Petroleum systems of the Simeulue fore-arc basin, offshore Sumatra, Indonesia

Rdiger Lutz, Christoph Gaedicke, Kai Berglar, Stefan Schloemer, Dieter Franke, and Yusuf S. Djajadihardja

AUTHORS Rdiger Lutz  Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, 30655 Hannover, Germany; ruediger.lutz@bgr.de Rdiger Lutz obtained a Ph.D. in geology from the RWTH (Rheinisch-Westflische Technische Hochschule) Aachen University, Germany. He joined the Federal Institute for Geosciences and Natural Resources (BGR) in 2003 and works as a research scientist in the petroleum geology group. His main research interests include marine geology, seismic interpretation, and petroleum systems modeling. Christoph Gaedicke  Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, 30655 Hannover, Germany; christoph.gaedicke@bgr.de Christoph Gaedicke holds a Ph.D. in geology from the IFM-Geomar (Leibniz Institute of Marine Sciences) Research Center for Marine Geosciences at the University of Kiel, Germany. He joined BGR in 1999, where he is working on processes at active continental margins and in petroleum basins. He is the head of the sub-Department of Resource Geology, Polar Geology. Kai Berglar  Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, 30655 Hannover, Germany; kai.berglar@bgr.de Kai Berglar received a Ph.D. in geology from the Leibniz University, Hannover, Germany. He joined BGR in 2006 and works as a research scientist in the marine seismic unit. His main research interests are in seismic stratigraphy, sequence stratigraphy, and seismic processing. Stefan Schloemer  Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, 30655 Hannover, Germany; stefan.schloemer@bgr.de Stefan Schloemer holds a Ph.D. in mineralogy from the RWTH Aachen University, Germany. He worked two years for EniTecnologie/AGIP in Milan, Italy, before he joined BGR in 2001. Stefan works on CO2 soil gas monitoring, carbon capture and storage (CCS), petroleum system analysis, gas shales, drill-bit metamorphism, and

ABSTRACT Forearc basins result from plate convergence. These basins are situated offshore between an outer-arc high and the mainland. Historically, these regions have not been considered important petroleum provinces partly because low heat flow may limit significant thermal hydrocarbon generation. The Simeulue forearc basin extends between Simeulue Island and northern Sumatra. It is a frontier shallow shelf area with few wells and no wells in the basin center; therefore, it is studied using geophysical data and geologic surface samples. Multichannel seismic data show bright spots above potential hydrocarbon reservoirs in carbonate buildups. Amplitude versus offset analyses indicate the presence of gas, and surface geochemical prospecting suggests thermal hydrocarbon generation. Heat flow in the Simeulue Basin ranges between 37 and 74 mW/m2, as deduced from one-dimensional petroleum system modeling and bottom-simulating reflector depths. Two possible source rocks (Eocene and lowermiddle Miocene) were assigned for three-dimensional petroleum system modeling in the Simeulue Basin. Because of a similar preMiocene geologic evolution of the present-day back arc and the fore arc, it can be assumed that the back-arc source rocks also occur in the fore arc. Modeling based on two heat-flow scenarios (40 and 60 mW/m2) reveals that oil and gas generation is possible within and below the main depocenters of the central and southern Simeulue Basin. This study shows that deep burial (>6 km [>3.7 mi]) of source rocks can compensate for low heat flow. Therefore, forearc basins may be more prolific for

Copyright 2011. The American Association of Petroleum Geologists. All rights reserved. Manuscript received June 3, 2010; provisional acceptance August 24, 2010; revised manuscript received November 29, 2010; final acceptance January 19, 2011. DOI:10.1306/01191110090

AAPG Bulletin, v. 95, no. 9 (September 2011), pp. 15891616

1589

coal-fire research. He is running the geochemistry laboratory including gas chromatography (GC), gas chromatography isotope ratio mass spectrometery (GC-IRMS), and other instruments. Dieter Franke  Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, 30655 Hannover, Germany; dieter.franke@bgr.de Dieter Franke holds a Ph.D. in geosciences from the Free University of Berlin, Germany. He joined BGR in 1996 and is presently the team leader of the petroleum geology group. His main research interests are rift processes and the formation of passive margins. Yusuf S. Djajadihardja  Agency for the Assessment and Application of Technology (BPPT), Jl. M. H. Thamrin No. 8, Jakarta 10340, Indonesia; iyung@ceo.bppt.go.id

hydrocarbons than previously considered, and each forearc basin should be studied carefully to evaluate its hydrocarbon potential.

INTRODUCTION Forearc basins are located at almost all convergent continental margins between the deep sea trench and the magmatic arc. Active margins occupy some 44,000 km (27,340 mi) (Von Huene and Scholl, 1991), mainly around the Pacific Ocean, but also along the Indonesian island arc and in the Caribbean, the Mediterranean, and the Arabian Sea. The widespread occurrence of this basin type makes it worthy to research to decipher its hydrocarbon potential. Forearc basins may overlie either continental crust or accreted rocks. Commonly, forearc basins are bounded seaward by an outer-arc high that is part of the accretionary prism. The tectonic evolution of forearc basins is controlled mainly by the subduction of an oceanic plate. Important factors controlling the basin dynamics are convergence rate, dip angle of the oceanic plate, obliquity of convergence, roughness and sedimentary cover of the oceanic crust, and the sedimentary input of the landward magmatic arc and continent (Jarrard, 1986; Dickinson, 1995; Van der Werff, 1996; Fuller et al., 2006). Forearc basins have an elongated shape extending parallel to the trench-arc system. Their width commonly ranges between 25 and 125 km (15.5 and 77.7 mi), but their length may reach 500 km (310 mi) (Busby and Ingersoll, 1995). Forearc basins are widespread, however, their hydrocarbon potential is believed to be low. This is caused by a complicated structural history with long-lasting and usually ongoing tectonic activity affecting the basins, mainly terrigenous sediment input from the arc and, thus, the questionable presence of prolific source rocks and generally low heat flow. Nevertheless, a tectonic basin classification is not sufficient to evaluate the petroleum potential of a sedimentary basin (Demaison and Huizinga, 1991). Heat-flow values of forearc basins range between 20 and 45 mW/m2, with a typical value of 40 mW/m2 (Allen and Allen, 1990; Dickinson, 1995). In contrast, passive margins typically exhibit a heat flow between 30 and 50 mW/ m2 or even more in cases of forced fluid flow (Gosnold, 2010). Relatively cold oceanic crust is subducted at continental margins and reduces the heat flow beneath the forearc basin. Heat flow is typically further reduced by a high sediment input from the volcanic arc and the adjacent continent (Benfield, 1949). The Simeulue Basin matches all features of a typical forearc

ACKNOWLEDGEMENTS The research projects were conducted with grants 03G0186A (SeaCause) and 03G0189A (SUMATRA) from the Federal Ministry of Education and Research (BMBF), Germany. We thank all colleagues from BGR who made this work possible through their help during the cruises, data acquisition, and processing. In particular, we thank the ships masters L. Mallon and O. Meyer and the crew of FS Sonne during the SeaCause and SUMATRA cruises. Thorough reviews by Debra K. Higley, Barry J. Katz, Jory A. Pacht, and Kenneth E. Peters helped improve the manuscript considerably. The AAPG Editor thanks the following reviewers for their work on this paper: Debra K. Higley, Jory A. Pacht, and Kenneth E. Peters.

1590

Simeulue Fore-arc

Figure 1. Regional tectonic setting of the Sunda subduction zone. The forearc basins off Sumatra lie between the Sumatra mainland and the outer-arc high that occasionally emerges above sea level, forming the island chain west of Sumatra. SFZ = Sumatran fault zone; MFZ = Mentawai fault zone; BF = Batee fault; WAF = West Andaman fault. Faults and the deformation front in the Sunda trench are based on Sieh and Natawidjaja (2000). Black arrows indicate relative plate motion based on continuous GPS data (Prawirodirdjo and Bock, 2004).

basin (e.g., low heat flow, tectonic setting controlled by subduction) and was therefore chosen for study using a multidisciplinary approach, with special emphasis on its hydrocarbon potential. The Simeulue Basin is a frontier basin having only few wells and unpublished well logs, limited borehole information, and seismic data. New multichannel seismic data, originally acquired to study the area of the December 26, 2004, earthquake, lithologic information from wells, and new surface gas-geochemical and heat-flow data were used to establish a conceptual model of the basin evolution. This model forms the basis for petroleum system modeling to assess the hydrocarbon potential of the basin.

