-
IPA93-1.1-115
PROCEEDINGS INDONESIAN PETROLEUM ASSOCIATION Twenty Second
Annual Convention, October 1993
N FORE-ARC ZONE OF SUMATRA: CAINOZOIC BASIN-FORMING TECTONISM
AND HYDROCARBON POTENTIAL
D.M. Hall* B.A. D u V
M.C. Courbe** B.W. Seubert** M. Siahaan**
A.D . Wirabudi**
ABSTRACT
In the Bengkulu PSC of onshore and offshore Southwest Sumatra,
localized basins containing four distinct seismic megasequences are
recognized.
The basal, Paleogene, megasequence was deposited as a syn-rift
unit within a series of northeast-trending half graben, probably
segmented by northwest-trending transfer faults. A major
unconformity separates this unit from a late Paleogene to early
Miocene megasequence and appears to mark a change in basin- forming
mechanism from orthogonal extension to possible oblique slip.
According to this model, the transfer faults of the rift system
were rejuvenated by right-lateral oblique slip in the late
Paleogene to early Miocene, thereby superposing local pull-apart
basins on the underlying graben.
These units are succeeded with strong unconformity by a middle
to late Miocene megasequence marking the onset of open marine
deposition within a unified forearc basin. Finally, this unit was
overlain by a dominantly regressive Pliocene to Recent syn-orogenic
megasequence resulting from the main period of uplift and erosion
of the Barisan Mountains. The associated basin inversion of the
older megasequences increases in intensity from offshore toward
this mountain belt.
These results imply that far from accommodating a simple,
homogeneous fore-arc basin, the fore-arc is tectonically
heterogeneous with considerable potential for localised Paleogene
and early Neogene basins.
* Fina Exploration Norway Inc. * * Previously Exploration
Members of Fina Bengkulu S.A
Recent exploration of the Bengkulu PSC, targetting the lower two
megasequences of Paleogene to early Miocene age, implies that such
localized basins within the fore-arc can be prospective for
hydrocarbons. Well results indicate the presence of mature source
rocks and migrated hydrocarbons, and therefore appear to contradict
the widespread assumption that heat flow values in fore-arc areas
are insufficient to allow expulsion and migration of
hydrocarbons.
INTRODUCTION
Fore-arc basins are commonly assumed to be unrewarding areas for
hydrocarbon exploration, a view that appeared to be confirmed by
the results of the first phase of exploration activity in the
Sumatran fore-arc in the late 1970s to early 1980s. During this
period, hydrocarbon indications were limited to uncommercial
methane gas discoveries made by Unocal in the northern part of the
fore-arc, and a minor oil show in a well drilled by Aminoil in the
Bengkulu area of the southern fore-arc. This exploration
concentrated almost entirely on shelfal Neogene plays located on
the basin margins.
Neogene basin development within the northern Sumatran fore-arc
(Figure 1) has also been the subject of a number of non-commercial
regional studies (eg. Karig et al., 1980; Beaudry and Moore, 1985;
Matson and Moore, 1992). Until recently, however, the southern
fore-arc (Figure 2) has not received the same attention, and more
significantly for exploration, even less has been known about
Paleogene basin history.
An exception to this were the seismic and aeromagnetic data
acquired in the Bengkulu area, which indicated the presence of a
localised depocentre of presumed
IPA, 2006 - 22nd Annual Convention Proceedings, 1993
-
320
Paleogene age (Howles, 1986). One possibility was that this
basin could represent the southward continuation of back arc graben
trends known north of the Barisan Mountains. This in turn had
obvious implications for hydrocarbon potential.
It was primarily to evaluate this concept that exploration was
carried out in the Bengkulu PSC from July 1989 to July 1992 by a
group comprising Fina (Operator), Enterprise and British Gas. The
exploration work programme included the acquisition of 3480
kilometers of onshore and offshore seismic, gravity and magnetic
data (Figure 3). Following this, the Arwana-1 exploration well was
drilled to a total depth of 4175m at an offshore location in the
southeast of the PSC.
It is the purpose of this paper to discuss the impact that
interpretation of this dataset has had on the understanding of
basin history, and hydrocarbon potental of the southern Sumatran
fore-arc. In this respect, the results of Arwana-1 are particularly
significant, as the well represents the first substantive
calibration of a basinal Paleogene section anywhere in the Sumatran
fore-arc. Furthermore, the presence in this well of mature source
rocks and significant oil shows, including indications of migrated
oil, challenges some of the conventional views of fore-arc
prospectivity .
REGIONAL SETTING
The Bengkulu PSC comprised an offshore-onshore coastal region
covering a pre-relinquishment area of 16,800 square kilometers in
the southeastern part of the Sumatran fore-arc (Figures 1 and 2).
The PSC was situated landward of the shelf-slope break which
separates the inner shelfal part of the fore-arc, here termed the
Inner Fore-Arc, from the bathymetric deep of the Quter Fore-Arc.
Consequently, water depths within the PSC average 50 meters, and
only exceed this near the southwest boundary of the contract area.
The northeastern part of the PSC included part of the Barisan
Mountains, which in turn are bounded to the northeast by the West
Sumatra Fault. The Barisan Mountains represent an uplifted and
folded complex of sedimentary, igneous and volcanic rocks (Figure
2), and cannot therefore be described solely as a volcanic arc. In
the Bengkulu area, the boundary between the Barisan Mountains and
the coastal plain is sharply defined by a dextral oblique-slip
fault, which appears to be a splay from the main trend of the West
Sumatra Fault.
