IPA11-G-009

PROCEEDINGS, INDONESIAN PETROLEUM ASSOCIATION Thirty-Fifth Annual Convention & Exhibition, May 2011

SUMBA AREA: DETACHED SUNDALAND TERRANE AND PETROLEUM IMPLICATIONS

Awang Harun Satyana*

Margaretha E.M. Purwaningsih** ABSTRACT Sumba Island is a terrane situated in the forearc setting of the Neogene-Quaternary Sunda-Banda volcanic arc. Sumba is considered a microcontinent and its origin has been a matter of debate. There are two main competing hypotheses: (1) provenance from NW Australia or (2) provenance from SE/Eastern Sundaland. We studied various considerations from previous authors and present a new interpretation and synthesis based on multidisciplinary methods including: stratigraphic succession, geochronology-geochemistry of magmatic rocks, paleomagnetism, isotope geology, and Eocene large foraminifera. The Paleogene stratigraphic succession of Sumba is similar to that of Southern Sulawesi. Extruded magmas display the characteristics of typical island arc environment at the margin of Sundaland. Paleomagnetic data of Sumba show the location of SE/Eastern Sundaland in the Late Cretaceous and has occupied its present position since the Early Miocene. Pb-Nd isotope characteristics of rocks from Sumba showisotopic signatures and affinities with rocks of Sundaland. Sumba contains typical Eocene low-latitude Sundaland fauna of Assilina, Pellatispira, and Biplanispira and no Eocene high-latitude Australian fauna of Lacazinella. We present new consideration on the detachment of Sumba from SE/Eastern Sundaland and its emplacement as continental sliver at its present position through escape tectonics. Petroleum accumulations occur and are produced from Eastern Indonesian foreland basins related to collision of 'Australoid' microcontinents like the Banggai and Bintuni Basins. However, this collision-related petroleum play does not typify the Sumba microcontinent due to the absence of Australian sediments and no collision in its history. Rifted structures, proven to be prolific in Western ∗ BPMIGAS ** ConocoPhillips Indonesia

Western Indonesian basins define the Sumba petroleum potential. Geology of Sumba Island, marine seismic and the presence ofpetroleum seeps/slicks show positive indications of a petroleum system in Sumba area. INTRODUCTION The last publication in IPA Proceedings discussing the Sumba Island was a paper written 30 years ago by Burollet and Salle (1981). In the last 30 years there have been studies on the geology and tectonics of the island published in various geologic and tectonic journals or other conference proceedings. This paper examines these studies to collect and expand the knowledge on tectonics and petroleum geology of Sumba area. This paper deals with tectonics and petroleum implications of Sumba area. The area of investigation in this paper covers the onshore and offshore Sumba Island and parts of the Savu/Sawu Basin. Sumba Island is famous in Indonesian geologic literature because of its tectonically enigmatic position. The island has been known as a detached terrane (microcontinent), a geologic block transported from its provenance and tectonically emplaced into its current position. There have been debates on the island's provenance/origin. Previous authors examined the origin of the island based on various methods discussed in detail in Results section. This paper firstly will address these previous debates, examine and synthesize them into the new interpretation covering the origin and emplacement of the Sumba microcontinent using more comprehensive methods that have never been available to previous researchers. Secondly, the paper will discuss petroleum implications of Sumba area given the results of this study. There are several detached terranes in Eastern Indonesia that are hydrocarbon productive. We will consider Sumba area similarities and differences to other microcontinents

and present petroleum play types unique to Sumba area. METHODS This paper comprises two parts: (1) discussion of the geology and tectonics of Sumba area (Sumba Island mainly) and (2) discussion on the petroleum implications of Sumba area. In the first step (1), we collected published literature from various journals, examined them, and put them in the new interpretation and synthesis of the origin and emplacement of the Sumba Island. A field survey to examine the geology of Sumba Island was conducted in August 2010. In step two (2), we referred to other detached terranes, especially in Eastern Indonesia, that produce hydrocarbons, seeking analogies with Sumba area. Special charactistics of Sumba area compared to other microcontinents were reviewed with attention to petroleum play types of Sumba area. Seismic lines offshore Sumba were examined for different play types. DISCUSSION & RESULTS Tectonic Position of the Sumba Island Sumba Island, a part of Lesser Sunda (Nusa Tenggara) islands in the southern part of Central Indonesia is an island located to the south of Quaternary volcanic islands comprising Sunda and Banda Arcs (Figure 1). The island is located in a forearc setting relative to the Sunda-Banda volcanic arcs comprising mainly islands of Bali-Lombok-Sumbawa-Flores-Alor-Wetar. Sumba Island is presently non-volcanic forming one belt of nonvolcanic arcs of Banda Arc with Timor, Tanimbar and Seram Islands. Banda Arc comprises two parallel arcs of inner volcanic islands comprising Flores-Alor-Wetar-Romang-Damar-Teun-Nila-Serua-Rozengain-Banda Islands and outer nonvolcanic islands of Sumba-Savu-Roti-Timor- Moa- Sermata- Babar -Tanimbar- Kei- Watubela- Gorong- Seram- Buru Islands. Sumba Island is tectonically significant because it is located at the border of the subduction zone to the west where the oceanic crust of the Indian Ocean subducts beneath the Sunda Arc and the collision zone to the east where the continental crust of the Australian Continent subducts beneath/underthrusts the outer Banda Arc from Timor to Seram (Figure 2). Sumba lies obliquely between two fore-arc basins, the Lombok Basin to the west and the triangular Savu Basin to the east. Bathymetrically,

Sumba stands out as a ridge that obliquely separates the Savu forearc basin (>3000 m depth) in the east and the Lombok forearc basin (>4000 m depth) in the west. Based on tectonic studies, complemented by paleomagnetism and geochemistry, several researchers considered Sumba to be a microcontinent or continental fragment/sliver (Hamilton, 1979; Chamalaun and Sunata, 1982; Wensink, 1994, 1997; Vroon et al., 1996; Soeria-Atmadja et al., 1998) detached from its provenance and transported to its present position as a terrane that is alien or exotic to its surrounding areas. The island of Sumba with a Bouguer gravity anomaly of +160 to +200 mgal is underlain by a continental type of crust with a thickness of 24 km (Chamalaun et al., 1981). The exact outline of the Sumba fragment is not fully known. The island of Sumba is some 220 kilometers long and about 60 kilometers wide. Seismic profiles show that the Sumba position is unique and its features do not extend more than 100 km from the island (Burollet and Salle, 1981). Towards the west in the Lombok Basin, Sumba extends below sea level for approximately 50 km until it is cut off by a NE-SW trending fault system. In the Savu Fore-Arc Basin east of Sumba, seismic reflections reveal a submarine ridge, the North Savu Ridge. The ridge begins at Sumba's easternmost end and runs to the ESE towards the island of Savu. Wensink (1994) estimated the dimension of the Sumba crustal fragment to be 400 km long and 200 km wide. Geology of Sumba Island Geology of Sumba has been investigated since the end of 19th century by Dutch geologists. Results of early geological investigations on Sumba Island before World War II were synthesized by van Bemmelen (1949). The first investigation was conducted by Verbeek in 1899. Several important works on the geology of the island were (all references can be found in van Bemmelen, 1949) from: Verbeek (1908), Witkamp (1912, 1913), Hühnerwadel et al. (1924, 1925) and Kinser and Dieperink (1940). The last authors published the geological map of the island at scale 1:200,000. After World War II, the geology of Sumba Island was studied and mapped by the Geological Survey of Indonesia and many other workers/scientists from various institutions. Several important works during this period were from Effendi and Apandi (1981), Burollet and Salle (1981), Chamalaun et al.

(1981), von der Borch et al. (1983), Fortuin et al. (1992), Fortuin et al. (1994), Effendi and Apandi (1994), Abdullah (1994), Wensink (1994), Wensink and van Bergen (1994), van der Werff et al. (1994), Vroon et al. (1996), Fortuin et al. (1997), Rampnoux et al. (1997), Soeria-Atmadja et al. (1998), Abdullah et al. (2000), and Abdullah (2010). A geological sketch map of Sumba is shown in Figure 3. The stratigraphy of the island begins with slightly to unmetamorphosed sediments of Mesozoic age, unconformably overlain by considerably less deformed Tertiary and Quaternary deposits; the total thickness of which is more than 1000 m (van Bemmelen, 1949). The Quaternary coral reef terraces, which cap the seaward edge of the Neogene Sumba Formation, are almost continuously exposed along the western, northern and eastern coasts of Sumba. Abdullah (1994) and Abdullah et al. (2000) distinguished four sedimentary cycles in Sumba (Figure 4). The first cycle (Late Cretaceous-Paleocene) is represented by marine turbidites of the Lasipu Formation. It was accompanied by two major calc-alkaline magmatic episodes, the Santonian-Campanian episode (86-77 Ma) and the Maastrichtian-Thanetian episode (71-56 Ma). The second cycle (Paleogene) was marked by volcaniclastic and neritic sedimentation accompanied by the third magmatic episode of Lutetian-Rupelian age (42-31 Ma) (Eocene-Oligocene, Paumbapa Formation). The following Neogene sedimentary cycle was a period of widespread transgression, characterized by rapid sedimentation in a deep sea environment (Kananggar/Sumba Formation) (Fortuin et al., 1992, 1994, 1997). This syn-tectonic turbiditic sedimentation containing reworked volcanic material also has been observed in neighboring Lombok and Savu basins. The fourth cycle (Quaternary) was marked by the uplift of terraces, beginning 1 Ma ago. Pictures of these rocks from a recent field survey can be seen in Figure 5. The Mesozoic sediments are typically carbonaceous siltstones with volcanogenic mudstones, sometimes showing signs of low-grade metamorphism in prehnite-pumpellyite meta-greywacke facies (Chamalaun et al. 1981), interbedded with sandstones, conglomerates, limestones and volcaniclastic debris. They are crosscut by Late Cretaceous intrusions which range in composition from microgabbro to quartz-diorite, and also by granodioritic and calc-alkaline dykes of Paleogene