GEOLOGIC SETTING The Sumatra-Java area is part of the Sunda arc that stretches from the Andaman Sea in the north-

west to the Banda Sea in the east. Along the Sunda arc, the Indo-Australian Plate subducts under the Eurasian Plate (Figure 1). The western Indonesian forearc basins extend for more than 1800 km (>1119 mi) from northwest of Aceh to southwest Java. The width of the basins ranges from less than 70 km (<44 mi) to the south of the Sunda Strait to about 120 km (75 mi) to the west off northern Sumatra. The basins form a strongly subsiding belt between the elevated Sumatra PaleozoicMesozoic arc massif cropping out along Sumatra and Java and the rising outer-arc high (Karig et al., 1980; Schlueter et al., 2002; Susilohadi et al., 2005). The studied area (Figure 1) is a classic example of a subduction system, composed of the downgoing Indo-Australian slab along the Sumatra trench, an accretionary wedge, the outer-arc ridge emerging above sea level with Simeulue Island (Pubellier et al., 1992; Samuel and Harbury, 1996), and the Simeulue forearc basin off Sumatra in front of the
Lutz et al. 1591

Figure 2. Location of multichannel seismic profiles covering the Simeulue Basin between Simeulue Island and Sumatra. Isolines depict bathymetric contours (Smith and Sandwell, 1997), and circles show locations of wells (Rose, 1983; Beaudry and Moore, 1985). Filled diamonds mark the locations of sediment core gas samples. Hollow diamonds mark sections of seismic lines used for AVO analyses. Bold lines highlight seismic profiles shown in this study. Bold numbers indicate wells used for 1-D modeling. 1 = Teunom-1; 2 = Bubon-1; 3 = Keudepasi-1; 4 = Tuba; 5 = Seunagan-1; 6 = Meulaboh-East2; 7 = Meulaboh-East-1; 8 = Meulaboh-1; 9 = Tripa-1; 10 = Palumat-1; 11 = Ujung-Batu-1; 12 = Pulau-Baru-1; 13 = Palambak; 14 = Singkel-1; 15 = Telaga. Unfilled circles = unknown status or dry hole.

volcanic arc (Barisan Range). Convergence along the Sunda arc becomes increasingly oblique from south to north, resulting in large-scale dextral strikeslip fault systems within the forearc basins and on Sumatra (Malod and Kemal, 1996; Sieh and Natawidjaja, 2000). The Simeulue forearc basin is bounded to the west by Simeulue Island. The Banyak Islands (Figure 2) separate the Simeulue Basin from the southerly located Nias forearc basin. The northern end of the Simeulue Basin is formed by a ridgelike structure and a change in water depth of more than 1500 m (4921 ft) (Izart et al., 1994) to the Aceh Basin. The basin itself is elongated parallel to the trench. It extends more than 260 km (162 mi) in northwest-southeast and 100 km (62 mi) in southwest-northeast directions. It contains Neogene sedimentary fill having as much as 5 s two-way traveltime (TWT) or approximately 7 km (4.3 mi). The basin is located in deep water with a maximum water depth of about 1300 m (4265 ft) (Figure 2).
1592 Simeulue Fore-arc

The basin slope and the shallow water inner shelf at the transition to the Sumatra mainland form the northeastern boundary of the basin (Berglar et al., 2008). The geologic evolution of the Simeulue Basin before the early Miocene shows striking similarities to the evolution of the North Sumatra Basin (Barber et al., 2005). North-south-oriented horsts and grabens developed in the North Sumatra Basin during a late Eocene rifting phase. A subsequent late Oligoceneearly Miocene basin sag phase resulted in widespread carbonate deposition and reef growth (Clure, 2005). Initiation of mid-Miocene wrench tectonics led to uplift of the coastal fold belt, the Barisan Range, and separated the North Sumatra Basin from the forearc region (Figure 1), including the Simeulue Basin. Ongoing compression since the PliocenePleistocene formed the coastal fold belt of Sumatra (Clure, 2005). The present-day back-arc area of Sumatra comprises the North Sumatra, Central Sumatra, and South Sumatra basins, where prolific petroleum

Table 1. Acquisition Parameters of Multichannel Seismic Data Streamer length No. channels Towing depth Near offset Maximum offset Tuned airgun volume Record length Sample rate Shot interval Receiver spacing 3 km (1.9 mi) 240 6 m (19.7 ft) 150 m (492 ft) 3137.5 m (10,294 ft) 50.8 L (14.5 MPa) 14 s 2 ms 50 m (164 ft) 12.5 m (41 ft)

systems exist (e.g., Matchette-Downes et al., 1994; Katz, 1995; Keeley et al., 1995; Cole and Crittenden, 1997; Todd et al., 1997). Oil in the North Sumatra Basin was generated from terrigenous and marine source rocks. Biomarkers indicative of both environments were found in oil samples (Schiefelbein and Cameron, 1997). The Central Sumatra Basin (Figure 1) is the most prolific petroleum system in Indonesia and contains the Brown Shale Formation source rock (Keeley et al., 1995; Doust and Noble, 2008). Source rocks in both basins are of EoceneOligocene age. Other source rocks consist of several coal seams of Oligocene, Miocene, and Pliocene age as reported by Hadiyanto (1992). He studied eight offshore wells, 18 onshore wells (<200 m depth [<650 ft]) and several outcrop samples, all of which are located west of the Barisan Range with some of them in the Simeulue Basin.

METHODS In 2006, a total of 1500 km (932 mi) of multichannel seismic (MCS) data coincident with highresolution sediment echo-sounding, bathymetry, gravity, and magnetic data were acquired in the Simeulue Basin (Figure 2) using the German research vessel Sonne. Parameters of MCS acquisition are listed in Table 1, and a detailed description of the processing flow is given in Berglar et al. (2008). Amplitude versus offset (AVO) studies are commonly used to detect gas- or oil-bearing sediments.

A reconnaissance AVO analysis was performed, including analyses of the seismic data for amplitude variations, creating near- and far-offset stacks and picking and displaying the amplitudes of selected common depth-point (CDP) gathers. No wells were available to constrain the modeling. The quality of the input seismic data is crucial for any AVO result. Data processing for the AVO analyses were different from the conventional processing flow. Here, parameters were carefully selected, and few processing steps were applied. The good quality of the field data was helpful. The raw data were smoothly bandpass filtered (zero phase; 4-8-80160 Hz) and a zero-phase spiking deconvolution was applied (40-ms operator length; white noise, 0.1; decon gates: water-bottom relative 1.0 s). Extensive testing resulted in a time and offset variant gain function that was mainly based on carefully selected stacking velocities. A spherical divergence correction was applied to compensate for loss of amplitudes based on waveform spreading. In addition, a receiver and source array correction was applied to compensate filtering effects of both the source and the receiver array (offset amplitude correction). All processes affecting the amplitudes were calibrated against the sea bottom reflection. A normal moveout correction with zero-percent stretch mute using stacking velocities finalized the processing sequence. No dip moveout or prestack migration was applied because the area under study is structurally simple. Furthermore, no multiple attenuation was needed because the water-bottom multiples appear much later in the data than the reflections of interest. Heat-flow values (Q) derived from the depth of bottom-simulating reflectors (BSRs) were calculated (Figure 3) using the equation Q = l grad(t) with grad(t) = (TBSR - Tsf)/(DBSR - Dsf). Interpreted BSRs were converted from TWT to depth (DBSR) using a velocity profile derived from wideangle reflection seismic data in the southern Simeulue >Basin (Franke et al., 2008). The depth of the seafloor (Dsf ) was calculated, assuming a sound speed in water of 1500 m/s (4921 ft/s). The temperature at BSR depth (TBSR) was then determined with a water-methane phase diagram (Kvenvolden and Barnard, 1982). Water temperature at the sea
Lutz et al. 1593

Figure 3. Sketch to illustrate the calculation of heat-flow values based on bottom-simulating reflector (BSR) depths. Dsf = depth of sea floor; Tsf = temperature at sea floor; DBSR = depth of BSR; TBSR = temperature at BSR depth; l = thermal conductivity.

floor (Tsf) was obtained from CTD (conductivitytemperature-depth) measurements for depths down to 1100 m (3609 ft); for greater depths, we assumed a temperature of 1C. For thermal conductivity (l), we used a published value of 1.23 W/(m K) from Delisle and Zeibig (2007), which was measured in this area. Basin and petroleum system modeling was performed to reconstruct the evolution of a sedimentary basin based on physical and chemical laws (Welte and Ykler, 1981; Tissot et al., 1987) and provide information on thermal evolution and the generation, migration, and accumulation of hydrocarbons. Principles and limitations of the calculated models have been described by various authors (Ungerer et al., 1990; Hermanrud, 1993; Yalcin et al., 1997). However, the technique can be applied to sedimentary basins at all stages of exploration, from unexplored or frontier basins to mature basins (Karlsen et al., 2004; Struss et al., 2008; Underdown and Redfern, 2008). The sediment-water interface temperatures through time were calculated based on the present-day latitude (Wygrala, 1989) and water depth within the software package (PetroMod). Surface geochemical exploration methods aim to detect surface alterations or indications of past
1594 Simeulue Fore-arc

or active microseepage of subsurface hydrocarbons toward the surface. Thus, positive indications of an active petroleum system can support frontier basin evaluation or prospect ranking. Analysis of vertically migrated oil and gas reveals information about the type and maturity of the possible source rocks. However, neither reservoir depth nor the amount of trapped hydrocarbons can be determined. Sediment samples for hydrocarbon analysis were collected from cores taken using gravity (SL) or piston (KL) corers and multicorers (MUC). Sampling sites were selected predominantly above carbonate buildups and faults based on interpretation of the seismic data. Typically, one sample for geochemical analysis was taken from the base of the core or core catcher and, if possible, at least one sample was taken from the upper part of the core (typically 2 m [6.6 ft] above core catcher). At several sites, additional sediment samples were taken from the lowermost part of a multicorer (typically 0.3- to 0.4-m [0.98- to 1.3-ft] depth). Twenty-six samples from 15 sites in the Simeulue Basin were taken, and nine samples were taken from five sites located at two distal reference locations on the outerarc high (140KL, 142KL, 139MUC southeast of Simeulue Island and 115KL, 116KL northwest of Simeulue Island, Figure 2; Table 2). Sediment samples were desorbed according to the technique described by Faber and Stahl (1983). This technique allows analyses of both composition and isotope ratios of adsorbed hydrocarbons in sediment samples. About 100 to 150 g of wet sediment was placed in a vacuum apparatus and treated with phosphoric acid. The carbon dioxide liberated was fixed in KOH or NaOH solution. The composition of desorbed hydrocarbons (methane through pentane) was evaluated without delay by means of a standard gas chromatography (GC) analysis (Shimadzu GC 14b, Porapaq Q column, 2 m [6.6 ft], 1/8, isothermal at 115C). The remaining hydrocarbon gases were compressed and displaced into evacuated glass sampling tubes for later analysis of isotope composition (GC-IRMS [isotope-ratio mass spectometry] with FinniganMAT Delta Plus). Concentrations of hydrocarbons are given in nanograms (109) hydrocarbons per gram of dry sediment (ppb) carbon