The setting of the PSC suggests two regional factors which may
have influenced initial basin development. The first relates to the
oblique convergence of the Indian Ocean Plate and Sunda Craton,
which may have
commenced prior to the middle Miocene. Evidence for this are the
Neogene pull-apart basins in the southernmost part of Sunda Straits
(Huchon and Le Pichon, 1984). The formation of these basins has
been explained by the northward movement of the Sumatra Sliver
Plate (Jarrard, 1986), a term which describes the large region of
fore-arc between the West Sumatra and Mentawi dextral strike-slip
faults (Figure 1).
Secondly, it has been noted that the southward trend of
Paleogene back-arc graben such as the Benakat Gulley (de Costa,
1974) align with the Bengkulu area if it is assumed that subsequent
dextral movement along the West Sumatra Fault has been of the order
of 100 kilometers (eg. Howles, 1986).
Consequently, the development of Paleogene to early Neogene
basins in the Bengkulu area was probably influenced by both
extensional and oblique slip tectonics.
PREVIOUS EXPLORATION
Between 1970 and 1972 a total of six offshore wells (Figure 3)
were drilled in the Bengkulu area: four by the Jenny Oil Group and
Marathon in the Mentawi PSC, and two by Aminoil in the
Banten-Lampung PSC. None of these wells reached total depths
greater than about 1960 metres, and in each case the proposed
objective of Miocene carbonate build-ups, overlying what was
interpreted to be volcanic or igneous basement, was water-wet or
absent. The only exception to this is the Bengkulu A-lx well, which
encountered oil shows in a basal carbonate, originally interpreted
as equivalent to the early Miocene Baturaja Limestone of South
Sumatra, but now thought to be earliest middle Miocene.
Subsequent interpretation has shown that only the Bengkulu A-lx
and A-2x wells were located on valid structural closures. The
Mentawi-A1 and Mentawi-C1 wells were drilled on velocity pull-ups
created by overlying late Miocene (Parigi Formation) reefs, whereas
the Bengkulu X-1 and Bengkulu X-2 wells were located on a gravity
high with no clearly-defined structural closure. Furthermore, all
of the wells were located outside the main Paleogene
depocentres.
It is therefore clear that this first phase of drilling did not
fully evaluate the hydrocarbon potential of the area.
DISTRIBUTION OF PALEOGENE- EARLY NEOGENE BASINS
Four Paleogene to early Neogene basins have been identified
within the limits of the original Bengkulu
-
321
PSC. Their location is shown by the basement depth structure in
Figure 4. Depth to basement within basins located in the offshore
area was estimated from the combined interpretation of seismic,
gravity and magnetic data. In the onshore area however, seismic
definition of basement structure is ambiguous owing to
surface-related signal-to-noise problems. Basement interpretation
onshore is therefore mostly based on gravity data.
The onshore North Manna Basin and adjacent offshore South Manna
Basin are located in the southeastern part of the PSC (Figure 4),
and were the prime objectives of data acquisition during the
1989-1992 exploration period. Consequently, these basins are the
main subject of this paper. Based on more limited data coverage,
two further depocentres are tentatively recognized: one located
east of Bengkulu and the other in the northern area of the PSC near
Ketahun.
The North Manna and South Manna Basins are broad half-graben,
which thicken to the northeast (Figure 5 ) . In addition, the North
Manna Basin has been tilted toward the southwest by younger
Plio-Pleistocene inversion. The associated uplift of the Barisan
Mountains has obscured the northern limit of the North Manna Basin,
although the apparent trend of the basin axis suggests that it may
have extended northeastward at least as far as the West Sumatra
Fault. In contrast. the depositional axis of the South Manna Basin
displays a clear northwest trend, offset to the southeast relative
to the North Manna Basin. The two basins are separated by a narrow
median high which also trends northwest, below the present
coastline. Basement depths in the South Manna Basin are interpreted
to exceed six kilometers, approximately the same level as the
subduction trench in the Outer Fore-Arc.
The nature of basement underlying the Inner Fore-Arc Paleogene
basin fill remains uncalibrated by drilling or outcrop exposure.
However, in places, a parallel-bedded seismic facies has been
recognized (Figures 5 and 8), possibly suggesting that the basement
has a sedimentary or metasedimentary rather than crystalline
origin. Possible origins include Cretaceous to Paleocene fore arc
basins or shelfal platform cover sediments deposited on continental
crust. Regardless of origin, it is clear from the contrasting
subsidence histories of the Inner (shelfal) and Outer (basinal)
Fore-Arc that the boundary between the two area5 coincides with a
significant contrast in basement rigidity.
GENERAL STRATIGRAPHY
The lithostratigraphy of the North and South Manna Basins
comprises a variety of volcanic-arc derived sediments interbedded
with marine claystones and
minor carbonate intervals (Figure 6). Biostratigaphic analysis
of the Arwana-1 well within the South Manna Basin indicates a
relatively complete Cainozoic section from the early Oligocene or
possibly late Eocene, which was deposited in an inner to outer
sublittoral environment. The Arwana-1 well provides the only
control of Paleogene stratigraphy in the southern fore- arc region,
as the previous exploration wells drilled in the early 1970s did
not penetrate below the base of the Middle Miocene. In the North
Manna Basin, the Paleogene section remains uncalibrated because the
onshore outcrop provides no reliable age determinations older than
early Miocene.
