515).
However, in the northern part of the Jampang Plateau, near the
northern edge of the Jampang Formation, these gentle southward dips are not observed and, instead, steep dips, folding and
thrusting are typical. Figure 4 is a geological map of the area
(Sukamto 1990); the Jampang Formation is exposed over a
large part of the map, terminating in the north along the
Cimandiri Valley and to the east beneath the volcanic products
of the Upper Miocene Beser Formation. Two large anticlines
are shown and are further highlighted by interbedded breccia
and tuff members of the formation. These structures trend
WNWESE and are associated with reverse faults that dip
towards the south and southwest. We suggest that these folds

164

B. Clements et al.

Fig. 3. West Java, Ciletuh region. (A) Interpretation of the Ciletuh Formation of Schiller et al. (1991) compared to (B) the new interpretation
of the Ciemas and Ciletuh formations discussed in this paper.

Thrusting of a volcanic arc: Java

165

Fig. 4. Geological map of the northern Jampang Plateau, West Java. Cikarang and Jampang members highlight EW-trending folds. These verge
northwards and are associated with reverse faults (after Sukamto 1975).

are close to the thrust front and formed as the result of

thrusting of the Jampang Formation northwards.
The Middle Miocene (van Bemmelen 1949; Marks 1957)
Lengkong Formation is a thin (c. 200 m), well-bedded sequence
of calcareous and non-calcareous tuffaceous epiclastic volcanogenic sandstones, mudstones, Globigerina marls and sandy limestones that were deposited above the Jampang Formation in
this area. These are marine deposits that were sourced from the
surrounding arc rocks and represent minor subsidence of the
volcanic arc once volcanism ceased. The Lengkong Formation
is preserved only in the cores of EW-trending synclines of the
Jampang Formation, and preserves the same dips as the
surrounding strata, suggesting that deformation post-dates their
deposition.

(Koolhoven 1933) and comprises predominantly OligoMiocene volcanic rocks. Eocene and Oligocene sedimentary
rocks are also exposed; no significant deformation is observed.
The NZ comprises a conformable, moderately deformed
Miocene sequence of volcanogenic, siliciclastic and carbonate
rocks; the zone is characterized by numerous diorite intrusions;
a number of asymmetric folds exist and structures trend EW
to ESEWNW (Koolhoven 1933). In the south of the area, the
relatively undeformed Lower Pliocene Cimanceuri Formation
and Malingping Tuff rest unconformably over formations that
were subject to this major phase of deformation (Clements
2008), and include the Middle Miocene Sareweh Formation.
This major phase of deformation is therefore considered to
have occurred during the Late Miocene.

Bayah region
Over much of the Bayah region steep dips and repetition of
lithologies due to faulting are common. The region can be
divided into three WNWESE-trending zones, the Southern
Zone (SZ), Central Zone (CZ) and Northern Zone (NZ). The
SZ comprises predominantly Eocene to Lower Miocene sedimentary rocks; Koolhoven (1933) drew attention to the lack of
andesites. These include the Upper Eocene quartz-rich sandstones and coals of the Bayah Formation. These coals are
worked locally but operations are always small because of
consistently steep dips (>40), and discontinuity of coals, some
of which are due to thrusting (Ziegler 1916). Oligocene coals of
the Cimandiri and Cikarang coal fields in the Bayah region are
part of the Cijengkol Formation, have similar steep dips and are
located within northerly verging asymmetric anticlines and
synclines. The SZ is intensely folded and a number of reverse
faults exist, typically trending EW to ESEWNW. The MZ is
separated from the SZ by fault and reverse fault contacts

Cimandiri Valley, Cikalong Formation

Lower Oligocene clastic sedimentary rocks are exposed over
an area of several square kilometres southwest of Sukabumi
near the village of Warungkiera. They were assigned to the
Rajamandala Formation by Sukamto (1975) and the Ciletuh
Formation by Kusumahbrata (1994); we assign them to the
Cikalong Formation (Clements & Hall 2007; Clements 2008).
There are pebbly conglomerates with highly rounded pebbles
up to 3 cm long, bedded quartz-rich sandstones, calcareous
sandstones and limestones, tuffaceous sandstones and dark
mudstones. Channels, load casts, normal grading and fluid
escape structures suggest rapid deposition. There are several
hundred metres of sediments, although observed structural
repetition by folding and faulting means that thickness estimates are uncertain. The rocks are deformed and, throughout
the area, bedding dips are consistently greater than 60; along
the Cimandiri River quartz-rich sandstones and mudstones
strike approximately 055.

166

B. Clements et al.

Similar rocks are exposed further east, to the south of

Cianjur and at Padalarang. At these localities the Cikalong
Formation lies beneath folded Oligo-Miocene limestones of the
Rajamandala Formation, in anticlinal cores (see below). Rare
limestone olistoliths are present both at Cikalong and in the
Cisukarama valley, south of Cianjur. These are typically coralline limestones with abundant foraminifera of Middle Eocene
age (Clements 2008). The presence of agglutinated foraminifera
in Cikalong Formation mudstones (P. Lunt, pers. comm. 2006)
suggests a deep-water environment. We suggest that the
shallow-water limestones were transported as olistoliths into
deeper water down a steep slope south of a narrow shelf.
Cikalong Formation rocks are exposed along the line of
the Cimandiri Fault, which extends from Pelabuhanratu to
Bandung (Schiller et al. 1991; Dardji et al. 1994; Martodjojo
1984, 2003). This is also the northern edge of the OligoMiocene Jampang Formation and a steep scarp of Southern
Mountains volcanic rocks rises abruptly from the southern
bank of the Cimandiri River. The line of the valley has been
interpreted as a strike-slip fault (Dardji et al. 1994) that
influenced sedimentation in the Eocene and/or Oligocene
(Martodjojo 1984, 2003; Schiller et al. 1991). It is often said to
be an active strike-slip fault and has been traced offshore
(Schlter et al. 2002; Susilohadi et al. 2005), although on SRTM
(NASA Shuttle Radar Topography Mission) images it does not
have the sharp linear character of strike-slip faults, such as the
Palu-Koro Fault in Sulawesi or the Sumatran Fault Zone. We
suggest it is a thrust, and is not an active fault. The field
relationships in the Cimandiri Valley are ambiguous and can
be interpreted to indicate either Early Miocene thrusting or
post-Middle Miocene thrusting.
Rajamandala Ridge
The Rajamandala Formation includes shallow-marine algal,
reefal and reworked reefal limestones with abundant shell, coral
and foraminiferal material of Late Oligocene to Early Miocene
age (Baumann et al. 1973; M. BouDagher-Fadel, pers. comm.
2006; Clements 2008). The limestones form a prominent ridge
that can be traced ENEWSW from just west of Bandung to
the south of Cianjur where the ridge terminates and the
limestones disappear under younger rocks. Similar limestones
along-strike in the Sukabumi area are generally accepted to be a
continuation of the Rajamandala Formation. The limestones
dip steeply and are tightly folded in the Rajamandala area west
of Bandung; the major fold is overturned to the north and the
northern limb is locally thrust northwards over younger rocks
(Fig. 5). The ridge was described by van Bemmelen (1949) as a
steep anticline or anticlinorium overturned to the north, and he
reported that the core of Oligocene flysch has been squeezed
out diapirically on the road to Cianjur to form a northward
overthrust. In places elsewhere, the structure may be more
complicated, with low dips on the limestones, but generally dips
are steep, folds verge towards the NW and there is thrusting of
Rajamandala Limestone over the LowerMiddle Miocene
Citarum Formation towards the NW.
Cimapag Formation
On the south coast, in the Bayah region, between Tanjung
Layar and Ujung Lengonwaru the Tuff Member of the Cimapag
Formation (Sudjatmiko & Santosa 1992) is well exposed. The
Cimapag Formation is the equivalent of the upper parts of the
Jampang Formation in the Southern Mountains. At Tanjung
Layar there is a spectacular coastal section with 100 m of
bedded units exposed in small cliffs and the modern wave-cut