Table 2. Geochemical Results for Gas Samples from the Simeulue Basin and Two Reference Locations on the Outer-Arc High* Station 70SL 73KL 77SL 79SL 85SL 90SL 97MUC 102SL 105SL 106SL 112MUC 113SL 119KL 122SL 131SL 115KL Latitude (North) 23349.8 24948.2 24909.0 25749.2 25923.4 24659.1 23348.2 22913.1 31142.2 31014.5 35228.3 35226.4 33132.4 31629.3 33346.9 32929.6 Longitude (East) 964524.6 962306.2 963649.8 962959.1 961312.5 962456.5 964524.0 97716.9 964641.6 964648.5 96026.6 96027.1 961852.3 96914.7 964525.7 951952.7 Sample ID 70SL1 70SL2 73SL1 73SL2 77SL1 77SL2 79SL1 79SL2 85SL1 85SL2 90SL1 90SL2 97MUC1 102SL1 102SL2 105SL1 105SL2 106SL1 106SL2 112MUC1 113SL1 119KL1 122SL1 122SL2 131SL1 131SL2 115KL1 115KL2 115KL3 116KL1 139MUC1 140KL1 140KL2 140KL3 142KL1 Depth (m) 1.3 1.5 2.7 4.7 2.7 4.7 2.7 4.7 2.7 4.7 2.7 4.7 0.3 2.7 4.7 5.4 6.4 2.7 4.7 0.4 1.2 7.7 3.7 4.7 0.2 3.7 5.6 7.7 9.7 8.7 0.4 6.8 6.8 7.8 8.6 CH4 (ppb) 92 108 60 100 68 66 52 61 125 41 75 98 512 60 51 42 53 101 48 37 27 92 75 104 76 113 82 73 78 71 124 11,748 277 20,823 10,281 C2H6 (ppb) 9 9 5 7 6 6 4 5 9 2 6 8 8 6 4 2 4 14 4 2 2 6 5 6 7 7 9 9 8 9 9 15 10 19 20 C2H4 (ppb) 1 2 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 2 C3H8 (ppb) 5 6 2 3 3 3 2 2 5 1 2 3 4 1 2 1 2 7 2 2 1 3 2 4 3 4 5 5 4 4 5 7 8 11 12 C3H6 (ppb) 1 1 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 1 0 1 1 1 d13C Methane 42.7 45.6 31.6 28.9 34.6 37.6 33.2 33.7 38.7 37.8 37.2 41.2 68.0 41.6 33.6 33.9 41.0 36.0 33.6 16.4 15.1 34.0 27.8 31.9 39.6 43.6 29.5 30.8 34.2 28.4 49.8 91.1 79.7 90.4 85.1

116KL 139MUC 140KL

32934.1 14534.4 14534.3

951953.1 964626.0 964626.3

142KL

14534.2

964626.4

*Gas concentrations are given in nanograms of gas per gram of dry sediment, d13C of CH4 in vs. PDB.

isotope data of methane in the standard delta notation versus PDB.

STRATIGRAPHY Published data from exploration wells (Karig et al., 1979; Rose, 1983; Beaudry and Moore, 1985)

drilled on the west Sumatra shelf in the 1970s (Figures 2, 4) provide lithologic information (rock units) along with a general age for the sediments. Berglar et al. (2008) reviewed existing data and established a detailed seismic and chronostratigraphic framework. The conceptual model of this study is based on the stratigraphy and results of these authors; therefore, we briefly summarize
Lutz et al. 1595

Figure 4. Location of Tuba-1A, Meulaboh-1, and Tripa-1 exploration wells on composite seismic section. The stratigraphy of the Meulaboh-1 well is used in this study for chronostratigraphic correlation of seismic horizons (Berglar et al., 2008). Two phases of carbonate buildup growth occurred in the early to middle Miocene and in the late Miocene to early Pliocene, respectively. Line-tie with profile BGR06-137 is indicated at the top. See Figure 2 for location of profiles. Green and blue lines denote major unconformities, and yellow lines mark the top of carbonate buildups. The inset shows a close-up of onlapping Pleistocene sediments on the upper MiocenePliocene unconformity.

their findings illustrated on a composite seismic section (Figure 4) and present a model of the Simeulue forearc basin evolution along this section (Figure 5). Three major unconformities occur throughout the basin and enable a correlation of the four main sedimentary units imaged in seismic data. Sub-Neogene The lowermost unconformity marks the top of the acoustic basement and separates the sub-Neogene from Miocene sediments. Some low-frequency, high-amplitude reflection bands are imaged below the unconformity. They may be concordant to the unconformity or steeply dipping. Acoustic basement was drilled in the Meulaboh-1 well and described as Eocene to Oligocene gray to black mud1596 Simeulue Fore-arc

stone with minor interbeds of silt and sandstones (Rose, 1983; Beaudry and Moore, 1985). Lower and Middle Miocene The lower and middle Miocene succession shows parallel to subparallel continuous reflectors and concordantly overlies the basal Neogene unconformity with onlapping contacts. This seismic reflection pattern is attributed to clastic sediments. In the Meulaboh-1 well, this unit is described as nearshore marine and nonmarine clastics. The thickness of the sediments reaches 3 s (TWT) (4200 m [13,800 ft]) in the deepest part of the basin. On topographic highs of the acoustic basement, several carbonate buildups developed on the paleoshelf (phase I). These carbonates are described as interbedded micrite and packstone. The packstone

Figure 5. Schematic evolution of the Simeulue Basin along a 2-D cross section based on seismic profiles shown in Figure 4. The thin, dark-pink layer represents a hypothetical Eocene source rock. First carbonate buildup growth during the early Miocene on topographic highs (bright blue color). Continued growth of carbonate buildups until middle Miocene. Subsequent formation of unconformity by short-lived uplift and initiation of second phase of carbonate buildup from late middle Miocene until late MiocenePliocene. Strong subsidence and clastic sediment supply from the rising Barisan Mountains during PliocenePleistocene shut off carbonate production and led to burial of the buildups and increasing water depth in the present-day basin center. Lutz et al. 1597

Figure 6. Structural contour (depth) maps for four different time steps of basin evolution. Top left, early to middle Miocene: Three small basins developed and first carbonate buildups. Top right, middlelate Miocene: Western part of the basin formed an elongated trough. Bottom left, late MiocenePliocene: Second carbonate buildup phase along the paleoshelf edge. Bottom right present-day water depth. Projection Universal Transverse Mercator 48S.

grains consist of corals, algae, foraminifera, pelecypods, and bryozoans. Locally, vuggy and moldic porosity exists (Rose, 1983). Three small basins are distinguished (Figure 6, top left), which developed along the western margin of the Simeulue Basin. These depressions likely formed in response to flexural subsidence caused by irregular load of the
1598 Simeulue Fore-arc

evolving accretionary prism (Matson and Moore, 1992; Berglar et al., 2008). Upper Miocene and Pliocene A phase of rapid subsidence of the forearc basin center is marked by an unconformity that separates

middle Miocene from upper Miocene sediments (Figure 4). The onlapping contact of upper Miocene sediments is best imaged in the center of the basin. The seismic appearance of this sequence is similar to that below the unconformity: highly continuous high-amplitude reflections having low to moderate frequency. After consolidation and uplift of the outer-arc high, the western part of the basin formed a subsiding elongated trough (Figure 6, top right), whereas carbonate buildups developed on the shallow shelf (phase II; Figure 6, bottom left). The carbonates are composed of packstones and grainstones. Biogenic particles are formed by red algae, foraminifera, bryozoans, corals, echinoids, and pelecypods (Rose, 1983). The carbonate buildups were subsequently buried by continuing sedimentation of clastic deposits (Figure 5). Pleistocene to Holocene The entire basin is draped with a cover of wellstratified Pleistocene sediments (Figure 5) reaching a thickness of 0.6 s (TWT) (500 m [1650 ft]). The sediments consist of sand, clay, and silt (Rose, 1983). This sequence onlaps the Pliocene sediments at the basin fringe (Figure 4, inset) and reflects the present-day depocenter in deep water (Figure 6, bottom right).