age. The sediments show large scale slump structures and significant fracturing. These sediments constitute the Lasipu Formation. Microfossil assemblages in some samples indicate Coniacian to Early Campanian ages (mid to Late Cretaceous) (Burollet and Salle, 1981). The detrital material suggest either a continental origin, or an island arc environment; essentially a Mesozoic submarine fan with shallow-water deposits (Von der Borch et al., 1983) or an open marine bathyal environment (Burollet and Salle , 1981). During the Paleogene, Sumba was a part of a magmatic arc (Abdullah et al., 2000) characterized by calc-alkaline volcanic rock series (Western Sumba) and shallow marine fossiliferous limestones and sandstones of the Paumbapa Formation and have an Eocene and Oligocene age (Effendi and Apandi 1981). The corresponding deposits include tuffs, ignimbrites, greywackes, intercalations of foraminiferal limestones, marls, micro-conglomerates and claystones. These rocks unconformably overlie Mesozoic rocks and are in turn unconformably overlain by the Neogene Series. In the early Miocene there is another period of volcanic activity (Wensink, 1994). This volcanism of the Jawila Formation, is restricted to West Sumba. Large areas are covered with tuff, tuff-agglomerates, tuff-sandstones and lahars while rather fresh basalts and basaltic andesites occur as well. There are small exposures of the Middle Miocene Pamalar Formation with claystone and limestone, the latter both in lagoonal and in reef facies. An enormous mass of sediments with a thickness of at least 800 m covers large areas on Sumba. These sediments, which slightly unconformably overlie older rocks, belong to the Sumba Formation and have a late Miocene to early Pliocene age (Fortuin et al. 1992). The deposits show a general shallowing from east to west. The eastern facies of the Sumba Formation, often called Kananggar Formation, comprises basal conglomerates, overlain by volcanoclastic turbidites, sands, gravels and intercalated white, pelagic chalks. In East Sumba the formation contains many slumps. The western facies is mainly shallow marine; here, deposits of the Waikabubak Formation are found with carbonate platform sediments of reef and lagoonal origins. The Quaternary is represented by coral-reef terraces which fringe the island on the west, north and east coasts. These terraces comprise sandstones, conglomerates, marls and prominent reef limestones.

In general, the pre-Tertiary sediments of the Lasipu Formation are mildly deformed, exhibiting broad, open folds. The pre-Tertiary basement of Sumba reveals faulting with rifted blocks (Wensink, 1994). Synsedimentary tectonism with normal faulting and large-scale slumping occurred during the Neogene. A section from south to north on Sumba shows all formations dipping to the north. Since the Pliocene the uplift of Sumba amounts to approximately 3 km. The elongated dome of Sumba represents an uplifted area of the fore-arc basin. The Neogene sediments of Sumba are gently folded and or warped, and broken by faults. Collision of a promontory of Australian continental lithosphere with Sumba at ca. 8 Ma contributed to the uplift of Sumba and created many structural features on the Australian side of the plate boundary and partitioned the lithosphere into structural domains (Keep et al., 2003). The outline of the geology of Sumba as given above shows that both the stratigraphy and the tectonics of the island are rather simple. Contrasting with that, the geology of Timor, another island of non-volcanic outer Banda Arc, is very complicated, both in stratigraphy and in tectonics (Hamilton, 1979). This is one of the reasons for relating or not relating Sumba to Timor. Origin of the Sumba Island: Australia, Timor, Tethys Sumba has been considered as a microcontinent and its origin has been a matter of debate. There are two dominant competing hypotheses: (1) provenance from the margin of NW Australia and (2) provenance from the margin of eastern Sundaland. Other considerations are (3) Sumba was a fragmented Tethys isolated continent/microcontinent and (4) Sumba was part of Timor which escaped into its present position by the opening of Savu Basin. (1) Sumba was originally a part of the Australian Continent which was detached when the Wharton basin was formed, drifted northwards and was subsequently trapped behind the eastern Java Trench (Audley-Charles, 1975; Norvick, 1979; Otofuji et al., 1981; Pigram and Panggabean, 1984; Hartono and Tjokrosapoetro, 1984; Nishimura and Suparka, 1986); (2) Sumba was once part of Sundaland which drifted southwards during the opening of the marginal seas in the eastern margin of Sundaland (at present, many authors are in favor of a northern provenance/Sundaland/SE Asia of the

Sumba fragment such as: Hamilton, 1979; Burollet and Salle, 1981; von der Borch et al., 1983; Rangin et al., 1990; Wensink, 1994; Abdullah, 1994; van der Werff et al., 1994; Wensink and van Bergen, 1995; Vroon et al., 1996; Fortuin et al., 1997; Soeria-Atmadja et al., 1998; Abdullah et al., 2000; Rutherford et al., 2001; Satyana, 2003; Abdullah, 2010; Rigg and Hall, 2010); (3) Sumba was either a microcontinent or part of a larger continent within the Tethys, which was later fragmented (Chamalaun and Sunata, 1982); and (4) Sumba was part of Timor and escaped to its present position after the collision of Timor-Australian continent and by the opening of the Savu Basin (Audley-Charles, 1985; Djumhana and Rumlan, 1992). (1) Australian origin - Sumba was part of the Australian continent, where it occupied a position near the Scott Plateau. Along a SW-NE fracture zone at the eastern side of the Wharton Basin; a fragment broke away. It is well known that along Australia's west coast, rifting began some 160 Ma ago, resulting in the opening of the Wharton Basin (Falvey and Mutter, 1981). Sumba became detached and rotated clockwise (Audley-Charles 1975). In a reconstruction of Eastern Indonesia, Norvick (1979) located Sumba near Australia to the west of a S-N trending fault that he called "Sumba Fracture". Otofuji et al. (1981) compared Jurassic paleomagnetic directions of Sumba and Timor. Comparison with the Permian paleomagnetic direction of Timor indicates that Sumba was subjected to a clockwise rotation through 79.4° relative to Timor since the Jurassic. When Sumba is restored by a counter-clockwise rotation, the Sumba pole is gradually approaching the Jurassic pole of Australia as well as Timor pole. This implies that, at least until the Jurassic, Sumba and Timor were situated at the Australian continental margin. Based on a seemingly confirmed Jurassic rift drift event in northwestern Australia, Hartono and Tjokrosapoetro (1983) believed the origin of Sumba to be from northwestern Australia. Probably the pre-rift drift position of Sumba was at the southeastern most part of northwest Australia compared with pre-breakup positions of the other continental fragments now occurring in eastern Indonesia. It is also noted that more towards the southeast the breakup becomes younger. Nishimura and Suparka (1986) interpreted the paleomagnetic evidence that Sumba was attached to the Australian continental margin up to Jurassic time and rotated 44° clockwise after separation. The main objection to Australian provenance for Sumba is that the pre-Tertiary and the Paleogene

stratigraphy of Sumba differs from that of NW Australian shelf. There are no Paleozoic and Mesozoic sediments in Sumba area as are found in NW Australia. No volcanic, volcaniclastic or magmatic rocks have been discovered in NW Australia from the Late Cretaceous and Paleogene as are found in Sumba area. Otofuji et al. (1981) analyzed 32 specimens from shales in West Sumba collected not far from the Tanadaro Granodiorite. They assigned a Jurassic age to these shales quoting van Bemmelen (1949); however, these sediments, which belong to the Lasipu Formation, are now assigned a Late Cretaceous age (von der Borch et al. 1983). Otofuji et al. (1981) did not give a straightforward answer to the question whether Sumba had an Australian or a Sundaland origin, although they have a slight preference for a southern provenance. Wensink (1994) found difficulty relating Sumba to Australian provenance due to the significant presence of granodiorite intrusives and related rocks which have an age of approximately 64 Ma, as well as the Paleocene volcanics of the Massu Formation. The rifting along Australia's coasts took place in the Jurassic and the early Cretaceous, thus the referred igneous rocks of Sumba are too young for correlation with the Australian rifting. (2) Timor origin - The stratigraphy of Sumba may be correlated with the Cretaceous to Miocene part of the Timor allochthon (Audley-Charles, 1985). The sedimentary and eruptive rock succession in Sumba shows remarkable similarities to the allochthonous Palelo, Wiluba and Cablac deposits of Timor. On both islands the Cretaceous members of these sequences are regarded as characteristic of fore-arc deposits built on thin continental crust. Djumhana and Rumlan (1992) suggested that Sumba was originally a part of Timor, with the present southwest part of Sumba conjoined to northwest Timor during Mesozoic-Oligocene time (Figure 6). Continental drift of Australia to the north during early-middle Miocene time is believed to have caused initial movement of Sumba by transcurrent faulting that reached a climax during middle-late Miocene time. Sumba moved to the southwest by escape tectonism, separated from Timor and rotated 60° clockwise. The Savu Basin developed as an extensional basin from early Pliocene time behind the escaping Sumba fragment. The islands of Sumba and Timor lie only some 400 km apart. Both islands are situated south of the Sunda-Banda volcanic arc and north of the

deformation front in the Java Trench and the Timor Trough. The outline of the geology of Sumba shows that both the stratigraphy and the tectonics of the island are rather simple. However, the geology of Timor is very complicated, both in stratigraphy and in tectonics (Wensink, 1994). The main objection to Timor provenance for Sumba is similar to that of relating Sumba to NW Australia provenance. The pre-Tertiary and the Paleogene stratigraphy of Sumba are different from that of NW Australian shelf. No Paleozoic or Mesozoic sediments are found in Sumba area as are found in Timor. There are no volcanic, volcaniclastic or magmatic rocks in Timor for the Late Cretaceous and Paleogene as are present in Sumba area. A comparison of the stratigraphic sequences of Sumba with those of Timor shows that there are some resemblances between the Sumba rocks and the North Timor Palelo Series (Audley-Charles 1985). In Northern Timor there are volcanics of the Metan Formation and nummulitic limestones of Eocene age which are overlain by Oligocene reefal limestones. These rock sequences of Northern Timor are considered to have a northern provenance (Audley-Charles' 'Banda Allochthon') with a Southeast Asian origin. The rocks may have been incorporated in the early Banda Arc and may have collided with the Australian margin approximately 3 Ma ago. The opening of Savu Basin by escape tectonism of Sumba may not be the mechanism for the opening of Savu Basin since Sumba may not be part of Timor, therefore it never escaped from Timor. Rigg and Hall (2010) suggested that the basin is underlain by continental crust and its normal faulting in the middle Miocene and rapid subsidence to several kilometers was driven by subduction rollback. (3) Tethys microcontinent origin - Sumba was either an isolated microcontinent or part of a larger continent within Tethys that was later fragmented (Chamalaun and Sunata, 1982). However, the geology of the island shows that it is likely that there were relationships with other continental units. The composition and structure of both the Lasipu and the Sumba sediments indicate such relationships, meaning that Sumba did not originate from an isolated microcontinent. Chamalaun and Sunata (1982) concluded that the combination of the results of Otofuji et al. (1981) with their own data did not provide a straightforward answer to the question whether Sumba had an Australian or a Sundaland origin (Wensink, 1994), although they had a slight preference for a southern provenance.