Fig. 5. West Java, Rajamandala Limestone. Map and cross-sections

showing folding and thrusting of limestones west of Bandung (see
Fig. 1 for location).

platform. At the base of the section are well-bedded volcaniclastic sediments which are typically laterally discontinuous and
channelized. There are many conglomeratic debris flows which
contain clasts of tuffaceous sandstones, calcareous sandstones,
and large boulders of basalt that appear to have been dumped
into finer material. The sandstones are cross-bedded, have
scoured bases with load casts and contain abundant plant and
charcoal fragments up to several centimetres long, and common vertical and horizontal burrows. There are many features
indicating rapid deposition, perhaps as lahars. These rocks pass
up into a number of thickly bedded ignimbrites. These are
1015 m thick and typically have a breccio-conglomeratic
pumice-rich basal unit with an erosional base, containing some
reworked angular lithic, and well-rounded siliciclastic and calcareous clasts up to 40 cm across. The ignimbrites grade up
into several metres of finer block and ash material with diffuse
lamination which, in turn, grade into several metres of fine ash
with very fine convolute lamination. Spectacular flame structures seen in the unit indicate the direction of transport was
broadly to the west. The rocks are pyroclastic flow deposits,
overlain by ash fall layers which were later deformed by
products of subsequent explosive events; there are at least four
of these events recorded at Tanjung Layar. These rocks have
been deformed since the Early Miocene since they are now
sub-vertical and strike eastwest parallel to the coast.
Cinambo and Halang formations
To the northeast of Bandung, and in a more extensive area that
extends further westwards towards Bumiayu in Central Java,

Thrusting of a volcanic arc: Java

the Cinambo and Halang formations are well exposed. The
Middle Miocene Cinambo Formation comprises well-bedded
volcanogenic and calcareous sandstones and mudstones that
are interpreted as turbidites. Calcareous sandstones contain
foraminiferal, coral, algal and shelly detritus. The Upper
Miocene Halang Formation comprises predominantly volcanogenic conglomerates, sandstones and siltstones interpreted as
turbidites and debrites. In an area that extends from Sumedang
to Bumiayu there are some excellent exposures of intensely
deformed Cinambo and Halang Formation rocks. These form
part of a more extensive zone of deformed rocks that extends
towards Bumiayu, Central Java. The formation has been
intensely folded and faulted and bedding is of variable dip and
typically strikes EW; reverse faults commonly dip southward.
All deformation is consistent with northerly tectonic transport
and this deformation must, therefore, be Late Miocene or
younger.
DEFORMATION ELSEWHERE IN SOUTH JAVA:
EAST JAVA
In the southern part of East Java the case for northward
thrusting is less obvious than in West Java. The Eocene to
Miocene arc rocks of the Southern Mountains are relatively well
exposed, but the equivalent age rocks to the north are generally
not exposed, neither is the contact between them and the
Southern Mountains Arc. The structure and stratigraphy of
East Java appears relatively simple (Smyth et al. 2005, 2008). In
the north the southern edge of the Palaeogene Sunda Shelf can
be traced from west to east along the Rembang Hills where
there is now a northward-vergent fold-and-thrust belt. To the
south of the shelf was the Kendeng Basin, largely filled with the
Eocene to Miocene volcaniclastic products of the Southern
Mountains Arc, which was situated south of the basin. The
Kendeng Basin is very poorly exposed, and there is little
subsurface information, but the basin sequence is interpreted
to thicken southwards towards the Southern Mountains
(Pertamina 1996b). The physiographic trough of the Kendeng
Basin deepened towards the south during the Palaeogene.
Today, the rocks of the Southern Mountains Arc dip at a low
angle southwards and their northern contact with the Kendeng
Basin is not exposed. Previously we have interpreted the
structure of the Southern Mountains Arc and southern
Kendeng Basin to be a largely undeformed stratigraphic transition from basin to arc (Smyth 2005; Smyth et al. 2005, 2008),
simply tilted to the south as the result of the uplift and
emergence of East Java, although the Pertamina (1996b)
cross-section shows thrusting at this contact. There are two
locations, at Prigi and Wonosari, near the northern edge of the
Southern Mountains that support a thrust interpretation.
Prigi
In the Southern Mountains around Prigi are carbonate rocks of
the Lower Miocene Campurdarat Formation, volcanic rocks of
the Mandalika Formation and many small hills which are small
andesite intrusions (Samodra et al. 1992a, b; Rahardjo et al.
1995). In a small quarry near Prigi, about 5 km from the nearest
outcrop of the Campurdarat Formation, there are black
limestones with volcaniclastic interbeds assigned to the Prigi
Member of the Campurdarat Formation (Smyth 2005) exposed
in a small quarry. The limestones contain volcaniclastic debris
as well as a diverse faunal assemblage, including corals, red
algae, ostracods, echinoid spines and benthic foraminifera
indicating a Middle Miocene (N8) age. They are interbedded
with graded crystal-lithic volcaniclastic sandstones with scoured