RESULTS The MCS data image several striking features, including carbonate buildups, bright spots, and bottom-simulating reflectors in the Simeulue Basin, which underpin our conceptual model and were incorporated into a petroleum system model. Carbonate Buildups Miocene carbonate platforms are common in Southeast Asia and form important petroleum reservoirs in the South China Sea, for example, Luconia Province, Malampaya and Pearl River Mouth Basin (Fournier and Borgomano, 2007). In the Simeulue Basin, more than 30 carbonate buildups

were identified. Distinct high-amplitude reflections are interpreted as the top and flanks of the carbonate buildups. Internal seismic reflectivity is generally weak, but individual reflectors are partially visible. The shape of the carbonate buildups varies. Some exhibit an asymmetrical geometry, whereas others are developed as pinnacle reefs. The slopes are always steeper than the surrounding strata that onlap the flanks. Some buildups have a plateau-like top, whereas others have an irregular uneven surface (Figures 7, 8). A chaotic seismic reflection pattern may occur at the shelfward side (Figure 7), which is interpreted as reef debris. The reef debris sequence was deposited in the carbonate platform interior. Two phases of carbonate buildup are recognized (Figure 7). Few buildups formed in a first phase (early to middle Miocene) immediately after the formation of the sub-Neogene/lower Miocene unconformity. Commonly, these grew on structural highs of the sub-Neogene basement. The major phase of carbonate buildup growth occurred after development of the upper middle Miocene to lower upper Miocene unconformity. The buildups developed in shallow-water areas between the paleoshelf break and the shoreline, some on top of older buildups of the first phase. The tops of the earlier buildups were eroded during the late middle Miocene to early late Miocene (Figure 7), whereas the basinward flank of the buildup is covered by middle Miocene sediments. The buildups of the second phase are covered by upper Miocene to Holocene sediments. Seven buildups of the first phase and 26 of the second phase were mapped. The horizontal extent of carbonate buildups along the two-dimensional (2-D) seismic profiles ranges from less than 1 km (<0.6 mi) to more than 10 km (>6 mi). One carbonate buildup was transected at different angles by lines BGR06-211 and BGR06-137, and we measured an extension of 7.5 km (4.7 mi) and 8.5 km (5.3 mi), respectively. The thickness of the carbonate buildups is variable. Along line BGR06-135 (Figure 7), the maximum thickness of carbonate buildups was identified: 0.25 s (TWT) (350 m [1150 ft]) for the lower buildup and 0.41 s (TWT) (600 m [1950 ft]) for the upper one.
Lutz et al. 1599

Figure 7. Part of multichannel seismic profile BGR06-135 (upper panel) and interpretation (lower panel). Two carbonate buildups grew at the same location on a sub-Neogene basement high. The growth phases were interrupted by a drowning phase, and the lower buildup was truncated by an unconformity. The carbonate buildups mark the paleoshelf edge. The upper Miocene to lower Pliocene paleoshelf is occupied by carbonate platform sediments with a chaotic reflection pattern. Arrows indicate onlapping strata at the flank of the buildups (orange lines separate the phases and the green line denotes the top of the reef debris unit). See Figure 2 for location of profile.

An example of the growth history of an individual buildup is shown in Figure 9. Here, we have adopted the evolutionary model for a carbonate buildup from Luconia (Zampetti et al., 2004). The upper Miocene to lower Pliocene buildup grew on a lower to middle Miocene buildup. The growth of the second buildup started in the late Miocene above the unconformity that separates the middle Miocene from the upper Miocene. High-amplitude, low-frequency reflector bands are characteristic of the initial phase of platform growth (unit A on Figure 9). The aggradation phase is documented by unit B, which shows the typical low-reflection pattern. A wedge with a low-reflection discontinuous pattern is interpreted as a lowstand wedge, which implies a short-lived relative sea level fall. High-amplitude reflections mark the change from aggradation to drowning (unit C), thus indicating a subsequent relative sea level rise. The growth of the buildup is accompanied by debris and platform
1600 Simeulue Fore-arc

deposits with a chaotic reflection pattern in the vicinity of the buildup. The final phase of submergence is characterized by a chaotic low-reflection seismic pattern (unit D). Drowning of the buildup resulted in backstepping of units C and D. Progressive drowning is exhibited by onlapping of the surrounding strata. A transgressive cap terminates the drowning phase, and the buildup is completely buried by sediments. At the steeper north-northwest flank of the buildup, normal faults cut through the buildup. These faults may originate by oversteepening and collapse of the slope or by abundant earthquakes related to the subduction zone. A distinct velocity pull-up at the base agrees with the interpretation of high-velocity carbonate rocks (Figure 9). Bottom-Simulating Reflectors Bottom-simulating reflectors are common phenomena in multichannel seismic profiles at continental

Figure 8. Part of multichannel seismic profile BGR06-139 (upper panel) and interpretation (lower panel). Carbonate buildups grew above a regional unconformity at the paleo-shelf edge. Faults cut through the upper sedimentary succession and occasionally reach the sea floor. Ellipsoids mark areas with bright spots. BSR = bottom-simulating reflector; arrow indicates onlapping strata above an unconformity at the slope of the basin. The locations for amplitude versus offset analysis (left, Figure 12; right, Figure 13) are marked at the top. See Figure 2 for location of profile. TWT = two-way traveltime.

margins. Occurrences of BSRs are well constrained, where BSRs cut through dipping layers. However, identification is difficult where the layering of the sediments parallels the sea floor and the BSR. At some places, BSRs are caused by the diagenetic transition from opal A to opal CT (Berndt et al., 2004). However, BSRs mostly evolve at the base of the gas hydrate stability zone and mark the negative impedance contrast between pore space that is filled by hydrate and the underlying gas-charged sediments (Max and Dillon, 1998). The BSRs exhibit a clear negative polarity (negative acoustic impedance) compared with the sea-floor reflection, and they cross-cut the sedimentary layering. The depths of BSRs are controlled mainly by temperature, pressure, and composition of the pore fluids (Pecher et al., 1998; Sloan, 1998).

Multiple BSRs occur close to Simeulue Island, that is, a succession of two (Figure 10) or more BSRs at different depths. The BSRs are observed at several locations, predominantly in the southern part of the Simeulue Basin (Figure 11). The lower ones might be caused by mineralogical changes (Berndt et al., 2004). However, the uppermost BSR marks the current base of the gas hydrate stability zone (Popescu et al., 2006). This upper BSR occurs in the Simeulue Basin between 217 m (712 ft) below sea floor (mbsf) and 642 mbsf (2106 ft) at water depths between 831 m (2726 ft) and 1157 m (3796 ft). The double BSRs in Figures 8 and 10 are separated by 70 to 80 ms (TWT). Heat-flow values were derived from BSR depths according to the method described above, ranging between 37 and 74 mW/m2.
Lutz et al. 1601

Figure 9. Part of multichannel seismic profile BGR06-138 (upper panel); the black-outlined box marks the section of the interpretation (lower panel). In the upper buildup, the buildup stages A and B and build-in stages (rate of carbonate production < pace of sea level rise) C and D are distinguished. The surrounding area is covered by debris and platform deposits; a lowstand wedge is interpreted on the north-northwest flank. Half arrows indicate onlapping strata at the flank of the buildups. See Figure 2 for location of profile. TWT = twoway traveltime.

Bright Spots High-reflection zones with strong amplitude that are visible in seismic profiles are often called bright spots. Such zones are caused by changes in the physical properties of the sediments (Lseth et al., 2009). They indicate either changing lithology (e.g., cementation) or changing fill of pore space (e.g., presence of gas or oil). Distinct bright spots are shown in Figure 8, where they occur above carbonate buildups at depths between 1.8 and 2.1 s ( TWT ). The highamplitude reflectors extend over a distance of more than 40 km (>25 mi). Above the northwestern carbonate buildup, the bright spot occurs in a gentle anticline. The other bright spots above the build1602 Simeulue Fore-arc

ups occur in layers that probably represent coarser grained sediments. At the northwestern rim of the carbonate buildup in Figure 8, a bright spot is evident in sediments that onlap the toe of the carbonate buildup at a depth of 2.6 s (TWT). Above the southeastern carbonate buildup, a young fault system cuts from the buildup through the bright spot to the sea floor. No amplitude anomalies are observed along these faults, indicating closed faults. The highest amplitudes of a bright spot are located above the carbonate buildup shown in Figure 10. The bright spot extends in a southwesterly direction for approximately 8 km (5 mi) as a layer of parallel high-amplitude reflectors and ascends within the inclined sediments from 1.9 to 1.7 s (TWT), crossing the BSRs.

Figure 10. Part of multichannel seismic profile BGR06-137 (upper panel) and interpretation (lower panel). Inset (upper panel) documents the crosscutting of bottom-simulating reflector (BSR) and sediment layers. A bright spot (ellipsoids) occupies sediments above carbonate buildups. It extends parallel to the layering above two distinct bottom-simulating reflectors. The arrow (bottom right) marks cutoff layers at an erosional truncation. Line-tie with profile BGR06-211 is indicated at the top. See Figure 2 for location of profile; TWT = two-way traveltime.

Bright spots occur in the Simeulue Basin, especially in the vicinity of the area having the thickest sediment, mostly above carbonate buildups. Carbonate buildups and bright spots are marked in Figure 11. Amplitude Versus Offset Analysis The following section describes the results of the AVO analysis for seismic line BGR06-139 (CDPs 20,00028,000). The maximum source-receiver offset was 3137.5 m (10,294 ft), and the water depth was approximately 1100 m (3609 ft;

1450 ms [TWT]). The target depth was 1950 ms (TWT), that is, 1500 to 1600 m (4921 to 5249 ft) depth. Although the reflections are not perfectly flat, distinct higher amplitudes are observable at 1900 to 1950 ms (TWT) (Figure 8). This becomes more obvious in offset-dependent stacks. Figure 12 shows an example of CDP from seismic line BGR06139. The upper image presents the raw amplitudes in the depth range 1880 to 2020 ms (TWT). Particularly interesting is the reflection at approximately 1930 ms (TWT) that depicts a pronounced increase in negative amplitudes with offset. The fact that the reflection bends upward at higher
Lutz et al. 1603

Figure 11. Map showing the location of carbonate buildups (green), bright spots (blue), and bottom-simulating reflectors (BSRs) (red) interpreted on multichannel seismic profiles. 1 = Teunom-1; 2 = Bubon-1; 3 = Keudepasi-1; 4 = Tuba; 5 = Seunagan-1; 6 = Meulaboh-East-2; 7 = Meulaboh-East-1; 8 = Meulaboh-1; 9 = Tripa-1; 10 = Palumat-1; 11 = Ujung-Batu-1; 12 = Pulau-Baru-1; 13 = Palambak; 14 = Singkel-1; 15 = Telaga.

offsets, that is, the picked average velocities are too high, may also suggest gas-bearing sediments at this location. The picked raw amplitudes increase from about 12 to 25, as shown in the panel below the seismic section. The lower image in Figure 12 shows the same CDP, but a wider depth area, ranging from 1300 to 2300 ms (TWT). Most reflections show decreasing amplitudes with increasing offset, indicating that amplitude corrections during processing were chosen accurately. However, the interpreted gas sand reflection at 1930 ms (TWT) shows increasing amplitudes with increasing offset and a phase reversal compared with the water-bottom reflection. The amplitudes normalized to those of the water-bottom reflection (Figure 12, lower panel) increase with offset from 0.4 to 1.5. Figure 13 shows range-limited stacks from a part of line BGR06-139 (CDPs 21,00022,000; near offset stack range: 648 to 1150 m (2126 to 3773 ft); far offset stack range: 26403137.5 m [866110,294 ft]). The sea-bottom amplitudes at
1604 Simeulue Fore-arc

approximately 1500 ms (TWT) are merely the same in the near-offset and far-offset stacks, but the reflections at approximately 1900 to 2000 ms (TWT) depth show higher amplitudes in the faroffset stack (Figure 13, bottom) than in the nearoffset stack (Figure 13, top). From these observations, we conclude that the observed bright spots have negative AVO intercept and gradient. These anomalies can therefore be interpreted to be caused by gas sands that are typical class III anomalies (Rutherford and Williams, 1989; Ross and Kinman, 1995). These sands exhibit the amplitude increase versus offset that is commonly used as a gas indicator.