The Sumba fragment seems to have occupied its present position at least since the early Miocene. Origin of the Sumba Island: Sundaland Many authors are in favor of a northern provenance Sundaland/SE Asia for the Sumba fragment/island/terrane/block. Various methods have been used to identify the provenance for the Sumba area, including: stratigraphic succession (Burollet and Salle, 1981; Simandjuntak, 1993; Abdullah, 1994); geochronology-geochemistry of magmatic rocks (Abdullah, 1994; Abdullah et al., 2000; Abdullah, 2010), paleomagnetism (Wensink, 1994; Wensink and van Bergen, 1995), isotope geology (Vroon et al., 1996) and Eocene larger foraminifera (Lunt, 2003). The theory that the Sumba micro-continent detached from SE Sundaland has been considered since Hamilton’s work (1979). The Cretaceous-Paleogene geology of the Sumba Platform correlates with the southern arm of Sulawesi and SE Kalimantan (Simandjuntak, 1993). Abdullah (1994) noted similarities in the Paleogene sedimentary facies and magmatism on Sumba and Sulawesi and concluded that the island was originally part of a Paleogene volcanic arc that was situated near western Sulawesi from Late Cretaceous time to the Paleogene. Paleomagnetic data of Sumba show the location of eastern Sundaland in the Late Cretaceous and has occupied its present position since the Early Miocene (Wensink, 1994). Potential Pb-Nd isotope characteristics of rocks from Sumba and its expected provenances show corresponding isotopic signatures and affinities with Sundaland (Vroon et al., 1996). Sumba contains typical Eocene low-latitude Sundaland fauna of Assilina, Pellatispira, and Biplanispira and no Eocene high-latitude Australian fauna of Lacazinella (Lunt, 2003). Burollet and Salle (1981) provided the first comprehensive geological study of Sumba Island that explained its geodynamic position. Based on Sumba stratigraphic succession, magmatic rocks, and structural episodes; Burollet and Salle (1981) concluded that in contrast to Timor, whose framework belongs to the Australian foreland, Sumba represents a borderland of the Sunda shelf. The first tectonic phase of Sumba at the end of the Cretaceous that was associated with Lower Paleocene (dated 59-66 Ma) calk-alkali trachyte with hypersthene and calk-alkali syenite, may be compared to one of the main tectonic phases known in East Kalimantan and Sulawesi, showing more or

less craton characteristics at the beginning of Paleocene. At the beginning of the Upper Eocene, andesitic and calc-alkali trachyandesitic lavas that were persistent though the Palaeogene, equivalent to a large extent to the submarine arc of the Sunda islands. The island drowned in the inter-arc trench during the Miocene and uplifted in the Plio-Quaternary as a result of the subduction of the front of the Australian shelf. Simandjuntak (1993), based on regional stratigraphic correlation, argued that the Cretaceous-Paleogene geology of Sumba Island is quite similar to the southern arm of Sulawesi and in some aspects to the southeastern part of Kalimantan (both areas are located in SE Sundaland) (Figure 7). Lithological association of flysch slope sediments containing Globotruncana sp of Late Cretaceous age (Praikajelu Formation) and the associated basaltic, andesitic and rhyolitic volcanics of the Massu Formation on Sumba Island is similar to sequences in southern arm and Central Sulawesi (Latimojong Formation and Langi Volcanics) and in Southeast Kalimantan (Pitap Formation). Late Cretaceous-Paleogene intrusives of syenite, diorite, granodiorite and granite occurring in those areas are similar to the Early Paleocene intrusives on Sumba Island. The Late Cretaceous is considered to be a time of thermal doming and plutonism, initiating cratonisation of the Sunda Shield. The Paleogene carbonate platform and greywackes of Sumba are correlative to SE Kalimantan and the southern arm of Sulawesi (Berai and Tonasa carbonates, respectively). In Late Paleogene time, there was a non-depositional period on Sumba Island while in SE Kalimantan sedimentation continued into the Neogene. During Oligo-Miocene time, on the southern arm of Sulawesi, deposition was dominated by a carbonate platform with a break in Middle Miocene time. Based on this, Simandjuntak (1993) considered that the detachment of Sumba Island from Sulawesi took place in Paleogene to Neogene time. Abdullah (1994), Soeria-Atmadja et al. (1998), Abdullah et al. (2000) and Abdullah (2010) studied in detail the stratigraphic succession and magmatic/volcanic rocks of Sumba and its expected provenance in SE Sundaland (Figures 3, 4). A set of 24 magmatic rock samples representing granitoid intrusions, lava flows and subvolcanic dykes of mafic to intermediate composition from various outcrops within the investigated area were selected for 40K-40Ar dating as well as chemical analyses (major and trace elements). Numerous other magmatic rock samples were studied

petrographically (Abdullah et al., 2000). Three periods of magmatic activity were recognized by Abdullah (1994) on the basis of most of these data, at ca 86-77 Ma (Santonian-Campanian), 71-56 Ma (Maastrichtian-Thanetian) and 42-31 Ma (Lutetian-Rupelian). Erupted magmas display the characteristics of a predominantly calc-alkaline (CA) and a minor potassic calc-alkaline (KCA) series; they are characterized by variable K2O content, relatively high Al2O3 and low TiO2 content, suggesting a typical island arc environment. Such affinity is consistent with their moderately to fairly enriched incompatible element patterns showing negative anomalies in Nb, Zr, and to a lesser extent in Ti, typical of subduction-related magmas. No evidence of Neogene magmatic activity has been recorded anywhere on Sumba. Similarities between Sumba and the Southwestern Sulawesi magmatic belt with respect to both the Late Cretaceous-Paleocene magmatism and stratigraphy, support the idea that Sumba was part of an 'Andean' magmatic arc near the Western Sulawesi magmatic belt (Abdullah, 1994; Soeria-Atmadja et al., 1998) and near the Southeast Kalimantan coast (Meratus Mountains) (Yuwono et al., 1988; Wensink, 1997) at the margin of Asiatic Plate. The southward migration of Sumba to its present frontal arc position in the Sunda–Banda arc has occurred since Late Cretaceous–Paleocene time (Soeria-Atmadja et al., 1998). Comprehensive paleomagnetic study of Sumba Island was first provided by Wensink (1994). Paleomagnetic investigation of suitable rocks can be a valuable tool for the unraveling of tectonic problems, i.e. to determine the rotation the translation of the Sumba fragment during transport. In this regard, it is important that the ages of the studied rocks are well known and the tectonics are properly understood. The geology of Sumba reasonably satisfies both conditions. Wensink (1994) collected two hundred hand samples from three formations: dark colored, siliciclastic mudstones of the Late Cretaceous Lasipu Formation; volcanics of the Paleocene Massu Formation comprising basalts and andesitic basalts; and basalts from the Early Miocene Jawila Formation in West Sumba. The sediments of the Lasipu Formation revealed a paleolatitude of 18.3°; the volcanics of the Massu Formation gave a paleolatitude of 7.4°; and the volcanics of the Jawila Formation a paleolatitude of 9.9 ° (Figure 8A). These paleomagnetic data have been interpreted in terms of an original position of the Sumba fragment

in the northern hemisphere in Late Cretaceous time. Between the Late Cretaceous and Paleocene, Sumba performed a counterclockwise (CCW) rotation of 50° and a drift of 11° to the south; between the Paleocene and early Miocene the fragment moved in a CCW rotation of 85° and a drift of 17° to the south. Since the early Miocene, Sumba has occupied its present position. Later paleomagnetic studies of Sumba were detailed by Wensink and van Bergen (1995) of the early Miocene Jawila volcanics followed by Wensink (1997) for the Late Cretaceous Lasipu Formation. It was concluded by Wensink and van Bergen (1995) that paleomagnetic and geochemical evidence from the early Miocene volcanics of the Jawila Formation in western Sumba constrain the final drift stage and tectonic emplacement of the island. The lavas range from predominantly andesites to dacites, and indicate textural evidence for weak metamorphism. The Sumba fragment has occupied approximately its present position since the Miocene where the island was located within the latter arc between Sumbawa and eastern Flores, but with a minor southward drift. Based on a later paleomagnetic study, Wensink (1997) interpreted that Eastern Sundaland with Borneo, west and south Sulawesi, and Sumba formed one continental unit in the Late Mesozoic, most likely attached to the Southeast Asian mainland. The Sumba microcontinent most likely became detached from eastern Sundaland soon after deposition of the Lasipu sediments. Based on Pb-Nd isotopic characteristics of sediments and volcanics, Vroon et al. (1996) evaluated provenances of continental fragments in Eastern Indonesia (Figure 8B). The evidence is based on a comparison of Pb-Nd isotopic signatures between meta-sedimentary or volcanic rocks from the microcontinents and possible provenance areas. Provenance areas considered were continental margins of Australia-New Guinea or Sundaland. Pb-Nd isotopic variations in possible provenances were studied. North Australia has very high 206Pb/204Pb (up to 19.57) and low 143Nd/144Nd (0.51190-0.51200). Western New Guinea has low 206Pb/204Pb (18.6-19.0) and relatively high 143Nd/144Nd (0.51218-0.51225). The Bird’s Head area has 206Pb/204Pb of 18.60-18.75. Southern New Guinea has 206Pb/204Pb of 18.75-19.0. Sundaland has less radiogenic Pb isotopes. Marine sedimentary rocks of the Late Cretaceous Lasipu Formation in Sumba were analyzed for the Pb-Nd isotopes. They display limited variations in 143Nd/144Nd (0.51244-0.51248) and Pb isotopes (206Pb/204Pb = 18.74-18.77). Vroon et al. (1996)