167

bases and tuffaceous muds, which show evidence of slumping

and dewatering. These rocks are interpreted to have been
deposited close to an active volcanic source, where volcanic
debris was mixed with shallow-marine bioclasts on a narrow
shelf, with debris flows carrying all this material downslope.
What is unusual is the strongly deformed character of these
rocks. Nearby exposures of the Mandalika Formation and the
Campurdarat Formation are undeformed and dip gently
towards the south. In contrast, in the Prigi Quarry the beds are
tightly folded, with overturned axial planes dipping eastwards,
and folds plunging at a low angle towards the northwest. No
such folding has been seen elsewhere. We suggest it may be a
result of deformation at or near a thrust at the base of the
Southern Mountains Arc.
Wonosari
At the eastern end of the Southern Mountains is an extensive
area of limestones, modern karst and caves in the Wonosari
Hills to the south of the Jiwo Hills and Batu Agung Escarpment. The Upper Oligocene to Lower Miocene volcanic rocks
of the Batu Agung Escarpment were deposited in a subaerial to
shallow-marine setting. Volcanic activity culminated in a major
explosive eruptive phase (Smyth et al. 2005, 2008) at about
20 Ma, and then ceased for about 10 Ma. During the period
when there was little or no volcanic activity, the Lower to
Middle Miocene Sambipitu, Kepek and Wonosari formations
were deposited unconformably on volcanic rocks of the Batu
Agung Group (Smyth 2005). The Sambipitu, Kepek and
Wonosari formations dip at less than 10 towards the south,
much less steeply than the underlying volcanic rocks of the
Batu Agung Group (2535).
The Sambipitu and Kepek formations interdigitate and have
been interpreted as the deposits of the Wonosari Trough
(Lokier 2000a, b). This trough is suggested to have been a
narrow (510 km wide and 45 km long), deep (c. 1 km) basin
bounded to the north by the Batu Agung Group volcanics and,
to the south, by a fault-bounded block on which the platform
carbonates of the Wonosari Formation were deposited (Lokier
2000a, b). The carbonate platform supplied abundant material
to the Kepek Formation, a sequence of turbidites containing
redeposited carbonate debris and volcaniclastic material, on its
northern flank. To the north of this was the Sambipitu
Formation, a series of calcareous volcanogenic turbidites that
now occupies the area of low topographic relief of the northern
Wonosari Plain, south of the Batu Agung Escarpment. These
rocks have more volcanic debris, less carbonate and have a
more distal character than the Kepek Formation.
There are some problems with this simple stratigraphic
interpretation. The southern side of the Wonosari Trough is
preserved and there is a transition to the Wonosari Platform,
but on the north side of the trough there is no obvious
transition from the deep-water deposits of the Sambipitu
Formation to the eroding volcanic source area inferred to the
north. The Kepek and Sambipitu formations record a significant increase in water depth at the site of the previously
emergent Early Miocene arc. The dimensions of the Wonosari
Trough imply rapid localized subsidence immediately south of
the arc following the cessation of arc activity. For its depth the
trough is exceptionally narrow and, although there are deep
elongate basins in the present-day Java forearc (van der Werff
1996; Kopp et al. 2002), they are much wider.
These problems could be explained if the Wonosari Platform has been thrust northwards, stacking the underlying
Kepek and Sambipitu formations, and reducing the apparent
width of the basin. This interpretation is consistent with the

168

B. Clements et al.

Fig. 6. Central Java, Karangsambung

region. Geological map and crosssection (simplified from Asikin
et al. 1992) showing north-directed
thrusting and folding of Eocene to
Pliocene formations.

structure of the area, with thrusts parallel to the low dips of

the sediments, but would be difficult to prove without very
detailed dating. Biostratigraphic studies are hampered by diagenetic alteration of carbonate-rich lithologies, lack of critical
age-diagnostic faunas, and faunal reworking.
DEFORMATION ELSEWHERE IN SOUTH JAVA:
CENTRAL JAVA
Central Java includes the most extensive area of basement rocks
known in Java, the Lok Ulo Complex (van Bemmelen 1949;
Asikin et al. 1992). This is usually regarded as the imbricated
product of Cretaceous deformation (e.g. Parkinson et al. 1998;
Wakita 2000) at a subduction margin that extended from West
Java, through the Lok Ulo Complex, to the Meratus Mountains.
Parkinson et al. (1998) suggested that subduction ceased in the
Cretaceous after collision of a Gondwana continental fragment
that is now beneath most of West Sulawesi. In contrast,
Sribudiyani et al. (2003) suggested that subduction was continuous from the Cretaceous to the Eocene. They proposed that
there is a continental fragment beneath most of East Java that
collided with a volcanic arc, represented by the Jatibarang
Formation, at the end of the Eocene, and they interpret
the Karangsambung Formation and associated rocks that overlie the Lok Ulo Complex as forearc deposits. Northward
thrusting and largely north-vergent folds are known from the
Karangsambung region (Fig. 6). Thrusts cut Middle Miocene
strata and Miocene and Pliocene formations are folded (Asikin
et al. 1992), suggesting that some of the deformation was more
recent. In marked contrast to West and East Java, in Central
Java there are no Southern Mountains and the Palaeogene arc
rocks are absent.
We too consider there is Gondwana crust beneath parts of
East Java, and dating of recycled zircons (Smyth et al. 2005,

2007) indicates that there is a continental crust beneath the

Southern Mountains that extends west almost to Yogyakarta.
However, like Parkinson et al. (1998), we suggest this fragment
collided in the Cretaceous and terminated subduction at that
time (Smyth et al. 2007). In East Java the oldest Cenozoic
sedimentary rocks above basement, in the Nanggulan area, are
fluviatile quartzose conglomerates and sandstones lacking volcanic debris, and it is difficult to reconcile such a terrestrial
fluviatile setting with the deep forearc basin a short distance to
the west in the Karangsambung area, proposed by Sribudiyani
et al. (2003). This would require a trench to be situated in
Central Java along-strike from a volcanic arc in the Southern
Mountains.
In East Java, the Eocene terrestrial rocks pass rapidly up
into volcaniclastic-rich marine deposits, indicating a deepening
basin to the north of the Southern Mountains Arc, the
Kendeng Trough. In Central Java, the EoceneOligocene
Karangsambung and Totogan formations are also deep-water
deposits, and include scaly clays with Nummulites limestone
blocks, polymict conglomerates and basalt blocks, interpreted
as olistostromes (A. H. Harsolumakso, pers. comm. 2005) or
mud volcanoes (A. J. Barber, pers. comm. 2007). However, in
contrast to similar age rocks further east, they do not contain
abundant volcanic material, but do contain large amounts of
debris derived from continental sources. The stratigraphic
observations in East and Central Java can be reconciled if the
Karangsambung rocks were deposited in a deep basin that
was the western continuation of the Kendeng Trough, but
supplied with sediment derived largely from the Sunda Shelf to
the north, whereas in East Java the Kendeng Trough was
supplied with debris derived predominantly from the Southern
Mountains Arc to the south. We interpret the Kendeng Trough
as a flexural response to volcanic loading by the Southern
Mountains Arc (Smyth et al. 2005; Waltham et al. 2008), which