Geochemistry Several exploration wells were drilled during an exploration phase from 1968 to 1978 (Rose, 1983) close to the coast along the northern shelf, in water depth mainly less than 100 m (<328 ft). Three wells

Figure 12. Supergather formed from CDPs 26,000 to 26,010 of line BGR06-139. Top: Raw amplitudes in the depth range from 1880 to 2020 ms (two-way traveltime [TWT]) and extracted raw amplitudes along the prominent negative reflection at 1920 to 1930 ms (TWT). Bottom: Raw amplitudes in the depth range from 1300 to 2300 ms (TWT) and extracted normalized amplitudes along the water bottom and the prominent negative reflection at 1920 to 1930 ms (TWT).

encountered noncommercial quantities of gas. Isotopic data on gas composition are not published, but the fact that only methane is reported suggests a biogenic origin of this gas. This study adds new gas geochemical data to the sparse database. Methane concentrations of desorbed sediment samples (see Methods section) from the Simeulue Basin range from 27 to 512 ppb, ethane and propane from 2 to 14 ppb and 1 to 7 ppb, respectively

(Table 2). A linear correlation between ethane and propane concentration can be observed (Table 2). However, the ethane concentrations are very low (mostly <10 ppb) and comparable to data from the southern Mentawai Basin, where samples were obtained during a previous cruise (SO139). In the Mentawai Basin, this concentration was interpreted to represent a background value (Faber et al., 2001). The d13C isotopic composition of methane ranges,
Lutz et al. 1605

Figure 13. The 6250-m (20,505-ft) long part of seismic line BGR06-139 (CDPs 21002200). Top: Near offset (6481150 m [2126 3773 ft]) stack with the sea bottom at 1500 ms (TWT). Bottom: Same part of line as on top, but displayed as far offset (26403137.5 m [866110,294 ft]) stack. High amplitudes at 1900 to 2000 ms (TWT) indicate gas-bearing sediments.

with the exception of two samples, generally between 27.8 and 68 (Figure 14; Table 2). Two samples reveal unusually heavy carbon isotopes, 112MUC 16.4 and 113SL 15.1. Three samples from reference cores 140KL and 142KL on the outer-arc high show methane
Figure 14. Gas wetness and carbon isotopes of adsorbed methane in the Bernard diagram. The line between the marine source rock (SR) and bacterial gas indicates hypothetical composition of a mixed bacterial and thermogenic gas.

concentrations higher than 10,000 ppb, the concentrations of the remaining samples from the reference sites (82277 ppb) are within the range for adsorbed methane in the Simeulue Basin. All cores of the reference sites exhibit similar ethane and propane concentrations (820 and 412 ppb,

1606

Simeulue Fore-arc

respectively) and are comparable to the samples from the Simeulue Basin. The carbon isotopic composition of methane ranges between 91.1 and 28.4. The origin of adsorbed gas can be determined by comparison of the chemical and isotopic composition of the major constituents. Figure 14 shows the data from the Simeulue Basin, the reference cores from the outer-arc high and, for comparison, data from the Mentawai Basin in the Bernard diagram. Typically, bacterial gas is rich in methane, and ethane occurs, if any, in traces, giving wetness ratios methane/S(ethane + propane) more than 250. Isotopic composition of bacterial methane is very light (<60) as compared with thermal gases. The different source rocks of the thermal gases (marine or terrigenous) can be distinguished by the Bernard diagram because of their different composition. Mixing of different gas types is easily recognized (compare the theoretical mixing line of two different end members in Figure 14). With a few exceptions, the wetness ratio methane/S(ethane + propane) of the adsorbed gases from the Simeulue Basin indicate thermal generation of the hydrocarbons (Figure 14). One sample from the Simeulue Basin (97MUC, d13C 68) shows a large contribution of isotopically light biogenic methane and can be interpreted as a mixed gas. Wetness data combined with the d13C isotopic composition of methane imply an active petroleum system with a marine source rock in the Simeulue forearc basin. This is in clear contrast to the results from the southern Mentawai Basin. Available gas data from the Mentawai Basin point toward almost exclusively biogenic gas without an active petroleum system. Data from the reference sites at the outer-arc high northwest of Simeulue Island (116KL and 115KL) do not show a distinctive difference from the results for the Simeulue Basin. The previously mentioned samples 140KL and 142KL (southeast of Simeulue Island) with exceptionally high methane concentrations evidently represent biogenic gas. This is supported by the wetness ratio (>700) and d13C values of these samples, which are generally lighter than 80. The adsorbed gas from sample 139MUC represents a mixture gas similar to sample 97MUC from the Simeulue Basin.

Conceptual Model A conceptual model was derived, which integrates the results from seismic stratigraphy, bright spot, carbonate buildup, and BSR occurrences, as well as AVO analysis and organic geochemistry. In the conceptual model, source rocks for oil and gas are assumed to occur at two stratigraphic levels, namely Eocene oil-prone source rock within the acoustic basement and lower to middle Miocene terrigenous source rock. Although the gas geochemical data favor a marine source rock, we also incorporated a lower to middle Miocene terrigenous source rock. The reason for doing this is that the bright spots are concentrated around the depocenter in the Simeulue Basin (Figure 15). It is possible that a source rock within the depocenter contributes to the petroleum system. If this is correct, the most likely candidate is lower to middle Miocene terrigenous source rock. This assumption of two candidate source rocks is based on the similar geologic evolution of the present-day back-arc region, where these source rocks are present and the forearc region is under study (Todd et al., 1997; Doust and Noble, 2008). From the Late Cretaceous to the early Eocene, before the emergence of the Barisan Mountains, northern Sumatra formed a stable platform, extending northward into the North Sumatra Basin and westward into a continental shelf in the area of the present forearc basins and outer-arc high (De Smet and Barber, 2005). The upper part of the acoustic basement was drilled at the Meulaboh-1 and Tuba-1A wells and consists of Eocene or Oligocene mudstones with minor interbeds of siltstone and sandstone, indicating sediments in the upper part of the acoustic basement. Lower Miocene sediments consist of nearshore marine and nonmarine mudstone, sandstone, siltstone, and minor coal beds (Rose, 1983). During basin subsidence (Paleogenelate Oligocene, early Miocenemiddle Miocene, late MiocenePliocene, PleistoceneHolocene) and burial of the source rocks, hydrocarbon generation took place in the deep Neogene depocenter and in the Paleogene sediments below the depocenter. Subsequent expulsion and migration resulted in hydrocarbon accumulations in Miocene carbonates.
Lutz et al. 1607

Figure 15. Conceptual model (bottom) depicted on seismic profile BGR06-139 (top). See Figure 2 for location of profile. TWT = twoway traveltime.

Figure 15 depicts the main elements of the conceptual model. The bright spots within the upper Miocene to lower Pliocene sediments (northwest of depocenter) might be caused by thermally generated hydrocarbon gas. The bright spots above the carbonate buildups may be related to hydrocarbon gas that bypassed the carbonates or that escaped from the carbonate buildups. The Pliocene clastic sediments form the seal above the upper Miocene carbonate buildups, which is the case in two localities (Meulaboh-1 and Keudapasi wells), where gas accumulations were found (Rose, 1983).