In both the North and South Manna Basins, an important
stratigraphic boundary occurs at the base of the Middle Miocene,
representing the downward change from regional to localized basin
geometries. This boundary coincides with the base of a widespread
carbonate interval informally referred to in this paper as the N9
Limestone after the equivalent Blow (1969) foram zone. The section
above the base of the N9 Limestone contains a relatively diverse
faunal assemblage, indicating essentially unrestricted access to
the open oceanic environment. Below this level however, the Lower
Miocene to Paleogene is characterized by a less diverse faunal
assemblage, suggesting deposition within a more restricted
basin.
SEISMIC STRATIGRAPHY AND LITHOSTRATIGRAPHY
The Recent to Paleogene stratigraphy of the Bengkulu area can be
further described in terms of four seismic megasequences (Figures 5
to 8), each characterizing a major tectonostratigraphic phase of
basin evolution (ie. sensu Hubbard et al., 1985). Megasequences are
bounded by major seismically-defined stratal surfaces which often
correlate with important changes in external basin controls such as
re-organization of plate movements. Each megasequence is subdivided
into component sequences, the boundaries of which also form
prominent seismic events interpreted as corresponding to changes in
regional relative sea level, basin subsidence or sediment
supply.
Megasequence I (? Late Eocene to early Oligocene)
Megasequence I represents the initial fill of the early Neogene
- Paleogene basins, which was deposited within a complex mosaic of
segmented half graben depocentres. The only direct evidence of
Megasequence I lithologies comes from. the basal 60 metres of
Arwana-1 , which comprise massive volcanogenic intervals,
interbedded with dark brown and grey-green claystones.
-
322
The volcanogenic lithologies are mainly volcanic litharenites,
which petrographic studies of sidewall cores indicate comprise
welded ignimbrite clasts, lithic volcanic clasts and vitriclasts.
Although diagenesis has obscured much of the original rock fabric,
there are some reworked intervals with reduced matrix content. This
is inferred both from thin sections and log-based interpretation of
permeability variations. Faunal and geochemical evidence indicate
that the interbedded dark brown claystones are of organic marine
origin, whereas the grey-green claystones are probably derived from
a volcanic source.
The tentative late Eocene date assigned to the basal part of
thewell is based on the recognition of nannoflora taxa Diacoster
cf. Saipanensis and Dicoaster cf. barbadiensis and also of
palynoflora taxa Proxapertites sp. As these specimens occur in very
low numbers, the possibility of reworking into sediments of
Oligocene age cannot be excluded. If in situ, the presence of
Proxapertites sp., which is thought to be derived from a mangrove
habitat, together with the marine nannoflora, indicate a near-shore
depositional environment.
Although the base of Megasequence I was not penetrated by
Arwana-1, seismic data suggest a section below TD of approximately
2000 metres overlying acoustic basement. Basement is estimated to
be at a total depth of approximately six kilometers (two way time
4.50 secs). The internal seismic character of the Megasequence
comprises a series of high amplitude events, possibly suggesting a
downward continuation of the interbedded volcanoclastic and
argillaceous units penetrated by Arwana-1. However, owing to
limited seismic resolution at these deeper levels, and absence of
well control, it has not been possible to subdivide the
Megasequence into component sequences.
The recognition of Megasequence I in the North Manna Basin is
less certain owing to the limited deep resolution of the onshore
seismic.
Megasequence I is probably, at least in part, equivalent to the
Lahat Formation of the South Sumatra Basin. In both cases the
sediments represent the initial fill of graben depocentres,
although if the late Eocene age of Megasequence 1 in Arwana-1 is
correct, deposition in the basins of the Bengkulu region may have
commenced earlier than in the South Sumatra Basin. It is also
possible that the Kikim volcanics which occur at the base of the
Lahat Formation are the time equivalent of the volcanics in
Megasequence I.
Megasequence I1 (early Oligocene to early Miocene)
In the South Manna Basin, deposition of Megasequence I1 occurred
within an elongate northwest-trending
depocentre, which was superimposed on the underlying system of
segmented half graben. Megasequence I1 is also recognized within
seismic traversing the North Manna Basin, and at outcrop within the
Barisan Mountains (Brown Series of Elber 1938). Detailed
depositional relationships within this onshore basin however are
not as clearly defined owing to poor seismic resolution and
imaging.
Megasequence I1 can be subdivided into the following four
sequences:
Sequence 11.1 (early Oligocene): The basal Sequence 11.1 is
confined to the deeper parts of the basin, and has a transparent
seismic character. In Arwana-1, the sequence comprises regular
interbeds of dark brown and grey-green claystone, with juvenile
volcanoclastic lithologies. Evidence from sidewall cores and
interpretation of logs indicate that the interbedding of these
different lithotypes ranges from millimeter scale laminations to
beds a few metres thick. The volcanoclastics in Core 3 of Arwana-1
contain a variety of lithologies, including vitric crystal tuffs,
tuffaceous sandstones, dark brown mudstones and polymict
conglomerates, which based on sedimentological evidence are
interpreted as being deposited as submarine mass flow deposits.
However, there is no evidence from micropaleontology that
deposition of these mass flows took place in a deep environment or
that sediments were transported any significant distance. Although
globigerine forams were recovered from the core, they were
concentrated in discrete horizons and could have been washed into a
shallow marine environment. A silled basin with limited open marine
access is one possible interpretation of this.
On the basis of age equivalence, Sequence IP.1 can be correlated
with the upper part of the Lahat Formation of the South Sumatra
Basin (Benakat Member).