Thrusting of a volcanic arc: Java

169

Fig. 7. Java: summary tectonic maps showing (A) present-day interpreted structure and (B) schematic pre-thrusting relationships inferred
between the Sunda Shelf, the Palaeogene volcanic arc and sedimentary basins between them.

raises the question: where is the arc that produced the flexural
basin in Central Java? We suggest that the Southern Mountains
arc was originally present in this area but was subsequently
thrust northwards and removed by erosion. Central Java
exposes a deeper structural level than either East or West Java;
the upper thrust sheet of the volcanic arc has been removed
and now exposes the rocks of the deep-marine trough, and the
underlying basement is exposed (Fig. 7).
DISCUSSION
Modern analogue
A model for the Eocene to Early Miocene setting is found at
the present day in the East JavaBali region. Figure 8 illustrates
the key features. To the north is the Sunda Shelf which, during
periods of low sea-level in the Pleistocene, has been an area of
large rivers carrying sediment from south Borneo and Java
towards the east and southeast. At the shelf edge are the islands
of Madura and Kangean, and shallow-water reefs distributed

between them. During periods of low sea-level these carbonates

are likely to be reduced in area, and possibly destroyed, whereas
in periods of highstands they become more widespread.
Immediately to the south of the shelf edge is an elongate
basin (the Bali basin) orientated westeast. South of Madura the
basin is almost full of Plio-Pleistocene sediments (Pertamina
1996b), but, to the east, north of Bali, the basin contains much
less sediment and is up to 1.5 km deep, with very steep slopes to
the north and south. This basin extends to the east north of Bali
and Lombok where it is an unfilled physiographic trough with
depths up to about 1.5 km. The basin is up to 100 km wide and
can be traced along the arc for at least 500 km. South of the
basin and trough are the volcanoes of East Java, Bali and
Lombok. We suggest this feature is the equivalent of the
Eocene to Miocene Kendeng Trough in East Java, i.e. predominantly a flexural response to volcanic arc loading to the south.
The basin runs parallel to the arc and has relatively steep
northern and southern slopes. To the south is a very narrow
shelf adjacent to the volcanoes of Bali and Lombok.

170

B. Clements et al.
and olistostromes, possibly including blocks of limestones
eroded from the emergent carbonate islands. All these deposits
would be very similar to those known from the Eocene to
Miocene of West Java, such as the Ciemas Formation submarine fan and the Cikalong Formation olistostromes, and of
Central Java, in the Karangsambung and Totogan formations.

Fig. 8. East Java: topography from SRTM and GTOPO30 and

bathymetry from Sandwell & Smith (1997) to show modern analogue of the relationships between Sunda Shelf, the volcanic arc and
sedimentary basins in the Palaeogene, discussed in text. Surmised
Pleistocene rivers are based on Voris (2000).

During periods of explosive volcanic activity, volcanic

debris would be fed rapidly into the basin as debris flows and
turbidites. South of Madura it is likely that the basin was filled
from the west by debris carried by rivers draining East Java,
from the south by volcanic debris derived straight offshore
from the volcanoes, and from the north by carbonate carried
downslope from Madura; the greater part of the basin fill is
likely to be reworked volcanic material. The basin fill is likely to
be very similar to the fill of the Eocene to Miocene Kendeng
Basin north of the Southern Mountains Arc.
On the north side of the basin is the wide shallow-marine
Sunda Shelf. During periods of highstands, such as the present
day, much of the sediment carried downslope into the basin is
probably carbonate derived from the islands, such as Kangean
and Madura, and the carbonate reefs between the islands.
During periods of lowstands, such as the Last Glacial Maximum, most of the shelf would have been emergent and there
would have been much more quartzose continental clastic
material carried by rivers, probably from greater distances
within Sundaland, such as south Borneo and West Java. When
these rivers carrying coarse rounded quartzose clastic material
reached the coast, there was a rather narrow shelf and a steep
slope, and much of this material would have been transported
rapidly downslope into deep water, particularly during storms.
There would be sandstones and conglomerates, debris flows

Thrusting of the arc

In West Java the flexural basin is almost completely missing
(Fig. 7), but its deposits are preserved, mainly in a narrow zone
between the Jampang Formation of the Southern Mountains
Arc and the shelf sequences north of the Cimandiri Valley.
They are commonly steeply dipping but, despite this, and
evidence of locally deep-water deposition, they are regarded as
autochthonous deposits, and variations in interpreted water
depths have been suggested to be due to contemporaneous
vertical tectonic movements and changing eustatic sea-level
(van Bemmelen 1949; Sukamto 1975; Martodjojo 1984; Susilohadi et al. 2005; P. Lunt pers. comm. 2006). We suggest that
such interpretations are improbable. They require unlikely rapid
vertical movements, invoke narrow elongate physiographic
features unlike anything observed around Java at present, and
ignore evidence of deformation. We suggest that all these
features can be explained using the modern analogue for the
setting outlined above, with an elongate flexural basin formed
north of the volcanic arc (which we term the Cimandiri Basin;
Fig. 7) that has subsequently been overthrust by the Southern
Mountains Arc. This suggests total northward displacement of
the order of 50 km to 100 km.
In Central Java, the deposits of the flexural basin, which we
term the Karangsambung Basin (Fig. 7), are preserved in the
Karangsambung and Totogan formations, and overlain by the
Miocene to Pliocene volcaniclastic turbidites that later filled
the basin. These rocks are thrust northwards, but the overlying
overthrust volcanic arc has been almost entirely removed by
erosion on land. It is possible that in Central Java there was no
equivalent of the Southern Mountains Arc of West and East
Java and hence no significant overthrust arc. Volcanic arcs do
often have significant gaps between volcanoes for uncertain
reasons, and the load imposed by the volcanoes in West and
East Java may have been sufficient to produce the flexural
basin in Central Java. However, we suggest that it is more likely
that in Central Java a deeper structural level is seen, as
suggested by the more extensive exposures of pre-Cenozoic
basement, and that the remnants of the arc are present offshore
to the south.
In East Java, the deposits of the flexural basin are preserved
in the Kendeng Basin (Fig. 7). Although the basin had a similar
form to that in West and Central Java, the basin fill is different.
In West and Central Java most of the fill was sediments carried
south from the Sunda Shelf, and the Cimandiri and Karangsambung basins contain mainly quartzose material with subordinate volcanic debris. In East Java there was an extensive
carbonate shelf to the north of the Kendeng Basin and
relatively small amounts of continental clastic material were
carried south into the basin, probably mainly during lowstands.
Most of the fill, therefore, came from the volcanic arc to the
south, which for much of the Oligocene and Early Miocene
was emergent and explosive, providing immense amounts of
volcanic debris, particularly abundant volcanic quartz. Because
of poor exposure of the Kendeng Basin sequences, it is difficult
to estimate the amount of thrusting along its southern edge. We
estimate that there was only a few kilometres of thrusting of the
Southern Mountains Arc in East Java.