PETROLEUM SYSTEM MODELING One-Dimensional Modeling Published vitrinite reflectance values (Hadiyanto, 1992) from six wells were used to calibrate the models. These wells were drilled during an exploration phase from 1968 to 1978 along the northern
1608 Simeulue Fore-arc

shelf (Figure 16). At the Meulaboh-1 and Tuba-1A wells, the sub-Neogene mudstones and siltstones were drilled. Thus, the 1-D models at these locations start in the Eocene and include the upper Oligocenelower Miocene unconformity. All other wells terminated in Neogene sediments (Tripa = lower Miocene; Keudepasi, Bubon, Teunom = middle Miocene). Below the drilled Neogene sediments, approximately 1000 m (3281 ft) of sedimentary basement was assumed from drilling results in the Meulaboh-1 and Tuba-1A wells. A constant heatflow model was assigned to each 1-D model, although the heat-flow history might have been more complex during the rifting phase. The model cannot be further refined based on the available data. However, it was possible to derive a general trend of heat flow across the wells. By adjusting the heat flow (and predicted vitrinite reflectance) to match the measured vitrinite reflectance values of the individual wells, we found a trend of increasing heat flow from the southeast to the northwest (Figure 16). At the Tripa well, the measured vitrinite reflectance

Lutz et al. Figure 16. One-dimensional models of nearshore wells (AF) with published vitrinite reflectance values (Ro) (Hadiyanto, 1992) used for calibration of the models. Left parts of the panels show measured (dots) and calculated (solid line) vitrinite reflectance values. Right parts of the panels show burial history of the respective well and the heat-flow value used for calculation. Wells Meulaboh (B) and Tuba-1A (C) reached sub-Neogene sediments, all other wells terminated in Neogene sediments. All results for each well were calculated using a constant heat-flow model. The red dots in the Meulaboh well are from the Meulaboh East-1 well. See Figure 2 for location of wells. 1609

values could be matched using a heat flow of 40 mW/m2. A heat flow of 50 mW/m2 was used at Meulaboh-1, 55 mW/m2 was used at Tuba-1A, and 60 mW/m2 was used for the Keudepasi and Bubon wells (Figure 16). The calibration of the model for the Teunom well required an exceptionally high heat flow of 100 mW/m2. Three-Dimensional Modeling The 3-D model area is approximately 240 km (~149 mi) in the northwest-southeast direction and approximately 100 km (~62 mi) in the northeastsouthwest direction. It is based on sediment isochron maps for the different stratigraphic levels that were gridded on the basis of 2-D seismic data. The data were depth converted using interval velocities as derived from stacking velocities. The imported maps have a regular grid spacing of 500 m (1640 ft). Imported maps include the recent sea floor, base Pleistocene, base Pliocene, base upper Miocene, and base lower Miocene. Thicknesses for the sedimentary layers below the acoustic basement were extrapolated from the two deep wells on the shallow shelf that reached the acoustic basement. We estimated the thicknesses as follows: 100 m (328 ft) Oligocene, 50 m (164 ft) Eocene source rock, 400 m (1312 ft) Eocene sediments, and 500 m (1640 ft) lower Eocene sedimentary basement. Erosion during late Oligoceneearly Miocene removed 500 m (1640 ft) of sediments in our models, and erosion during the late middle Miocenelate Miocene affected only the southeastern part of the basin, with removal of as much as 500 m (1640 ft) of sediments, whereas erosion did not occur on the present-day shelf or in the central part of the basin. According to the conceptual model, two source rocks were assigned. The Eocene source rock was classified as organofacies B, and the lower to middle Miocene source rock was classified as organofacies D/E (Pepper and Corvi, 1995). Based on the assigned organofacies, the respective reaction kinetic data sets were assigned (Pepper and Corvi, 1995), and both source rocks are capable of oil generation. A total organic carbon (TOC) content of 3% and a hydrogen index (HI) of 300 mg HC/g TOC were used.
1610 Simeulue Fore-arc

The heat-flow values derived from the 1-D models of the nearshore wells are in the range of 40 to 60 mW/m2, with the exception of the Teunom well. This range is further constrained by calculation of recent heat-flow values as derived from BSR depths. To acknowledge this range, we calculated two different models or scenarios, one with a heat flow of 40 mW/m2 and a second with 60 mW/m2. The final model was calculated with cell sizes of 1.5 1.5 km (0.9 0.9 mi). In the 40-mW/m2 model, large parts of the Simeulue Basin remain thermally immature. Only deeper levels in the center and in the southern part of the basin reach the oil and wet gas zone. In the 60-mW/m2 model, the zones of hydrocarbon generation enclose a wider area. Sediments in the deepest part of the basin are presently overmature (Figure 17). Figure 18 illustrates the results of a pseudowell at approximately the location of the thickest Neogene sedimentary succession (location in Figure 2). Burial history at this location and the calculated transformation ratios (TR) and temperatures are shown in Figure 18 for the 40-mW/m2 and 60-mW/ m2 models. In the 40-mW/m2 model, hydrocarbon generation from the Eocene source rock began at 16 Ma. Expulsion (10% TR) started at 15 Ma, and peak expulsion (50% TR) occurred at 10 Ma. Hydrocarbon generation at the base of the lower Miocene sediments started at 15 Ma, and onset and peak expulsion occurred at 11 and 5 Ma, respectively. The Eocene source rock reaches transformation ratios of 85% at the base, whereas the lower Miocene sediments reach 65% transformation ratio (Figure 18). In the model with a higher heat flow (60 mW/ m2), hydrocarbon generation from the Eocene source rock began at 19.5 Ma. Expulsion (10% TR) started at 16 Ma, and peak expulsion (50% TR) occurred at 14 Ma. Hydrocarbon generation at the base of the lower Miocene sediments started at 17 Ma, and onset and peak expulsion occurred slightly later than in the Eocene source rock at 15 and 13 Ma, respectively. Oil expulsion ceased (95% TR) at 10 Ma for the Eocene source rock and at 5 Ma for the basal Miocene sediments. The

Figure 17. Calculated vitrinite reflectance (Ro) (Sweeney and Burnham, 1990) and respective zones of hydrocarbon generation are shown for two models; one with 40 mW/m2 (left) and one with 60 mW/m2 (right). The white line and hatched plane indicate the location of the 2-D cross sections (bottom). The deepest parts of the present-day depocenter reach well into the zone of gas formation in both models.

Figure 18. Calculated transformation ratio (TR) for two source rocks (thin Eocene source rock at the bottom; lower middle Miocene source rock) at the pseudowell (for location see Figure 2). Isolines show temperature evolution. Left: Calculated transformation ratio for the 40-mW/m2 model. Right: Calculated transformation ratio for the 60-mW/m2 model. Lutz et al. 1611

Eocene source rock and the base lower Miocene sediments have lost their hydrocarbon-generation potential present-day at this location, but large parts of the lower to middle Miocene sediments are still capable of hydrocarbon generation and expulsion (Figure 18).

DISCUSSION South China Sea carbonate buildups are prolific hydrocarbon reservoirs (Zampetti et al., 2004; Fournier and Borgomano, 2007), which highlights their potential as hydrocarbon reservoirs in other areas. However, well data (cores, logs) for the buildups situated in the present deep-water area of the Simeulue Basin do not exist. Therefore, we must rely on seismic data and analog examples for interpretation of the paleo-environment and evolution of the carbonates. Reservoir properties, for example, porosity and permeability, can only be assumed from analogs. In addition, intracarbonate seals may have developed, which might limit migration. Potential seals above the carbonate buildups are unproven. Candidates are the upper Miocene and Pliocene clastic sediments deposited during basin subsidence. These form the seal above the gas reservoirs at the Meulaboh-1 well. The presence of bright spots can be explained by migration of thermally generated hydrocarbon gas from the depocenter (location of pseudowell) into the upper Miocene and Pliocene sediments to the northwest and southeast. Horizontal migration distances are mostly less than 30 km (<19 mi). The leakage of hydrocarbon gas out of the carbonate buildups may also contribute to the formation of bright spots, but gas chimneys were not observed. Multiple BSRs occur close to Simeulue Island in the southwestern part of the basin. Multiple BSRs are explained by compositional changes in the gas hydrates that affect their stability fields, a transitional zone between gas hydrates and free gas, or some BSRs may represent past (relict) positions of the hydrate stability zone (Foucher et al., 2002; Popescu et al., 2006). Multiple BSRs are located on the slope of the uplifted Simeulue Is1612 Simeulue Fore-arc

land, which supports the interpretation that the lower BSRs in a BSR sequence are the result of rapid uplift caused by tectonic activity and therefore represent paleo-BSRs. The Simeulue Basin was subject to recent strong tectonic activity and basin inversion (Berglar et al., 2008). An important parameter for hydrocarbon generation is the temperature history of the sedimentary basin. This is controlled by the basal heat flow, a critical input parameter in the numerical model. Several sources for heat-flow determination were available. Measurements of recent heat flow at several locations in the Simeulue Basin provided values between 47 and 107 mW/m2 (Delisle and Zeibig, 2007). In particular, the higher values were measured at active faults, where fluid expulsion occurs, thus documenting values that are not representative of the entire basin. Calculations of heat flow based on the depth of BSRs are useful, where wells are absent (Grauls, 2001) and deliver information on a larger area of the basin. The calculated heat-flow values in this study range between 37 and 74 mW/m2. The third source of heat-flow values was derived by calculating 1-D models for the six wells having published vitrinite reflectance values. Calibration using the measured vitrinite data resulted in heat-flow values between 40 and 100 mW/m2. The very high heat flow of 100 mW/ m2 at the Teunom well might be biased by its proximity to the volcanic chain on Sumatra. Another explanation might be the location on a compressional ridge at the northern end of the Simeulue Basin (Izart et al., 1994; Mosher et al., 2008; Berglar et al., 2010), where strike-slip faults might act as fluid conduits, and hence warm fluids could have locally heated the sediments. In any case, we do not consider this value as representative. All other heat-flow values used in the 1-D models (40, 50, 55, 60 mW/m2) are in the range of the BSR-derived heat flows. Based on these results, we calculated two heat-flow scenarios for the 3-D model, with 40 and 60 mW/m2, respectively. The Eocene source rock lost its present-day hydrocarbon generation potential in the deepest part of the basin (Figure 18) in the 60-mW/m2 model; but at shallower depth, hydrocarbon generation is still possible. In the 60-mW/m2 model,

hydrocarbon generation started in late early Miocene, and expulsion peaked in the deepest parts of the depocenter in the late middle Miocene when the first phase of carbonate buildup growth occurred. These buildups were partially eroded at their tops (Figures 7, 10) by an erosional event in the middle to late Miocene erosional event, which resulted in an unconformity and may have caused the loss of significant amounts of hydrocarbons. Subsequent generation of oil and gas from medium depths near the depocenter and migration of hydrocarbons from deeper parts resulted in the filling of the carbonate buildups during the Pliocene and Pleistocene (Figure 18). In the 40-mW/m2 model, the Eocene source rock and the lower-middle Miocene sediments retain present-day hydrocarbon generation potential. Hydrocarbon generation initiated during the early middle Miocene and peaked in the late Miocene for the deeply buried Eocene source rock and in the early Pliocene for the base of the lower Miocene sediments. In this model, the amount of generated hydrocarbons is too low to significantly charge the carbonate buildups. However, the bright spots may be caused by relatively small quantities of gas. Most of the hydrocarbons remain within the source rocks, thus significant expulsion and migration from the lower middle Miocene sediments did not occur in this model. The calculated transformation ratios (Figure 18) show that in both models, hydrocarbons could have been generated and may still be generated in the Simeulue Basin. From these results, it is obvious that the location of the source rocks in the sedimentary column is critical for the timing of hydrocarbon generation. Expulsion and migration are furthermore strongly dependent on the petrophysical properties of the overburden rock. However, identification of reservoir zones in the carbonate rocks is difficult because of strong heterogeneities of their petrophysical properties (Borgomano et al., 2008).

