ABSTRACT The Tarakan Basin is one of three major Tertiary age deltaic depocentres located around the margin of Kalimantan. 14 oil and gas fields have been discovered, the largest of which were discovered before world war II. Almost all of the production has occurred on dip oriented arches in a very proximal (mostly non-marine) depositional environment. However, better seismic data suggests that forced regressions have caused deposition of deltaic reservoirs far downdip in present day deep water. This depositional model has resulted in a new exploration fairway for delta front sandstone reservoirs and has increased the hydrocarbon potential of a basin having a productive history dating from 1899. The application of deep water technology pioneered in the Gulf of Mexico will be essential in developing potential hydrocarbon accumulations in this fairway. The Tarakan Basin was initiated simultaneously with the formation of the Sulawesi Sea by rifting of north and west Sulawesi from east Kalimantan between the Middle to Late Eocene and the end of the Early Miocene on east hading enechelon block faults. Accommodation space for an increased volume of deltaic sediments was created during the Middle Miocene to Pliocene period due to a combination of basin subsidence and gravity induced listric faulting. The productive dip oriented arches were formed during the uppermost Pliocene to recent period by transpression on wrench faults crossing the Makassar Strait. INTRODUCTION The Tarakan Basin is generally a passive deltaic margin with a minor wrench tectonic overprint. Oligocene and Miocene sediments thin and eventually onlap onto an Eocene Rift sequence. Magnetic anomalies imply sea-floor spreading with associated NW trending transform faults. The Tarakan basin is divided into four sub-basins (Figure 1): 1. Muara sub-basin, the southernmost depocentre developed exclusively offshore. 2. Berau sub-basin mostly onshore and located in the south. 3. Tarakan sub-basin, mostly offshore but including Bunyu and Tarakan Islands. 4. Tidung sub-basin, the most northerly basin and mainly onshore. The NW-SE trending Muara sub-basin is bounded to the SW by a coast parallel wrench fault zone along the north shore of the Mangkalihat Peninsula. Towards the NE the Muara sub-basin is bounded by another wrench fault zone which causes a basement high upon which the Maratua reefal islands have developed (Figure 1). Seismic data suggests the presence of 5000 metres of Oligocene to Recent carbonate rift and passive margin sediments in the Muara Sub-basin, resting on older volcanics. Little structuration is present in the Middle Miocene to Recent post rift portion of the section. The Eocene rift section probably contains source rocks. Reservoirs in the Muara Sub-basin are predominantly in carbonates. The Berau sub-basin is bounded to the north and south by Pre-Tertiary outcrops. The southern margin, is the igneous Suikerbrood Ridge. Eastwards the basin extends into the Tarakan Sub-basin (Figure 1). The division between the Tarakan and Berau Sub-basins is based upon the pinch-out of the Tarakan Formation of Pliocene age. Several NNW-SSE trending compressional features are present within the Berau Sub-basin. This structuration is related to left lateral movement along wrench zones accommodating the ongoing spreading of the Makassar Strait. The thick clastic fill in the Tarakan sub-basin is an amalgamation of numerous Plio-Pleistocene clastic depocentres located below Bunyu and Tarakan islands and prograding further offshore. The Pliocene thins regionally to the west and south, onlapping Miocene highs and eventually pinching out. The Tidung Sub-basin is separated from the Tarakan Sub-basin by a carbonate platform, as defined by the northern pinchout of the Tarakan Formation. The Tidung sub-basin contains several NW trending tightly folded anticlines. Thrusting also occurs along the coast. In the north it is bounded by the Semporna fault zone exhibiting sinistral transform movement. EXPLORATION AND PRODUCTION Oil was first discovered on Tarakan Island in 1899 and exploration has continued to the present day. In 1967 Pertamina awarded the first PSC contract in offshore Tarakan to Japex. Between 1973 and 1976 the Mengatal, South Pamusian and Selatan fields were discovered by Tesoro. The last field discovered was Mamburungan in 1986. ARCO acquired the onshore Sembakung PSC in 1974 and had three significant discoveries: Sembakung and Bangkudulis oil fields and gas at Sesayap. A portion of the Sembakung PSC is presently held by Pertamina-Teikoku who have drilled four wells in the