Sequence 11.2 (late Oligocene to earliest Miocene): This
sequence contains a number of clearly-defined, parallel seismic
events corresponding to volcanoclastic interbeds which are thicker
than those present in the underlying Sequence 11.1. A further
significant contrast between the two Sequences is the absence of
dark brown claystones in Sequence 11.2. In Arwana-1, the upward
change in lithology across the lower boundary of Sequence 11.2 is
abrupt. It is associated with an upward change from a slightly
overpressured section into siltier beds which display a
characteristic invasion separation on the resistivity logs.
The boundary between Sequences 11.1 and 11.2 in Arwana-1 is also
approximately coincident with the top of the early Oligocene which,
in turn, is based
-
323
primarily on palynological evidence (last appearance of ?
Corrudinium incompositum). The late Oligocene to earliest Miocene
age assigned to Sequence I1 is based on the combined evidence of
palynology and micropaleontology .
Within Sequence 11.2, reworked early Cretaceous marine
palynomorphs were also recognized within a thin calcareous unit.
These perhaps suggest the nature of pre-rift basement lithology in
the Bengkulu area.
Results from sidewall cores indicate that the interbedded
volcanoclastics are comprised of tuffaceous deposits with variable
matrix and crystal content. The gamma-ray curve defines probable
sediment supply cycles, characterized by an upward-coarsening motif
into the main clastic bed, overlain by an upward-fining unit. These
cycles probably reflect variations in volcanic activity, and are
probably independent of changes in relative sea level.
Sequence 11.2 is likely to be the time equivalent of the Talang
Akar Formation of the South Sumatra Basin.
Sequence 11.3 (early Miocene): In Arwana-1, the basal part of
Sequence 11.3 is characterized by the re- appearance of dark brown
claystones, which within 70 metres pass upward into an argillaceous
dolomite. This dolomite can be correlated with the Baturaja
Limestone of equivalent age in South Sumatra. Restricted outcrop of
the same limestone in a basin margin, skeletal wackestone/packstone
facies, also occurs close to the onshore, southeastern boundary of
the former contract area (upper part of the Air Saung river). At
outcrop, the biofacies of the limestone is distinctive, containing
both the key benthonic forams Lepidocyclina and Spiroclypeus .
The upper part of Sequence 11.3 comprises massive dark brown
claystone, which in turn passes up into a series of thin (less than
5 metre thick) sandy intervals. Cores 1 and 2 of Arwana-1 suggest
that these feldpathic arenites were deposited as storm/flood
laminae, or thoroughly mixed by bioturbation with claystones and
siltstones. The sandstones are commonly cemented by an early
pore-filling calcite cement. Framework grains include bipyramidal
beta quartz, indicating derivation from a volcanic provenance, and
also unaltered sub-angular to sub-rounded feldspars, suggesting
limited transport or exposure to weathering processes. The
framework texture is under-compacted, owing primarily to the early
calcite cementation.
Sequence 11.4 (latest early Miocene): The unconformity
separating Sequences 11.3 and 11.4 is associated with a phase of
mild, localized inversion tectonism. In
Arwana- 1, this boundary possibly accounts for missing section
between the nannofossl NN2 and "4 zones (in terms of the Blow foram
zonation, the missing section would correspond to the N6 to basal
N7 zones). The lithostratigraphy comprises a continuation of
claystones with occasional interbeds of feldspathic arenites.
Within the South Manna Basin, the seismic facies associated with
Sequence 11.4 displays low-angle, progradational clinoforms.
The dark brown claystones of Sequences 11.3 and 11.4 can be
correlated on the basis of both biostratigraphy and
lithostratigraphy with the Gumai Formation of the South Sumatra
Basin. Furthermore, the sandstones in the uppermost part of
Sequence 11.3 and within the lower part of Sequence 11.4 can be
correlated with similar age sandstones in the South Sumatra
Basin.
Megasequence 111 (middle to late Miocene):
Megasequence I11 represents deposition within the regional
fore-arc basin, which in the Bengkulu platform area occurred from
middle Miocene times onward. The abundance and diversity of middle
Miocene forams within Megasequence I11 clearly indicate deposition
in an open marine environment.
As it is beyond the scope of this paper to discuss the regional
correlation of fore-arc sequence stratigraphy, we have not
sub-divided Megasequence I11 into sequences as has been attempted
for northern areas of the fore-arc by Beaudry and Moore (1985).
The lower part of Megasequence I11 lithostratigraphy is
characterized by a series of high frequent,
transgressive-regressive cycles, comprising claystones, siltstones
and minor limestones. urthermore, it is clear from Arwana-1 logs
that the periodicity of these cycles is irregular. possibly owing
to contemporary non-linear subsidence of the Inner Fore-Arc
shelf.
In contrast, the upper part of Megasequence I11 is characterized
by more massive shelfal limestones, including major reefal
build-ups (Parigi Limestone equivalent).
Megasequence IV (early Pliocene to Recent):
Following a major marine transgression in earliest Pliocene,
differential subsidence between the Inner and Outer Fore-Arc areas
increased. Deposition within the rapidly-subsiding Inner and Outer
Fore-Arc areas comprised marine clays, interbedded with massive,
prograding siltstone wedges derived from the coeval
-
324
uplift of the Barisan Mountains and associated Plio- Pleistocene
volcanic activity.
TECTONIC HISTORY
We interpret the distinctive lithological character of
Megasequences I, I1 111 and IV, and the spatial relationships
between each of these Megasequences within the Bengkulu Inner
Fore-Arc region to reflect their deposition as separate
tectonostratigraphic units in four distinct, superposed basin
types. At least three and possibly all four of these units is
present in both the North and South Manna Basins, which should
therefore be regarded as composite basins in the sense of Hubbard
et al. (1985). On the platform areas outside the North and South
Manna Basins, and outside two other probable Paleogene depocentres
tentatively recognized in the Bengkulu area, only the youngest
basin-forming units, Megasequences 111 and IV, are present.