analyses and petroleum systems modeling to predict the timing and locations of hydrocarbon generation in a frontier area situated in a forearc setting. The main results are as follows: Bright spots are widespread in the central Simeulue Basin in upper MiocenePliocene sediments and are most likely caused by trapped hydrocarbon gases, as supported by AVO analyses. Heat flow in the basin ranges between 37 and 74 mW/m2, according to 1-D modeling of six wells and BSR depths. Higher heat flows were measured at active fault zones. Hydrocarbons migrated from the depocenter into potential reservoir rocks. Surface geochemical data suggest a thermal origin for hydrocarbon gas generated from marine source rock. Carbonate buildups likely serve as reservoirs and were charged with oil and gas based on a petroleum system model that assumes a heat-flow model of 60 mW/m2; upper MiocenePliocene sediments might act as reservoirs. If the modeled heat flow is 40 mW/m2, significantly less hydrocarbon generation is predicted. The amount of generated hydrocarbons is sufficient to explain the bright spots, but the carbonate reservoirs are not charged with significant amounts of hydrocarbons.

REFERENCES CITED
Allen, P. A., and J. R. Allen, 1990, Basin analysis: Principles and Applications: Oxford, Blackwell, 451 p. Barber, A. J., M. J. Crow, and M. E. M. de Smet, 2005, Tectonic evolution, in A. J. Barber, M. J. Crow, and J. S. Milsom, eds., Sumatra: Geology, resources and tectonic evolution: Geological Society (London) Memoir 31, p. 234259. Beaudry, D., and G. F. Moore, 1985, Seismic stratigraphy and Cenozoic evolution of west Sumatra forearc basin: AAPG Bulletin, v. 69, p. 742759, doi:10.1306/AD4627FE -16F7-11D7-8645000102C1865D. Benfield, A. E., 1949, The effect of uplift and denundation on underground temperatures: Journal of Applied Physics, v. 20, p. 6670, doi:10.1063/1.1698238. Berglar, K., C. Gaedicke, R. Lutz, D. Franke, and Y. S. Djajadihardja, 2008, Neogene subsidence and stratigraphy of the Simeulue forearc basin, northwest Sumatra: Marine Geology, v. 253, p. 113, doi:10.1016/j.margeo .2008.04.006.

CONCLUSIONS This study integrates geophysical, gas-geochemical, heat-flow, and well data with results from AVO

Lutz et al.

1613

Berglar, K., C. Gaedicke, D. Franke, S. Ladage, F. Klingelhoefer, and Y. S. Djajadihardja, 2010, Structural evolution and strike-slip tectonics off north-western Sumatra: Tectonophysics, v. 480, p. 119132, doi:10.1016/j.tecto.2009 .10.003. Berndt, C., S. Bunz, T. Clayton, J. Mienert, and M. Saunders, 2004, Seismic character of bottom simulating reflectors: Examples from the mid-Norwegian margin: Marine and Petroleum Geology, v. 21, p. 723733, doi:10.1016/j .marpetgeo.2004.02.003. Borgomano, J., F. Fournier, S. Viseur, and L. Rijkels, 2008, Stratigraphic well correlations for 3-D static modeling of carbonate reservoirs: AAPG Bulletin, v. 92, p. 789 824, doi:10.1306/02210807078. Busby, C. J., and R. V. Ingersoll, 1995, Tectonics of sedimentary basins: Oxford, Blackwell Science, 579 p. Clure, J., 2005, Fuel resources: Oil and gas, in A. J. Barber, M. J. Crow, and J. S. Milsom, eds., Sumatra: Geology, resources and tectonic evolution: Geological Society (London) Memoir 31, p. 131141. Cole, J. M., and S. Crittenden, 1997, Early Tertiary basin formation and the development of lacustrine and quasilacustrine/marine source rocks on the Sunda Shelf of SE Asia, in A. J. Fraser, S. J. Matthews, and R. W. Murphy, eds., Petroleum geology of Southeast Asia: Geological Society (London) Special Publication 126, p. 147183. Delisle, G., and M. Zeibig, 2007, Marine heat-flow measurements in hard ground offshore Sumatra: EOS Transactions American Geophysical Union, v. 88, p. 3839, doi:10.1029/2007EO040004. Demaison, G., and B. J. Huizinga, 1991, Genetic classification of petroleum systems: AAPG Bulletin, v. 75, p. 1626 1643. De Smet, M. E. M., and A. J. Barber, 2005, Tertiary stratigraphy, in A. J. Barber, M. J. Crow, and J. S. Milsom, eds., Sumatra: Geological Society (London) Memoir 31, p. 8697. Dickinson, W. R., 1995, Forearc basins, in C. J. Busby and R. V. Ingersoll, eds., Tectonics of sedimentary basins: Oxford, Blackwell Science, p. 221261. Doust, H., and R. A. Noble, 2008, Petroleum systems of Indonesia: Marine and Petroleum Geology, v. 25, p. 103 129, doi:10.1016/j.marpetgeo.2007.05.007. Faber, E., and W. Stahl, 1983, Analytical procedure and results of an isotope geochemical surface survey in an area of the British North Sea, in J. Brooks, ed., Petroleum geochemistry and exploration of Europe, International Congress: Geological Society (London) Special Publication 12, p. 5163. Faber, E., J. Poggenburg, and W. Stahl, 2001, Methane in seawater, in H. Beiersdorf, ed., Geoscientific investigations at the active convergence zone between the eastern Eurasian and Indo-Australian plates off Indonesia. Summary and synthesis of cruises SO137-139, and final report of cruise SO139: Hannover, Federal Institute for Geosciences and Natural Resources (BGR), p. 125. Foucher, J.-P., H. Nouze, and P. Henry, 2002, Observation and tentative interpretation of a double BSR on the Nankai slope: Marine Geology, v. 187, p. 161175, doi:10.1016 /S0025-3227(02)00264-5.

Fournier, F., and J. Borgomano, 2007, Geological significance of seismic reflections and imaging of the reservoir architecture in the Malampaya gas field (Philippines): AAPG Bulletin, v. 91, p. 235258, doi:10.1306 /10160606043. Franke, D., M. Schnabel, S. Ladage, D. R. Tappin, S. Neben, Y. S. Djajadihardja, C. Mueller, H. Kopp, and C. Gaedicke, 2008, The great Sumatra-Andaman earthquakes: Imaging the boundary between the ruptures of the great 2004 and 2005 earthquakes: Earth and Planetary Science Letters, v. 269, p. 118130, doi:10.1016/j.epsl.2008 .01.047. Fuller, C. W., S. D. Willett, and M. T. Brandon, 2006, Formation of forearc basins and their influence on subduction zone earthquakes: Geology, v. 34, p. 6568, doi:10 .1130/G21828.1. Gosnold, W., 2010, The global heat flow database, The International Heat Flow Commission: <http://www .heatflow.und.edu/index2.html> (accessed February 15, 2010). Grauls, D., 2001, Gas hydrates: Importance and applications in petroleum exploration: Marine and Petroleum Geology, v. 18, p. 519523, doi:10.1016/S0264-8172(00) 00075-1. Hadiyanto, 1992, Organic petrology and geochemistry of the Tertiary formations at Meulaboh area, west Aceh Basin, Sumatra, Indonesia: Ph.D. thesis, University of Wollongong, Wollongong, 219 p. Hermanrud, C., 1993, Basin modeling techniquesAn overview, in A. G. Dor, J. H. Auguston, C. Hermanrud, D. S. Stewart, and O. Sylta, eds., Basin modeling: Advances and applications: Norwegian Petroleum Society Special Publication 3, p. 134. Izart, A., B. Mustafa Kemal, and J. A. Malod, 1994, Seismic stratigraphy and subsidence evolution of the northwest Sumatra forearc basin: Marine Geology, v. 122, p. 109 124, doi:10.1016/0025-3227(94)90207-0. Jarrard, R. D., 1986, Causes of compression and extension behind trenches: Tectonophysics, v. 132, p. 89102, doi:10.1016/0040-1951(86)90027-2. Karig, D. E., S. Suparka, G. F. Moore, and P. E. Hehanussa, 1979, Structure and Cenozoic evolution of the Sunda arc in the central Sumatra region, in J. S. Watkins, L. Montadert, and P. W. Dickerson, eds., Geological and geophysical investigations of continental margins: AAPG Memoir 29, p. 223237. Karig, D. E., M. B. Lawrence, G. F. Moore, and J. R. Curray, 1980, Structural framework of the fore-arc basin, NW Sumatra: Journal of the Geological Society (London), v. 137, Part 1, p. 7791, doi:10.1144/gsjgs.137.1.0077. Karlsen, D. A., J. E. Skeie, K. Backer-Owe, K. Bjorlykke, R. Olstad, K. Berge, M. Cecchi, E. Vik, and R. G. Schaefer, 2004, Petroleum migration, faults and overpressure: Part II. Case history: The Haltenbanken petroleum province, offshore Norway, in J. M. Cubitt, W. A. England, and S. R. Larter, eds., Understanding petroleum reservoirs: Towards an integrated reservoir engineering and geochemical approach: Geological Society (London) Special Publications 237, p. 305372. Katz, B. J., 1995, A survey of rift basin source rocks, in J. J.