area. The Bunyu Tapa gas field on Bunyu Island was discovered by Pertamina in 1975. This led to the construction of a methanol plant to utilize the gas. Bunyu Nibung and Barat Fields were discovered in 1974 and 1979 respectively and are essentially satellite fields of the Bunyu Field. In 1985 a new offshore PSC was awarded to a consortium with Sceptre as operator. After the drilling of five wells, four of which contained subcommercial quantities of hydrocarbons, the PSC was relinquished in 1993. So far, only one well has been drilled in a down dip position, the Vanda-1, in 348 m of water in a distal position of the central part of the offshore. Petrocorp's Maratua and Karang Besar PSCs were awarded in 1990 covering the on- and offshore Berau and Muara sub-basins, south of Sceptre's Bunyu PSC. These two blocks were subsequently awarded to Maersk in 1995. Shell's Sebawang PSC was awarded in 1995 covering the offshore Tarakan Sub-basin. Exploration in the Tarakan Basin has resulted so far in the discovery of 14 oil and gas fields. Cumulative production from these fields is approximately 320 MMBO in a basin of 7,000 km², with ultimate recovery from the proven fields estimated at 500 MMBO. Total gas produced to date from 13 wells has been 81 BCGF. Nearly 86% of the oil production to date has come from two fields: the Pamusian Field on Tarakan Island and Bunyu Field on Bunyu Island. Most of the remaining production comes from a series of very small fields located in individual fault blocks on Tarakan and Bunyu Islands. TECTONIC HISTORY Middle Eocene-Early Miocene The Tarakan Basin was initiated simultaneously with the formation of the Sulawesi Sea by rifting of north and west Sulawesi from east Kalimantan (Hamilton, 1979). Extension and subsidence began during the Middle to Late Eocene and had stopped by the end of the Early Miocene (Burollet and Salle, 1981; Situmorang, 1982, 1983; Figure 2). This extensional tectonic phase opened the Tarakan Basin eastward as indicated by the predominance of enechelon block faulting hading to the east. The opening of the Sulawesi Sea has been interpreted as being related to the same tectonic episode responsible for the opening of the South China Sea (Rangin, 1991). Middle Miocene-Pliocene The Tarakan Basin was more tectonically stable from the Middle Miocene until the Pliocene with deltaic sedimentation (Figure 2) from the west through several drainage systems. During this phase the combination of basin subsidence and gravity induced listric faulting (growth faults) created accommodation space for an increased volume of deltaic sediments. The growth faults are the most common structure in the Tarakan Sub-basin. The faulting trends north-south in the southern portion of the basin, and changes to a more northeasterly orientation north of Bunyu Island. This change in orientation corresponds to the edge of the northerly Plio-Pleistocene depocentre. Here the faults are more laterally continuous and exhibit maximum displacement. A regional seismic dip line (Figure 3) shows the typical expression of the extensional regime and the north-south fault system. Pliocene-Recent The latest tectonic phase is a reactivation of transform movement along the wrench faults crossing the Makassar Strait beginning in the uppermost Pliocene and continuing to present day (Figure 2 & 3). Three major dextral wrench fault zones, and several smaller scale wrench faults are found in the Tarakan Basin. The Semporna fault zone is the most northern wrench zone. It separates the volcanics of the Semporna Peninsula from the Neogene sediments on Sebatik Island (Figure 1). Further onshore it becomes a boundary between Cretaceous sediments in the north and Neogene sediments in the south. The Maratua Fault zone occurs as a complex transpressional zone. It forms the boundary between the Tarakan and Muara Sub-basins (Figure 1). The third major wrench fault forms the southern boundary of Muara Sub-basin, along the north shore of the Mangkalihat Peninsula. It appears as the extension of the Palu-Koro fault in Sulawesi (Figure 4). Transpression during this period resulted in the formation of major dip oriented arches. Vitrinite reflectance, porosity and seismic data suggest 1000-1500 m of structural inversion occurred during arch formation. Five major arches dominate the western area, informally named from north to south, the Sebatik, Ahus, Bunyu, Tarakan and Latih arches (Figure 1 and 5). They are broad SE plunging flexures formed by NE-SW transpression and are oriented roughly north-northwest - south-southeast, changing to northwest - southeast