The geophysical results and the results from Arwana-1 suggest
that Megasequence I was probably deposited during the Paleogene as
a syn-rift unit within a system of northeast-trending half graben,
which were probably segmented by northwest-trending transfer faults
(Figure 5) . Tilted fault blocks bounded by northeast-trending
faults are well imaged in some of northwest-southeast oriented
seismic lines over the South Manna Basin (Figure 8). These growth
faults clearly indicate the syn- tectonic deposition of
Megasequence I (Figure 9).
A major unconformity between Megasequences I and I1 is
interpreted as marking a change in the basin- forming mechanism
from Paleogene extension to possible pull-aparts associated with
oblique slip. According to this model, some of the northwest-
trending transfer faults segmenting the older rift basin were
rejuvenated by right-lateral oblique slip in the late Paleogene to
early Miocene, thereby superimposing local pull-apart basins on the
underlying Mega- sequence I graben. A transtensional pull-apart
origin for the Megasequence I1 basin-fill within the composite
South Manna Basin is consistent with its narrow, elongate
depocentre (Figures 10 and l l ) , acd the presence of mild, coeval
inversion structures along the approximately rectilinear,
northwest-trending basin margins. Furthermore, the basal seismic
sequence of egasequence I1 (11.1) is clearly offset in places by
reactivation of the older northeast-trending, Megasequence I faults
(Figure 8), consistent with apull- apart interpretation for the
younger, superposed basin.
Megasequence I1 is succeeded with strong unconformity by
Megasequence 111, marking deposition in aunified fore-arc
basin.
Megasequence IV was deposited during Pliocene to Recent uplift
and erosion of the Barisan Mountains, and can therefore be
described as syn-orogenic. The associated inversion of
Megasequences I and I1 increases in intensity from offshore toward
this mountainbelt.
SOURCE ROCK AND RESERVOIR POTENTIAL
Source Rock Potential
Source rock lithofacies are present as dark brown marine
claystones in Megasequences I and 11. Within Arwana-1, two main
intervals are recognized: an upper unit within Sequence 11.3 and a
lower interval, corresponding to Sequence 11.1 and the uppermost
part of Megasequence I (Figure 12).
The upper source rock interval displays incipient (threshold)
maturity equivalent to a vitrinite reflectance (VR) of 0.5. This
maturity level is consistent with the estimated thermal gradient in
Arwana-1 of 2.8 degrees celsius/100 metres. Total organic carbon
(TOC) values of selected claystone samples are ca. 2%, hydrogen
index (HI) values range from 300 to 400, and pyrolysis yields of up
to 10 kg/ton were recorded. As most of the interval is
lithologically homogeneous, these richnesses also represent bulk
rock characteristics.
The lower source rock interval is within the oil window, with a
VR of 0.6 estimated at a depth of 3645m. TOC values of selected
claystone samples are between 1% and 2%, with hydrogen index values
decreasing from 300 to between 100 and 200 (Figure 12). On a bulk
rock basis however, these richnesses are significantly reduced by
variable interlamination of volcanoclastic lithologies.
Despite this, the reduction in TOC and HI values compared with
the upper interval suggests that the lower source rock interval is
partially spent. It is therefore reasonable to assume that the TOC,
and HI of the lower (mature) interval were originally as good as
the upper, incipiently mature interval.
Both the upper and lower intervals can be classified as Type 11,
with oil and gas generation capacity.
The distribution of oil shows in Arwana-1 corresponds to the
main source rock intervals. Biomarker analysis of an extract from
the Baturaja Limestone equivalent (basal Sequence 11.3) suggests
low maturity and probable sourcing from the adjacent interbedded,
early-mature source rocks. On the other hand, analysis of an oil
show in the voIcanic sandstone near the top of Megasequence I
(Figure 13) indicates derivation from a
-
325
parent source rock with a maturity of about 0.87% VR. In
contrast, the extract from the deepest source rock in Arwana-1
indicates a maturity of 0.67% VRE. This evidence is based on an
extract from shows and is ,therefore not conclusive.
However, this contrast in maturities suggests that the
hydrocarbons in the volcanic sandstone may have migrated a vertical
distance of up to one kilometer from the parent source levels. This
in turn suggests a depth to the top of the oil expulsion window of
about five kilometers. As the maximum depth to basement is
estimated to be greater than six kilometers, it follows that the
gross thickness of the oil expulsion window may exceed one
kilometer in the basin depocentres, implying the possibility of a
substantial hydrocarbon kitchen.
Although the presence of oil shows is encouraging, the
hydrocarbon prroducing potential of the basins in the Bengkulu area
will depend on bulk rock generative capacity of source rocks, which
in turn will be controlled by the extent of heterogeneous
interbedding of the source intervals with the non-organic,
volcanoclastic lithologies. Other factors such as the effectiveness
of migration routes also need to be considered.
Reservoir Potential.
The overall quality of the reservoir lithologies encountered in
Arwana-1 is pool'. The volcanoclastic sandstones in Megasequence I
exhibited log porosities of ca. 10%, and effective permeability was
inferred from a marked invasion profile in the resistivity logs.
Unfortunately, however, effective permeabilities could not be
confirmed by RFT measurements.