interpreted that these isotopic signatures do not correspond to the Australian or New Guinean continental domains, and thus favor a northern rather than a southern origin. Because of stratigraphic indications for a paleoposition of Sumba near SW Sulawesi (Simandjuntak, 1993), Late Cretaceous flysch sedimentary rocks from the Balangbaru Formation of SW Sulawesi (Hasan, 1991) were analyzed for comparison. They yielded 143Nd/144Nd of 0.51246-0.51255 and Pb isotopes (206Pb/204Pb) of 18.67-18.74, which implies a close isotopic similarity with the Lasipu Formation. Based on this, it is considered that Sumba originated from SE Sundaland. Provenance of Sumba Island can also be investigated using certain Eocene larger foraminifera. Indo-Pacific Eocene carbonate sediments can be divided into two groups based on the presence of certain larger foraminifera (Lunt, 2003). One of these faunal groups is associated with the Sundaland craton, the geological core of western Indonesia and is also found on low latitude Pacific islands as well as low latitude western Tethyan regions. The second fauna is found on the Australian Plate, and the micro-plate terranes have been derived from it since the Eocene. This correlation leads to the hypothesis that the Middle and Late Eocene Sundaland fauna, identified by three, probably related genera: Assilina, Pellatispira, and Biplanispira [hereafter abbreviated "APB"] indicate a low latitude, shallow marine fauna, able to cross oceanic migration barriers but restricted from migrating far outside the tropics (Figure 9). In contrast, the fauna identified by the genus Lacazinella, which has about the same stratigraphic range as the APB lineage, is thought to be a higher latitude fauna centered on the Australian continent. This faunal difference occurred at a time of maximum separation of the Sunda and Australian plates. Therefore, subsequent Tertiary collision of these plates can be identified by the present complex distribution of previously separate faunas. Many islands that make up the Banda Arc, including Sumba, parts of Timor, and Seram have records of the APB fauna. This similarity indicates that they were separate from the Australian plate and at low latitudes in Eocene times. Assuming the palaeomagnetic anomaly "M25", end mid-Jurassic rift-drift event was the last known cause for separation of micro-plate fragments from the Australian margin (Veevers et al. 1988), then these fragments would have been carried north and become accreted onto the subducting margin south of Sundaland within the Cretaceous. Sumba as a

fragment of Sundaland based on geological criteria, is consistent with the faunal data. Caudri (1934, described in Lunt, 2003) reported and illustrated Assilina orientalis Douvillé and several species of Pellatispira from southern Sumba in the mid Eocene through Oligocene shallow marine Tanah Roong series. The presence of two typical Eocene low-latitude Sundaland fauna of three APB Assilina, Pellatispira, and Biplanispira and no Eocene high-latitude Australian fauna of Lacazinella shows that the provenance of Sumba Island was Sundaland. Based on our studies of stratigraphic succession, geochronology-geochemistry of magmatic rocks, paleomagnetism, isotope geology, and large Eocene foraminifera and earlier objections to Australian/Timor/Tethys provenances for the Sumba terrane, we believe that the origin/provenanse for Sumba terrane was SE/Eastern Sundaland. Sumba has a basement of Upper Cretaceous turbidites overlain unconformably by gently dipping Paleogene shallow water sediments and volcanic rocks and resembles the stratigraphy of the adjacent Asian margin in SW Sulawesi and offshore east Java (Packham, 1996). Tectonic reconstruction for SE Asia and SW Pacific during Cenozoic by Hall (2002) shows that Sumba was originally part of Eastern Sundaland located between East Java and South Sulawesi. Detachment and Emplacement of Sumba Terrane The southward movement of Sumba took place during pre-Neogene time by transcurrent/ transformal displacement and the island has occupied its present position in the forearc basin in front of Sunda-Banda volcanic arc since the early Neogene. The beginning of Sumba dispersion is various from the Late Cretaceous (Wensink, 1994, 1997) to Middle Miocene (Simandjuntak, 1993). Wensink (1994) and Wensink and van Bergen (1995) argued that based on the recent paleomagnetic evidence, there are indications that Sumba started to drift in the Late Cretaceous and had already arrived at or near its present position in the Early Miocene. Most of the authors (such as Parkinson et al, 1998) suggested the Paleogene as the period of Sumba dispersion. Detailed K-Ar chronology of Sumba magmatism shows its beginning during Late Cretaceous and its vanishing in Late Eocene-Early Oligocene. Regionally as well as chronologically, these results constrain the

geodynamical evolution of this area during the Oligocene and Miocene. During this period, Sumba Island shifted to its forearc position. Rutherford et al. (2001) presented another interpretation on periods of detachment and emplacement of Sumba. During the Late Cretaceous, Sumba formed part of a Great Indonesian Volcanic Arc system near southeastern Eurasia. Between Late Cretaceous and Early Miocene time, Sumba remained part of that arc system, which ceased to be volcanically active by 31 Ma. During the late Miocene (18 Ma), Sumba, as part of the relict arc system, was situated near the present site of Alor and Wetar. At 16 Ma, Sumba was torn from the relict arc and began to move velocity of 50 mm/yr in a west-south-westerly direction. Volcanism along the modern Banda Arc soon followed. By 7 Ma, Sumba finally came to rest at its present location, after having moved 450 km into the forearc. The history of detachment of Eastern Sundaland terrane is complicated and may involve a number of mechanisms (Satyana, 2003; Satyana, 2010), including: (1) crustal breakdown to the west of South Sulawesi volcanic arc by Plio-Pleistocene diastrophism, (2) back-arc spreading of marginal basins of Southwest Pacific areas, (3) rotation of the continental Southeastern Sundaland, (4) back-arc spreading due to subduction rollback related to India-Eurasia collision at 50 Ma, (5) southern extension related to sea-floor spreading of the Sulawesi Sea, (6) tectonic escape due to India-Eurasia collision and (7) mantle delamination by upwelling plume under Eastern Sundaland. The eastern margin of Sundaland is fragmented and tectonically complicated. The rock specimens comprise variably metamorphosed accretionary complexes, imbricated terranes, melange, turbidite and broken formations, and ophiolites. These rocks have suffered considerable dismemberment, tectonic and structural modification, and thermal overprinting due to tectonic and metamorphic activity throughout the Tertiary, related to the convergence of the Indo-Australian, Eurasian and western Pacific microplates (Parkinson et al, 1998). The provenance and way of detachment of some fragments believed once part of Eastern Sundaland are also complex and variably interpreted. Post-accretionary dispersion/detachment is a usual case in the Circum-Pacific region (Howell et al., 1985). The main period of accretionary activity ended by Early Tertiary time in the Cordillera and

in northeastern Siberia. These accretionary episodes have been followed by a history of complex strike-slip faulting, folding, and thrust faulting resulting in the breakup of some terranes. In Japan, left-slip faults are smearing out and dispersing the terranes while accretion is still occurring, and in eastern China, east-west trending left-slip faults resulting from the northeastward movement of India are fragmenting the collection of terranes in that area. At the eastern margin of Sundaland, the accretion stopped at around 50 Ma (Middle Eocene) and the accreted crust started to detach beginning with the opening of the Makassar Straits. The dispersion of terranes, by either rifting or sliding, results in the diminution of continents. Based on the tectonostratigraphy, Satyana (2003) described the accretion of SE/Eastern Sundaland by a number of terranes during the Late Jurassic to the earliest Tertiary. This resulted in the growing of Sundaland through amalgamated terranes and accreted mass associated with subduction and collision (Figure10). This growing continent by amalgamated terranes blocked the mantle circulation in the astenosphere. Buoyant mantle material unroofed the amalgamated terranes beginning around 50 Ma; in the Middle Eocene, some of the accreted mass of SE Sundaland rifted and drifted eastward, southward and southeastward slivering the continent. The dispersed mass includes: SW Sulawesi through opening of the Makassar Strait, Flores Sea Islands, and Sumba Island. Satyana (2003) suggesting that the uprise of buoyant metasomatized mantle in connection with the initial opening of Makassar Strait in the Early Tertiary was responsible for the separation of Sumba from the mainland of Sulawesi, as well as the islands of Doang and Salayar which now lie to the south of Sulawesi. Translation of these continental fragments occurred along the N-S trending proto-Paternoster-Walanae-Salayar fault zone between the Late Cretaceous and the early Miocene accompanied by crustal rifting and the left-lateral fault system facilitating the southward migration of Sumba and its counter-clockwise rotation (Soeria-Atmadja et al., 1998) (Figure 11). Another mechanism of migration was movement of the block/terrane by major strike-slip faults related to escape tectonics. Satyana (2006) considered the role of escape tectonism in western Indonesia following the collision of India with Eurasia in the Paleogene as motive for fashioning the present tectonic configuration. The term tectonic escape/escape tectonics/extrusion tectonics as