Thrusting of a volcanic arc: Java

Age of thrusting
The thrusting is Middle Miocene or younger, but at present we
cannot be certain of its exact age, nor if there was more than
one episode of thrusting. This is largely due to the absence of
critical dates at thrust contacts; in most cases it is possible only
to determine that rocks older than Early Miocene have been
thrust. A Middle Miocene age could indicate that thrusting is
linked to the termination of the Palaeogene phase of arc activity
on Java. There was regional plate reorganization at this time,
following Australian continental collision in eastern Indonesia,
which is suggested to have initiated counter-clockwise rotation
of Borneo and Java (Hall 1996, 2002), and contributed to
cessation or diminution of arc volcanic activity (Macpherson &
Hall 1999, 2002) from Java eastwards in the Sunda Arc.
In the Cimandiri Valley, field relationships indicate either
Early Miocene or post-Middle Miocene thrusting. In the
Jampang Plateau, field relationships indicate post-Middle
Miocene thrusting. In the Bayah Dome and at Padalarang, field
relationships indicate post-Middle Miocene thrusting and, in
Sumedang to Bumiayu and in the Bayah Dome, field relationships indicate thrusting occurred since the mid-Late Miocene.
In northern Java, van Bemmelen (1949) gives several examples
of thrusting or associated deformation that are Late Miocene or
Pliocene. Thrusting at this time could be associated with the
shift of the location of arc activity to the north from the
Southern Mountains to its present-day position (Smyth et al.
2005; Hall & Smyth 2008). In East Java this is particularly clear
since the modern arc volcanoes follow a simple linear trend,
and the modern arc is parallel to, and about 50 km north of, the
Palaeogene arc. In Central and West Java the arc also moved
north by a similar distance but the modern volcanoes are
distributed over a much wider zone than in East Java. The
cause of the arc shift is unclear. Simandjuntak & Barber (1996)
suggested that a change from extension to contraction could be
a consequence of subduction of the Roo Rise on the Indian
Plate. There are certainly features south of Java that suggest
that collision of the Roo Rise, or similar topographic features, is
important (Figs 1 and 9); the forearc basins to the west and east
disappear, the forearc is elevated, the trench shallows, and there
is a change from subduction accretion to subduction erosion
close to the trench (Kopp et al. 2006). However, it is not known
when the Roo Rise arrived at the trench, nor why this would
have caused the arc to shift north throughout Java.
Finally there are a number of mud volcanoes towards the
east of the Kendeng Basin. Best known perhaps is the Lusi mud
volcano, which is located at the western margin of the Strait of
Madura, at Sidoarjo (Fig. 1) and has erupted unabated for more
than two years, flooding an area in excess of 7 km and
displacing more than 25 000 people (Mazzini et al. 2007; Davies
et al. 2008; Tingay et al. 2008). Mud volcanoes are the surface
expression of overpressure, which commonly reflects rapid
sedimentation or thrusting. Mud volcanism in East Java is an
indication of active tectonic change in the recent past.
Structural division of Java
We suggest Java can be separated structurally into three distinct
parts that broadly correspond to the regions of West, Central
and East Java (Fig. 7). As explained above, in West and East
Java the overthrust volcanic arc is still preserved, whereas
Central Java displays deeper structural levels below the volcanic
arc, which has been largely removed by erosion. However, the
most important differences are between West-Central and East
Java. The East Java Palaeogene Southern Mountains volcanic
arc was built on continental crust (Fig. 10; Smyth 2005; Smyth
et al. 2005, 2007, 2008). This had two important consequences:

171

Fig. 9. (A) Major features of the Java and forearc region. Topography from SRTM and GTOPO30 and bathymetry from Sandwell &
Smith (1997). Position of MuriaProgo lineament is marked by the
line of black dots and 2006 Yogyakarta earthquake shown. (B) Sharp
change in S-wave velocities beneath East and Central Java at
3570 km depth range (indicated by black arrows; S-wave tomographic model). Question mark conveys uncertainty in interpretation
of S-wave anomalies from Widiyantoro (2006). (C) Bouguer
gravity map of Java and (D) our shelf edge and that of Hamilton
(1979).

first, volcanoes were built from a base on land or close to

sea-level and, hence, were emergent at their earliest stages; and,
secondly, the volcanic activity was explosive Plinian-type for
most of the Eocene to Early Miocene. The volcanic arc
provided large quantities of debris in the form of volcanic ash
to the Kendeng Basin from the Late Eocene until the Early
Miocene. In contrast, the evidence from Ciletuh suggests that
the volcanoes of West and Central Java were built on ophiolitic
basement, the earliest volcanic activity was in a deep-water
setting and the volcanic activity was basaltic and non-explosive.
It was not until the Late Oligocene that the volcanoes had
become emergent and were more explosive, as the crust

172

B. Clements et al.
younger thinner ophiolitic crust and their underlying mantle.
The orientation of the lineament is quite similar to many
cross-arc normal faults in East Java, and the normal fault that
resulted in the 2006 Yogyakarta earthquake (Widiyantoro 2006;
Fig. 10) suggests active arc-parallel extension at the present day.
Satyana (2007) suggested that a NESW fault in a similar
position to the MuriaProgo lineament is one of a conjugate
pair of strike-slip faults that bound Central Java. There is no
surface evidence of strike-slip movement on either of the faults,
but they have a similar orientation to the cross-arc normal faults
in East Java, and could be extensional faults. Normal faulting
on these faults would be consistent with the deeper structural
levels exposed in Central Java and their position could be
determined by changes in the character of the deeper crust, as
suggested above.