1614

Simeulue Fore-arc

Lambiase, ed., Hydrocarbon habitat in rift basins: Geological Society (London) Special Publication 80, p. 213 242. Kelley, P. A., B. Mertani, and H. H. Williams, 1995, Brown Shale Formation: Paleogene lacustrine source rocks of central Sumatra, in B. J. Katz, ed., Petroleum source rocks: Casebooks in Earth Sciences, Berlin, SpringerVerlag, p. 283308. Kvenvolden, K. A., and L. A. Barnard, 1982, Hydrates of natural gas in continental margins, in J. S. Watkins and C. L. Drake, eds., Studies in continental margin geology: AAPG Memoir 34, p. 631640. Lseth, H., M. Gading, and L. Wensaas, 2009, Hydrocarbon leakage interpreted on seismic data: Marine and Petroleum Geology, v. 26, p. 13041319, doi:10.1016/j.marpetgeo .2008.09.008. Malod, J. A., and B. M. Kemal, 1996, The Sumatra margin; oblique subduction and lateral displacement of the accretionary prism: Geological Society (London) Special Publication 106, p. 1928, doi:10.1144/GSL.SP.1996 .106.01.03. Matchette-Downes, C. J., A. E. Fallick, Karmajaya, and S. Rowland, 1994, A maturity and palaeoenvironmental assessment of condensates and oils from the North Sumatra Basin, Indonesia, in A. C. Scott and A. J. Fleet, eds., Coal and coal-bearing strata as oil-prone source rocks?: Geological Society (London) Special Publications 77, p. 139 148. Matson, R. G., and G. F. Moore, 1992, Structural influences on Neogene subsidence in the central Sumatra fore-arc basin: AAPG Memoir 53, p. 157181. Max, M. D., and W. P. Dillon, 1998, Oceanic methane hydrate; the character of the Blake Ridge hydrate stability zone, and the potential for methane extraction: Journal of Petroleum Geology, v. 21, p. 343358, doi:10.1111 /j.1747-5457.1998.tb00786.x. Mosher, D. C., J. A. Austin Jr., D. Fisher, and S. P. S. Gulick, 2008, Deformation of the northern Sumatra accretionary prism from high-resolution seismic reflection profiles and ROV observations: Marine Geology, v. 252, p. 89 99, doi:10.1016/j.margeo.2008.03.014. Pecher, I., C. R. Ranero, R. V. Huene, T. A. Minshull, and S. C. Singh, 1998, The nature and distribution of bottom simulating reflectors at the Costa Rican convergent margin: Geophysical Journal International, v. 133, p. 219 229, doi:10.1046/j.1365-246X.1998.00472.x. Pepper, A. S., and P. J. Corvi, 1995, Simple kinetic models of petroleum formation: Part I. Oil and gas generation from kerogen: Marine and Petroleum Geology, v. 12, p. 291 319, doi:10.1016/0264-8172(95)98381-E. Popescu, I., M. De Batist, G. Lericolais, H. Nouze, J. Poort, N. Panin, W. Versteeg, and H. Gillet, 2006, Multiple bottom-simulating reflections in the Black Sea: Potential proxies of past climate conditions: Marine Geology, v. 227, p. 163176, doi:10.1016/j.margeo.2005.12.006. Prawirodirdjo, L., and Y. Bock, 2004, Instantaneous global plate motion model for 12 years of continuous GPS observations: Journal of Geophysical Research, v. 109, p. B08405, doi:10.1029/2003JB002944. Pubellier, M., C. Rangin, J. P. Cadet, I. Tjashuri, J. Butterlin,

and C. M. Mueller, 1992, LiIle de Nias, un edifice polyphase sur la bordure interne de la fosse de la Sonde (archipel de Mentawai, Indonesie): Comptes Rendus de lAcademie des Sciences de Paris, Serie 2, Mecanique, Physique, Chimie, Sciences de lUnivers, Sciences de la Terre, v. 315, p. 10191026. Rose, R. R., 1983, Miocene carbonate rocks of Sibolga Basin, northwest Sumatra: Proceedings of the 12th Annual Convention of the Indonesian Petroleum Association, June 1983: Jakarta, Indonesia, Indonesian Petroleum Association, v. 12, p. 107125. Ross, C. P., and D. L. Kinman, 1995, Nonbright-spot AVO: Two examples: Geophysics, v. 60, p. 13981408, doi:10 .1190/1.1443875. Rutherford, S. R., and R. H. Williams, 1989, Amplitudeversus-offset variations in gas sands: Geophysics, v. 54, p. 680688, doi:10.1190/1.1442696. Samuel, M. A., and N. A. Harbury, 1996, The Mentawai fault zone and deformation of the Sumatran forearc in the Nias area: Geological Society (London) Special Publication 106, p. 337351, doi:10.1144/GSL.SP.1996 .106.01.22. Schiefelbein, C., and N. Cameron, 1997, Sumatra/Java oil families, in A. J. Fraser, S. J. Matthews, and R. W. Murphy, eds., Petroleum geology of Southeast Asia: Geological Society (London) Special Publication 126, p. 143146. Schlueter, H. U., C. Gaedicke, H. A. Roeser, B. Schreckenberger, H. Meyer, C. Reichert, Y. Djajadihardja, and A. Prexl, 2002, Tectonic features of the southern Sumatrawestern Java forearc of Indonesia: Tectonics, v. 21, p. 1047, doi:10 .1029/2001TC901048. Sieh, K., and D. Natawidjaja, 2000, Neotectonics of the Sumatran fault, Indonesia: Journal of Geophysical Research, v. 105, p. 28,29528,326, doi:10.1029/2000JB900120. Sloan, E. D., 1998, Clathrate hydrates of natural gases: New York, Marcel Dekker, 705 p. Smith, W. H. F., and D. T. Sandwell, 1997, Global sea-floor topography from satellite altimetry and ship depth soundings: Science, v. 277, p. 19561962, doi:10.1126/science .277.5334.1956. Struss, I., V. Artiles, B. Cramer, and J. Winsemann, 2008, The petroleum system in the Sandino forearc basin, offshore western Nicaragua: Journal of Petroleum Geology, v. 31, p. 221244, doi:10.1111/j.1747-5457.2008 .00418.x. Susilohadi, S., C. Gaedicke, and A. Ehrhardt, 2005, Neogene structures and sedimentation history along the Sunda forearc basins off southwest Sumatra and southwest Java: Marine Geology, v. 219, p. 133154, doi:10.1016 /j.margeo.2005.05.001. Sweeney, J. J., and A. K. Burnham, 1990, Evaluation of a simple model of vitrinite reflectance based on chemical kinetics: AAPG Bulletin, v. 74, p. 15591570. Tissot, B. P., R. Pelet, and P. Ungerer, 1987, Thermal history of sedimentary basins, maturation indices, and kinetics of oil and gas generation: AAPG Bulletin, v. 71, p. 14451466, doi:10.1306/703C80E7-1707-11D7 -8645000102C1865D. Todd, S. P., M. E. Dunn, and A. J. G. Barwise, 1997, Characterizing petroleum charge systems in the Tertiary of SE

Lutz et al.

1615

Asia, in A. J. Fraser, S. J. Matthews, and R. W. Murphy, eds., Petroleum geology of Southeast Asia: Geological Society (London) Special Publication 126, p. 2547. Underdown, R., and J. Redfern, 2008, Petroleum generation and migration in the Ghadames Basin, north Africa: A two-dimensional basin-modeling study: AAPG Bulletin, v. 92, p. 5376, doi:10.1306/08130706032. Ungerer, P., J. Burrus, B. Doligez, P. Y. Chenet, and F. Bessis, 1990, Basin evaluation by integrated two-dimensional modeling of heat transfer, fluid flow, hydrocarbon generation, and migration: AAPG Bulletin, v. 74, p. 309335. Van der Werff, W., 1996, Variation in forearc basin development along the Sunda arc, Indonesia: Journal of Southeast Asian Earth Sciences, v. 14, p. 331349, doi:10 .1016/S0743-9547(96)00068-2. Von Huene, R., and D. W. Scholl, 1991, Observations at convergent margins concerning sediment subduction, subduction erosion, and the growth of continental crust:

Reviews of Geophysics, v. 29, p. 279316, doi:10.1029 /91RG00969. Welte, D. H., and M. A. Ykler, 1981, Petroleum origin and accumulation in basin evolution: A quantitative model: AAPG Bulletin, v. 65, p. 13871396, doi:10.1306 /03B59553-16D1-11D7-8645000102C1865D. Wygrala, B., 1989, Integrated study of an oil field in the southern Po Basin, northern Italy: Ph.D. thesis, University of Cologne, Kln, 217 p. Yalcin, M. N., R. Littke, and R. F. Sachsenhofer, 1997, Thermal history of sedimentary basins, in D. H. Welte, B. Horsfield, and D. R. Baker, eds., Petroleum and basin evolution: Berlin, Springer, p. 71167. Zampetti, V., W. Schlager, J.-H. van Konijnenburg, and A.-J. Everts, 2004, 3-D seismic characterization of submarine landslides on a Miocene carbonate platform: Luconia Province, Malaysia: Journal of Sedimentary Research, v. 74, p. 817830, doi:10.1306/040604740817.

1616

Simeulue Fore-arc