related to ‘indentation tectonics’, refers to the lateral motion of fault bounded geological blocks following collision (indentation). The motion is away from the collision zone and towards free oceanic zone. Strike-slip and extensional/rifting structures accommodate the lateral motion. In Kalimantan, major shear related to the India collision is the Lupar-Adang/Paternoster Fault (Satyana, 1994; Satyana, 1996; Satyana et al., 1999). It is a transverse trending major structural element shearing the island of Kalimantan from the Natuna Sea through to the Strait of Makassar as long as 1350 km. The trace of this fault may also continue or attach to the major faults in South Sulawesi such as Walanae Fault, Palu-Koro Fault until the Sumba Fracture. In Late Cretaceous-earliest Paleogene, Sumba and other terranes amalgamated SE/Eastern Sundaland. It is postulated that following the collision of India with Eurasia beginning in the Eocene, the major strike-slip fault of Adang-Paternoster-Walanae-Sumba Fracture resulted in escaped terranes (one of which was Sumba) southeastward/southward to the free oceanic edge which at that time was the ocean between the Sundaland and Australia. Another mechanism considered for Sumba detachment was back-arc spreading (Abdullah et al., 2000) (Figure 12). During the Paleogene, the rate of movement of the Indian oceanic crust subducting eastern Sundaland decreased, leading to the generation of back-arc basins and the formation of a marginal sea due to roll-back movement of the subducted plate. Back-arc spreading resulted in the southward/southeastward migration of Sumba. Back-arc spreading in SE/Eastern Sundaland caused the opening of the Makassar Strait separating western Sulawesi from SE/Eastern Kalimantan. Simandjuntak (1993) suggested displacement of the Sumba terrane could be kinematically related to one of the following tectonic movements: (1) Sumba detached from SE Kalimantan and rifted southwards by transcurrent-transformal displacement prior to the development of the late Neogene volcanic arcs in the Lesser Sunda region. (2) Sumba terrane detached from the rift zone subsequent to the extensional faulting leading to the break up and formation of the Makassar Strait during the separation of South Sulawesi from SE Kalimantan prior to the development of the late Neogene volcanic arcs in the Lesser Sunda region. (3) Mid-Miocene successions of turbidites in Sumba are quite different to the volcanic, carbonates, and molassic sediments in South Sulawesi. The

detachment of Sumba from near Bone Bay, or from the Walanae depression in the south arm of Sulawesi seems to have taken place in the middle Miocene by reactivated sinistral wrenching of the Palu-Koro Fault or the Walanae Fault prior to the development of the volcanic arcs in Lesser Sunda. Simandjuntak (1993) proposed that the northern part of Bone Bay is more likely to be the original site of the Sumba terrane as indicated by the geological similarity and a relatively good fitting of topography of Sumba with the northern part of the Bone Bay region. Based on paleomagnetic studies Wensink (1994, 1997) and Wensink and van Bergen (1995) described the emplacement of Sumba terrane from the northern hemisphere into its present position. The paleolatitudes of Sumba in Late Cretaceous and Paleocene times are 18.4°N and 7.4°N, respectively, therefore Sumba drifted 11° to the south; between the Paleocene and early Miocene Sumba drifted of 17° to the south. Since the early Miocene the island of Sumba has been in approximately its present position. During its drifting, Sumba underwent several periods of counter-clockwise rotation. Total drift of Sumba from its provenance at 18.4°N to its present position at 9.9°S, moved as far as 28.3° cross latitudinal. Petroleum Implications Sumba Island has been investigated for the possibility of petroleum accumulation since 1940 when geologists of NPPM (Nederlandsche Pacific Petroleum Maatschappij) spent six months of reconnaissance work on the island (van Bemmelen, 1949). No further petroleum exploration on Sumba Island and surrounding seas was carried out until Burollet and Salle, geologists from Total, published their geological study of Sumba Island in 1981 for the main purpose of gaining a better understanding of the geodynamic position of Sumba and gathering more information to prepare a program for the 1980-1981 offshore seismic campaign. After that work, Sumba Island and surrounding seas have been unexplored. The forearc basin of the Savu Sea, located to the east of Sumba Island has been more explored than Sumba area. The Savu Basin was explored from 1968-1975 by International Oil Exploration and Woodside-Burmah Oil N.L. These activities included field work in Savu Island, offshore 2D-seismic surveys (1968, 1970, 1975) and drilling of one exploration well Savu-1 by Woodside-Burmah Oil N.L. in 1975. The well (TD 1225 m) was drilled

at the Savu Ridge, penetrating Neogene claystones and carbonates and undifferentiated highly deformed pre-Neogene section. The well was dry and no reservoir was encountered. Because the Savu Basin is located near the prolific NW Shelf of Australia, new 2D seismic was shot in 2002 (2,740 km) and 2007 (3,000 km). Gravity data as acquired along with both seismic surveys. Recent publications on Savu Basin are from Tampubolon and Saamena (2009), Toothill and Lamb (2009) and Rigg and Hall (2010). Some seismic lines of Savu Basin include the offshore areas of Sumba Island. These seismic lines will be used as the basis to review the prospectivity of Sumba area. The geologic settings of Savu Basin and Sumba area are different. Sumba area is an exotic terrane and prospectivity review should include consideration of this matter. As a terrane, it is interesting to review the petroleum implications for Sumba area. Recent publication by Satyana et al. (2008) discusses some terranes/microcontinents in Indonesia where there are petroleum accumulations in sedimentary basins formed by collision of these terranes. The best examples of this are petroleum accumulations related to the collision of the Banggai-Sula microcontinent. The Banggai-Sula microcontinent has Australian-origin. The collision of the Banggai-Sula micro-continent with East Sulawesi ophiolites was responsible for the formation of the foreland Banggai Basin, its foredeep and fold-thrust belt. Gas and oil fields have been discovered in the Banggai Basin, including: the Minahaki, Matindok, Senoro, Donggi, Sukamaju, and Maleo Raja gas fields; and the Tiaka oil field sourced by Miocene lagoonal carbonates and shales and reservoired by syn-drifting Miocene reefal and platform carbonates. Australian Mesozoic sediments were deposited as a syn-rift sequence in grabens of the Banggai-Sula microcontinent. When collision of the micro-continent took place in the Late Miocene, the rift grabens subsided and were partly overprinted by compressional tectonics resulting in thrusted anticlines. Some thermogenic gas seeps expected to be sourced by Mesozoic sequences occurred in this area. Collision and post-collision tectonic escape in the Banggai-Sula collision significantly affected: (1) basin formation due to isostatic subsidence and underthrusting of the micro-continent, and postcollision extension, (2) sedimentation of postcollision/molassic deposits, (3) subsidence of the basins due to deposition of molasses and/or

thrust sheet of post-collision sequences, (4) generation of hydrocarbons in Miocene and Mesozoic sources due to isostatic subsidence and/or burial by multiple thrust sheets, and (5) trap formation related to collisional thrusting and post-collision wrenching. However, we consider that the proven collision-related petroleum play of Banggai-Sula microcontinent will not typify Sumba microcontinent due to: (1) the absence of collision in the history of Sumba tectonic transport and (2) the absence of Australian Mesozoic source rock and reservoirs such as Australian originated microcontinents (Banggai, Buton, Seram, Bird's Head of Papua). Other petroleum plays, not typical of Eastern Indonesia microcontinents, should be applied in exploring Sumba area. Based on the origin of Sumba terrane, we consider that play types of Sumba area will be equal with proven play types of Western Indonesian basins, especially in Eastern Sundaland basins (East Java Basin or western Sulawesi basins). Basically, Western Indonesian basins demonstrate gross similarities in terms of structure and stratigraphy reflecting common regional controls throughout their Cenozoic histories (Netherwood, 2000). A common middle to late Eocene timing for initial basin rifting and associated fluvio-lacustrine fill, including the main source rocks for the majority of Western Indonesian basins. Transgression from the middle Oligocene to middle Miocene with fluvial reservoirs being succeeded by the main deltaic and carbonate reservoirs occurred in the late Oligocene to early Miocene, with regional seals deposited in the Middle Miocene at maximum transgression. Late Miocene through Pliocene compression resulted in structuring events and increased heat flow associated with the collision of the Australian craton with the Asian plate, 8-3 Ma, and collision of the Luzon arc with the Asian plate at about 5 Ma. Although there are gross geological similarities between Western Indonesia basins, there are also significant geological differences. These are primarily controlled by basin position on Sundaland promontory in relation to present-day and Cenozoic subduction of the Indo-Pacific plate northwards beneath Sundaland. Considering the origin of Sumba terrane from western Sulawesi area/Eastern Sundaland, the analogy is made to the South Makassar and Bone Basins, especially during the Paleogene. During the Paleogene, the South Makassar and Bone basins show elements and processes of petroleum systems

related to rifting. Potential source rocks are syn-rift Early Tertiary lacustrine, woody terrigenous to marine lagoonal source rocks in buried half grabens. Potential reservoirs are syn-rift fluvial and paralic sands, late syn-rift paralic to nearshore marine sand and early sag phase (Eocene to early Oligocene) carbonate reef and sand reservoirs. Potential seals are interbedded claystones in syn-rift and early sag phase deposits and interbedded hemipelagic claystones in basinal deposits. Paleogene play types of South Sulawesi Basin are: rotated fault blocks, synrift section, deformed/fractured carbonates, thrust fold features, slope channels, and inverted pinchouts. Paleogene play types of Bone Basin are: faulted anticlines, tilted fault blocks, stratigraphic subcrop plays, drape over basement highs, turbidite fans, slope channel fill, and stratigraphic pinchout (Sudarmono, 2000; Yulihanto, 2004). Elements of the system in Sumba area, as analogue to South Makassar and Bone Basin can be inferred from seismic data of Sumba area. New seismic data around Sumba show the presence of these Paleogene rifts and their possible play types. The existence of elements of petroleum system in Sumba area can be examined by reviewing the related rocks in Sumba Island which may continue into the offshore areas composing elements of the petroleum system. Field observations and laboratory analyses reported by Burollet and Salle (1981) provided this examination. Potential reservoir rocks can be referred to: (1) Eocene neritic facies (littoral to middle shelf) that comprise mainly medium to coarse greywackes. These greywackes correspond to the destruction of the Lower Paleocene volcanic massifs. They are dated Middle Eocene. They are locally associated with Nummulitic mark, probably deposited in a slightly shallower environment. During the Upper Eocene, these facies pass progressively to greywackes with abundant planktonic foraminifera and with rare large benthic foraminifera. This outer shelf environment indicates a progressive deepening which culminates in bathyal facies observed in the form of sandy shale with abundant radiolaria and diatoms. (2) The Neogene formation (coarse conglomerates, sandstones and limestones) which covers most of Sumba Island, unconformably overlies all the older formations and seems to drape away from the Pre-Tertiary massifs. Early-middle Miocene limestones comprising bioclastic packstone with abundant micritized and rolled red algae fragments and scattered quartz. In the Western part of the island, mainly around Waikabubak and more precisely between this town and the south