Fig. 10. (A) The character of crust beneath East Java and (B) the
extent of the E JavaW Sulawesi basement block at 45 Ma. Note
that latitudes and longitudes are those for 45 Ma (based on reconstructions by Hall 2002).

thickened and the volcanic edifice became more substantial.

Thus, the flexural character of the Kendeng Basin is more
obvious because the basin formed on crust initially at sea-level.
In West and Central Java, the flexural contribution by volcanic
loading to subsidence of the basin north of the arc would have
increased with time, but the Karangsambung and Cimandiri
basins formed south of the shelf edge where there was already
a slope and deepening to the south. Because of this morphology these basins were fed mainly by quartzose debris from
Sundaland until the arc became sufficiently mature to be an
important source of material.
We therefore consider that the most fundamental structural
division in Java is between Central and East Java (Fig. 10), at
the western limit of the old continental basement (Smyth et al.
2008). Smyth (2005) interpreted a major NNESSW lineament
at approximately 110.5E (Figs 9 and 10), which we call the
MuriaProgo lineament. This lineament links several major
features onshore and offshore (Fig. 9): a structural high in the
Java forearc, the centres of three Oligo-Miocene volcanoes (van
Bemmelen 1949) in the West Progo Mountains, Mount Merapi
and the unusual K-rich volcano of Mount Muria. To the east of
this line the basement of the Southern Mountains is old
continental crust (Fig. 10), whereas to the west it is Cretaceous
ophiolitic rocks. Widiyantoro (2006) has observed a sharp
boundary between low and high S-wave velocities in the depth
range of 3570 km below Central Java in his S-wave tomographic model (Fig. 9B). He suggested that this was the
approximate edge of continental shelf during the OligoMiocene, based on Hamilton (1979; Fig. 9D). However, we do
not agree that there was a sudden change in orientation of the
shelf edge and we suggest this seismic discontinuity is more
likely to reflect the difference between old continental crust and

Significance for hydrocarbon exploration

The hypothesis of significant northward thrusting offers some
new play concepts in southern Java. In East Java, Eocene clastic
sedimentary rocks containing coals would be buried more
deeply near the thrust front, but the abundance of volcanic
material in the Kendeng Basin could make reservoir rocks
unattractive for exploration. However, in West Java the arc is
thrust further northwards, the flexural basin sequence is likely
to be shortened and stacked beneath the arc rocks and at the
base of this thrust pile would be quartz-rich clastic rocks with
abundant coals of Eocene and Oligocene age belonging to the
southern Sunda Shelf (Fig. 11). The seismic data from the
offshore Malingping block suggest that the shelf margin
sequences include a number of stratigraphic and structural
traps, and these could be preserved below the thrusts. More
subsurface information is required to test these suggestions.
CONCLUSIONS
After Cretaceous collision of an Australian microcontinental
fragment with the JavaMeratus subduction system, subduction
ceased and there was a passive margin south of Java until the
Eocene (Fig. 10). In the Middle Eocene, subduction resumed
and a new arc developed south of the Sunda Shelf. The load of
volcanoes contributed to the development of a flexural basin
between the arc and the shelf that was between 50 km and
100 km wide, and followed the shelf edge running ENE the
length of Java. From the Late Eocene to Early Miocene the
basin was supplied with quartz-rich clastic sediments by rivers
draining the Sunda Shelf in West and Central Java, with
subordinate amounts of volcanic debris supplied from the arc
volcanoes, which were largely submarine and non-explosive. In
East Java, volcanoes were emergent earlier and erupted explosively and supplied greater amounts of volcanic debris to the
basin to the north. Arc activity ceased for a period in the Early
Miocene and resumed again in the Late Miocene at a location
north of the site of the Palaeogene arc. At some time between
the Early Miocene and Pliocene, the Palaeogene sequence was
thrust northwards, by more than 50 km in West Java, but with
probably much smaller displacements in East Java. Java can be
separated into three structural sectors of West, Central and East
Java. Central Java displays the deepest structural levels of
thrusting and Cretaceous basement is exposed; the overthrust
volcanic arc has been largely removed by erosion but may be
present to the south of Java. In West and East Java the
overthrust volcanic arc is still preserved. In West Java the arc is
now thrust onto the shelf sequences that formed on the
Sundaland continental margin. Northward thrusting of the
Palaeogene volcanic arc rocks of the Southern Mountains offers

Thrusting of a volcanic arc: Java

173

Fig. 11. Regional cross-section through the West Java margin. Forearc structure is from Kopp et al. (2002). Estimates of crustal thickness are
from Hamilton (1979). Inset is a schematic cross-section showing potential hydrocarbon plays (hypothetical accumulations shown in green)
discussed in the text.

a simpler explanation than autochthonous models for stratigraphic and structural relationships in southern Java. Overthrusting of the arc may have contributed to overpressure in the
subsurface, as now well documented for East Java. Structural
and stratigraphic traps beneath the overthrust arc may also offer
new hydrocarbon exploration possibilities as both source and
reservoir lithologies are present.
The authors are grateful to the consortium of oil companies who
have supported projects in SE Asia for many years. Thanks go to
Agus Harsolumakso, Benyamin Sapiie and other ITB geologists for
introducing us to the Karangsambung area, for their co-operation in
our work in Java and for many interesting discussions; Djadjang
Sukarna and the Pusat Survei Geologi (former GRDC); and Eko
Lelono and Lemigas for their co-operation, help and support. Ivan
Yulianto and Edy Slameto and numerous Lemigas counterparts
provided invaluable field support. Peter Lunt helped with discussions
of Java geology, many suggestions for research and practical assistance. Marcelle BouDagher-Fadel, Bob Morley, Bernhard Seubert,
Colin Macpherson and Moyra Wilson have all assisted in different
ways.

REFERENCES
Asikin, S. Handoyo, A. Busono, H. & Gafoer, S. 1992. Geology of the Kebumen
Quadrangle, Jawa. Quadrangle 1401-1. Scale 1:100,000. Geological Research
and Development Centre, Bandung, Indonesia.
Baumann, P. de Genevraye, P. Samuel, L. & Mudjito, S.S. 1973. Contribution
to the geological knowledge of south west Java. In: Indonesian Petroleum
Association, Proceedings 2nd Annual Convention, 105108.
Clements, B. 2008. Paleogene to Early Miocene tectonic and stratigraphic evolution of
West Java, Indonesia. PhD thesis, University of London, London.
Clements, B. & Hall, R. 2007. Cretaceous to Late Miocene stratigraphic and
tectonic evolution of West Java, Indonesia. In: Indonesian Petroleum Association, Proceedings 31st Annual Convention, 87104.
Clements, B. & Hall, R. 2008. UPb dating of detrital zircons from West Java
show complex Sundaland provenance. In: Indonesian Petroleum Association,
Proceedings 32nd Annual Convention, IPA08-G-115, 118.