further northwards. The age of compression appears to get younger northwards and is coeval with, but decoupled from, the major extensional system to the east. The intensity of folding also increases northwards where the broad arches give way to tight folds in the onshore area. The Latih arch and minor anticlines develop in the southern part of the Tarakan Basin. They are NW-SE oriented. These anticlines have been drilled and had oil shows and gas flows from thin turbidite sandstones. OIL PLAY CONCEPTS The existing oil and gas fields are a combination of structural and stratigraphic trapping along the axis of the Tarakan and Bunyu arches. Hydrocarbon charge has been focused from the basin up the plunging arches into Plio-Pleistocene sandstone reservoirs deposited in a very proximal/delta plain depositional setting (Figure 6). Unfortunately, several oil and gas discoveries in this setting have poor lateral reservoir continuity, poor recovery efficiency and contain hydrocarbon volumes that are often too small to be economically produced. The two largest fields in the basin, Tarakan (215 MMBO) and Bunyu (140 MMBO and 350 BCF) are faulted anticlines. All of the other accumulations are substantially dependent upon fault closure and contain less than 25 MMBO recoverable each. In such a proximal setting the combination of numerous thin sands in the section, high net-to-gross and the very high likelihood of sand / sand juxtapositions across faults have contributed to the high failure rate. As the southeasterly plunging arches become progressively younger from south to north hydrocarbon charge may have occurred before trap formation in the northernmost part of the Tarakan Basin. Little exploration has occurred in areas further offshore which are characterized by growth faulting. Only the Vanda-1 well has been drilled by Sceptre in 1990 (Figure 6). Prospective fairways have been identified in the southeastern offshore in water depths greater than 200 m. One is the outboard extension of the crestal collapse/graben trend (Figure 4). The offshore fairways rely upon deltaic progradation to the east and forced regressions during drops in sea level to get deposition of reservoir far down dip of existing well control (Figure 3). Traps could exist containing thick delta front reservoirs in the hanging walls of large scale listric growth faults. In this setting, reservoirs and seals could be thicker and more continuous than those in proximal settings. A Miocene clastic reservoir play has been proven in the onshore Berau and Tidung Sub-Basins with the establishment of production at Sembakung Field (25 MMBO) in 1976 by Arco (Figure 6). This play type is located in a half graben tectonic province (Pertamina DAK-SSG, 1993). Traps are found in rollover anticlines within down thrown blocks of major growth faults. Play objectives of this tectonic province are Santul, Tabul, and Meliat Formation with charge from carbonaceous source rocks of the Meliat Formation. The Bangkudulis Field (2.4 MMBO), which is located in the west of Bangkudulis Island has a similar play type as the Sembakung Field (Pertamina DAK-SSG, 1993). Oligo-Miocene carbonate reefs of the Tabalar Formation comprise the most prospective play in the Muara Sub-basin (Figure 6). No commercial hydrocarbon discoveries have been made, however, oil and gas shows were reported in exploration wells which penetrated reefal built-ups at Karang Besar-1, Segitiga-1 and Tabalar-1. This sub-basin is structurally less complex being characterized by uniform subsidence between basement ridges and persistent carbonate deposition on basement highs and platform areas. Trap types include pinnacle reefs, fore-reef talus deposits, and shelf edge reefs. Onlap and pinchout stratigraphic traps of Eocene sandstones are possible but poorly defined on seismic data (Pertamina-BEICIP, 1992). The primary risks of plays in the Muara sub-basin include charge, because the basin is not deeply buried, and trap integrity because the build-ups are large along with the potential for thief zones onlapping the build-ups. DEEPWATER TECHNOLOGY APPLICATION Shallow water developments are expected to be strait forward, utilizing technology readily available in Indonesia. However, most of the prospectivity resides downdip of existing well control in increasing water depth. In moderately deep water (± 200 - 400 metres) oil developments would utilize a number of wellhead platforms located above the field and would be tied back to processing platforms located west of the fields in shallower water. Beyond 400 metres, conventional bottom-supported