coast, reefal and bioclastic limestones (corals, algae, foraminifera, etc.) overlay a thick volcano-detrital series with tuff, cinerite, clay, tephra, etc. The reefal or bioclastic limestone extends east of Waikabubak as far as the Tanadaro mountains and seems to be there at the lower part of the carbonate series. Its age is indicated by calcareous nannoplankton as early Miocene to basal Middle Miocene (see Figure 5B). (3) Lower section of the Waingapu Formation (late Miocene) consisting mainly of fine to coarse grained sandstones, locally microconglomeratic with boulders of volcanic rocks, marls and limestones, locally cemented by crystalline calcite. Reworked large benthonic foraminifera from Miocene were observed in several samples. These sandstones are interbedded with sandy shales. The rocks are exposed at the eastern part of the island. Potential source rocks of Sumba are Neogene rocks with abundant plant remains (coal). Conglomerates are overlain by black, pyritous organic shale with a very rich fauna. The combined presence of fresh water algae (Botryococcus) indicates a restricted brackish environment, first step of the transgression. This shale is the only sample from Sumba that contains a significant amount of organic matter with TOC: 4.86%, (excellent). The organic matter is a typical humic material with a lower maturation stage. Botryococcus are fresh water algae species living in restricted lacustrine environments (an example of proven lacustrine kitchen is excellent quality-Oligocene Pematang source rocks from Central Sumatra Basin). Based on this, it is considered that Sumba area indicates the presence of both oil-prone (lacustrine) and gas-prone (humic coal) source rocks. Presence of pyritous organic shale and excellent TOC shows very good-excellent preservation of organic matter in restricted-reduction environments. Seals for reefal and coralline limestones can be provided by shaly planktonic foraminiferal wackestone-packstone with abundant coccoliths, radiolaria and sponge spicules associated with fine volcanic glass fragments. The microfauna is characteristic of bathyal environment and the age is middle Miocene (Langhian-Serravallian). The rocks are exposed on the western side of Tanadaro. So the subsidence of the reefal environment to deep bathyal zones is precisely located at the limit of lower to middle Miocene (analogous to East Java Basin where prolific Mudi reefs are sealed by Tuban shales). Late Miocene chalky pelagic limestones (typical of bathyal environment) can be seals for Late Miocene Waingapu sandstones. In the entire series, the mineralogy is characterized by the

association montmorillonite and mixed layers of illite/montmorillonite then by exclusive montmorillonite. These associations are directly related to the transformation of detrital volcanic rocks in deep water environment. Recent seismic lines in offshore Sumba areas (Toothill and Lamb, 2009) (Figures 13-15) show the presence of rifted structures as commonly recorded in Eastern Sundaland areas (Satyana, 2010). Potential traps related to rifted structures are: tilted fault blocks related to rifted basin and drape channel sands overlying the basement high, reefal buildups over the horst, faulted anticlines, traps in synrift sections, stratigraphic subcrop plays, turbidite fans, slope channel fill, stratigraphic pinchout, and fractured basement highs. Hydrocarbon kitchens may exist in the synrift sections. Seismic lines prove the absence of structures related to collision, like those developed in Timor area. Hydrocarbon potential of the Sumba area is enhanced by the recording of seeps on satellite images, many of which show good correlation with geological features seen on seismic data (Toothill and Lamb, 2009) (Figure 14). The seeps and their strong correlation with geological features show that a hydrocarbon system is active in the basin. These seepages are indicative of migrating hydrocarbons. Also, some of the mapped slicks show clustering, which may relate to multiple vents associated with the same geological feature. Basically, the presence of source, reservoir, and sealing-quality rocks are exposed in Sumba Island; rifted structures in Sumba offshore areas with the configuration of typical graben kitchens and traps of rifted structures as revealed by recent seismic lines; and a number of oil seeps/slicks offshore Sumba could indicate the presence of an active petroleum system. Figure 16 shows the petroleum system of South Sulawesi, stratigraphic succession of Sumba, typical of Sundaland or South Sulawesi, and possible elements and processes of petroleum systems in Sumba. CONCLUSIONS Sumba Island is a terrane (microcontinent) presently located in a forearc setting of the Sunda-Banda volcanic arcs. Tectonically, the island is important since it is located at the border between the subduction zone of Indian oceanic crust beneath Sunda Arc to the west and the collision zone of Australian continental crust with Timor island arc to

the east. Sumba lies obliquely between two fore-arc basins, the Lombok Basin to the west and the Savu Basin to the east. The origin of Sumba terrane has been a matter of considerable debate in geologic literature. There are two main competing hypotheses: (1) Northwest Australian provenance and (2) SE/Eastern Sundaland provenance. In addition to these, other provenances argued by previous workers are: (3) Timor Island and (4) Tethys Sea isolated microcontinent. We examined the possibilities of these four provenances, and believe that SE/Eastern Sundaland is the most plausible origin for the Sumba terrane. SE/Eastern Sundaland as the origin of the Sumba terrane is supported by examinations and interpretations of stratigraphic succession, geochronology-geochemistry of magmatic rocks, paleomagnetism, isotope geology, and large Eocene foraminifera. Detachment of Sumba terrane from SE/Eastern Sundaland could be a result of a number of mechanisms such as: mantle delamination by upwelling plume under the Eastern Sundaland or back-arc spreading due to subduction rollback related to India-Eurasia collision at 50 Ma. The movement of Sumba terrane to its present position may be through regional strike-slip faults of Paternoster-Walanae-Selayar-Sumba Fracture as a response of escape tectonics due to the India-Eurasia collision. The movement involved southward drift of Sumba terrane as far as 28.3° cross latitudinal and several episodes of counter-clockwise rotation. The detachment and emplacement took place during the Paleogene. Petroleum prospectivity of Sumba microcontinent cannot be inferred from proven plays of other microcontinents in Eastern Indonesia (Banggai-Sula, Buton, Kepala Burung). Sumba was an Asian microcontinent which means it has no Mesozoic and Upper Paleozoic sediments which are prolific in Australian continent/microcontinents, and Sumba has no history of collision for its emplacement. There was no foreland basin developed due to collision in Sumba area, foreland basins are proven petroleum provinces within Eastern Indonesian microcontinents subjected to collisional tectonics. Based on newly acquired seismic lines, rifted structures typical of South Sulawesi, southern Makassar Straits, East Java, and Bone Bay can be analogous for petroleum plays of Sumba. Sumba area has requisite characteristics for a petroleum producing province and is worthy of

further exploration. These characteristics are: (1) Sumba Island has source, reservoir, and sealing rocks extending from Sumba into offshore areas; (2) recently acquired seismic lines show the presence of rifted structures in offshore area with many possible traps, and (3) there are a number of hydrocarbon seeps/slicks offshore indicating the presence of an active petroleum system. ACKNOWLEDGMENTS We thank the Technical Program Committee of IPA for selecting this study to be published and presented, and for reviewing this paper. BPMIGAS and ConocoPhillips Management are acknowledged for supporting the the authors to conduct and publish this study. Dr. Chalid Idham Abdullah and Benyamin Sapiie, Ph.D. (ITB), Dr. Carolus Prasetyadi (UPN), Chandra Tiranda (Mitra Energy, presently with Talisman Energy), Yunan Muzaffar (Directorate General of Oil and Gas) contributed significant literature, unpublished data, and discussions. REFERENCES CITED Abdullah, C.I., 1994, Contribution á l’étude géologique de I’lle de Sumba : Apports a La Connaissance de La Géodynamique de L’Archipel Indonésien Orientale, Thése de Doctorat, Université de Savoie, Chambery, France, 255 ps, unpublished. Abdullah, C.I., 2010, Evolusi magmatisme pulau Sumba, Proceedings of Indonesian Association of Geologists, 39th Annual Convention and Exhibition. Abdullah, C.I., Rampnoux, J.P., Bellon, H., Maury, R.C., Soeria-Atmadja, R., 2000, The evolution of Sumba Island (Indonesia) revisited in the light of new data on the geochronology and geochemistry of the magmatic rocks, Journal of Asian Earth Sciences 18, 533-546. Audley-Charles, M.G., 1975, The Sumba Fracture : a major discontinuity between western and eastern Indonesia, Tectonophysics, 26, 213-228. Audley-Charles, M.G., 1985, The Sumba Enigma: Is Sumba a Diapiric Fore-arc Nappe in process of Formation?, Tectonophysics, Abstract, 119, 1-4, 435-449.