Dardji, N. Villemin, T. & Rampnoux, J.P. 1994. Paleostress and strike-slip

movement: the Cimandiri Fault Zone, West Java, Indonesia. Journal of
Southeast Asian Earth Sciences, 9, 311.
Davies, R.J. Brumm, M. Manga, M. Ubiandini, R. Swarbrick, R. & Tingay, M.
2008. The East Java Mud Volcano (2006 to Present): An earthquake or
drilling trigger? Earth and Planetary Science Letters, 272, 627638.
Doust, H. & Sumner, H.S. 2007. Petroleum systems in rift basins a
collective approach in Southeast Asian basins. Petroleum Geoscience, 13,
127144.
Endang Thayyib, S. 1977. The status of the melange complex in Ciletuh area,
Southwest Java. In: Indonesian Petroleum Association, Proceedings 6th Annual
Convention, 241254.
Hall, R. 1996. Reconstructing Cenozoic SE Asia. In: Hall, R. & Blundell, D.J.
(eds) Tectonic Evolution of SE Asia. Geological Society, London, Special
Publications, 106, 153184.
Hall, R. 2002. Cenozoic geological and plate tectonic evolution of SE Asia
and the SW Pacific: computer-based reconstructions, model and animations. Journal of Asian Earth Sciences, 20, 353434.
Hall, R. & Smyth, H.R. 2008. Cenozoic arc processes in Indonesia: identification of the key influences on the stratigraphic record in active volcanic
arcs. In: Draut, A.E. Clift, P.D. & Scholl, D.W. (eds) Formation and
Applications of the Sedimentary Record in Arc Collision Zones. Geological Society
of America, Special Publications, 436, 2754.
Hamilton, W. 1979. Tectonics of the Indonesian region. USGS Professional Paper,
1078.
Hamilton, W. 1988. Plate tectonics and island arcs. Geological Society of America
Bulletin, 100, 15031527.
Keetley, J.T. Cooper, G.T. Hill, K.C. Kusumabrata, Y. OSullivan, P.B. &
Saefudin, I. 1997. The structural development of the Honje High, Bayah
High and adjacent offshore areas, West Java, Indonesia. In: Howes, J.V.C.
& Noble, R.A. (eds) Proceedings of the International Conference on Petroleum
Systems of SE Asia & Australia, 2123 May, Jakarta, Indonesia. Indonesian
Petroleum Association, 655665.
Koolhoven, W.C.B. 1933. Geological Map of Java, Explanatory Notes to Sheet 14
(Bayah), 1:100,000. Mining Bureau of the Netherlands Indies.
Kopp, H. Klaeschen, D. Flueh, E.R. Bialas, J. & Reichert, C. 2002. Crustal
structure of the Java margin from seismic wide-angle and multichannel
reflection data. Journal of Geophysical Research, 107, DOI: 10.1029/
2000JB000095.

174

B. Clements et al.

Kopp, H. Flueh, E.R. Petersen, C.J. Weinrebe, W. Wittwer, A. & Meramax

Scientists. 2006. The Java margin revisited: Evidence for subduction
erosion off Java. Earth and Planetary Science Letters, 242, 130142.
Kusumahbrata, Y. 1994. Sedimentology and stratigraphy of the Bayah, Walat and
Ciletuh Formations, SW Java Basins, Indonesia. PhD thesis, University of
Wollongong, NSW, Australia.
Lokier, S.W. 2000a. The development of the Miocene Wonosari Formation,
south Central Java. In: Indonesian Petroleum Association, Proceedings 27th Annual
Convention, 217222.
Lokier, S.W. 2000b. The Miocene Wonosari Formation, Java, Indonesia: volcaniclastic
influences on carbonate platform development. PhD thesis, University of London,
London.
Longley, I.M. 1997. The tectonostratigraphic evolution of SE Asia. In: Fraser,
A.J. Matthews, S.J. & Murphy, R.W. (eds) Petroleum Geology of Southeast Asia.
Geological Society, London, Special Publications, 126, 311339.
Macpherson, C.G. & Hall, R. 1999. Tectonic controls of geochemical
evolution in arc magmatism of SE Asia. In: Proceedings 4th PACRIM
Congress, Bali Indonesia. Australian Institute of Mining and Metallurgy,
359368.
Macpherson, C.G. & Hall, R. 2002. Timing and tectonic controls on
magmatism and ore generation in an evolving orogen: evidence from
Southeast Asia and the western Pacific. In: Blundell, D.J. Neubauer, F. &
von Quadt, A. (eds) The Timing and Location of Major Ore Deposits in an
Evolving Orogen. Geological Society, London, Special Publications, 204,
4967.
Marks, P. 1957. Stratigraphic lexicon of Indonesia. Publikasi Keilmuan Djawatan
Geologi Bandung, Seri Geologi, 31.
Martodjojo, S. 1984. Evolusi Cekungan Bogor. Thesis, ITB, Bandung.
Martodjojo, S. 2003. Evolusi Cekungan Bogor. ITB Press, Bandung.
Martodjojo, S. Suparka, S. & Hadiwasastra, S. 1978. Status Formasi Ciletuh
dalam evolusi Jawa Barat. Geologi Indonesia, 5, 2938.
Mazzini, A. Svensen, H. Akhmanov, G.G. Aloisi, G. Planke, S.
Malthe-Srenssen, A. & Istadi, B. 2007. Triggering and dynamic evolution
of the LUSI mud volcano, Indonesia. Earth and Planetary Science Letters, 261,
375388.
McCaffrey, R. 1996. Slip partitioning at convergent plate boundaries of SE
Asia. In: Hall, R. & Blundell, D.J. (eds) Tectonic Evolution of SE Asia.
Geological Society, London, Special Publications, 106, 318.
Parkinson, C.D. Miyazaki, K. Wakita, K. Barber, A.J. & Carswell, D.A. 1998.
An overview and tectonic synthesis of the pre-Tertiary very-high-pressure
metamorphic and associated rocks of Java, Sulawesi and
Kalimantan, Indonesia. Island Arc, 7, 184200.
Pertamina, B.P.P.K.A. 1996a. Petroleum Geology of Indonesian Basins, Principles,
Methods and Application. Volume III: West Java Sea Basins. Pertamina,
BPPKA, Jakarta.
Pertamina, B.P.P.K.A. 1996b. Petroleum Geology of Indonesian Basins, Principles,
Methods and Application. Volume IV: East Java Basins. Pertamina, BPPKA,
Jakarta.
Rahardjo, W. Sukandarrumidi & Rosidi, H.M.D. 1995. Geological Map of the
Yogyakarta Quadrangle, Java. Geological Research and Development Centre,
Bandung.
Samodra, H. Gafoer, S. & Tjokrosapoetro, S. 1992a. Geology of the Pacitan
Quadranglem, Jawa (1507-4). Geological Research and Development Centre,
Bandung.
Samodra, H. Suharsono, G.S. & Suwarti, T. 1992b. Geology of the Tulugagung
Quadrangle, Jawa. (quadrangle 1057-5). scale 1:100,000. Geological Research
and Development Centre, Bandung.
Sandwell, D.T. & Smith, W.H.F. 1997. Marine gravity anomaly from Geosat
and ERS 1 satellite altimetry. Journal of Geophysical Research, 102, 10 03910
054.
Satyana, A.H. 2007. Central Java, Indonesia a terra incognita in petroleum
exploration. New considerations on the tectonic evolution and petroleum
implications. In: Indonesian Petroleum Association, Proceedings 31st Annual
Convention, 105129.
Schiller, D.M. Garrard, R.A. & Prasetyo, L. 1991. Eocene submarine fan
sedimentation in southwest Java. In: Indonesian Petroleum Association, Proceedings 20th Annual Convention, 125182.
Schlter, H.U. Gaedicke, C. & Roeser, H.A. et al. 2002. Tectonic features of
the southern Sumatrawestern Java forearc of Indonesia. Tectonics, 21,
10471062.