Other clastic intervals exhibited porosities mostly in the range
10 to 15%. Low permeabilities were indicated throughout
Megasequence II, with the exception of a crystal-rich tuff bed in
Sequence 11.2 and a feldspathic arenite in Sequence 11.4, both of
which delivered RFT water samples (Figure 14).
Porosities in Megasequence I and the lower part of Megasequence
II in the well were created by an aggressive secondary dissolution
process, which appears to be linked to oil migration. Good
permeabilities in the volcanoclastics however, depends additionally
on original sorting (textural maturity), or the presence of
extensive fracture systems.
The reservoir potential of early and middle carbonate build-ups,
overlying the Paleogene basin depocentres, remains
under-explored.
SUMMARY OF BASIN DEVELOPMENT
Based on the structural and stratigraphic results, basin
development can be summarised as follows (Figures 15 to 17):
Megasequence I Time
Within the South Manna Basin, and probably the North Manna Basin
deposition of pro-delta marine claystones within the segmented
rifts (Figure 15) was periodically interrupted by the input of
reworked volcanic sandstones, derived from coeval volcanic
activity. It is also possible that the basal sections of some half
graben were isolated from marine influenco, and were characterized
instead by lacustrine deposition.
The detailed relationship between the North and South Manna
graben and the South Sumatra graben of the back-arc area is
unknown. The Paleogene Bengkulu and South Sumatra graben may have
allowed a continuous depocentre to develop, with a northward
transition from marine conditions to the paralic/ lacustrine
environments of the back-arc basins. It is however more likely that
this trend was segmented by possible transfer or relay fault
systems associated with the regional northwest trending structural
grain.
Megasequence II Time
Megasequence II time was characterized by arestricted marine
environment in which depositional conditions were influenced by
variations in subsidence rate and sediment supply, probably within
an evolving pull- apart basin. These variations in turn are
represented in the contrasting character of the constituent
Sequences.
Consequently, Sequence 11.1 represents the initial deepening of
the basin, and resulting deposition of rhythmically-interbedded
argillaceous and clastic slope deposits. The absence in Arwana-1 of
marine claystones in Sequence 11.2 implies that an additional
restriction of the marine environment occurred during this time,
with deposition of volcanogenic sediments dominating (Figure 16).
This is consistent with the regional late Oligocene sea level
lowstand recognized in Paleogene basins throughout the Sunda
Shield.
Sequences 11.3 and 11.4 represent the final infill of the
localized early Neogene basins (Figure 4), and their deposition was
associated with the reworking of mature volcanoclastic sandstones
in a shallow shelf environment (Figure 17).
The lithological and biostratigraphical similarity of Sequences
II.3 and II.4 in Arwana-1 with the Lower
-
326
Miocene outcrop in the Barisan Mountains and the Gumai Shale of
the South and Central Sumatra suggests that by the late early
Miocene a continuous depositional fairway existed between the
Bengkulu area and the South Sumatra Basin.
Megasequences I11 and IV Time (Regional Fore-Arc Basin)
During the middle Miocene, the Inner Fore-Arc shelf was a
more-or-less uniformly subsiding surface characterized by the
deposition of the transgressive cycles of Megasequence 111. During
the Plio-Pleistocene the rate of shelfal subsidence increased
significantly coincident with the deposition of the prograding
synorogenic sediments of Megasequence IV.
CONCLUSIONS AND IMPLICATIONS FOR STRUCTURE AND PROSPECTIVITY OF
THE SUMATRAN' FORE-ARC
We have identified two quite distinct Paleogene to early Neogene
basin styles which are superposed within the present Inner Fore-Arc
region of the Bengkulu area. An earlier, Paleogene basin type
(corresponding to Megasequence I) developed as a result of
northeast-trending rifting, and was probably tectonically
overprinted by a pull-apart basin (Megasequence 11) when
northwest-southeast directed extension changed to
northwest-directed oblique slip or transtension.
It therefore follows that the South Manna Basin cannot be
described simply as a back-arc basin in a fore-arc setting. Rather,
our results highlight the influence of two distinct tectonic
systems: a continuation of extensional trends within the Sunda
Shield, modified by the onset of right-lateral oblique slip within
the Sumatra Sliver Plate. The superposition of the two associated
basin types and their Megasequences (I and 11) suggests that zones
of structural weakness coincident with the Paleogene graben trends
influenced the initial break up of the Sumatra Sliver Plate.
These results imply that far from accommodating a simple,
homogeneous fore-arc basin, the Sumatran fore-arc is tectonically
heterogeneous, with considerable potential for localized Paleogene
and early Neogene depocentres. This in turn has obvious
implications for basin development in other fore-arcs where the
effects of oblique subduction are apparent.
The results of Arwana-1 have a significant impact on the
hydrocarbon potential of fore-arc basins in general. The presence
of mature source rock lithofacies and migrated oil in this well
contradicts the traditional
assumption that heat flow values in fore-arc basins are
insufficient to allow expulsion and migration of hydrocarbons.
However, despite this encouragement, the volumetric
hydrocarbon-producing potential of these basins remains to be
proven. The bulk generative potential of the source rock prism has
been identified as a critical factor, and this is determined in
turn by the degree of interbedding with non-organic lithologies
derived from volcanic sources.
The presence of reservoir clearly also represents a significant
risk in the further exploration of fore-arc areas. The reservoir
potential of volcanoclastic sediments depends on processes such as
secondary dissolution and fracturing, as well as the primary
depositional rock fabric. Consequently, these lithofacies types
should not be completely dismissed as reservoir targets. There also
remains the possibility that qon- volcanogenic lithofacies, not
penetrated by Arwana-1, provide good reservoirs elsewhere in these
basins.