Burrolet, P.F., and Salle, C.I., 1981, A Contribution to the Geological Study of Sumba (Indonesia), Proceedings of Indonesian Petroleum Association 10th Annual Convention, p. 331-344. Chamalaun, F.H. and Sunata, W., 1982, The paleomagnetism of the Western Banda Arc system : Sumba, in Paleomagnetic research in Southeast and East Asia, Proceedings of a Workshop, Kuala Lumpur, Malaysia, March 1992, Joint Prospecting for Mineral Resources in Asian Offshore Areas (CCOP), Bangkok, p. 162-194. Chamalaun, F.H., Grady, A.E., von der Borch, C.C., and Hartono, H.M.S., 1983, Banda Arc tectonics – the significance of Sumba Island in Watkins, J.S. and Drake, C.L. eds, Studies in Continental Margin Geology – American Association of Petroleum Geologists Memoir No. 34, Tulsa, p. 361-376. Chamalaun, F.H., Grady, A.E., Von der Borch, C.C., Hartono, H.M.S., 1981, The tectonic significance of Sumba, Bulletin Geological Research and Development Centre, Bandung, 5, 1-20. Djumhana, N. and Rumlan, D., 1992, Tectonic concept of the Sumba continental fragment, Eastern Indonesia, Proceedings of Indonesian Association of Geologists, the 21th Annual Scientific Meeting, p. 585-598. Effendi, A.C. & Apandi, C. 1980, Geological Map of Sumba Quadrangle, Nusa Tenggara, Geological Research & Development Centre, Ministry of Mines and Energy, Bandung, Indonesia. Effendi, A.C. and Apandi, T., 1994, Geological report of Waikabubak and Waingapu Quadrangle, Scale 1:250.000, Geological Research and Development Centre, Bandung, Indonesia. Falvey, D. A. and Mutter, J. C. 1981, Regional plate tectonics and the evolution of Australia's passive continental margins, Journal of Australian Geology and Geophysics, 6, 1 29. Fortuin, A.R., Roep, Th. B., Sumosusastro, P.A., van Weering, T.C.E, van der Werff, W., 1992, Slumping and sliding in Miocene and Recent developing arc basins, onshore and offshore Sumba (Indonesia), Marine Geology, 108, 345-363. Fortuin, A.R., Roep, Th., Sumosusastro, P.A., 1994, The Neogene sediments of east Sumba, Indonesia -

products of a lost arc?, Journal of Southeast Asian Earth Sciences, 9 (1/2), 67-79. Fortuin, A.R., van der Werff, W., Wensink, H., 1997, Neogene basin history and paleomagnetism of a rifted and inverted forearc region, on and offshore Sumba, Eastern Indonesia, Journal of Asian Earth Sciences, 15 (1), 61-88. Hall, R., 2002, Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions, model and animations, Journal of Asian Earth Sciences, 20, 353-434. Hamilton, W., 1979, Tectonics of the Indonesian Region, Geological Survey Professional Papers No. 1078, US Government Printing Office, Washington DC. Hartono, H.M.S. and Tjokrosapoetro, S., 1984, Preliminary account and reconstruction of Indonesian terranes, Proceedings of Indonesian Petroleum Association, 13th Annual Convention, p. 185-226. Hasan, K., 1991, The Upper Cretaceous Flysch Succession of the Balangbaru Formation, Southwest Sulawesi, Proceedings of Indonesian Petroleum Association, 20th Annual Convention, p. 183-208. Howell, D.G., Jones, D.L., Schermer, E.R., 1985, Tectonostratigraphic terranes of the Circum-Pacific Region in Howell, D.G., ed., Tectonostratigraphic Terranes of the Circum-Pacific Region, Circum-Pacific Council for Energy and Mineral Resources, Houston, p. 3-23. Keep, M., Longley, I. and Jones, R., 2003, Sumba and its effect on Australia’s northwestern margin, in Hillis, P. R. and Muller, R. D., eds., Evolution and Dynamics of the Australian Plate, Geological Society of Australia Special Publication 22. Lunt, P., 2003, Biogeography of some Eocene larger foraminifera, and their application in distinguishing geological plates, Palaeontologia Electronica, 6, 1, http://palaeo-electronica.org/paleo/2003_2/geo/issue2_03.htm Netherwood, R., 2000, The Petroleum geology of Indonesia, Reservoir Optimization Conference, Schlumberger, Jakarta.

Nishimura, S. and Suparka, S., 1986, Tectonic development of East Indonesia, Journal of Southeast Asian Earth Sciences, 1, 45-57. Norvick, M. S. 1979, The tectonic history of the Banda Arcs, eastern Indonesia: a review, Journal of Geological Society London, 136, 519-527. Otofuji, Y., Sasajima, S., Nishimura, S., Dharma, A., and Hehuwat, F., 1981, Paleomagnetic evidence for clockwise rotation of the northern arm of Sulawesi, Indonesia, Earth Planetary Science Letters, 54, 272-180. Packham, G., 1996, Cenozoic SE Asia : reconstructing its aggregation and reorganization, in Hall, R. and Blundell, D., eds., Tectonic Evolution of Southeast Asia, Geological Society Special Publication No. 106, p. 123-152. Parkinson, C.D., Miyazaki, K., Wakita, K., Barber, A.J., and Carswell, D.A., 1998, An overview and tectonic synthesis of the pre-Tertiary very-high-pressure metamorphic and associated rocks of Java, Sulawesi and Kalimantan, Indonesia, The Island Arc (1998) 7, 1-17. Pigram, C. J., Panggabean, H., 1984, Rifting of the northern margin of the Australian continent and the origin of some microcontinents in eastern Indonesia, Tectonophysics, 107, 331-353. Rangin, C., Jolivet, L., Pubellier, M. and the Tethys Pacific Working Group, 1990, A simple model for the tectonic evolution of Southeast Asia and Indonesia region for the past 43 M.Y., Bulletin Society of Geological French, 8, 6, 889-905. Rigg, J.W.D. and Hall, R. 2010, Structural and Stratigraphic Evolution of the Savu Basin, Indonesia, in Hall, R., Cottam, M.A. and Wilson, M.E.J., eds., The SE Asian Gateway: History and Tectonics of Australia-Asia Collision, Geological Society of London Special Publication. Rutherford, E., Burke, K., and Lytwyn, J., 2001, Tectonic history of Sumba island, Indonesia, since the Late Cretaceous and its rapid escape into the forearc in the Miocene, Journal of Asian Earth Sciences, 19, 4, 453-479. Satyana, A.H., 1994, The Northern massifs of the Meratus Mountains, South Kalimantan: nature, evolution and tectonic implications to the Barito structures, Proceedings of Indonesian Association of Geologists, 23rd Annual Convention, p. 457-470.

Satyana, A.H., 1996, Adang-Lupar Fault, Kalimantan: controversies and new observations on the trans-Kalimantan megashear, Proceedings of Indonesian Association of Geologists, 25th Annual Convention, p. 124-143. Satyana, A.H., Imanhardjo, D.N., and Surantoko, 1999, Tectonic controls on the hydrocarbon habitats of the Barito, Kutei, and Tarakan basins, Eastern Kalimantan, Indonesia: major dissimilarities in adjoining basins, Journal of Asian Earth Sciences, 17, 99-122. Satyana, A.H., 2003, Accretion and dispersion of Southeast Sundaland: the growing and slivering of a continent, Proceedings of Joint Convention of Indonesian Association of Geologists and Indonesian Association of Geophysicists, Jakarta. Satyana, A.H., 2006, Post-collisional tectonic escapes in Indonesia: fashioning the Cenozoic history, Proceedings of Indonesian Association of Geologists, 35th Annual Convention. Satyana, A.H., Armandita, C., and Tarigan, R., 2008, Collision and post-collision tectonics in Indonesia: roles for basin formation and petroleum systems, Proceedings of Indonesian Petroleum Association, 32nd Annual Convention, Jakarta. Satyana, A.H., 2010, Crustal structures of the Eastern Sundaland’s rifts, Central Indonesia: geophysical constraints and petroleum implications, Bali2010 International Geosciences Conference and Exposition, HAGI-SEG Joint Convention, Nusa Dua. Sikumbang, N., 1986, Geology and Tectonics of the Pre-Tertiary Rocks in the Meratus Mountains, SE Kalimantan, Indonesia, PhD Thesis, University of London, unpublished. Simanjuntak, T.O., 1993, Tectonic Origin of Sumba Platform, Jurnal Geologi dan Sumberdaya Mineral, III/22, 10-29. Soeria-Atmadja, R., Suparka, S., Abdullah, C.I., Noeradi, D., Sutanto, 1998, Magmatism in western Indonesia, the trapping of the Sumba block and the gateways to the east of Sundaland, Journal of Asian Earth Sciences, 16, 1, 1-12. Sudarmono, 2000, Tectonic and stratigraphic evolution of the Bone Basin, Indonesia: insights to

the Sulawesi collision complex, Proceedings of Indonesian Petroleum Association, 27th Annual Convention, p. 531-544. Tampubolon, B.T. and Saamena, Y., 2009, Savu Basin: a case of frontier basin area in Eastern Indonesia, Proceedings of Indonesian Petroleum Association, 33rd Annual Convention, Jakarta. Toothill, S. & Lamb, D. 2009, Hydrocarbon prospectivity of the Savu Sea basin, Proceedings of Indonesian Petroleum Association, 33rd Annual Convention, Jakarta. Van Bemmelen, R.W., 1949, The Geology of Indonesia, vol. 1A, Governmen Printing Office, The Hague. Van der Werff, 1995, Cenozoic evolution of the Savu Basin, Indonesia: forearc basin response to arc-continent collision, Marine and Petroleum Geology, 12, 3, 247-262. Van der Werff, W., Kusnida, D., Prasetyo, H., and van Weering, T.C.E., 1994, Origin of the Sumba forearc basement, Marine and Petroleum Geology, 11, 363-374. Veevers, J. J., 1988, Morphotectonics of Australia's northwestern margin - a review, in Purcell, P. G. and Purcell, R. R., eds., The North West Shelf Proceedings of Petroleum Exploration Society of Australia Symposium, p. 19-27. Von der Borch, C.C., Grady, A.E., Hardjoprawiro, S., Prasetyo, H., and Hadiwisastra, S., 1983, Mesozoic and Late Tertiary submarine fan sequences and their tectonic significance, Sumba, Indonesia, Sedimentary Geology, 37, 113-132. Vroon, P.Z., van Bergen, M.J., and Forde, E.J., 1996, Pb and Nd isotope constraints on the provenance of tectonically dispersed continental fragments in east Indonesia, in Hall, R. and Blundell, D., eds., Tectonic Evolution of Southeast Asia, Geological Society Special Publication No. 106, p. 445-453. Wensink, H. and van Bergen, M.J., 1995, The tectonic emplacement of Sumba in the Sunda-Banda Arc: paleomagnetic and geochemical evidence from the early Miocene Jawila volcanic, Tectonophysics, 250, 15-30. Wensink, H., 1994, Paleomagnetism of rocks from Sumba: tectonic implications since the late

Cretaceous, Journal of Southeast Asian Earth Sciences, 9 (1/2), 51-65. Wensink, H., 1997, Paleomagnetic data of Late Cretaceous rock from Sumba, Indonesia: the rotation of the Sumba continental fragment and its relation with eastern Sundaland, Geologie en Mijnbouw, 76, 57-71. Yulihanto, B., 2004, Hydrocarbon play analysis of the Bone Basin, South Sulawesi, Proceedings of

International Geoscience Conference on Deepwater and Frontier Exploration in Asia and Australasia, Indonesian Petroleum Association, p. 333-348. Yuwono, Y.S., Priyomarsono, S., Maury, R.C., Rampnoux, J.P., Soeria-Atmadja, R., Bellon, H., and Chotin, P., 1988, Petrology of the Cretaceous magmatic rocks from Meratus Range, Southeast Kalimantan, Journal of Southeast Asian Earth Sciences, 2, 1, p. 15-22.