Simandjuntak, T.O. & Barber, A.J. 1996. Contrasting tectonic styles in the
Neogene orogenic belts of Indonesia. In: Hall, R. & Blundell, D.J. (eds)
Tectonic Evolution of SE Asia. Geological Society, London, Special Publications, 106, 185201.
Smyth, H. 2005. Eocene to Miocene basin history and volcanic activity in East Java,
Indonesia. PhD thesis, University of London, London.
Smyth, H. Hall, R. Hamilton, J.P. & Kinny, P. 2005. East Java: Cenozoic
basins, volcanoes and ancient basement. In: Indonesian Petroleum Association,
Proceedings 30th Annual Convention, 251266.
Smyth, H.R. Hamilton, P.J. Hall, R. & Kinny, P.D. 2007. The deep crust
beneath island arcs: Inherited zircons reveal a Gondwana continental
fragment beneath East Java, Indonesia. Earth and Planetary Science Letters,
258, 269282.
Smyth, H.R. Hall, R. & Nichols, G.J. 2008. Early Cenozoic volcanic arc
history of East Java, Indonesia: the stratigraphic record of eruptions on a
continental margin in a tropical setting. In: Draut, A.E. Clift, P.D. & Scholl,
D.W. (eds) Formation and Applications of the Sedimentary Record in Arc Collision
Zones. Geological Society of America Special Publication, 436, 199222.
Sribudiyani, Muchsin, N. & Ryacudu, R. et al. 2003. The collision of the East
Java Microplate and its implication for hydrocarbon occurrences in the
East Java Basin. In: Indonesian Petroleum Association, Proceedings 29th Annual
Convention, 335346.
Sudjatmiko & Santosa. 1992. Geological map of the Leuwidimar quadrangle, Java,
Geological Research and Development Centre, Bandung.
Sukamto, R. 1975. Geologic Map of the Jampang and Balekambang quadrangles, Java
1:100,000. Geological Research and Development Centre, Bandung.
Sukamto, R. 1990. Geology of the Jampang and Balekambang Sheet area, West Jawa
1:100,000. Geological Research and Development Centre, Bandung.
Sukarna, D. Mangga, S.A. & Brata, K. 1993. Geology of the Bayah area:
implications for the Cenozoic evolution of West Java, Indonesia. Bulletin of
the Geological Society of Malaysia, 33, 163180.
Sukarno, Wantoro, Pane, Y.A. Noon, S. & Hulsbos, R. 1999. Well
Cipatulah-1, Malingping Block, Offshore Southwest Java. Unpublished,
confidential well report.
Susilohadi, S. Gaedicke, C. & Ehrhardt, A. 2005. Neogene structures and
sedimentation history along the Sunda forearc basins off southwest
Sumatra and southwest Java. Marine Geology, 219, 133154.
Tingay, M. Heidbach, O. Davies, R. & Swarbrick, R. 2008. Triggering of the
Lusi mud eruption: Earthquake versus drilling initiation. Geology, 36,
639642.
van Bemmelen, R.W. 1949. The Geology of Indonesia. Government Printing
Office, Nijhoff, The Hague.
van der Werff, W. 1996. Variation in the forearc basin development along the
Sunda Arc, Indonesia. Journal of Southeast Asian Earth Sciences, 14, 331349.
Verbeek, R.D.M. & Fennema, R. 1896. Geologische beschrijving van Java en
Madoera. J. G. Stemler Cz, Amsterdam.
Voris, H.K. 2000. Maps of Pleistocene sea levels in Southeast Asia:
shorelines, river systems and time durations. Journal of Biogeography, 27,
11531167.
Wakita, K. 2000. Cretaceous accretionary-collision complexes in central
Indonesia. Journal of Asian Earth Sciences, 18, 739749.
Waltham, D. Hall, R. Smyth, H.R. & Ebinger, C. 2008. Basin formation by
volcanic arc loading. In: Draut, A.E. Clift, P.D. & Scholl, D.W. (eds)
Formation and Applications of the Sedimentary Record in Arc Collision Zones.
Geological Society of America Special Paper, 436, 1126.
Widiyantoro, S. 2006. Learning from the May 27, 2006 Yogya-Central Java
Destructive Earthquake. In: Proceedings, 31st HAGI Annual Convention and
Exhibition, Semarang.
Yulianto, I. Hall, R. Clements, B. & Elders, C. 2007. Structural and
stratigraphic evolution of the offshore Malingping Block, West Java,
Indonesia. In: Indonesian Petroleum Association, Proceedings 31st Annual Convention, 171183.
Ziegler, K.G.J. 1916. Report on the results of mining geological investigations
in South Bantam. Jaarboek Mijnwozen, 1, 97.

Received 12 November 2008; revised typescript accepted 2 March 2009.