Our understanding of Sumatran fore-arc basin development and
associated hydrocarbon potential is therefore clearly still at an
early stage. In addition to the uncertainties in hydrocarbon
potential, the distribution of units equivalent to Megasequences I
and I1 in other Sumatran fore-arc basins, including other basins in
the Bengkulu area, requires attention. This advancement will come
from further exploration in what should still be regarded as an
under-explored, frontier province.
ACKNOWLEDGEMENTS
We wish to thank the management of Petrofina, Pertamina and
Partners British Gas and Enterprise Oil for permission to publish
this paper. Pusat Penelitian Dan Pengembangan Geologi (GRDC)
provided very helpful assistance and logistical support. We are
particularly indebted to Ir. Nana Ratman and Ir. Thamrin Cobrie
Amin for their help with field sampling.
We are grateful to all those in Petrofina who assisted with the
Bengkulu project. In particular special thanks are extended to Dr.
Paul Baumann, who provided valuable technical input during the term
of the Bengkulu PSC, Dr. Ralph Burwood for reviewing the results of
the geochemical analyses and Serge Froment for his work on the
well-site and also for producing the post-well geological report.
Biostratigraphical and geochemical analyses are based mostly on the
work of P.T. Corelab Indonesia and in particular we would like to
thank .Brown, R.E. Hulsbos, J. Harrington and S. Hindmarsh.
Finally, we gratefully acknowledge the efforts of Fina Exploration
Norway in helping us to
-
327
produce the manuscript. The interpretations presented in this
paper are those of the authors and do not necessarily represent the
views of all the co-ventures in the Bengkulu PSC.
REFERENCES
Beaudry, D. and Moore, G.F., 1985. Seismic stratigraphy and
Cenozoic evolution of West Sumatra, Bulletin American Association
of Petroleum Geologists, 69, 5, p. 742-759.
Blow, W.H., 1969. Late middle Eocence to recent planktonic and
foraminifera1 biostratigraphy: Proceedings of the First
International Conference on Planktonic Microfossils, Geneva (1967)
, p. 199-422.
de Costa, G.L. , 1974. The geology of central and south Sumatra
basins, Zndonesian Petroleum Association, 3rd Annual Convention
Proceeding, p. 77-110.
Elber, R., 1938. Geologie des Kuestengebietes von Benkoelen
zwischen Seblat (NW) und Bintoehan (SE), (Westkueste von
Sued-Sumatra): BPM (Shell) Unpub., p. 24.
Howles, A.C., 1986. Structural and Stratigraphic Evolution of
the Southwest Sumatran Bengkulu Shelf,
Indonesian Petroleum Association, 15th Annual Convention
Proceeding, p. 215-243.
Hubbard, R.J., Pape, J., and Roberts, 1985. Depositional
sequence mapping as a technique to establish tectonic and
stratigraphic framework and evaluate hydrocarbon potential on a
passive continental margin, in O.R. Berg and D. Wolverton eds.,
seismic stratigraphy 11: an integrated approach to hydrocarbon
exploration, AAPG Memoir, 39, p. 79-91.
Huchon, P. and Le Pichon X., 1984. Sunda Strait and Central
Sumatra fault, Geology, 12, p. 668-672.
Jarrard, R.D., 1986. Terrace Motion by Strike-Slip Faulting of
Forearc Slivers, Geology, 14, p. 780-783.
Karig, D.E., Lawrence, M.B., Moore, G.F. and Curray, J.R., 1980.
Structural framework of the fore-arc basin, NW Sumatra, J . Geol.
SOC. London, 137, p. 77-91.
Matson, R.G. , Moore, G.F. , 1992. Structural Influences on
Neogene Subsidence in the Central Sumatra Fore- Arc Basin, AAPG
Memoir, 53, p. 157-181.
Rose, R., 1983. Miocene Carbonate Rocks of Sibolga Basin
Northwest Sumatra, Indonesian Petroleum Association, 12th Annual
Convention Proceeding, p. 107-125.
-
328
I! G 5 I 0 cn
u@ 7 L
8 3 8 6 m
w
E
$1 0
s a 3 a a
w
v)
v)
..... .....
.....
..... ..... ..... .....
.....
M
%
C .3 Y Y
u d 0 0 0
.3
u
Y H a C 0 M
.3
E
-
c j a c/)
3 Y E
8; d
3
rn
329
-
4" s
[ ME
NTA
WA
I A- I
]
I AR
WA
NA
-11
\ e
\w
W
ELL
S D
RIL
LED
DU
RIN
G 1
97
0-7
2
EX
PLO
RA
TIO
N W
ELL
199
2
SE
ISM
IC 1
99
0-9
1
LOC
ATI
ON
OF
RE
GIO
NA
L C
RO
SS
-SE
CTI
ON
S (
FIG
.5)
LOC
ATI
ON
OF
SE
ISM
IC P
AN
ELS
(F
IG.7
+8)
I O
FB
90
-67
@ F
B90
-83
w
w
0
3 -
Ben
gkul
u PS
C: E
xplo
ratio
n st
atus
.
-
33 1
. . .. m r
d I
v, d l I
-
sw
NE
I @
S
OU
TH M
AN
NA
BA
SIN
N
OR
TH M
AN
NA
BA
SIN
69
I I
L I
tt
t
PRO
BA
BLE
TR
AN
SFE
R F
AU
LTS
NW
0
so
l
I .