steel jackets, or concrete platforms, are rarely technically or economically viable. This is also the practical safe limit of saturation diving techniques for underwater intervention and maintenance of seabed facilities. Deepwater development is expensive, due to the high costs of the facilities and the specialized equipment required to install them. Thus designing wells to maximize productivity, by using horizontal or multilateral completions through the reservoir, is an essential aspect of any deepwater field development. CONCLUSIONS Four tectonic stages are recognized in the Tarakan Basin. The tectonic evolution progressed from a stable cratonic stage prior to the Middle Eocene to an extensional stage. This resulted in the opening of the Tarakan Basin, and was followed by uplift of the hinterland in the interior of Kalimantan beginning in the Middle Miocene. The emerging mountains are the provenance of large volumes of clastics into the Tarakan Basin. The last tectonic stage is indicated by the reactivation of the transform movement along the wrench faults across the Makassar Strait resulting in the formation of several diporiented arches. Oil and gas accumulated in structural and stratigraphic traps in the up-dip portions of the arches. Potentially productive fairways have been identified in a graben trend in the southeastern offshore portion of the basin, characterized by growth faults. Oligo-Miocene carbonate reef comprise the most prospective play in the southern part of the basin. However, no commercial hydrocarbon discoveries have been made. Deep water technology is required for exploring the down-dip growth faults fairways. ACKNOWLEDGMENTS We would like to thank Pertamina-BPPKA and Shell Companies in Indonesia for permission to publish this paper. The manuscript has been improved by comments from Frank Musgrove, Mobil Oil Indonesia. REFERENCES Achmad, Z. & Samuel, L., 1984, Stratigraphy and Depositional Cycles in the NE. Kalimantan Basin, Proceeding of the Indonesian Petroleum Association, 13/1, 109-120. Burollet, P. F. & Salle, C., 1981, Histoire Geologique de I'ile de Sumba (Indonesie), Bull. Soc. Geol. France, v. 7, no. 3, 573-580. Haddad, G. A. & Vail, P. R., 1992, Pliocene and Quaternary Sealevel Change: Linking Eustasy (from Sequence Stratigraphy) to the Oxygen Isotope Record of Global Ice Volume, In: Dijon et al. (eds) Conf. Seq. Strat. European Basins. Haq,B. U., Hardenbol, J. & Vail, P. R., 1988, Mesozoic and Ceonozoic Chronostratigraphy and Eustatic Cycles, In: Wilgus et al. (eds) Sea-Level changes: An Integrated Approach, SEPM Special Publication No. 42, 71-108. Hamilton, W., 1979, Tectonic of Indonesian Region, US. Geological Survey, Professional Paper 1078. Heriyanto, N., Satoto, W., & Sardjono, S., 1992, An Overview of Hydrocarbon Maturity and Its Migration Aspects in Bunyu Island, Tarakan Basin, Proceeding of the Indonesian Petroleum Association, 21/1, 1-22. Hidayat, S., Amiruddin, Nas, D. S., 1992, Geologi Lembar Tarakan dan Sebatik, Kalimantan Timur, Proyek Pemetaan Geologi dan Geofisika, Bidang Pemetaan Geologi, Pusat Penelitian dan Pengembangan Geologi. Nandang, H. & Wahyudin, M., 1994, Reflectance Gradient and Shale Compaction, Their Relationship to Basin Configuration During Early Neogen: A NE Kalimantan Basin Reassessment, Pertemuan Ilmiah Tahunan IAGI ke 23 Proceeding, 927-942. Netherwood, R. & Wight, A., 1993, Structurally-Controlled, Linear Reefs in a Pliocene Delta-Front Setting, Tarakan Basin, Northeast Kalimantan, In: Siemers et al. (eds) , Carbonate Rocks and Reservoirs of Indonesia, IPA Core Workshop, 3/1 - 3/16. Pertamina - BEICIP, 1992, Global Geodynamics, Basin Classifiacation and Exploration Play Tyoes in Indonesia, Pertamina. Pertamina DAK-SSG (Sembakung Study Group), 1993, Studi Regional Terpadu Blok Sembakung Cekungan Tarakan Kalimantan Timur Bagian Utara, Eksplorasi Pertamina DAK. Rangin, C., 1991, The Philippine Mobile Belt: a Complex Plate Boundary, Journal of Southeast Asian Earth Sciences, v. 6, no. 3/4, 209-220. Situmorang, B., 1982, The Formation of the Makassar Basin as Determined from Subsidence Curve, Proceeding of the Indonesian Petroleum Association, 11/1, 83-107. Situmorang, B., 1983, Formation, Evolution, and Hydrocarbon Prospect of the Makassar Basin, Indonesia, Transaction 3rd Circum Pacific Conference, 227-232. Situmorang, R.L., & Burhan, G., 1992, Peta Geologi Lembar Tanjung Redeb, Kalimantan Timur, Geological Research and Development Centre. Tongkul, F., 1991, Tectonic Evolution of Sabah, Malaysia, Journal of Southeast Asian Earth Sciences, v. 6, no. 3/4, 395-405.