Figure 1 - Location of Sumba Island, in front of Sunda-Banda volcanic arc and in between Lombok and

Savu basins.

Figure 2 - Sumba Island in regional tectonic setting of Eastern Indonesia. The island is located in the

forearc of the Sunda-Banda volcanic arc at the border between Java Trench (subduction zone of Indian Ocean) and Timor Trough (collision zone of Australian Continent). Sumba is bordered by major faults that could have transported the island during the Paleogene to its present position. Lombok and Savu basins are forearc basins. Blocks colored in purple in SE Kalimantan (Borneo) and southern Sulawesi (Celebes) are argued as the origin of the Sumba terrane (after Hamilton, 1979; Burollet and Salle, 1981; Abdullah et al., 2000).

Figure 3 - Geological sketch map of Sumba. Box A, B, C are profiled in Figure 4 (Abdullah et al., 2000;

Abdullah, 2010).

Figure 4 - Stratigraphic columns/profiles of Sumba from west to east. Area of columns as A (west), B (central), E (east) is shown at Figure 3 (Abdullah et al.,

2000, Abdullah, 2010).

Figure 5 - Some outcrop photographs of Central Sumba, showing various rocks comprising the central part of Sumba Island. Geological map is taken from Abdullah (2010), legends of rock unit see Figure 3. Field survey took place in September 2010. The rocks are: (A). Mio-Pliocene chalky carbonates of Waingapu Formation, (B). Early Miocene-basal middle Miocene coralline limestones of Waikabubak Formation, (C). Eocene foraminiferal marlstone/limestone, brecciated of Paumbapa Formation, and (D). Late Cretaceous turbiditic interbedded sandstones and claystones of Lasipu Formation.

Figure 6 - Schematic diagram showing Cenozoic reconstruction for Sumba when considered as part of Timor during the Late Cretaceous-Paleocene, detached and emplaced at its present position during Mio-Pliocene time due to post-collision tectonic escape relating to the collision of Australia into Timor. The Savu Basin was opened due to the escape of Sumba from Timor. The escape was accommodated by major strike slip faults (Djumhana and Rumlan, 1992).

Figure 7 - Stratigraphic correlation of Sumba, South Sulawesi and SE Kalimantan (after Simandjuntak, 1993). Based on stratigraphic succession, it is obvious that Sumba is very similar to South Sulawesi, indicating that Sumba shared same place with South Sulawesi before dispersion.

Figure 8A - Paleolatitudinal positions for the island of Sumba derived from paleomagnetic data of three different formations (Wensink, 1994). The sediments of the Lasipu Formation revealed a paleolatitude of 18.3° N; the volcanics of the Massu Formation gave a paleolatitude of 7.4° N; the volcanics of the Jawila Formation presented a paleolatitude of 9.9 ° S. Between the Late Cretaceous and Paleocene, Sumba performed a counterclockwise (CCW) rotation of 50° and a drift of 11° to the south; between the Paleocene and early Miocene the fragment moved a CCW rotation of 85° and a drift of 17° to the south. Since the early Miocene, Sumba has occupied its present position.

Figure 8B - Comparison of Pb-Nd isotopic signatures between meta-sedimentary or volcanic rocks from

the microcontinents and possible provenance areas (Vroon et al., 1996). North Australia has very high 206Pb/204Pb and low 143Nd/144Nd. Sundaland has less radiogenic Pb isotopes. Late Cretaceous Lasipu Formation of Sumba displays limited variations in 143Nd/144Nd and Pb isotopes. A close isotopic similarity occurs between samples from Sulawesi and Sumba, showing close relationship. Based on paleomagnetism and isotope geology, it is concluded that Sumba originated from SE/Eastern Sundaland.

A

B

Figure 9A - Summary of large Eocene foraminifera of Assilina/Pellatispira/Biplanispira (APB) and Lacazinella faunas in the Indo-Pacific realm (Lunt, 2003). The APB faunal group is associated with the Sundaland Craton and is also found on low latitude Pacific islands as well as low latitude western Tethyan regions. The Lacazinella fauna is found on the Australian Plate, and the micro-plate terrains have been derived from it since the Eocene. Note Sumba is included into the Sundaland APB group.

Figure 9B - Middle and Late Eocene plates and climatic zones with locations of APB fauna in red,

Lacazinella fauna in blue. In shallow marine carbonate facies the APB fauna appears to dominate the tropics but can occur locally at higher latitudes, whereas Lacazinella is a high latitude Austral index.

Figure 10 - Paleotectonic reconstruction of the SE/Eastern Sundaland and its accreted crust during the Cretaceous (Satyana, 2003). Present-day outlines of Java, parts of Sumatra and Kalimantan, and Sumba are shown for reference. The Sumba microcontinent accreted to SE/Eastern Sundaland. The accreted masses to SE/Eastern Sundaland had blocked mantle circulation. Upwelling of buoyant metasomatized mantle delaminated the upper accreted crust, resulting in initial opening of the Makassar Strait in the Early Tertiary causing the separation of Sumba from the mainland of Sulawesi, as well as the islands of Doang and Salayar that now lie to the south of Sulawesi [Late Cretaceous fore-arc basin from Hasan (1991), Alino Arc from Sikumbang (1986).Gondwanan micro-continents from Parkinson et al. (1998). Cretaceous island arc from Hamilton (1979)].

Figure 11 - Dispersion reconstruction of eastern margin of SE Sundaland. The dispersion took place in

response to back-arc spreading behind the magmatic arcs of Java-West Sulawesi and displacement by regional transform faulting. Sumba occupied its present position before the formation of Late Miocene arc. (after Soeria-Atmadja et al., 1998).

Figure 12 - Schematic 3D-diagrams depicting the four main stages of tectonic evolution of Sumba: (A) Late Cretaceous-Paleocene, (B) Paleogene, (C) Middle Miocene-Pliocene, (D) Quaternary (Abdullah et al., 2000; Abdullah, 2010).

Figure 13 - NW-SE Seismic Section across the western sector of the Savu Basin (Toothill and Lamb, 2009). Situated at the northwest end of the 400km seismic line, the section crosses the western end of the Savu Sea Basin. This demonstrates the development of the basin with sedimentary section thickening towards the northwest. Six seismic horizons have been interpreted within the basin including the rift event. The deepest visible reflector (pink horizon) shows significant faulting. A thick sediment section is shown at the northwest end of the line. The deepest sediment is deposited into rifted half grabens that measure approximately 10km across. Two basement highs occur in the section and may be associated with promontories along the coast of Sumba Island, some 15 km to the southwest. In the southeast, as the line draws closer to the coast of this island, the section thins significantly and major basement uplift is present. Sediment deposited after the main rift event appears typical of that expected post rifting, with high and low amplitude alternating sequences probably representing sands and shales infilling the local post-rift topography. The section between the sage green and deep blue horizons appears quiescent in the deeper part of the basin but, as it climbs to the southeast, appears to have undergone slumping, possibly associated with increasing proximity to Sumba Island. Above the deep blue horizon, depositional patterns appear to vary laterally and deep channels up to 12 km across are seen at various locations. Above the light green horizon deposition becomes even more varied and in places chaotic, possibly indicating the rapid uplift of Sumba island immediately to the west and the sudden influx of sediments associated with it.

Figure 14 - NW-SE oriented seismic line that runs along the western edge of the Savu Sea Basin, close to

the Island of Sumba (Toothill and Lamb, 2009). A number of seeps are located above this line (shown by arrows), all of which show clustering (see index map of hydrocarbon seeps/slicks - light green, red, yellow small circles, courtesy of Mitra Energy and Directorate General of Oil and Gas, Indonesia). The most north-westerly seep, which in fact comprises a cluster of five seeps, is approximately above a basement outcrop from either side of which sedimentary section and the sea floor dips steeply away. This is the highest point along the entire length of the section and would be a natural point of leakage for hydrocarbons migrating along carrier beds in the sedimentary section. Seep data acquired by satellite. Mapping have shown that a number of the identified seeps which fall above or very close to seismic lines appear to be associated with geological features where hydrocarbons might migrate and escape to the sea floor.

Figure 15 - Play types of Sumba offshore, Savu Basin and northwestern limit of Australia Shelf (courtesy of CGGVeritas and Directorate General of Oil and Gas, Indonesia). The play types of Sumba area among others: tilted fault block related to rifted basin and draped channel sands overlying the basement high related to promontory of Sumba Island.

Figure 16 - Stratigraphic succession between Sumba area (western part, basically similar to the central part and it has deeper facies for eastern part of Sumba) and South Sulawesi, an expected provenance for Sumba terrane. Based on the geology of the island (see discussions in the text on petroleum implications), Sumba has potential source rocks, reservoirs, and seals. The petroleum system for Sumba area can be referred to South Sulawesi or South Makassar petroleum systems with similar processes of volcanism, rifting, postrifting, and related sedimentation.