.
r B
AR
ISA
N U
PLI
FT ,.=
0
5 10
km
-
SE
I, II,
IU,a
: BA
SIN
-FO
PM
ING
ME
GA
SE
QU
EN
CE
S
PF.
: "
8AS
EM
EN
T" P
AP
ALL
EL
FAC
ES
TH
n - A
1 0
B
AS
IN
FIG
UR
E
5 ~
Sout
h an
d N
orth
Man
na B
asin
s: R
egio
nal c
ross
-sec
tions
(for
loca
tion
see
fig. 3
).
w
w
N
-
333
STRATIGRAPHIC SUMMARY OF NORTH AND SOUTH MANNA BASINS
I SOUTHMANNA NORTH MANNA BASIN BASIN
\ - - --c - -+ - 3 a a w
...........................................................................
~~ .. 1 :::-::'.::::.::.-.~::::-::::::::::-.:::::-:~
-:::::::.- :: :: 2 :::: .-. ::::::::.-:::::::.-: ::: : :: ~ ~ .
.
- - t - ---c - - - c - - - - -
..... ::::: v :::I v ...... --- v..........v...
@ N9 LIMESTONE @ BATURAJA LIMESTONE EQUIVALENT /// PROGRADING
CLINOFORMS
FI ORGANIC (DARK BROWN) F] ORGANIC LEAN I COALS CLAYSTONES - -
CLAYSTONES 0
(u t=i
- MEGA SEQ. - El
FIGURE 6 ., Stratigraphical summary of North and South Manna
Basins.
-
334
0.0
1 .o
2.0
3.0
4.0
5.0
FIGURE 7 - Seismic panel from line FB 90-67 showing position of
Arwana-1 and main sequence boundaries (for location refer to Fig.
3) .
-
335
m M X 3
-
336
/ v, 0 =+- /a
E
0 :I w a- z 0
hl
-
337
I
-
338
-
2500 -
3000 -
3500 -
AR
WA
NA
# 1
SO
UR
CE
RO
CK
SU
MM
AR
Y
T.O
.C.
* H
YD
RO
GE
N IN
DE
X *
0
1 2
3 4
0 10
0 200
300
40
0
SEIS
M.
SE
Q.
1 I
I I
I
DARK
BR
OW
N
CLA
YSTO
NES
-1-1
+--c
...
......
......
.. ....
........
.....
1 ............
......
.:_.I --
...
......
......
...
++
--
-c
-
-+
...
......
......
.. ...
......
......
.. Ir. 3
........
........
. ....
........
.....
- -
._
-.-
-.-
........
........
. ....
........
.....
--
.-
-. -
.-
........
........
. &
...
......
......
.. .-
........
. 11
. 1
: : : : : - : :T
.- ....
......
. ....
........
.....
-
---c
........
........
....
........
....
'"1
-
4000
- * T
OC
AN
D H
I VA
LUES
FO
R SE
LEC
TED
DAR
K B
RO
WN
CLA
YSTO
NE
LITH
OLO
GIE
S.
EXTR
AC
T SA
MPL
ES
- @ B
ATU
RA
JA L
IMES
TON
E EA
RLY
MA
TUR
E VR
E 0.
5
OIL
SH
OW
EX
TRA
CT
FRO
M
VOLC
AN
IC S
ST.
VRE 0.87
@
SO
UR
CE
RO
CK
EX
TRA
CT
DA
RK
BR
OW
N
CLA
YS2'
ON
E V
RE
0.7
OIL
SH
OW
S
FIG
UR
E 1
2 -
Sum
mar
y of
sour
ce ro
ck c
hara
cter
istic
s.
-
OIL SHOW EXTRACT FROM VOLCANOCLASTIC
131 Lt. EOC. SOURCE ROCK 4155 M SOWICE ROCK EXTRACT FROM DARK
BROWN CLAYSTONE
0.4 0.5 0 .6 0.7 0 . 8 0.9 I .o t.1 MATURITY (VI+RINITE
REFLECTANCE (%) E M . )
FIGURE 13 - Arwana-1, Megasequence I: Biomarker comparison of
oil show and adjacent source rock.
-
341
2500
LITHOLO6Y ] (~ Av (~o)
m . 12 % (L)
SEQ.
I I ! 1
II .4
3000
"--"'?i ~ THIN BEDS :ul,2
-'-_._. 11% (c) II .3B
11.3A
-V
m
V-
3500
I e
I
I 4000
_ - - I 12% (L)
V
~ V -
. . . . . :13 . . . . . : THIN . - . . . . . - .
- : - BEDS 13% (C) . '~ . -7 . ".~
:T ~ I 12% (L)
11.2
I1.1
E
I I
CORES
CALCULATED FROIV] CORES (C) OR LOGS {L)
PERMEABIUTY INOICATED BY RFT.
I O~
ENVlRO~IENT
OPEN MARINE
RESTRICTED
SUBLITTORAL
VARIABLE
SEDIMENT INPUT
RESTRICTED
HIGH
SEDIMENT
INPUT
RESTRICTED
SUBLITTORAL
VARIABLE SEDIMENT INPUT
RESTRICTED
SUBUTTORAL
TO LITTORAL
S. SUMATRA BASIN EQUIV.
BATU - RA3A
Z /- < /
/5 0~ O
[-,
FIGURE 14 - Arwana-1, Early Neogene to Paleogene reservoir
stratigraphy.
-
342
-
.. ..
.. a U
343
-
344
/ . ~o"
.~ .
f : , : " /
:l:: I ?} ///
t%t
~Q Q~
:
%"
S;
>
ZZ